The effect of CP 68,722, a thiozolidinedione derivative, on insulin sensitivity in lean and obese Zucker rats

The .Effect of CP 68,722,

a Thiozolidinedione Lean and Obese

Derivative, Zucker Rats

on Insulin

Sensitivity

in

Lorrie Bowen, Peter P. Stein, Ralph Stevenson, and Gerald I. Shulman The effect of a new drug (CP 68,722, Pfizer) on parameters of insulin sensitivity in an established insulin-resistant animal model was examined. Rates of hepatic glucose production (HGP) and peripheral glucose uptake in obese Zucker (fa/fa) rats treated with a IO-day course of the medication using a two-step (2 and IO mU/ kg/min) euglycemic hyperinsulinemic clamp technique were measured. In addition, changes in substrate concentrations after drug treatment were examined. Basal HGP rates were similar in the lean versus the obese animals (37 * 3 v 39 2 3 pmol/kg/min); however, the obese animals had impaired insulin-induced suppression of HGP at both 2 mU/kg/min (36 + 3 v 23 2 4 kmol/kg/min) and IO mU/ kg/min (18 t 5 v2 2 1 pmol/kg/min). Insulin stimulation of glucose disposal was also defective in the obese animals (37 -c 2 v 88 -t 7 rmol/kg/min at 2 mW/ kg/min and 98 t 9~219 f 18 pmol/kg/min at IO mU/kg/min). In addition, obese animals had elevated free fatty acid (FFA) and ketone levels, both of which were resistant to insulin-induced suppression. After drug treatment, few alterations in glucose or lipid metabolism were found in the lean animals. In the obese animals, insulin suppression of HGP was normalized during the higher insulin infusion rate (0 v 18 _f 5 pmol/kg/min at IO mU/kg/min), and peripheral glucose disposal was enhanced at both steps of the insulin clamp (54 2 4 v 37 ? 2 pmol/kg/min at 2 mU/kg/min and 134 ? 12 v 98 ? 9 pmol/ kg/min at IO mU/ kg/min). Drug treatment lowered both fasting and insulin-stimulated suppression of FFA and ketone concentrations in the obese rat. An unexpected finding was that drug treatment resulted in a marked acceleration of insulin clearance, such that basal insulin levels were reduced, as were insulin levels during the first steps of the insulin clamp. These results suggest that CP 68,722 acts at both the liver and skeletal muscle tissues to enhance insulin sensitivity, and thus may prove to be a useful medication in patients with insulin-resistant diabetes mellitus. Copyright 0 1991 by W.B. Saunders Company

H

YPERGLYCEMIA in patients with non-insulindependent diabetes mellitus (NIDDM) is a result of both impaired insulin release and peripheral resistance to the action of insulin.’ The present generation of oral hypoglycemic agents, the sulfonylurea compounds, enhance pancreatic p-cell release of insulin as their major mechanism of action.’ However, the insulin resistance in both liver and skeletal muscle tissues in NIDDM patients suggests the need to develop drugs that enhance insulin sensitivity at these sites. CP 68,722 (CP) is a novel compound that has some structural similarities to ciglitazone and CS-045. These compounds have been reported to have a prominent hypoglycemic effect’.‘; however, their mechanism of action is poorly defined. To examine the site of action of this class of drugs, we studied the effect of CP on obese and lean Zucker rats. The obese (fa/fa) Zucker rat is a wellestablished model of insulin resistance. While these obese animals are not hyperglycemic, prior studies have shown defects in sensitivity to insulin, similar to those found in NIDDM patients, in both the liver and skeletal muscle tissues.’ To examine the site of action of CP, we studied groups of lean and obese Zucker rats with and without drug treatment, using the hyperinsulinemic euglycemic clamp technique.’ METHODS Animals

thyl] thiazolidine-2.4.dione, (Pfizer, Groton. CT), was suspended in 0.25% methylcellulose and administered via gavage to lean (n = 9) and obese (n = 11) Zucker rats at a dose of 50 mg/kg/d for a IO-day course. Untreated lean (n = 6) and obese (n = 7) mice were used as control groups. Etglycemic

Hyperinsulinemic

Clamps

After an overnight fast, the animals underwent a euglycemic hyperinsulinemic clamp, as previously described.” In the present study. a two-step insulin infusion protocol was used: 2 mU/kg/min was infused for the first 90 minutes. followed by a an infusion of 10 mU/kg/min for the 90- to 180-minute period. Both insulin steps were preceded by a priming insulin infusion to rapidly increase plasma insulin concentration to levels close to the plateau level of the constant infusion. A prime (6 PCi) continuous (0.1 PCiimin) infusion of 3-‘H-glucose was started 60 minutes before the insulin infusion to measure rates of hepatic glucose production (HGP).” Anafytic Procedures Plasma glucose was measured by the glucose oxidase method (Beckman Instruments. Palo Alto. CA). Plasma insulin was measured by radioimmunoassay using rat insulin standards. Plasma 3-‘H-glucose radioactivity was measured in the supernatants of barium hydroxide-zinc sulfate precipitates of plasma samples after evaporation to dryness to eliminate tritiated water.X 3-P-hydroxybutyrate concentrations were determined by enzymatic assays using a kit purchased form Sigma Chemical (St Louis, MO), and nonesterified fatty acids were determined using a kit (Nescauto NEFA Kit-U) purchased from Nippon-Shoji Kaisha (Osaka, Japan).

Zuck;er obese (faifa) and lean (fai-) rats were obtained from Charlex River (Raleigh, NC), housed on a 12.hour light/l2-hour dark cycle, and given free access to water and rat chow (Prolab 3000. &way, Syracuse, NY). Lean (74 to 110 days old) and obese (54 to 60 days old) animals were weight-matched. Four to 5 days before insulin clamp study, flexible plastic catheters were placed in an internal jugular vein and carotid artery. as previously described.’ These catheters were tunneled subcutaneously and fixed in place on the back of the animals head using tape and surgical clips. CP, (t)-5-[(3.4-dihydro-2-phenylmethyl-2H-l-benzopyran-6-yl) me-

Metabo//sm,

Vol 40, No 10 (October), 1991: pp 1025-1030

From the Secfion of Endocrinology and Metabolism, Yale Universit) School of Medicine, New Haven, CT; and Central Research Section, Pfizer Inc, Groron, CT Address reprint requests to Gerald I. Shulman, PhD, MD, Fitkin 104, Section of Endocrinology and Metabolism, 3.33 Cedar St, New Haven, CT 06510. Copyright 0 1991 by W.B. Saunders Companv 0026-0495191 /JOlO-0006$03.0010 1025

BOWEN ET AL

1026

Calculations

Table 1. Baseline Weights, and Glucose and Insulin Values

Steady-state rates of HGP and peripheral glucose uptake (GU) were taken during the last 30-minute periods of each step of the clamp, ie, 60 to 90 minutes for the 2.mU/kg/min infusion rate and 150 to 180 minutes for the 10.mU/kg/min infusion rate. The rate of HGP was calculated by subtracting the exogenous glucose infusion rate necessary to maintain euglycemia from the rate of appearance of glucose measured by tritiated glucose.’ All values are given as the mean value ? SEM. Comparisons within groups at different insulin concentrations were made with a paired t test, while comparisons between groups were made with an unpaired twotailed t test (Statview II, Abacus Concepts); P values less than .05 were considered significant RESULTS Glucose Metabolism in Lean and Obese Untreated Animals

Baseline characteristics. Fasting insulin (Fig 1) and glucose concentrations (Table 1) were slightly, but not significantly, higher in the obese animals. Pretreatment fasting insulin levels were 93 t 22 in the obese animals versus 50 ? 10 uU/mL in the lean animals. Insulin levels significantly increased in the obese and the lean animals during both the 2-mU/kg/min (189 + 54 uU/mL v 174 + 11 kU/mL) and the lo-mU/kg/min insulin infusion steps (770 ? 143 kU/mL v 612 * 40 uU/mL). Hepatic g&ose production. Basal HGP was similar in the lean (37 ? 3 p,moI/kglmin) and obese animals (39 ? 3 umol/kg/min) (Fig 2). During the 2-mU/kg/min insulin infusion rate, the obese animals had significantly less suppression of HGP compared with the lean animals (36 ? 3 v 23 * 4 ~moI/kg/min,P < ,025). Similarly, during the lo-mU/kg/min insulin infusion step, obese animals had a marked defect in insulin suppression of HGP (18 f 5 v 2 -+ 1 umol/kg/min, P < .005). Thus, at both intermediate and high insulin concentrations, obese animals displayed a marked hepatic resistance to insulin with incomplete suppression of HGP (Fig 2). Whole body glucose uptake. As noted above, basal glucose and HGP were similar in the obese and lean animals,

Lean/No

Fasting glugose (mg/dL) Fasting insulin (kU/mL)

Rx Obese/No

102 ? 5 50 f 10

Rx

116k9 92 k 22

Lean/Rx

106 -t 3

Obese/Rx

101+3

10 -t 2’

17 2 2’

352 2 14

317 k 11

Weights (g) Pre

275 + 6

Post

263 k 11 -

320 + 16* 311 + 13

*P < .05 v pretreatment weights. ‘P < ,005 v untreated groups.

indicating that basal whole body glucose uptake was also similar (Fig 3). In contrast, at both rates of insulin infusion, whole body glucose disposal was significantly less in the obese (37 ? 2 pmol/kg/min at 2 mU/kg/min and 98 +- 9 umol/kg/min at 10 mU/kg/min) compared with the lean animals (88 ? 7 kmol/kg/min at 2mU/kg/min and 219 * 18 pmol/kg/min at 10 mU/kg/min, both P < .OOOl v obese). In the obese animals, glucose uptake was not significantly increased during the 2-mU/kg/min insulin infusion, but did significantly increase during the lo-mU/kg/min insulin infusion (P < .Ol). Nonetheless, glucose uptake was still less than half of that seen in the lean animals during both steps of the insulin clamp. Substrate concentrations. Basal I3-hydroxybutyratc levels were slightly, but not significantly, higher in the obese compared with the lean animals. During both insulin infusion steps, the obese animals had significantly (P < .02) greater ketone levels than the lean animals, and the 2-mU/kg/min insulin infusion rate failed to decrease ketone levels below fasting concentrations (Table 2). Free fatty acid (FFA) levels were significantly higher in the obese compared with the lean animals at basal and during both insulin infusion steps (Fig 4). FFA levels during the 2-mU/kg/min insulin infusion rate were not suppressed below fasting concentrations, and even at the higher insulin infusion rate, FFA levels declined by less than 30%. In contrast, the 2-mU and 10 mU/kg/min infusion rates induced a 40% and 75% suppression FFA levels. respectively, in the lean animals. Response to Drug Treatment

Inaulln

+ pso.05n untrMt.d

CU/ml)

0

2

10

lnrulln lnfurlon Rat0 (mu/kg-mln) Fig 1. Insulin concentrations at fasting and during insulin infusion. Drug treatment significantly reduced fasting insulin concentrations and during both steps of the insulin clamp.

As shown in Table 1, the lean animals had a significant loss of weight over the lo-day course of treatment. In comparison, the weight of the obese animals did not significantly change with drug treatment. Insulin concentrations. After drug treatment, fasting insulin concentrations decreased markedly in both lean (11 +2 kU/mL. P < .005 v untreated) and obese (19 t 2 yU/mL, P < ,001 v untreated) animals (Fig I), but remained higher in the obese compared with the lean animals (P < .05). During the 2-mU/kg/min insulin infusion, insulin levels slightly increased in the lean group (21 f 2 p,U/mL, P < .Ol v basal insulin levels), but did not signiticantly increase in the obese animals (22 ? 2 pU/mL). In contrast, the lo-mU/kg/min insulin infusion rate produced a significant increase in insulin concentrations in both the lean (255 t 31 uU/mL, P < .OOl) and obese (391 t 46 yU/mL, P < ,001) animals. Nonetheless, insulin levels

THlOZOLlDlNEDlONE

EFFECTS ON INSULIN SENSITIVITY

l

Glucose

3o

1027

PC0.001M ““vealed

# p’ 0.025M. ObeS#

Drug treatment resulted in a significant decline in basal FFA levels in the obese (P < ,002) animals (Fig 4). Moreover, FFA levels were lower in the treated animals at both insulin infusion rates compared with the untreated group (P < .0025). However, the 2-mU/kg/min infusion rate did not decrease FFA levels below basal levels in either the untreated or treated groups. DISCUSSION

2

0

huh

10

lnfurion Rate (mllkg-mln)

Fig 2. Hepatic glucose production as measured by ‘H-glucose infusion during the euglycemic insulin clamp. Basal HGP was similar in the obese and lean animals, but insulin infusion reduced HGP significantly less in the obese animals. Drug treatment normalized HGP in the obese animals during the lo-mU/kg/min insulin infusion.

were still only about 50% of the levels found in the untreated animals during the same steps of the insulin clamp. Glucose metabolism. As shown in Fig 2, drug treatment appeared to have no effect on basal HGP or insulin suppression of HGP in the lean animals. In addition, CP did not alter insulin-stimulated glucose uptake or insulin suppression of ketone concentrations (Table 2). Basal FFA levels were reduced in the treated animals compared with the untreated group (P < .05); however, drug treatment did not reduce FFA levels at either step of the insulin clamp (Fig 4). In obese animals, after drug treatment, both basal HGP (37 + 6 umol/kg/min) and HGP at 2 mU/kg/min (17 ? 10 p,mol/kg/min) were slightly, but not significantly, decreased (Fig 2). In contrast, drug treatment normalized insulin suppression of HGP in the obese animals during the 10 mU/kg/min insulin infusion rate. Drug treatment resulted in a significant increase in glucose uptake during the first step of the insulin clamp (54 L 4 umol/kg/min, P < .OOl v basal, P < .00.5 v untreated), while in the untreated obese animals the same insulin infusion rate failed to increase glucose uptake (Fig 3). Similarly, glucose uptake was 134 2 12 pmol/kg/min during the second step of the clamp, significantly increased compared with untreated obese rats (P < .05). Despite the increase in insulin-stimulated glucose uptake induced by drug treatment, obese animals still had diminished uptake compared with drug-treated lean animals. Drug treatment did not alter the elevated basal P-hydroxybutyrate concentrations in the obese animals (Table 2). The 2-mU/ kg/min insulin infusion resulted in a significant decline in ketone levels compared with the untreated obese animal:s. Nonetheless, ketone concentrations remained elevated compared with lean animals. At the lo-mU/kg/min insulin infusion rate, ketone levels were not different after drug treatment.

Thiazolidinedione derivatives are a group of drugs that have previously been shown to have a prominent hypoglycemic effect in a number of animal mode1s7’ In the present study, we have examined the site of action of a new compound, CP 68,722, in an insulin-resistant animal, the obese Zucker rat. As noted above, the Zucker obese rat is a model of insulin resistance. While these animals are not hyperglycemic, they have sites of insulin insensitivity similar to those previously reported in patients with NIDDM, and hence provide a model of how medications might improve sensitivity to insulin in other insulin-resistant states such as NIDDM. Prior studies of obese Zucker rats have identified several defects in insulin action. Terrettaz et al’ performed euglycemic insulin clamps on anesthetized lean and obese Zucker rats. These investigators found that basal HGP and glucose metabolism were similar in the lean and obese animals. Insulin infusion failed to stimulate any increase in glucose metabolism in the obese animals, while increasing glucose uptake in lean animals by about threefold to fourfold. HGP was also resistant to insulin, with a marked shift to the right of the insulin dose-response relationship; however, normal insulin suppression of HGP was achieved at high insulin concentrations. Other investigators studying isolated soleus muscle preparations have also reported markedly depressed insulin-stimulated glucose metabolism in obese Zucker rats.‘“,” In the present study, we found that insulin resistance was

I Perlphoral Glucow Uptrko

a

#T -r,

Lean-umrulad

150

(woVkg-mln)

0

2

10

In8ulln Infudon Rate (mu/kg-mln) Fig 3. Peripheral glucose disposal during the euglycemic insulin clamp. Insulin-stimulated glucose disposal was markedly impaired in the obese animals, but was improved by drug treatment.

BOWEN

1028

Table 2. p-Hydroxybutyrate

Concentrations

(pEq/ L),

Fasting and During Euglycemic Insulin Clamp 2 mllikglmin

Fasting

Groups

Lean

4,709

? 1,056

317 + 96

Lean-treated

3,460

+ 380

750 + 231

10 mU/kglmin

48 ‘- 19 86 k 38

Obese

7,082

+ 1,922

5,795

+ 2,306

451 + 144

Obese-treated

7,082

+ 721

3,460

2 384’

250 ? 48

*P < .05 v untreated.

present in both liver and skeletal muscle tissues in the obese Zucker rat. In contrast to the results of Terrettaz et al, we found that peripheral glucose uptake is significantly stimulated in both obese and lean animals, albeit with a markedly lower insulin responsiveness in the obese rats. Moreover, while Terrettaz et al found that maximal insulin levels increased glucose uptake by threefold to fourfold, we noted an increase of about sevenfold. The difference between our results and those previously reported may relate to the conditions under which the animals were studied. Our animals were awake and were not acutely stressed. In contrast, Terrettaz et al studied anesthetized animals in whom catheters were placed immediately prior to insulin clamp study. Both stress and anesthesia can depress glucose metabolism, and probably explain the differences between these studies. In addition, Terretaz et al examined lean and obese animals that were age-matched rather than weight matched as in the present study. Such differences in comparison groups could certainly have modified the differences in glucose metabolism between these groups of animals. Another difference between the present study and the study by Terrettaz lies in insulin suppression of HGP. We found that maximum suppression of HGP was approximately 50%, whereas Terrettaz report a shift to the right of the insulin dose-response relationship for HGP with normal suppression at high insulin levels. Since Terrettaz et al achieved much higher insulin levels, the dose-response

3ow FFA Conoentratlon

Wp/mL) x00

loco

0

2

10

Ineulln Infuolon Rate (mu/kg-mln) Fig 4. FFA concentrations in the basal state and during insulin infusion. In the basal state and during both insulin infusion rates, drug-treated obese animals had lower FFA levels than untreated obese animals.

ET AL

relationship previously reported and reported here may be similar. Thiazolidinedione derivatives have not. as yet, been extensively evaluated as to their site of action. Prior studies have examined the effect of both acute and chronic administration of these drugs on glucose, insulin and lipid metabolism in a number of different animal model systems. Chang et al studied ob/ob genetically obese mice and found that treatment with ciglitazone reduced fasting glucose towards normal and improved the hypoglycemic response to insulin.‘,’ In a separate study, Fujita et al’ studied genetically obese yellow KK mice and obese Zucker rats. Steady-state blood glucose (SSBG) and steady-state plasma insulin (SSPI) were measured as an index of insulin sensitivity during constant infusion with insulin, glucose. epinephrine and propranolol. These studies showed that SSBG declined with drug treatment, suggesting enhanced peripheral sensitivity. Tritiated glucose was infused in this study; however, HGP was not reported. These investigators suggested that, since specific activity was unchanged, no effect of cightazone on HGP was likely. Fujiwara ct al” examined the effect of a thiozolidinedione derivative (CS045) in diabetic KK mice, obiob mice, and streptozotocindiabetic mice. The insulin-resistant KK and obiob mice responded to drug treatment with a significant lowering of plasma glucose levels. The insulin-deficient streptozotocin mice showed no effect of drug therapy. These studies, then, document a hypoglycemic effect of thiozolidinedione derivatives, and suggest a prominent peripheral site of action. In addition, a recent study by Caro et al examining the effect of ciglitazone on parameters of glucose and fat metabolism in isolated hepatocytes failed to show a significant response to the drug in the presence or absence of insulin.” In the present study, CP 68,722, a thiozolidinedione derivative. induced a significant improvement in both insulin suppression of HGP and in insulin-stimulated glucose uptake in obese, insulin resistant rats. HGP was normalized in the obese animals during the higher insulin infusion rate at insulin concentrations, which were substantially lower than in the untreated obese animals. Suppression of HGP during the 2-mU/kg/min insulin infusion was not enhanced after drug treatment. Nonetheless. drug treatment resulted in a marked lowering of insulin levels during the 2-mu/kg/ min insulin infusion and, hence, HGP was suppressed to an extent similar to that seen in the untreated animals, but at much lower insulin levels, suggesting enhanced insulin sensitivity. Drug treatment induced a significant increase in glucose disposal at both intermediate and high insulin infusion rates. However, a significant defect still was present in the obese treated animals in comparison to treated or untreated lean animals. These results suggest that the insulin resistance in obese animals is more readily reversed in liver than in skeletal muscle by CP treatment. No effect of drug therapy was evident for either fasting HGP or glucose disposal in the lean animals. Again, insulin levels were much lower in the treated animals; the lower insulin concentrations during both steps of the insulin clamp in the drug-treated lean animals obscures conclusions about alterations in insulin sensitivity. As the same

THIOZOLIDINEDIONE

1029

EFFECTS ON INSULIN SENSITIVITY

HGP and glucose uptake occurred at lower insulin concentrations, enhanced insulin sensitivity, even in the lean animals, may be induced by drug treatment. Drug treatment induced a favorable alteration in FFA in the obese animals, with both fasting and insulin-stimulated FFA levels lower in the drug-treated animals. A similar improvement in FFA levels has been reported previously with ciglitazone’ and CS-045.” This response of FFA levels to drug treatment could explain some of the beneficial effect of the medication on glucose disposal. Since the studies by Randle et al.” it has been known that high levels of FFA can impair glucose metabolism, while lowering FFA levels can enhance insulin sensitivity. This mechanism could account for the improvement in glucose disposal seen during the insulin clamp; however, fasting FFA levels were significantly lowered in the obese animals without an improvement in glucose disposal. The lower insulin levels after drug treatment could account for this apparent lack of increase in glucose disposal despite lower FFA levels. Drug t.reatment also resulted in a significant decrease in ketone levels in the obese animals during the first step of the insulin clamp. This improvement could reflect enhanced insulin action at the level of the liver with diminished ketogenesis, decreased substrate (FFA) delivery to the liver, or enhanced peripheral utilization. Since turnover studies were not performed, differentiating increased utilization from diminished production is not possible. Since hepatic sensitivity to insulin, as reflected by HGP, was not significantly improved at the 2-mU/kg/min infusion rate, enhanced hepatic effect of insulin seems to be an unlikely explanation. Since drug treatment resulted in a significant increase in glucose uptake at the 2-mU/kg/min infusion rate, increased utilization could explain the diminished ketone 18evels. Further studies are necessary to define the mechanism of this response. The effect of thiozolidinedione derivatives on insulin levels has been seen in previous studies. Fujita et al” reported a decrease in fasting insulin levels of from 30% to 50% in Sprague-Dawley rats, and an approximately 60% decrease in obese Zucker rats. In this same report, no effect on fasting insulin levels was found in dogs. Chang et al found an approximately 60% decrease in fasting insulin levels in obiob mice, but no significant decrease in control animals.’ Finally, Fujiwara et al found an approximately 75% decline in insulin concentrations after 10 days of drug treatment with CS-045.” In the present report, we found a

similar decline in basal insulin concentrations in both lean and obese Zucker rats. In addition, we found that during both steps of the insulin clamp, insulin concentrations were significantly lowered in the drug-treated animals. The 2-mU/kg/min insulin infusion did not produce a significant increment in insulin concentrations in either the lean or obese animals. Enhanced glucose disposal and decreased HGP may partially explain the decline in insulin concentrations in the fasting state; however, the lower levels during insulin infusion strongly suggest an effect of the drug on insulin clearance. The site of this alteration cannot be determined from the present results. However, the process increasing insulin clearance appears to be saturable, since the step up from 2- to lo-mU/kg/min infusion rate increased insulin levels by an amount similar to the increment seen in untreated animals. Normally, the liver metabolizes approximately 50% of insulin, while the kidney and insulin-target organs (ie, adipocytes. skeletal muscle, etc) account for the rest. Hence, increased insulin clearance at any, or all, of these sites could account for our observations. Whether the effect the drug has on insulin clearance and the drug’s effect on insulin sensitivity are linked is also unknown. However, enhanced receptor internalization. for example. could account for both alterations.“‘,‘5 An unexpected finding in the present study was the drug-induced weight loss that occurred in the lean but not the obese animals. This effect was seen in all of the lean-treated animals. Prior studies of ob/ob mice have suggested a slowing of weight gain in the treated animals, possibly due to a depression of food intake.’ However, a decrease in weight in a control group of animals has not been reported. Our lean-treated animals appeared healthy, and continued to eat and drink during the period of treatment. We do not have data on actual food consumption, and so conclusions regarding the etiology of the weight loss are not possible. In conclusion, the present study has shown that the obese Zucker rats has defects in both insulin suppression of HGP and in insulin stimulation of glucose disposal. Treatment with CP resulted in improved hepatic and peripheral sensitivity to insulin. In addition, with both fasting and during insulin infusion, FFA and ketones were diminished in the treated animals. Finally, CP had a dramatic effect to decrease insulin levels in the fasting state and during insulin infusion in both groups of animals.

REFERENCES 1. DrFcronzo R: The triumvirate: Beta cell, muscle, liver. A collusion responsible for NIDDM. Diabetes 37:667-685. 1988 2. Gutniak M, Karlander S, Efendic S: Glyburide decreases insulin requirement, increases beta-cell response to mixed meal, and does not affect insulin sensitivity: Effects of short- and long-term treatment in secondary failure to sulfonylurea. Diabetes Care 10:5’45-554. 1987 3. Chang A. Wyse B. Gilchrist B. et al: Ciglitazone, a new hypoglycemic agent. 1. Studies in ob/ob mice, diabetic Chinese hamsters, and normal and streptozotocin-diabetic rats. Diabetes 32:830-838. 1983 4. Chang A. Wyse B, Gilchrist B: Ciglitazone, a new hypoglyce-

mic agent. II. Effect on glucose and lipid metabolisms and insulin binding in the adipose tissue of C57BL/6J-obiob and ?/‘? mice. Diabetes 32:839-845, 1983 5. Fujita T. Sugiyama Y, Taketomi S, et al: Reduction of insulin resistance in obese and/or diabetic animals by 5-[4-( l-methylcyclohexyl-methoxy)benzyl]-thiazolidine-2.4 dione (ADD-3878, U-63,287, Ciglitazone), a new antidiabetic agent. Diabetes 32:804810.1983 6. Fujiwara T, Yoshioka S, Ushiyama I, et al: Characterization of new oral antidiabetic agent CS-045. Diabetes 37: 1549-1558, 1988 7. Terrettaz J, Assimacopoulos-Jeannet F. Jeanrenaud B: Severe hepatic and peripheral insulin resistance as evidenced by

BOWEN ET AL

euglycemic clamps in genetically obese faifa rats. Endocrinology 118:674-678, 1986 8. Rossetti L. Smith D, Shulman G, et al: Correction of hyperglycemia with phlorizin normalizes tissue sensitivity to insulin in diabetic rats. J Clin Invest 79:1510-1515, 1987 9. Steele R: Influences of glucose loading and of injected insulin on hepatic glucose output. Ann NY Acad Sci 82:420-430, 1959 10. Crettaz M, Prentki M. Zaninetti D, et al: Insulin resistance in soleus muscle from obese Zucker rats. Involvement of several defective sites. Biochem J 186:525-534, 1980 11. Czech M, Richardson D, Becker S, et al: Insulin response in skeletal muscle and fat cells of the genetically obese Zucker rat. Metabolism 27:1967-1980, 1978 (suppl2)

12. Caro J, Ittoop 0, Sinha M: Giyburide but not ciglitazone enhances insulin action in the liver independent of insulin receptor kinase activation. Metabolism 38:606-611, 1989 13. Randle P. Garland P, Hales C, et al: The glucose-fatty acid cycle: Its role in insulin sensitivity and the metabolic disturbance of diabetes mellitus. Lancet 1:785-789, 1963 14. Jochen A, Berhanu P: Insulin-stimulated glucose transport and insulin internalization share a common postbinding step in adipocytes. Diabetes 36:542-545. 1987 15. Veda M, Robinson F. Smith M. et al: Effects of monensin on insulin processing in adipocytes: Evidence that the internalized insulin-receptor complex has some biological activities. J Biol Chem 260:3941-3946.1985