Search

The efficacy of the calcium channel blocker nitrendipine in preventing early renal and glomerular hypertrophy and increased urinary albumin excretion (UAE) was studied in experimental diabetes in rats, starting treatment at the onset of diabetes. Female Wistar rats were randomised into four groups: diabetic and non-diabetic rats were given either placebo or nitrendipine (250 mg/kg) in the diet for 8 weeks. After 8 weeks the kidneys in the diabetic animals had increased significantly compared to the non-diabetic controls. In the diabetic nitrendipine-treated animals renal and glomerular hypertrophy was significantly smaller than in the diabetic placebo-treated group (p < 0.05). After an initial increment within the first week, the UAE remained constant throughout the study period in the diabetic nitrendipine-treated animals, while a steady increase was seen in the diabetic placebo-treated group (p < 0.05). No differences were seen in systemic blood pressure between calcium- channel-blocker-treated groups and placebo-treated groups. In conclusion, administration of nitrendipine to diabetic rats for 8 weeks had a significant inhibitory effect on renal and glomerular hypertrophy, and showed a tendency towards a reduction in UAE (p = 0. 06) without affecting metabolic control or systemic blood pressure.

Renal and glomerular growth is inherent in early human and experimental diabetes frequently followed by later increase in urinary albumin excretion (UAE). Treatment with angiotensin converting enzyme (ACE) inhibitors has proven effective in delaying progression of human and experimental diabetic renal changes, and so has somatostatin analog treatment in experimental diabetes. The aim of the present study was to investigate three weeks of octreotide and captopril treatment alone or in combination following three months of untreated experimental diabetes, and compare the effects to those of insulin treatment. Diabetes induced significant increases in renal and glomerular growth and urinary albumin excretion. Octreotide and captopril alone and in combination reduced renal but not glomerular size, and the combined administration reduced UAE. None of these schedules affected blood glucose levels. Insulin treatment inducing euglycemia significantly reduced renal and glomerular size and UAE. In conclusion, insulin treatment with normalization of the diabetic metabolic derangement nearly normalizes renal and glomerular growth and UAE after three months of untreated diabetes. The combined treatment of octreotide and captopril was also followed by a significant decrease in renal growth and reduction in UAE compared to placebo treatment without affecting the metabolic control of the diabetic animals.

Growth hormone (GH) and insulin-like growth factor I (IGF-I) may play a role in early diabetic renal and glomerular growth and in the later development of experimental diabetic kidney disease. Rats from a genetic GH/IGF-I-deficient dwarf rat strain were made streptozotocin diabetic and were compared with nondiabetic dwarf rats. GH/IGF-I-intact rats with and without diabetes served as controls. After 6 mo of diabetes, kidney weight and total glomerular volume increased significantly in GH/IGF-I-intact diabetic rats compared with the nondiabetic GH/IGF-I-intact rats (P < 0.05), whereas the diabetic dwarf rats had insignificant changes compared with dwarf control rats. By the end of the study, urinary albumin excretion (UAE) increased from similar base levels of approximately 15-20 micrograms/24 h to 473 +/- 52 (SE) micrograms/24 h in GH/IGF-I-intact diabetic rats compared with 151 +/- 32 micrograms/24 h in diabetic dwarf rats (P < 0.01). In conclusion, isolated GH/IGF-I deficiency reduces the degree of renal and glomerular hypertrophy and the increase in UAE after 6 mo of experimental diabetes in GH/IGF-I-deficient rats.

Initial diabetic renal hypertrophy is preceded by a transient increase in kidney insulin-like growth factor I(IGF-I). In the present study streptozotocin-diabetic rats were treated with a new somatostatin analogue (lanreotide), insulin or placebo for 7 days and compared to non-diabetic control rats. Kidney IGF-I changes were examined in the renal cortex, medulla, and whole kidney homogenates. The renal cortex contained approximately 5.5 times more IGF-I than the renal medulla (p < 0.01); further, IGF-I increased transiently and more pronouncedly in the renal cortex compared to the medulla. Lanreotide treatment significantly inhibited diabetic renal and glomerular growth compared to placebo-treated diabetic rats. Further, lanreotide treatment was followed by a significantly lower medulla IGF-I by day 2, while lanreotide had no effect on cortex IGF-I accumulation. In conclusion, IGF-I accumulated transiently and was more pronounced in the real cortex compared to the renal medulla and, further, lanreotide prevented diabetic renal and glomerular growth bringing new evidence that intervention with somatostatin analogues may have a role in the prevention of experimental diabetic kidney disease.

Female nondiabetic and streptozotocin diabetic Wistar rats received 200 units heparin two times daily by subcutaneous injection for 6 mo. Mesangial volume fraction was reduced in heparin-treated control (CH) compared with untreated control (C) animals (CH 0.18 +/- 0.02 vs. C 0.24 +/- 0.02, P < 0.05), but other histological parameters were similar. In the heparin-treated diabetic (DH) group, wet kidney weight was increased compared with the untreated diabetic (D) group (DH 1,156 +/- 39 vs. D 1,050 +/- 30 mg, P < 0.05), as were absolute, but not fractional, glomerular volume (P < 0.05) and capillary volume (P < 0.05). Basement membrane thickness (DH 193 +/- 3 vs. D 231 +/- 9 nm, P < 0.01) and mesangial/glomerular volume fraction (P < 0.001) were decreased. Urinary albumin excretion was increased in the heparin-treated control animals compared with control animals [CH 980 (range 150-4,448) vs. C 221 (range 86-654) micrograms/24 h, P < 0.001] and in the heparin-treated diabetic animals compared with the diabetic animals [DH 12,785 (range 4,495-29,520) vs. D 899 (range 450-1,335) micrograms/24 h, P < 0.001]. Thus the possibly deleterious increases in glomerular capillary volume and albumin excretion may negate the beneficial effects of heparin on mesangial and basement membrane structures.

IGF-I acts as a renotropic factor in early streptozotocin-induced diabetes. Somatostatin analogue (octreotide) treatment initiated at the onset of diabetes prevents kidney IGF-I accumulation and renal growth. Seven days of octreotide treatment initiated after 3, 5, 7 or 9 days of untreated diabetes was investigated. Diabetic renal hypertrophy was followed by renal hyperplasia. Compared with placebo-treated diabetic rats, the earliest octreotide intervention was followed by a greater reduction in renal growth compared with intervention later on (days 3 to 10, 12%; days 5 to 12, 10%; days 7 to 14, 9%; days 9 to 16, 6%; P < 0.05). Octreotide treatment was unable to reduce protein accumulation and kidney DNA increase consistently. No difference in glomerular volume fraction or total glomerular volume was observed between placebo- and octreotide-treated diabetic rats. Octreotide treatment was followed by reduced kidney and serum IGF-I especially following early intervention, while no effect over that of diabetes was observed in the later intervention periods. The results confirm the notion that initial renal IGF-I accumulation is a prerequisite for early diabetic kidney hypertrophy in rats and show that delayed octreotide treatment cannot reverse renal and glomerular growth which is already manifest.

Treatment of diabetic rats with octreotide can inhibit early diabetic renal hypertrophy. Octreotide administration for 6 months from the day of diabetes induction inhibits renal hypertrophy and diminishes increase in urinary albumin excretion. To investigate the effect of octreotide on manifest diabetic renal changes, octreotide treatment was given for 3 weeks after an untreated diabetic period of 3 or 6 months. In addition, following 6 months of diabetes, a group of diabetic rats was treated with insulin for 3 weeks. Renal and glomerular hypertrophy, and increased urinary albumin excretion were observed in diabetic rats compared to non-diabetic control rats from 3 months and throughout the study period. Octreotide treatment did not affect body weight, food intake, blood glucose or serum fructosamine levels. We observed no effect of octreotide treatment on renal and glomerular hypertrophy or urinary albumin excretion compared to placebo-treated diabetic rats. Insulin treatment for 3 weeks after 6 months of untreated diabetes normalized blood glucose and serum fructosamine levels, and furthermore renal hypertrophy was significantly diminished compared to the placebo-treated diabetic rats. However, insulin treatment had no effect on glomerular hypertrophy or urinary albumin excretion. In conclusion, octreotide treatment for 3 weeks following an untreated diabetic period of 3 or 6 months is unable to reduce the increased renal and glomerular volume or urinary albumin excretion. However, insulin treatment for 3 weeks with induction of euglycaemia diminishes the renal hypertrophy but has no effect on glomerular volume or urinary albumin excretion.

Insulin-like growth factor I (IGF-I) has been proposed as a renotropic factor in initial diabetic kidney growth. To examine the effects of an isolated growth hormone (GH) and IGF-I deficiency on diabetic renal hypertrophy, dwarf rats were made diabetic and studied over a period of 7 days. Diabetic dwarf rats treated with human GH (hGH) or insulin and diabetic rats with intact pituitary were used as controls. In diabetic control animals kidney weight had increased by day 2 (P less than 0.01), and the increase amounted to 27% after 7 days, whereas untreated diabetic dwarf rats had a slower and lesser degree of kidney weight increase, reaching significance on day 7 only, amounting to 8%. hGH administration in diabetic dwarf rats increased the kidney weight on day 7 when compared with untreated diabetic dwarf rats (P less than 0.05) and was 19% over that of insulin-treated diabetic dwarf rats. The glomerular volume had increased by 43% in untreated diabetic control rats at day 7, compared with a 29% increase in untreated diabetic dwarf rats (P less than 0.05). hGH administration in diabetic dwarf rats increased the glomerular volume by 46%, comparable to the increase seen in diabetic control animals. Kidney IGF-I was increased on day 2 by 51 and 46% in saline- and hGH-injected diabetic dwarf rats, respectively, but a significantly higher increase in kidney IGF-I amounting to 96% was seen in diabetic control rats.(ABSTRACT TRUNCATED AT 250 WORDS)

The insulin-like growth factors, insulin-like growth factor I and insulin-like growth factor II are bound to six distinct classes of insulin-like growth factor binding proteins (IGFBPs) in the circulation and in extracellular fluids. Diabetic renal hypertrophy is preceded by a transient increase in kidney insulin-like growth factor I suggestive of a renotropic function for insulin-like growth factor I. In order to examine a possible involvement of IGFBPs in initial diabetic kidney growth and in kidney insulin-like growth factor I accumulation, we studied rat kidney IGFBPs by ligand blotting during the first 4 days after induction of diabetes. Six distinct bands were identified in kidney and liver tissue with apparent molecular weight values of 38-47 (doublet), 34, 30, 24 and 20 kDa. The 38-47 kDa doublet band probably corresponds to the insulin-like growth factor binding subunit of IGFBP-3, the 24 kDa band to IGFBP-4 and the 30 kDa band to IGFBP-1 and/or IGFBP-2, as these IGFBPs in rats have similar molecular weight. In untreated diabetic rats a transient increase in the kidney 30 kDa band was demonstrable 24 h after induction of diabetes with a maximal rise (two-fold) after 48 h, followed by a decrease to baseline values after 4 days. In untreated diabetic rats the 38-47 kDa doublet band also increased (two-fold) in kidney during the first 2 days after induction of diabetes, followed by a subsequent decrease. Insulin-treatment prevented both the increase in the 30 kDa and in the 38-47 kDa bands.(ABSTRACT TRUNCATED AT 250 WORDS)

Initial renal hypertrophy in experimental diabetes is prevented by administration of a long-acting somatostatin analogue octreotide (SMS). To investigate the long-term effects of SMS on renal hypertrophy and urinary albumin excretion (UAE), streptozotocin-diabetic and non-diabetic rats were treated with two daily subcutaneous injections of SMS (100 micrograms x 2) for six months. Untreated diabetic and non-diabetic animals were used as reference groups. No differences were seen between the two diabetic groups in respect to body weight, food intake, blood glucose levels, urinary glucose output, hemoglobin A1C(HbA1C), fructosamine, serum growth hormone (rGH) or creatinine clearance, but kidney weight (896 +/- 36 vs. 1000 +/- 24 mg, P less than 0.02), UAE (417 +/- 131 vs. 1098 +/- 187 micrograms/24 hr, P less than 0.02), kidney insulin-like growth factor I (IGF-I) (167 +/- 16 vs. 239 +/- 17 ng/g, P less than 0.01) and serum IGF-I (301 +/- 26 vs. 407 +/- 17 micrograms/liter, P less than 0.01) were all reduced in the SMS-treated diabetic animals when compared to the untreated diabetic group. In non-diabetic rats SMS reduced body weight (274 +/- 3 vs. 293 +/- 5 g, P less than 0.01), kidney weight (695 +/- 9 vs. 764 +/- 17 mg, P less than 0.01), UAE (83 +/- 29 vs. 364 +/- 114 micrograms/24 hr, P less than 0.02), kidney IGF-I (202 +/- 12 vs. 280 +/- 12 ng/g, P less than 0.01), serum IGF-I (428 +/- 21 vs. 601 +/- 54 micrograms/liter, P less than 0.01) and serum rGH (67 +/- 6 vs. 126 +/- 27 micrograms/liter, P less than 0.05) when compared to untreated controls. When kidney weights were expressed in relation to body weight no difference was found between SMS-treated and untreated controls, while the difference between SMS-treated and untreated diabetic animals was still present (P less than 0.01). In conclusion, chronic administration of SMS has abating effects on diabetic renal hypertrophy and UAE, and thus indicates that SMS may reduce development of diabetic kidney lesions in experimental diabetes. The long-term suppressive effects of SMS on renal enlargement and UAE may in part be mediated through reduction in circulating and kidney IGF-I levels.

Initial diabetic renal hypertrophy is preceded by a transient increase in kidney insulin-like growth factor I suggesting that insulin-like growth factor I may be implicated in diabetic kidney growth. The present study was undertaken to examine the effects of exogenous insulin-like growth factor I infusion on diabetic renal hypertrophy at a time when renal insulin-like growth factor I concentration had returned to normal and the initial steep kidney growth rate had diminished to a much slower rate. Groups of rats with diabetes duration of 5 days were infused s.c. for 4 subsequent days with equimolar concentrations of insulin-like growth factor I (36 nmol/day) or insulin (35 nmol/day). Insulin infusion lowered blood glucose to a normal level within 2 days and induced an average body-weight gain of 9.3 +/- 0.6 g/day. Insulin-like growth factor I infused diabetic rats maintained the original diabetic state with blood glucose levels comparable to those of 0.154 mol/l NaCl-infused diabetic rats, but had nevertheless an average body-weight gain of 6.8 +/- 1.0 g/day while untreated diabetic rats had a lower body-weight gain amounting to 3.3 +/- 0.8 g/day (p less than 0.01). The kidney weight at day 9 in untreated diabetic animals was about 25% greater than that of non-diabetic control animals, while in insulin-like growth factor I treated diabetic rats a further increase (p less than 0.05) was seen, amounting to 36% above control level. No increase was seen in the insulin-treated diabetic group.(ABSTRACT TRUNCATED AT 250 WORDS)

We have measured specific binding of insulin-like growth factor I and growth hormone to renal plasma membranes from control, streptozotocin-diabetic, insulin-treated diabetic, uninephrectomised and combined diabetic-uninephrectomised male Wistar rats. Control, insulin-treated and uninephrectomised rats had similar body weights after 7 days (243 +/- 2 g), whereas diabetic and diabetic-uninephrectomised animals were significantly lighter (219 +/- 4 and 203 +/- 4 g, p less than 0.05). Blood glucose concentrations were similar in the diabetic and diabetic-uninephrectomised animals (around 26 mmol/l) but significantly lower in the insulin-treated group. Right kidney weight increased by 14% in the control, insulin-treated and sham-nephrectomised animals, by 33% in the diabetic group, 38% in thenephrectomised animals and 60% in the diabetic-nephrectomised group. The renal content of insulin-like growth factor I was similar and stable in the control, insulin-treated and sham-nephrectomised animals (208 +/- 14 ng/g wet weight) but rose to a peak of 669 +/- 35 ng/g in the diabetic group (p less than 0.001), 871 +/- 34 ng/g in the nephrectomised animals (p less than 0.001) and 1012 +/- 43 ng/g in the diabetic-uninephrectomised group (p less than 0.001). Maximum binding of insulin-like growth factor I fell on day 1 in the diabetic group (8.3 +/- 1.4 vs 5.2 +/- 0.71 x 10(-11) mol/l; p less than 0.01) but thereafter was identical to control animals.(ABSTRACT TRUNCATED AT 250 WORDS)

It has recently been demonstrated that immunoassayable kidney insulin-like growth factor I concentration increases 24-48 h after induction of diabetes, preceding the initial renal hypertrophy. To elucidate whether this increase is due to increased local production we studied rat kidney insulin-like growth factor I gene expression during the first four days after induction of streptozotocin diabetes. Eighteen hours after injection with streptozotocin the diabetic animals were divided into two groups, one of which was treated with insulin, and daily for four days animals from each group were taken out for investigation. After four days the wet kidney weight had increased from baseline by 20% (from 687 +/- 23 to 827 +/- 6 mg (mean +/- SEM), p less than 0.01) in the untreated diabetic group, while no significant increase occurred in the insulin-treated group (687 +/- 23 vs 732 +/- 21 mg, NS). Kidney insulin-like growth factor I increased rapidly from baseline, the rise amounting to 52% after 48 h (from 271 +/- 11 to 411 +/- 32 ng/g, p less than 0.01) with a decline to control level on day four in the untreated diabetic group. Kidney insulin-like growth factor I remained unchanged in the insulin-treated diabetic group. Insulin-like growth factor I mRNA was measured by solution-hybridization assay. No differences were found in kidney insulin-like growth factor I mRNA between the two diabetic groups over the study period, while in liver, insulin-like growth factor I mRNA tended to be lower on day four in diabetic rats when compared to insulin-treated rats (p = 0.07).(ABSTRACT TRUNCATED AT 250 WORDS)

The initial renal hypertrophy in experimental diabetes is dependent on the prevailing blood glucose level and is associated with renal accumulation of insulin-like growth factor I. To investigate the relationship of blood glucose to kidney IGF-I, a graded range of diabetic aberration was established in young rats by iv injection of increasing amounts of streptozotocin (25-80 mg/kg) at day 0. In 30 diabetic rats the mean of day 1 and day 2 blood glucose concentrations ranged from 6.2 to 32.0 mmol/l and 24-h urinary glucose excretion (24-48 h) from 0.04 to 43.3 mmol/24 h. The right kidneys were removed after 48 h, weighed and their IGF-I concentration analysed by radioimmunoassay. Kidney IGF-I was positively correlated to blood glucose (r = 0.66, p less than 0.0001) as well as to 24-h urinary glucose output (r = 0.54, p less than 0.005). At this early stage, kidney weight already correlated to blood glucose (r = 0.60, p less than 0.0005). No relationship between kidney IGF-I and kidney weight was found. However, if animals with severe diabetes were excluded, a significant correlation could be established (r = 0.51, p = 0.01, N = 24). The results support the hypothesis that IGF-I plays a causal role in the initial renal hypertrophy of experimental diabetes.

In a previous study we demonstrated that the kidney content of somatomedin C was maximal one to two days after uninephrectomy or induction of diabetes, and that insulin treatment prevented an increase in kidney somatomedin C as well as kidney growth in diabetic animals. In the present study we have examined the effect of a somatostatin analogue on kidney somatomedin C and initial renal growth in the two experimental situations. The kidney hypertrophy in untreated diabetic animals amounted to 23% four days after streptozotocin injection and followed an increase in kidney somatomedin C content of 60% reaching the maximum after 48 h. In young and old uninephrectomized rats kidney growth was 19% and 16% after four days. In young animals a prompt increase of 50% in kidney somatomedin C was seen as reaching the maximum after 24 h, while the somatomedin C content in kidneys from old animals was maximal after 48 h (increase of 58%) in good accordance with the slightly slower kidney growth. The new findings of the present study are that administration of a long-acting somatostatin analogue (Sandostatin) effectively prevented the obligatory increase in kidney somatomedin C content as well as kidney growth both in experimental diabetes and after uninephrectomy. It is noteworthy that Sandostatin administration did not alter the metabolic state in diabetic animals indicating that the inhibition of kidney hypertrophy could not be attributed to improved metabolic control. The results thus support the concept that somatomedin C is involved in initial diabetic and post-nephrectomy renal growth.

Somatostatin release from the isolated pancreas of 3 normal and 6 streptozotocin diabetic dogs has been measured in response to various stimuli to determine whether abnormalities in somatostatin release are present in the diabetic pancreas. Simultaneous measurement of glucagon secretion was also made. In the pancreas from normal dogs increases in perfusate glucose from 25 to 200 mg/100 ml induced a 2--3 fold increase in somatostatin release and a two thirds decrease in glucagon secretion. In contrast, in the diabetic pancreas glucose caused no change in the secretion of the two hormones. In the diabetic pancreas addition of insulin to the perfusate (25,000 microU/ml) for periods from 10 to 75 minutes aimed at restoring normal extracellular insulin levels in the islets failed to restore either somatostatin or glucagon secretion to normal. In contradistinction to the lack of effect of glucose, the somatostatin and glucagon responses to arginine (5 mmol/l), isoproterenol (2 ng/ml) and calcium (5 mmol/l) were normal in the diabetic pancreas. The data suggests the presence of a selective glucoreceptor abnormality of the D as well as of B and A cells in the streptozotocin diabetic dog.

Insulin-deficient diabetes in man as well as in experimental diabetes is associated with islet cell insensitivity to glucose. The present study was designed to determine whether this abnormality could be counteracted either by increasing the intraislet insulin level or by normalizing the diabetic state by a glucose-controlled insulin infusion system (GCIIS: Biostator, Life Science Instruments, Elkhart, Indiana). Using the isolated, perfused pancreas of dogs with moderate, untreated alloxan diabetes of 4 days duration, we found that 5 mM arginine (N = 4) and 5 mM calcium (N = 4) stimulated D- and A-cell secretion, whereas an increment in glucose from 1.3 to 11 mM (N = 4) had no effect on islet hormone secretion. In the pancreas from untreated alloxan-diabetic dogs, acute infusion of large amounts of insulin (25 mU/ml) in vitro simultaneously with an elevation of perfusate glucose from 1.3 to 11 mM failed to restore the glucose from 1.3 to 11 mM failed to restore the glucose sensitivity. In contrast, treatment of alloxan-diabetic dogs (N = 3) by a GCIIS for 24 h revived some responsiveness of the glucagon, insulin, and somatostatin to glucose (1.3-11 mM) of the subsequently perfused pancreas. It is concluded that the insensitivity to glucose of islet cells in insulin-deficient diabetes is not ascribed to an intra-islet insulin deficiency per se but rather to an abnormal metabolic state secondary to insulin deficiency. The results also indicate that the glucose receptor dysfunction is not due to a direct lesion by the diabetogenic drug.

Kidney growth after induction of experimental diabetes in rats was compared to compensatory renal growth in response to unilateral nephrectomy. After 4 days of diabetes, kidney weight had increased from 816 +/- 21 mg (SEM) to 940 +/- 42 mg (15%). In insulin-treated diabetic rats kidney weight was unchanged at the end of the study, namely 828 +/- 15 mg. In unilaterally nephrectomised rats kidney weight increased from 840 +/- 20 mg (SEM) to 1050 +/- 60 mg during 4 days (24%). We observed increased kidney content of somatomedin C in both diabetic and uninephrectomized rats. In untreated diabetic rats it was maximal after 48 h, with an increase of 77% (3469 +/- 312 ng/g (SEM) versus 1961 +/- 173 ng/g). After 4 days the somatomedin C content had returned to initial levels. In insulin-treated rats somatomedin C content did not increase during the observation period. The somatomedin C content of the remaining kidney after unilateral nephrectomy was maximal after 24 h with an increase of 58% (from 1340 +/- 203 ng/g (SEM) to 2122 +/- 214 ng/g). The somatomedin C content returned to normal at day 4. Serum somatomedin C declined insignificantly in diabetic animals during the experimental period, but a significant decrease (p less than 0.02) was found in uninephrectomized rats. This study demonstrates that kidney somatomedin C peaks during the first or second day after uninephrectomy or induction of diabetes, respectively, and that insulin treatment sufficient to prevent kidney growth abolishes the increase. These similar rapid initial hypertrophies/hyperplasies may thus be dependent on local somatomedin C formation.

The hormonal and metabolic responses to an intraduodenal glucose load (0.5 g/kg) were first determined in eight normal dogs before diabetes (alloxan/streptozotocin) was induced and then comparison made of the responses to the glucose load when normal plasma glucose profiles were recreated by preprogrammed infusion of insulin via the portal or peripheral circulations. Basal intraportal and peripheral insulin infusions at 0.021 +/- 0.001 and 0.022 +/- 0.000 U . kg-1 . h-1, respectively, for 16 h to fasting diabetic dogs normalized peripheral plasma levels of glucose (5.5 +/- 0.3 and 5.6 +/- 0.6 mmol/liter, respectively), immunoreactive insulin (IRI) (11.5 +/- 1.2 and 16.4 +/- 1.6 microU/ml), glucagon (65 +/- 7 and 62 +/- 5 pg/ml), lactate (0.63 +/- 0.04 and 0.54 +/- 0.03 mmol/liter), and alanine (0.236 +/- 0.037 and 0.191 +/- 0.008 mmol/liter). However, peripheral but not intraportal infusion of insulin depressed levels of glycerol, nonesterified fatty acids (NEFA), and 3-hydroxybutyrate (0.074 +/- 0.006 vs. 0.109 +/- 0.013, P less than 0.01; 0.67 +/- 0.04 vs. 0.84 +/- 0.09, P less than 0.05; and 0.018 +/- 0.004 vs. 0.059 +/- 0.015 mmol/liter, P less than 0.01, respectively). With the preprogrammed insulin infusions used to normalize plasma glucose profiles to the intraduodenal glucose load, all hormonal and metabolic responses were normalized during intraportal infusion (IRI, 72.5 +/- 4.2 microU/ml; glucagon, 66 +/- 10 pg/ml; lactate, 1.06 +/- 0.10 mmol/liter; alanine, 0.251 +/- 0.042 mmol/liter; glycerol, 0.043 +/- 0.013 mmol/liter; NEFA, 0.24 +/- 0.03 mmol/liter; and 3-hydroxybutyrate, 0.012 +/- 0.007 mmol/liter) but marked hyperinsulinemia (103.2 +/- 6.1 microU/ml) and depressed glycerol, NEFA, and 3-hydroxybutyrate responses at 2 h (0.056 +/- 0.005, 0.52 +/- 0.10, and 0.019 +/- 0.010 mmol/liter, respectively) resulted during peripheral infusion. Therefore, only the portal route of insulin infusion achieved complete metabolic normalization during glucose loading in diabetic dogs.