Mechanisms by which Bis(Maltolato)Oxovanadium(IV) Normalizes Phosphoenolpyruvate Carboxykinase and Glucose-6-Phosphatase Expression in Streptozotocin-Diabetic Rats in Vivo

Abstract

Vanadium treatment normalizes plasma glucose levels in streptozotocin-diabetic rats in vivo, but the mechanism(s) involved are still unclear. Here, we tested the hypothesis that the in vivo effects of vanadium are mediated by changes in gluconeogenesis. Diabetic rats were treated with bis(maltolato)oxovanadium(IV) (BMOV) in the drinking water (0.75–1 mg/ml, 4 wk) or, for comparison, with insulin implants (4 U/d) for the final week of study. As with insulin, BMOV lowered plasma glucose and normalized phosphoenolpyruvate carboxykinase (PEPCK) and glucose-6-phosphatase (G-6-Pase) mRNA in the liver and kidney of diabetic rats. To determine the importance of reducing hyperglycemia per se, diabetic rats were treated either with a single ED₅₀ dose of BMOV (0.1 mmol/kg, ip) or with phlorizin (900 mg/kg·d, 5 d). BMOV rapidly restored PEPCK and G-6-Pase mRNA and normalized plasma glucose in responsive (50%) diabetic rats but had no effect on the nonresponsive hyperglycemic rats. Phlorizin corrected plasma glucose but had no effect on PEPCK mRNA and only partially normalized G-6-Pase mRNA. In conclusion, 1) BMOV inhibits PEPCK mRNA expression and activity by rapid mechanisms that are not reproduced simply by correction of hyperglycemia; and 2) BMOV inhibits G-6-Pase expression by complex mechanisms that depend, in part, on correction of hyperglycemia.

INCREASED HEPATIC GLUCOSE production is a major factor contributing to fasting hyperglycemia in both type 1 and type 2 diabetes (1–3). In type 1 diabetes, insulin deficiency results in an increase in gluconeogenesis, which together with impaired glucose transport in the peripheral tissues, leads to dramatic hyperglycemia. In type 2 diabetes, insulin resistance leads to decreased glucose clearance and excessive glucose production despite availability of insulin (2, 3). Several studies have shown that increase in endogenous glucose production correlates closely with elevated levels of two key gluconeogenic enzymes, phosphoenolpyruvate carboxykinase GTP (PEPCK) and glucose-6-phosphatase (G-6-Pase) (4–6), supporting the idea that up-regulation of these enzymes may contribute to the development of diabetes.

PEPCK (EC 4.1.1.32), which catalyzes the conversion of oxaloacetate to phosphoenolpyruvate, plays a crucial role in gluconeogenesis. PEPCK activity is principally controlled at the level of gene expression and is sensitive to a number of hormones (4). This enzyme is expressed primarily in the liver, kidney cortex, small intestine, and adipose tissue, although low levels of the enzyme have been detected in other tissues (4, 7). Of the two known isoforms of PEPCK, the mitochondrial isoform (PEPCK-M) is constitutively expressed, whereas the cytosolic form (PEPCK-C) is tightly regulated by hormonal and dietary factors and is the form of this enzyme that is elevated in diabetes (4, 8, 9). In the liver, PEPCK expression is enhanced by glucagon (acting via cAMP), glucocorticoids, thyroid hormone, and fasting, whereas insulin and high-carbohydrate diet decrease PEPCK-C synthesis (4, 10). In the kidney, transcription of PEPCK-C gene is stimulated by glucocorticoids, fasting, and metabolic acidosis (4, 10, 11). Physiological suppressors of basal renal PEPCK expression have not yet been described, and insulin, the main regulator of PEPCK in the liver, has been reported to have no effect on PEPCK in the kidney (10, 12). The rate of change in PEPCK-C gene transcription is rapid (∼20 min in the liver). PEPCK mRNA and protein have half-lives of about 0.5 h and 6–8 h, respectively (4, 13). Because PEPCK is a key enzyme in gluconeogenesis, it is not surprising that PEPCK enzyme activity and mRNA levels are elevated in the liver and kidney of most animal models of diabetes and insulin resistance (10, 11, 14–16).

G-6-Pase (EC 3.1.3.9) catalyzes the hydrolysis of glucose-6-phosphate to glucose, the last step of both hepatic gluconeogenesis and glycogenolysis. It is mainly expressed in the liver, kidney, and small intestine (6). G-6-Pase is thought to be a multicomponent protein complex, which consists of a catalytic subunit (P36) and putative accessory transport proteins (5, 6). Activity and mRNA levels of the catalytic subunit of G-6-Pase are increased by glucose, free fatty acids, fasting, glucocorticoids, and glucagon (via cAMP) (5, 6, 17–19), whereas insulin strongly inhibits both basal and glucocorticoid-induced G-6-Pase gene expression (5, 6, 20). Unlike PEPCK, metabolic acidosis does not increase renal G-6-Pase (21). Furthermore, suppression of G-6-Pase mRNA level is rapid in the liver (less than 90 min), whereas it takes a few hours in the kidney (22). In several animal models of diabetes, G-6-Pase mRNA and activity are elevated in the liver and kidney (14, 17, 21, 23).

Vanadium compounds are candidates for oral therapy in diabetes, based on studies with experimental models of type 1 diabetes, in which they normalize plasma glucose homeostasis, carbohydrate and lipid metabolism, and ketogenesis (24, 25). However, the exact mechanism of action of vanadium in vivo is still not clearly understood. A growing body of evidence suggests that vanadium might act on insulin target tissues. Several potential sites in the insulin-signaling pathway, including both receptor and post-receptor mechanisms, have been proposed for the insulin-like effects of vanadium compounds (25–27). It has also been suggested that the glucose-lowering effects of vanadium are due to an improvement in β-cell function (28, 29), inhibition of protein tyrosine phosphatases (25, 30) and alterations in gene expression and activity of the key metabolic enzymes (25).

In the present study, we examined the in vivo effects of bis(maltolato)oxovanadium(IV) (BMOV), an organic vanadium compound (31), on PEPCK and G-6-Pase expression in streptozotocin (STZ)-diabetic rats, a model of poorly controlled type 1 diabetes, in which the counterregulatory hormones also contribute to the complex regulation of these enzymes. The advantages of BMOV over the inorganic vanadium compounds are greater potency, lower toxicity, and improved tolerance (32, 33). Our results showed that 4 wk of treatment with BMOV (0.75–1 mg/ml) or 1 wk of treatment with insulin (4 U/d) normalized PEPCK enzyme activity and mRNA and G-6-Pase mRNA levels in both liver and kidney of STZ-diabetic rats. Further experiments with the ED₅₀ dose of BMOV (0.1 mmol/kg, ip) showed that BMOV at this dose corrected plasma glucose in 50% of diabetic rats, which was associated with normalization of PEPCK and G-6-Pase mRNA in the liver, whereas the mRNA levels of both enzymes remained high in the hyperglycemic BMOV-treated diabetic rats (50%). Correction of hyperglycemia with phlorizin (900 mg/kg·d, ip, 5 d) partially restored the elevated levels of G-6-Pase mRNA but had no effect on PEPCK mRNA levels. These results suggest that the hypoglycemic effects of BMOV in STZ-diabetic rats are at least partially mediated by decreasing endogenous glucose production via suppression of G-6-Pase and PEPCK mRNA expression.

Materials and Methods

Materials

Regular/zinc insulin implants were obtained from LinShin Canada, Inc. (Ontario, Canada). Phosphoenolpyruvate, malate dehydrogenase (484 U/mg protein), β-nicotinamide adenine dinucleotide (β-NADH), deoxy (d)-GTP, STZ, phlorizin, chloroform, isopropanol, EDTA, diethylpyrocarbonate (DEPC), 10 × buffer II PCR reagent, and Tris-HCl were purchased from Sigma (St. Louis, MO). RNAZol B reagent was from Tel-Test Inc. (Friendswood, TX). Moloney murine leukemia virus reverse transcriptase (200 U/μl), random primers were from Life Technologies, Inc. (Burlington, Ontario, Canada). Deoxynucleotide (dNTP) [dATP, dCTP, dGTP, deoxythymidine triphosphate (dTTP)] was from Promega Corp. (Madison, WI). Deoxyribonculease I-ribonuclease (RNase) free (10 U/μl) was from Roche (Laval, Québec, Canada). HotStar Taq DNA polymerase (5 U/μl), MgCl₂, and PCR buffer were from QIAGEN (Miami, FL). RNaseZap, RNA later, RNase inhibitor, and 18S internal standards were from Ambion, Inc. (Austin, TX). The RIA kit for rat insulin was obtained from Linco Research, Inc. (St. Charles, MO).

Experimental protocols

Chronic treatment with BMOV or insulin

Male Wistar rats, weighing 190–220 g, were obtained from Charles River Laboratories, Montr\|[eacute]\|al. The animals were cared for in accordance with the principles and guidelines of the Canadian Council on Animal Care. Rats were housed individually in the treated groups and in pairs in the control groups, on a 12-h light, 12-h dark schedule (lights on 0600–1800 h) and given food and fluid ad libitum. Animals were monitored for their health during the first week after arrival and then were randomly assigned to two groups: control (C) and diabetic (D). Experimental diabetes was induced in rats by a single iv injection of STZ dissolved in 0.9% saline (60 mg/kg, iv) under halothane anesthesia. Control rats were injected by 0.9% saline. Three days post STZ injection, rats with blood glucose levels higher than 14 mm were considered diabetic. One week after STZ injection, control and diabetic rats were divided into further subgroups: control (C), control treated with BMOV (CB), diabetic (D), diabetic treated with BMOV (DB) and diabetic treated with insulin (DI) (n = 5 per group). BMOV treatment was started 7 d after STZ injection at an initial concentration of 0.25 mg/ml in the drinking water, which was gradually raised by increments of 0.25 mg/ml during the first week of treatment until each animal reached euglycemia and the maximum concentrations (0.75–1 mg/ml) were maintained for 3 wk. Diabetic animals were monitored individually during the study. The mean BMOV dose during the last week of treatment in the BMOV-treated control and diabetic groups was 0.24 mmol·kg⁻¹·d⁻¹ and 0.60 mmol·kg⁻¹·d⁻¹, respectively, the difference being mainly due to different body weights in control and diabetic rats. Body weights and food and fluid intakes were measured daily during the study and plasma glucose and insulin levels were monitored weekly. Blood was collected from the tail vein between 1000 and 1100 h from fed animals and centrifuged (17,500 × g, 20 min, 4 C) and plasma was stored at −20 C until assayed. Seven days before termination, rats in the insulin-treated group were anesthetized with halothane and insulin implants [4 U/d (14% regular and 0.4% zinc insulin in palmitic acid)] were inserted sc in the back of the neck. After 4 wk of treatment with BMOV, rats were anesthetized with pentobarbital (65 mg/kg, ip) and killed between 0900 and 1100 h (absorptive state). Blood was collected by cardiac puncture for measurement of plasma parameters (glucose, triglycerides, insulin, and glucagon). Liver and kidney were removed immediately, rinsed with sterile PBS and frozen quickly in liquid nitrogen. Tissues were stored at −70 C for subsequent mRNA extraction and measurement of enzyme activity.

Acute treatment with BMOV or phlorizin

Male Wistar rats weighing 190–220 g were randomly divided into two groups: control (C) and diabetic (D). One week after STZ injection, control and diabetic rats were divided into further subgroups: control (C, n = 5), control treated with BMOV (CB, n = 5), diabetic (D, n = 10), diabetic treated with BMOV (DB, n = 14) and diabetic treated with phlorizin (DPH, n = 6). Blood was collected from the tail vein at 1000 h from fed animals before starting the treatments. Animals in CB and DB groups were treated with a single ED₅₀ dose of BMOV (0.1 mmol/kg, ip) dissolved in 1% carboxymethylcellulose. The ED₅₀ dose of BMOV used in the present study was based on several published (32) and unpublished studies performed on STZ-diabetic rats in this laboratory. This dose of BMOV (0.1 mmol/kg, ip) normalizes plasma glucose levels in 50% of the BMOV-treated diabetic rats. DPH group received phlorizin (900 mg/kg·d, ip) dissolved in 1% carboxymethylcellulose, in two divided doses. This allowed for continuous inhibition of renal tubular reabsorption of glucose and maintenance of a tighter glycemic control during 24 h, which was assessed by measuring plasma glucose levels twice per day during the treatment. Rats were killed after 24 h (C, CB, D, and DB groups) or after 5 d (DPH group) between 0900 and 1100 h (absorptive state). Half of the diabetic rats (n = 5) were killed with DPH group on d 5. At the end of study, blood was collected by cardiac puncture from all groups for measurement of plasma insulin and glucose levels. Liver was removed immediately and processed as described.

Plasma parameters

Plasma glucose and triglyceride levels were measured using a Beckman Glucose Analyzer 2 (Beckman Instrumentals, Inc., Brea, CA) and Roche Molecular Biochemicals kit (Roche Diagnostic Corp., Indianapolis, IN), respectively. Plasma insulin and glucagon levels were determined with a double antibody RIA using kits from Linco Research, Inc.

PEPCK enzyme activity

PEPCK activity in the liver and kidney was determined as previously described (34) with a few modifications. Briefly, kidney (cortex) or liver tissues (∼500 mg) were homogenized in 4 vol ice-cold buffer using a polytron (Brinkmann Instruments, Inc., Westbury, NY; model: PT 3100) for 2 × 15 sec at 6000 rpm. The homogenates were centrifuged at 10,000 × g for 30 min and the supernatants were filtered (through a 53-μm mesh Nitex Netting nylon monofilament) and stored at −70 C until assayed. Homogenization buffer contained 10 mm Tris-HCl (pH 7.2), 1 mm EDTA, 0.25 m sucrose, and 50 mm KCl. Protein concentration in the homogenates was measured by Bradford method. For PEPCK enzyme assay, aliquots of kidney or liver cytosols (0.3 mg of protein) were added to the reaction mixture containing 50 mm Tris-HCl (pH 7.2), 2 mm MnCl₂, 2.5 mm phosphoenolpyruvate, 10 mm NaHCO₃ (freshly prepared), 5 U malate dehydrogenase, and 0.15 mm NADH, in a final volume of 1 ml. The reaction was initiated by adding 0.4 mm dGDP (final concentration), and the decrease in absorbance was monitored at 340 nm, 25 C for 3 min. A reaction mixture without dGDP was used as control for each sample. Reaction rates were proportional to protein concentration and linear for at least 5 min. One enzyme unit converts 1 μmol NADH to NAD per minute.

RNA extraction

Total cellular RNA was extracted from excised kidney (cortex) or liver tissues using RNAZol B reagent (Tel-Test Inc.), according to the manufacturer’s instructions. Briefly, about 100 mg of tissue was homogenized in 2 ml ice-cold RNAZol B reagent using a polytron (model 398, BioSpec Products, Inc., Bartleville, OK). RNA was extracted from the homogenates by adding chloroform (1 vol homogenate + 0.1 vol chloroform). After 5 min (4 C) the suspensions were centrifuged at 12,000 × g (4 C, 15 min). The RNA was precipitated from the aqueous phase by addition of an equal volume of isopropanol. Samples were incubated (4 C, 15 min) and centrifuged at 12,000 × g (4 C, 15 min). The supernatant was removed and the RNA was washed once with 75% ethanol and centrifuged at 7,500 × g (4 C, 8 min). After removing the supernatant, the RNA pellet was dried briefly and dissolved in 50 μl diethylpyrocarbonate (DEPC)-treated distilled water. RNA was quantified by measuring absorbance spectrophotometrically at 260 nm, and its integrity was assessed after electrophoresis on nondenaturing 1% agarose gel stained with ethidium bromide (5 μg/ml).

Semiquantitative RT-PCR

Reverse transcription of 5 μg total RNA was performed in a 60-μl reaction volume containing 200 U of Moloney murine leukemia virus reverse transcriptase (Life Technologies, Inc.), 20 U RNase inhibitor, 3 mm MgCl₂, 1 × buffer II from Sigma, 0.3 μg random primers (Life Technologies, Inc.) and 1 mm dNTP for 50 min at 42 C. Contaminating genomic DNA present in the RNA preparations was removed by digesting the reaction with 5 U of deoxyribonuclease I for 45 min at 37 C before the addition of reverse transcriptase. The PCR amplifications were performed in 100 μl reaction mixture. The PCR mixture contained 250 μm dNTP, 2 mm MgCl₂, 0.5 U HotStar Taq DNA polymerase (QIAGEN), 1 μl of sense and antisense primers, 5 μl of the RT product, and 1× QIAGEN buffer. The reaction mixtures were subjected to 32 cycles of PCR amplification consisting of denaturation for 60 sec at 94 C, annealing for 60 sec at 55 C and elongation for 60 sec at 72 C. The final extension was completed at 72 C for 7 min. The oligonucleotide primers (5′-AGCCTCGACAGCCTGCCCCAGG-3′ sense and 5′-CCAGTTGTTGACCAAAGGCTTTT-3′ antisense) for PEPCK were designed from published reports (7) and the amplified product was a 575-bp cDNA. The primers (5′-TAAGTGGATTCTTTTTGGACA-3′ sense and 5′-GAAGAGGCTGGCAAAGGGTGT-3′ antisense) for G-6-Pase (35) amplified a 562-bp cDNA. PEPCK and G-6-Pase mRNA expression levels were normalized to 18S ribosomal RNA expression (rRNA) (Ambion, Inc.). Ten microliters of 6× loading buffer [containing 0.25% bromothymol blue, 0.25% xylene cyanol FF, and 15% Ficoll type 400 (Amersham Pharmacia Biotech, Piscataway, NJ) in DEPC-treated distilled water] was added to the PCR samples. Twenty microliters of PCR products were electrophoresed on 2% agarose gel stained with ethidium bromide and gels were photographed under UV light. The intensity of mRNA bands was analyzed by densitometry and the amplified products of PCR were purified with a QIAGEN PCR purification kit (QIAquick) and sequenced to assure the accuracy of the PCR amplifications.

Statistical analysis

Values are expressed as the mean ± sem, and n = number of rats in each group. Statistical analyses were performed using one-way ANOVA followed by a Newman-Keuls test. P < 0.05 was taken as level of significance.

Results

Chronic treatment with BMOV or insulin

General characteristics and plasma parameters

General characteristics and plasma parameters of the animals in different treatment groups are summarized in Tables 1–3. Animals were monitored daily for changes in body weight during the experiment. At the end of the study, body weights were significantly lower in the diabetic and BMOV-treated diabetic groups compared with the control group (Table 1). Treatment with BMOV did not improve the growth retardation in diabetic rats, whereas 1 wk after receiving insulin there was a significant increase in body weight in the insulin-treated diabetic rats compared with the untreated rats, indicating the anabolic effects of insulin. BMOV treatment in control rats led to a slight but significant decrease in body weight gain compared with untreated control rats. The attenuation of hyperglycemia in BMOV-treated diabetic rats was accompanied by normalization of food and fluid intakes (Table 1). There was no significant difference in food and fluid intakes between control rats and BMOV-treated diabetic rats at the end of study. Insulin treatment in diabetic rats resulted in a significant decrease in the food and fluid intakes.

Table 1.

General characteristics of animals in different treatment groups

	Body weight (g)		Fluid intake (ml/d)		Food intake (g/d)
	Beginning	Termination	Beginning	Termination	Beginning	Termination
Control	319 ± 8	462 ± 15	43 ± 1	55 ± 3	29 ± 1	32 ± 1
Control + BMOV	324 ± 5	408 ± 8a	44 ± 2	46 ± 3	30 ± 7	28 ± 1
Diabetic	292 ± 11	317 ± 21b	210 ± 6b	320 ± 49b	48 ± 4b	59 ± 5b
Diabetic + BMOV	306 ± 7	323 ± 35b	250 ± 14b	67 ± 7a	49 ± 1b	27 ± 13a
Diabetic + insulin	298 ± 10	404 ± 19c	190 ± 12c	105 ± 3c	46 ± 3c	53 ± 1c

Values are presented as the mean ± sem. Measurements were performed in the fed rats (between 1000 and 1100 h) at the beginning (7 d after STZ injection) and the end of study (4 wk after treatment with BMOV). Significant changes are indicated by superscripts:

^a

BMOV-treated vs. corresponding untreated group;

^b

diabetic vs. corresponding control;

^c

diabetic + insulin vs. control (P < 0.05, ANOVA).

Table 2.

Plasma glucose and insulin levels in different treatment groups

	Beginning	Wk 1	Wk 2	Wk 3	Wk 4
Plasma glucose (mmol/liter)
Control	8.3 ± 0.1	7.7 ± 0.1	7.7 ± 0.4	7.4 ± 0.2	9.9 ± 0.5
Control + BMOV	8.8 ± 0.2	7.7 ± 2.2	6.9 ± 0.1	7.0 ± 0.1	10.0 ± 0.7
Diabetic	27.9 ± 1.1a	29.5 ± 1.7a	26.9 ± 1.9a	26.5 ± 0.9a	29.2 ± 1.9a
Diabetic + BMOV	28.8 ± 0.9a	24.1 ± 4.3a	18.2 ± 4.5a,b	17.0 ± 2.4a,b	16.3 ± 2.7a,b
Diabetic + insuline	30.8 ± 1.2a	29.0 ± 1.5a	28.8 ± 1.0a	24.5 ± 1.2a	5.7 ± 1.8c,d
Plasma insulin (ng/ml)
Control	2.2 ± 0.4	2.7 ± 0.2	2.4 ± 0.2	3.0 ± 0.3	3.5 ± 0.5
Control + BMOV	2.4 ± 0.5	1.3 ± 0.3b	1.1 ± 0.1b	1.2 ± 0.2b	2.8 ± 0.3
Diabetic	0.7 ± 0.1a	0.6 ± 0.1a	0.4 ± 0.1a	0.3 ± 0.1a	0.5 ± 0.2a
Diabetic + BMOV	0.6 ± 0.1a	0.5 ± 0.1a	0.4 ± 0.1a	0.5 ± 0.2a	0.5 ± 0.2a
Diabetic + insuline	0.6 ± 0.1a	0.4 ± 0.1a	0.3 ± 0.1a	0.3 ± 0.1a	7.9 ± 1.8c,d

Values are presented as the mean ± sem. Measurements were performed in the fed rats every week (between 1000 and 1100 h). Significant changes are indicated by superscripts:

^a

Diabetic vs. corresponding control;

^b

BMOV-treated vs. corresponding untreated group;

^c

diabetic + insulin vs. control;

^d

diabetic + insulin vs. diabetic + BMOV (P < 0.05, ANOVA);

^e

Rats in the diabetic + insulin group received insulin implants at the end of wk 3 (1 wk before the termination).

Table 3.

Effects of BMOV on plasma glucagon and triglycerides

Group	Glucagon (pg/ml)	Triglycerides (mmol/liter)
Control	49.1 ± 2.9	0.5 ± 0.1
Control + BMOV	60.5 ± 4.8	0.7 ± 0.1
Diabetic	100.0 ± 3.8a	5.6 ± 1.6a
Diabetic + BMOV	71.9 ± 1.0b	1.0 ± 0.1b
Diabetic + insulin	63.6 ± 5.3b	0.3 ± 0.1b

Measurements were performed in the fed rats (between 1000 and 1100 h) after 4 wk of treatment with BMOV in the drinking water (0.75–1 mg/ml) or 1 wk of treatment with insulin (4 U/d). Values are presented as the mean ± sem. Significant changes are indicated by superscripts:

^a

Diabetic vs. control;

^b

diabetic + BMOV or diabetic + insulin vs. diabetic (P < 0.05, ANOVA).

Plasma glucose and insulin levels were measured weekly by sampling blood from the tail vein between 1000 and 1100 h in fed rats (Table 2). STZ injection resulted in a profound increase in the plasma glucose concentration and a significant decrease in plasma insulin levels in Wistar rats. Two weeks after initiation of BMOV treatment, plasma glucose levels were significantly lower in the BMOV-treated diabetic rats compared with untreated rats [DB (wk 2): 18.2 ± 4.5 vs. D (wk 2): 26.9 ± 1.9 mmol/liter]. Glucose levels were further lowered after 4 wk of treatment [DB (wk 4): 16.3 ± 2.7 vs. D (wk 4): 29.2 ± 1.9 mmol/liter]. There was no significant difference in plasma insulin levels between BMOV-treated and untreated diabetic rats. Consistent with our previous studies, BMOV treatment had no effect on plasma glucose levels in the control rats. However, plasma insulin levels were slightly lower in BMOV-treated control rats as compared with untreated rats. One week after treatment with insulin plasma glucose levels dropped from 24.5 ± 1.2 to 5.7 ± 1.8 mmol/liter. Plasma glucagon levels were increased about 2-fold in the diabetic rats compared with controls and treatment with either BMOV or insulin markedly lowered plasma glucagon (Table 3). Elevated levels of plasma triglycerides were also normalized after treatment with either BMOV or insulin (Table 3).

PEPCK activity and mRNA expression

Measurement of enzyme activity in fed rats showed that PEPCK activity was significantly higher in the liver of diabetic rats compared with control animals (D: 52.3 ± 2.5 vs. C: 31.8 ± 1.9 mU/mg protein, P < 0.05) (Fig. 1A). Four weeks treatment with BMOV completely normalized PEPCK activity in the diabetic rats (DB: 32.8 ± 2.7 vs. C: 31.8 ± 1.9 mU/mg protein, P > 0.05). Results from an RT-PCR assay indicated that in parallel with the elevated enzyme activity, PEPCK mRNA levels were significantly increased in the diabetic rats and were restored following BMOV treatment (Fig. 2A). BMOV did not have any effect on PEPCK activity or mRNA level in the liver of control rats. As with liver, renal PEPCK activity was significantly higher in the diabetic rats compared with controls (D: 63.4 ± 3.0 vs. C: 40.2 ± 4.9 mU/mg protein, P < 0.05) (Fig. 1B). This increase in enzyme activity was accompanied by an increase in mRNA levels in the kidney of diabetic rats (Fig. 2B). Both PEPCK expression and activity were normalized by BMOV treatment (C: 40.2 ± 4.9 vs. DB: 43.9 ± 2.7 mU/mg protein, P > 0.05). One week of treatment with insulin restored PEPCK mRNA and activity in the liver and kidney of diabetic rats.

Figure 1.

Effects of chronic treatment with BMOV and insulin on PEPCK activity in the liver (A) and kidney (cortex) (B) of control and STZ-diabetic Wistar rats. BMOV-treated animals received BMOV (0.75–1 mg/ml) in the drinking water for 4 wk. Rats in the insulin-treated group received insulin implants (4 U/d) 1 wk before termination. Animals were killed between 0900 and 1100 h in the fed state. Liver or kidney homogenates were prepared and PEPCK activity was measured as explained in the Materials and Methods section. C, CB, D, DB, and DI denote control, control-treated with BMOV, diabetic, diabetic treated with BMOV, and diabetic treated with insulin, respectively (n = 5 per group). Results are presented as the mean ± sem of PEPCK activity in each group. *, P < 0.05 vs. all other groups (ANOVA).

Figure 2.

Effects of chronic treatment with BMOV and insulin on PEPCK mRNA levels in the liver (A) and kidney (cortex) (B) of control and STZ-diabetic Wistar rats. Four weeks after treatment with BMOV or 1 wk after treatment with insulin, rats were killed in the fed state. Total RNA was extracted from the tissues and used for RT-PCR as explained in Materials and Methods. C, CB, D, DB, and DI denote control, control treated with BMOV, diabetic, diabetic treated with BMOV, and diabetic treated with insulin, respectively (n = 5 per group). PEPCK mRNA abundance was normalized against 18S rRNA levels, which did not vary significantly between samples. Results are presented as the mean ± sem of the relative optical densities of scanned images (PEPCK/18S internal control) in each group. *, P < 0.05 vs. all other groups (ANOVA). The lower panels show representative analysis of PCR products from different treatment groups following electrophoresis on 2% agarose gel.

G-6-Pase mRNA expression

STZ-diabetes resulted in a profound increase in G-6-Pase mRNA levels in the liver of diabetic rats and to the same extent in the kidney (Fig. 3). The increase in G-6-Pase mRNA was much more profound compared with the elevation of PEPCK mRNA in the STZ-diabetic rats after 5 wk of diabetes. Treatment with BMOV completely normalized the elevated levels of G-6-Pase mRNA expression in the liver and kidney of diabetic animals, whereas it did not have any effect on G-6-Pase mRNA expression in the control rats. Treatment with insulin for 1 wk restored the diabetes-induced increase in G-6-Pase mRNA to the same extent as was seen with BMOV.

Figure 3.

Effects of chronic treatment with BMOV and insulin on G-6-Pase mRNA in the liver (A) and kidney (cortex) (B) of control and STZ-diabetic rats. G-6-Pase mRNA levels were assessed by RT-PCR after 4 wk of treatment with BMOV or 1 wk of treatment with insulin in the kidney and liver from fed rats. C, CB, D, DB, and DI denote control, control treated with BMOV, diabetic, diabetic treated with BMOV, and diabetic treated with insulin, respectively (n = 5 per group). The G-6-Pase mRNA abundance was normalized against 18S rRNA levels, which did not vary significantly between samples. Results are presented as the mean ± sem of the relative optical densities of scanned images (G-6-Pase/18S internal control) in each group. *, P < 0.05 vs. all other groups (ANOVA). The lower panels show representative analysis of PCR products from different treatment groups following electrophoresis on 2% agarose gel.

Acute treatment with BMOV or phlorizin

Plasma glucose and insulin levels

To investigate whether the inhibitory effects of BMOV on PEPCK and G-6-Pase were mediated by suppression of gene expression and/or indirectly by the correction of hyperglycemia, STZ-diabetic rats were treated acutely with an ED₅₀ dose of BMOV (0.1 mmol/kg, ip) or for 5 d with phlorizin (900 mg/kg·d, ip). Treatment of diabetic rats with this dose of BMOV normalized plasma glucose levels in 50% of animals. At termination (after 24 h), plasma glucose levels in responsive (BR, n = 7) and nonresponsive (BNR, n = 7) BMOV-treated diabetic rats were 8.7 ± 0.8 and 22.4 ± 2.2 mmol/liter, respectively (Fig. 4). Plasma glucose levels in the nonresponsive BMOV-treated diabetic rats were not significantly different from untreated diabetic rats (D: 26.1 ± 1.0 vs. BNR: 22.4 ± 2.2 mmol/liter, P > 0.05). This allowed us to look at the effects of BMOV on PEPCK and G-6-Pase in the presence and absence of hyperglycemia. Previous studies in our laboratory have shown that BMOV, when administered as a single dose (ip), lowers plasma glucose levels in responsive animals within 2–6 h, and euglycemia is sustained following withdrawal of the drug (32). Therefore, animals were killed 24 h after administration of BMOV when they were in the euglycemic state. As with BMOV, treatment with phlorizin normalized plasma glucose levels in the diabetic rats (D: 26.1 ± 1.0 vs. DPH: 10.8 ± 1.1 mmol/liter) without any significant effect on plasma insulin levels. One of the phlorizin-treated rats did not reach the euglycemic state at the end of treatment and therefore, was eliminated from the study.

Figure 4.

Plasma glucose levels at basal state and 24 h after treatment with a single ED₅₀ dose of BMOV (0.1 mmol/kg, ip) or 5 d after treatment with phlorizin (900 mg/kg·d, ip). Blood was collected between 1000 and 1100 h (fed state). Both control and diabetic rats received the same dose of BMOV. As explained in the Results section, acute treatment with an ED₅₀ dose of BMOV normalized plasma glucose levels in 50% of BMOV-treated diabetic rats (BR, n = 7 vs. BNR, n = 7). C, CB, D, BR, BNR, and DPH denote control, control treated with BMOV, diabetic, diabetic treated with BMOV (responsive), diabetic treated with BMOV (nonresponsive), and diabetic treated with phlorizin, respectively. *, P < 0.05 vs. untreated diabetic group (ANOVA).

PEPCK and G-6-Pase mRNA expression

One week after STZ injection, PEPCK and G-6-Pase mRNA levels were significantly higher in the liver of diabetic rats as compared with controls (Fig. 5). Treatment with a single ED₅₀ dose of BMOV (0.1 mmol/kg, ip) completely normalized plasma glucose levels and restored the elevated levels of PEPCK and G-6-Pase mRNA expression in the liver of BMOV-responsive diabetic rats. However, expression of both enzymes remained high in the nonresponsive hyperglycemic rats. As in the chronic study, BMOV had no effect on PEPCK and G-6-Pase mRNA expression in the control rats. Treatment with phlorizin for 5 d corrected plasma glucose levels but did not have any detectable effect on PEPCK mRNA expression and only partially normalized G-6-Pase mRNA levels in the liver.

Figure 5.

Effects of acute treatment with BMOV (0.1 mmol/kg, ip) or phlorizin (900 mg/kg·d, ip) on PEPCK (A) and G-6-Pase (B) mRNA levels in the liver of control and STZ-diabetic rats. mRNA levels were assessed by RT-PCR after 24 h of treatment with BMOV or 5 d of treatment with phlorizin in the fed rats. C, CB, D, BR, BNR, and DPH denote control, control treated with BMOV, diabetic, diabetic treated with BMOV (responsive), diabetic treated with BMOV (nonresponsive), and diabetic treated with phlorizin, respectively. The mRNA abundance was normalized against 18S rRNA levels. Results are presented as the mean ± sem of the relative optical densities of scanned images (PEPCK/18S or G-6-Pase/18S internal control) in each group. ★, P < 0.05 vs. untreated diabetic (ANOVA). The lower panels show representative analysis of PCR products from different treatment groups following electrophoresis on 2% agarose gel.

Discussion

Several studies have shown that vanadium compounds mimic/enhance most of the metabolic effects of insulin both in vitro and in vivo (25, 36). However, despite numerous studies during the past decade the molecular mechanism(s) by which vanadium mediates its metabolic effects are still not well understood, and a growing body of evidence suggests that the effects of vanadium on glucose homeostasis, in vivo, may not be completely explained by enhancing glucose transport and glycogen synthesis in the peripheral tissues (37–40). Accordingly, the studies described here were undertaken to investigate the in vivo actions of vanadium on enzymes involved in gluconeogenesis and glucose production.

Insulin restores PEPCK activity and mRNA in diabetic rats

In agreement with previous reports on animal models of diabetes, PEPCK activity and mRNA levels were significantly higher in the liver of STZ-diabetic rats compared with control rats (4, 10, 41, 42) and were restored after insulin treatment (10, 41), indicating that the insulin signaling pathway was not impaired. Recent in vitro studies have shown that the inhibitory effects of insulin on PEPCK gene expression in the liver are mediated by the effects of phosphatidylinositol 3-kinase (PI3-K) (43, 44) on an insulin-responsive region in the PEPCK gene promoter (4). The links between PI3-K and PEPCK promoter are still undefined. Some studies (45), but not all (43), support a role for protein kinase B (PKB) in the repression of glucocorticoid and cAMP induction of PEPCK. However, increased levels of plasma glucagon observed in STZ-diabetic rats by ourselves and others (46) suggests that, in vivo, the inhibitory effects of insulin on PEPCK may also be mediated by suppression of secretion and/or actions of glucagon.

The contribution of kidney to the elevated levels of gluconeogenesis in diabetes is important to consider, in view of the finding that kidney may account for more than 40% of total endogenous glucose production (6). As in other studies, PEPCK activity and mRNA levels were significantly higher in the kidney of diabetic rats compared with nondiabetic controls (10, 11, 15). It is believed that diabetes induces renal PEPCK mRNA indirectly by causing acidosis (10–12) because correction of acidosis prevents the increase in PEPCK gene expression (11). Treatment with insulin normalized both PEPCK activity and expression in the kidney. In contrast to liver, metabolic acidosis is the primary regulator of renal PEPCK (4, 11, 12). Hence, it appears that the effects of insulin on renal PEPCK observed in this study were due to improvement in the metabolic state.

Insulin restores G-6-Pase mRNA in diabetic rats

STZ-diabetes markedly increased mRNA levels of the catalytic subunit of G-6-Pase in the liver and to almost the same extent in the kidney, and levels in both tissues were normalized with insulin treatment. These results are in agreement with those obtained from other studies performed on experimental models of diabetes (14, 17, 21, 23, 47). Interestingly, the increase in G-6-Pase mRNA in both tissues was more profound than the increase in PEPCK mRNA, emphasizing the likely significance of G-6-Pase in control of gluconeogenesis. Insulin plays an important role in the regulation of G-6-Pase in both liver and kidney in vivo (21), indicating that hypoinsulinemia and unopposed hyperglucagonemia may contribute to the elevated levels of G-6-Pase mRNA in these tissues. A multicomponent insulin responsive sequence identified in the promoter region of the G-6-Pase gene (48) appears to be regulated via PI3-K and downstream PKB-dependent and -independent pathways (35, 49). However, the finding that correction of hyperglycemia per se is able to suppress the marked diabetes-induced increase in G-6-Pase mRNA in the liver (17) suggests a possible role for hyperglycemia in up-regulation of G-6-Pase in STZ-diabetic rats.

BMOV normalizes PEPCK and G-6-Pase in diabetic rats

In the present study, we showed that normalization of plasma glucose levels in BMOV-treated diabetic rats was accompanied by inhibition of hepatic PEPCK and G-6-Pase mRNA expression. These results are consistent with in vitro studies performed on PEPCK in hepatoma cells (50) and in vivo studies with inorganic vanadium compounds in experimental models of diabetes (14, 42, 51). It has been shown that vanadate treatment leads to a decrease in G-6-Pase activity in the diabetic liver (14, 52–54). One of the main aims of this study was to show whether the inhibitory effects of vanadium on G-6-Pase, in vivo, are associated with suppression of its mRNA expression. Vanadium is a potent protein phosphatase inhibitor (25) and direct inhibitory effects of vanadium on G-6-Pase activity, in vitro, have been documented (22, 55), but these effects appear to occur only at high concentrations that are not usually achieved in vivo, indicating that a direct inhibitory effect of vanadium on G-6-Pase activity is not likely to account for the effects of vanadium seen at low (micromolar) levels generated during in vivo treatment. An important finding of this study was that the inhibitory effects of vanadium on G-6-Pase, in vivo, are mediated by suppression of its expression.

As with the effects of insulin, inhibitory effects of BMOV on PEPCK and G-6-Pase mRNA in the liver could involve direct effects on the gene promoters, and/or indirect effects via alteration in counterregulatory hormones. The first hypothesis is supported by the finding that vanadate inhibits both basal and cAMP-stimulated expression of PEPCK in hepatocytes in vitro in the absence of endogenous hormones and that there is a vanadate response region in the PEPCK gene promoter (50). However, the observation that BMOV lowered the elevated levels of plasma glucagon in diabetic rats indicates that BMOV may also inhibit PEPCK and G-6-Pase indirectly by restoring plasma glucagon levels. This idea is further supported by the interesting finding in the present study that BMOV did not have any effect on basal plasma glucose levels or PEPCK and G-6-Pase mRNA expression in the fed control rats, which have reduced endogenous glucose production, whereas it was reported that vanadium inhibits glucagon-mediated increase in plasma glucose in control rats (56). Furthermore, vanadium might act directly on the cAMP system, enhancing cyclic nucleotide phosphodiesterase (57) or inhibiting cAMP-dependent protein kinase (58). Finally, because high concentrations of free fatty acids and glucose are known to increase G-6-Pase expression (5, 17–19), it could therefore be suggested that BMOV may decrease G-6-Pase mRNA levels by normalizing hyperglycemia and/or hyperlipidemia.

As with liver, treatment with BMOV normalized PEPCK activity and mRNA expression and restored G-6-Pase mRNA in the kidney of STZ-diabetic rats. Whether the inhibitory effects of BMOV on PEPCK mRNA in the kidney are primary (via PEPCK promoter) or are secondary to a decrease in ketone bodies and improvement of the metabolic acidosis is not clear. However, the observation that vanadate normalized blood ketone body levels and improved metabolic acidosis in diabetic rats (59), suggests that the effects of BMOV on PEPCK might be related to an improvement in the metabolic acidosis.

Direct effects of vanadium on PEPCK and G-6-Pase mRNA

To investigate whether the inhibitory effects of BMOV on PEPCK and G-6-Pase are mediated by its effects on their gene expression or via correction of metabolic state, two experimental approaches were used. In the first approach, rats were treated with a single ED₅₀ dose of BMOV (0.1 mmol/kg, ip) to induce rapid normalization of plasma glucose so that effects on gene expression, in vivo, could be assessed without any sustained changes in plasma glucose levels. Normalization of plasma glucose by acute BMOV treatment in the responsive diabetic rats (BR, n = 7) was associated with correction of PEPCK and G-6-Pase mRNA levels in the liver while in the nonresponsive hyperglycemic animals (BNR, n = 7) expression of both enzymes remained high (Fig. 5). These acute effects of BMOV therefore demonstrate that the effects of diabetes on PEPCK and G-6-Pase expression can be rapidly reversed by vanadium treatment. As a second approach to further investigate the importance of reducing hyperglycemia per se, animals were treated with phlorizin, an antihyperglycemic agent that inhibits renal tubular glucose transport and blocks glucose reabsorption when the plasma glucose is increased above the basal state. Thus, it normalizes plasma glucose without causing hypoglycemia or altering plasma insulin levels (60, 61). Phlorizin treatment normalized plasma glucose levels in the diabetic rats and partially restored the elevated levels of G-6-Pase mRNA, suggesting that correction of hyperglycemia per se results in partial normalization of this enzyme. Phlorizin treatment had no effect on PEPCK mRNA levels in the diabetic liver. Taken together, these results suggest that suppression of G-6-Pase expression with BMOV is mediated both by a combination of direct effects on gene expression and by indirect effects via correction of hyperglycemia. In contrast, effects of BMOV on PEPCK expression cannot be explained by reducing levels of circulating glucose.

Selective actions of vanadium

The interesting finding that BMOV had no detectable effect on PEPCK and G-6-Pase mRNA levels in the control rats supports the idea that vanadium has selective effects on the mechanisms responsible for exaggerated enzyme expression in STZ-diabetes. Selective effects of vanadium have been shown before, for example, it has been consistently found that vanadium does not reproduce the effects of insulin replacement on the growth pattern of STZ-diabetic rats, a finding confirmed in the present study (Table 1). Also in this study, although insulin and BMOV had equivalent effects on the expression of PEPCK and G-6-Pase, insulin had a more rapid and profound effect in lowering plasma glucose levels. Similarly, vanadium has been found to inhibit lipolysis at concentrations much lower than those required for the stimulation of glycogen or fatty acid synthesis in white adipose tissue (58).

The differential effects of vanadium observed on tissue metabolism might be explained by selective actions at the signal transduction level. Recently, we demonstrated that, at doses sufficient to lower plasma glucose levels in the STZ-diabetic rats, BMOV had no detectable effect on PI3-K activity in the skeletal muscle (39) or on PKB activity in the skeletal muscle and liver (38). On the basis of these findings, it appears that the effects of BMOV are not mediated via the PI3-K/PKB axis but might involve more selective downstream effects, perhaps mediated by changes in the concentration of cAMP or cell responses to this cyclic nucleotide. Overall, our observations imply that some, but not all, of the regulatory effects of insulin are mimicked by BMOV in vivo.

In summary, results of this study suggest that the hypoglycemic effects of BMOV in STZ-diabetic rats are at least partially mediated by suppression of the key gluconeogenic enzymes via both inhibitory effects on their expression (PEPCK and G-6-Pase) and correction of hyperglycemia (G-6-Pase).

Acknowledgments

This study was supported by a National Sciences & Engineering Research Council of Canada/Technology Partnership Program grant. L.M. was a recipient of the Canadian Institutes of Health Research traineeship. Technical support of Ms. Mary Battell, Ms. Violet Yuen, and Dr. Jerzy Kulpa is gratefully acknowledged. The authors would like to thank Dr. Chris Orvig and Dr. Katherine Thompson, Department of Chemistry, The University of British Columbia, for providing the BMOV and Dr. Bruce McManus, Department of Pathology and Laboratory Medicine, The University of British Columbia, for his kind support and providing equipment for the RT-PCR assays.

^*

Present Address: Department of Pathology and Laboratory Medicine and the British Columbia Research Institute for Children’s & Women’s Health, The University of British Columbia, Vancouver, British Columbia, Canada.

^†

Present Address: Department of Pharmacology, University of the Pacific, Stockton, California.

Abbreviations

 
• BMOV

  Bis(maltolato)oxovanadium(IV)

 
• BR

  diabetic treated with BMOV (responsive)

 
• BNR

  diabetic treated with BMOV (nonresponsive)

 
• C

  control

 
• CB

  control treated with BMOV

 
• d

  deoxy

 
• D

  diabetic

 
• DB

  diabetic treated with BMOV

 
• DI

  diabetic treated with insulin

 
• DEPC

  diethylpyrocarbonate

 
• dNTP

  deoxynucleotide

 
• DPH

  diabetic treated with phlorizin

 
• dTTP

  deoxythymidine triphosphate

 
• G-6-Pase

  glucose-6-phosphatase

 
• β-NADH

  β-nicotinamide adenine dinucleotide

 
• PEPCK

  phosphoenolpyruvate carboxykinase GTP

 
• PEPCK-C

  cystolic form

 
• PEPCK-M

  mitochondrial isoform

 
• PI3-K

  phosphatidylinositol 3-kinase

 
• PKB

  protein kinase B

 
• RNase

  ribonuclease

 
• STZ

  streptozotocin