Journal Logo

Articles

Effects of Aldose Reductase Inhibition on Responses of the Corpus Cavernosum and Mesenteric Vascular Bed of Diabetic Rats

Keegan, Alan; Jack, Alison M.; Cotter, Mary A.; Cameron, Norman E.

Author Information
Journal of Cardiovascular Pharmacology: April 2000 - Volume 35 - Issue 4 - p 606-613
  • Free

Abstract

Diabetes mellitus causes vascular-based complications including nephropathy, retinopathy, and neuropathy, which have a major impact on quality of life and morbidity in patients. (1-3). The vascular endothelium appears to be particularly vulnerable. Impaired nitric oxide (NO)-mediated endothelium-dependent relaxation has been noted for the vasculature in experimental diabetes, including aorta (4-6), basilar artery (7), vasa nervorum (8,9), and mesenteric resistance vessels (10). Male impotence is a common diabetic complication. In animal models and patients, defective NO release from corpus cavernosum nonadrenergic noncholinergic (NANC) nerves and endothelium impairs control of penile smooth muscle relaxation, compromising erectile performance (11-13).

Diabetes also has deleterious effects on vasorelaxation caused by other mediators, including prostanoids (14) and endothelium-derived hyperpolarising factor (EDHF). A recent study showed that streptozotocin-induced diabetes markedly impaired smooth muscle hyperpolarisation and vasodilation mediated by EDHF in rat mesenteric vessels (10). The identity of EDHF is not established; however, it may be predominant in the small resistance vessels (15-17) controlling tissue nutritive perfusion, which could be particularly relevant for the pathophysiology of diabetic microvascular complications.

Hyperglycaemia-driven flux through the polyol pathway is probably the most intensively studied of diabetes-induced metabolic perturbations in the aetiology of complications. The polyol pathway is active in many cells, including vascular endothelium (18). Early studies showed that aldose reductase inhibitors (ARIs), which inhibit the conversion of glucose to the sugar alcohol, sorbitol, prevent reduced nerve conduction velocity in diabetic rats (1,19). More recently, the importance of ARI vascular actions has been emphasised; for example, improvements in nerve blood flow parallel the correction of nerve function in diabetic rats (20). ARI treatment prevents defective aorta NO-mediated endothelium-dependent relaxation in experimental diabetes (5,21,22); however, there are conflicting reports on efficacy for mesenteric resistance vessels (23,24). The aim was to further examine polyol pathway involvement in vascular dysfunction in diabetic rats by assessing the effect of preventive ARI treatment with WAY121509 on corpus cavernosum and mesenteric vascular bed.

MATERIALS AND METHODS

Rats, diabetes induction, and treatment

Male Sprague-Dawley rats (Aberdeen University colony), 19 weeks old at the start of experiments, were used. Diabetes was induced by intraperitoneal injection of streptozotocin (Zeneca Pharmaceuticals, Macclesfield, Cheshire, U.K.) at a dose of 42-45 mg/kg freshly dissolved in sterile saline. Diabetes was verified 24 h later by hyperglycaemia and glycosuria (Visidex II and Diastix; Ames, Slough, U.K.). Plasma glucose (GOD-Perid method; Boehringer Mannheim, Mannheim, Germany) was determined on tail-vein samples. Groups comprised nondiabetic controls and 2-month untreated diabetic controls. A further group of diabetic rats was treated with the ARI, WAY121509 [spiro[isoquinoline-4(1H), 3′-pyrrolidine]-1,2′,3′,5′(2H)-tetrone; Wyeth-Ayerst, Princeton, NJ, U.S.A.] starting 2 days after diabetes induction. The drug was administered by dietary supplementation, being mixed with powdered rat chow such that rats received a daily dose of ∼10 mg/kg. This WAY121509 dose and administration method completely corrects impaired nerve conduction velocity and blood flow, and suppresses diabetic tissue sorbitol to below the nondiabetic level (25,26).

Corpus cavernosum

Experiments were carried out as previously described (27). In brief, rats were anaesthetised (5% halothane in air), the penis was excised, and the penile shaft was dissected free of connective tissue. Corpora cavernosa were separated and a longitudinal slit was made in each to facilitate drug access. Cavernosa were mounted in baths containing modified Krebs-Ringer solution (144.0 Na+, 5.0 K+, 1.25 Ca+, 1.1 Mg2+, 25.0 HCO, 1.1 PO43−, 1.1 SO42−, 5.5 glucose, in mM), at 37°C, gassed with 95% O2/5% CO2 (pH 7.35). Isometric tension was monitored, and resting tension was set at 1.5 g. Tissues were equilibrated for 60 min and then precontracted with 30 μM phenylephrine (PE), which results in >80% of the maximal contraction in tissues from diabetic and nondiabetic rats (28). Cumulative concentration-response curves were determined for endothelium-dependent relaxation to acetylcholine (ACh) or endothelium-independent relaxation to sodium nitroprusside (SNP). Phenylephrine-precontracted cavernosa from nondiabetic and diabetic rats showed similar (>95%) relaxation to 300 μM papaverine (28).

Repetitive supramaximal transmural electrical field stimulation (TEFS) of autonomic nerves was achieved by using platinum wire electrodes (pulse-train duration, 15 s; pulse-width, 5 ms; 50-90 V; 2-20 Hz). Frequency-response curves were first determined for noradrenergic vasoconstrictor nerves. Subsequently, the tissues were preincubated for 30 min with guanethidine (5 μM) to abolish noradrenergic nerve responses, and atropine (1 μM) to eliminate cholinergic fibre-mediated responses. After precontraction with 30 μM PE, TEFS caused responses mediated by the NANC innervation.

Mesenteric vascular bed

The abdominal mesentery was exposed under 5% halothane anaesthesia, the superior mesenteric artery was cannulated and the mesenteric bed was freed from the small intestine and placed in a tissue chamber at 37°C. The preparation was perfused with modified Krebs-Ringer solution gassed with 95% O2 and 5% CO2. The flow rate was 5 ml/min. Perfusion pressure and drug effects were monitored by using a pressure transducer. At least 30-min equilibration was allowed, and preparations were first contracted with 100 μM PE and then allowed to relax for 30 min. Preparations were then precontracted with a dose of PE (3-100 μM) adjusted to give an increase in perfusion pressure of 30-100 mm Hg over the basal pressure. Cumulative concentration-response curves to ACh alone and after preincubation with a sufficiently high dose (30 mM) of NG-nitro-L-arginine (NOLA) to ensure a maximal blockade of NO synthase (NOS), and to SNP in the presence of NOLA were obtained. Pressor responses to cumulatively increasing PE concentrations also were determined before and after preincubation with 30 mM NOLA.

Statistical analysis

Results are expressed as mean ± SEM. Data were subjected to Bartlett's test for homogeneity of variances, followed by log transformation if necessary before one-way analysis of variance (ANOVA). Where significance (p < 0.05) was reached, between-group differences were established by using the Student-Neuman-Keuls multiple comparison test. If variances were not homogeneous, data were analysed by an appropriate nonparametric test, such as Kruskal-Wallis one-way analysis of variance followed by Dunn's multiple comparison test. Within-group serial comparisons were made by using paired Student's t tests. Concentration-response curves were fitted by sigmoid curves with the least squares method to estimate median effective concentration (EC50; Graphpad, San Diego, CA, U.S.A.).

RESULTS

Unfasted plasma glucose levels were similarly elevated (p < 0.001) in control (39.9 ± 1.6 mM) and ARI-treated (45.5 ± 1.2 mM) diabetic groups compared with the nondiabetic control group (9.0 ± 0.4 mM). Body weights at the start of the investigation were in the range of 435 ± 4 to 460 ± 5 g. Diabetes caused a weight loss of ∼33% after 2 months; final body weights for diabetic control (287 ± 6 g) and ARI-treated (330 ± 12 g) diabetic rats were not significantly different.

Corpus cavernosum experiments

Relaxation to ACh at ≥10 μM was impaired by 2 months of diabetes (Fig. 1A). Maximal relaxation showed a 35.5 ± 4.3% (p < 0.001) deficit compared with the nondiabetic group, without a significant change in (−log)EC50. ARI treatment from diabetes onset completely prevented these relaxation deficits (p < 0.01). There were no significant between-group differences in the tension levels attained during PE precontraction of corpus cavernosum (Table 1). Relaxation to SNP after PE precontraction was similar in nondiabetic, untreated diabetic, and ARI-treated diabetic rats (Fig. 1B): maximal relaxation and (−log)EC50 values did not vary significantly between groups.

FIG. 1
FIG. 1:
Effects of diabetes and long-term aldose reductase inhibitor treatment on phenylephrine-precontracted corpus cavernosum relaxation to acetylcholine (A) and sodium nitroprusside (SNP) (B). Nondiabetic control (○), 2-month diabetic control (•), 10 mg/kg/day WAY121509-treated 2-month diabetic (▪) groups. Data expressed as group mean ± SEM, n = 12 to 15; **p < 0.01 vs. diabetic control and WAY121509-treated groups.
TABLE 1
TABLE 1:
Effects of diabetes and aldose reductase inhibitor treatmenta

TEFS of corpus cavernosum elicited contractions (Fig. 2A) that could be eliminated by 1 μM tetrodotoxin or 5 μM guanethidine (27). Neither diabetes nor ARI treatment caused significant changes in responses compared with the nondiabetic control group. TEFS in the presence of guanethidine and atropine, after PE precontraction, elicited NANC relaxations (Fig. 2B) that could be completely abolished by 10 μM NOLA (27). Eight weeks of untreated diabetes impaired this nitrergic response at stimulation frequencies of ≥8 Hz. Maximal relaxation (Table 1) was 32.9 ± 3.8% (p < 0.001) reduced by diabetes. These diabetic deficits were prevented by ARI treatment; the impairment of maximal relaxation was attenuated by 84.0 ± 18.3% (p < 0.001).

FIG. 2
FIG. 2:
Effects of diabetes and long-term aldose reductase inhibitor treatment on corpus cavernosum responses to transmural electrical field stimulation. Contractile responses to stimulation (A) and relaxation responses of phenylephrine-precontracted cavernosum in the presence of atropine and guanethidine (B). Nondiabetic control (○), 2-month diabetic control (•), 10 mg/kg/day WAY121509-treated 2-month diabetic (▪) groups. Data expressed as group mean ± SEM, n = 12; **p < 0.01 vs. diabetic control and WAY121509-treated groups.

Mesenteric vascular bed experiments

Precontraction with PE (10-100 μM) caused pressure development of 45.1 ± 3.5, 42.6 ± 5.8, and 34.2 ± 5.9 mm Hg in mesenteric vascular beds from nondiabetic, diabetic, and ARI-treated diabetic groups, respectively, which were not significantly different. Vasodilation in response to ACh at concentrations ≥3 nM(Fig. 3A) was impaired by diabetes. The concentration-response curve for ARI treatment showed greater vasorelaxation for ACh concentrations ≥30 nM compared with untreated diabetes, but deficits remained compared with the nondiabetic group for ≥10 nM ACh. Diabetes reduced maximal vasodilation (Table 2) by 25.2 ± 4.0% (p < 0.001); this was partially (50.6 ± 9.0%; p < 0.05) attenuated by ARI treatment. ACh (−log)EC50 was markedly reduced by diabetes (p < 0.001), corresponding to a 5.8-fold reduction in sensitivity. This was partially prevented by ARI treatment (p < 0.001), although a twofold sensitivity reduction remained (p < 0.01) compared with the nondiabetic group.

FIG. 3
FIG. 3:
Effects of diabetes and long-term aldose reductase inhibitor treatment on vasodilation of the phenylephrine-preconstricted mesenteric vascular bed. Acetylcholine responses in the absence (A) and presence (B) of the NOS inhibitor, N G-nitro-L-arginine; sodium nitroprusside responses in the presence of N G-nitro-L-arginine (C). Nondiabetic control (○), 2-month diabetic control (•), 10 mg/kg/day WAY121509-treated 2-month diabetic (▪) groups. Data expressed as group mean ± SEM, n = 12; *p < 0.05 vs. nondiabetic control group; #p < 0.05 effects of WAY121509 vs. diabetic control group.
TABLE 2
TABLE 2:
Effects of diabetes and aldose reductase inhibitor (ARI) treatment with WAY121509 on vasodilation to acetychloline and sodium nitroprusside in the mesenteric vascular bed

During coperfusion with NOLA (Fig. 3B), PE precontraction (3-10 μM) caused increased of perfusion pressure in mesenteric vascular beds from nondiabetic (69.8 ± 9.3 mm Hg), diabetic (61.6 ± 11.2 mm Hg), and ARI-treated diabetic (58.9 ± 11.2 mm Hg) groups, which did not differ significantly. ACh-induced vasodilation was depressed for ≥3 nM in nondiabetic and ARI groups, ≥10 nM in the diabetic group (p < 0.01), and for maximal vasodilation (Table 2) in all groups (p < 0.0001). Compared with the nondiabetic group, decreased vasorelaxation was noted with diabetes (≥3 nM ACh) and ARI treatment (≥30 nM ACh); however, the ARI-treated group had greater ACh (≥0.3 μM) responses than those of the diabetic group. Compared with controls, maximal vasodilation was 73.5 ± 5.0% reduced by diabetes (p < 0.001). ARI treatment attenuated this diabetic deficit by 43.5 ± 6.8% (p < 0.001), although maximal vasodilation remained partially depressed (p < 0.001) compared with the nondiabetic group. Relaxation to ACh in all groups was unaffected by coperfusion with the cyclooxygenase inhibitor, flurbiprofen (3 μM; data not shown).

ACh (−log)EC50 values (Table 2) were reduced by NOLA (p < 0.0001) coperfusion in nondiabetic and ARI-treated diabetic groups by ∼0.8 log10 units, corresponding to ∼6.5-fold attenuation of sensitivity. In contrast, the ∼0.3 log10 unit shift for the diabetic group was not statistically significant. The values given in Table 2 were estimated from sigmoid curve fits for pooled data from all rats in each group because the level of vasorelaxation was too low in three of 12 of the diabetic rats to permit individual estimation of (−log) EC50. However, conventional estimates of group (−log)EC50 for nondiabetic and ARI-treated diabetic (all rats) and diabetic (n = 9) groups were in close agreement, being 6.97 ± 0.10, 6.62 ± 0.06, and 6.70 ± 0.15, respectively (nondiabetic or ARI-treated diabetic vs. diabetic group, p < 0.001; nondiabetic vs. ARI-treated diabetic group, p < 0.01).

Figure 3C shows endothelium-independent vasodilation to SNP after PE precontraction in the presence of NOLA. There were no significant between-group differences in maximal vasodilation or (−log)EC50(Table 2).

Basal pressure at the 5 ml/min perfusion rate was 18.6 ± 1.7 mm Hg in mesenteries from nondiabetic rats, and this was reduced (p < 0.05) in the untreated (13.2 ± 1.4 mm Hg) and ARI-treated (12.8 ± 1.1 mm Hg) diabetic groups. Pressor responses over these basal values in response to PE (Fig. 4A) were somewhat reduced by diabetes at 30 and 100 nM. Although the maximal pressor response was not significantly affected by diabetes (Table 3), two-way ANOVA revealed that the whole concentration-response curve was significantly depressed (p < 0.0001). ARI treatment further reduced the overall response (p < 0.01 vs. diabetic control group), although the maximal perfusion pressure did not differ statistically from that in untreated diabetes. PE (−log)EC50 values (Table 3) did not differ between groups. During coperfusion with NOLA (Fig. 4B), the pressor response to PE (p < 0.001) and (−log)EC50 values (p < 0.05) were increased in all groups compared with the response before NOLA (Table 3). Two-way ANOVA showed that whereas the concentration-response curves for untreated and ARI-treated diabetic groups did not differ significantly with NOLA coperfusion, both curves were depressed (p < 0.001) compared with the nondiabetic group.

FIG. 4
FIG. 4:
Effects of diabetes and long-term aldose reductase inhibitor treatment on pressor responses of the mesenteric vascular bed to phenylephrine in the absence (A) and presence (B) of the NOS inhibitor, N G-nitro-L-arginine. Nondiabetic control (○), 2-month diabetic control (•), 10 mg/kg/day WAY121509-treated 2-month diabetic (▪) groups. Data expressed as group mean ± SEM, n = 12; *p < 0.05, **p < 0.01 vs. nondiabetic control group.
TABLE 3
TABLE 3:
Effect of diabetes and aldose reductase inhibitor treatment with WAY121509 on pressor responses to phenylephrine and in the mesenteric vascular bed.

DISCUSSION

The data reveal a marked defect in ACh-induced endothelium-dependent relaxation of vascular tissue in 2-month diabetic rats. In contrast, endothelium-independent relaxation to the NO donor, SNP, was not affected by diabetes in either tissue. Endothelium-dependent relaxation in rat corpus cavernosum depends largely on NO (13,27), suggesting that the diabetic defect in ACh-induced relaxation relates to diminished or altered endothelial function as opposed to insensitivity of vascular smooth muscle to NO. This is the first demonstration that ARI treatment prevents diabetic deficits in NO-mediated endothelial responses of rat corpus cavernosum. Assuming EDHF is not NO, the impairment of mesenteric EDHF-mediated responses is compatible with the notion of a rather generalised endothelial dysfunction in diabetes. WAY121509 partially prevented this EDHF deficit, suggesting a relatively broad protective effect of ARI treatment.

Several ARIs have been shown to prevent defective endothelium-dependent relaxation in aortas from diabetic rats (5,22) and rabbits (21). Furthermore, long-term feeding of nondiabetic rats with a high-galactose diet, which is metabolised to galactitol by aldose reductase, caused an ARI-preventable depression of endothelium-dependent relaxation in a similar fashion to experimental diabetes (6,24). The galactosaemic model is interesting because rats develop diabetes-like complications (neuropathy, nephropathy, and retinopathy), but unlike diabetes, the second half of the polyol pathway is not activated because galactitol is a poor substrate for sorbitol dehydrogenase (19). Thus a substantial part of the diabetic deficit is attributable to hyperactivity in the first polyol pathway step, which is compatible with findings for peripheral nerve perfusion and function (26).

Corpus cavernosum NANC nerve activation causes vasorelaxation that depends primarily on NO, as it is completely blocked by NOS inhibition in rats (13,27). Diabetes caused a defect in NANC-mediated relaxation similar in magnitude to that for endothelium, which was largely prevented by WAY121509 treatment. This is the first report of ARI action on corpus cavernosum innervation. Similar diabetic NANC relaxation deficits have previously been reported for other tissues including anococcygeus muscle, gastric fundus, and duodenum (29-31). ARI effects have been examined only for the anococcygeus preparation, where sorbinil treatment was ineffective (29). However, at the sorbinil dose used, the level of polyol pathway inhibition would be considerably less than for WAY121509. Given that an extremely high blockade is required to prevent neural and vascular dysfunction in diabetic rats (20), this could explain the discrepancy between anococcygeus and corpus cavernosum studies. Compared with NANC nerves, tension production to TEFS of cavernosal noradrenergic nerves was unaffected by diabetes or ARI treatment. Thus 2 months of diabetes does not have a universal deleterious effect on corpus cavernosum nerve supply; nitrergic transmission is particularly vulnerable.

Interpretation of the ACh data from the mesenteric vascular bed based on the effects of NOS inhibition is complicated because of interactions between NO and EDHF mechanisms, the former suppressing the latter (15). However, the data suggest that there are diabetic NO and EDHF deficits, and that ARI treatment promotes higher levels or activity of these mediators. Thus diabetic dysfunction of the NO component, indicated by reductions in ACh (−log)EC50 and the abolition of between-group differences by NOS inhibition, was protected by ARI treatment. When the NO system was blocked to reveal EDHF responses, there was also a very marked diabetic deficit that was partially attenuated by the ARI. Previous reports on the effects of ARI treatment on mesenteric vessels in diabetes have not discriminated between NO- and EDHF-mediated responses and also make conflicting claims. One study (24) found that endothelium-dependent diabetic deficits in histamine and bradykinin responses were attenuated by the ARI, tolrestat, whereas in another study (23), ponalrestat had little effect on the deficit in relaxation to ACh. However, in a subsequent investigation, short-term exposure of mesenteric resistance vessels to high glucose (20-45 mM) caused impaired endothelial function, which was partially prevented when ponalrestat also was added to the tissue bath (32). WAY121509 is considerably more potent than ponalrestat, and it is plausible that the negative data in the long-term ponalrestat study resulted from insufficient polyol pathway blockade (20).

The striking diabetic deficit in mesenteric EDHF-mediated responses agrees with a previous report (10). This is the first study showing partial prevention by ARI treatment. The mechanism to account for this is unclear, in part due to a lack of information on EDHF. Current hypotheses suggest that EDHF may be an arachidonic acid metabolite, perhaps cytochrome P450 derived; a cannabinoid-like substance; or K+(16,17,33). The data from this investigation do not shed further light on EDHF identity; however, they may be indirectly pertinent for the relation between EDHF and NO.

Aldose reductase requires the cofactor, NADPH (19). Thus during hyperglycaemia-enhanced polyol pathway flux, competition between aldose reductase and other NADPH-requiring reactions could occur. NADPH is a cofactor for NOS (34); therefore, it was postulated that a "NADPH-sparing" effect of ARI treatment would improve NO synthesis (5). This argument may also apply to corpus cavernosum NANC innervation. If EDHF production depends on cytochrome P450, which also requires NADPH, then as proposed by Quilley et al. (17), NADPH-competition may explain a dual protective effect of ARIs on NO and EDHF systems in diabetes. However, this hypothesis also has problems. It not certain that cytochrome P450 is involved in EDHF production; the identity of EDHF may differ between vascular beds, and several candidates have been suggested (16). Furthermore, the affinity of NOS for NADPH is an order of magnitude greater than that of aldose reductase; therefore the latter would have to be greatly in excess in endothelial cells to impair NO-mediated vasodilation.

NADPH also is a cofactor for glutathione reductase, which converts glutathione from oxidised (GSSG) to reduced (GSH) form. GSH is a major antioxidant, which is decreased by diabetes (and galactosemia) in several tissues, including lens, nerve, and endothelial cells, all of which contain aldose reductase (19,35-37). However, simple interference with glutathione reductase action does not account for diabetes and ARI effects. Thus the GSH/GSSG ratio would be expected to diminish, whereas total glutathione (GSSG + GSH) would be unchanged. Nevertheless, in peripheral nerve and lens, diabetes does not alter the GSH/GSSG ratio but causes a reduction in total glutathione. This is ARI-reversible, implicating effects on glutathione synthesis (35,36). Consequently, as well as inhibiting tissue polyol accumulation, ARIs have an indirect antioxidant action via the glutathione redox cycle.

Oxidative stress contributes to vascular and neural complications in diabetes. Antioxidant treatment protects endothelial function (38-40) and improved nerve blood flow and conduction velocity (36,41,42). Superoxide and NO combine to form peroxynitrite, inactivating NO and reducing vasodilation. Peroxynitrite rapidly decays, liberating cytotoxic hydroxyl radicals (43). Thus elevated oxidative stress in diabetes could affect endothelium-dependent vasodilation in at least two ways; neutralisation of NO and hydroxyl radical-mediated endothelial damage. These arguments are probably relevant to NANC nerve dysfunction. Corpus cavernosum NANC and endothelium-dependent relaxation deficits in diabetic rats were prevented and partially reversed by treatment with the antioxidant, α-lipoic acid (27). It is unclear whether or how EDHF dysfunction in diabetes could be linked to oxidative stress; however, endothelial damage is a likely candidate, by a peroxynitrite mechanism or other sources of reactive oxygen species such as transition metal-catalysed autoxidation reactions (42).

The PE-mediated increase in perfusion pressure of mesentery from untreated diabetic rats was blunted compared with that for nondiabetic controls, with no change in EC50. This agrees with previous studies on aorta and the mesenteric vascular bed from mature diabetic rats (5,44). However, some studies in isolated mesenteric vessels also revealed diabetes-induced increases in sensitivity and/or maximal responses that were attributed to an impaired NO system (45). The differences between studies probably relate to the vessels examined and variations in the model used, including diabetes duration and severity. Diabetes and the attendant hyperphagia cause hypertrophy of the mesenteric vasculature in rats (46). Under the constant-flow conditions used, this would cause reduced basal pressures, and decreased pressor responses in mesentery from diabetic rats that would remain, even when flow-induced vasodilation was blunted by NOLA. The further reduction in perfusion pressure in WAY121509-treated diabetic rats in the absence but not the presence of NOLA could therefore be explained by preservation of NO-mediated responses that oppose the constrictor action of PE.

In conclusion, inhibition of aldose reductase in diabetic rats has beneficial effects for corpus cavernosum and mesentery endothelium-dependent vasodilation, as well as corpus cavernosum neurogenic vasodilation. This suggests that there are potential therapeutic benefits for erectile dysfunction and diabetic vasculopathy in line with the, albeit modest, effects of ARIs on vascular based complications such as neuropathy observed in clinical trials (47).

Acknowledgment: This research was supported by grants from the British Diabetic Association and the British Heart Foundation. We thank Dr. Tom Hohman of Wyeth-Ayerst Inc. for the gift of WAY121509.

REFERENCES

1. Cameron NE, Cotter MA. The relationship of vascular changes to metabolic factors in diabetes mellitus and their role in the development of peripheral nerve complications. Diabetes Metab Rev 1994;10:189-224.
2. Tesafaye S, Malik R, Ward JD. Vascular factors in diabetic neuropathy. Diabetologia 1994;37:847-54.
3. Koya D, King GL. Protein kinase C activation and the development of diabetic complications. Diabetes 1998;47:859-66.
4. Oyama Y, Kawasaki H, Hattori Y, Kanno M. Attenuation of endothelium-dependent relaxation in aorta from diabetic rats. Am J Physiol 1986;261:H1086-94.
5. Cameron NE, Cotter MA. Impaired contraction and relaxation in aorta from streptozotocin-diabetic rats: role of polyol pathway. Diabetologia 1992;35:1011-9.
6. Cameron NE, Cotter MA. Contraction and relaxation of aortas from galactosaemic rats and the effects of aldose reductase inhibition. Eur J Pharmacol 1993;243:47-53.
7. Mayhan WG. Impairment of endothelium-dependent dilation of the basilar artery during diabetes mellitus. Brain Res 1992;580:297-302.
8. Kihara M, Low PA. Impaired vasoreactivity to nitric oxide in experimental diabetic neuropathy. Exp Neurol 1995;132:180-5.
9. Maxfield EK, Cameron NE, Cotter MA. Effects of diabetes on reactivity of sciatic vasa nervorum in rats. J Diabetes Complications 1997;11:47-55.
10. Fukao Y, Hattori Y, Kanno M, Sakuma I, Kitabatake A. Alterations in endothelium-dependent hyperpolarization and relaxation of mesenteric arteries from streptozotocin-induced diabetic rats. Br J Pharmacol 1997;121:1383-91.
11. Azadzoi KM, Saenz De Tejada I. Diabetes mellitus impairs neurogenic and endothelium-dependent relaxation of rabbit corpus cavernosum smooth muscle. J Urol 1992;147:220-5.
12. Saenz De Tejada I, Goldstein I, Azadzoi K, Krane RJ, Cohen RA. Impaired neurogenic and endothelium-mediated relaxation of penile smooth muscle from diabetic men with impotence. N Engl J Med 1989;320:1025-30.
13. Ignarro LJ, Bush PA, Buga GM, Wood KS, Fukoto JM, Rajfer J. Nitric oxide and cyclic GMP formation upon electrical field stimulation cause relaxation of corpus cavernosum smooth muscle. Biochem Biophys Res Commun 1990;170:843-50.
14. Cameron NE, Cotter MA. The role of linolenic acid in diabetic polyneuropathy. In: Dyck PJ, Thomas PK, eds. Diabetic neuropathy. 2nd ed. Philadelphia: Saunders, 1999:359-67.
15. Zygmunt PM, Ryman T, Hogestatt ED. Regional differences in endothelium-dependent relaxation in the rat: contribution of nitric oxide-independent mechanisms. Acta Physiol Scand 1995;155:257-66.
16. Félétou M, Vanhoutte PM. The alternative: EDHF. J Mol Cell Cardiol 1999;31:15-22.
17. Quilley J, Fulton D, McGiff JC. Hyperpolarizing factors. Biochem Pharmacol 1997;54:1059-70.
18. Hawthorne GC, Bartlett K, Hethrington CS, Alberti KGMM. The effect of high glucose on polyol pathway activity and myoinositol metabolism in cultured human endothelial cells. Diabetologia 1989;32:163-6.
19. Dvornik D, Porte D, eds. Aldose reductase inhibition: an approach to the prevention of diabetic complications. New York: McGraw-Hill, 1987.
20. Cameron NE, Cotter MA, Dines KC, Maxfield EK, Carey F, Mirlees DJ. Aldose reductase inhibition, nerve perfusion, oxygenation and function in streptozotocin-diabetic rats: dose-response considerations and independence from a myo-inositol mechanism. Diabetologia 1994;37:651-63.
21. Tesfamariam B, Palacino JJ, Weisbrod RM, Cohen RA. Aldose reductase inhibition restores endothelial cell function in diabetic rabbit aorta. J Cardiovasc Pharmacol 1993;21:205-11.
22. Otter DJ, Chess-Williams R. The effects of aldose reductase inhibition with ponalrestat on changes in vascular function in STZ-diabetic rats. Br J Pharmacol 1994;113:576-80.
23. Taylor PD, Wickden AD, Mirrlees DJ, Poston L. Endothelial function in the isolated perfused mesentery and aortae of rats with streptozotocin-induced diabetes: effects of treatment with the aldose reductase inhibitor, ponalrestat. Br J Pharmacol 1994;111:42-8.
24. Fortes ZB, Becker C, Oliveira MA, Scivoletto R. Influence of aldose reductase inhibition on the microvascular reactivity in experimental diabetes. Gen Pharmacol 1996;27:917-21.
25. Cameron NE, Cotter MA, Dines KC, Hohman TC. Reversal of defective peripheral nerve conduction velocity, nutritive endoneurial blood flow and oxygenation by a novel aldose reductase inhibitor, WAY-121,509, in streptozotocin-diabetic rats. J Diabetes Complications 1996;10:43-53.
26. Cameron NE, Cotter MA, Basso M, Hohman TC. Comparison of the effects of inhibitors of aldose reductase and sorbitol dehydrogenase on neurovascular function, nerve conduction and tissue polyol pathway metabolites in streptozotocin-diabetic rats. Diabetologia 1997;40:271-81.
27. Keegan A, Cotter MA, Cameron NE. Effect of diabetes and treatment with the antioxidant, α-lipoic acid, on endothelial and neurogenic responses of corpus cavernosum in rats. Diabetologia 1999;42:343-50.
28. Keegan A. Mechanisms underlying defective aorta and corpus cavernosum function in experimental diabetes. PhD thesis. University of Aberdeen, 1997.
29. Way KJ, Reid JJ. The aldose reductase inhibitor sorbinil does not prevent the impairment in nitric oxide-mediated neurotransmission in anococcygeus muscle from diabetic rats. Eur J Pharmacol 1996;318:101-8.
30. Jenkinson KM, Reid JJ. Effect of diabetes on relaxations to nonadrenergic, non-cholinergic nerve stimulation in longitudinal muscle of the rat gastric fundus. Br J Pharmacol 1995;116:1551-6.
31. Martinezcuestav MA, Massuda H, Whittle BJR, Moncada S. Impairment of nitrergic-mediated relaxation of rat isolated duodenum by experimental diabetes. Br J Pharmacol 1995;114:919-24.
32. Taylor PD, Poston L. The effect of hyperglycaemia on function of rat isolated mesenteric resistance artery. Br J Pharmacol 1994;113:801-8.
33. Edwards G, Dora KA, Gardener MJ, Garland CJ, Weston AH. K+ is an endothelium-derived hyperpolarizing factor in rat arteries. Nature 1998;396:269-71.
34. Ignarro LJ. Biosynthesis and metabolism of endothelium-derived nitric oxide. Annu Rev Pharmacol Toxicol 1990;30:535-60.
35. Lou MF, Dickerson JE, Garadi R, York BM. Glutathione depletion in the lens of galactosemic and diabetic rats. Exp Eye Res 1988;46:517-30.
36. Hohman TC, Banas D, Basso M, Cotter MA, Cameron NE. Increased oxidative stress in experimental diabetic neuropathy. Diabetologia 1997;40:A549.
37. Tagami S, Kondo T, Yoshida K, Hirokawa J, Ohtsuka Y, Wakami Y. Effects of insulin on impaired antioxidant activities in aortic endothelial cells from diabetic rats. Metabolism 1992;41:1053-8.
38. Keegan A, Walbank H, Cotter MA, Cameron NE. Chronic vitamin E treatment prevents defective endothelium relaxation in diabetic rat aorta. Diabetologia 1995;38:1475-8.
39. Archibald V, Cotter MA, Keegan A, Cameron NE. Contraction and relaxation of aortas from diabetic rats: effects of chronic antioxidant and aminoguanidine treatments. Naunyn Schmiedebergs Arch Pharmacol 1996;53:584-91.
40. Pieper GM, Siebeneich W. Diabetes-induced endothelial dysfunction is prevented by long term treatment with the modified iron chelator, hydroxyethyl starch conjugated-deferoxamine. J Cardiovasc Pharmacol 1997;30:734-8.
41. Cameron NE, Cotter MA, Archibald V, Dines KC, Maxfield EK. Anti-oxidant and pro-oxidant effects on nerve conduction velocity, endoneurial blood flow and oxygen tension in non-diabetic and streptozotocin diabetic rats. Diabetologia 1994;37:449-59.
42. Cameron NE, Cotter MA. Neurovascular dysfunction in diabetic rats: potential contribution of autoxidation and free radicals examined using transition metal chelating agents. J Clin Invest 1995;96:1159-63.
43. Darley-Usmar V, Wiseman H, Halliwell B. Nitric oxide and oxygen radicals: a question of balance. FEBS Lett 1995;369:131-5.
44. Longhurst PA, Head RJ. Responses of the isolated perfused mesenteric vasculature from diabetic rats: the significance of appropriate control tissues. J Pharmacol Exp Ther 1985;235:45-9.
45. Taylor PD, McCarthy AL, Thomas CR, Poston L. Endothelium-dependent relaxation and noradrenaline sensitivity in mesenteric resistance arteries of streptozotocin-induced diabetic rats. Br J Pharmacol 1992;107:393-9.
46. Cooper ME, Rumble J, Komers R, He-cheng D, Jandeleit K, Sheung-to C. Diabetes-associated mesenteric vascular hypertrophy is attenuated by angiotensin-converting enzyme inhibition. Diabetes 1994;43:1221-8.
47. Boulton AJM, Levin S, Comstock J. A multicentre trial of the aldose-reductase inhibitor, tolrestat, in patients with symptomatic diabetic neuropathy. Diabetologia 1990;33:431-7.
Keywords:

Diabetes; Aldose reductase; Vasculature

© 2000 Lippincott Williams & Wilkins, Inc.