Treatment of Streptozotocin-Induced Diabetic Rats With AVE7688, a Vasopeptidase Inhibitor: Effect on Vascular and Neural Disease

Unless stated otherwise, all chemicals used in these studies were obtained from Sigma Chemical (St. Louis, MO). AVE7688 was a generous gift from Dr. Jurgen Punter of Sanofi Aventis.

Male Sprague-Dawley (Harlan Sprague Dawley, Indianapolis, IN) rats 13–14 weeks of age were housed in a certified animal care facility, and food (#7001; Harlan Teklad, Madison, WI) and water were provided ad libitum. All institutional (approval ACURF #0210257) and National Institutes of Health guidelines for use of animals were followed. Diabetes was induced by intravenously injecting streptozotocin (55 mg/kg in 0.9% NaCl, adjusted to pH 4.0 with 0.2 mol/l sodium citrate). Control rats were injected with vehicle alone. Diabetes was verified 48 h later by evaluating blood glucose levels with the use of glucose-oxidase reagent strips (Lifescan, Milpitas, CA). Rats with a blood glucose level ≥300 mg/dl (16.7 mmol/l) were considered diabetic.

Four weeks after the verification of diabetes, two groups of rats were established: an untreated group and a group treated with AVE7688 (450 mg/kg in the diet). The AVE7688 was mixed in the meal form of the diet and the diet pelleted and dried in a vacuum oven for 16 h at 37°C. The treatment phase of the study lasted 8–10 weeks. The 12- to 14-week duration of diabetes was used in order to study the effect of AVE7688 on thermal response latency. This dose of AVE7688 has been shown to ameliorate diabetic nephropathy in Zucker diabetic fatty rats and in our studies blocked serum ACE activity (22,23). For this study, any diabetic rat that lost >10% of its initial body weight was treated with 1–2 units of insulin every other day in order to maintain weight. This dose of insulin did not correct hyperglycemia.

On the day of the experiment, nonfasting blood glucose was determined and the rats anesthetized with Nembutal (50 mg/kg i.p.; Abbott Laboratories, North Chicago, IL). Nerve conduction and endoneurial blood flow studies were determined, and afterward the abdominal aorta was isolated and occluded 1–2 cm above the branch of the common iliac artery. The rat was then killed by exsanguination and body temperature lowered with topical ice followed by isolation and removal of the epineurial arterioles for vascular studies.

MNCV was determined as previously described using a noninvasive procedure in the sciatic-posterior tibial conducting system (1–4). Sensory nerve conduction velocity (SNCV) was determined using the digital nerve to the second toe as described by Obrosova et al. (29). The MNCV and SNCV were reported in meters per second.

Immediately after determination of conduction velocities, sciatic nerve endoneurial nutritive blood flow was determined as previously described (1–4). The hydrogen clearance data were fitted to a mono- or bi-exponential curve using commercial software (Prism; GraphPad, San Diego, CA), nutritive blood flow (ml · min⁻¹ · 100 g⁻¹) was calculated, and vascular conductance (ml · min⁻¹ · 100 g⁻¹/mmHg) was determined by dividing nutritive blood flow by the average mean arterial blood pressure. Two recordings were made for each rat at different locations along the nerve and the final blood flow value averaged.

Videomicroscopy was used to investigate in vitro vasodilatory responsiveness of arterioles vascularizing the region of the sciatic nerve (branches of the superior gluteal and internal pudendal arteries) as previously described (1–4). Cumulative concentration-response relationships were evaluated for acetylcholine (10⁻⁸–10⁻⁴ M) and CNP (10⁻⁷ −3 × 10⁻⁶ M). At the end of each dose response determination, sodium nitroprusside (10⁻⁴ M) was added. Afterward, we added papaverine (10⁻⁵ M) to determine maximal vasodilation, which was consistently the same as the vascular tone of the resting vessel. For some studies vessels were preincubated with charybdotoxin and apamin or N-ω-nitro-l-arginine (LNNA) before the addition of acetylcholine or CNP.

The day before the terminal studies, thermal nociceptive response in the hind paw was measured using the Hargreaves method with instrumentation provided by IITC Life Science (model 390G; Woodland Hills, CA). The rat was placed in the observation chamber on top of the thermal testing apparatus and allowed to acclimate to the warmed glass surface (30°C) and surroundings for a period of 15 min. The mobile heat source was maneuvered so that it was under the heel of the hind paw and then activated, a process that activates a timer and locally warms the glass surface: when the rat withdrew its paw, the timer and the heat source were turned off (30). Following an initial recording, which was discarded, four measurements were made for each hind paw, with a rest period of 5 min between each measurement. The mean of the measurements reported in seconds were used as the thermal nociceptive response.

Hydroethidine (Molecular Probes, Eugene, OR), an oxidative fluorescent dye, was used to evaluate in situ levels of superoxide (O₂⁻) in epineurial vessels as described previously (1–3). This method provides sensitive detection of O₂⁻ in situ. Superoxide levels were also measured in the aorta by lucigenin-enhanced chemiluminescence as described previously (1–3).

One of two mechanisms by which acetylcholine can mediate vascular relaxation in arterioles that provide circulation to the sciatic nerve is through the production of nitric oxide (31). The chemistry of nitric oxide is complex, and several biochemical pathways other than nitric oxide production can influence nitric oxide action. For example, superoxide anion can interact with nitric oxide to form peroxynitrite, reducing nitric oxide bioactivity. To determine whether formation of superoxide by arterioles that provide circulation to the sciatic nerve promotes the formation of peroxynitrite, we measured 3-nitrotyrosine (a stable biomarker of tissue peroxynitrite formation). Briefly, frozen tissue segments of arterioles were cut into 5-μm sections and then incubated in PBS solution containing 1% Triton X-100 and 0.1% BSA for 30 min at room temperature. Afterward, the samples were incubated in this buffer solution containing mouse anti-nitrotyrosine antibody (Upstate, Lake Placid, NY) overnight at 4°C. After washing, the sections were incubated for 2 h with Alexa Fluor 555 goat anti-mouse IgG (Molecular Probes, Eugene, OR). Sections were then rinsed and mounted with VectorShield. The labeled vessels derived from these studies were visualized with an Olympus IX71 inverted research microscope interfaced with a PC containing SimplePCI imaging software.

We analyzed for CNP by immunohistochemically staining 10-μm sections of epineurial arterioles of the sciatic nerve. We generally followed the methods described by Nakanishi et al. and Woodard et al. (32,33). Epineurial arterioles of the sciatic nerve were collected as described above with minimal preparation embedded in Tissue-Tek OCT compound (Sakura Finetek, Torrance, CA) and sectioned. These sections were then incubated with the primary antibody 40 μg/ml (T-4218 rabbit anti–C-type natriuretic peptide [1–29] [porcine] IgG; Peninsula Laboratories, San Carlos, CA) for 16 h in 0.01 mol/l PBS containing 0.1% BSA and 0.1% Triton X-100. The sections were washed and incubated with the secondary antibody Alexa Fluor-546–conjugated IgG in buffer for 2 h (Molecular Probes). Following this incubation, vessels were washed with 0.01 mol/l PBS and water and mounted with VectorShield. The sections were then visualized using an Olympus IX71 inverted research microscope. For a negative control the incubation step with the primary antibody was omitted. Optimal settings for the microscope and exposure was determined and left constant for recording of all the samples. Pixel intensity for CNP immunostaining was determined for each vessel segment and averaged for each condition. The same procedures were generally used to analyze for neutral endopeptidase. For the primary antibody, we used anti–CD-10 rabbit polyclonal IgG (Santa Cruz Biotechnology, Santa Cruz, CA). The sections were washed as described above and incubated with the secondary antibody Alexa Fluor-546–conjugated IgG in buffer for 2 h (Molecular Probes).

ACE activity in the serum was quantitated using a colorimetric assay kit from ALPCO Diagnostics (Windham, NH) and the data presented as milliunits per milliliters of serum (4). One unit of ACE activity is defined as the amount of enzyme required to release one micromole of hippuric acid per minute and per liter of serum at 37°C. Serum TBARS (thiobarbituric acid reactive substance) was determined as previously described (1–3).

The results are presented as means ± SE. Comparisons between the groups for MNCV, SNCV, endoneurial blood flow (EBF), thermal algesia, serum TBARS, and serum ACE activity were conducted using a one-way ANOVA and Newman-Keuls test for multiple comparisons (Prism software; GraphPad). Statistical analyses of these data were also performed using Bonferroni-Dunn test, and results were similar. Dose-response curves for acetylcholine- and CNP-induced relaxation were compared using a two-way repeated-measures ANOVA with autoregressive covariance structure using the proc mixed program of SAS (1–4). Whenever significant interactions were noted, specific treatment-dose effects were analyzed using Bonferroni adjustment. A P value <0.05 was considered significant.

Data in Fig. 1 demonstrate that epineurial arterioles express CNP primarily in the endothelial layer. The top potion of Fig. 1 demonstrates that no immunostaining can be detected when the primary antibody is omitted from the procedure.

We next examined the effect of CNP on vascular relaxation in epineurial arterioles. Data in Fig. 2 demonstrate that CNP caused a dose-dependent vascular relaxation with a maximum affect occurring at 3 × 10⁻⁶ M. Compared with acetylcholine, CNP-induced relaxation was ∼60% of the maximum acetylcholine response. We previously reported that acetylcholine-induced vascular relaxation in epineurial arterioles is mediated by two mechanisms involving the production of nitric oxide and EDHF (9,31). The former is blocked by LNNA and the latter by the combination of charybdotoxin and apamin (9,31). Data in Fig. 3 demonstrate that the combination of charybdotoxin and apamin, as previously described, inhibited acetylcholine-mediated vascular relaxation by ∼60% (9,31), whereas charybdotoxin and apamin inhibited the relaxation mediated by 10⁻⁶ M CNP by almost 90%. In studies using LNNA, we found that it had no effect in inhibiting CNP-mediated vascular relaxation (data not shown).

Diabetes causes a decrease in acetylcholine-mediated vascular relaxation in epineurial arterioles (31). Data in Fig. 4 demonstrate that streptozotocin-induced diabetes of 8 weeks’ duration caused a decrease in CNP-mediated vascular relaxation of ∼60%.

It has been demonstrated that neutral endopeptidase can degrade natriuretic peptides and that this can be prevented by vasopeptidase inhibitors, which are a form of drug that contains ACE inhibitor and neutral endopeptidase inhibitor activities (12,20).

Our studies demonstrate that epineurial arterioles express neutral endopeptidase activity primarily in the outer muscle and/or adventitia layer of the vascular wall (Fig. 5). Streptozotocin-induced diabetes caused a modest increase (35%) in the expression of neutral endopeptidase. Analysis of multiple sections from 12 different vessel segments taken from four control and four streptozotocin-induced diabetic (8 weeks’ duration) rats indicated an increase in staining for neutral endopeptidase from 68.8 ± 8.7 to 91.5 ± 5.9 for control and diabetic rats, respectively (data are relative light units, P < 0.05). Based on these data we were interested in determining the effect of treatment of diabetic rats with AVE7688, a vasopeptidase inhibitor, on vascular and neural function.

Table 1 and Fig. 6 provide data on the effect of AVE7688 treatment of diabetic rats on vascular and neural function. Treatment was initiated 4 weeks after the onset of diabetes and lasted 8–10 weeks. Diabetic rats failed to gain weight, and at the time of experimentation the weight of diabetic rats and diabetic rats treated with AVE7688 was significantly less than that of control rats. Blood glucose levels in diabetic rats and diabetic rats treated with AVE7688 were significantly increased compared with control rats. There was no difference in body weight or blood glucose levels between diabetic rats and diabetic rats treated with AVE7688. Treating diabetic rats with AVE7688 significantly improved EBF, MNCV, and SNCV; prevented the development of hypoalgesia; and reduced aorta superoxide levels. The data in Table 1 for EBF is the nutritive flow. Conductance values for control rats, untreated diabetic rats, and diabetic rats treated with AVE7688 were 0.144 ± 0.016, 0.066 ± 0.012, and 0.201 ± 0.034*, respectively (*P < 0.05 compared with untreated diabetic rats). As previously reported, diabetes caused a significant decrease in acetylcholine-mediated vascular relaxation, and this was completely corrected with treatment of diabetic rats with AVE7688 (Fig. 6).

Data in Fig. 7 demonstrate that diabetes caused a significant increase in superoxide and nitrotyrosine in epineurial arterioles of the sciatic, and this was reduced by treatment with AVE7688.

The major findings of this study were: 1) epineurial arterioles of the sciatic nerve express CNP; 2) CNP is vasoactive in these vessels with characteristics consistent with EDHF; 3) diabetes decreased CNP-mediated vascular relaxation; 4) epineurial arterioles express neutral endopeptidase, and its expression is increased with diabetes; and 5) AVE7688 (a vasopeptidase inhibitor) treatment of diabetic rats ameliorates vascular and nerve dysfunction and reduces the diabetes-induced increases in superoxide and nitrotyrosine levels in epineurial arterioles.

The natriuretic peptide family consists of ANP, BNP, and CNP. Once in the circulation, ANP and BNP migrate to their target tissues such as the kidney, adrenal gland, and the peripheral vasculature. Binding to natriuretic peptide receptor-A elevates intracellular cGMP, which results in renal sodium and water excretion, decreased aldosterone secretion, and vascular smooth muscle relaxation (34). The biological role of CNP is less well known.

We demonstrated that CNP is expressed in endothelial cells of epineurial arterioles that provide circulation to the sciatic nerve. It has been demonstrated that CNP is secreted from endothelial cells and acts on vascular smooth muscle cells to exert vasodilator and antiproliferative effects (35–37). The amount of vasodilation caused by CNP is dependent on the balance of peptide production and breakdown. One explanation for this is the presence of neutral endopeptidase (7). We demonstrated that exogenous CNP caused a concentration-dependent vasodilation of epineurial arterioles that was decreased in vessels isolated from diabetic rats. We also demonstrated that epineurial arterioles express neutral endopeptidase in the outer layer of the vessel wall and adventitia and that neutral endopeptidase expression is moderately increased in diabetes. The expression pattern of neutral endopeptidase may also change in diabetes, with more staining appearing throughout the vessel wall in epineurial arterioles from diabetic rats. One possible explanation for the decrease in vascular relaxation by exogenous CNP in epineurial arterioles from diabetic rats is the increased degradation of CNP due to the increased expression of neutral endopeptidase. In this regard, in human forearm resistance vessel vasodilation to CNP was significantly increased when activity of neutral endopeptidase was blocked, suggesting that neutral endopeptidase activity limits CNP bioactivity (7). Inhibition of neutral endopeptidase has also been shown to improve endothelial function and to protect against atherogenesis in an animal model (38).

It has been reported that CNP may account for the biological activity of EDHF in some vascular tissue (5,6). Using mesenteric resistance arteries, Chauhan et al. (5) demonstrated that CNP-mediated vascular relaxation had characteristics consistent with EDHF. Our results are consistent with CNP having EDHF-like properties. We previously demonstrated that acetylcholine-mediated vascular relaxation in epineurial arterioles was mediated by the release of nitric oxide and EDHF (9,31). The former was blocked by LNNA and the latter attenuated by high potassium concentrations or the combination of charybdotoxin and apamin (9,31). In this study, we found that CNP-mediated relaxation was blocked by ∼90% by the combination of charybdotoxin and apamin but unaffected by LNNA. Charybdotoxin and apamin are inhibitors of the large-conductance Ca²⁺-activated K⁺ channel and small- conductance Ca²⁺-activated K⁺ channel, respectively, and combined they have been widely shown to be effective inhibitors of EDHF activity (9). However, additional studies are necessary to unconditionally prove that CNP is EDHF in the epineurial arterioles.

Our studies highlight the potential importance of CNP in regulating vascular tone and that protecting CNP-mediated vascular function may be an important approach for treating diabetes-related vascular impairment. Thus, we were interested in determining whether treating diabetic rats with a vasopeptidase inhibitor would protect vascular and neural function. Use of a vasopeptidase inhibitor, which inhibits both ACE and neutral endopeptidase, would be expected to protect CNP and perhaps other vasoactive compounds, such as calcitonin gene-related peptide (CGRP), from degradation by neutral endopeptidase. Neutral endopeptidase metabolizes bradykinin, the natriuretic peptides, and adrenomedullin (12,20). These peptides counter the adverse effects of angiotensin II by their vasodilator, natriuretic, and diuretic and autonomic neural actions (39). We recently demonstrated that treating streptozotocin-induced diabetic rats with the ACE inhibitor Enalapril provided an effective approach to preventing/reversing diabetes-induced vascular and neural dysfunction (4). Enalapril treatment provided significant protection against diabetes-induced superoxide formation in epineurial arterioles, and we previously reported that antioxidant treatment of diabetic rats improved vascular and neural dysfunction (1,2,4). However, vascular and neural dysfunction in diabetic rats was not totally corrected with Enalapril treatment, suggesting that more comprehensive treatments or combination therapy is necessary (4).

Vasopeptidase inhibitors have been previously shown to provide greater renoprotection than Enalapril in a rat model subjected to nephrectomy and in lowering blood pressure in hypertensive rats (40). In other rat models of renal vascular disease and hypertension, vasopeptidase inhibitor treatment restored renovascular endothelial function, prevented vascular hypertrophy, and lowered systolic blood pressure (41,42). In Zucker diabetic fatty rats, vasopeptidase inhibitor treatment prevented nephropathy (22,23).

In these studies, we have shown for the first time that treatment of streptozotocin-induced diabetic rats with a vasopeptidase inhibitor significantly improved vascular and neural function. Correction of acetylcholine-mediated vascular relaxation in epineurial arterioles, EBF, and MNCV was nearly 100%, while SNCV was improved and thermal hypoalgesia prevented by ∼75%. Vasopeptidase inhibitor treatment also significantly reduced oxidative stress in the aorta and epineurial arterioles. Overall, the level of efficacy with AVE7688 was the greatest we have observed compared with a variety of other interventions ranging from antioxidants, ACE inhibitors, angiotensin receptor antagonists, aldose reductase inhibitors, aminoguanidine, or myo-inositol (1,2,4,43,44). There are likely several reasons for this level of efficacy. First, we have demonstrated that epineurial arterioles of the sciatic nerve express both CNP and neutral endopeptidase. Moreover, in diabetes the expression of neutral endopeptidase was increased. It has been demonstrated that neutral endopeptidase decreases the bioactivity of CNP and that local concentrations of vasoactive peptides in the vessel wall are regulated by the neutral endopeptidase cleavage pathway in the immediate vicinity of their target cells (7,17). Therefore, in epineurial arterioles of diabetic rats, increased expression of neutral endopeptidase could decrease CNP bioactivity, and treatment of diabetic rats with AVE7688 protected CNP from degradation. Second, we previously demonstrated that treatment with antioxidants protects vascular and neural function in streptozotocin-induced diabetic rats (1,2). Since vasopeptidase inhibitors contain ACE inhibitor activity and we have demonstrated that the ACE inhibitor Enalapril improved diabetes-induced vascular and neural dysfunction and reduced oxidative stress in vascular tissue, it is likely that AVE7688 efficacy is partially due to antioxidant properties. This was supported by studies demonstrating that increased formation of superoxide and nitrotyrosine in epineurial arterioles of diabetic rats was significantly reduced by treatment with AVE7688. Third, it has been demonstrated that another target for neutral endopeptidase is CGRP (45). Neutral endopeptidase plays a major role in the inactivation of CGRP released from sensory fibers and actively degrades CGRP in human skin, whereas inhibition of neutral endopeptidase increases circulating plasma levels of CGRP (46–48). We have demonstrated that epineurial arterioles of the sciatic nerve are innervated by sensory nerves that contain CGRP, and CGRP is an important regulator of vascular tone and blood flow in the sciatic nerve (49). Furthermore, we have demonstrated that in type 1 and type 2 diabetic rats, the bioactivity of exogenous CGRP and CGRP expression was decreased (49,50). Since neutral endopeptidase expression is increased in diabetes, the decrease in CGRP bioactivity in diabetes could partially be due to increased degradation of CGRP. Overall, a decrease in CNP and CGRP bioactivity in vivo would likely compromise vascular tone, impair blood flow, induce ischemia, and subsequently decrease neural function.

In summary, we have demonstrated that epineurial arterioles of the sciatic nerve express CNP and neutral endopeptidase and that exogenous CNP caused vasodilation of epineurial arterioles with characteristics that were similar to EDHF. Streptozotocin-induced diabetes caused a decrease in exogenous CNP-mediated relaxation of epineurial arterioles that may be due to an increased expression of neutral endopeptidase. Since after 4 weeks of untreated diabetes vascular and neural dysfunction already exists, our studies suggest that treating streptozotocin-induced diabetic rats with AVE7688 reversed the diabetes-induced impairment in vascular and neural function. The beneficial affects of AVE7688 are likely due to several factors including protection of the vasoactive effects of CNP, and perhaps also CGRP, and reduction of oxidative stress. Overall, these studies suggest that use of vasopeptidase inhibitors may be a valuable approach for the treatment of diabetic vascular and neural dysfunction.

This work was supported by a grant from the Order of the Amaranth through the American Diabetes Association, a Merit Review Grant from the Veterans Affairs Administration, and National Institute of Diabetes and Digestive and Kidney Diseases Grant DK073990 from the National Institutes of Health (NIH). The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

We extend our appreciation to Sanofi Aventis for supplying AVE7688 for these studies.