Upregulation of AT₂ receptor and iNOS impairs angiotensin II-induced contraction without endothelium influence in young normotensive diabetic rats

Abstract

Diabetes and insulin resistance are associated with an increased risk of hypertension and cardiovascular disease. Recent evidence demonstrates that AT₂ receptors (AT2R) play an important role in the hemodynamic control of hypertension by vasodilation. The quantitative significance of AT2R in the establishment of diabetic vascular dysfunction, however, is not well defined and needs further investigation. Goto-Kakizaki (GK) rats, a polygenic model of spontaneous normotensive type 2 diabetes, were used to examine any abnormalities in cardiovascular function associated with AT2R at the early stage of the disease without endothelium influence. Using a myograph to measure the isometric force, we observed that ANG II-induced contraction was impaired in denuded GK aorta compared with control Wistar-Kyoto (WKY) aorta and exhibited a retarded AT1R antagonist response and enhanced Rho kinase signaling. When AT1R were blocked, ANG II induced a significant vasodilation of precontracted GK aorta via AT2R. The protein and mRNA of AT2R were increased in diabetic GK denuded aorta. Blocking AT2R restored the ANG II-induced contraction in the GK vasculature to control levels, demonstrating a counteractive role for AT2R in AT1R-induced contraction. Inhibition of inducible nitric oxide synthase (iNOS) by N^G-monomethyl-l-arginine mimicked AT2R inhibition in denuded GK aorta, suggesting that AT2R-induced vasodilation was dependent on iNOS/NO generation. The protein and mRNA of iNOS were also increased in GK aorta. In conclusion, these results clearly demonstrate that enhanced AT2R and iNOS-induced, NO-mediated vasodilation impair ANG II-induced contraction in an endothelium-independent manner at the early stage of type 2 diabetes.

hypertension, diabetes, and insulin resistance often coexist and are leading risk factors for cardiovascular disease (23). One important mechanism responsible for the defective vasorelaxation in diabetes has been linked to defective insulin-mediated relaxation of the vasculature (29, 40, 48). It has been reported that angiotensin II (ANG II) inhibits insulin signaling, induces insulin resistance (15, 53), and is an important mediator in the pathophysiology of diabetic complications, acting via at least two different types of receptors, type 1 (AT1R) and type 2 (AT2R).

AT2R is upregulated in certain pathological conditions such as hypertension, vascular injury, and inflammation (1). Recent evidence implicates AT2R in the regulation of renal and cardiovascular function, including vasodilation and natriuresis (26). The AT2R mediates cellular differentiation and growth, opposing the actions of AT1R stimulation (11, 25, 26), and therefore is very important in preventing tissue remodeling and disease progression. AT2R stimulation produces an autacoid vasodilator cascade composed of bradykinin (BK), nitric oxide (NO), and guanosine cyclic 3′,5′-monophosphate (cGMP) that mediates vasodilation, counteracting AT1R-induced contraction (46, 47). Indeed, AT2R knockout mice have higher blood pressure and an exaggerated response to ANG II infusion on blood pressure (27). AT2R-mediated relaxation of isolated resistance arteries as well as aorta has been reported in spontaneously hypertensive rats (SHR) (3, 13, 54, 56). These data support a role for AT2R in the hemodynamic control of hypertension. The quantitative significance of AT2R in the establishment of diabetic vascular dysfunction, however, is not well defined and needs further investigation.

Understanding vascular abnormalities without the influence of the endothelial dysfunction clarifies the abnormalities in AT2R function on diabetic hemodynamic control. The effect of endothelium on hemodynamic control has been well studied (12). Few studies, however, focus on the dysfunction of smooth muscle cells in contraction and relaxation and measure contraction independently from the effects of the endothelium dysfunction in diabetes. Interestingly, a study by Siddiqui and Hussain (45) demonstrated that an abnormal ANG II-induced response was only detected with the endothelium-denuded aorta of obese Zucker rats, since the ANG II-induced contractile response was compensated by the endothelial influence. Insulin resistance in SHR is associated with endothelial dysfunction characterized by the imbalance between NO and endothelin-1 (ET-1) production, suggesting that ET-1 acts as the vasoconstrictive component that is released from the endothelium to change the contractility of smooth muscle cells (37). In this study, we uniquely used denuded aortic rings to measure ANG II-induced contraction to better understand the defects in Goto-Kakizaki (GK) non-obese type 2 diabetes.

Recently, iNOS acquired attention as an important molecule in the mechanism of insulin resistance, linking the overproduction of NO and insulin resistance (34). Since increased inflammatory actions may play an important role in the pathophysiology of cardiovascular abnormalities in hypertension, atherosclerosis, or diabetes, inducible NO synthase (iNOS) may be an important mediator of cardiovascular disease. Vascular smooth muscle (VSM)-derived NO may be of importance in vivo for the local control of vascular function directly at the site of injury. In healthy blood vessels, the prominent source of NO is derived from constitutively expressed NOS in the endothelium. This may no longer be the case under pathological conditions due to vascular dysfunction (43). iNOS produces NO from VSM, whereas Ca²⁺-dependent NOS, such as endothelial (eNOS) or neuronal NOS (nNOS), is not present in VSM cells (VSMC) (6, 43). Expression of iNOS during systemic inflammation (20) or by adenoviral infection (22) contributes to impaired vascular contraction. Nevertheless, the role of medial expression of iNOS in diabetic vasculature and the impact of medial iNOS on diabetic vascular dysfunction have not been investigated. Thus, independently from the endothelium, the role of iNOS on the AT2R-induced vasodilation was examined using endothelium-denuded GK diabetic aorta.

We report that unlike the hypertensive model, spontaneous non-obese, normotensive (35, 52) GK diabetic rats develop impaired ANG II-induced contraction due to increased AT2R and iNOS expression as a counterregulatory reaction to the AT1R-ROK (Rho kinase) contractile signaling. To our knowledge, this is the first study demonstrating a protective role of AT2R counteracting AT1R-dependent contraction via iNOS in the early stage of diabetes without the influence of endothelium dysfunction.

MATERIALS AND METHODS

Chemicals.

ANG II, AT2R antagonist, PD 123,319, the NOS inhibitor N^G-monomethyl-l-arginine (l-NMMA), and specific antibody for β-actin were obtained from Sigma (St. Louis, MO). The AT1R antagonist losartan was a generous gift from EMD-Merck (San Diego, CA). Y-27632 was obtained from EMD (San Diego, CA). Specific antibodies to AT1R and AT2R were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). iNOS antibody was acquired from BD Bioscience (San Jose, CA). Enhanced chemiluminescence (ECL), anti-rabbit IgG, and anti-mouse IgG were obtained from Amersham Biosciences (Amersham, UK). Western blot reagents were obtained from Bio-Rad Laboratories (Hercules, CA).

Animal and tissue preparation.

A colony of type 2 diabetic GK rats was established at Winthrop University Hospital with animals originally supplied by Dr. Robert V. Farese (James A. Haley Veterans Affairs Medical Center, Tampa, FL) as detailed earlier (6, 31). Wistar-Kyoto (WKY) rats were purchased from Taconic Farms (Germantown, NY) and used as controls as described previously (6, 31, 40). Both weight-matched WKY and GK rats were euthanized at the age of 7–8 wk with 95% CO₂ inhalation, which was approved by the Institutional Laboratory Animal Care and Use Committee of Winthrop University Hospital. The thoracic aortas of rats were rapidly and carefully excised, placed in ice-cold physiological saline solution (PSS), and cut into 3-mm-long rings using a multisectioning tool. Endothelium removal was performed by gentle rotation around a thin, polished metal stick as described previously (33). Rings from denuded aortas were mounted simultaneously in a Multi-Chamber myograph (model DMT610M; ADInstruments) filled with PSS of the following composition (in mM): 119.0 NaCl, 24 NaHCO₃, 4.7 KCl, 1.6 CaCl₂·2H₂O, 1.17 MgSO₄·7H₂O, 5.5 glucose, 0.026 EDTA, and 1.18 KH₂PO₄. In the case of GK tissue, PSS contained 20 mM glucose to mimic hyperglycemic condition. The solution in the baths was constantly aerated with 95% O₂ and 5% CO₂ and was kept at 37°C (pH 7.4).

Measurement of aortic ring surface size.

At the conclusion of each experiment, aortas were dissected and cross sections were cut and then dried on a plastic surface. Measurements were performed on a dried aorta centered in the optical field of a ×10 objective of a Nikon SMZ 10 microscope.

Isometric force measurement.

Contractile force, measured with isometric transducers within a Multi-Chamber myograph (DMT610M; ADInstruments), was processed with a data acquisition system (PowerLab 8SP; ADInstruments) and analyzed using Chart 5 software (ADInstruments). After 45 min of equilibration with a resting tension of 2 g for rat aorta, the rings were primed by exposure to activation solution, which substituted 60 mM KCl with 119 mM NaCl, until the contractile response reached its plateau. The ring segments were then incubated with phenylephrine (PE; 1 × 10⁻⁶ M) to get the maximal contractile response. The PE response from each bath was considered as 100% for comparison with each agonist-induced contractile response. Some tissues were pretreated with the AT1R antagonist losartan, the AT2R antagonist PD 123,319, or the NOS inhibitor l-NMMA for 30 min, followed by serial dilutions of ANG II.

Western blotting.

After denudation, aortic tissue was minced and lysed in a buffer as we described recently (31). Lysates were then homogenized with a Dounce homogenizer and centrifuged for 30 min at 14,000 g. Thereafter, Western blotting was performed as previously described (31).

Quantitative real-time PCR.

To amplify iNOS and GAPDH cDNA, sense and antisense oligonucleotide primers were designed based on the published cDNA sequences (38). Since only the AT_1A subtype of AT1R is known to be present in rodent aorta (44), we used AT_1A sequence for AT1R, following a previous publication (17). Oligonucleotides were obtained from Sigma-Genosys (St. Louis, MO). Sequences for the real-time PCR were as follows: iNOS sense, 5′-AGACGCACAGGCAGAGGT-3′; iNOS antisense, 5′-AGGCACACGCAATGATGG-3′; GAPDH sense, GGAGAAACCTGCCAAGTATGA-3′, GAPDH antisense, CCCTGTTGCTGTAGCCATATT-3′; AT1R sense, 5′-CTGAAGCCTGTCTACGAAAATGAG-3′; AT1R antisense, 5′-TAGATCCTGAGGCAGGGTGAAT-3′; AT2R sense, 5′-ACCTTTTGAACATGGTGCTTTG-3′; and AT2R antisense, 5′-GTTTCTCTGGGTCTGTTTGCTC-3′. Aortic tissues dissected from WKY and GK rats were first denuded, and total RNA was isolated using an RNA isolation kit (Qiagen, Valencia, CA). cDNA was synthesized with the 1st Strand cDNA synthesis kit for RT-PCR (Roche Applied Science, Indianapolis, IN) for the first-strand synthesis of single-stranded cDNA from RNA for use as a PCR template by using 2–3 μg of total RNA in a 20-μl reaction volume. For real-time PCR, the cDNA was amplified using a LightCycler 480 SYBR green I master for PCR with the LightCycler 480 system (Roche Applied Science). The double-strand DNA-specific dye SYBR green I was incorporated in the PCR reaction buffer LightCycler 480 to allow for quantitative detection of the PCR product in a 25-μl reaction volume. The temperature profile of the reaction was 95°C for 10 min, 40 cycles of denaturation at 95°C for 30 s, annealing at 62°C for 45 s, and extension at 72°C for 60 s. An internal housekeeping gene control, GAPDH, was used to normalize differences in RNA isolation, RNA degradation, and the efficiencies of the RT. The size of the PCR product was first verified on a 1.5% agarose gel, followed by melting curve analysis.

Immunofluorescence.

Aortic tissues isolated from 8-wk-old WKY and GK rats were denuded and then flash-frozen with liquid nitrogen. The frozen aortas were embedded in OCT compound in cryomolds, cut into 6-μm-thick cryostat sections, and mounted on polylysine-coated slides (Polyscience). Mounted tissue sections were then fixed in ice-cold acetone for 5 min, air-dried for 30 min, and incubated in normal serum block solution for 1 h at room temperature. Sections were incubated with iNOS primary antibody at the appropriate dilution overnight at 4°C. After being washed, they were incubated in FITC-conjugated secondary antibody in PBS for 30 min at room temperature and counterstained with 4,6-diamidino-2-phenylindole (DAPI) for 20–30 min at room temperature for nuclei staining. Finally, they were viewed and photographed with Nikon fluorescent microscopy (Melville, NY).

Data analysis.

Contraction was expressed as a percentage of the response to 1 × 10⁻⁶ M PE. Data are means ± SE. Student's t-test or ANOVA were used to determine the statistical significance. A P value <0.05 was considered statistically significant.

RESULTS

Confirmation of endothelium denudation.

The endothelium influences vascular vasodilation as well as contractility by producing ET-1 and NO (33, 37). To exclude the involvement of the endothelium and only determine the abnormalities in medial smooth muscle function on tone regulation, we removed the endothelium. We routinely confirmed the absence of the endothelium-dependent 5 μM acetylcholine (ACh; Novartis; East Hanover, NJ)-mediated vasodilation. Endothelium denudation was also confirmed by electron microscopy (data not shown).

ANG II-induced contraction is impaired in denuded GK aortic rings.

To determine whether the ANG II-induced contraction was augmented or impaired due to vascular dysfunction in type 2 diabetic GK denuded aorta, we compared the ANG II-induced contraction of aortas in WKY controls with those of the diabetic GK model using a myograph system. The body weight, age, and systolic blood pressure (SBP) of both strains were similar (data not shown). The SBP of WKY and GK rats was not significantly different, which is consistent with previous studies (35, 52). The dose-response curve of ANG II-induced contraction was shifted toward the right and downward, demonstrating impaired ANG II-induced contraction (Fig. 1A). We did not detect any significant differences in ANG II-induced contractile responses in high- (20 mM) and low-glucose (5 mM) PSS in either WKY or GK rats (data not shown). The effect of high glucose on the ANG II-induced contractile response may be detectible only after long-term exposure. Since force is affected by aortic size, we confirmed that there were no significant differences between WKY and GK aortic ring sizes (Fig. 1B). To determine any general defects in aorta contractile function, we measured the response to PE (1 × 10⁻⁶ M) and KCl in WKY control and GK diabetic aortic rings. Contraction in response to both PE and KCl was not significantly different between GK and WKY aortas (Fig. 1C), implying that the impaired ANG II-induced contraction in GK aorta was due to changes in ANG II-specific signaling.

Fig. 1.

Impaired ANG II-induced contraction in denuded Goto-Kakizaki (GK) rat aortic rings. Aortic rings from either Wistar-Kyoto (WKY) or GK rats were mounted on a myograph to measure contraction stimulated by ANG II for 5 min. A: ANG II-induced contraction in WKY and GK aortic rings (n = 16). PE, phenylephrine. B: aortic rings were cut open and dried on a plastic surface after each experiment. The width and height of WKY and GK aortas were measured to determine the difference in aortic surface area. C: the contractile response stimulated by PE and KCl were measured with the myograph, to compare any difference in PE- and KCl-induced contractile response between WKY control and GK diabetic aortas. *P < 0.05, WKY vs. GK.

AT1R signaling in GK aorta.

Since the AT1R is known to be responsible for ANG II-induced contraction, impaired ANG II-induced contraction may be due to the defective AT1R signaling. To investigate the involvement of AT1R in impaired ANG II-induced contraction in GK aorta, we used the AT1R antagonist losartan. Losartan (100 nM) decreased ANG II-induced contraction by ∼83% in WKY aorta but did not significantly reduce the ANG II-induced contraction in GK aorta over the range of ANG II tested (Fig. 2A). At 10 μM, however, losartan did reduce ANG II-induced contraction significantly (Fig. 2B), demonstrating the AT1R dependency of ANG II-induced contraction in GK aorta.

Fig. 2.

Enhanced AT₁ receptor (AT1R) signaling in GK aortic rings. Aortic rings from either WKY or GK rats were mounted on a myograph and pretreated with the AT1R antagonist losartan for 30 min. ANG II was stimulated to cause contraction. A: the effect of AT1R antagonist (1 × 10⁻⁷ M) on ANG II-induced contraction (n = 3–5). B: the effect of AT1R antagonist (1 × 10⁻⁵ M) on ANG II-induced contraction (n = 4–5). *P < 0.05, control vs. losartan treatment.

Enhanced Rho kinase signaling in GK aorta.

The small GTPase RhoA is known to be activated by contractile agents and activates ROK. ROK inhibitors have been shown to restore normal blood pressure in several hypertensive rat models and demonstrate a vital role for ROK in vasoconstriction (49). To determine whether the impaired ROK activity is the cause of impaired ANG II-induced contraction in GK aorta, we used Y-27632, a selective ROK inhibitor. Y-27632 at 1 μM significantly lowered the contractile response in GK aorta but did not cause any reduction of ANG II-induced contraction in WKY aorta (Fig. 3), demonstrating the increased dependency of ANG II-induced contraction on ROK activation. We conclude that ROK activity is augmented in GK diabetic aorta compared with WKY aorta.

Fig. 3.

Enhanced Rho kinase (ROK) signaling in GK aortic rings. Aortic rings from either WKY or GK rats were mounted on a myograph and pretreated with the ROK inhibitor Y-27632 for 30 min. The effect of ROK inhibitor (1 × 10⁻⁶ M) on ANG II-induced contraction is shown (n = 4). *P < 0.05, control vs. Y-27632 treatment.

Increased AT2R expression and a lower AT1R/AT2R ratio causes impaired ANG II-induced contraction in GK aorta.

ANG II effects are thought to be regulated by the balance of AT1R and AT2R expression in target tissues, with AT2R upregulation observed in certain pathological conditions (1). Given that the ANG II AT2R is involved in vasodilation, we decided to determine its level and impact on the ANG II-induced contraction in GK aorta. Western blotting with the use of specific antibodies to angiotensin receptors showed no significant change in AT1R expression in lysates obtained from GK and WKY control aortas (Fig. 4A). However, the expression of AT2R in GK diabetic aorta was increased 3.6-fold over that of WKY aorta (Fig. 4, A and B), resulting in an overall decrease in the AT1R/AT2R ratio. We then measured the mRNA level of AT1R and AT2R using quantitative real-time PCR. Denuded diabetic GK aorta expressed 14-fold higher AT2R mRNA than WKY aorta (Fig. 4C), whereas AT1R mRNA levels were similar in both strains. Thus AT2R protein as well as gene expression is upregulated in denuded GK diabetic aorta.

Fig. 4.

Upregulation of AT₂ receptor (AT2R) causes defective ANG II-induced contraction in GK aortic rings. A and B: upregulation of AT2R protein in GK aortic rings (n = 7). *P < 0.05, WKY vs. GK. C: upregulation of AT2R mRNA (n = 3). Aortic rings from either WKY or GK rats were homogenized with lysis buffer. The protein or mRNA quantification was performed using Western blotting or real-time PCR. *P < 0.05, WKY vs. GK. D: AT2R-dependent ANG II-induced relaxation. To determine AT2R-dependent vasorelaxation, we pretreated aortic rings with the AT1R antagonist losartan (1 × 10⁻⁴ M; 30 min) followed by the contractile agent PE (1 × 10⁻⁷ M). ANG II was then added to cause relaxation (n = 3). *P < 0.05, WKY vs. GK. E: the effect of AT2R antagonist PD 123,319 on ANG II-induced contraction. Aortic rings from either WKY or GK rats were mounted on a myograph and pretreated with AT2R antagonist PD 123,319 (1 × 10⁻⁷ M; 30 min) followed by ANG II stimulation for contraction (n = 6–10). *P < 0.05, control vs. PD 123,319 treatment.

To determine AT2R-dependent vasorelaxation, we pretreated aortic rings with the AT1R antagonist losartan (100 μM) for 30 min to mask AT1R, followed by the addition of the contractile agent PE (100 nM). ANG II (1 μM) was then added to induce a 15% relaxation of PE-induced contraction (Fig. 4D). We then examined whether AT2R-mediated vasodilation caused the impaired contraction in GK aorta using the AT2R antagonist PD 123,319. Pretreatment with PD 123,319 (100 nM) did not significantly increase the ANG II-induced contraction in WKY aorta. However, it did significantly increase the ANG II-induced contraction levels in GK aorta, restoring the impaired contraction back to control levels (Fig. 4E), indicating that vasodilation resulting from AT2R overexpression most likely caused the impaired ANG II-induced contraction in GK diabetic aorta as shown in Fig. 1A.

NOS inhibition mimics the effect of AT2R antagonist on ANG II-induced contraction.

The AT2R is known to cause vasodilation via the cGMP/NO pathway (4, 18, 41, 47). Insulin-resistant GK diabetics with vascular dysfunction (Fig. 3) may have developed a compensatory pathway through vasodilation properties via AT2R. Since we used endothelium-denuded aortas, we next determined whether any changes in expression of iNOS, known to be expressed in VSM only (6, 43), were present in diabetic GK aorta. iNOS protein expression in GK aorta was twofold higher than in WKY controls (Fig. 5, A and B). Quantitative real-time PCR confirmed that iNOS gene expression in GK diabetic aorta was fourfold higher than in WKY controls (Fig. 5C).

Fig. 5.

Inhibition of nitric oxide synthase (NOS) with N^G-monomethyl-l-arginine (l-NMMA) mimics the AT2R antagonist effect and restores defective ANG II-induced contraction in GK aortic rings. Aortic rings from either WKY or GK rats were mounted on myograph and pretreated with the NOS inhibitor l-NMMA for 30 min. ANG II was stimulated to cause contraction. A and B: upregulation of inducible NOS (iNOS) protein in GK aortic rings (n = 3). *P < 0.05, WKY vs. GK. C: upregulation of iNOS mRNA (n = 3). *P < 0.05, WKY vs. GK. D: iNOS protein expression without stimulation in GK rat aorta. WKY and GK aortas dissected from 7- to 8-wk-old rats were washed in PBS, quickly frozen in isopentane, and sectioned. Frozen sections were stained with a specific antibody against iNOS and secondary antibody conjugated with Alexa 555, counterstained with 4,6-diamidino-2-phenylindole (DAPI) to recognize nuclei, and then visualized with a fluorescence microscope. Da: Alexa 555 staining positive for iNOS protein expression. Db: DAPI staining for nuclei. Dc: tissue section. Top images show the series of GK aorta sections; bottom images show the series of WKY aorta. Data are representative of 3 experiments. E: the effect of NOS inhibitor on ANG II-induced contraction (n = 4–10). *P < 0.05, control vs. l-NMMA.

To exclude the possibility that an artificial effect of cell culture in high glucose caused iNOS expression, we carefully examined aortas dissected directly from GK diabetic rats for the increased expression of iNOS. Immunofluorescent labeling of frozen aorta sections was performed as described previously with slight modification (8, 39). As shown in Fig. 5D, iNOS (Fig. 5A) was expressed in the GK aorta medial layer but was not present in WKY aorta, demonstrating iNOS upregulation in GK diabetic aorta. Counterstain of DAPI (Fig. 5B) showed the localization of nuclei. iNOS expression in unstimulated GK aorta was present and localized in the nucleus as well as the cytoplasm. This confirmed the expression of iNOS in GK tissue samples and proved that iNOS overexpression was not attributable to the artificial cell culture conditions.

To examine the involvement of NO in AT2R-mediated vasodilation, we used l-NMMA, a specific inhibitor of NOS, to inhibit the generation of NO. Whereas l-NMMA at 3 mM did not change the ANG II contraction dose-response curve in WKY aorta, ANG II-induced contraction was significantly increased in GK aorta (Fig. 5E). NOS inhibition mimicked the effect of AT2R antagonist (Fig. 4B). This result implies that AT2R-mediated vasodilation may be via NOS stimulation in the GK diabetic model.

Aortic length-tension curves.

Alterations in vasotension, commonly observed in diabetic vascular tissue, could result from changes in wall elasticity. Active tension is derived from the interaction between myosin and actin, whereas passive tension can develop in the muscle's complex connective tissue. Therefore, we decided to determine the elasticity of GK and WKY rat aortas by measuring passive tension. When stretched, the tension of GK and WKY aortas was measured as a function of their circumferential length, and length-tension curves are presented in Fig. 6. Clearly, the aortas of GK rats were more rigid than those of WKY controls.

Fig. 6.

Length-tension curves of the GK and WKY rat aortas. Aortic rings from either WKY or GK rats were mounted on a myograph and stretched to measure the elasticity (n = 6). *P < 0.05, WKY vs. GK. Delta S/S, change in vessel circumferential length.

DISCUSSION

In this study, we demonstrated that the early stage of normotensive (35, 52) non-insulin-dependent diabetes mellitus (NIDDM) is accompanied by abnormal AT2R and iNOS upregulation in endothelium-denuded aorta, which depresses ANG II-induced contraction due to simultaneous stimulation of AT2R, causing vasodilation via NO. AT2R is upregulated in certain pathological conditions such as vascular injury and inflammation (1). In these pathological states, hemodynamic effects mediated by AT2R (32, 51) may be more apparent. For example, we observed that AT2R induces NO-mediated vasodilation in diabetic GK but not WKY aorta (Figs. 4 and 5). Our data suggest that the respective 3.6-fold and 14-fold increases in GK AT2R protein and mRNA expression (Fig. 4, A and B) play an important role in diabetic counteracting of AT1R-induced contraction. Antagonizing AT2R restored ANG II-induced contraction in GK aorta to control levels. In vitro, the vasodilatory role of AT2R is supported by evidence based on enhanced ANG II-mediated vasoconstriction in the presence of AT2R obstruction or in AT2R knockout mice (25, 27). Moreover, high concentrations of ANG II with AT1R blockade indeed caused vasodilation of precontracted GK diabetic aorta (Fig. 4D), demonstrating a vasodilatory role for AT2R. The upregulation of AT2R and enhanced AT2R-induced vasodilation have been observed in other studies in the brush-border and basolateral membranes in obese Zucker rat (24), in the mesenteric arteries of young SHR (50), and in the thoracic aorta of SHR (5, 32). Therefore, consistent with our current finding, most previous studies using normotensive or hypertensive animals (5, 18), as well as knockout or transgenic mice for the AT2R gene (27), have demonstrated that AT2R mediates a depressed response to ANG II.

AT2R-mediated vasodilation has been reported in both an endothelium-dependent and -independent manner, sharing similar signaling pathways for vasodilation (2, 14, 16, 46, 47). AT2R stimulation activates an autacoid vasodilator cascade composed of BK, NO, and cGMP, in an endothelium-dependent manner (46, 47, 51). Furthermore, AT2R-mediated, endothelium-dependent relaxation also has been found to be via a cytochrome P-450-dependent, NO-independent pathway, which may involve the production of epoxyeicosatrienoic acid and the subsequent opening of large-conductance, Ca²⁺-activated K⁺ channels (BK_Ca) (2). Interestingly, endothelium-independent relaxation in rat aorta was reported recently with the use of l-NMMA and the AT2R antagonist PD 123,319 (14, 16). These studies functionally demonstrated the existence of AT2R located on smooth muscle of rat aortic rings that mediated vasorelaxation via the stimulation of B2 receptors by bradykinin, which in turn resulted in the activation of the NO-cGMP pathway, vasodilator cyclooxygenase product(s), and voltage-dependent and Ca⁺-activated large-conductance K⁺ channels (14, 16). It is assumed that ANG II can induce vasodilation by direct stimulation of AT2R on VSMC (42), which play a role in vasodilation of the vasculature. We believe that the upregulation of AT2R and iNOS is in the medial layer, given the confirmation of denudation by electron microscopy, as well as the increased AT2R and iNOS mRNA and protein expression in denuded GK diabetic aorta (Figs. 4, A–C, and 5, A–C). We also confirmed this using immunocytochemistry, showing that the iNOS is expressed on the GK aortic tissue medial layer (Fig. 5D). Our current finding also reports AT2R-mediated vasodilation to be mediated via NO production in the smooth muscle layer in an endothelium-independent manner. Current studies were performed with denuded vessels to remove the effect of the endothelium in these studies. Although this is a good approach to study the effect of agonists/antagonists on VSM function, it may not truly represent the physiological condition. The intact aorta may behave differently due to the combined effects of the endothelium and smooth muscle. Vascular endothelial cell dysfunction leads to reduction in endothelium-derived relaxing factors such as NO, prostacyclin, and endothelium-derived hyperpolarizing factor or increased production of contracting factors such as ET-1 and thromboxane A₂ (9). Thus the diabetic endothelium may produce increased contractile properties such that the smooth muscle layer may have developed the protective increased vasodilatory system such as AT2R and iNOS, as found in our study. Thus the overall contractile response to ANG II in intact aorta may be different compared with what we found in denuded aorta of GK diabetic rat.

The lower losartan concentration (100 nM) did not antagonize the response to ANG II in GK rat aorta, whereas it did so in WKY rats (Fig. 2A). This could be due to two reasons: either ANG II has higher affinity to the AT1R in GK compared with WKY rats, or losartan has a lower affinity for the AT1R in GK rats. From Fig. 2, the pEC₅₀ (−log EC₅₀) of ANG II in WKY and GK rats was 8.8 and 8.6 M, respectively. Thus the affinity of ANG II in WKY and GK was similar, and the first possibility can be ruled out. Therefore, a lower affinity of losartan to AT1R in GK rats, roughly calculated from the IC₅₀ of losartan in WKY and GK rats with the given data, may partially be the explanation.

We also demonstrated enhanced ROK activities associated with diabetic vascular contraction. These data are in agreement with previous studies in which vascular dysfunction was characterized by increased contractility of VSMC, an abnormal vascular tone (33), and defective vasorelaxation (48), all common abnormalities observed in atherosclerosis, diabetes, and hypertension (48). In vitro studies demonstrated by our laboratory using GK VSMC revealed that ROK activity was increased compared with normal control WKY VSMC (40). This implies that the mechanism of reduced ANG II-induced contraction lies distal to abnormalities in AT1R and ROK-mediated contractile signaling. Thus the ANG II dose response was decreased due to the lower AT1R /AT2R ratio, which resulted in less contraction because of increased AT2R-induced vasodilation. AT2R upregulation may result from the increased inflammation observed in diabetes and compensates for the contractile response, such as AT1R signaling and augmented ROK activity (Figs. 2 and 3), in the diabetic vasculature.

The GK rat is a strain that spontaneously developed NIDDM via repeated inbreeding of glucose-intolerant Wistar rats over several generations (19). The genetic rat model is particularly relevant to understanding human type 2 diabetes because the defects in glucose-stimulated insulin secretion, peripheral insulin resistance, hyperinsulinemia, and hyperglycemia are seen as early as 4 wk after birth. In contrast to many other rodent models of NIDDM, GK rats do not exhibit hypertension (35, 52), hyperlipidemia, or obesity (28). Therefore, the model provides a valuable tool for dissecting the pathogenesis of insulin-resistant normotensive diabetic vascular dysfunction to understand the progression of vascular dysfunction in severe diabetes combined with hypertension (28). Thus we chose to perform these studies with young 8-wk-old GK rats to report the progression of vascular dysfunction.

It is interesting that normotensive diabetic GK rats are accompanied by the compensatory mechanism to contractile function, such as AT2R and iNOS upregulation. Mild hypertensive insulin-resistant obese Zucker rats expressed higher AT1R as well as AT2R to cause an increased ANG II contractile response and high blood pressure (45). The normotensive insulin-resistant type 2 diabetic GK model, however, expressed similar AT1R levels (Fig. 4A) with increased AT2R expression, resulting in impaired ANG II-induced contraction, which in turn may possibly result in similar blood pressure compared with the WKY control (35, 52). Further study using a resistant vessel and in vivo study using AT2R antagonist may clarify the role of AT2R in the blood pressure control in GK diabetic rats.

We found that although defective vasorelaxation is generally accepted in diabetes (40, 48), progression of the dysfunction in relaxation may accompany the compensatory enhanced relaxation stage in both an endothelium-dependent (30) and -independent manner. Similar to our finding, Kobayashi et al. (30) reported enhanced ACh-induced relaxation and impaired norepinephrine-induced contraction due to NO overproduction via eNOS in early-stage GK rats (12 wk old) and determined that the impaired ACh-induced relaxation in later-stage GK rats (36 wk old) was due to reductions in both NO production and NO responsiveness but not eNOS expression. In the type 1 streptozotocin (STZ)-induced diabetic rat model, Pieper (36) showed that there is an early (1 day) increase in endothelium-dependent relaxation, followed by a reversion phase (for 1–2 wk) in which relaxation is normal, and then a final phase (for 8 wk) of impaired relaxation. The current study also found enhanced endothelium-independent AT2R-dependent vasodilation in the early stage of GK due to upregulation of AT2R and iNOS. Gene transfer of iNOS was used to demonstrate that expression of iNOS in blood vessels produces impairment of NO-dependent relaxation as well as contraction (21), as with our current study finding. Impaired contraction and relaxation were improved in carotid arteries expressing iNOS by inhibitors of iNOS (21). Thus impaired ANG II-induced contraction in GK diabetic aorta may be partly due to the increased expression of iNOS. Moreover, unlike the aged animal model, in which the compensation mechanism may have been lost to result in the augmented contractile response, in 8-wk-old GK, ANG II-induced contraction is impaired due to the compensatory upregulation of AT2R and iNOS/NO-dependent vasodilation.

For the first time, we have demonstrated the upregulation of iNOS in the medial layer of aorta playing an important role in hemodynamic control in young type 2 diabetic rat aorta. Previous studies from our laboratory demonstrated the role of iNOS in vasorelaxation via NO and cGMP in VSMC (40). Interestingly, iNOS upregulation is known not only to be vasodilatory on vasculature but also to cause insulin resistance (34), implying an important role of iNOS in dysfunction of the diabetic vasculature that has impaired insulin-induced vasodilation. Age, duration, and severity of disease states such as hypertension and diabetes in rodent models may determine the ANG II-induced vascular hemodynamic control in relation to iNOS. Active guanylyl cyclase is impaired in all ages of GK rat (55), and oxidative stress is increased (7) in GK diabetic rats. In GK diabetic (28) and STZ-induced diabetic mice (10, 30), even eNOS is upregulated in all stages of diabetes, and the endothelium dysfunction is severe in old age (28), implying other factors such as oxidative stress reducing the NO bioavailability (10). The relationship between superoxide (O₂⁻) and NO was associated with a marked increase in the protein expression of NOS and a decrease in the level of its cofactor, tetrahydrobiopterin (BH₄), in diabetic aortas (7). Thus the increase in iNOS protein expression may be linked to the increased generation of NO in GK diabetic vasculature. The iNOS adenoviral overexpression caused the impaired contraction as well as impaired vasorelaxation (22). An enhancement in O₂⁻ production, which may result in a diminution in NO bioavailability in aortic tissues of GK diabetic rats (7) at the later stage of diabetes, may rule out the protective enhanced vasodilation with increased AT2R/iNOS/NO signaling, resulting in an enhanced vasocontractile status.

Perspectives and Significance

This is the first study to demonstrate that endothelium-independent AT2R-mediated vasorelaxation via increased expression of iNOS plays an important role in protecting diabetic vessels from vascular dysfunction. Our data emphasize the important and protective role of AT2R-iNOS on the insulin-resistant diabetic vasculature in which the insulin-responsive vasodilatory mechanism is severely impaired. We postulate that this compensatory mechanism may be therapeutically exploited. Our data also provide important information for understanding the progression of diabetic vascular dysfunction with an enhanced protective vasodilatory mechanism against early vascular dysfunction in a young type 2 diabetic rat model. We intend to study this phenomenon further to clarify the abnormalities in a severe or later stage of diabetes combined with hypertension based on the current finding.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute Grant 5 R01 HL067953-04.