Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Although first introduced as an angina therapy over 100 years ago, the organic nitrate vasodilators, such as nitroglycerin (NTG), remain an important class of cardiovascular agents (Bauer et al., 1995). Organic nitrates are prodrugs of the potent vasorelaxant nitric oxide (NO), requiring vascular metabolism for their pharmacological activities (Chung and Fung, 1990; Feelisch and Kelm, 1991). Their unique hemodynamic profile (i.e., predominant relaxation of capacitance vessels at low doses; Bassenge and Zanzinger, 1992; Bauer and Fung, 1996) makes these drugs effective agents for treating both angina pectoris and congestive heart failure (Murrell, 1879; Abrams, 1992). However, rapid tolerance development to both the hemodynamic and therapeutic effects of organic nitrates limits the utility of this otherwise useful drug class (Ahlner et al., 1991; Bauer and Fung, 1994).

Although the clinical problem of nitrate tolerance has been well described, a unifying mechanism of this phenomenon has not been established (Bauer and Fung, 1994). Previous studies by us and others have suggested that physiological responses to selective preload reduction and/or metabolic alterations in vascular tissues may participate in the pharmacodynamic tolerance observed (Needleman and Johnson, 1973; Parker et al., 1991; Feelisch and Kelm, 1991). A recently proposed mechanism of nitrate tolerance implicates increased vascular superoxide anion formation (O₂), reducing NO bioavailability and its relaxation effects (Munzel et al., 1995). O₂ interacts with NO in a diffusion rate-limited reaction to form peroxynitrite (ONOO), a potent cellular toxicant (Pryor and Squadrito, 1995; Beckman and Koppenol, 1996). This aggressive oxidant is known to irreversibly nitrate cellular proteins, forming 3-nitro-L-tyrosine (3NT) (Beckman, 1996). Several recent studies have shown that ONOO is apparently formed in disease settings, including vascular dysfunction during renal transplant rejection (MacMillan-Crow et al., 1996), and that this agent is a potent inhibitor of enzyme processes in vitro (Zou et al., 1997; Mihm and Bauer, 1998). ONOO induced protein nitration has been demonstrated as a mechanism of this enzymatic inhibition (Yamakura et al., 1998). Finally, recent evidence suggests that ONOO is formed in vascular smooth muscle homogenates during NTG tolerance development, as measured by electron spin resonance spectroscopy (Dikalov et al., 1998).

Given the recent evidence that O₂ may mediate vascular tolerance to organic nitrates together with its established chemical interactions with NO, here we hypothesized that nitrate tolerance is associated with increased protein nitration, a stable biomarker of ONOO formation. Additionally, we evaluated the potential contribution of vascular protein nitration as a mediator of nitrate vascular tolerance.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Isolated Vascular Function. Healthy male Sprague-Dawley rats (300-400 g) were euthanized by pentobarbital overdose, and the thoracic aorta was rapidly isolated and prepared as described previously (Bauer and Fung, 1991; Bauer et al., 1997). Vascular segments (2-3 mm) were suspended by stainless steel hooks in 10-ml tissue baths containing Krebs' buffer at 37°C, oxygenated by constant bubbling of a 95/5% O₂/CO₂ mixture. In vitro vascular tolerance was induced by a 30-min preincubation period with 22 µM NTG (EC₉₅ concentration, Nitrocine; Schwarz Pharma, Mannheim, Germany). Vehicle control consisted of Krebs' buffer + 100 µl 5% dextrose vehicle to account for time course and solvent effects. After a series of five rapid washes, vessels were precontracted with 2.5 µM phenylephrine to a magnitude between 1.3 and 1.9 g tension. After precontraction, NTG concentration-effect data were obtained by cumulative addition of NTG in 100-µl aliquots over a concentration range of 4 × 10¹¹ to 2 × 10⁴ M. Vessel tension data were collected by DigiMed Tissue Force Analyzer and System Integrator model 210 (Micro-Med, Louisville, KY).

Vascular Immunohistochemistry. Immediately after functional assessment, vessel segments were immersed in 10% formaldehyde (Formalin) for 48 h, then transferred to tissue cassettes for standard dehydration and paraffin infiltration in an automated tissue processor (Fisher Histomatic model 166; Fisher Scientific, Pittsburgh, PA). After paraffin embedding, 5-µm sections were mounted onto Fisher Scientific ProbeOn Plus slides for immunohistochemical treatment. Tissue sections were heated to 60°C for 30 min followed by immersion in clearing fluid (Hemo-D; Fisher Scientific) to remove paraffin wax. The tissue was partially rehydrated by alcohol gradient (100 to 70%) then immersed in 3% hydrogen peroxide/methanol solution for 10 min to block endogenous peroxidase activity. Slides were rinsed and reheated in citrate buffer (pH 6.0) to recover antigenicity. Tissues were blocked in 10% goat serum (Vector Laboratories, Burlingame, CA)/PBS blocking solution for 30 min, then incubated with rabbit anti-mouse polyclonal antibodies directed against 3NT (1:400 dilution; Upstate Biotechnology, Lake Placid, NY). Staining (isotypic) control tissues were exposed for the same duration to nonimmune rabbit IgG (1:200; Vector Labs) in place of primary antibody. After primary antibody incubation, tissues were washed, then exposed to biotinylated goat anti-rabbit secondary antibody (Vector Labs) for 20 min. Peroxidase enzymes were linked to the antibody complex by incubation in a 1:200 dilution of ABC Elite reagent (Vector Labs) in PBS. Diaminobenzidine (0.06% w/v) was used to provide visualization of immunoreactivity, followed by methyl green counterstaining.

Digital Photomicroscopy and Image Analysis. Immunostained tissues were visualized using light microscopy at 200× magnification (Olympus BX40), digitally captured (Pixera digital camera, 1440 × 760 pixel resolution), and transferred into Image Pro Plus software (Media Cybernetics, Silver Spring, MD) for analysis. Image analysis was conducted using methods similar to Russ (1998). Cross-sectional images of a particular vascular segment encompassed at least 75% of the total circumference. Relative immunoreactivity of the vascular smooth muscle layer was then evaluated by applying intensity thresholding analysis using identical total tissue areas (75,000 µm²). Colored images were converted to gray-scale via extraction of the green channel to remove interference from methyl green counterstaining. Intensity threshold of 150 was predetermined as background staining in control tissues; therefore, the percentage of total pixels in an image falling into the 0 to 150 range was used as a relative measure of immunoreactivity.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Correlation of 3NT immunoprevalence and image area analyzed. TNM-treated tissue sections were used to validate image capture and analysis methods. Number of "positively stained" pixels (i.e., above predetermined intensity threshold) was linearly related to the total area examined, r2 = 0.999.

Data Analysis. Cumulative relaxation data were expressed as a percentage of initial precontraction; relaxation data from each vessel segment was fitted to a sigmoidal E_max model using GraphPad Prizm Software (San Diego, CA). EC₅₀, E_max, and Hill slope were determined for each treatment group. Statistical analyses were performed using one-way ANOVA. Statistical significance was assigned at p < .05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Isolated Vascular Function. NTG tolerance was rapidly induced by 30-min preincubation in EC₉₅ concentration of NTG. Figure 2A illustrates the average concentration-effect data for NTG in control versus tolerant vessels. NTG preincubation resulted in a significant loss of NTG potency, but no loss in the maximal ability of NTG to relax the vessels, as illustrated in Fig. 2A. The fitted vascular relaxation parameters are presented in Table 1. Tolerant vessels demonstrated a 13-fold loss of NTG potency compared with control, as determined by comparison of fitted EC₅₀ values. No difference in fitted E_max was observed among treatment groups.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Vascular responses to NTG. Average concentration-effect data during cumulative additions of NTG in vitro (mean ± S.E.). Tolerance induced by 30-min preincubation in EC95 concentrations of NTG. A: , Control (n = 6); , NTG preincubation (NTG = 22 µM, n = 14). B: , TYR Control (500 µM, n = 6); , NTG preincubation + TYR (NTG = 22 µM, TYR = 500 µM, n = 8). *, statistically significant difference from control EC50, p < .05.

3NT Immunohistochemistry. After functional assessment, vessel segments were fixed and prepared for immunohistochemical analysis to determine the extent of smooth muscle protein nitration. Control segments demonstrated negligible 3NT immunoprevalence, and TYR incubation alone did not influence this parameter; therefore, the two control groups were merged for further statistical comparisons. Representative images are displayed in Fig. 3. Using digital image analysis, control tissues exhibited only 0.08 ± 0.04% positive immunoreactive area (n = 60 images from 12 total vascular segments).

View larger version (66K): [in this window] [in a new window]   Fig. 3.   Representative 3NT immunohistochemistry on vascular segments. After functional assessment, aortic segments were probed for 3NT immunoprevalence. Images (400× original magnification) are grayscaled for intensity thresholding analysis, dark staining indicates immunoprevalence for 3NT. A, control tissue; B, NTG preincubation; C, NTG preincubation + TYR. Orientation: Top, vessel lumen; bottom, adventitial layer.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Digital image analysis of 3NT immunoprevalence determined by intensity thresholding analysis. 3NT prevalence expressed as a percentage of total vascular smooth muscle area demonstrating positive staining for 3NT; *, statistically significant difference from control, p < .05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Organic nitrate tolerance limits the long-term efficacy of an otherwise valuable cardiovascular drug class (Zimrin et al., 1988; Fung et al., 1998). Despite decades of investigation, the mechanisms by which nitrate tolerance develops are poorly understood (Munzel and Harrison, 1997). Recent evidence suggests that increased vascular free radical formation may participate in this phenomenon (Munzel et al., 1995). For example, in vivo experiments and isolated vascular studies have shown that antioxidants such as ascorbic acid (Bassenge et al., 1998) and -tocopherol (Watanabe et al., 1997) can prevent or reduce nitrate tolerance development. In addition, recent studies by Munzel et al. (1995) have shown that vascular tissues from NTG tolerant rabbits have elevated basal formation of O₂, and that this change functionally reduces the vasoactivity of NTG and other NO donors in vitro. These investigators also demonstrated that NTG vascular tolerance was reversible via addition of superoxide dismutase to the incubation medium. Finally, Dikalov et al. (1998) have provided indirect evidence, using electron spin resonance spectroscopy, of ONOO formation in vascular smooth muscle homogenates after acute, high concentration NTG (500 µM) incubation. Although these previous studies suggest a role for O₂, the contribution of ONOO, the toxic end-product of NO destruction by O₂ during nitrate tolerance has not been investigated. Here we hypothesized that vascular protein nitration, indicative of ONOO formation, occurs during nitrate tolerance.

Increased O₂ formation could contribute to nitrate tolerance by a number of mechanisms. O₂ might simply serve to shunt NO produced from NTG away from guanylate cyclase and its vascular smooth muscle activities. Moreover, ONOO is an established inhibitor of several enzymatic and biochemical processes (Freeman, 1994). Because conversion of NTG to NO requires enzymatic participation (Ignarro et al., 1981; Chung and Fung, 1990), we hypothesized that the ONOO formed during nitrate tolerance (due to increased vascular O₂ formation) may play a functional role in the phenomenon by inhibiting the enzyme(s) required for NO formation.

Using conditions similar to those previously used, we observed significant vascular tolerance to NTG after only 30-min incubation with 22 µM NTG. Associated with this tolerance development was extensive evidence of vascular protein nitration when compared with control tissues. These findings provide the first experimental evidence of vascular protein nitration during organic nitrate tolerance.

Although recent studies have suggested the existence of other potential biological pathways of TYR nitration, these studies have demonstrated that the chemistries responsible are dependent on neutrophil infiltration and activation (Sampson et al., 1998; Eiserich et al., 1998). Because tolerance induction occurred so rapidly and our histochemical analysis of the tissues demonstrated no immune cell infiltration into the vascular smooth muscle, the protein nitration in these studies were most likely derived from ONOO formation in the vasculature. Additionally, previous studies demonstrate that cardiovascular tissue homogenates require exogenous addition of neutrophil myeloperoxidase (5 µM), 1 mM concentrations of nitrite and hydrogen peroxide, and long incubation times (>1 h) to produce detectable protein nitration (Sampson et al., 1998). In contrast, NO and O₂ are known to form ONOO at a diffusion-limited rate, even at very low concentrations. Therefore, under our experimental conditions the observed protein nitration may be most simply explained by increased ONOO formation, rather than more complicated or less efficient chemical processes.

Studies by Munzel et al. (1995) have suggested that the vascular endothelial cell layer is a major production site for O₂ during nitrate tolerance. In our studies, protein nitration (suggestive of ONOO formation) was observed consistently throughout the entire smooth muscle layer after NTG incubation. Given the high instability and rapid reactivity of ONOO with intracellular proteins, the observed staining pattern suggests that ONOO was likely formed throughout the vascular wall during tolerance development. Identification of the primary cellular site(s) of O₂ formation during nitrate tolerance (endothelial layer versus vascular smooth muscle) may provide further insight into the biochemical mechanisms involved, as well as provide basic understanding of the intra-and intercellular reactivity of ONOO.

We evaluated the potential role of protein nitration in nitrate tolerance development by the addition of 500 µM TYR to the vascular incubation media. Free TYR was added as a competitive nitration site for ONOO at a concentration 5- to 10-fold in excess of physiologic intracellular levels (Thalhammer et al., 1982). TYR (500 µM) had no effect on NTG potency in control or tolerant vessel segments. However, TYR addition completely protected vascular smooth muscle from protein nitration, most likely by scavenging ONOO away from vascular smooth muscle sites. These results indicate that protein nitration apparently does not directly participate in NTG tolerance. Investigating protein nitration as the functional endpoint of ONOO formation does not account for alternative biochemistries by which ONOO may exert effects (e.g., sulfhydryl oxidation). However, we were able to completely reverse the extensive protein nitration associated with nitrate tolerance (via TYR addition), while leaving NTG potency in the vessel unchanged. By eradicating all protein nitration in the vasculature, it is likely that a majority of the ONOO formed during tolerance was scavenged by TYR addition, with no alteration in functional response to NTG. Therefore, it is unlikely that alternative ONOO-mediated biochemistries participate in tolerance development.

Skatchkov et al. (1997) recently demonstrated evidence of ONOO formation during nitrate tolerance in vivo. In a dog model of nitrate tolerance, these investigators demonstrated that urinary levels of 3NT were 4-fold higher in nitrate-tolerant dogs compared with control (Skatchkov et al., 1997). Our observations are consistent with these in vivo studies and consistent with the hypothesis that O₂ contributes to nitrate tolerance. Additionally, Dikalov et al. (1998) have demonstrated ONOO formation during NTG incubation, and have suggested that formation of reactive oxidative species such as ONOO is a key mediator of NTG tolerance development. Although NO-related biochemistry may be altered during tolerance development, our observed dissociation between extent of protein nitration and NTG potency suggests that vascular protein nitration does not mediate this phenomenon.