Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Nitric oxide, the radical metabolite formed from the metabolism of L-arginine by nitric oxide synthase (NOS²), has been shown to be involved in a variety of physiological processes, including neurotransmission, vasorelaxation, platelet aggregation, and penile erection as well as in a variety of pathological conditions including septic shock, reperfusion injury, arthritis, atherosclerosis, diabetes, and graft rejection (Burnett et al., 1992; Forstermann et al., 1994; Moncada et al., 1991; Schmidt and Walter, 1994). NOS was recently shown to be a hemoprotein of the cytochrome P450 type (McMillan et al., 1992; Stuehr and Ikeda-Saito, 1992; White and Marletta, 1992). In accord with this finding, suicide inactivators of liver P450 cytochromes, such as phencyclidine (Hoag et al.1984; Osawa and Coon, 1989; Osawa and Pohl, 1989) and CBrCl₃ (Osawa and Pohl, 1989), are suicide inactivators of NOS (Osawa and Davila, 1993; Osawa et al., 1994). Suicide or metabolism-based inactivators of liver microsomal P450 cytochromes have been important tools for probing the mechanism of catalysis, identifying active site residues, and for assessing the biological roles of P450 metabolites as well as for studying their turnover (Halpert et al., 1994). Suicide inactivators of NOS may provide similar insights.

In the course of such studies, we discovered that guanabenz (Wytensin, for structure see fig. 1), a clinically used antihypertensive agent, caused the inactivation of cytosolic penile NOS in vitro and in vivo. The current paper describes our findings and characterization of the kinetics of this inactivation process. The inactivation caused by guanabenz in vivo was comparable with that caused by N^G-nitro-L-arginine, a known inhibitor of NOS (Dwyer et al., 1991). Guanabenz may be a useful tool for future studies on the mechanisms of inactivation of NOS. In addition, we are not aware of any articles describing the inactivation of liver microsomal P450 cytochromes by guanabenz, in contrast to that for phencylidine or CBrCl₃ described above. Moreover, as NO plays a key role in a variety of physiological processes, the inactivation of NOS by xenobiotics, such as drugs, is of potential biomedical importance.

View larger version (6K): [in this window] [in a new window]   Fig. 1.   Structure of arginine, guanabenz, and NG-nitro-L-arginine.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. Guanabenz was purchased from Research Biochemicals International (Natick, MA). N^G-Nitro-L-arginine, L-arginine, D-arginine, and NADPH were purchased from Sigma. Male Wistar rats were purchased from Charles River Laboratories (Wilmington, MA). (6R)-5,6,7,8-tetrahydro-L-biopterin (tetrahydrobiopterin) was purchased from Dr. Schirk's Laboratory (Jona, Switzerland). L-[¹⁴C(U)]arginine (330.0 mCi/mmol) was purchased from Du Pont NEN.

Treatment of Animals and Sample Preparation. Guanabenz or N^G-nitro-L-arginine was dissolved in physiological saline and administered to male Wistar rats (150-250 g) at the indicated doses by intraperitoneal injection at 9:00 a.m. and 6:00 p.m.. Rats were sacrificed by decapitation 16 hr after the last injection. Whole deskinned penis was removed, washed with ice-cold physiological saline, cut into 1-2-mm pieces, and homogenized in 1 ml of ice-cold homogenization buffer [10 mM Hepes, pH 7.5, containing 320 mM sucrose, 100 µM EDTA, 1.5 mM dithiothreitol, 10 µg/ml trypsin inhibitor, 10 mg/ml leupeptin, 2 µg/ml aprotinin, 1 mg/ml phenylmethanesulfonyl fluoride, and 100 µM tetrahydrobiopterin (Knowles et al., 1990)] with the use of a metal tissue mincer (SDT Tissumizer, Tekmar, Cincinnati, OH). The homogenate was centrifuged at 245,000g for 10 min at 4°C. The supernatant fraction was collected and frozen in liquid nitrogen and stored at 80°C for later analysis. Protein concentrations of these samples were determined by the method of Bradford (Bio-Rad) with the use of bovine serum albumin as a standard. Rat brain cytosol was prepared with the use of the same buffers as for penile tissue as described previously (Knowles et al., 1990).

Enzyme Assays. The NOS activity of samples from the in vivo studies were determined by adding the supernatant fraction (0.6 mg) to an "assay mixture" containing 1 mM CaCl₂, 1 mM NADPH, 30 µM [¹⁴C]arginine (60 mCi/mmol), 100 µM tetrahydrobiopterin, 10 µg/ml calmodulin in a total volume of 200 µl of 40 mM potassium phosphate, pH 7.4. The assay mixture was incubated at 37°C for 10 min, and the amount of [¹⁴C]citrulline was determined as previously described (Osawa et al., 1994). The formation of [¹⁴C]citrulline was linear over the 10-min period. In control experiments, supernatant fractions (0.6 mg) from untreated and treated rats were mixed into the assay mixture to show that the inhibition of NOS was not due to changes in endogenous arginine levels or to the formation of an inhibitory metabolite. For experiments on the inactivation of NOS in vitro, the supernatant fraction from untreated rats was loaded onto a Sephadex G-25 M column (PD-10, Pharmacia Biotech Inc.) preequilibrated in 10 mM Hepes, pH 7.5, containing 320 mM sucrose, 100 µM EDTA, 1.5 mM dithiothreitol, 10 µg/ml trypsin inhibitor, 10 mg/ml leupeptin, 2 µg/ml aprotinin, and 1 mg/ml phenylmethanesulfonyl fluoride to remove endogenous arginine and excess tetrahydrobiopterin. An aliquot (1.2 mg) of the gel-filtered fraction was placed in a "first reaction mixture" containing 1 mM CaCl₂, 1 mM NADPH, 10 µg/ml calmodulin, and the desired concentration of guanabenz in a total volume of 1 ml of 40 mM potassium phosphate, pH 7.4. Aliquots (150 µl) were taken from the first reaction mixture and placed in the assay mixture, and the activity was determined as described above.

Western Blot. The penile supernatant fraction (15 µg of protein) was analyzed with the use of SDS-polyacrylamide gel electrophoresis (4-12% gradient gel). The gels were blotted onto a nitrocellulose membrane (Schleicher & Schuell), blocked with 0.2 mg/ml thimerosal in Blotto solution (Advanced Biotechnologies Inc., Columbia, MD), and probed (1:250) with a mouse monoclonal antibody against brain NOS (Transduction Laboratories, Lexington, KY). An anti-mouse IgG antibody (1:10,000) conjugated to peroxidase (Boehringer Mannheim) was used as a secondary antibody. An ECL reagent (Amersham) and X-OMAT film (Kodak, Rochester, NY) was used to detect the peroxidase conjugate as described by the manufacturer. The intensity of the bands was evaluated by a laser densitometer (Molecular Dynamics). Differing amounts of cytosol prepared from rat brains or insect cells overexpressing neuronal NOS were analyzed to ensure that the density was linearly dependent on the amount of NOS over the relevant concentration range.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Inactivation of NOS Activity by Guanabenz in Vitro. The presence of guanabenz in the first reaction mixture containing the supernatant fraction prepared from penile tissue caused the time-dependent inactivation of NOS activity (fig. 2A). This inactivation was dependent on the concentration of guanabenz, and the loss of activity seemed to follow pseudo first-order kinetics with an apparent dissociation constant for the initial reversible complex (K_i) and an inactivation constant (k_inact) of 38.5 µM and 0.179 min^-1, respectively (fig. 2B). A complete dependence on the presence of NADPH in the first reaction mixture was observed for the inactivation process (fig. 2C). The presence of L-arginine, the natural substrate for NOS, protected from this inactivation (fig. 2D). The D-isomer of arginine had no effect on the inactivation. The addition of excess tetrahydrobiopterin, calmodulin, FMN, and FAD in the assay mixture did not restore NOS activity after treatment with guanabenz. Moreover, gel filtration of the guanabenz-treated sample did not restore activity. The per cent residual specific activity was 27% before filtration and 31% after gel filtration. In experiments for which data are not presented, the addition of oxyhemoglobin, which would react with NO, had no effect on the inactivation.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Time-dependent inactivation of NOS activity in penile cytosolic fraction by guanabenz in vitro. The inactivation of NOS was determined with the use of a "first reaction mixture" and an "assay mixture" as described in Materials and Methods. (A) Extent of inhibition of NOS activity at the following concentrations of guanabenz in the first reaction mixture:  , untreated; , 5 µM; , 10 µM; , 20 µM; , 50 µM; , 100 µM. (B) Replot of the data from A. (C) Extent of inhibition of NOS with guanabenz (30 µM) in the absence of NADPH () or in the presence of NADPH () in the first reaction mixture. A control without guanabenz and NADPH is also shown (). (D) Extent of inhibition of NOS with guanabenz (30 µM, ) and the effect of L-arginine (30 µM,) or D-arginine (30 µM, ) in the first reaction mixture.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Gel filtration of guanabenz-treated brain cytosolic fraction. The brain cytosolic fraction (7.2 mg/ml) was placed in a reaction mixture containing guanabenz (100 µM) and NADPH (0.1 mM) in 0.6 ml of 40 mM potassium phosphate, pH 7.4, similar to that described previously (Osawa and Davila, 1993). The mixture was incubated at room temperature for 1 hr. The mixture was then placed onto a Sephadex G-25 column as described in Materials and Methods. The NOS activity was determined by the oxyhemoglobin method as previously described (Osawa and Davila, 1993). The values are the mean and standard deviation (N = 3).

Effect of Guanabenz on NOS in Vivo. Guanabenz (5.0 mg/kg/day) caused a decrease of approximately one-half in the rat penile NOS activity after 4 days of treatment similar to that found for N^G-nitro-L-arginine (5 mg/kg/day), a known inhibitor of NOS (fig. 4A). One day of treatment with guanabenz had no effect on NOS activity (fig. 4A) even though this was of sufficient duration for the effects of N^G-nitro-L-arginine (data not shown). In comparison, guanabenz caused a decrease of approximately one-quarter in the NOS activity present in brain cytosol of the treated rats (fig. 4B). The inhibition of brain NOS by N^G-nitro-L-arginine was approximately 50% (fig. 4B), consistent with that reported previously (Dwyer et al., 1991), as well as closely matching the inhibition found in penile samples.

View larger version (55K): [in this window] [in a new window]   Fig. 4.   Time-dependent inhibition of NOS activity in penile (A) or brain (B) cytosolic fractions by guanabenz in vivo. Rats were treated with guanabenz or NG-nitro-L-arginine at a dose of 5 mg/kg/day for 1 or 4 days and the penile or brain NOS activities measured as described in Materials and Methods. The values are the mean and standard deviation (N = 3).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Dose-dependent inhibition of penile NOS activity by guanabenz in vivo. The rats were treated for 4 days at the indicated doses of guanabenz as described in Materials and Methods. The values are the mean and standard deviation (N = 3).

View larger version (48K): [in this window] [in a new window]   Fig. 6.   Levels of immunodetectable NOS in penile cytosolic fraction from rats treated with guanabenz or NG-nitro-L-arginine. Penis samples from fig. 4 were analyzed. Immunoblotting was as described in Materials and Methods. The density of the signal on the X-ray film was quantified by laser densitometry with the use of a Molecular Dynamics scanner. The values represent the mean and SD of three samples.

View larger version (49K): [in this window] [in a new window]   Fig. 7.   Levels of immunodetectable NOS in penile cytosolic fraction after treatment with guanabenz in vitro. Penile cytosolic fraction was treated with guanabenz (100 µM) as described in fig. 2. Immunoblotting was as described in Materials and Methods. The density of the signal on the X-ray film was quantified by laser densitometry with the use of a Molecular Dynamics scanner. The values represent the mean and SD of three samples.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Guanabenz was shown to inactivate NOS activity in the cytosolic fraction of brain and penile tissue by a process that required NADPH, a necessary cofactor for enzyme catalysis. This suggested a metabolic component to the inactivation process. The inactivation seemed to be an active site-directed event as the natural substrate of NOS stereospecifically protected the enzyme from the inactivation. The inactivation was dependent on time as well as the concentration of guanabenz, with an apparent K_i and k_inact for the penile NOS activity of 38.5 µM and 0.179 min^-1, respectively. The inactivation was not due to formation of NO, which is known to inhibit the activity of NOS in vitro (Griscavage et al., 1994), as oxyhemoglobin had no effect on the inactivation (data not shown). Thus, the kinetics are consistent with guanabenz being a metabolism-based inactivator of cytosolic NOS activity, which can be attributed to the neuronal isoform (Bredt et al.; 1991; Burnett et al., 1992; Magee et al., 1996). Guanabenz contains a guanidine moiety as do other known metabolism-based inactivators of NOS, including several derivatives of arginine. For example, the N^G-methyl, N^G-allyl, and N^G-amino analogs of arginine inactivate NOS isolated from stimulated macrophages with apparent K_i and k_inact values of 4.2 µM and 0.05 min^-1, 3.4 µM and 0.026 min^-1, and 3 µM and 0.26 min^-1, respectively (Olken et al., 1991; Olken and Marletta, 1992; Wolff and Lubeskie, 1996). The N^G-methyl-L-arginine is also a metabolism-based inactivator of the neuronal NOS with apparent K_i and k_inact values of 2 µM and 0.022 min^-1, respectively (Feldman et al., 1993; Reif and McCreedy, 1995). Other guanidine-containing compounds, such as aminoguanidine (Wolff and Lubeskie, 1995) and diaminoguanidine (Wolff and Lubeskie, 1996), have been shown to be metabolism-based inactivators of neuronal, macrophage, and endothelial isoforms of NOS. However, methylguanidine and 1,1'-dimethylguanidine are not metabolism-based inactivators (Wolff and Lubeskie, 1996). Based on this evidence, it was suggested that the hydrazine moiety was critical for inactivation. The hydrazone moiety on guanabenz may also suffice in conferring the ability to be a metabolism-based inactivator. The exact mechanism of inactivation is currently not known for any metabolism-based inactivator of NOS. Studies on the suicide inactivation of liver microsomal P450 cytochromes (Correia et al., 1981; Halpert et al., 1994; Osawa and Pohl, 1989) and recent studies with NOS (Gerber and Ortiz de Montellano, 1995; Olken et al., 1994) suggest that covalent alteration of the heme prosthetic group and/or the protein are likely mechanisms.

Unlike the other guanidine-containing compounds that were studied, guanabenz is already clinically used. Although the inactivation of NOS is not the pharmacological basis of the antihypertensive action of guanabenz, the drug is known to cause impotence as a side effect (Brock and Lue, 1993; Weiss, 1991). This may be of relevance as NO has been shown to play a key role in penile erection (Burnett et al., 1992; Bush et al., 1992; Rajfer et al., 1992) and that inhibitors to NOS can cause impotence (Burnett et al., 1992). In the current study, guanabenz was found to inactivate penile NOS activity in vivo. The NOS activity of the penile cytosolic fraction from rats treated with guanabenz (5 mg/kg/day) for 4 days was approximately one-half of that of the control animals, a level that is comparable with that seen with N^G-nitro-L-arginine (5 mg/kg/day) (Dwyer et al., 1991). The nitro compound was chosen as a positive control as previous studies indicate irreversible inactivation of rat brain activity in vivo (Dwyer et al., 1991), although more recent in vitro evidence indicates that the inhibition is slowly reversible (Klatt et al., 1994). The lack of complete inhibition of NOS by N^G-nitro-L-arginine at this dose is consistent with that found for the rat brain activity (Dwyer et al., 1991; Iadecola et al., 1994). The inhibition of penile NOS activity by guanabenz after a 4-day treatment was dose dependent with comparable inhibition at 0.5 mg/kg/day. This dose is near the recommended pharmacological dose (0.06 to 0.45 mg/kg/day) for use as an antihypertensive agent in man. Because the inhibition of NOS is time dependent, it is possible that treatment of rats for a period greater than 4 days may decrease the dose needed for observing the inhibition of NOS and may more closely reflect the conditions for use in man. Further work on the physiological effects of guanabenz on penile erection are necessary to determine the significance of these effects. In addition, effects on other isoforms of NOS are warranted, as studies on mice with targeted deletion of the neuronal isoform indicate that the endothelial form can play a prominent role in penile erection (Burnett et al., 1996). It is noteworthy that higher levels of the endothelial isoform were found in the knockout mice in comparison with that of the wild type, perhaps indicating a compensatory mechanism (Burnett et al., 1996). Recently, it has been shown that penile tissue contains an alternatively spliced variant of the brain NOS (Magee et al., 1996), but the impact of these differences on the activity or susceptibility to inactivators of the NOS enzyme has also not been determined. Nonetheless, the current studies clearly indicate that inhibition of NOS by guanabenz can occur in vivo.

The activity loss in vivo was concomitant with a decrease of approximately 50% in the level of immunodetectable NOS protein in the cytosolic fraction prepared from the penis, as visualized by Western blot with the use of a monoclonal antibody that recognizes the amino-terminal portion of neuronal NOS, which is conserved in the penile form. There were no detectable losses in the amount of immunodetectable NOS protein after inhibition with N^G-nitro-L-arginine. Thus, the inhibition per se seems not to be the cause for the decrease in steady state levels, in analogy to that found for the liver P450 cytochromes (Bornheim et al., 1987; Correia et al., 1987; Correia et al., 1992; Lunetta et al., 1989; Tierney et al., 1992). We are interested in determining if the metabolism-based inactivation of NOS involves covalent alteration of the protein and subsequent affects on turnover, as has been indicated by studies with liver microsomal P450 cytochromes (Correia et al., 1987; Noguchi et al., 1982; Tierney et al., 1992). Moreover, it would be of interest to determine if long-term treatment with guanabenz leads to a greater loss of NOS or if compensatory mechanisms give rise to induction of the protein.

We report here that drug-induced inactivation of NOS may be a mechanism for toxicological side effects of drugs. Because NO plays a key role in a multitude of important physiological processes, including sexual function, such interactions may explain a variety of toxic effects associated with many drugs. Moreover, the biochemical mechanisms by which this occurs may lead to insight on the regulation of NOS by chemicals, including drugs and other xenobiotics.