The Role of Tetrahydrobiopterin in the Regulation of Neuronal Nitric-oxide Synthase-generated Superoxide*

Tetrahydrobiopterin (H4B) is a critical element in the nitric-oxide synthase (NOS) metabolism ofl-arginine to l-citrulline and NO⋅. It has been hypothesized that in the absence of or under nonsaturating levels of l-arginine where O2 reduction is the primary outcome of NOS activation, H4B promotes the generation of H2O2 at the expense of O 2 ⨪ . The experiments were designed to test this hypothesis. To test this theory, two different enzyme preparations, H4B-bound NOS I and H4B-free NOS I, were used. Initial rates of NADPH turnover and O2 utilization were found to be considerably greater in the H4B-bound NOS I preparation than in the H4B-free NOS I preparation. In contrast, the initial generation of O 2 ⨪ from the H4B-free NOS I preparation was found to be substantially greater than that measured using the H4B-bound NOS I preparation. Finally, by spin trapping nearly all of the NOS I produced O 2 ⨪ , we found that the initial rate of H2O2 production by H4B-bound NOS I was considerably greater than that for H4B-free NOS I.

In 1992, we discovered that NOS I generates O 2 . in the absence of L-arginine (13). More recently, NOS II and NOS III, like NOS I, have been found to generate O 2 . during enzymic cycling (14 -16). In the presence of L-arginine, NOS I generates NO ⅐ and O 2 . ; the ratio of these free radicals is dependent upon the concentration of L-arginine (17,18). Thus, L-arginine is one of the controlling factors that dictate the selectivity of free radicals produced by NOS. However, in the absence of substrate, NOS uses O 2 as the terminal electron acceptor, generating O 2 . and H 2 O 2 by sequential one-electron reductive steps (see Fig. 1). Under these conditions, there is undoubtedly an alternative mechanism by which NOS regulates the formation of each of these cell-signaling products of O 2 reduction. One possibility is that H 4 B controls production of O 2 . by increasing the reduction rate of the NOS-Fe 2ϩ O 2 species (Refs. 19 -22 and Fig. 1). Evidence to support this theory comes from experiments where the addition of H 4 B to purified NOS I diminished the spin trapping of O 2 . (17,18,23). However, these findings must be viewed with caution, because H 4 B in aqueous solution has been reported to scavenge O 2 . (24 -26) with a rate constant of 3.9 ϫ 10 5 M Ϫ1 s Ϫ1 (26). Thus, the conclusion drawn from the earlier studies (17,18,23)
Purification of NOS I-NOS I was expressed and purified essentially as described by Roman et al. (30), with the modification that the culture * This work was supported in part by National Institutes of Health Grants RR-12257 (to G. M. R.), T32-ES07263 (to P. T.), R25-GM-55036 (to J. W.), NS-34152 (to G. F.), and GM-52419 and Robert A. Welch Foundation Grant AQ1192. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
NADPH Consumption-Oxidation of NADPH was performed in a reaction using potassium phosphate buffer (50 mM, pH 7.4, 1 mM DTPA, 1 mM EGTA), CaCl 2 (2 mM), calmodulin (100 units/ml), and NADPH (150 M) at room temperature. The reaction was initiated by the addition of NOS I (0.40 M). A UV-visible spectrophotometer (Uvikon, model 940, Research Instruments International, San Diego, CA) was used to monitor the reaction spectrophotometrically at 340 nm. The initial rate of NADPH oxidation was estimated using an extinction coefficient of Oxygen Consumption-Oxygen consumption was measured with a commercial oxygen monitoring system (Hansatech). The system was composed of a membrane-coated Clark-type electrode fitted in a glass body reaction chamber and equipped with a Teflon-coated stirring bar and an air-tight stopper. Data acquisition was performed with proprietary hardware and software (Hansatech  (32). Ferricytochrome c (0 -23 M) was used as a competitive inhibitor (33). The reaction mixtures were immediately transferred to an EPR flat quartz cell and introduced into the cavity of the EPR spectrometer (model E-109; Varian Medical Systems, Inc.). EPR spectra were recorded at room temperature 3 min after the reaction was initiated by the addition of xanthine oxidase. Instrument settings were: microwave power, 20 mW; modulation frequency, 100 kHz; modulation amplitude, 0.5 G; sweep time, 12.5 G/min; and response time, 0.5 s.

Estimation of the Half-life of BMPO-OOH-
The half-life of BMPO-OOH was determined by monitoring the decrease in the first line of the EPR spectrum of BMPO-OOH as a function of time. The reaction mixture contained BMPO (50 mM) and hypoxanthine (400 M) in potassium phosphate buffer (chelexed, 50 mM, pH 7.4, 1 mM DTPA) for 10 min, and then SOD (30 units/ml, as defined in Ref. 34) was added. The reaction mixture was immediately transferred to an EPR flat quartz cell and introduced into the cavity of the EPR spectrometer (model E-109; Varian Medical Systems, Inc.). EPR spectra were recorded at various time intervals for 60 min.
Rate of Hydrogen Peroxide Formation-Estimation of H 2 O 2 production was obtained by fluorometric analyses (fluorometer, Hitachi model F2500, High Technologies America, Inc., San Jose, CA). A modified method utilizing the dye Amplex Red was adopted (35)(36)(37). The incubation medium was supplemented with Amplex Red (1 M) and horseradish peroxidase (5 units/ml) in sodium phosphate buffer (50 mM, 1 mM EGTA, pH 7.4). The reaction mixture contained NADPH (160 M), CaCl 2 (0.5 mM), calmodulin (100 units/ml), BMPO (100 mM), and SOD (0.04 -80 units/ml). SOD (0.04 unit/ml) was added to each reaction to suppress initial fluorescence seen from the inclusion of NADPH, and SOD (0.14 -80 units/ml) was used in control experiments described under "Results and Discussion." The reaction was initiated by the addition of purified H 4 B-free NOS I (4 nM) or H 4 B-bound NOS I (4 nM) into the reaction mixture. The initial rate of H 2 O 2 generation was recorded as an increase in fluorescence of the dye at 585 nm with the excitation set at 550 nm. The fluorescence was calibrated by generating a standard curve with known concentrations of H 2 O 2 . The concentration of the commercial 30% H 2 O 2 solution was calculated from light absorbance at 240 nm employing an extinction coefficient of 0.0436 mM Ϫ1 cm Ϫ1 ; the stock solution was diluted to 50 M with water and used for calibration immediately. The specificity of horseradish peroxidase/ Amplex Red toward H 2 O 2 was confirmed, because tert-butyl hydroperoxide was not found to be a substrate.
NOS I Activity by [ 14 C]L-Citrulline Formation Assay-The enzymatic activity of purified NOS I was determined by its ability to catalyze the formation of L-citrulline from L-arginine as previously reported (17) (42). Second, the half-life of BMPO-OOH was considerably greater than for that of DMPO-OOH at 53 s (43) and in the same range as that for DEPMPO-OOH at 18 min (44). Third, the EPR spectrum of BMPO-OOH exhibited a greater signal-to-noise ratio than that found for DEPMPO-OOH. The small signal-to-noise ratio of DEPMPO-OOH was the result of additional hyperfine splitting associated with the phosphorous atom located at the ␣-carbon on the pyrroline ring.
Next spectral peak height as a function of time (see inset in Fig. 2). Fig. 2 (Fig. 1). Alternatively, this peroxide can arise from the dismutation of O 2 . , in which the rate constant at pH 7.4 is 3.0 ϫ 10 5 M Ϫ1 s Ϫ1 (45). It is therefore by no means a trivial task to separate these disparate pathways. After considering several options, we settled on an approach that required increasing the concentration of BMPO to a level so that this nitrone would spin trap most, if not all, of the O 2 . produced (41). Thereupon, the only source of NOS-derived H 2 O 2 would be from the one-electron reduction of the NOS-Fe 2ϩ O 2 species. We then had to find a method that would meet the following criteria. First, the assay had to detect H 2 O 2 in real time, not at some arbitrary time after the reaction had commenced. Second, the method must not interfere with the spin trapping of O 2 . .
Third, given that NOS can easily transfer electrons to a wide variety of one-electron acceptors, such as ferricytochrome c (13), the assay had to be an oxidative process. Fourth, the method had to be sensitive and selective for H 2 O 2 . Given these limitations, we settled on a fluorometric assay developed by Zhou et al. (36,37). The overall mechanism, shown below, involves three distinct reactions, as presented by Chance (46 Although the rate constant for the reaction of H 2 O 2 with horseradish peroxidase to form Compound I is 10 ϫ 10 6 M Ϫ1 s Ϫ1 (46), the rate constants for the sequential one-electron reduction of Compounds I and II to the fluorescent resorufin from Amplex Red are unknown. However, based on similar reactions reported in the literature (46) involving Compound I and II with other donor molecules, we estimate that the rate constant would not be above 1 ϫ 10 5 M Ϫ1 s Ϫ1 , and most likely it is close to 1 ϫ 10 3 M Ϫ1 s Ϫ1 , too slow to allow quantitative estimates of the initial rate of H 2 O 2 production. Therefore, the initial rate of NADPH consumption measured does not have a ml), and SOD (ranging from 0.14 to 80 units/ml). We found that the rate of H 2 O 2 production, as measured by an increase in fluorescence, was constant over this range of SOD used in the experiment (data not shown). Thus, where appropriate SOD (0.14 units/ml) was included in each reaction. When BMPO (100 mM) was added to the above reaction mixture, without and with SOD (0.14 units/ml), we found that BMPO spin-trapped ϳ90% of the O 2 . generated by NOS.
Based on these control experiments, we were confident that by inclusion of BMPO in the reaction mixture, most of the NOS-produced O 2 . released from NOS was considerably less when H 4 B was bound to the enzyme than in the absence of this pterin (Fig. 2). After several min, however, the EPR spectral peak height of BMPO-OOH (Scheme 1) from H 4 B-bound NOS I out-paced that observed with H 4 B-free NOS I (Fig. 2 Although it may be premature to speculate as to the physiologic significance of our findings, we offer one possible scenario. In the absence of or under low levels of L-arginine, where O 2 reduction is the primary end product of NOS activation, H 4 B will undoubtedly play a critical role in regulating the generation of H 2 O 2 and O 2 . . Because each of these reduction products of O 2 activates a different cell signal pathway (47,48), the importance of H 4 B in the regulation of NOS-derived O 2 . and H 2 O 2 and its diversity of physiological functions cannot be underestimated.