Introduction

Top Summary Introduction Procedures Results Discussion References

Nitric oxide is a widespread messenger molecule regulating biological processes as diverse as blood vessel relaxation, neuronal cell-to-cell communication and immune function (Mayer and Hemmens, 1997). The major physiological target of NO is sGC (GTP pyrophosphate-lyase (cyclizing), EC 4.6.1.2), which catalyzes the formation of cGMP from GTP. The enzyme contains a regulatory heme group that binds NO at diffusion-controlled rates and confers the pronounced NO stimulation of the enzyme (Ignarro, 1991; Wedel et al., 1994).

NO is synthesized by three NOS isoforms, which all catalyze an NADPH- and O₂-dependent oxidation of L-arginine to form L-citrulline and NO (Hemmens and Mayer, 1997; Stuehr, 1997). Two NOS isoforms constitutively expressed in cells such as neurons (nNOS) and endothelium (eNOS) are activated by Ca²⁺-dependent calmodulin binding, whereas expression of a third, Ca²⁺-independent isoform is induced by cytokines (iNOS). All three NOS isoforms are homodimeric proteins whose subunits are comprised of an amino-terminal oxygenase domain that binds heme, L-arginine, and H₄biopterin and a carboxyl-terminal reductase domain that binds calmodulin, FMN, FAD, and NADPH. The reductase domain shuttles electrons from NADPH via the flavins to the oxygenase domain, which is the site of heme iron reduction, O₂ activation and NO synthesis. At low concentrations of L-arginine or in its absence, the enzymatic reduction of O₂ uncouples from substrate oxidation leading to the production of O₂ and H₂O₂ instead of NO (Heinzel et al., 1992; Pou et al., 1992). In addition to the amino acid substrate, the pteridine cofactor H₄biopterin is also required for the tight coupling of O₂ reduction to L-arginine oxidation (Mayer and Werner, 1995).

Recently manganese complexes of substituted mesoporphyrins have been described as a new class of SOD mimetics which are cell permeable and stable in the presence of metal ion chelators. These Mn-porphyrins were shown to protect Escherichia coli against paraquat-induced oxidative stress (Liochev and Fridovich, 1995) and to facilitate the growth of SOD-deficient E. coli strains (Faulkner et al., 1994; Batinic-Haberle et al., 1997). Substituted metalloporphyrins also potently inhibit peroxynitrite-induced oxidation of dihydrorhodamine-123 (Zingarelli et al., 1997). It has been demonstrated that peroxynitrite reacts rapidly with MnTMPyP in a 1:1 stoichiometry, thereby generating an oxomanganese intermediate, which catalyzes plasmid DNA strand breaking under physiological conditions (Groves and Marla, 1995). By the use of Mn-porphyrins, the involvement of peroxynitrite has been demonstrated in the depression of cellular respiration (Szabo et al., 1996) and in the oxidation of mitochondrial as well as nuclear proteins in immunostimulated macrophages (Szabo et al., 1997). In addition, Mn-porphyrins were reported to ameliorate vascular contractile and cellular energetic failure in endotoxin-treated rats (Zingarelli et al., 1997).

The present study was designed to investigate the interference of MnTMPyP with the NO/cGMP signaling pathway. Our results confirm that the Mn-porphyrin acts as an efficient scavenger of peroxynitrite but provide no evidence for a SOD-mimetic effect of this compound in cells. Instead, MnTMPyP was found to be a potent NO scavenger in the presence of GSH and to directly inhibit both NOS and sGC.

	Experimental Procedures

Top Summary Introduction Procedures Results Discussion References

Materials. MnTMPyP was purchased from Alexis (Läufelfingen, Switzerland). Stock solutions (0.1 M) were prepared with Nano-pure water (Barnstead ultrafiltered type I, resistance > 18 M cm¹) and kept at 20° before use. DEA/NO was from NCI Chemical Carcinogen Repository (Kansas City, MO). Tenfold concentrated stock solutions of the NO donor were prepared daily in 10 mM NaOH. L-[³H]Arginine was from Amersham, supplied by MedPro (Vienna, Austria). NO solutions were prepared by dissolving NO gas (99% pure; Linde München, Germany) in deoxygenated water as described previously (Kukovetz and Holzmann, 1989). All other chemicals were obtained from Sigma (Vienna, Austria).

Culture of endothelial cells and determination of endothelial cGMP and L-citrulline formation. Porcine aortic endothelial cells were cultured as previously described (Mayer et al., 1993). Before experiments, the cells were subcultured in 24-well plastic plates and grown to confluence (~2 × 10⁵ cells/well). The culture medium was removed, and the cells were washed once and equilibrated in buffer A (isotonic phosphate buffer, pH 7.4, containing 8 mM Na₂HPO₄, 1.5 mM KH₂PO₄, 137 mM NaCl, 2.7 mM KCl, 0.9 mM CaCl₂, 1 mM MgCl₂, 1 mM 3-isobutyl-1-methylxanthine, and 1 µM indomethacin) in the absence or presence of MnTMPyP at the concentrations indicated. After 15 min, Ca²⁺ ionophore A23187 or DEA/NO was added to give initial final concentrations of 0.3 and 1 µM, respectively. Reactions were terminated 4 min later by removal of buffer and addition of 1 ml of 0.01 N HCl. After incubating for 1 hr, intracellular cGMP was measured in the supernatants of the lysed cells by radioimmunoassay.

Determination of NOS activity. Rat nNOS and bovine eNOS were purified from recombinant baculovirus-infected Sf9 cells as described previously (Harteneck et al., 1994; List et al., 1996). Formation of L-[³H]citrulline from L-[³H]arginine was determined by incubation of 0.2-0.3 µg of enzyme at 37° for 10 min in a 50 mM triethanolamine/HCl buffer (pH 7.0), containing 0.1 mM L-[³H]arginine (~50,000 cpm), 0.2 mM NADPH, 5 µM flavin adenine dinucleotide, 10 µM H₄biopterin, 0.5 mM CaCl₂, and 10 µg/ml calmodulin, followed by isolation of L-[³H]citrulline by cation exchange chromatography (Mayer et al., 1994). Uncoupled reductive activation of oxygen was determined as Ca²⁺/calmodulin-dependent oxidation of NADPH in the absence of L-arginine and H₄biopterin by continuously monitoring the decrease in absorbance at 340 nm (Mayer et al., 1991). The cytochrome P450 reductase activity of NOS was assayed as reduction of cytochrome c (0.2 mM) by continuously monitoring the increase in absorbance at 550 nm against calmodulin-deficient blanks (Klatt et al., 1992). MnTMPyP was added in 10-fold concentrated stock solutions.

Determination of sGC activity. For direct activation with the NO donor DEA/NO, purified sGC (50 ng; V_max ~6-8 µmol/mg/min) was incubated at 37° for 10 min in a total volume of 0.1 ml of a 50 mM K₂HPO₄/KH₂PO₄ buffer, pH 7.4, containing 0.5 mM [-³²P]GTP (200,000-300,000 cpm), 3 mM MgCl₂, 1 mM cGMP, and 0.05 mg/ml bovine serum albumin. Reactions were started by adding DEA/NO (1 µM final concentration) and 10-fold concentrated stock solutions of MnTMPyP or vehicle to the assay mixtures and transferring the samples from 4 to 37°. For reconstitution with nNOS, purified sGC (50 ng) was incubated at 37° for 10 min in a total volume of 0.1 ml of a 50 mM K₂HPO₄/KH₂PO₄ buffer, pH 7.4, containing 0.5 mM [-³²P]GTP (200,000-300,000 cpm), 3 mM MgCl₂, 1 mM cGMP, 0.05 mg/ml bovine serum albumin, 0.1 mM L-arginine, 10 µM CaCl₂, 10 µg/ml calmodulin, 50 µM NADPH, 40 µM CHAPS, and 1 mM GSH. Reactions were started by adding nNOS (2 µg/ml; specific activity 0.06-0.08 µmol/mg/min under these conditions; V_max ~0.8 µmol/mg/min) and 10-fold concentrated stock solutions of MnTMPyP or vehicle to the assay mixtures, followed by transferring the samples from 4° to 37°. Reactions were terminated by ZnCO₃ precipitation, followed by isolation of [-³²P]cGMP as described (Schultz and Böhme, 1984). Results were corrected for enzyme-deficient blanks and recovery of cGMP.

Electrochemical detection of NO. NO was measured with a Clark-type electrode (Iso-NO, World Precision Instruments, Berlin, Germany), which was connected to an Apple Macintosh computer via an analog to digital converter (MacLab, World Precision Instruments). The output current was recorded at 0.6 Hz under constant stirring at 37°. Two-point calibration of the electrode was performed daily according to the procedure recommended by the manufacturer. Solutions to be tested were applied by injection into 1.8-ml glass vials completely filled with 50 mM KH₂PO₄/K₂HPO₄ buffer, pH 7.4, containing additives as indicated and sealed with a septum.

Light absorbance spectroscopy of MnTMPyP. Light absorbance spectra were recorded at ambient temperature with a Hewlett-Packard 8452A diode array spectrophotometer in 50 mM KH₂PO₄/K₂HPO₄ buffer, pH 7.4, containing additives as indicated. Tenfold concentrated stock solutions of MnTMPyP were added to give the final concentrations indicated in the text and figures. The Mn(II)-NO-TMPyP complex (Soret band at 434 nm) was prepared by addition of sodium dithionite to convert Mn(III)TMPyP (Soret band at 462 nm) to Mn(II)TMPyP (Soret band at 450 nm) (Faulkner et al., 1994) and addition of 10 µl of a 2 mM aqueous NO solution.

Data evaluation. Unless indicated otherwise, data are mean values ± standard errors of three experiments performed in duplicate. Parameters of the concentration-response curves were calculated according to the Hill equation.

	Results

Top Summary Introduction Procedures Results Discussion References

To test for a possible SOD-mimetic effect of MnTMPyP, we studied the effect of this compound on endothelial cGMP accumulation induced by the Ca²⁺ ionophore A23187. Basal cGMP levels of unstimulated cells were 2.0 ± 0.3 pmol/10⁶ cells. Increasing concentrations of A23187 led to a pronounced stimulation of cGMP formation (up to 19.2 ± 1.1 pmol/10⁶ cells) with an EC₅₀ of 51.0 ± 13.2 nM (Fig. 1A, filled symbols). This effect of A23187 was completely inhibited by the NOS inhibitor N^G-nitro-L-arginine (10 µM; not shown). MnTMPyP (10 µM) did not potentiate the effect of the Ca²⁺ ionophore (EC₅₀ = 69.0 ± 9.0 nM) but significantly decreased its maximal effect (15.1 ± 2.1 pmol of cGMP/10⁶ cells) (open symbols). Similar data were obtained when the NO donor DEA/NO was used to directly activate endothelial sGC (data not shown). Fig. 1B (filled symbols) shows that MnTMPyP inhibited A23187-induced cGMP accumulation in a concentration-dependent manner with an IC₅₀ of 75.0 ± 10.4 µM. The effect of the porphyrin was virtually complete at 0.3 mM (2.52 ± 0.2 pmol cGMP/10⁶ cells). This pronounced inhibition was not explained by inhibition of NOS, as evident by the much lower sensitivity to MnTMPyP of endothelial arginine-to-citrulline conversion (~30% inhibition at 0.3 mM; Fig. 1B, open symbols).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Effect of MnTMPyP on A23187-induced cGMP accumulation and L-citrulline formation in endothelial cells. A, Endothelial cells were preincubated in the absence (filled symbols) or presence (open symbols) of MnTMPyP (10 µM) for 15 min, followed by incubation with A23187 at the indicated concentrations for 4 min and determination of cellular cGMP levels by radioimmunoassay. Data are mean values ± standard error of three experiments performed in duplicate. B, Endothelial cells were preincubated in the presence of increasing concentrations of MnTMPyP for 15 min, followed by incubation with 0.3 µM A23187 for 10 min and determination of cellular cGMP levels by radioimmunoassay (filled symbols) or conversion of L-[3H]arginine to L-[3H]citrulline (open symbols; see Experimental Procedures for details). Results are expressed as percent inhibition (100% = formation of cGMP or conversion of incorporated L-[3H]arginine to L-[3H]citrulline in the absence of MnTMPyP). Data are mean values ± standard error of three experiments performed in duplicate.

Although NOS inhibition was not apparent in the experiments with intact cells, MnTMPyP was a fairly potent inhibitor of purified recombinant nNOS. As shown in Fig. 2A, MnTMPyP inhibited formation of L-citrulline in a concentration-dependent manner with an IC₅₀ of 5.5 ± 0.8 µM. The porphyrin inhibited L-citrulline formation of the inducible isoform with a similar potency (IC₅₀ = 9.0 ± 1.4 µM), but was a slightly less potent inhibitor of the endothelial enzyme (IC₅₀ = 23 ± 1.1 µM; data not shown). In the light of our previous results with SOD mimetic copper complexes (Mayer et al., 1996), it was likely that MnTMPyP behaved similarly and interfered as a parasitic electron acceptor with the cytochrome P₄₅₀ reductase activity of NOS, which can be assayed as Ca²⁺/calmodulin-dependent reduction of cytochrome c (Klatt et al., 1992). The cytochrome c reductase activity of purified nNOS was 10.0 ± 0.3 and 5.9 ± 0.4 µmol/mg/min in the absence and presence of 50 µM MnTMPyP, respectively. Further, MnTMPyP led to a pronounced, approximately 4-fold stimulation of NOS-catalyzed NADPH oxidation (measured in the absence of L-arginine). As shown in Fig. 2B, MnTMPyP increased the NADPH oxidase activity of the enzyme in a concentration-dependent manner from 0.64 ± 0.08 µmol/mg/min to 2.7 ± 0.4 µM; the EC₅₀ was 2.69 ± 0.2 µM, maximal effects were obtained with 0.1 mM of the porphyrin. Similar data were obtained with purified recombinant bovine eNOS and murine macrophage iNOS (data not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effect of MnTMPyP on L-citrulline formation (A) and NADPH (B) oxidation by purified nNOS. A, Purified rat nNOS (0.2-0.3 µg) was incubated for 10 min at 37° in 0.1 ml of a 50 mM triethanolamine/HCl buffer (pH 7.0) containing 0.1 mM L-[3H]arginine (~50,000 cpm), 0.2 mM NADPH, 5 µM flavin adenine dinucleotide, 10 µM H4biopterin, 0.5 mM CaCl2, and 10 µg/ml calmodulin in the presence of increasing concentrations of MnTMPyP, followed by isolation of L-[3H]citrulline by cation exchange chromatography. Data are mean values ± standard error of three experiments performed in duplicate. B, Purified rat nNOS (1 µg) was incubated at 37° in 0.2 ml of a 50 mM triethanolamine/HCl buffer (pH 7.0) containing 0.2 mM NADPH, 0.5 mM CaCl2, and 10 µg/ml calmodulin in the presence of increasing concentrations of MnTMPyP. Oxidation of NADPH was measured by continuously monitoring the decrease in absorbance at 340 nm against calmodulin-deficient blanks. Enzyme activity was calculated using an extinction coefficient of 6.34/mM/cm. Data are mean values ± standard error of three experiments.

The cell culture experiments indicated that effects unrelated to NOS inhibition may account for the potent interference of MnTMPyP with endothelial cGMP accumulation. Therefore, we tested the porphyrin for inhibition of purified sGC and scavenging of NO. Incubation of sGC with the NO donor DEA/NO (1 µM) or Ca²⁺/calmodulin-activated NOS (0.2 µg/0.1 ml) led to a pronounced stimulation of cGMP formation (from 0.043 ± 0.009 to 7.89 ± 0.35 and 3.44 ± 0.01 µmol/mg/min, respectively). It should be pointed out that sGC stimulation by donors of pure NO, e.g., DEA/NO, is GSH-independent (Mayer et al., 1995a), whereas stimulation of the enzyme by NOS or other NO/O₂-generating systems requires the presence of a thiol (Mayer et al., 1998). Therefore, the experiments shown in Fig. 3 were performed either in the presence (NOS) or in the absence (DEA/NO) of GSH. MnTMPyP inhibited the formation of cGMP in a concentration-dependent manner with IC₅₀ values of 0.8 ± 0.09 and 0.6 ± 0.2 µM when sGC was activated with NOS and DEA/NO, respectively. Full inhibition was observed with porphyrin concentrations as low as 10-100 µM. Identical results were obtained with DEA/NO-activated GC in the presence of GSH (data not shown). These results showed that MnTMPyP potently inhibited NO stimulation of sGC in a thiol-independent manner.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Effect of MnTMPyP on sGC activity. Purified sGC (50 ng) was incubated with Ca2+/calmodulin-activated nNOS (filled symbols) at 37° for 10 min in a total volume of 0.1 ml of a 50 mM K2HPO4/KH2PO4 buffer, pH 7.4, containing 0.5 mM [-32P]GTP, 3 mM MgCl2, 1 mM cGMP, 0.05 mg/ml bovine serum albumin, 0.1 mM L-arginine, 10 µM CaCl2, 10 µg/ml calmodulin, 50 µM NADPH, 40 µM CHAPS, and 1 mM GSH. Reactions were started by adding nNOS (2 µg/ml) and 10-fold concentrated stock solutions of MnTMPyP. For direct activation with the NO donor DEA/NO (open symbols), purified sGC (50 ng) was incubated at 37° for 10 min in a total volume of 0.1 ml of a 50 mM K2HPO4/KH2PO4 buffer, pH 7.4, containing 0.5 mM [-32P]GTP, 3 mM MgCl2, 1 mM cGMP, and 0.05 mg/ml bovine serum albumin. Reactions were started by adding DEA/NO (1 µM final concentration) and 10-fold concentrated stock solutions of MnTMPyP to the assay mixtures. Incubations were terminated by ZnCO3 precipitation, followed by isolation of [-32P]cGMP. Results were corrected for enzyme-deficient blanks and recovery of cGMP. Data are mean values ± standard error of three experiments performed in duplicate.

To study whether this inhibitory effect of MnTMPyP was the result of a direct interaction with sGC and/or scavenging of free NO, we measured NO autoxidation kinetics electrochemically in the absence and presence of the porphyrin under various conditions. Fig. 4 shows representative traces obtained with authentic NO in 50 mM phosphate buffer (pH 7.4) at 37°. Addition of 3.6 µl of an NO solution (~2 mM) to a total volume of 1.8 ml of buffer led to a pronounced response of the electrode, followed by a decrease of the signal with second order kinetics. The NO oxidation kinetics were identical in the absence (a) and presence (b) of 2 µM MnTMPyP, yielding third order rate constants of 2.39 ± 0.54 × 10⁷/M²/sec (n = 3) and 2.14 ± 0.20 × 10⁷/M²/sec (n = 4), respectively. However, the porphyrin was converted to an efficient NO scavenger under reducing conditions. The presence of 1 mM GSH did not appreciably affect the autoxidation of NO (k = 3.2 ± 0.1 × 10⁷/M²/sec; n = 3; c), but MnTMPyP (2 µM) led to a rapid disappearance of the NO signal with an apparent zero order rate constant of 145 ± 23 nM/sec under these conditions (n = 4; d). Thus, MnTMPyP scavenged NO in the presence but not in the absence of GSH. Identical data were obtained in the presence of the reducing compound NADPH (0.2 mM) (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effect of MnTMPyP on NO autoxidation in the presence or absence of GSH. An aliquot (3.6 µl) of a saturated aqueous NO solution (~2 mM) was added to 1.8 ml of a 50 mM K2HPO4/KH2PO4 buffer, pH 7.4 at 37°. Changes in NO concentrations were monitored over 5 min with a Clark-type NO electrode. Conditions were: buffer (a), 1 mM GSH (b), 2 µM MnTMPyP (c), and 1 mM GSH plus 2 µM MnTMPyP (d). MnTMPyP was injected at the peak level of the NO concentration (arrow). The traces shown are representative of four experiments.

The reactivity of MnTMPyP toward peroxynitrite and NO was studied by light absorbance spectroscopy. The parent complex exhibited a Soret absorbance maximum at 462 nm that was shifted to 420 nm upon addition of peroxynitrite (Fig. 5A). A previous study identified this 420 nm species as the corresponding oxo-Mn(IV) complex which rapidly decays back to the starting Mn(III)porphyrin (Groves and Marla, 1995). Treating MnTMPyP (2 µM) with a solution of NO (final concentration 0.1 mM) did not alter the absorbance spectrum (Fig. 5B; solid line). However, when the same experiment was performed in the presence of 1 mM GSH, a pronounced shift of the Soret band from 462 to 434 nm was observed, indicative of the formation of the corresponding Mn(II)-nitrosyl complex (Yonetani et al., 1972). Identical spectral changes were observed when NADPH (0.2 mM) was present as reducing agent instead of GSH (data not shown). As observed with the oxidized complex formed by peroxynitrite, the Mn(II)-NO species fairly rapidly decayed back to the Mn(III)porphyrin (t_1/2 ~ 2 min). These experiments provided a reliable explanation for our electrochemical observations on scavenging of NO by MnTMPyP in the presence of GSH.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Perturbations of the MnTMPyP light absorbance spectra induced by peroxynitrite (A) and NO/GSH (B). A, Light absorbance spectra of MnTMPyP (2 µM) were recorded before (solid line) and after (dotted line) addition of authentic peroxynitrite (0.1 mM final) in 50 mM K2HPO4/KH2PO4 buffer, pH 7.4, at ambient temperature. The experiment shown is representative of three. B, Light absorbance spectra of MnTMPyP (2 µM) were recorded in the presence of 0.1 mM authentic NO (solid line), 1 mM GSH (dashed line), and in the presence of GSH plus NO (dotted line) at ambient temperature in 50 mM K2HPO4/KH2PO4 buffer, pH 7.4. The experiments shown are representative of three.

	Discussion

Top Summary Introduction Procedures Results Discussion References

The present data confirm previous reports on the rapid reaction of the cell permeable Mn-porphyrin MnTMPyP with peroxynitrite (Groves and Marla, 1995; Hunt et al., 1997) _, but we question the usefulness of this compound and related drugs as specific tools to probe the involvement of peroxynitrite in biological processes. We obtained no evidence for a SOD-mimetic effect of the porphyrin. Scavenging of O₂ by SOD is known to prolong the half-life of NO (Gryglewski et al., 1986), resulting in a potentiation of NO-mediated effects, including the accumulation of cGMP in vascular endothelial cells (Mayer et al., 1993), but MnTMPyP inhibited cGMP formation in this cell culture system. Because there is solid evidence that MnTMPyP has SOD activity in vitro (Faulkner et al., 1994; Liochev and Fridovich, 1995), our data suggest that the SOD-mimetic effect of MnTMPyP was overcome by inhibition of endothelial sGC at low concentrations of the drug.

In an earlier study, we found that the presence of SOD is essential for the detection of the formation of NO by purified NOS and concluded that, in the absence of SOD, NO is converted to peroxynitrite by enzymatically produced O₂ (Mayer et al., 1995b). This conclusion was recently questioned in an article claiming that the effect of SOD is not a consequence of O₂ scavenging but results from oxidation of the postulated NOS product NO to NO (Schmidt et al., 1996). Synthetic compounds with SOD-mimetic activity could be useful tools to unequivocally settle this issue. However, like the Cu(II) complexes studied previously (Mayer et al., 1996), MnTMPyP turned out to be a potent inhibitor of purified NOS. The inhibition of enzymatic cytochrome c reduction, together with the pronounced stimulation of uncoupled NADPH oxidation, clearly demonstrates that the porphyrin interferes as a parasitic electron acceptor with the transfer of electrons from the flavin-containing reductase domain to the catalytic heme site of NOS. Inasmuch as we obtained virtually identical data with all three NOS isoforms, the low sensitivity of Ca²⁺-activated arginine-to-citrulline conversion in the cell culture experiments hints at the existence of protective mechanisms that prevent the interaction of the Mn-porphyrin with NOS in intact cells. Alternatively, this compound may not so easily gain access to mammalian cells as to E. coli (Faulkner et al., 1994; Liochev and Fridovich, 1995). It remains to be clarified whether the inhibition of NO₂/NO₃ accumulation in the medium of endotoxin-treated macrophages by a related Mn-porphyrinic drug (Szabo et al., 1996) was the result of radical scavenging or inhibition of iNOS.

It was surprising to find that MnTMPyP potently scavenged NO in the presence of reducing compounds such as GSH, the most abundant intracellular thiol in mammalian tissues (Meister, 1994). Together with light absorbance spectroscopy to monitor redox changes of MnTMPyP, the measurement of NO with a Clark-type electrode allowed us to elucidate the mechanism underlying this effect. Our data suggest that the parent Mn(III) complex does not bind NO but becomes reduced by GSH or NADPH to the Mn(II) species, the high NO binding affinity of which may have resulted in formation of the corresponding nitrosyl complex with a typical Soret band at 434 nm (Yonetani et al., 1972; Dierks et al., 1997). The scavenging of NO occurred at substoichiometric concentrations of MnTMPyP, pointing to redox cycling of the Mn-porphyrin. This may involve rapid dissociation of NO, followed by reoxidation of the Mn(II) complex by O₂ (Hoffman, 1979). Thus, although our results confirm that MnTMPyP does not scavenge NO in nonreducing buffers (Szabo et al., 1996), this drug may act as a potent, catalytically active NO scavenger in the reducing environment of cells.

Finally, MnTMPyP proved to be a potent inhibitor of sGC, the major physiological target of NO. The inhibitory effect was most likely direct, i.e., not the result of scavenging of NO because MnTMPyP exhibited virtually identical potency in the absence and presence of GSH even though it did not scavenge NO in the absence of the thiol. Further, no scavenging of NO was observed in the sGC assay buffer in the absence of GSH (electrochemical data; not shown). These findings indicate that MnTMPyP, like other metalloporphyrins such as Zn- and Sn-protoporphyrin IX (Ignarro, 1992; Luo and Vincent, 1994), is a potent inhibitor of sGC, presumably because of interference with heme-dependent NO stimulation of the enzyme.