|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 278, Issue 49, 48602-48610, December 5, 2003
Single-turnover of Nitric-oxide Synthase in the Presence of 4-Amino-tetrahydrobiopterinPROPOSED ROLE FOR TETRAHYDROBIOPTERIN AS A PROTON DONOR*![]() ![]() ![]() ![]() ![]()
From the
Received for publication, May 30, 2003 , and in revised form, August 26, 2003.
Tetrahydrobiopterin (BH4) is an essential cofactor of nitric-oxide synthase (NOS) that serves as a one-electron donor to the oxyferrous·heme complex. 4-Aminotetrahydrobiopterin (4-amino-BH4) is a potent inhibitor of NO synthesis, although it mimics all allosteric and structural effects of BH4 and exhibits comparable redox properties. We studied the reaction of reduced endothelial NOS oxygenase domain with O2 in the presence of 4-amino-BH4 at 30 °C by optical and electron paramagnetic resonance (EPR) spectroscopy. With Arg as the substrate, we observed a trihydropteridine radical with a corresponding heme species that was oxyferrous, with a Soret maximum at 428 nm and no EPR signal. With NG-hydroxy-L-arginine (NHA) no pterin radical appeared, whereas an axial ferrous heme·NO complex was formed. The corresponding optical spectra, with Soret bands at 417/423 nm, suggest that the proximal sulfur ligand is protonated. Accordingly, 4-amino-BH4 serves as a one-electron donor to Fe(II)·O2 with both Arg and NHA, but the reaction cycle cannot be completed with either substrate. We propose that protonation of is inhibited in the presence of 4-amino-BH4. With Arg, dissociation of and binding of O2 yields Fe(II)·O2 and a pteridine radical; with NHA, reaction of the substrate with heme-bound eventually yields Fe(II)·NO and reduced 4-amino-BH4. These results suggest that BH4 donates a proton to during catalysis and that inhibition by 4-amino-BH4 may be due to its inability to support this essential protonation step.
Nitric-oxide synthase (NOS,1 EC 1.14.13.39 [EC] ) converts L-arginine into L-citrulline and NO in two consecutive reaction cycles with intermediate formation of NG-hydroxy-L-arginine (NHA) (see Refs. 13 for recent reviews). Both reactions consume one equivalent of O2 as well as electrons in stoichiometries of 2 and 1 equivalent for the first and second cycle, respectively. Oxygen binding and NO synthesis take place at a P450-type heme in the oxygenase domain; the electrons are provided by NADPH via two flavin moieties in the reductase domain. The molecular mechanism of neither cycle is fully known yet but it is generally believed that hydroxylation of Arg follows a similar pathway as proposed for mono-oxygenation by cytochrome P450. For the conversion of NHA to citrulline and NO, a range of different models was suggested. In both cycles the ferrous oxy complex is the ultimate unequivocally identified intermediate.
NOS is unique among P450-type enzymes in requiring tetrahydrobiopterin (BH4) as a cofactor (see Refs. 4 and 5 for reviews). The main function of BH4 is to serve as a transient one-electron donor to the ferrous oxy·heme complex (6, 7), as was demonstrated for the first cycle by the detection of the BH32 radical by electron paramagnetic resonance (EPR) spectroscopy (812). Most likely, BH4 performs the same function in the second cycle as well (9, 13, 14), although little or no BH3 radical is observed (810), probably because the different electronic requirements of the second cycle allow rapid regeneration of BH4 (4, 13). As yet it is unclear whether BH4 also participates in later reaction steps such as proton donation or stabilization of the putative ferryl·oxy complex (4).
In line with the key role of BH4 as a one-electron donor, other tetrahydropteridines are generally able to substitute BH4 as cofactors, whereas 7,8-dihydropteridines, which are fairly resistant to redox reactions (15), are BH4-competitive inhibitors (4, 1618). One notable exception is 4-aminotetrahydrobiopterin (4-amino-BH4), which is a strong inhibitor of NOS catalysis (16, 17, 19) despite being a tetrahydropteridine with electrochemical properties similar to those of BH4 (15). Here, we report the effects of 4-amino-BH4 on the oxidation of reduced NOS by O2 in the presence of Arg and NHA, studied by EPR and optical spectroscopy at 30 °C.
MaterialsThe oxygenase domain of bovine endothelial NOS (eNOSoxy) was expressed in and purified from Escherichia coli (10). The cDNA encoding the domain was a kind gift from Dr. W. C. Sessa. Recombinant rat brain full-length BH4-containing (1 BH4 per dimer) and BH4-deficient (<0.1 BH4 per dimer) forms of neuronal NOS (nNOS) were purified from baculovirus-infected Sf9 cells (20, 21). Recombinant human neuronal NOS (hnNOS) was obtained from the yeast Pichia pastoris expression system using essentially the same techniques and procedures as described for the human endothelial isoform (22). The cDNA was a kind gift from Dr. J. F. Parkinson. All chemicals were from Sigma or Merck except for NHA (Alexis; Lausen, Switzerland), all pteridines (Dr. B. Schircks, Jona, Switzerland), Ar and O2 (Air Liquide, Schwechat, Austria), CO, and NO (Linde Gas, Graz, Austria, Montpellier, France, and Oslo, Norway). Low Temperature Optical SpectroscopyLow temperature UV-visible absorption spectra of the reaction between reduced eNOSoxy and O2 were measured in a Cary 3E (Varian, Palo Alto, CA) spectrophotometer, adapted for low temperature studies, according to previously published procedures (6, 13, 23). EPR SpectroscopySamples were prepared as described previously (10). Samples in 50% ethylene glycol, 50 mM potassium Pi (pH 7.5) were made anaerobic by gassing with argon for 30 min at 0 °C. Just enough anaerobic dithionite solution was added to the samples at room temperature to shift the Soret band from 396 nm to 412 nm, indicating the reduction of ferric heme to the ferrous state. Samples were cooled to 30 °C, and the reaction was initiated by blowing 250 µl of pure O2 through the 200-µl solution. The samples were allowed to react for 20 s before flash-freezing in 130 °C n-pentane. EPR spectra were recorded at 9.65 GHz on a Bruker ESP300E EPR spectrometer equipped with an Oxford Instruments cryostat 900 (0.5- or 2-milliwatt microwave power, 100-kHz modulation frequency, 1-, 3-, and 10-G modulation amplitude, and temperatures between 4 and 100 K). Radicals and low spin heme signals were quantified by double integration and comparison with 0.5 mM Cu(II) under nonsaturating conditions as a standard. High spin ferric heme was quantified by comparison with 100% ferric high spin NOS and with high spin/low spin NOS mixtures of known composition. EPR simulations were performed using the SimFonia program provided by Bruker. Hyperfine coupling constants were varied along with line widths in x, y, and z direction until optimal fitting was achieved. A Lorentzian/Gaussian line shape ratio of 1.0 was always used. P1/2 values of the microwave power saturation were determined over a range of 0.3 microwatt to 200 milliwatts and least-squares fitted to I = 1/(1 + P/P1/2)b/2, in which I is the EPR intensity, P is the microwave power, and P1/2 and b are fitting parameters formally representing the microwave power at half-saturation and the inhomogeneity factor. NOS Activity DeterminationsNADPH oxidation was monitored at 340 nm and 37 °C in a total volume of 250 µl with a Hewlett-Packard 8452A diode array UV-visible absorption spectrophotometer (24). NO synthesis was quantified as nitrite/nitrate formation by the Griess assay according to published procedures (25) in 96-well microplates with an Anthos HT3 microplate reader (Fresenius Kabi, Linz, Austria). Calibration curves were constructed with NaNO2. The effect of 4-amino-BH4 on the integrity of the heme in single-turnover was investigated by measuring the Soret absorbance of the ferrous·CO complex (24). Typically, samples of 1-ml total volume, containing 3 µM eNOSoxy, 2 mM Arg, 0.2 mM CHAPS, 2.4 mM 2-mercaptoethanol, and 50 mM potassium Pi (pH 7.4) in the presence or absence of 50 µM BH4 or 4-amino-BH4, were prepared in quartz cuvettes equipped with screw-on rubber septa (Hellma type 117.104-QS, Müllheim/Baden, Germany). Samples were made anaerobic by flushing with argon for 5 min. Subsequently, samples were reduced by titration with sodium dithionite, added with a Hamilton syringe from 50 mM stock solutions that were prepared in 50 mM de-aerated potassium Pi (pH 7.4). After complete reduction, single-turnover was induced by flushing with pure O2 for 1 min. The progress of reduction and oxidation was monitored by absorption spectroscopy. After complete oxidation, samples were flushed with CO for 1 min and reduced with excess dithionite. Absorption spectra were recorded after each step in the procedure.
Effect of 4-Amino-BH4 on NO SynthesisPrevious studies have shown that 4-amino-BH4 is a strong BH4-competitive, reversible inhibitor of NO synthesis, measured as conversion of radiolabeled arginine and NHA to citrulline (16, 17, 19). To elucidate how 4-amino-BH4 affects the second step of NOS catalysis, we compared the formation of nitrite/nitrate from NHA by NOS in the presence and absence of BH4 and 4-amino-BH4 (Fig. 1). In the presence of BH4, 10-min incubation of hnNOS with NHA in complete assay reaction mixture yielded 42.2 ± 1.8 µM nitrite (n = 5). Nitrate was barely detectable ( 3 µM). In the presence of superoxide dismutase (SOD) and catalase (CAT), nitrite formation decreased by about 50% (18.2 ± 0.5 µM). Partial inhibition of NOS activity in the presence of SOD/CAT has been observed before and is apparently due to unmasking of NO-induced autoinhibition by scavenging of (26, 27). As expected, nitrite formation decreased in the absence of BH4, but the remaining yield (22.4 ± 5.7 µM) was still remarkably high. The oxidation of NHA in the presence of BH4-free NOS has been reported before and was proposed to involve the NOS heme directly (28, 29) or, alternatively, to be mediated by NOS-derived (30). In agreement with the latter hypothesis, we found that the activity in the absence of pterin was almost completely abolished by SOD/CAT (3.1 ± 0.9 µM). With Arg instead of NHA the yield of nitrite was 16.1 ± 2.6 µM in the presence of BH4; hardly any nitrite formation (2.1 ± 0.5 µM) occurred in the absence of pterin (not shown).
With 4-amino-BH4 results were very similar to those obtained in the absence of pterin. With NHA as the substrate, the yield was 17.6 ± 5.0 µM and SOD/CAT almost completely inhibited nitrite formation (1.3 ± 0.8 µM remaining). With Arg as the substrate, nitrite formation was hardly detectable (1.6 ± 0.4 µM, not shown). These results show that 4-amino-BH4 does not support SOD/CAT-resistant (i.e. genuinely enzymatic) NO synthesis from either substrate, whereas it allows uncoupled production, resulting in indirect SOD/CAT-sensitive formation of nitrite from NHA. Integrity of the Heme-thiolate Bond in the Presence of 4-Amino-BH4 To establish whether catalysis in the presence of 4-amino-BH4 affects the heme coordination structure, we determined if single-turnover has any effect on the formation of the characteristic absorption band at 450 nm in the presence of CO under reducing conditions (Fig. 2). In all cases, addition of dithionite to anaerobic eNOSoxy caused a shift of the Soret band from 395 nm (high spin ferric heme) to 414 nm (ferrous heme). Upon addition of O2 to fully reduced eNOSoxy, the Soret maximum was shifted back to 395 nm, indicating complete reoxidation. Addition of CO to the reoxidized compound, with both BH4 and 4-amino-BH4, resulted in a Soret band at 450 nm with little or no evidence of P420 formation (Fig. 2, A and B). An absorbance maximum at 450 nm was also formed in the absence of pterin, but it was unstable and slowly converted to a 420-nm species (Fig. 2C), in line with earlier reports (31). These results indicate that, like BH4, 4-amino-BH4 stabilizes the coordination structure of the heme.
Effect of 4-Amino-BH4 on NADPH OxidationTo study the effects of 4-amino-BH4 on uncoupled O2 reduction, we compared NADPH oxidation by hnNOS in the presence and absence of BH4 and 4-amino-BH4 and of Arg and NHA (Fig. 3). In the absence of substrate, both BH4 and 4-amino-BH4 moderately increased NADPH oxidation. Whereas Arg alone did not significantly affect NADPH oxidation, it counteracted the activity increase caused by BH4, probably due to NO-induced autoinhibition. Surprisingly, even stronger inhibition of NADPH oxidation occurred in the combined presence of Arg and 4-amino-BH4, even though NO is not formed under these conditions. The presence of Arg decreased the activity observed with 4-amino-BH4 by 78 ± 14%, which suggests that inhibition of NADPH oxidation involves catalytic turnover, thus providing the first indication of a mechanism-based component of NOS inhibition by 4-amino-BH4.
In the presence of NHA, NADPH oxidation was decreased, probably because the higher NO/NADPH ratio for this substrate (2 as compared with 2/3 for Arg) results in stronger NO-induced autoinhibition. Consequently, a similar decrease of NADPH oxidation by 4-amino-BH4 as obtained with Arg is not to be expected (inhibition of NADPH oxidation by 4-amino-BH4 would decrease the rate of NO synthesis and hence partially relieve NO-induced inhibition). Nevertheless, with NHA as well, NADPH oxidation became slower by 46 ± 13% in the presence of 4-amino-BH4.
EPR Spectroscopy: The Reaction with ArgPrevious studies have demonstrated the formation of a relatively stable BH3 radical intermediate in single-turnover of reduced NOS with O2 and Arg (812). Similar radical intermediates are formed with other tetrahydropteridines that support NO synthesis (10, 32). We surmised that, despite the fact that 4-amino-BH4 is an inhibitor of NO synthesis, it might nevertheless yield a one-electron oxidized radical, because its electrochemical properties resemble those of BH4 (15). Fig. 4A shows that the 4-amino-BH3 radical was indeed formed in single-turnover of reduced eNOSoxy with O2 and Arg. For the turnover sample, the radical amounted to 30% of the heme concentration, which corresponds to
The main difference with the BH3 radical at 10 K was that the EPR spectrum of 4-amino-BH3 exhibited more resolved hyperfine splitting, which might suggest line broadening of the BH3 radical signal (Fig. 4B). Line broadening might be caused by dipolar coupling between the pterin radical and high spin ferric heme, which is to be expected in view of the proximity of the two paramagnets. Such broadening should disappear at higher temperature because of fast relaxation of the heme. Indeed, at 100 K the BH3 radical signal was sharper, with hyperfine splitting similar to that of 4-amino BH3 at 10 K (Fig. 4C). This confirms that at 10 K the BH3 radical signal is broadened by dipolar coupling with the heme and that similar broadening does not occur for the 4-amino-BH3 radical. Strong interaction between the spins of the pterin and the adjacent heme can also be deduced from the microwave power saturation behavior, with all pterin radicals reported thus far exhibiting remarkably high P1/2 values (8, 10, 12, 32). We derived fitting parameters of 1.0 milliwatts for P1/2 and 0.77 for b, suggesting strong dipolar coupling between BH3 and high spin ferric heme (33). In contrast, the 4-amino-BH3 radical showed signs of partial saturation already at 10 microwatts (Fig. 5), with fitting parameters of 65 microwatts and 0.42 for P1/2 and b, respectively. The low value of P1/2 indicates that the majority of the radical is magnetically isolated, resulting in slow relaxation. However, the extremely low b value is suggestive of heterogeneity, with a minor part experiencing dipolar coupling with the heme (33).
It was demonstrated previously that successful simulation of the BH3 radical EPR spectrum requires the assumption that the unpaired electron resides mainly at N5, and that N5 is protonated (10). Accordingly, the BH3 spectrum observed in the present study at 100 K was satisfactorily fitted (Fig. 4C) with hyperfine interactions from N5, N5-H , and C6-H , in line with a recent study (12). Most of the features of the spectrum of 4-amino-BH3 observed at 10 K could be reproduced with similar fitting parameters, in line with the conclusion that most of the 4-amino-BH3 radical is magnetically isolated (Fig. 4A). However, a truly adequate fit required the introduction of a minor fraction with a much broader signal, in accordance with the microwave power saturation curve. It should be noted that the fitting parameters of the broad signal do not represent true hyperfine constants, because they are affected by coupling with the heme. Furthermore, the corresponding pterin fraction may be considerably smaller than the 33% that it contributes to the radical signal (see legend to Fig. 4), because the narrower signal is already saturated under the applied conditions. The fact that we can simulate the 4-amino-BH3 EPR spectrum with hyperfine interactions from one nitrogen and two hydrogens strongly suggests that, like BH3, the 4-amino-BH3 radical is an N5 centered cation with the highest spin density on N5.
EPR Spectroscopy: The Reaction with NHAWe found no pterin radical signal in the presence of 4-amino-BH4 and NHA. Instead, we observed formation of a signal in the g = 2 region with g values of 2.061, 2.010, and 1.982 and with 3-fold hyper-fine splitting of g2 and g3 by 17 and 10 G, respectively (Fig. 6). The parameters of the spectrum, which is typical of an Fe(II)·NO complex, are compared in Table I with those of other heme Fe(II)·NO complexes. Hyperfine splitting of 17 G at g2 is characteristic for a five-coordinated complex, whereas six-coordinated complexes generally display splitting constants of 20 G. This suggests profound changes at the proximal side of the heme after reaction of ferrous NOS with NHA and 4-amino-BH4. In a typical experiment we observed
The Fe(II)·NO complex was not observed in the absence of pterin, or with BH2 or 4-amino-BH2 (not shown). With BH4 at pH 7.5, we observed only trace amounts of the Fe(II)·NO complex in about 1 out of 10 samples. To exclude dithionite-induced artifacts, samples where prepared in which the heme centers were reduced by substoichiometric amounts of dithionite. The mixed ferric-ferrous state of the enzyme was confirmed by UV-visible spectroscopy. Under these conditions we still observed the Fe(II)·NO complex. Moreover, samples showing the Fe(II)·NO complex no longer exhibited the signal after transiently warming to room temperature, indicating that the species is an intermediate on the pathway to complete oxidation.
UV-visible Absorption SpectroscopyEarlier we used optical spectroscopy to study the same reactions (13). In the present study we repeated those experiments with similar results. In the presence of Arg the initial spectrum after O2 addition at 30 °C exhibited a strongly red-shifted Soret peak (
With NHA, the first spectrum after O2 addition at 30 °C had a maximum at 423 nm, which slowly shifted to 417 nm (Fig. 7B), in line with prior observations (13). This compound was stable at low temperature but converted fully to high spin ferric heme ( max 394 nm) after transient warming to room temperature (Fig. 7B). Characterization by UV-visible and EPR Spectroscopy of Ferrous-nitrosyl eNOSoxy in the Presence of 4-Amino-BH4 To ascertain that the unusual spectra of the ferrous·NO complex obtained in single-turnover with NHA and 4-amino-BH4 are not due to an effect of 4-amino-BH4 per se on the spectral properties and stability of the complex, we performed a number of control experiments. Addition of NO gas to an anaerobic sample of dithionite-reduced eNOSoxy (1.5 µM) in the presence of 0.5 mM NHA and 50 µM 4-amino-BH4 resulted in immediate (within 0.3 min) formation of a typical ferrous·NO spectrum with absorbance maximum at 436 nm (not shown). The spectrum was stable (no change in 15 min) and showed no trace of a compound absorbing at 417/423 nm. Essentially identical results were obtained with BH4. It has been demonstrated that Arg, NHA, and BH4 stabilize the six-coordinate form of the ferrous·NO complex, whereas O2 has a destabilizing effect (34). Thus, we determined whether the five-coordinate species observed in single-turnover with NHA and 4-amino-BH4 originated from O2-induced destabilization of the six-coordinate complex. Because the combined presence of NHA, 4-amino-BH4, and O2 might generate the five-coordinate species by single-turnover, we performed these control experiments in the presence of Arg instead of NHA. When NO was added to an anaerobic EPR sample of dithionite-reduced eNOSoxy (55 µM) in the presence of Arg and 4-amino-BH4 (1.5 mM each), a spectrum typical of six-coordinate Fe(II)·NO was generated with shape and parameters (gx 2.083, gz 2.005, gy 1.968; Az 20.2 G, Ay 11.3 G, Fig. 8A) virtually identical to previously published values (34). Under aerobic conditions the same species dominated the EPR spectrum (Fig. 8B), although the signal intensity was reduced, the shape became distorted, and a minor signal at g = 2.038 appeared. A minor signal at g = 2.03 has been observed before under similar conditions and may be due to a species with different Fe·NO geometry (34). We found no trace of the Fe(II)·NO complex observed with 4-amino-BH4 and NHA after single-turnover.
Identity of the Intermediate in the Reaction with ArgThe differences in line width and microwave power saturation of the EPR signals imply that the 4-amino-BH3 radical interacts much weaker with the heme than BH3. Simulation of the spectra indicates that the structures of the BH3 and 4-amino-BH3 radicals are similar (see "Results"), and both pteridines occupy virtually identical binding sites in the enzyme (35), precluding structural differences as the origin of the divergent power saturation. Dissociation of the pterin radical from the enzyme or oxidation of excess 4-amino-BH4 in solution by NOS-derived could give rise to magnetically isolated 4-amino-BH3. However, the observed radical is protonated at N5 (10), whereas the radical would be expected to rapidly deprotonate in solution (36). Furthermore, rapid disproportionation of the radical in solution (36), as well as rapid rereduction and further oxidation of 4-amino-BH3 by , would keep its concentration below the EPR detection limit. Moreover, SOD did not affect the intensity of the radical signal. Hence, 4-amino-BH3 in solution can be ruled out as the origin of the EPR signal.
We propose that the different power saturation is due to a different state of the heme. Whereas the catalytic cycle is rapidly completed in the reaction with BH4, yielding BH3·-H+/Fe(III), the reaction with 4-amino-BH4 generates an EPR-silent heme species. The corresponding optical spectrum (
Identification of the Intermediate in the Reaction with NHAWith NHA we observed no 4-amino-BH3 radical but formation of a ferrous·NO complex in significant yields. At first sight the EPR spectrum suggests a five-coordinate NO complex with a small admixture of six-coordinate Fe(II)·NO (34, 37, 38). As already reported previously (13), no intermediate with a Soret band at 436 nm was observed, confirming the absence of six-coordinate thiolate-ligated ferrous·NO heme. However, the absorbance maxima of five-coordinate ferrous·NO complexes, including that of P420-type NOS, are at EPR spectra of six-coordinate nitrosyl-hemoglobin and other N-coordinated nitrosyl·heme complexes consist of varying mixtures of a rhombic and an axial signal, with hyperfine splitting constants of 2122 and 1617 G, respectively (43, 44). The axial signal is thought to derive from a compound with greatly increased Fe-N(His) distance (44) and has EPR parameters resembling those of a five-coordinate nitrosyl heme (see Table I). Because protonation of the thiolate ligand is expected to cause an increase of the Fe·S and a decrease of the Fe·NO distance, a similar shift from a rhombic to an axial spectrum with parameters resembling those of the five-coordinate nitrosyl·heme is anticipated after protonation of the proximal Cys in NOS. We propose that the observed intermediate is a 4-amino-BH4-H+/Fe(II)·NO species, in which the heme has six coordinates with a protonated thiol as a weak proximal ligand. Implications of the Key Observations: A Role for BH4 as a Proton Donor in NO SynthesisThe observations permit us to draw important conclusions concerning the role of BH4 in catalysis with some confidence. The crucial results and their implications are as follows. (i) In the reaction with Arg, trihydropteridine radicals are formed in similar yields with the inhibitor 4-amino-BH4 and the activator BH4. This suggests that, in addition to its role as an electron donor to the oxyferrous complex, BH4 plays another essential role in catalysis that is not mimicked by 4-amino-BH4. (ii) In contrast to the results with BH4, the reactions with Arg and NHA in the presence of 4-amino-BH4 do not rapidly regenerate the ferric enzyme, but yield long-lived ferrous compounds. This suggests that in both cycles the additional function of BH4 involves its participation in a reaction step between formation of the reduced oxyferrous (ferrous-superoxy) complex and regeneration of ferric heme.
The identity of the affected reaction step can be inferred by elimination. The large separation from the site of catalysis precludes any direct mechanistic involvement of BH4. Long-range allosteric effects are conceivable, but in view of the close resemblance of the crystal structures of inducible NOS oxygenase domain with BH4 and 4-amino-BH4 (35), it is highly unlikely that such effects would differ between the two pteridines. In view of these facts, the only obvious way in which BH4 and 4-amino-BH4 might affect the catalytic cycle differently is by the participation of BH4, but not of 4-amino-BH4, in protonation of the ferrous superoxy complex. The fact that protonation of
Proposed Reaction Sequence with BH4 Current models of NO synthesis are based on the reaction cycle describing cytochrome P450 catalysis. Although the identity of the final oxidant in cytochrome P450 is still debated and may vary with the identity of the substrate and the type of reaction catalyzed (for reviews on the reaction mechanism of cytochrome P450 see Refs. 4648), considerable progress has been made in the detection of intermediates. Using cryoreduction techniques the formation of ferric peroxy and hydroperoxy complexes has been reported (49). The same technique was recently applied successfully to eNOSoxy, which demonstrated the formation of the peroxy complex but not of its protonated counterpart (45). However, in that study 4-amino-BH4 was used as a catalytically noncompetent pteridine, without considering the possibility that proton transfer to A hypothetical model of NOS catalysis that takes into account the putative role of BH4 as a combined electron/proton donor is presented in Scheme 1A. Accordingly, in both reaction cycles oxygen binding to ferrous heme, followed by electron and proton transfer by BH4, results in the formation of a hydroperoxy complex. A similar suggestion was made by Davydov et al. (45) on the basis of cryoreduction studies. In the crystal structure of eNOSoxy (with heme-bound NO as a surrogate for the oxy complex (50)) a potential proton relay pathway can be discerned leading from BH4 to heme-bound oxygen, involving one of the heme propionates, the side-chain carboxylate of Glu-363, and both the amino acid and guanidinium moieties of the substrate (Fig. 9).
Proposed Reaction Sequence with 4-Amino-BH4 According to our hypothesis, the catalytic cycle cannot continue beyond the state in the presence of 4-amino-BH4, and decay to ferric heme and hydrogen peroxide will be impossible as well. Under such conditions, the complex may dissociate instead, generating deoxyferrous heme and . Although there appear to be no published precedents for that particular reaction, autoxidation of to and ferric heme is assumed to be the main pathway for uncoupling of NOS (4). After dissociation of , ferrous heme will rapidly bind a second equivalent of O2, yielding the experimentally observed Fe(II)·O2/4-amino-BH3·-H+ species (Scheme 1B). Although we favor this scenario, the mechanistic details remain unclear, and alternative routes cannot be ruled out. For instance, O2 may expel from the heme, with O2 binding preceding dissociation of , or O2 may be reduced directly by the Fe(II)· complex.
The same reaction sequence can be expected to occur with NHA, with the Status of the Proximal Heme Ligand in the Ferrous NO ComplexThe combined EPR and optical spectra of the ferrous·NO complex in the reaction with NHA and 4-amino-BH4 strongly suggest that the heme remains six-coordinate but with a considerably weaker proximal ligand than a thiolate. Although the possibility that the proximal thiolate is displaced cannot be ruled out, the more likely option is that the cysteinyl sulfur becomes protonated, maintaining a weak bond to the heme.
One may speculate what might cause protonation of the proximal thiolate in the presence of 4-amino-BH4. Possibly, the presence of the unprotonated electron-donating
Inhibition of NO Synthesis by 4-Amino-BH4 According to our hypothesis reversible inhibition by 4-amino-BH4 is primarily due to its inability to transfer a proton to the
The detection of the 4-amino-BH3· radical in the reaction of eNOSoxy with Arg and O2 implies that 4-amino-BH4 can function as a one-electron donor to the Fe(II)·O2 complex. Yet the reaction of reduced eNOSoxy with O2 was halted at a stage before completion of the cycle with either substrate, strongly suggesting that BH4 must have an additional function in catalysis. The results are best rationalized by assuming that BH4 serves as a proton donor to heme-bound in both reaction cycles and that 4-amino-BH4 inhibits NO synthesis, because it does not support this crucial step of the NOS reaction. Further studies using more sophisticated spectroscopic techniques (electron-nuclear double resonance, pulsed EPR, magnetic circular dichroism, and resonance Raman) are necessary to resolve the issues raised in the present study.
* This work was supported by the Human Frontier Science Program (RGP0026/2001-M), by the Fonds zur Förderung der Wissenschaftlichen Forschung in Österreich (to B. M.), and by the Research Council of Norway (to K. K. A.). 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. || To whom correspondence should be addressed. Tel.: 43-316-380-5569; Fax: 43-316-380-9890; E-mail: antonius.gorren{at}uni-graz.at.
1 The abbreviations used are: NOS, nitric-oxide synthase; eNOSoxy, the oxygenase domain of recombinant bovine endothelial NOS, expressed in and purified from E. coli; nNOS, neuronal NOS, used here also to refer to the recombinant rat brain enzyme purified from baculovirus-infected Sf9 cells; hnNOS, human neuronal NOS, expressed and purified from P. pastoris; BH4, tetrahydrobiopterin ((6R)-5,6,7,8-tetrahydro-6-(L-erythro-1',2'-dihydroxypropyl)pterin); 4-amino-BH4, 4-aminotetrahydrobiopterin ((6R)-2,4-diamino-5,6,7,8-tetrahydro-6-(L-erythro-1',2'-dihydroxypropyl)pteridine); BH3 and 4-amino-BH3, the one-electron oxidized radical compounds corresponding to BH4 and 4-amino-BH4, respectively; NHA, NG-hydroxy-L-arginine; CaM, calmodulin; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; SOD, superoxide dismutase; CAT, catalase; EPR, electron paramagnetic resonance.
2 The abbreviations BH3 and 4-amino-BH3 are used here to refer to any one-electron oxidized species of BH4 and 4-amino-BH4, irrespective of the protonation state. The same is true for the abbreviations BH4 and 4-amino-BH4.
3 The impetus for the present studies was furnished by the striking resemblance of the absorbance spectrum of this compound to the spectrum of a cytochrome P450 intermediate that was originally assigned to a
We thank Dr. John Salerno for helpful discussion. The excellent technical assistance of F. Henning Cederkvist and Hans-Petter Hersleth with enzyme preparation is gratefully acknowledged.
This article has been cited by other articles:
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||