Chimeras of Nitric-oxide Synthase Types I and III Establish Fundamental Correlates between Heme Reduction, Heme-NO Complex Formation, and Catalytic Activity*

Neuronal nitric-oxide synthase (nNOS or NOS I) and endothelial NOS (eNOS or NOS III) differ widely in their reductase and nitric oxide (NO) synthesis activities, electron transfer rates, and propensities to form a heme-NO complex during catalysis. We generated chimeras by swapping eNOS and nNOS oxygenase domains to understand the basis for these differences and to identify structural elements that determine their catalytic behaviors. Swapping oxygenase domains did not alter domain-specific catalytic functions (cytochrome c reduction or H2O2-supportedN ω-hydroxy-l-arginine oxidation) but markedly affected steady-state NO synthesis and NADPH oxidation compared with native eNOS and nNOS. Stopped-flow analysis showed that reductase domains either maintained (nNOS) or slightly exceeded (eNOS) their native rates of heme reduction in each chimera. Heme reduction rates were found to correlate with the initial rates of NADPH oxidation and heme-NO complex formation, with the percentage of heme-NO complex attained during the steady state, and with NO synthesis activity. Oxygenase domain identity influenced these parameters to a lesser degree. We conclude: 1) Heme reduction rates in nNOS and eNOS are controlled primarily by their reductase domains and are almost independent of oxygenase domain identity. 2) Heme reduction rate is the dominant parameter controlling the kinetics and extent of heme-NO complex formation in both eNOS and nNOS, and thus it determines to what degree heme-NO complex formation influences their steady-state NO synthesis, whereas oxygenase domains provide minor but important influences. 3) General principles that relate heme reduction rate, heme-NO complex formation, and NO synthesis are not specific for nNOS but apply to eNOS as well.

Nitric oxide (NO) 1 is generated by nitric-oxide synthases (NOSs) and has multiple functions in physiology and pathology (1)(2)(3). Animals express three main NOS isoforms: one is cytokine-inducible and Ca 2ϩ -independent (iNOS or NOS II), and the two others are expressed constitutively (nNOS or NOS I; eNOS or NOS III) and become activated by Ca 2ϩ -dependent calmodulin (CaM) binding. All NOSs are bi-domain enzymes comprised of an N-terminal oxygenase domain that binds iron protoporphyrin IX (heme), (6R)-5,6,7,8-tetrahydro-L-biopterin (H 4 B), and L-arginine (Arg) and a C-terminal reductase domain that binds FMN, FAD, and NADPH (4 -6). A CaM binding motif is located between the oxygenase and reductase domains, and its occupancy triggers electron transfer between the reductase domain FMN and the oxygenase domain heme (7). This enables NOSs to catalyze the NADPH-and O 2 -dependent oxidation of Arg to generate NO and citrulline, with N -hydroxyarginine (NOHA) being formed as an enzyme-bound intermediate (4 -6).
The constitutive NOSs share many features but differ markedly in their catalytic profiles. For example, nNOS is 3-4 times more active than eNOS in steady-state NO synthesis (8 -13). This holds true even though a majority of nNOS partitions into an inactive ferrous heme-NO complex immediately after initiating NO synthesis (14). This lowers the concentration of active nNOS molecules by 4-or 5-fold, creates a condition in which oxidation of the ferrous-NO complex becomes rate-limiting, and shifts the apparent K m (O 2 ) value of the enzyme to a much higher value (15,16). In contrast, a very minor percentage of eNOS accumulates as a heme-NO complex during steady-state NO synthesis (13), and its slow catalysis is associated with a much slower heme reduction rate than nNOS (13,17,18).
To better understand how the heme reduction rate controls NO complex formation and NO synthesis, we developed a kinetic simulation model for nNOS catalysis (19). Our model incorporates the key finding that newly synthesized NO binds to the ferric heme before it leaves the enzyme active site (20). The kinetic model accurately simulates initial and steady-state features of nNOS catalysis including heme-NO complex formation, a concomitant deflection in NADPH oxidation and NO synthesis, and an increase in apparent K m (O 2 ) value. Experimental evidence and additional simulations revealed that slowing down heme reduction in nNOS decreased the percentage of heme-NO complex and the rate of NO synthesis achieved in the steady state (21). On this basis, we hypothesized that eNOS behavior might fit the kinetic model and suggested that the differences between nNOS and eNOS might be explained by their divergent heme reduction rates (19,21).
To help test this hypothesis, we created chimeras by exchanging the oxygenase domains of nNOS and eNOS (N ox E red and E ox N red ). In pioneering work, Ortiz de Montellano and co-workers (12,22) generated similar chimeras from eNOS and nNOS and characterized their steady-state NO synthesis, cytochrome c reduction, and NADPH oxidation in response to Arg and H 4 B. The authors concluded that the reductase domain controlled the rates of NO synthesis and cytochrome c reduction, whereas oxygenase domains controlled NADPH oxidation in response to Arg and H 4 B. In our case, we hoped the chimeras would reveal how heme reduction, NO complex formation, and NO synthesis are related in eNOS and help us gauge to what extent reductase and oxygenase domains control these parameters in either NOS isoform. We examined flavin and heme reduction rates, heme-NO complex formation, and initial and steady-state catalytic behaviors of each chimera and compared these to data obtained with eNOS and nNOS. The results show how individual reductase and oxygenase domains regulate heme reduction and NO complex formation in eNOS and nNOS and how these two factors combine to regulate catalysis.

EXPERIMENTAL PROCEDURES
Materials-All regents and materials were obtained from Sigma or sources reported previously (21,23).
Molecular Biology-Restriction digestions, cloning, bacterial growth, and the transformation and isolation of DNA fragments were performed using standard procedures. Rat nNOS DNA and bovine eNOS DNA were inserted into the 5Ј-NdeI and 3Ј-XbaI sites of the pCWori vector (23,24). To create the chimeras we used site-directed mutagenesis to generate a unique restriction site between the end of the oxygenase domain and the beginning of the CaM binding domain in both bovine eNOS and rat nNOS. The unique restriction site Eco47III was incorporated at S 485 -A 486 of eNOS and H 714 -V 715 of nNOS. This created a silent mutation in eNOS and an His-Val3 Ser-Ala mutation in nNOS. Sequence alignment using MacVector revealed that both Eco47III sites were located in identical positions in nNOS and eNOS. For making the Eco47III restriction site in eNOS, we used the QuikChange TM sitedirected mutagenesis kit from Stratagene. The oligonucleotides used to construct the Eco47III site (underlined) in eNOS were synthesized by Integrated DNA Technologies, and their corresponding oligonucleotides were as follows: S 485 -A 486 -Eco47III sense, TGGAAAGGGAGCGCTAC-CAAGGGCGCCGGCATCA and S 485 -A 486 -Eco47III antisense, TGAT-GCCGGCGCCCTTGGTAGCGCTCCCTTTCCA. A RoboCycler gradient 96 from Stratagene was employed. The standard PCR cycling parameters were 3 min for denaturing of the template at 95°C and 16 cycles for amplification (30 s for melting at 95°C, 1 min for annealing at 60°C, and 18 min for extension at 68°C) followed by a 7-min extension at 68°C. The protocol used ϳ50 ng of template, 20 pmol of each primer, 2 l of 10 mM dNTPs, and 1 M 2.5-unit Pfu polymerase in a final volume of 100 l. The PCR product was digested by 1 l of DpnI endonuclease and then transformed into Epicurian Coli® XL1-Blue supercompetent cells. The Eco47III restriction site in the nNOS cDNA was constructed by subcloning a PCR-generated fragment from pCWori/nNOS using a 3Ј-oligo containing a newly engineered Eco47III site. The nNOS fragment was obtained by PCR amplification using Pfu Turbo DNA polymerase (Stratagene), which possesses higher fidelity than other polymerases. The nNOS cDNA fragment coding from the BlpI unique restriction site 622 to the SanDI restriction site 2162 was amplified using the following primers: primer 1, CCTGTGCTGAGCATCCTCAA; primer 2, TGGGGGTCCCGTTGGTGCCCTTCCAAGCGCTGGTGTTC-CATGGATCAGG. Here the PCR cycling parameters were 3 min for denaturing of the template at 95°C and 28 cycles for amplification (30 s for melting at 95°C, 1 min for annealing at 58°C, and 6 min for extension at 68°C) followed by a 12-min extension at 68°C. The protocol used ϳ10 ng of template, 50 pmol of each primer, 2 l of 10 mM dNTPs, and 1 M 2.5-unit Pfu polymerase in a final volume of 100 l. The PCR product and wild-type pCWori vector containing nNOS DNA were digested by both BlpI and SanDI restriction endonuclease enzymes, and fragments were isolated by 1% agarose gel. The double-digested fragment of wild-type NOS pCWori plasmid was replaced by the double-digested PCR fragment and transformed into JM109 cells to generate the recombinant plasmid. Both chimera proteins were constructed by interchanging the double restriction (NdeI and Eco47III)-digested fragments. Chimeric DNA constructs were confirmed by DNA sequencing at the Cleveland Clinic sequencing facility. Chimeric cDNAs in the pCWori plasmid were transformed into Escherichia coli strain BL21(DE3) for protein expression.
Expression and Purification of Wild-type and Chimera Proteins-Wild-type rat nNOS, bovine eNOS, and both chimera proteins (E ox N red and N ox E red ) had a His 6 tag attached to their N termini to aid purification. They were overexpressed in E. coli strain BL21(DE3) and purified by sequential chromatography on Ni 2ϩ -nitrilotriacetic acid and 2Ј,5Ј-ADP-Sepharose resins as described (15,23). The ferrous-CO adduct absorbing at 444 nm was used to quantitate heme protein content using an extinction coefficient of 74 mM Ϫ1 cm Ϫ1 (A 444 -A 500 ).
NO Synthesis, NADPH Oxidation, and Cytochrome c Reduction-Steady-state activities of wild-type and chimera proteins were determined separately at 25°C using spectrophotometric assays that were described previously in detail (15,23). Absorbance at 436 nm was monitored to follow ferrous heme-NO formation and absorbance at 340 nm was monitored to follow NADPH oxidation (15,21). The concentration of ferrous heme-NO complex formed during NO synthesis was estimated from the absorbance change at 436 nm using an extinction coefficient of 49,800 M Ϫ1 cm Ϫ1 (14), and the amount of NADPH oxidation was determined using an extinction coefficient of 6,220 M Ϫ1 cm Ϫ1 at 340 nm. Signal/noise ratios were improved by averaging six consecutive scans. Each experiment was performed three separate times.

RESULTS
Expression and Physical Properties-Purified N ox E red and E ox N red both exhibit the expected molecular mass (Fig. 1). Spectroscopic analysis showed that their heme shifted to a high-spin state in the presence of 20 M H 4 B and 1 mM Arg. Dithionite reduction of each chimera in the presence of Arg, H 4 B, and CO produced the expected 444-nm absorbance peak for the ferrous-CO complex in all cases (data not shown). These data confirm that exchanging the oxygenase domains of eNOS and nNOS did not alter protein expression or the physical properties of the oxygenase domains.
Reductase-independent Catalysis-We first compared reductase domain-independent catalysis by the chimeras and wildtype NOSs by measuring H 2 O 2 -dependent NOHA oxidation (Table I). The chimeras catalyzed NOHA oxidation to different degrees such that the activity of each chimera matched with the wild-type NOS that provided its oxygenase domain. This is consistent with the reaction not requiring electrons from the reductase domain (25,26) and indicates that, under this circumstance, reductase domain identity did not influence catalysis by the oxygenase domains.
Steady-state Catalysis-NADPH-dependent cytochrome c reductase activity of each chimera in the presence or absence of CaM matched the activity of the NOS that provided its reductase domain (Table II). Thus, swapping oxygenase domains did not influence the reductase domain catalysis or response to CaM, which is consistent with previous results (12,22). Steadystate NO synthesis activities of E ox N red and nNOS were identical and well coupled to NADPH oxidation (2.1 and 1.9 NADPH oxidized/NO formed, respectively) (Table II). In contrast, the steady-state NO synthesis activity of N ox E red was about one-third that of nNOS but was 33% greater than eNOS. NO synthesis by N ox E red and eNOS was also less coupled to their NADPH oxidation (4.4 and 4.0 NADPH oxidized/NO formed, respectively). Thus, the rates of NO synthesis and NADPH oxidation by each chimera equaled or approached the NOS isoform that provided its reductase domain.
Kinetics of Flavin and Heme Reduction-We measured the rates of NADPH-dependent flavin and heme reduction in CaMbound chimeras using stopped-flow spectroscopy under anaerobic conditions. Fig. 2 (left panels) depicts flavin reduction as an absorbance decrease at 485 nm versus time. Flavin reduction was biphasic in both chimeras and was somewhat faster in the chimera containing the eNOS reductase domain (Table III). This rate difference was also observed when comparing flavin  reduction in eNOS and nNOS (13). In contrast, heme reduction (as measured by CO binding) 2 in N ox E red was 380 times slower than in E ox N red (Fig. 2, right panels). The heme reduction rate observed for E ox N red closely matched that reported for nNOS, and the rate seen with N ox E red was twice as fast as eNOS (Table III).
Heme-NO Complex Buildup-We compared heme-NO complex buildup in the chimeras during steady-state NO synthesis. Fig. 3 contains wavelength scans of nNOS, eNOS, and the two chimeras before and during NO synthesis at 15°C. nNOS and E ox N red exhibited strong Soret absorbance positioned near 436 nm during steady-state NO synthesis, indicating their significant partitioning into a heme-NO complex. In contrast, N ox E red had a less prominent Soret absorbance at 436 nm in the steady state, indicating it formed less heme-NO complex, and eNOS showed very little absorbance gain in this region of the spectrum. Difference spectroscopy (Fig. 3, insets) confirmed that the heme-NO complex had a Soret peak at 436 nm and a broad visible absorbance at 560 nm in all cases, indicating that the heme-NO complex was predominantly ferrous. The estimated percentage of ferrous heme-NO complex present at steady state was ϳ70% in nNOS and E ox N red , ϳ25% in N ox E red , and ϳ12% in eNOS.

Kinetics of NADPH Oxidation and Heme-NO Complex
Formation-We next utilized stopped-flow spectroscopy to investigate the kinetics and extent of heme-NO complex formation and their relationship to NADPH oxidation during the initial and steady-state phases of NO synthesis. In Fig. 4, absorbance changes at 436 and 340 nm were monitored versus the time to follow heme-NO complex formation and NADPH oxidation, respectively, in reactions run at 10°C. In all cases, heme-NO complex buildup was best described as a biphasic process (Table IV, k 1 , k 2 ). The rates of heme-NO complex formation were essentially the same in nNOS and E ox N red , whereas they were somewhat faster in N ox E red compared with eNOS. The apparent k 1 values for nNOS and E ox N red were four and six times faster than k 1 values for N ox E red and eNOS, respectively. The apparent k 2 values for nNOS and E ox N red were 14 and 8 times faster than k 2 values for N ox E red and eNOS, respectively. The percentage of heme-NO complex at steady state was estimated from the stopped-flow data and displayed the rank order nNOS Ͼ E ox N red Ͼ Ͼ N ox E red Ͼ eNOS (Table IV).
The initial rates of NADPH oxidation during heme-NO complex formation (Table IV, m 1 values) followed a rank order of nNOS Ͼ E ox N red Ͼ Ͼ N ox E red Ͼ eNOS. Subsequent heme-NO complex buildup in nNOS and in E ox N red was associated with a slowing down of their NADPH oxidation rates (Fig. 4) such that they were ϳseven and five times slower than the initial rates, respectively, once reaching the steady state (Table IV, m 2 values). In contrast, N ox E red showed no discernable change in its NADPH oxidation rate between initial and steady-state phases of NO synthesis, and for eNOS the NADPH oxidation rate slightly increased (Table IV; Fig. 4). DISCUSSION Although nNOS and eNOS are both expressed constitutively and become activated by Ca 2ϩ /CaM binding, they differ markedly in their reductase and NO synthesis activities, electron transfer rates, and propensities to form a heme-NO complex during catalysis (12-14, 17, 18, 27). We generated nNOS-eNOS chimeras with swapped oxygenase domains to understand how electron transfer, heme-NO complex formation, and NO synthesis activities are related in eNOS and nNOS and to identify structural features that underpin their different catalytic behaviors.  Steady-state cytochrome c reductase and NO synthesis activities of each chimera most closely matched the NOS isozyme that provided its reductase domain. This implies that each reductase domain maintained its native catalytic functions and was the primary determinant of NO synthesis activity. Nishida and Ortiz de Montellano (12,22) reached identical conclusions using eNOS-nNOS chimeras that were similar but not identical to ours. 3 Our study extends their work by providing data on electron transfer rates, pre-steady-state behaviors, and heme-NO complex formation, which considered together can explain the catalytic profiles of nNOS, eNOS, and their chimeras.
A central finding is that eNOS and nNOS reductase domains essentially maintained their native electron transfer rates to NOS ferric heme in both chimeras. Thus, in this regard reductase domains of eNOS and nNOS can interact equally well with either oxygenase domain. In eNOS the reductase domain catalyzes slow electron transfer from its FMN group to the ferric heme, whereas the nNOS reductase domain is much faster (13,17). Heme reduction is not limited by slow flavin reduction in either eNOS (13,17), nNOS (27), or the chimeras (see Table III) when CaM is bound. Together, this establishes that no distinct structural or electronic features exist in the oxygenase domains of eNOS and nNOS to control their different heme reduction rates. Our data support the concept of a common docking site for the FMN module being present on nNOS and eNOS oxygenase domains. A putative docking site has been suggested on the basis of surface homology mapping and is made up of basic and nonpolar surface residues (28,29). Our data also establish that electron transfer from FMN to heme is controlled almost exclusively by structural and/or electronic features inherent in each NOS reductase domain. Conceivably, these could include nonconserved regions both in and away from the FMN module. We are making second generation chimeras to help identify the key structural elements.
Our previous work (13)(14)(15) showed that NO synthesis by nNOS causes significant heme-NO complex buildup, whereas in eNOS very minor heme-NO complex formation is observed. The chimeras help to identify what factors control different heme-NO complex formation in nNOS and eNOS. First, we found that the four enzymes had the same rank order regarding their rates of heme reduction and rates of heme-NO complex formation (nNOS ϭ E ox N red Ͼ Ͼ N ox E red Ͼ eNOS; see Tables III and IV). This is consistent with nNOS single turnover experiments that show its ferric heme binds newly formed NO before releasing it (20) and suggests that this process also occurs for the eNOS heme. The data also provide a direct indication that ferric heme reduction limits the rate of NO synthesis in all four proteins. Indeed, the speeds of biosynthetic steps that follow ferric heme reduction (O 2 binding, transfer of a second electron, chemical transformations, and product release), when measured individually, have all been faster than this initial step (19). The percentage of heme-NO complex observed at steady state also correlated with the rate of ferric heme reduction in all four proteins. The percentages calculated from stopped-flow traces ranged from 12% of total enzyme for eNOS to 65% for nNOS, with the chimeras falling in between. These percentages generally agree with estimates we derived from our steady-state spectra. The difference in chimera heme-NO complex formation can be interpreted based on a recent model we developed for nNOS catalysis (Fig. 5) (19). As the ferric heme reduction rate increases (Fig. 5, kr), the NOS proteins generate NO faster, and the rate of ferric heme-NO complex formation increases as discussed above. However, faster heme reduction also increases the probability that the ferric heme-NO product will become reduced (krЈ) before NO can dissociate (kd). Reduction generates a ferrous heme-NO complex from which NO dissociates very slowly (19). The steady-state level of the ferrous heme-NO species depends on its relative rate of formation versus O 2 -dependent decay (Fig. 5, kox). In nNOS, the different rates are set such that in an air-saturated buffer a majority of enzyme is present as a ferrous heme-NO complex during the steady state. Previous work with nNOS (21, 27) and model simulations (19) shows that the percentage of its ferrous TABLE III Observed rate constants for NADPH-dependent flavin and heme reduction Measurements were done with CaM-bound enzymes at 10°C in a stopped-flow spectrophotometer under anaerobic conditions as described under "Experimental Procedures." Heme reduction was measured in the presence of CO. Rates are the averages obtained with two or three enzyme preparations. The data were best fit to a monophasic rate for heme reduction and a biphasic rate for flavin reduction.  heme-NO complex varies in proportion to the rate of ferric heme reduction (kr, krЈ) within the range of 0 -4 s Ϫ1 . This is precisely what we observed for the chimeras. N ox E red had slower ferric heme reduction than nNOS and displayed less heme-NO complex during the steady state. On the other hand, the faster ferric heme reduction in E ox N red relative to eNOS was associated with greater heme-NO complex accumulation.
Our results with E ox N red are the first to indicate how speeding eNOS ferric heme reduction will affect its catalytic profile. The data predict that eNOS will behave more like nNOS under this circumstance, increasing both its NO synthesis rate and degree of heme-NO complex buildup. Thus, different rates of ferric heme reduction seem to primarily determine the catalytic profiles of eNOS, nNOS, and the chimeras. The different NO synthesis activities of eNOS, nNOS, and the chimeras can best be appreciated when one also considers how heme-NO complex formation affects steady-state NO synthesis. Fast heme reduction caused a majority of nNOS and E ox N red to be present as their ferrous heme-NO complex during steady-state NO synthesis. This slowed down their activities to about one seventh and one fifth of the initial rates, respectively, as inferred from their initial versus steady-state NADPH oxidation rates (Table IV, m 1 and m 2 values). Multiplying their steady-state NO synthesis activities in Table II by factors of seven and five eliminates the effect of enzyme partitioning and provides a better estimate of their true activity (400 and 280 NO/min, respectively, at 25°C). On the other hand, eNOS and N ox E red had minor heme-NO complex formation, and thus their steady-state activities (15-20 NO/min, respectively, at 25°C in Table II) are decent estimates of their intrinsic activities. The analysis suggests that the intrinsic NO synthesis activities of nNOS and E ox N red are actually 26 and 14 times greater than those for eNOS and N ox E red , respectively, instead of the 3-4fold difference indicated by steady-state NO synthesis measurements. This reveals how heme-NO complex formation can blunt intrinsic differences in NOS activity.
In CaM-bound NOS, cytochrome c competes with the NOS oxygenase domain for electrons from the FMN group. However, the estimated differences in intrinsic NO synthesis rates as described above are somewhat greater than our measured rate differences in cytochrome c reduction (11-fold greater for nNOS versus eNOS, and 9-fold greater for E ox N red versus N ox E red ; see Table II). This may reflect some imprecision in the values we used for calculating estimates. Alternatively, it may reflect inherent differences in kinetics or mechanisms that control electron transfer to cytochrome c versus the NOS oxygenase domain. Indeed, chimeras of CaM and cardiac troponin C differentially activate cytochrome c reduction and NOS heme reduction (27). Thus, the two processes diverge in some aspects and cannot be presumed equivalent.
Although heme reduction rate is the major parameter distinguishing catalytic behaviors of eNOS, nNOS, and the chimeras, it is not the sole parameter. Related work indicates that the rate at which O 2 reacts with the ferrous heme-NO complex (Fig. 5, kox) differs between NOS isoforms 4 and is solely a function of the oxygenase domain. This parameter helps to control the proportion of enzyme that is present as a ferrous heme-NO complex during the steady state and becomes more important as the heme reduction rate in the system increases ( Fig. 5, krЈ) (19). For example, a faster kox for eNOS could explain why E ox N red accumulates a somewhat smaller amount of heme-NO complex than nNOS despite their identical steadystate NADPH consumption and NO synthesis rates (see Tables  II and IV). It might also explain why E ox N red and nNOS display equivalent NO synthesis in the steady state although the initial rate of NADPH consumption by E ox N red is only 75% that of nNOS (see Table IV). NOS oxygenase domains also differ in their extent of NADPH-dependent heme reduction, with eNOS being the poorest (7,13,30). This may help explain why N ox -E red is faster than eNOS regarding its initial rate of NADPH oxidation (Table IV) and its steady-state rates of NADPH consumption and NO synthesis (Table II). These and other differences probably also underpin the different rates of H 2 O 2 -promoted NOHA oxidation for eNOS and nNOS in Table I. However, it is apparent from our analysis that oxygenase domain-specific effects only "fine tune" the catalytic behavior of eNOS and nNOS. The most dramatic effects are controlled by the heme reduction rate, which is a function of their reductase domains.
There is an interesting difference between eNOS and nNOS regarding the effect of O 2 on heme reduction. When measured under anaerobic conditions, even in the presence of CO, heme reduction in eNOS is much slower than the rate that can be inferred from the initial rates of NADPH oxidation or NO synthesis in oxygenated buffer (compare Tables III and IV). This discrepancy does not occur in nNOS (19) or in iNOS. 4 We also observed an O 2 effect on heme reduction for N ox E red but not for E ox N red . Thus, the behavior seems specific for the eNOS reductase domain and operates independent of oxygenase domain identity.
To summarize, chimeras of nNOS and eNOS help to show how their reductase and oxygenase domains support different heme reduction, heme-NO complex formation, and NO synthesis. The heme reduction rate is controlled almost exclusively by the reductase domain and is the major parameter controlling heme-NO complex formation and NO synthesis, with oxygenase domains providing minor but measurable influences. Increasing the heme reduction rate in a chimera containing the eNOS oxygenase domain resulted in a catalytic profile approaching nNOS, whereas slowing the heme reduction in a chimera containing the nNOS oxygenase domain resulted in a catalytic profile approaching eNOS. Thus, general principles 4 J. Santolini and D. J. Stuehr, manuscript in preparation.
FIG. 5. Kinetic model for NO synthesis. The reduction of ferric enzyme (Fe 3ϩ ) to ferrous (Fe 2ϩ ) enables O 2 to bind and initiates NO synthesis from Arg. The immediate product of catalysis is the ferric heme-NO complex (Fe 3ϩ -NO), which can either release NO or become reduced to generate a ferrous heme-NO complex. The ferrous heme-NO complex can regenerate active ferric enzyme by reacting with O 2 . kr, kcat, krЈ, kd, and kox represent the rates of each indicated step. Adapted with permission from Refs. 19 and 21. governing heme-NO complex formation and NO synthesis activity in nNOS apply to eNOS as well.