An active site mutation induces oxygen reactivity in D-arginine dehydrogenase: A case of superoxide diverting protons

Enzymes are potent catalysts that increase biochemical reaction rates by several orders of magnitude. Flavoproteins are a class of enzymes whose classification relies on their ability to react with molecular oxygen (O2) during catalysis using ionizable active site residues. Pseudomonas aeruginosa D-arginine dehydrogenase (PaDADH) is a flavoprotein that oxidizes D-arginine for P. aeruginosa survival and biofilm formation. The crystal structure of PaDADH reveals the interaction of the glutamate 246 (E246) side chain with the substrate and at least three other active site residues, establishing a hydrogen bond network in the active site. Additionally, E246 likely ionizes to facilitate substrate binding during PaDADH catalysis. This study aimed to investigate how replacing the E246 residue with leucine affects PaDADH catalysis and its ability to react with O2 using steady-state kinetics coupled with pH profile studies. The data reveal a gain of O2 reactivity in the E246L variant, resulting in a reduced flavin semiquinone species and superoxide (O2•ˉ) during substrate oxidation. The O2•ˉ reacts with active site protons, resulting in an observed nonstoichiometric slope of 1.5 in the enzyme’s log (kcat/Km) pH profile with D-arginine. Adding superoxide dismutase results in an observed correction of the slope to 1.0. This study demonstrates how O2•ˉ can alter the slopes of limbs in the pH profiles of flavin-dependent enzymes and serves as a model for correcting nonstoichiometric slopes in elucidating reaction mechanisms of flavoproteins.

Enzymes are potent catalysts that increase biochemical reaction rates by several orders of magnitude.Flavoproteins are a class of enzymes whose classification relies on their ability to react with molecular oxygen (O 2 ) during catalysis using ionizable active site residues.Pseudomonas aeruginosa D-arginine dehydrogenase (PaDADH) is a flavoprotein that oxidizes Darginine for P. aeruginosa survival and biofilm formation.The crystal structure of PaDADH reveals the interaction of the glutamate 246 (E 246 ) side chain with the substrate and at least three other active site residues, establishing a hydrogen bond network in the active site.Additionally, E 246 likely ionizes to facilitate substrate binding during PaDADH catalysis.This study aimed to investigate how replacing the E 246 residue with leucine affects PaDADH catalysis and its ability to react with O 2 using steady-state kinetics coupled with pH profile studies.The data reveal a gain of O 2 reactivity in the E 246 L variant, resulting in a reduced flavin semiquinone species and superoxide (O 2 ˉ) during substrate oxidation.The O 2 ˉreacts with active site protons, resulting in an observed nonstoichiometric slope of 1.5 in the enzyme's log (k cat /K m ) pH profile with D-arginine.Adding superoxide dismutase results in an observed correction of the slope to 1.0.This study demonstrates how O 2 ˉcan alter the slopes of limbs in the pH profiles of flavin-dependent enzymes and serves as a model for correcting nonstoichiometric slopes in elucidating reaction mechanisms of flavoproteins.
Flavin-dependent enzymes are a class of enzymes that often rely on ionization processes during catalysis (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15).Flavoproteins can be classified as oxidases, monooxygenases, and dehydrogenases, depending on their ability to use molecular oxygen (O 2 ) as an electron acceptor and the products of their reactions (16,17).Oxidases use O 2 as an electron acceptor to produce H 2 O 2 .Monooxygenases insert an oxygen atom from O 2 into the substrate with H 2 O as a product.Dehydrogenases do not react with O 2, or if they do, they reduce O 2 to superoxide radicals (O 2 ˉ) without the production of H 2 O 2 or H 2 O (18).In flavin-dependent enzymes, the spin-forbidden reaction of the triplet state O 2 with the singlet state reduced flavin is overcome by successive electron transfers in a step-wise process to yield a caged O 2 ˉ/flavin semiquinone radical pair (16,17,(19)(20)(21).Studies on flavoproteins show that enzymes that react with O 2 , such as oxidases and monooxygenases, overcome the thermodynamically unfavorable generation of the O 2 ˉ/flavin semiquinone radical pair by stabilizing the transition state of the O 2 ˉ/flavin semiquinone radical pair through electrostatic catalysis using a positive charge close to the flavin C(4a) atom (16,17,(21)(22)(23)(24).
From structural and mechanistic analysis, a structural motif comprising nonpolar residues in the active site and a positive charge, either in the protein or from the enzyme's substrate or product, has been identified as a requirement for flavin reactivity with O 2 to yield the O 2 ˉ/flavin semiquinone radical pair (24).In principle, a flavoprotein can gain the ability to react with O 2 , provided these minimum requirements are met.The highly reactive O 2 ˉcan then undergo an ionization process to acquire a proton from the enzyme or bulk solvent, yielding a hydroperoxyl radical (HO 2 ) with a pK a value of 4.8 (25,26).The key players of ionization processes in several enzymes are solvent water, metal ions, and the side chains of ionizable active site residues (10,16,17,(27)(28)(29)(30)(31)(32)(33)(34)(35)(36).For several decades, pH profile and mutagenesis studies have been widely employed to identify and assign pK a values to the ionizing groups during enzyme catalysis (37)(38)(39)(40)(41)(42)(43).However, the pH profiles used to identify ionizing groups and assign pK a values could be misinterpreted if an enzyme produces highly reactive species such as O 2 ˉthat can react with ionized protons during catalysis.
Hence, the question remains whether the formation and leakage of O 2 ˉin flavoproteins that react poorly with O 2 to yield O 2 ˉcan affect the pH profiles used in assessing the residues important for enzyme catalysis.
PaDADH has broad substrate specificity.The enzyme oxidizes all D-amino acids, except D-aspartate and D-glutamate, to their corresponding a-imino acids, followed by nonenzymatic hydrolysis of the a-imino acid products to yield a-keto acids and ammonia (Fig. 1) (44,(67)(68)(69).PaDADH is a strict dehydrogenase that does not react with O 2 and follows a pingpong bi-bi steady-state kinetic mechanism (44,66).Structural analysis of the enzyme's active site reveals only polar amino acid residues, including H 48 , Y 53 , E 87 , R 222 , E 246 , Y 249 , and R 305 , and a lack of nonpolar residues (69).The enzyme contains four flexible loops, L1, L2, L3, and L4 (69), with residues Y 53 of loop L1 and E 246 of loop L2 involved in a gating interaction that secures the substrate in the active site for catalysis (Fig. 2) (67,69).Residue E 246 of loop L2 is located 7.6 Å from the flavin N5 atom, 8.2 Å from the flavin C(4a) atom, and does not participate in flavin reduction (70).
In this study, the PaDADH residue E 246 has been mutated to leucine to yield the glutamate 246 to leucine mutant (E 246 L) variant enzyme.The effects of the E 246 L mutation on the enzyme's catalysis and ability to react with O 2 have been investigated using rapid-reaction kinetics, steady-state kinetics, and pH profile studies.Additionally, the mutation's effects on the slopes of the pH profile plots have been explored.The resulting stopped-flow traces showed three distinct reaction phases at [D-leucine] ≤10 mM.As expected for flavin reduction, the absorbance at 446 nm decreased; however, there was a subsequent transient increase in the 446 nm absorbance (Fig. 3A).The stopped-flow traces were fit with triple exponentials (Equation 1).

Rapid-reaction kinetics of the
Analysis of the time-resolved UV-visible absorption spectra of the various flavin species generated during the aerobic reduction of the PaDADH E 246 L variant enzyme with Dleucine showed an oxidized flavin spectrum at 0.01 s with a l max at 446 nm and a second peak at 367 nm.After 1.2 s, there  was an observed quenching of the l 446 nm and l 367 nm peaks to yield a spectrum with a l max at 367 nm, consistent with a reduced flavin being present.After 8 s, there was an observed slight increase in the reduced flavin absorbance between 367 nm and 500 nm, although the overall spectral characteristics of the reduced flavin at 1.2 s were maintained.After 120 s, there were three observed peaks in the flavin spectrum at 367 nm, 394 nm, and 480 nm, consistent with the formation of a reduced flavin semiquinone species (Fig. 3B).
The first phase of the time-resolved aerobic reduction of the PaDADH E 246 L variant with D-leucine, characterized by the bleaching of the oxidized flavin absorption at 446 nm, was assigned to flavin reduction.The next phase, characterized by the transient gain of absorbance at 446 nm, was assigned to the imino acid product release.The last phase, showing a quenching of absorbance at 446 nm, was assigned to the reaction of the reduced flavin with O 2 (Fig. 3C To obtain the observed rate constant (k obs1 ) for flavin reduction, the rate constants for the first phase at any given substrate concentration were fit with Equation 2, yielding a zero y-intercept hyperbolic dependence of the k obs1 parameter on D-leucine concentration (Fig. 3D), allowing for the determination of the limiting rate constant for flavin reduction (k red ) and the apparent equilibrium constant for the dissociation of the substrate from the Michaelis complex (K d ).The resulting kinetic parameters are shown in Table 1.There was   The ability of the PaDADH E 246 L variant enzyme to react with O 2 as an electron acceptor was explored by measuring the initial velocities of the enzymatic reaction at various enzyme concentrations with 25 mM D-leucine as substrate and O 2 as the electron acceptor.The data yielded a positive-sloped linear dependence of the initial velocities of enzyme activity as a function of enzyme concentration.Similar results were obtained when the assay was repeated with D-arginine as substrate (Fig. 4A).
A separate experiment to investigate the dependence of the O 2 reactivity on D-arginine concentration yielded a flat line with a rate constant for the oxygen-driven dehydrogenase activity † of 0.23 s −1 , irrespective of the substrate concentration tested (Fig. 4B).When superoxide dismutase was introduced into the reaction assay to determine the O 2 species generated during the enzyme's turnover with O 2 as the electron acceptor, there was an observed decrease in the initial velocity of the enzymatic reaction with D-arginine (Fig. 4C).Similar results were obtained when the above assays were repeated with Dleucine as substrate (data not shown).
The steady-state kinetic studies of the PaDADH E246L variant enzyme reported in a previous study yielded a k cat value of 265 ± 5 s −1 , a k cat /K m value of 871,000 ± 35,000 M −1 s −1 , and a K m value of 0.30 ± 0.02 mM with D-arginine (70).Hence, to determine the effect of the enzyme-generated O 2 ˉon PaDADH E 246 L turnover with D-arginine, assays of the variant enzyme with O 2 as electron acceptor and 0.5 mM D-arginine as substrate were carried out under steady-state conditions.
The PaDADH E 246 L variant underwent multiple turnover cycles spanning 2 h upon reintroduction of D-arginine into the reaction mixture (Fig. 5).For all enzyme turnover cycles, the initial rate of the oxygen-driven dehydrogenase activity was 0.4 s −1 .Similar rates for D-arginine oxidation were observed when the experiment was repeated using 0.  region above pH 9.0, and an observed pK a value for a basic group between 8.2 and 8.8 (Fig. 7 and Table 2).
For the D-arginine substrate, the k cat /K m pH profile at 25 o C fit with Equation 3 yielded a plot with a nonstoichiometric slope of +1.5 for the increasing limb, with an observed pK a value for a basic group of 8.2.When the assay was repeated at 12 o C to investigate the effect of temperature on the slope of the k cat /K m pH profile plot, similar results were obtained, suggesting that temperature does not contribute to the observed slope.When the effect of the enzyme-generated O 2 ōn the slope of the k cat /K m pH profile plot was investigated by adding 200 to 500 units of superoxide dismutase to each apparent steady-state reaction mixture at 25 o C, the k cat /K m pH profile plot fit with Equation 3 yielded a stoichiometric value of +1 (Table 2).
The effect of pH on the k cat /K m pH profile of PaDADH E 246 L with D-leucine at 25 o C fit with Equation 4 yielded a kinetic plot with a slope of +1 for the increasing limb, with an observed pK a value for a basic group of 8.7 (Table 3) and a pH-independent region above pH 9.0 (Fig. 7).The PaDADH E 246 L variant enzyme reacts with O 2 to form a flavin semiquinone during substrate oxidation.Evidence supporting this conclusion comes from the enzyme-monitored aerobic flavin reduction with D-leucine and oxygen (Fig. 3), showing the formation and decay of an enzyme intermediate between 1.2 and 120 s.Analysis of the spectral data of the flavin reduction revealed a flavin spectrum with a peak at 367 nm, characteristic of a flavin semiquinone species at 120 s (Fig. 3) (73)(74)(75).By comparison, this feature is not observed with the PaDADH wildtype enzyme under both aerobic and anaerobic conditions (42,44,66,67,69,76,77).In the same way, the flavin semiquinone species was not observed with the E 246 L variant enzyme under anaerobic conditions, although similar kinetic parameters as reported for the aerobic flavin reduction in this study were observed: k red = 50 s −1 and K d = 12 mM (Table 1) (70).The data are consistent with the PaDADH E 246 L variant reacting with O 2 to yield the flavin semiquinone after flavin reduction and product release, as evidenced by the observed transient increase in the l 466 nm absorbance of the reduced flavin at 8 s (Fig. 3).The presence or absence of the product in the active site alters the flavin oscillator strength, yielding different absorbance intensities of the flavin 446 nm peak upon product release (78).The observation that the flavin semiquinone was not formed at [D-    HO 2 ˉ, HOˉ, and O 2 (Fig. 6) (vide supra).Moreover, the conversion of O 2 ˉto HO 2 ˉfollowing a hydrogen transfer from the flavin semiquinone during the O 2 -driven dehydrogenase mechanism cannot be ruled out.

Discussion
The O 2 -driven dehydrogenase activity of PaDADH E 246 L with a rate of 0.23 s −1 , irrespective of the substrate and concentration tested, can be explained as a likely saturation of the PaDADH E 246 L variant with the amino acid substrate during turnover with O 2 as the electron acceptor.The data agree well with a recent report investigating the role of the E 246 residue in PaDADH catalysis in which a small O 2 -driven dehydrogenase activity of 0.2 s −1 was reported for the E 246 L, E 246 G, and E 246 Q variant enzymes (70).In the same study, the kinetic parameters were investigated using PMS as an artificial electron acceptor for PaDADH since the physiological electron acceptor is unknown.The study reported a k cat value of 270 s −1 with PMS and a maximum rate of 0.2 s −1 with O 2 for the E 246 L variant enzyme with D-arginine (70).In this way, the previous study demonstrates that the PaDADH E 246 L variant enzyme turns over primarily through a PMS-driven dehydrogenase mechanism, not an O 2 -driven dehydrogenase mechanism (Fig. 8).Hence, the O 2 -driven dehydrogenase activity does not affect the PMS-driven dehydrogenase activity of PaDADH E 246 L.
The E 246 L variant-generated O 2 ˉdiverts protons from the active site during D-arginine oxidation.This conclusion is supported by the steady-state pH profiles of the PaDADH E 246 L variant with D-leucine and D-arginine in the presence and absence of superoxide dismutase (Fig. 7).The pH profile of the k cat /K m parameter with D-leucine yielded a plot with a +1 slope for the increasing limb from low to high pH.Conversely, with D-arginine, the k cat /K m pH profile yielded a nonstoichiometric slope of +1.5 that was corrected to +1 after superoxide dismutase addition.Conventionally, the slope of a kinetic parameter's pH profile has been used as an indicator for the number of ionizable processes required to complete the catalytic processes probed by the kinetic parameter (10-14, 89-102).Thus, the observed differences in the PaDADH E 246 L k cat /K m pH profile slope with D-arginine and D-leucine can be explained in light of different ionization processes being essential for the catalysis of the two substrates.
With D-arginine, the slope of +1 for the log (k cat /K m ) pH profile is assigned to the ionization of the substrate's a-NH 3 + during enzyme catalysis.The observed slope of +1 compared to the slope of +2 previously reported for the wildtype enzyme suggests that the unprotonated E 246 residue is important for  binding D-arginine.However, in a previous study investigating the role of residue E 87 in PaDADH, the observation of only a single ionizable group with a slope of +1 in the PaDADH E 87 L variant led to the assignment of residue E 87 as one of the two unprotonated groups required for D-arginine binding, with the substrate's a-NH 3 + group being the other (77).Altogether, the data portray the requirement for three groups: the substrate's a- 87 , and E 246 for the k cat /K m kinetic parameter during Darginine oxidation.With the substrate being the commonality between all enzyme forms under discussion: wildtype, E 87 variant, and E 246 variant, the first ionizable group can be unequivocally assigned to the substrate's a-NH 3 + group.Thus, given that the wildtype enzyme reports only two ionizable groups for the k cat /K m pH profile, the assignment of the second group must be critically assessed.From the PaDADH crystal structure (Fig. 9), both E 87 and E 246 interact with the guanidinium side chain and are important for D-arginine binding and imino arginine release (70,77).Since the guanidinium charge is delocalized, we propose that the second ionizable group reflects not one but the joint ionization of the E 87 and E 246 residues to ensure maximal binding of D-arginine in the wildtype enzyme.Thus, mutation of either E 87 or E 246 yields similar and nonadditive effects on the log (k cat /K m ) pH profile as reported for His 256 and Asp 266 of the carbohydrate-binding domain in rat hepatic lectin-1 (103).
The correction of the log (k cat /K m ) pH profile's slope from +1.In contrast, the D-leucine substrate has a short and nonpolar side chain and does not require ionization of the E 87 residue for binding.Thus, there is no observed effect of O 2 ōn the log (k cat /K m ) pH profile plot.The observed slope of +1 with an intrinsic pK a value of 8.5 and common to both Dleucine and D-arginine corresponds to the ionization of the substrate's a-NH 3 + group that initiates hydride transfer for amine oxidation (42,77).group renders the amino acid substrate or imino acid product with a net positive charge in the ligand-bound state (Fig. 9).Nonetheless, in the free form of the enzyme, which is the form that reacts with O 2 , the positive charge likely arises from either R 222 or R 305 .However, the wildtype enzyme contains only polar amino acid residues, negating one requirement for O 2 reactivity (66

Experimental procedures Materials
Escherichia coli strain Rosetta(DE3)pLysS was purchased from Novagen.The QIAprep Spin Miniprep Kit and the QIAquick Polymerase Chain Reaction Purification Kit were obtained from Qiagen.Pfu DNA polymerase was purchased from Stratagene, and DpnI was obtained from New England BioLabs.Oligonucleotides were purchased from Sigma Genosys for sitedirected mutagenesis and sequencing of the variant genes.Bovine liver superoxide dismutase and PMS were from MilliporeSigma.D-Amino acids were obtained from Alfa-Aesar.All other reagents used were obtained at the highest purity commercially available.

Site-directed mutagenesis, protein expression, and purification
The E 246 L variant gene of PaDADH was engineered by mutagenic polymerase chain reaction (PCR) with the pET20b(+)/PA3863 plasmid harboring the wildtype gene (dauA) as a template.A concentration of 5% dimethyl sulfoxide was added to the PCR reaction mixture to ensure proper separation of the high GC-rich, double-stranded DNA template.Site-directed mutagenesis amplicons were purified using the QIAquick PCR Purification Kit following the manufacturer's protocol.The purified plasmid was then subjected to endonuclease activity using DpnI at 37 o C for 2 h.The resulting plasmid was used to transform the DH5a strain of E. coli cells.The success of the mutation was confirmed by sequencing the gene using the services of Humanizing Genomics Microgen USA Corp in Maryland.The E 246 L variant enzyme of PaDADH was then expressed in E. coli Rosetta(DE3)pLysS and purified to homogeneity as previously described for the PaDADH wildtype enzyme in the presence of 10% (v/v) glycerol for enzyme stability and to prevent the loss of the bound FAD cofactor (44).The purified enzyme was stored at −20 o C in 20 mM Tris-Cl, pH 8.0, and 10% glycerol and was found to be active for at least 6 months.

Rapid-reaction kinetics of the PaDADH E 246 L variant enzyme
To establish the effect of the E 246 L mutation on the rapidreaction kinetic parameters of the PaDADH E 246 L variant enzyme and whether the mutation resulted in an acquired reactivity of the enzyme with O 2 , the reductive half-reaction of the PaDADH E 246 L variant enzyme was carried out under aerobic conditions.Using an SF-61DX2 Hi-Tech KinetAsyst performance stopped-flow spectrophotometer, the reaction was followed in 20 mM NaPP i , pH 10.0, and 25 o C and compared to the anaerobic reductive half-reactions of the wildtype and E 246 L variant enzymes that were previously investigated (42,70).Since 80% of flavin reduction occurs in the mixing time (2.2 ms) of the stopped-flow spectrophotometer with D-arginine as a substrate (42,71), the flavin reduction of the E 246 L variant enzyme was investigated with D-leucine as the reducing substrate.Substrate solutions (1-40 mM) were loaded into syringes and mounted onto the stopped-flow spectrophotometer.The reaction was followed by observing the spectroscopic decay of the 446 nm flavin peak over time upon reacting 10 mM enzyme with substrate solutions under pseudo-first-order conditions in single mixing mode.

Data analysis
The time-resolved flavin reductions were fit to Equation 1 amplitudes of the absorption changes, t is time, and C is the absorbance at an infinite time that accounts for the nonzero absorbance of the fully reduced enzyme-bound flavin.
The resulting kinetic parameters of the reductive halfreaction were determined after fitting the observed rate constants for flavin reduction at various D-leucine concentrations with Equation 2. The equation defines a hyperbolic saturation of the enzyme with the D-leucine substrate, yielding a yintercept value of zero.The data were fit with the Kaleida-Graph software (Synergy Software).Here, k obs1 represents the observed first-order rate constant for reducing the enzymebound flavin at any substrate concentration (S).k red is the limiting first-order rate constant for flavin reduction at saturating substrate concentrations.K d is the apparent equilibrium constant for dissociating the enzyme-substrate complex into the free substrate and enzyme.The same data were obtained when an equation that defines a hyperbolic saturation with a finite y-intercept was used.
O 2 reactivity studies of the PaDADH E 246 L variant enzyme To investigate the effect of the E 246 L mutation on the ability of the PaDADH E 246 L variant enzyme to react with O 2 under steady-state conditions, the initial velocities of the enzyme reaction were measured with D-leucine as a substrate and O 2 as an electron acceptor at pH 8.5.The reduction of O 2 was followed using a Clark-type oxygen electrode in a 1 ml reaction volume containing final enzyme concentrations of 0.48 mM to 9.7 mM and fixed D-leucine concentration at 25 mM in 20 mM NaPP i at 25 o C. Substrate solutions were prepared in the reaction buffer, and the pH was readjusted after the amino acid substrate was dissolved.The experiment was repeated with Darginine as substrate at pH 8.5, in a 1 ml reaction volume containing final enzyme concentrations of 0.17 To determine the effects of pH on the steady-state kinetics of the E 246 L variant enzyme of PaDADH, the apparent steady-state kinetic parameters of the enzyme with D-arginine or D-leucine as a substrate and PMS as an artificial electron acceptor were obtained by monitoring the initial PMS-driven O 2 consumption rates with a computer-interfaced Oxy-32 oxygen-monitoring system (Hansatech Instruments Ltd) under similar conditions.D-Arginine concentrations were between 0.1 and 80 mM, Dleucine concentrations were between 1.25 and 62.5 mM, the enzyme concentration ranged from 1.15 mM to 5.75 mM, PMS concentration was fixed at 1 mM, and the pH ranged from 5.0 to 10.5 at 25 o C. Temperature effects on the steady-state pH profiles for D-arginine were investigated by repeating the assays at 12 o C. All assays were carried out to ensure that the K m value was within the range of the substrate concentrations used at each pH.To ensure that the variant enzyme was fully saturated with PMS, the steady-state kinetic parameters were also determined at 1.5 mM PMS, yielding similar results.
Due to the observation of a nonstoichiometric slope of +1.5 for the log (k cat /K m ) pH profile with D-arginine at 12 o C and 25 o C, the assay was repeated with 200 to 500 units of superoxide dismutase in each apparent steady-state reaction mixture to investigate the effect of the enzyme-O 2 reactivity on the Darginine pH profile.The data analyses and interpretation focused only on the k cat /K m pH profiles due to the observation that superoxide dismutase affected only the k cat /K m pH profiles with D-arginine.
For the D-arginine substrate, the log values of the k cat /K m parameters under varying conditions of temperature and superoxide dismutase were plotted by fitting the log values of the k cat /K m parameters with Equation 3. The equation describes a curve that increases with increasing pH with a slope of S and a pH-independent limiting value (C H ) at high pH.
The plot of the k cat /K m parameter for the D-leucine substrate was made by fitting the log values with Equation 4, which describes a curve that increases with increasing pH with a slope of +1 and a pH-independent limiting value (C H ) at high pH.
PaDADH E 246 L variant enzyme with D-leucine as a substrate Since 80% of the flavin reduction of the PaDADH E 246 L variant enzyme with D-arginine occurs in the mixing time (2.2 ms) of the stopped-flow spectrophotometer (42, 71), the flavin reduction of the enzyme was investigated with D-leucine as the reducing substrate.The time-resolved aerobic reduction of the PaDADH E 246 L variant enzyme by D-leucine was investigated under pseudo-first-order conditions by monitoring the loss of the oxidized flavin's absorbance at 446 nm.

Figure 2 .
Figure 2. Gating of PaDADH active site by loop L1 and L2 residues Y 53 and E 246 .Y 53 and E 246 are shown in gray.All N atoms are shown in blue, and all O atoms are in red.The FAD cofactor is represented by its isoalloxazine ring with the C atoms in gold.IAR represents the iminoarginine product and is shown in cyan.The E 246 residue's hydrogen bond interaction with Y 53 and distance from the flavin are shown as dashed lines.Loop L2 is shown in coral, and loop L1 is shown in green.The PDB file 3NYE was visualized and analyzed using the UCSF Chimera software (110).PaDADH, Pseudomonas aeruginosa D-arginine dehydrogenase.
), hinting at the formation of a caged O 2 ˉ/flavin semiquinone radical pair during the aerobic reduction of the PaDADH E 246 L variant enzyme.The lack of a transient increase in the 446 nm absorbance at [D-leucine] ≥10 mM can be explained as a likely formation of enzyme-substrate complexes between the reduced enzyme and excess substrate following the imino acid product release, which prevents the reduced flavin from reacting with O 2 .

Figure 3 .
Figure 3. Aerobic reductive-half reaction of the PaDADH E 246 L variant enzyme.A, stopped-flow traces of the absorbance changes at 446 nm at different D-leucine concentrations (1-25 mM) fit with Equation 1.Each trace is the average of triplicate runs at each substrate concentration.For clarity, one out of every 100 experimental points is shown (vertical lines).Note the log time scale.B, time-resolved UV-visible absorption spectra of the various flavin species generated during the aerobic reduction of the PaDADH E 246 L variant enzyme with D-leucine.The reaction was monitored over 120 s (s) upon mixing 1 mM D-leucine with the E 246 L variant enzyme of PaDADH in the presence of atmospheric O 2 .The black spectrum represents the oxidized enzyme; the red spectrum represents the reduced flavin semiquinone; the green and the blue spectra represent the fully reduced flavin.C, time map for the various flavin species generated during the aerobic reductive-half reaction of the PaDADH E 246 L variant enzyme with 1 mM D-leucine.D, the observed rate constant for flavin reduction as a function of D-leucine concentration under aerobic conditions fit with Equation 2. The single point shown at each substrate concentration is the k obs value obtained from the fit of the average of triplicate runs with Equation 1 yielding an error of ≤5%.The assay was performed in 20 mM NaPP i , pH 10.0, using an SF-61DX2 Hi-Tech KinetAsyst high-performance stopped-flow spectrophotometer thermostated at 25 o C and equipped with a photomultiplier detector under aerobic conditions.The instrumental dead time is 2.2 ms.E246L, glutamate 246 to leucine mutant; PaDADH, Pseudomonas aeruginosa D-arginine dehydrogenase.

75 a
k red /K d (M −1 s −1 ) 4600 ± 400 2900 ± Reductive-half reaction kinetics were measured at varying concentrations of Dleucine under aerobic conditions.Assays were performed in 20 mM NaPP i , pH 10.0, at 25 o C. The kinetic parameters' values were obtained after fitting the kinetic data with Equation 2. b Previously reported data for the PaDADH E 246 L variant enzyme under anaerobic conditions (70).no observed dependence of the k obs2 and k obs3 parameters on [D-leucine].O 2 reactivity studies of the PaDADH E 246 L variant enzyme 15 and 0.24 mM Darginine (data not shown).Additionally, O 2 regeneration was observed in every enzyme turnover cycle before the D-arginine substrate was reintroduced (Fig. 5).The observed oxygen regeneration can be explained by a proton-dependent O 2 ˉdisproportionation to yield hydroperoxide anion HO 2 ˉ, hydroxide HOˉ, and O 2 (Fig. 6) (72), followed by a slow re-equilibration of the solution in the O 2 electrode chamber toward atmospheric oxygen.The observed nonzero O 2 level after enzyme turnover with Darginine can be explained by the O 2 ˉdisproportionation reaction leading to O 2 accumulation that gradually overturns the O 2 consumption of the D-arginine oxidation.Effects of pH on the k cat /K m parameter of PaDADH E 246 L Due to the O 2 reactivity and formation of the flavin semiquinone species at [D-leucine] ≤10 mM during the aerobic reduction of the PaDADH E 246 L variant, the kinetic investigations for the steady-state pH effects focused only on the k cat /K m parameter, which reports on the enzyme's behavior at low substrate concentrations and probes the free enzyme.Thus, to understand the effects of pH on PaDADH E 246 L's substrate capture, the steady-state reactions of the enzyme were investigated with D-arginine or D-leucine as the substrate from pH 5.0 to 10.5, with phenazine methosulfate (PMS) as an artificial electron acceptor since the physiological electron acceptor for PaDADH activity is not known.The plots of the log values of the k cat /K m parameter showed an increase in the k cat /K m parameter with increasing pH, a pH-independent

Figure 4 .
Figure 4. O 2 reactivity studies of the PaDADH E 246 L variant enzyme.A, plot of the initial velocity of the PaDADH E 246 L variant enzyme's reactivity with O 2 as a function of enzyme concentration at fixed 5 mM D-arginine in black and 25 mM D-leucine in purple .B, the dependence of the rate of O 2 reactivity of the PaDADH E 246 L variant enzyme as a function of D-arginine concentration.C, the effect of superoxide dismutase on the PaDADH E 246 L variant enzyme's reaction with O 2 with 5 mM D-arginine as substrate.The black trace represents the O 2 reactivity without superoxide dismutase.The blue trace represents the experiment performed in the presence of superoxide dismutase.The assays were carried out in 20 mM NaPP i , pH 8.0, using a Clark-type oxygen electrode system, thermostated at 25 o C. E246L, glutamate 246 to leucine mutant; PaDADH, Pseudomonas aeruginosa D-arginine dehydrogenase.
This study aimed to investigate the effects of the E 246 L mutation on the ability of PaDADH to react with O 2 .The study also investigated the effect of the E 246 L mutation on the catalysis of PaDADH using steady-state kinetics coupled with pH profile studies.The data demonstrate that upon the E 246 L mutation, PaDADH, which is a strict dehydrogenase that does not react with O 2 , gains the ability to react with O 2 , although poorly, to yield a reduced flavin semiquinone and O 2 ˉ(66, 67).Consequently, the PaDADH E 246 L variant turns over with O 2 as an electron acceptor through an alternative dehydrogenase pathway.Following the acquired O 2 reactivity, the variant enzyme yields a nonstoichiometric slope in the plot of the log (k cat /K m ) parameter as a function of pH with D-arginine as substrate.Details on the gain of function and the implications on PaDADH catalysis are discussed below.

Figure 5 .
Figure 5.Effect of the PaDADH E 246 L-generated O 2 ˉon the turnover ability of the enzyme.O 2 consumption and regeneration cycles were carried out with fixed 6 mM enzyme and 0.5 mM D-arginine.The steady-state regions are shown in black dotted lines and the initial rates of D-arginine oxidation for each turnover cycle is shown below as v o /e.The assay was performed in 20 mM NaPP i , pH 8.0, with O 2 as the electron acceptor, using a Clark-type oxygen electrode system at 25 o C. E246L, glutamate 246 to leucine mutant; PaDADH, Pseudomonas aeruginosa D-arginine dehydrogenase.

Figure 7 .
Figure 7. Effects of superoxide dismutase and pH on the k cat /K m parameter of the PaDADH E 246 L variant with D-arginine or D-leucine as substrate.A, pH dependence of the k cat /K m parameter with D-arginine.B, pH dependence of the k cat /K m parameter with D-leucine.Activity assays were carried out with varying concentrations of D-arginine or D-leucine as a substrate and fixed PMS as an artificial electron acceptor at 1 mM from pH 5.0 to 10.5 in 20 mM NaPP i .Assays without superoxide dismutase are shown in black and red for D-arginine at 25 o C and 12 o C, respectively, and purple for D-leucine at 25 o C. The D-arginine assay with superoxide dismutase at 25 o C is shown in blue .The D-arginine plots were obtained by fitting the kinetic data to Equation 3 for the k cat /K m parameter without superoxide dismutase at 12 o C and 25 o C, and Equation 4 for the k cat /K m parameter with superoxide dismutase.The Dleucine plot was obtained by fitting the kinetic data with Equation 4. The observed pK a values for the various assays are recorded in Table 2 for D-arginine and Table 3 for D-leucine.PaDADH, Pseudomonas aeruginosa D-arginine dehydrogenase; PMS, phenazine methosulfate.

Figure 8 .
Figure 8. Proposed reaction scheme of the PaDADH E246L dehydrogenase activity with PMS or O2 as an electron acceptor during turnover.A hydride transfer from the amino acid substrate reduces the enzyme-bound oxidized flavin to yield the reduced flavin and the imino acid product.The reduced flavin is then re-oxidized through a PMS-driven or an O2-driven dehydrogenase activity.For the PMS-driven turnover, the reduced flavin is reoxidized by PMS to yield PMSH2, restoring the enzyme to its resting state.For the O2-driven turnover, the reactivity of O2 with the reduced flavin yields the highly reactive caged O2ˉ/flavin semiquinone radical pair.The flavin semiquinone then donates a hydrogen atom to the O2ˉ, yielding HO2ˉand the re-oxidized flavin in the resting state of the enzyme.E246L, glutamate 246 to leucine mutant; O2ˉ, superoxide radicals; PMS, phenazine methosulfate.
5 to +1 following superoxide dismutase addition suggests that the O 2 ˉproduced by PaDADH E 246 L's reaction with D-arginine and O 2 is responsible for the observed nonstoichiometric slope.As the enzyme turns over with D-arginine, the highly reactive O 2 ˉgenerated during flavin reduction progressively accumulates in the solution due to multiple enzyme turnovers.However, during PaDADH E 246 L oxidation of D-arginine, the proton released from the a-NH 3 + ionization is unavailable for O 2 ˉreactivity due to the H 48 -mediated proton relay network to the bulk solvent (Fig. 10), as previously described for p-hydroxybenzoate hydroxylase (66, 77, 104).Thus, the accumulated and highly reactive O 2 ˉmost likely reacts with the E 87 proton to yield HO 2 during PaDADH E 246 L turnover with low concentrations of D-arginine, thereby favoring E 87 ionization.The observed steeper and nonstoichiometric slope of +1.5 (Fig. 7) in the log (k cat /K m ) pH profile, therefore, likely reflects inflated values for the k cat /K m parameter, which probes the free forms of enzymes at low substrate concentrations rather than enzyme-substrate complexes.Due to the O 2 ˉdiversion of the E 87 ionized protons, there is a partial detection of the E 87 ionization in the log (k cat / K m ) pH profile with D-arginine despite the absence of E 246 (103).
The acquired O 2 reactivity of the E 246 L variant yielding a caged O 2 ˉ/flavin semiquinone radical pair can be explained by the modified active site topology of the E 246 L variant.Studies on flavoproteins demonstrate that for an enzyme to react with O 2 , there must be a nonpolar residue and a positive charge close to the flavin cofactor to favor the electrostatics required for O 2 reactivity (18, 20, 21, 81, 83).In PaDADH, the interaction between the R 222 /R 305 network and the a-carboxylate

Figure 9 .
Figure 9.The active site topology of PaDADH showing substrate interactions and the highly polar active site pocket.The PDB file 3NYE was visualized and analyzed using the UCSF Chimera software (110).IAR, iminoarginine product; PaDADH, Pseudomonas aeruginosa D-arginine dehydrogenase.
, which describes a triple exponential process for flavin reduction.Here, k obs1 , k obs2 , and k obs3 represent the observed firstorder rate constant for reducing the enzyme-bound flavin at any given substrate concentration at 446 nm.A represents the absorbance at 446 nm at any given time, B 1 , B 2 , and B 3 are the PaDADH E 246 L reacts with O 2 to produce O 2 J .Biol.Chem.(2024) 300(6) 107381 9 mM to 8.5 mM and fixed D-arginine concentration at 5 mM in 20 mM NaPP i at 25 o C. In a separate experiment, the dependence of the PaDADH E 246 L variant enzyme's O 2 reactivity on substrate concentration was investigated at fixed 0.5 mM enzyme and varying concentrations of D-arginine (1 mM -20 mM) or D-leucine (1 mM-10 mM).To probe the oxygen species generated by the E 246 L variant O 2 reaction, the enzymatic assay was carried out with 6 mM E 246 L variant enzyme and fixed 5 mM D-arginine or 25 mM D-leucine.The reaction was then repeated by adding 200 to 500 units of superoxide dismutase to the reaction mixture.To investigate the effect of the PaDADH E 246 L-generated O 2 ˉon the enzyme's ability to turnover with D-arginine, the steady-state kinetic properties of the variant enzyme with O 2 as an electron acceptor were investigated using 6 mM enzyme with 0.15 mM, 0.24 mM, or 0.5 mM D-arginine.Upon reaching a plateau in the reaction cycle, D-arginine was reintroduced to the reaction mixture to yield a total of four reaction cycles.pH effects on the steady-state kinetics of the PaDADH E 246 L variant enzyme log Y ¼ log C H ð1þ 10 −pH 10 −pK b Þ (Eq 4)PaDADH E 246 L reacts with O 2 to produce O 2 1

Table 1
Rapid-reaction kinetic parameters of the PaDADH E 246 L variant enzyme with D-leucine as substrate

Table 2
Effects of superoxide dismutase on the pH effects on the steady-state kinetic parameters of the PaDADH E 246 L variant enzyme with D-arginine as substrate a a Enzymatic activities were measured at varying concentrations of D-arginine and fixed 1 mM PMS.Reactions were carried out in 20 mM sodium pyrophosphate.The k cat /K m parameter values were obtained after fitting the kinetic data with Equation3.PaDADH E 246 L reacts with O 2 to produce O 2 6

Table 3
Effects of pH on the steady-state kinetic parameters of the PaDADH E 246 L variant enzyme with D-leucine as substrate at 25 o C a Enzymatic activities were measured at varying concentrations of D-leucine and fixed 1 mM PMS.Reactions were carried out in 20 mM sodium pyrophosphate.The k cat / K m parameter values were obtained after fitting the kinetic data with Equation 4. a (105)(106)(107)(108)esence of the nonpolar L246in PaDADH E 246 L provides a suitable active site topology for O 2 reactivity, having a positive charge and a nonpolar residue(24).Thus, the E 246 L variant enzyme gains the ability to react with O 2 .Similarly, PaDADH reactivity with O 2 has been recently reported for the Y 249 F variant enzyme after replacing the polar tyrosine 249 residue with the nonpolar phenylalanine residue in the active site(105,106).However, unlike this study, the Y 249 F variant's O 2 reactivity yielded either a flavin N 5 adduct or a green 6-OH-FAD species.Nonetheless, these studies exemplify how substrates and protein residues dictate the versatility of flavin reactivity to favor specific reactions(105)(106)(107)(108).In conclusion, this study has used steady-state and rapidreaction kinetic approaches to investigate the effects of the E 246 L mutation on the ability of PaDADH to react with O 2 and the role of the generated O 2 ˉin the catalysis of the PaDADH E 246 L variant enzyme.The study demonstrates a mutation-induced gain of PaDADH reactivity with O 2 .pH studies on the steady-state kinetic parameters of the variant enzyme demonstrate that the O 2 ˉgenerated during