Article pubs.acs.org/biochemistry
Kinetic Mechanism and Intrinsic Rate Constants for the Reaction of a Bacterial Phenylalanine Hydroxylase Bishnu P. Subedi and Paul F. Fitzpatrick* Department of Biochemistry, University of Texas Health Science Center at San Antonio, San Antonio Texas 78229, United States S Supporting Information *
ABSTRACT: The pterin-dependent aromatic amino acid hydroxylases are non-heme iron enzymes that catalyze the hydroxylation of the aromatic side chain of their respective substrates using an FeIVO intermediate. While the eukaryotic enzymes are homotetramers with complex regulatory properties, bacterial phenylalanine hydroxylases are monomers that lack regulatory domains. As a result, the bacterial enzymes are more tractable for mechanistic studies. Using single turnover methods, the complete kinetic mechanism and intrinsic rate constants for Chromobacterium violaceum phenylalanine hydroxylase have been determined with both tetrahydrobiopterin and 6methyltetrahyropterin as substrates. In addition the kinetics of formation of the enzyme−pterin complex have been determined with the unreactive 5-deaza, 6-methyltetrahydropterin. For all three pterins, binding of phenylalanine and pterin occurs in random order with binding of the pterin first the preferred pathway. The reaction of the ternary enzyme−phenylalanine− tetrahydropterin complex can be described by a mechanism involving reversible oxygen binding, formation of an early intermediate preceding formation of the FeIVO, and rate-limiting product release.
P
substrate.13−15 The HO-pterin dehydrates in solution to form a quinonoid dihyropterin.16 The eukaryotic AAHs are homotetramers, with each monomer containing discrete regulatory, catalytic, and tetramerization domains. The presence of the regulatory domains contributes to the complex regulatory properties of these enzymes.17 In contrast, PheHs from prokaryotes such as Chromobacterium violaceum (CvPheH) are monomers with a single catalytic domain18 that have the same structure as the catalytic domains of the eukaryotic AAHs4 and lack the complex regulatory properties of liver PheH.17,19 As a result the bacterial PheHs provide a simpler system for detailed studies of catalysis by this family of non-heme iron enzymes. Here we describe a comprehensive analysis of the kinetics of enzyme− substrate binding and of catalysis with several tetrahydropterins using a bacterial PheH.
henylalanine hydroxylase (PheH) is an iron-dependent non-heme mononuclear enzyme that hydroxylates the aromatic amino acid phenylalanine to form tyrosine using tetrahydrobiopterin as the reducing substrate (Scheme 1).1,2
Scheme 1. Reaction Catalyzed by Phenylalanine Hydroxylase
■
This reaction is essential for production of tyrosine in the body as well as catabolism of excess dietary phenylalanine. PheH belongs to the family of aromatic amino acid hydroxylases (AAHs) that also includes tyrosine hydroxylase (TyrH) and tryptophan hydroxylase (TrpH). All three AAHs have similar active sites in which the iron is coordinated by two histidines and one glutamate,3−5 a facial triad motif present in many nonheme mononuclear iron enzymes.6 The present understanding of the chemical mechanism of the AAHs is shown in Scheme 2.7,8 The ferrous form of the enzyme reacts with molecular oxygen and a tetrahydropterin to yield a 4ahydroxypterin and a FeIVO intermediate.9−12 Aromatic amino acid hydroxylation occurs via an electrophilic aromatic substitution mechanism as the highly reactive ferryl-oxo species undergoes an electrophilic attack by the aromatic ring of the © 2016 American Chemical Society
MATERIALS AND METHODS Reagents. Tetrahydrobiopterin (BH4) and 6-methyltetrahydropterin (6MPH4) were purchased from Schircks Laboratories (Jona, Switzerland). 5-Deaza, 6-methyltetrahydropterin (5dMPH4) was synthesized using the method of Moad et al.20 Ampicillin and isopropyl β-D-1-thiogalactopyranoside were from Research Products International Corp. (Mount Prospect, IL). Lysozyme was from MP Biomedicals, LLC (Santa Ana, CA). All other major chemicals/media were purchased from Sigma-Aldrich Co. LLC (St. Louis, MO) or Thermo Fisher Received: October 4, 2016 Revised: November 15, 2016 Published: November 16, 2016 6848
DOI: 10.1021/acs.biochem.6b01012 Biochemistry 2016, 55, 6848−6857
Article
Biochemistry Scheme 2. Catalytic Mechanism of Phenylalanine Hydroxylase
Data Analysis. KaleidaGraph 4.5 (Synergy) was used for initial kinetic analyses using the equations indicated in the text. Global fitting of all kinetic data and construction of mechanistic pathways was done using KinTek Explorer Pro (KinTek Corp, Austin, TX).24 To increase the speed of the analyses, only 20% of the points out of 400 in each stopped-flow trace were used. Analyses using all 400 points showed that this did not affect the results. Calculation of confidence intervals was done using FitSpace Explorer25 with a typical X2 threshold of 1.1, that is, the amount a value can be changed until refitting of the entire data set increases the X2 value by more than 10%.
Scientific, Inc. (Waltham, MA). Expression of CvPheH in Escherichia coli and purification of the enzyme were carried out as previously reported.21 Apo CvPheH was prepared by incubating 100 μM enzyme with 5 mM EDTA and 1 mM nitrilotriacetic acid in 20 mM NaCl, 50 mM HEPES (pH 7.2) on ice for 2 h. The chelators were removed by dialysis against the same buffer without chelators for 36 h. The iron content was determined using o-phenanthroline as described previously.22 Rapid-Reaction Kinetics. Stopped-flow kinetic measurements were carried out using an Applied Photophysics SX-20 stopped-flow spectrophotometer (Leatherhead, UK) capable of both single and sequential mixing. All data were collected at 5 °C. Data collection was done with a logarithmic time scale to allow reliable analyses of multiple phases. The instrument was made anaerobic by overnight incubation with glucose and glucose oxidase as previously described.23 For a typical experiment, a solution of apoenzyme (∼100 or ∼200 μM for single or double mixing, respectively) in 20 mM NaCl, 50 mM HEPES (pH 7.2) was placed in a tonometer. Two equivalents of a ferrous ammonium sulfate solution (0.5 mM in 1 M HCl) was placed in the side arm. The contents of the tonometer were made anaerobic using vacuum-argon cycles; argon gas was passed through an activated copper catalyst to remove residual oxygen. Solutions of pterin and/or phenylalanine were prepared by bubbling argon directly into each solution in a Hamilton gastight syringe. Solutions containing different concentrations of oxygen were prepared by bubbling with mixtures of N2 and O2 from a MaxBlend oxygen mixer. The O2 concentrations were determined using a Yellow Springs Instrument model 5300 oxygen monitor. Chemical quench experiments were carried out using a BioLogic QFM-400 quench-flow instrument (Claix, France). Oxygen was removed from the instrument using the same procedure as for the stopped-flow apparatus. In a typical experiment, an aerobic mixture of 40 μM CvPheH plus 80 μM ferrous ammmonium sulfate and 2 mM tetrahydropterin was rapidly mixed with 4 mM phenylalanine in oxygen-saturated buffer followed by quenching with 2 M HCl. The amount of tyrosine formed at each time point was determined by HPLC on a Phenomenex Gemini-NX column with fluorescence detection (excitation 275 nm/emission 303 nm).22
■
RESULTS Absorbance Changes upon Phenylalanine and Tetrahydropterin Binding. The binding of phenylalanine and several tetrahydropterins to CvPheH was followed using anaerobic stopped-flow spectroscopy at 5 °C. An anaerobic protein solution in one syringe was rapidly mixed with anaerobic substrate(s) from another syringe. The absorbance was monitored between 240 and 400 nm. Binding of 6methyltetrahydropterin (6MPH4) to the enzyme resulted in a shift in the near-UV absorbance over about 100 ms, with a maximum change at 330−340 nm (Figures 1 and S1). The absorbance increased further in the presence of phenylalanine (Phe) to form the ternary CvPheH−6MPH4−Phe complex.
Figure 1. Absorbance changes upon mixing 100 μM CvPheH with 0.5 mM phenylalanine (- - -), 0.5 mM 6MPH4, ( ), or 0.5 mM phenylalanine plus 0.5 mM 6MPH4 (). All concentrations are after mixing. 6849
DOI: 10.1021/acs.biochem.6b01012 Biochemistry 2016, 55, 6848−6857
Article
Biochemistry Addition of phenylalanine alone to the enzyme did not result in any detectable absorbance change. The wavelength at which the absorbance changes were greatest, 340 nm, was selected for further analyses of the binding kinetics. The results with BH4 were qualitatively similar to those with 6MPH4, but the absorbance changes with this pterin were about twice those seen with 6MPH4. In contrast, with the redox-inactive 5-deaza6-methyltetrahydropterin (5dMPH4), there was no detectable absorbance change upon binding of the pterin alone to the enzyme, whereas binding of both phenylalanine and 5dMPH4 did result in a change in absorbance similar to that seen with 6MPH4 plus phenylalanine, but with about half the magnitude. Kinetics of Formation of Enzyme−Tetrahydropterin Binary Complexes. The kinetics of binding of BH4 and 6MPH4 to CvPheH were analyzed by following the absorbance at 320−340 nm after mixing the enzyme anaerobically with various concentrations of each tetrahydropterin. With the natural substrate BH4, the kinetic traces were monophasic at 320 nm (Figure 2A), but biphasic at 340 nm (Figure S2). The
Scheme 3. Kinetic Mechanisms for the Binding of a Tetrahydropterin to CvPheH
Phe complexes. To examine the kinetics of phenylalanine binding to the enzyme−tetrahydropterin complex, the enzyme was first mixed with an excess of a tetrahydropterin in the stopped-flow spectrophotometer. After allowing 1 s for binding to equilibrate, the solution was then mixed with various concentrations of phenylalanine, and the absorbance was monitored at 340 nm. When the enzyme was mixed with BH4 or 6MPH4 before mixing with phenylalanine, there was a biphasic increase in absorbance (Figures 3A and 4A). For both pterins the rate constant for the fast phase exhibited a linear dependence on the phenylalanine concentration, while the rate constant for the slow phase was independent of the concentration of phenylalanine (Figures S4A and S5A). Although formation of the CvPheH−5dMPH4 complex does not result in an absorbance change, formation of the CvPheH− 5dMPH4−Phe complex is detectable at 340 nm. When the enzyme was mixed with 5dMPH4 before mixing with phenylalanine, there was a monophasic increase in absorbance at this wavelength (Figure 5A). The rate constant exhibited a linear dependence on the phenylalanine concentration (Figure S6A). The kinetics of binding of tetrahydropterins to the E−Phe complex were examined in a separate set of experiments. When the enzyme was mixed with phenylalanine before mixing with BH4 (Figure 3B,C), the increase in absorbance was biphasic. The rate constant for the fast phase exhibited a linear dependence on the concentration of both pterins, while the second slow phase was independent of the concentration of the pterin (Figure S4B). In contrast, when the enzyme was mixed with phenylalanine before mixing with 6MPH4 (Figure 4B,C) or 5dMPH4 (Figure 5B,C), the increase in absorbance was monophasic. In both cases the rate constant exhibited a linear dependence on the concentration of the pterin (Figures S5B and S6B). Global Analysis of the Binding Kinetics. The stoppedflow data from all the binding experiments for each pterin were fit globally with KinTek Explorer to likely kinetic mechanisms (Schemes 4, 5, 6). The possibilities considered included both ordered and random binding of phenylalanine and tetrahydropterin. Because the binding of tetrahydropterins to form the binary complex exhibited biphasic kinetics, nonproductive binding of the tetrahydropterin and an isomerization of the initial binary complex were also considered. Initially, Scheme 3A was expanded to include phenylalanine binding by assuming that only one of the binary enzyme−tetrahydropterin complexes could bind phenylalanine, resulting in Scheme 4A. Scheme 3B was expanded by considering two different E− pterin binary complexes. In the first, phenylalanine binds to the initial E−PH4 complex (Scheme 4B). In the second, phenylalanine binds to the second E−PH4 complex (Scheme 4C).
Figure 2. Absorbance changes during formation of CvPheH− tetrahydropbterin binary complexes. (A) CvPheH (100 μM) was mixed with 300 (○), 500 (◊), or 750 (Δ) μM BH4; the lines are from the kinetic mechanism in Scheme 6B and the corresponding rate constants in Table 2. (B) CvPheH (100 μM) was mixed with 200 (○), 400 (◊), or 600 (Δ) μM 6MPH4; the lines are from the kinetic mechanism in Scheme 6B and the corresponding rate constants in Table 3. All concentrations are after mixing. In each trace only every fourth point is shown for clarity.
effect of the concentration of BH4 on the rate constant at 320 nm could be described by a straight line with a nonzero intercept (Figure S3A). With 6MPH4 the kinetic traces were biphasic in all cases (Figure 2B). The effect of the 6MPH4 concentration on the observed rate constant for the first phase could also be described by a straight line with a nonzero intercept (Figure S3B). The rate constant for the slower second phase was independent of the 6MPH4 concentration. The biphasic kinetics of tetrahydropterin binding are consistent with two minimal mechanisms. In the mechanism in Scheme 3A, there is independent binding of the pterin to the enzyme to form two different binary complexes. In the mechanism in Scheme 3B, the initial CvPheH−tetrahydropterin complex undergoes a rearrangement to a second binary complex. The stopped-flow data for both pterins were fit globally to each possibility using KinTek Explorer. With BH4, fitting to the branched mechanism in Scheme 3A resulted in a marginally higher X2 value than fitting to the sequential mechanism in Scheme 3B; with 6MPH4, Scheme 3B also fit slightly better than 3A (Table 1). For both pterins the values of k2 and k−2 for Scheme 3B were poorly restrained. Kinetics of Ternary Complex Formation. A doublemixing stopped-flow approach was taken to characterize the kinetics of formation of the ternary enzyme−tetrahydropterin− 6850
DOI: 10.1021/acs.biochem.6b01012 Biochemistry 2016, 55, 6848−6857
Article
Biochemistry Table 1. Kinetic Parameters for Formation of Enzyme−Tetrahydropterin Binary Complexes BH4 parameter −1
−1
k1 (mM s ) k−1 (s−1) k2 (mM−1 s−1 or s−1) k−2 (s−1) X2 a
Scheme 3A 240 (230−250) 12 (11−13) 70 (62−80) 4.8 (3.9−5.8) 39
a
6MPH4 Scheme 3B
Scheme 3A
Scheme 3B
320 (310−330) 11 (9.9−12) 1.6 (0.14−7.5) 16 (8.4−26) 35
240 (150−250) 72 (54−144) 11 (9−38) 11 (8.6−33) 620
240 (200−300) 68 (42−192) 4.8 (1.1−69) 14 (3.4−73) 590
Confidence intervals based on a X2 threshold of 1.1
Figure 3. Absorbance changes during formation of the CvPheH− BH4−Phe complex: (A) Enzyme (50 μM) was first mixed with 500 μM BH4 for 1 s and then with 200 (○), 400 (Δ), or 600 (□) μM phenylalanine. (B) CvPheH (50 μM) was first mixed with 200 (○), 400 (Δ), or 600 (□) μM phenylalanine for 1 s and then with 500 μM BH4. (C) CvPheH (50 μM) was first mixed with 400 μM phenylalanine for 1 s and then with 250 (◊), 500 (Δ), or 750 (∇) μM BH4. The lines are from the kinetic mechanism in Scheme 6B and the corresponding rate constants in Table 2. All concentrations are after the final mixing. In each trace only alternate points are shown for clarity.
Figure 4. Absorbance changes during formation of the CvPheH− 6MPH4−Phe complex. (A) Enzyme (50 μM) was first mixed with 500 μM 6MPH4 for 1 s and then with 200 (○), 400 (Δ), or 600 (□) μM phenylalanine. (B) Enzyme (50 μM) was first mixed with 200 (○), 400 (Δ), or 600 (□) μM phenylalanine for 1 s and then with 500 μM 6MPH4. (C) CvPheH (50 μM) was first mixed with 600 μM phenylalanine for 1 s and then with 500 (□), 750 (◊), or 1000 (∇) μM 6MPH4. The lines are from the kinetic mechanism in Scheme 6B and the corresponding rate constants in Table 3. All concentrations are after the final mixing. In each trace only every third point is shown for clarity.
The rate constants for formation of the E−BH4−Phe complex resulting from optimized fits of the mechanisms in Scheme 4A−C to the data are listed in Table 2. The X2 values for the mechanisms in Scheme 4A,B are significantly lower (
