Article pubs.acs.org/biochemistry
Revisiting the Val/Ile Mutation in Mammalian and Bacterial Nitric Oxide Synthases: A Spectroscopic and Kinetic Study Marine Weisslocker-Schaetzel, Mehdi Lembrouk, Jérôme Santolini, and Pierre Dorlet* Institute for Integrative Biology of the Cell (I2BC), CEA, CNRS, Univ Paris-Sud, Université Paris-Saclay, F-91198 Gif-sur-Yvette cedex, France S Supporting Information *
ABSTRACT: Nitric oxide is produced in mammals by the nitric oxide synthase (NOS) isoforms at a catalytic site comprising a heme associated with a biopterin cofactor. Through genome sequencing, proteins that are highly homologous to the oxygenase domain of NOSs have been identified, in particular in bacteria. The active site is highly conserved except for a valine residue in the distal pocket that is replaced with an isoleucine in bacteria. This switch was previously reported to influence the kinetics of the reaction. We have used the V346I mutant of the mouse inducible NOS (iNOS) as well as the I224V mutant of the NOS from Bacillus subtilis (bsNOS) to study their spectroscopic signatures in solution and look for potential structural differences compared to their respective wild types. Both mutants seem destabilized in the absence of substrate and cofactor. When both substrate and cofactor are present, small differences can be detected with Nωhydroxy-L-arginine compared to arginine, which is likely due to the differences in the hydrogen bonding network of the distal pocket. Stopped-flow experiments evidence significant changes in the kinetics of the reaction due to the mutation as was already known. We found these effects particularly marked for iNOS. On the basis of these results, we performed rapid freeze-quench experiments to trap the biopterin radical and found the same results that we had obtained for the wild types. Despite differences in kinetics, a radical could be trapped in both steps for the iNOS mutant but only for the first step in the mutant of bsNOS. This strengthens the hypothesis that mammalian and bacterial NOSs may have a different mechanism during the second catalytic step.
N
only difference in the distal environment of the heme is the replacement of a conserved valine residue in mammalian NOS with an isoleucine in bacterial enzymes.13 The study of the mutation of that specific amino acid has been reported previously for the inducible isoform of mammalian NOS (iNOS, V346I mutant) and the NOS from the bacterium Bacillus subtilis (bsNOS, I224V mutant) with kinetic characterization under single-turnover conditions.13,14 The mutation was found to have a significant impact on the kinetics of the reaction, in particular the final dissociation of NO from the active site but also the formation and decay of the oxyferrous species.13 The effect of the mutation on NO release was also studied on the NOS protein from Geobacillus stearothermophilus with similar results.15 The iNOS V346I mutant was further studied with respect to its NO binding properties, with respect to its steady-state kinetics, and in NO geminate recombination experiments.14,16 More recently, this mutation was found to be important in the design of selective inhibitors toward bacterial NOSs.17,18 Structurally, the comparison between iNOS and bsNOS was performed on the wild-type (WT) X-ray structures, and no significant changes were found.19,20 With regard to the
itric oxide synthases (NOSs) are the enzymes responsible for the production of NO in mammals. They catalyze the two-step oxidation of L-arginine (Arg) into L-citrulline and NO via the formation of Nω-hydroxy-L-arginine (NOHA).1,2 They are homodimeric enzymes containing an oxygenase and a reductase domain. The active site, located in the oxygenase domain, comprises a heme molecule bound to the protein by a cysteine residue and associated with a tetrahydrobiopterin cofactor (H4B). The precise and detailed mechanism of NOS is still under debate, but the first events are common to both steps of the catalysis (Scheme 1).2 The high-spin ferric heme is first reduced by an electron from the reductase domain and binds dioxygen to form an oxyferrous moiety.3 The biopterin cofactor is directly involved in the catalysis: it gives an electron to this oxyferrous complex before autoxidation can occur, allowing the catalysis to proceed toward the oxidation of the substrate.4,5 The biopterin thus forms a radical that is later reduced to regenerate the cofactor.2,6,7 Extensive genome sequencing has led to the identification of highly homologous proteins compared to the oxygenase domain of mammalian NOS throughout the living kingdoms. In particular, NOSs have been identified in many bacterial phyla, mostly Gram-positive, such as Deinococcus, Firmicutes, Actinobacteria, etc.8,9 Bacterial NOSs present an overall structure similar to that of the mammalian NOS oxygenase domain, and the heme active site is highly conserved.10−12 The © XXXX American Chemical Society
Received: October 5, 2016 Revised: January 10, 2017 Published: January 11, 2017 A
DOI: 10.1021/acs.biochem.6b01018 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry Scheme 1. NOS Molecular Mechanisma
Simplified mechanism detailing the first events of the catalysis; other heme intermediates occur during steps 1 and 2.2 Numbers in circles indicate the species that can be detected by ultraviolet−visible spectroscopy in stopped-flow experiments. a
Ferric Samples. Ferric FeIII samples were prepared by conditioning the protein in a 100 mM KPi, pH 7.4 buffer containing 10% glycerol in the presence of Arg (5 mM) or NOHA (500 μM), H4B (500 μM), and DTT (3 mM) by four successive cycles of dilution and centrifugation using MicroCon membrane concentrators with a 30 kDa cutoff (Millipore, Bedford, MA). Samples were directly frozen in EPR tubes. Ferrous Samples. Ferrous FeII samples were prepared from the ferric samples. Anaerobic FeIII NOS was obtained by 100 cycles of alternate vacuum and argon refilling, in a quartz cuvette. Ferric samples were reduced by small additions of a dithionite solution (100 mM, prepared with anaerobic KPi, pH 7.4 buffer) directly to the cuvette by using a gastight syringe (Hamilton, Reno, NV). Reduction was monitored by ultraviolet (UV)−visible absorption spectroscopy (Uvikon spectrometer from Serlabo Technologies). Nitrosyl Ferrous Samples. {FeNO}7 samples were prepared from the ferrous samples. A NO-saturated solution (∼3.3 mM) was prepared by flushing NO gas (filtered by a KOH solution) through a previously degassed KPi, pH 7.4 buffer. Small volumes of the NO solution were then added by using a gastight syringe to the ferrous protein sample to form the {FeNO}7 complex. This was monitored by UV−visible absorption spectroscopy. Samples were then transferred into argon-filled EPR tubes and frozen in liquid nitrogen. Stopped-Flow Experiments. A ferrous protein sample was transferred with a gastight syringe to the anaerobic channel of a stopped-flow SFM-300 setup (BioLogic Science Instruments) and quickly mixed at 4 °C with an aerobic buffer (airsaturated, dissolved O2 concentration of >300 μM). The reaction was monitored by UV−visible absorption spectroscopy using a diode array detector (Tidas, J&M Analytik AG, Aalen, Germany). Spectra were recorded with a minimal time interval of 3 ms. The final protein concentration was ∼10 μM. The formation and decay kinetics of hemic intermediates were determined by monitoring the absorbance change versus time at 396, 440, or 650 nm. For each condition, two or three separate experiments were performed, and for each of them, five series of spectra were averaged. The mean value and standard deviation of the kinetic constants were determined from all experiments.
mutants, only an inhibitor bound bsNOS I224V was crystallized.17 The changes in reaction kinetics observed for these mutations potentially provide a good opportunity to further study the mechanism of NOS and the structure−function relationship for mammalian and bacterial enzymes. Indeed, by using rapid freeze-quench experiments, it is in principle possible to trap the transient biopterin radical that is formed as the cofactor reduces the oxyferrous moiety at the beginning of catalysis (Scheme 1). Such a radical was detected and characterized by electron paramagnetic resonance (EPR).4,7,21−28 In performing rapid freeze-quench experiments on the WT of iNOS and bsNOS, our group reported a potential difference in mechanism for the second step.28 We were able to trap the radical for both steps in the case of iNOS but only for the first step in the case of bsNOS.28 The absence of radical trapped for the second step in bsNOS for any quenching times led to questions about whether the cofactor is involved at all in the electron transfer in that case. If it is, then significant changes in the radical reduction rate must occur between iNOS and bsNOS. The use of the Val/Ile mutants for which the kinetics are significantly affected compared to their respective WT seems therefore appropriate in further studying this question. In this report, we have used EPR and Raman spectroscopies to probe potential structural changes in the heme active site for the WT and mutant proteins of iNOS and bsNOS in solution. We have checked their kinetic properties by stopped flow to use rapid freeze-quench methods to probe the formation of a biopterin radical during catalysis. The results obtained are discussed with respect to the initial events of the catalytic mechanism proposed for NOS.
■
MATERIALS AND METHODS Chemicals. All chemicals were purchased from SigmaAldrich (St. Louis, MO). H4B was purchased from Schricks Laboratories (Jona, Switzerland) and NOHA from Enzo Life Sciences Inc. (Farmingdale, NY). NO was purchased from Messer France SA (Asnières, France). Sample Preparation. bsNOS I224V and iNOS V346I were overexpressed in Escherichia coli as described previously.10 B
DOI: 10.1021/acs.biochem.6b01018 Biochemistry XXXX, XXX, XXX−XXX
Article
Biochemistry Rapid Freeze-Quench Experiments. Ferrous samples were prepared as described above. To prevent unwanted reactions, the excess of dithionite was eliminated on protein desalting spin columns (Thermo Scientific) in a glovebox. The ferrous protein sample was then transferred with a gastight syringe to the anaerobic channel of a Freeze-Quench SFM-300 setup (BioLogic Science Instruments) and quickly mixed at 4 °C with an aerobic buffer (air-saturated, dissolved O 2 concentration of >300 μM). The reaction was stopped at a given time by rapid freezing in an isopentane bath at 150 K. The sample was then collected in an EPR tube and frozen at 77 K. The final protein concentration was ∼100 μM. EPR Spectroscopy. X-Band EPR spectra were recorded on a Bruker ELEXSYS 500 spectrometer equipped with a continuous-flow ESR 900 cryostat and an ITC504 temperature controller (Oxford Instruments, Abingdon, U.K.). Simulations were performed by using the Easyspin software package29 and routines written in the lab. Resonance Raman Spectroscopy. Samples (50 μL) were placed into a gastight quartz spinning cell, at room temperature, to prevent local heating and to prevent photodissociation and degradation. Raman excitation at 441.6 nm was obtained with a He−Cd laser (Kimmon, Tokyo, Japan). Resonance Raman spectra were recorded using a modified single-stage spectrometer (Jobin-Yvon T64000, Jobin-Yvon, Longjumeau, France) equipped with a liquid N2-cooled back-thinned CCD detector. Stray scattered light was rejected using a holographic notch filter (Kaiser Optical Systems, Ann Arbor, MI). Spectra were recorded as the co-addition of 40−240 individual spectra with CCD exposure times of 5−30 s each. Three to six successive sets of such spectra were then averaged. The laser power at the sample was
