Dehaloperoxidase–hemoglobin (DHP), a multifunctional globin protein, not only functions as an oxygen carrier as typical globins such as myoglobin and hemoglobin, but also as a peroxidase, a mono- and dioxygenase, peroxygenase, and an oxidase. Kinetics of DHP binding to NO, CO, and O2 were characterized for wild-type DHP A and B and the H55D and H55V DHP A mutants using stopped-flow methods. All three gaseous ligands bind to DHP significantly more weakly than sperm whale myoglobin (SWMb). Both CO and NO bind to DHP in a one-step process to form a stable six-coordinate complex. Multiple-step NO binding is not observed in DHP, which is similar to observations in SWMb, but in contrast with many heme sensor proteins. The weak affinity of DHP for O2 is mainly due to a fast O2 dissociation rate, in accordance with a longer εN–Fe distance between the heme iron and distal histidine in DHP than that in Mb, and an open-distal pocket that permits ligand escape. Binding affinities in DHP show the same 3–4 orders separation between the pairs NO/CO and CO/O2, consistent with the ‘sliding scale rule’ hypothesis. Strong gaseous ligand discrimination by DHP is very different from that observed in typical peroxidases, which show poor gaseous ligand selectivity, correlating with a neutral proximal imidazole ligand rather than an imidazolate. The present study provides useful insights into the rationale for DHP to function both as mono-oxygenase and oxidase, and is the first example of a globin peroxidase shown to follow the ‘sliding scale rule’ hypothesis in gaseous ligand discrimination.

Introduction

Dehaloperoxidase–hemoglobin (DHP) is a heme-containing globin, which accounts for ∼3% the soluble protein in crude extracts of the terebellid polychaete Amphitrite ornata (A. ornata) [1]. DHP has two forms, DHP A and B, which are coded by different genes and distinguished by five amino acids in the peptide sequences [2]. In its environment in a tube buried in shallow coastal mudflats and estuaries, A. ornata is known to use DHP as a primary enzyme for detoxification of naturally occurring toxins, including brominated phenols, indoles, and pyrroles among other halogenated aromatic hydrocarbons. Both DHP isozymes exhibit peroxidase activity, catalytically dehalogenating trihalophenols, which are widely found in benthic ecosystems to dihaloquinones in the presence of co-substrate H2O2 [1,3,4]. The phenols are potentially toxic to sediment-dwelling invertebrates due to their tendency to react to form dioxins and other reactive molecules that can interfere with replication. The quinone products of the DHP peroxidation reactions are much less harmful to A. ornata. More recently, DHP has been shown to exhibit peroxygenase and oxidase activities toward a range of substrates including dibromophenols, brominated indoles, and pyrroles [5,6]. Isolated DHP isozymes are also shown to have oxygen storage/transfer capabilities, which are consistent with their abundance in the coelom of A. ornata. DHP A and B are also both present in erythrocruorin, the giant hemoglobin (Hb) in the small tentacles of A. ornata that are usually visible above the surface of mudflat, where the organism is resident [5]. Both the abundance of DHP in the coelom and the existence of erythrocruorin strongly suggest that DHP is the principal protein responsible for oxygen transport and storage in A. ornata as well.

The peroxidase activity of DHP apparently contradicts its identity as a globin. Peroxidases are known to activate O–O bond cleavage of bound H2O2 while globins have a primary function to reversibly bind O2. Peroxidases normally have a ferric resting state while globins have a ferrous resting state. The functional differences between peroxidases and globins would seem to be so great that the idea of globin peroxidase seems like a contradiction. The resolution of this contradiction starts with the measurement of the high redox potential of 221 mV for DHP [7,8]. This redox potential is 100–150 mV higher than the average values of globins, stabilizing ferrous heme and supporting the oxygen storage/transport function observed in myoglobin (Mb) and Hb. The shift in redox potential in DHP is opposite to that of typical peroxidases, which are more negative by ∼250–300 mV relative to the globins. The negative redox potentials support stable ferric resting states in typical peroxidases [9]. The apparently paradoxical dual functions of DHP can be explained in part by a hypothesis that the peroxidation cycle is shifted, which means that it goes from the oxyferrous state directly to compound II (instead from the ferric state to compound I) when H2O2 binds. The flexibility of the distal histidine (His55) appears to have evolved to play a crucial role in H2O2 binding in the ferrous form of DHP. There is evidence that His55 exists in multiple conformations. In the first conformation (globin form), His55 stabilizes the Fe(II)–O2 species, but then switches to a different conformation enforced by substrate binding to facilitate the deprotonation of H2O2 and subsequent heterolytic O–O bond cleavage in the peroxidase cycle [5,10,11]. Consistent with the functional switch, the Nε–Fe distance in DHP (5.4–5.5 Å) [12] is intermediate between that of horseradish peroxidase (HRP, 6.0 Å) and Mb (4.3–4.5 Å) [1315]. The dynamics of His55, switching between the ‘open’ (inactive) and ‘closed’ (active) conformations, is proposed to be regulated by the binding of mono- and trihalophenol substrates, respectively [16,17].

In its peroxidase reactions, DHP has been proposed to react with H2O2 in its Fe(III), Fe(II), and even Fe(II)–O2 states to generate Compound I, Complex ES, and Compound II high oxidation state intermediates [5,10,1820]. These intermediates can be cycled back to either Fe(II) or Fe(III) heme by various halophenol co-substrates or their secondary products, or by some yet to be identified source(s) of reducing equivalents, facilitated by the intrinsic high midpoint potentials of the Fe(IV)/Fe(III) and Fe(III)/Fe(II) half reactions [5]. The versatility of DHP in catalysis is further demonstrated by peroxygenase and oxidase reactivities using haloindoles and nitrophenols as substrates, confirmed by 18O-labeled co-substrate analogs [21,22]. Moreover, isolated DHP is recently reported to exhibit P450 like mono-oxygenase and lipoxygenase-like dioxygenase activities using 5,5′-dibromo-indigo and 2,3-dimethylindole as co-substrates, respectively [23].

In addition to the uncommon multifunctional enzymatic activities, the other important aspect of multifunctional DHP is its oxygen storage/transfer function. To understand the oxygen storage/transfer function of DHP, a careful characterization of its binding to oxygen is essential. The oxygen dissociation constant of DHP, KD(O2), was recently measured: 7.7 µM for DHP A and about half of this value for DHP B [24], both about one order of magnitude weaker than that of sperm whale Mb (SWMb) [2527]. However, the method used for the KD(O2) measurement is based on a competition binding assay with t-butyl isocyanide and does not provide any kinetic information for O2 binding. Direct monitoring of the O2 reaction with DHP should reveal the kinetics/affinity of O2 binding to DHP. The binding kinetics/affinity of DHP for other gaseous ligands, NO and CO, can also be measured by directly monitoring their reactions with DHP. These measurements provide insights into whether DHP binds NO, CO, and O2 selectively and may reveal other, as yet unknown, function(s) of DHP.

The ‘sliding scale rule’ hypothesis, recently derived from bioinformatic graphical analyses based on a large database of hemeproteins containing a five-coordinate (5c) heme ligated to a proximal neutral imidazole, elucidates the 3–4 orders of difference in the affinities of these hemeproteins for the NO/CO and CO/O2 pairs [28,29]. One prominent example is the explanation for one of the most intriguing issues in the gaseous ligand selectivity of mammalian NO sensor soluble guanylyl cyclase (sGC): why sGC is able to bind NO with an apparent KD of pM but totally exclude O2 [29]. On the other hand, typical peroxidases, such as HRP and cytochrome c peroxidase (CCP), which commonly have an imidazolate proximal heme ligand, exhibit very poor gaseous ligand selectivity due to the high-ligand field strength. Given this background, it is of interest to examine whether DHP, as an unusual peroxidase, is capable of selectively binding NO, CO, and O2.

In the present study, we conducted systematic kinetic binding measurements of all three gaseous ligands with the multifunctional DHP A and B, and specific His55 mutants of DHP A. The measured kinetic/affinity parameters were subjected to bioinformatic graphical analysis. Both DHP A and B bind NO, CO, and O2 selectively and their KDs for the gaseous ligands follow the ‘sliding scale rule’. On the other hand, the data obtained for His55 mutants of DHP A clearly showed the effect of the distal histidine in regulating selectively the O2 binding via specific H-bonding with O2 ligand.

Experimental

Materials

All chemicals were purchased from Sigma–Aldrich (St. Louis, MO) and Fisher Scientific (Hampton, NH) and used without further purification. The E.Z.N.A plasmid DNA mini kit was from Omega Bio-Tek (Norcross, GA). The QuikChange II site-directed mutagenesis kit was purchased from Agilent Technologies (Santa Clara, CA). The designed oligonucleotide primers were synthesized by Integrated DNA Technologies (Coralville, IA). CO and NO gases were from Matheson-TriGas, Inc. (Houston, TX). NO was pre-purified by passing through a NaOH trap to remove nitrous and nitric acid contaminants.

Site-directed mutagenesis and protein purification

DHP A mutations were generated with the QuikChange II site-directed mutagenesis kit using the plasmids encoding wild-type DHP A (wt DHP A with a tag of six histidines at N-terminus) as the templates, and the primers used are listed in Table 1 with mutation sites underlined. The plasmids with the mutations were extracted using a E.Z.N.A plasmid DNA mini kit from the transformed BL21(DE3) Escherichia coli, and their integrities were verified by sequencing (Genewiz, South Plainfield, NJ).

Table 1
DNA sequences of primers for DHP A mutant constructs
Primer Sequence 
H55D forward 5′- GGCCAAGTTCGGTGATGACACTGAGAAAGTGTTCAACC-3′ 
H55D reverse 5′- GGTTGAACACTTTCTCAGTGTCATCACCGAACTTGGCC-3′ 
H55V forward 5′-GGCCAAGTTCGGTGATGTTACTGAGAAAGTGTTCAACC-3′ 
H55V reverse 5′-GGTTGAACACTTTCTCAGTAACATCACCGAACTTGGCC-3′ 
Primer Sequence 
H55D forward 5′- GGCCAAGTTCGGTGATGACACTGAGAAAGTGTTCAACC-3′ 
H55D reverse 5′- GGTTGAACACTTTCTCAGTGTCATCACCGAACTTGGCC-3′ 
H55V forward 5′-GGCCAAGTTCGGTGATGTTACTGAGAAAGTGTTCAACC-3′ 
H55V reverse 5′-GGTTGAACACTTTCTCAGTAACATCACCGAACTTGGCC-3′ 

WT DHP A, wt DHP B, H55D, and H55V DHP A were expressed and purified as previously described [3]. Briefly, E. coli cell pellets were suspended and lysed in the lysis buffer for 4 h. The cell slurry was then centrifuged at 28 000×g for 30 min, and the supernatant was applied to an Ni–NTA column. The protein was eluted, concentrated, and oxidized by K3[Fe(CN)6]. Unreacted K3[Fe(CN)6] was removed by a NAP-25 size exclusion column, and the buffer was simultaneously exchanged into 20 mM potassium phosphate (pH 6). Ferric DHP was then loaded onto a CM-52 ion exchange column and eluted with 150 mM potassium phosphate (pH 7). The concentrations of the purified ferric proteins were determined using the Soret extinction coefficients: wt DHP A and B both with ε406 nm = 116 400 M−1 cm−1, H55D DHP A with ε398 nm = 82 500 M−1 cm−1, and H55V DHP A with ε394 nm = 121 300 M−1 cm−1, respectively. The extinction coefficients of Soret bands of H55D and H55V DHP A were determined via quantification of the heme content by the pyridine hemochrome method [30].

Measurements of gaseous ligand bindings

The association and dissociation kinetics of NO, CO, and O2 to ferrous DHP were studied with an Applied PhotoPhysics SX-18MV stopped-flow instrument (Leatherhead, U.K.), with its sample handling unit kept in an anaerobic chamber (COY Lab Products, Grass Lake, MI). The anaerobic chamber was filled with a H2/N2 mixture (typically 7–10% H2) and any residual O2 was removed by reaction with H2 catalyzed by a platinum catalyst. The anaerobic condition in the chamber was indicated by the constant 0 ppm [O2] reading of the O2 meter placed inside the chamber. DHP was reduced by anaerobic Na2S2O4 titration in a tonometer after five alternating cycles of vacuum and argon displacement. The CO or NO stock solutions were prepared by long-time flushing of anaerobic buffer with CO or NO gas, respectively.

The time courses of different wavelengths were followed by single-wavelength stopped-flow using a monochromator, and the whole spectral changes were monitored by rapid-scan stopped-flow using a diode-array accessory. Global analysis of the rapid-scan data was conducted using the Pro-Kineticist (Applied PhotoPhysics). The association rate constants of CO and NO, kon(CO), and kon(NO), respectively, were measured by different methods.

In the case of CO binding, the observed rates, kobs, of single-wavelength data under pseudo-first-order conditions, were obtained by fitting time courses of absorbance to the standard exponential function:

 
formula
1

where Af, t, and a are the final optical absorbance, reaction time, and amplitude of the total optical change, respectively. The second-order association rate constant in CO binding, kon(CO), was derived from the slopes in the secondary plots of kobs versus [CO]:

 
formula
2

where the y-intercept corresponds to koff(CO).

The much faster formation rate of a six-coordinate (6c) NO–heme–His complex in DHP, kon(NO), was measured under the conditions of true second-order reaction [31], by rapidly mixing ferrous DHP with stoichiometric NO, and obtaining the kobs by fitting the time course to the following equation:

 
formula
3

where Af is the final optical absorbance and ΔA0 = Ai  − Af (Ai is the initial absorbance); t is the reaction time. Second-order rate constant, kon(NO) in unit of M−1 s−1, was obtained by multiplying kobs with the difference extinction coefficient, Δε, between the reactant DHP and the product, 6c NO–heme–His complex. [NO] did not decay noticeably during the experiments due to the well-maintained anaerobic conditions, as demonstrated by the high reproducibility of the multiple stopped-flow reactions. Moreover, there is little concern for residual oxygen in the anaerobic chamber due to the slow third-order reaction between NO and O2 [32]. Any residual Na2S2O4 left in the anaerobic DHP sample did not affect the kinetic measurement of the NO binding to DHP either, since NO binds to DHP at a rate faster by several orders of magnitude than it reacts with Na2S2O4.

The koff(CO) constants in DHP were very small, and the accurate values based on the y-intercepts of secondary plots were inaccurate due to large potential errors obtained by fitting. The koff(CO) constants were determined using an NO competition method by reacting the 6c CO complex with 1 mM NO. On the other hand, the koff(NO) constants were determined by sequential mixing stopped-flow spectrophotometry. DHP was reacted with 1× NO in the first mixing to form the 6c NO complex; the reaction mixture was allowed to age for 100 ms, and then reacted with 500 µM CO to replace NO, in the presence of 12.5 mM Na2S2O4 to consume dissociated NO to minimize its reassociation.

EPR spectroscopy

EPR spectra were recorded on a Bruker EMX spectrometer (Billerica, MA) at 115 K. Data analyses and spectral simulations were conducted using WinEPR and SimFonia programs furnished with the EMX system. EPR was conducted using the following parameters: frequency, 9.28 GHz; power, 4 mW; modulation frequency, 100 kHz; modulation amplitude, 2.5 G, and time constant, 0.33 s.

EPR samples of 6c NO complex were prepared by injecting desired volumes of 2 mM NO stock into and mixing with anaerobic ferrous DHP inside EPR tubes. The reactions were conducted inside the anaerobic chamber, and the EPR samples were then transferred outside the chamber, quickly frozen in a dry ice/ethanol mixture. The time between the NO injection and sample freezing was less than 1 min.

Computer fitting

Multiple time courses of A431 during the ferrous DHP reactions with different [O2] under pseudo-first-order conditions were fit simultaneously using the SCoP Program (Simulation Resources, Inc., Redlands, CA), based on the global minimization following the principal axis method (PRAXIS) [33]. The model used was based on the one-step reaction:

 
formula

The parameters allowed to float in the fittings were the forward/reverse rate constants, and the preset floating ranges were 10–1000 µM−1 s−1 and 1–100 000 s−1 for the forward and reverse rate constants, respectively.

Results

CO binding to ferrous DHP

The rapid-scan stopped-flow reaction of ferrous DHP A with CO exhibited a spectral change consistent with two-state (single-step) kinetics. The Soret peak shifted from 429 to 422 nm with no spectral intermediate (Figure 1A), indicating the formation of a six-coordinated low-spin (6cLS) CO–heme–His complex at 422 nm starting from a deoxy five-coordinated high-spin (5cHS) heme at 429 nm. The spectrum of the 6c CO–heme–His complex was resolved from the data based on a simple A → B model (Figure 1A, inset). Observed rates, kobss, obtained by fitting the time courses of A422 at various [CO], exhibited a linear dependence on [CO] (Figure 1B). The second-order association rate constant, kon(CO) as the slope, was 1.5 × 106 M−1 s−1 (Figure 1B and Table 2). The kon(CO) constants of DHP B, H55D, and H55V DHP A measured by the same method were 0.8 × 106, 2.7 × 106, and 2.4 × 106 M−1 s−1, respectively (Table 2). On the other hand, the dissociation rate constant of the CO complex in DHP A, koff (CO), was measured with the NO competition method following the time course of A422 (Figure 1C). The koff(CO) of DHP A was determined as 1.0 s−1 (Table 2). Similarly measured, the koff(CO)s of DHP B, H55D, and H55V DHP A were 0.7, 0.9, and 0.5 s−1, respectively (Table 2). The binding constants, calculated as KD(CO) = koff(CO)/kon(CO), were 6.7 × 10−7, 3.4 × 10−7, 2.1 × 10−7, and 8.3 × 10−7 M for wt, H55D and H55V DHP A, and DHP B, respectively (Table 3).

Kinetics of CO binding to DHP A.

Figure 1.
Kinetics of CO binding to DHP A.

(A) During the rapid-scan reaction of 5 µM DHP A with 50 µM CO, the spectra at 1.28, 6.4, 14.1, 24.3, 49.9, 75.5, and 101.1 ms are presented. Arrows represent the directions of the spectral changes. (inset) Deconvoluted optical species from the rapid-scan data based on a single-step reaction model A → B: A (black), ferrous DHP A; B (red), CO–heme–His complex. (B) The linear fit (line) of kobs of 5 µM ferrous DHP A reactions with CO to [CO] (circle). The kobss were from fitting the time courses of A422 to eqn (1). (C) Time course of A422 during the reaction of 0.25 µM DHP A CO complex with 1 mM NO (black line) is shown. The red line represents the fit to eqn (1).

Figure 1.
Kinetics of CO binding to DHP A.

(A) During the rapid-scan reaction of 5 µM DHP A with 50 µM CO, the spectra at 1.28, 6.4, 14.1, 24.3, 49.9, 75.5, and 101.1 ms are presented. Arrows represent the directions of the spectral changes. (inset) Deconvoluted optical species from the rapid-scan data based on a single-step reaction model A → B: A (black), ferrous DHP A; B (red), CO–heme–His complex. (B) The linear fit (line) of kobs of 5 µM ferrous DHP A reactions with CO to [CO] (circle). The kobss were from fitting the time courses of A422 to eqn (1). (C) Time course of A422 during the reaction of 0.25 µM DHP A CO complex with 1 mM NO (black line) is shown. The red line represents the fit to eqn (1).

Table 2
Association and dissociation rate constants of gaseous ligand binding to ferrous DHP
 kon (M−1 s−1koff (s−1
NO CO NO CO 
DHP A 1.1 × 108 1.5 × 106 3.6 × 10−3 1.0 
H55D DHP A 1.5 × 108 2.7 × 106 4.6 × 10−3 0.9 
H55V DHP A 2.7 × 108 2.4 × 106 4.0 × 10−3 0.5 
DHP B 1.7 × 108 0.8 × 106 4.0 × 10−3 0.7 
 kon (M−1 s−1koff (s−1
NO CO NO CO 
DHP A 1.1 × 108 1.5 × 106 3.6 × 10−3 1.0 
H55D DHP A 1.5 × 108 2.7 × 106 4.6 × 10−3 0.9 
H55V DHP A 2.7 × 108 2.4 × 106 4.0 × 10−3 0.5 
DHP B 1.7 × 108 0.8 × 106 4.0 × 10−3 0.7 
Table 3
Dissociation constants of gaseous ligands to ferrous DHP
 KD(NO), M KD(CO), M KD(O2), M 
DHP A 3.3 × 10−11 6.7 × 10−7 1.1 × 10−5 
H55D DHP A 3.1 × 10−11 3.4 × 10−7 1.2 × 10−4 
H55V DHP A 1.5 × 10−11 2.1 × 10−7 4.6 × 10−4 
DHP B 2.4 × 10−11 8.3 × 10−7 1.8 × 10−5 
 KD(NO), M KD(CO), M KD(O2), M 
DHP A 3.3 × 10−11 6.7 × 10−7 1.1 × 10−5 
H55D DHP A 3.1 × 10−11 3.4 × 10−7 1.2 × 10−4 
H55V DHP A 1.5 × 10−11 2.1 × 10−7 4.6 × 10−4 
DHP B 2.4 × 10−11 8.3 × 10−7 1.8 × 10−5 

NO binding to ferrous DHP

When ferrous DHP A was reacted with stoichiometric NO, the Soret peak quickly shifted from 429 to 420 nm, indicating the formation of a 6c NO–heme–His complex (Figure 2A). The optical spectrum of the 6c NO–heme–His complex was resolved from the rapid-scan data based on model A → B (Figure 2A, inset). The EPR spectrum of the DHP A reacting with stoichiometric NO exhibited features which were readily simulated using nitrogen hyperfine splitting constants, typical of a 6c NO complex (Figure 3A,B), in line with the observation based on the optical data. However, it was difficult to obtain accurate kobs for NO complex formation under pseudo-first-order conditions due to the much faster formation rate of NO complex compared with that of CO complex. The formation rate constant of the 6c NO complex, kon(NO), was determined with a single-wavelength stopped-flow measurement at 430 nm under true second-order reaction conditions, reacting 0.7 µM DHP A with stoichiometric NO. About 90% of the time course was captured, enabling a good fit to eqn (3), kon(NO) = 1.1 × 108 M−1 s−1 (Figure 2B and Table 2). The kon(NO)s of DHP B, H55D, and H55V DHP A measured with the same method were 1.7 × 108, 1.5 × 108, and 2.7 × 108 M−1 s−1, respectively (Table 2). On the other hand, the dissociation rate constant of the 6c NO–heme–His complex, koff(NO), was obtained by the CO competition method in the presence of excess Na2S2O4 and fitting the time course of A422 to eqn (1) (Figure 2C). The rate was much slower than kon(CO) of DHP A (Table 2), confirming that formation of the CO complex under these experimental conditions was rate-limited by koff(NO), 3.6 × 10−3 s−1. The koff(NO) was measured in the same way for H55D, H55V DHP A, and DHP B, 4.6 × 10−3, 4.0 × 10−3, and 4.0 × 10−3 s−1, respectively (Table 2). The KD(NO)s calculated as koff(NO)/kon(NO) were 3.3 × 10−11, 3.1 × 10−11, 1.5 × 10−11, and 2.4 × 10−11 M, for wt, H55D and H55V DHP A, and DHP B, respectively (Table 3).

Kinetics of NO binding to DHP A.

Figure 2.
Kinetics of NO binding to DHP A.

(A) The spectra captured at 1.28, 3.84, 6.4, 8.96, 11.5, 16.6, 24.3, 37.1, and 49.9 ms during the rapid-scan reaction of 1.1 µM DHP A with 1× NO are shown. Arrows represent the directions of the spectral changes. (inset) Deconvoluted optical species from the rapid-scan data based on a single-step reaction model A → B: A (black), ferrous DHP A; B (red), 6c NO–heme–His complex. (B) Time course of A430 during the reaction of 0.7 µM ferrous DHP A with 1× NO (black line). The association rate constant of DHP A 6c NO–heme–His complex, kon(NO), was obtained by fitting to eqn (3) (red line). (C) Time course of A422 in the dissociation of the 6c NO–heme–His complex in DHP A (black). DHP A was first reacted with 1× NO; after 100 ms, the reaction mixture was further reacted with 500 µM CO plus 12.5 mM Na2S2O4 (black line). Red line: fit to eqn (1).

Figure 2.
Kinetics of NO binding to DHP A.

(A) The spectra captured at 1.28, 3.84, 6.4, 8.96, 11.5, 16.6, 24.3, 37.1, and 49.9 ms during the rapid-scan reaction of 1.1 µM DHP A with 1× NO are shown. Arrows represent the directions of the spectral changes. (inset) Deconvoluted optical species from the rapid-scan data based on a single-step reaction model A → B: A (black), ferrous DHP A; B (red), 6c NO–heme–His complex. (B) Time course of A430 during the reaction of 0.7 µM ferrous DHP A with 1× NO (black line). The association rate constant of DHP A 6c NO–heme–His complex, kon(NO), was obtained by fitting to eqn (3) (red line). (C) Time course of A422 in the dissociation of the 6c NO–heme–His complex in DHP A (black). DHP A was first reacted with 1× NO; after 100 ms, the reaction mixture was further reacted with 500 µM CO plus 12.5 mM Na2S2O4 (black line). Red line: fit to eqn (1).

EPR spectra of DHP A 6c NO–heme–His complexes.

Figure 3.
EPR spectra of DHP A 6c NO–heme–His complexes.

(A) EPR spectrum of the NO complex formed by reacting DHP A with 1× NO. (B) Simulation of A using the following parameters: gx = 2.075, gy = 1.984, gz = 2.003; AN1: Ax = 10 G; Ay = 15 G and Az = 22.9 G; AN2: Ax = Ay = Az = 6 G. (C) EPR spectrum of the NO complex formed by reacting DHP A with 11× NO.

Figure 3.
EPR spectra of DHP A 6c NO–heme–His complexes.

(A) EPR spectrum of the NO complex formed by reacting DHP A with 1× NO. (B) Simulation of A using the following parameters: gx = 2.075, gy = 1.984, gz = 2.003; AN1: Ax = 10 G; Ay = 15 G and Az = 22.9 G; AN2: Ax = Ay = Az = 6 G. (C) EPR spectrum of the NO complex formed by reacting DHP A with 11× NO.

The heme–His bond was stable in the 6c NO–heme–His complex formed in the reaction of DHP A with stoichiometric NO, as indicated by the lack of further optical change after its formation (data not shown). The stability of the 6c NO–heme–His complex is corroborated by the EPR spectrum of the DHP A reaction with stoichiometric NO, which exhibits features typical of the 6c NO complex (Figure 3A). EPR data also demonstrated that DHP A does not exhibit multiple-step NO binding, which is distinct from the behavior of gas heme sensors such as sGC [34]. In the reaction of DHP A with up to 11 equivalents of NO, the 6c NO–heme–His remained stable as shown by the identical EPR spectra under these conditions (Figure 3A,C). The anisotropic EPR spectrum exhibited a nine-line hyperfine splitting of the center gz component (2.003), which was best simulated by a three-line hyperfine splitting (A = 22.9 G) from the NO nitrogen and three-line super hyperfine (A = 6 G) from the proximal histidine ε-nitrogen (Figure 3B). The hyperfine features for the gx (2.075) and the gy (1.984) components were not easily observed due to smaller hyperfine splittings and broader linewidth [35].

Reaction of O2 with ferrous DHP

The rapid-scan stopped-flow reaction of ferrous DHP A with 30 µM O2 proceeded rapidly and finished within 4 ms (Figure 4A). No optical intermediate was observed. However, the α/β bands were poorly resolved at 556 and 549 nm, indicating that the oxyferrous complex did not fully develop at [O2] of 30 µM. When ferrous DHP A was reacted with 100 µM O2, the first spectrum captured was that typical of an oxyferrous complex with better resolved α and β bands at 541 and 577 nm, respectively (Figure 4B), indicating more complete formation of the oxyferrous complex (Figure 4B). The rapid-scan data, therefore, indicated a sizeable koff(O2) for the oxyferrous complex. In single-wavelength stopped-flow measurements, when DHP A was reacted under pseudo-first-order conditions with O2 of various concentrations, all the kobss except for that with the lowest concentration of O2 were ≥500 s−1, which was sufficiently large and that majority of the reactions occurred in the dead time of the stopped-flow apparatus (Figure 4C), corroborating the large kobs = kon[O2] + koff of DHP A. Therefore, only very small factions of reactions were captured (Figure 4C), and kobss could not be measured to construct a secondary plot of kobs versus [O2], in contrast with the case of kon(CO) measurement (Figure 1B). Moreover, the kon(O2) to DHP A cannot be measured under true second-order reaction conditions by reacting DHP A with stoichiometric O2, like that for NO binding (Figure 2B). The method of fitting to a second-order reaction eqn (3) fails due to the very large koff(O2), which could not be determined using the CO or NO competition method. Actually, the ΔA422 observed in the reaction of the oxyferrous DHP A with 500 µM CO plus 12.5 mM Na2S2O4 only accounted for ∼5% of the total ΔA422 by CO complex formation and the reactions were rate-limited by kon(CO) (data not shown).

Kinetics of O2 binding to DHP A.

Figure 4.
Kinetics of O2 binding to DHP A.

(A) The spectra captured at 1.28, 3.84, 6.4, 8.96, and 11.5 ms during the rapid-scan reaction of 2.4 µM DHP A with 30 µM O2. Arrows represent the directions of the spectral changes. (inset) Deconvoluted optical species from the rapid-scan data based on a single-step reaction model A → B: A (black), ferrous DHP A; B (red), oxyferrous complex. (B) The spectrum captured at 1.28 ms during the reaction of 5.3 µM DHP A with 100 µM O2 (red), in comparison with the spectrum of ferrous DHP A (black). (C) Time courses of A431 in the 4.1 µM ferrous DHP A reactions with various concentrations of O2. The arrow indicates the increase in [O2]: 0 (anaerobic buffer, top), 15, 30, 45, 60, 90, 150, 240, 360, 480, and 600 µM. (D) Normalized ΔA431 versus [O2]. and each A431 was averaged over 10 – 20 ms. To compare the ΔA431 in different proteins, ΔA431 was normalized against the A431,max obtained by fitting to eqn (4). Circle: wt DHP A; triangle: H55D DHP A; diamond: H55V DHP A, and lines: the fits.

Figure 4.
Kinetics of O2 binding to DHP A.

(A) The spectra captured at 1.28, 3.84, 6.4, 8.96, and 11.5 ms during the rapid-scan reaction of 2.4 µM DHP A with 30 µM O2. Arrows represent the directions of the spectral changes. (inset) Deconvoluted optical species from the rapid-scan data based on a single-step reaction model A → B: A (black), ferrous DHP A; B (red), oxyferrous complex. (B) The spectrum captured at 1.28 ms during the reaction of 5.3 µM DHP A with 100 µM O2 (red), in comparison with the spectrum of ferrous DHP A (black). (C) Time courses of A431 in the 4.1 µM ferrous DHP A reactions with various concentrations of O2. The arrow indicates the increase in [O2]: 0 (anaerobic buffer, top), 15, 30, 45, 60, 90, 150, 240, 360, 480, and 600 µM. (D) Normalized ΔA431 versus [O2]. and each A431 was averaged over 10 – 20 ms. To compare the ΔA431 in different proteins, ΔA431 was normalized against the A431,max obtained by fitting to eqn (4). Circle: wt DHP A; triangle: H55D DHP A; diamond: H55V DHP A, and lines: the fits.

Although no kinetic parameters were measured, the affinity of DHP A for O2 was experimentally determined due to that the decrease in DHP A Soret signal was dependent on [O2], and all the reactions reached equilibrium after 10 ms (Figure 4C). It was therefore possible to measure KD(O2) of ferrous DHP A by titrating it with O2 and monitoring the [O2]-dependent Soret absorbance change at equilibrium using single-wavelength stopped-flow measurements. The plot of the observed absorbance change versus [O2] was successfully fit with the following equation:

 
formula
4

The KD(O2) of DHP A thus obtained was 11 μM (Figure 4D and Table 3).

The time courses of A431 in the reactions of DHP B with various levels of O2 were quite similar to those of DHP A (Supplementary Figure S1B), and the KD(O2) of DHP B, 18 μM (Table 3), was determined using the same O2-titration method. Almost no kinetic change was captured during the reactions of H55D and H55V DHP A even with lowest concentration of O2 (Supplementary Figure S1C,D), likely due to larger koff(O2)s. Nevertheless, KD(O2)s of H55D and H55V DHP A were determined using the O2-titration method. Mutation of His55 to polar aspartate apparently led to a significantly lower affinity of H55D DHP A for O2, KD(O2) = 120 µM (Supplementary Figure S1C and Table 3). The affinity for O2 was even lower for DHP A with His55 conversion to apolar valine, KD(O2) = 460 µM (Supplementary Figure S1D and Table 3).

Discussion

The CO rebinding kinetics of DHP triggered by flash photolysis have been studied as a function of temperature using the FTIR temperature derivative spectroscopy [36]. To build further on the low temperature data, the full set of CO and NO association/dissociation kinetics of DHP at ambient temperature has been measured in the present study. The kinetic studies are supported by many structural studies of DHP. The crystallographic data of CO complex of ferrous DHP and varieties of spectroscopic characterizations revealed two conformations of His55, ‘open’ and ‘closed’ [11,16,37]. Unlike the relative strong hydrogen bonding between CO and distal residue(s) that is observed in truncated Hbs from Bacillus subtilis [38] and truncated-II Hb from Mycobacterium tuberculosis [39], the hydrogen bonding interaction between axially bound CO and distal His55 in DHP is relatively weak. Nevertheless, previous structural, spectroscopic, and QM/MM studies all suggest that the structural perturbation in these Hbs is a result of enhanced flexibility of distal residue sidechain in the distal pocket [37]. Although the hydrogen bonding scheme changes from the ‘closed’ to the ‘open’ conformation of distal His55 in the CO complex, only a single set of kon(CO) and koff(CO) values was obtained in our binding kinetic measurements, indicating that the dynamic His55 conformational switch is in fast equilibrium and has a minimal effect on the CO-binding kinetics at ambient temperature. The minimal effect of the ‘closed’ ↔ ‘open’ conformational switch on CO binding is corroborated by the molecular dynamics (MD) calculation on the His55 mutants [37]. Although MD is limited to 20 ns, too short for the trajectory of any gaseous ligand in its binding to DHP, it indicates that residue 55 is locked in a single conformation in the two His55 DHP A mutants, ‘closed’ in H55V and ‘open’ in H55D mutants, respectively [37]. Nonetheless, the kinetics of CO binding in these two mutants remains unchanged compared with that of wt DHP A (Table 2).

In the crystallographic structure of DHP A CO complex (PDB 4GZG), the distal histidine of 6cLS DHP CO complex is primarily in the solvent exposed conformation, due to the interaction between His55 and the two propionate groups on the heme [37]. Resonance Raman data for the CO complex or DHP also showed one, not two, vibrational frequencies for Fe(II)–CO [40]. FTIR data also show one dominant C–O stretching at room temperature [36]. Similar to the cases of SWMb and model compound Fe(II)PP(1-MeIm) [41,42], the binding of both CO and NO to DHP A and B is a single-step process that leads to a stable 6c complex. The stable 6c NO complex in DHP is in contrast with those in sGC and some bacterial heme nitric oxide and oxygen-binding proteins (H-NOXs), which are unstable and subsequently convert into stable 5c NO–heme complexes [28,4346].

The binding parameters, kon/koff/KD of DHP (Tables 2 and 3), are very different from those of Mb and a model heme [28,29]. Despite their common globin protein folds and similar heme pockets (Supplementary Figure S2), the koff and kon constants of DHP are ∼2‒3 orders larger and ∼1 order smaller than those of SWMb, respectively. These differences lead to KDs of DHP that are ∼3 orders of magnitude larger than those of SWMb. On the other hand, the binding parameters of the first-binding step to form 6c complexes are quite similar to those of the bacterial H-NOXs [28,4346], even though they have totally different structural folds. Bioinformatic graphical analysis of the KD from many hemeproteins concludes that the main factors that caused the 8‒9 order range spread in KD values are heme distal steric hindrance and the proximal constraint [28,29]. Thus, such similarities in binding parameters suggest that the proximal strains and distal steric hindrance around heme are similar in DHP and H-NOXs. This point will be further discussed below.

The values of KD(O2) of DHP are 11 and 18 µM for DHP A and B, respectively (Table 3), and are reasonably close to the published values 7.7 and 4 µM, respectively, measured with a different method [24]. The differences can arise from the stopped-flow titration method used in the present study and the previous competition method between O2 ligand and t-butyl isocyanide. The relatively larger difference in KD(O2) of DHP B from the two studies is not clear at this time. The different KD(O2) between DHP A and B may be due to steric differences in the distal pockets, probably caused by the subtle effects of the amino acid differences between the two isozymes.

Single-wavelength stopped-flow data revealed that O2-binding to DHP has a large dissociation rate, kon[O2] + koff ≥ 500 s−1, in sharp contrast with that of SWMb, koff ∼ 15 s−1. Though not determined experimentally, the kon(O2) and koff(O2) constants of DHP were estimated by multiple nonlinear regression analysis of kinetic data acquired at various [O2] (Supplementary Figure S1). The association rate constants, kon(O2), obtained from nonlinear regression were similar for both DHP isozymes and His55 mutants, 3.0 × 107, 2.0 × 107, 1.9 × 107, and 9.9 × 106 M−1 s−1, for wt DHP A and B, H55D, and H55V DHP A, respectively (Supplementary Table S1). On the other hand, koff(O2)s of His55 mutants are noticeably larger than those of wt DHP A and B, 4.1 × 103 and 2.1 × 103 s−1 for H55V and H55D DHP A, respectively, and 2.7 × 102 and 2.9 × 102 s−1 for wt DHP A and B, respectively (Supplementary Table S1).

The kon(O2) and koff(O2) constants of DHP estimated by the computer fitting corroborates the H-bonding strength between O2 and εN of His55 in the distal heme pocket, implicated by the εN–Fe distance in DHP, which is intermediate between those of Mb and HRP. The O2 ligand is oriented in a ‘side on’ conformation in Mb to form a stable Fe(II)–O2 complex for oxygen storage or transfer and an ‘end on’ conformation in HRP to facilitate the heterolytic cleavage of O–O bond in its peroxidase catalysis [47]. The intermediate εN–Fe distance and the very flexible His55 switching dynamics in DHP not only provide it the versatility as a multifunctional hemeprotein, but also lead to a less stable Fe(II)–O2 intermediate than that of Mb. The capability in forming a Fe(II)–O2 intermediate affords DHP activities, including mono-oxygenase, oxidase, and dioxygenase; and its interaction with H2O2 enables it both peroxidase and peroxygenase activities.

Based on the rate constants and binding constants obtained for all three gaseous ligands, we are in a position to assess whether DHP follows the ‘sliding scale rule’. The rule based on the bioinformatic graphical analyses conducted for more than 100 hemeproteins and model compound provides trends that are conserved within various functional classes of hemeproteins [28,29].

The bioinformatic analysis provides five important conclusions regarding NO, CO, and O2 bindings to hemeproteins [28]. (1) The large KD(CO)/KD(NO) and KD(O2)/KD(CO) ratios, 103‒104, apply to 5c hemeproteins that have a neutral proximal histidine ligand, whereas much smaller differences in ligand affinities occur in other hemeproteins with proximal ligands of imidazolate, cysteine thiolate, or tyrosine phenolate anion. (2) Distal steric hindrance (typified by cytochrome c′) and proximal strain (illustrated by sGC) are the two dominating factors that can lead to dramatic 108‒109-fold changes in KD for all three gaseous ligands, but remarkably cause little change in the KD(CO)/KD(NO) and KD(O2)/KD(CO) ratios. These nearly constant ratios determine that log KD(NO)–log KD(CO)–log KD(O2) lines of different heme sensors remain parallel to each other in the ‘sliding scale rule’ plot. (3) As a result, for a heme sensor, determination of the KD for one ligand (i.e. CO) allows estimation of the KD values for the other two ligands, which is particularly useful for hemeproteins with very high (>10−3 M) KD values for O2 binding. (4) Although KD(CO) in a hemeprotein is determined by the distal steric hindrance and proximal strain, some hemeproteins have other structural elements which specifically enhance their affinities for O2 and/or NO. The presence of an H-bond donor(s) on the distal side of heme preferentially enhances(enhance) O2 affinity, but have little or no effect on CO or NO binding, leading to deviation from the ‘sliding scale rule’ at the log KD(O2) end. (5) The ultra-high apparent affinities of some heme sensors for NO, such as sGC and certain bacterial H-NOXs, are due to multiple-step NO binding in which the unstable 6c NO–heme–His complex either promptly converts to a 5c NO–heme complex or react rapidly with secondary NO to generate a 5c NO–heme complex. The KDs of the initial 6c Fe(II)–NO complexes in these heme sensors follow the ‘sliding scale rule’; however, the subsequently generated 5c NO complexes have significantly smaller koff constants, leading to much lower apparent KDs [28,29].

The graphical analysis shown in Figures 5 and 6 not only reveals the gaseous ligand selectivity of DHP, but also its binding behavior relative to other hemeproteins, some of them with very different functions and protein folds. The log KDs of wt, H55D and H55V DHP A, and wt DHP B are plotted versus ligand type, together with either those of various globins (Figure 5) or nonglobins (Figure 6). The nonglobins include H-NOXs from Nostoc sp. (Ns H-NOX) and Thermoanaerobacter tengcongensis (Tt H-NOX), cytochrome c′, and model heme (Figure 6). The lines connecting log KD(NO) and log KD(CO) for each hemeprotein are parallel aside from slight deviations due to the random variations during the measurements or slight differences in the experimental conditions.

Relationship of log KD, log kon, and log koff of NO, CO, and O2 for DHP and other globins versus ligand type.

Figure 5.
Relationship of log KD, log kon, and log koff of NO, CO, and O2 for DHP and other globins versus ligand type.

The measured values of KD (A), kon (B), and koff (C) for NO/CO/O2 of ferrous DHP are plotted on a logarithmic scale versus ligand type. Parameters measured for different globins, including SWMb and mutants, soybean leghemoglobin (Lb) and its H61L mutant, and flavohemoglobin (FHb), are plotted for comparisons. The values of the parameters are from [28] and the references therein. DHP A, black star; DHP B, red star; H55D DHP A, dark yellow star; H55V DHP A, green star; SWMb, black circle; H64Q Mb, red circle; H64V Mb, green circle; Lb, black triangle; H61L Lb, red triangle; FHb, blue square.

Figure 5.
Relationship of log KD, log kon, and log koff of NO, CO, and O2 for DHP and other globins versus ligand type.

The measured values of KD (A), kon (B), and koff (C) for NO/CO/O2 of ferrous DHP are plotted on a logarithmic scale versus ligand type. Parameters measured for different globins, including SWMb and mutants, soybean leghemoglobin (Lb) and its H61L mutant, and flavohemoglobin (FHb), are plotted for comparisons. The values of the parameters are from [28] and the references therein. DHP A, black star; DHP B, red star; H55D DHP A, dark yellow star; H55V DHP A, green star; SWMb, black circle; H64Q Mb, red circle; H64V Mb, green circle; Lb, black triangle; H61L Lb, red triangle; FHb, blue square.

Relationship of log KD, log kon, and log koff of NO, CO, and O2 for DHP and nonglobin hemeproteins versus ligand type.

Figure 6.
Relationship of log KD, log kon, and log koff of NO, CO, and O2 for DHP and nonglobin hemeproteins versus ligand type.

The measured values of KD (A), kon (B), and koff (C) for NO/CO/O2 on ferrous DHP are plotted on a logarithmic scale versus ligand type, in comparisons with those of wt and L16A cyt c′, Ns and Tt H-NOXs, and heme model Fe(II)PP(1-MeIm) to show the large dynamic range of KD, kon, and koff values modulated by the different heme-binding environments. DHP A, black star; DHP B, red star; H55D DHP A, green star; H55V DHP A, dark yellow star; human sGC, black square; Ns H-NOX, red diamond; Tt H-NOX, green diamond; Fe(II) PP(1-MeIm), cyan hexagon; cyt c′, red triangle; L16A cyt c′, black circle.

Figure 6.
Relationship of log KD, log kon, and log koff of NO, CO, and O2 for DHP and nonglobin hemeproteins versus ligand type.

The measured values of KD (A), kon (B), and koff (C) for NO/CO/O2 on ferrous DHP are plotted on a logarithmic scale versus ligand type, in comparisons with those of wt and L16A cyt c′, Ns and Tt H-NOXs, and heme model Fe(II)PP(1-MeIm) to show the large dynamic range of KD, kon, and koff values modulated by the different heme-binding environments. DHP A, black star; DHP B, red star; H55D DHP A, green star; H55V DHP A, dark yellow star; human sGC, black square; Ns H-NOX, red diamond; Tt H-NOX, green diamond; Fe(II) PP(1-MeIm), cyan hexagon; cyt c′, red triangle; L16A cyt c′, black circle.

For H64V SWMb and Ns H-NOX, which do not have a H-bond donor to stabilize Fe(II)–O2, their log KD(NO), log KD(CO), and log KD(O2) almost fall on straight lines (Figures 5A and 6A). On the other hand, Fe(II)–O2 complexes in SWMb and Tt H-NOX are stabilized by distal His64 or Tyr140, respectively, leading to downward sloped log KD(CO)–log KD(O2) lines relative to their upward log KD(NO)–log KD(CO) lines (Figures 5A and 6A). In wt DHP A and B, Fe(II)–O2 is also stabilized by H-bonding to distal His55, resulting in lower log KD(O2)s which deviates from the extrapolation of the corresponding log KD(NO)–log KD(CO) lines (Figure 5A). In H55D DHP A, the Fe(II)–O2 is less stabilized compared with wt DHP A due to the loss of distal His55, reflected by the ∼8-fold increase in koff(O2) of H55D DHP A compared with that of wt DHP A (Supplementary Table S1). Nonetheless, a hydrogen bonding network through H2O may still provide some stabilization for the Fe(II)–O2 complex [48], lowering KD(O2) for H55D DHP A from the extrapolation of its log KD(NO)–log KD(CO) line to 117 µM. In H55V DHP A, KD(O2) further increases to 460 µM, presumably due to the substitution of His55 with a nonpolar valine residue, which is not capable of supporting hydrogen bonding through its sidechain, as indicated by another 2-fold increase in koff(O2) in H55V DHP A compared with that of the H55D mutant (Supplementary Table S1).

Thus, from H55V, H55D, to wt DHP A, a progressively increasing effect of the H-bond donor strength in stabilizing the Fe(II)–O2 complex is clearly illustrated. A similar systematic effect on O2 affinity can be observed by manipulating the H-bonding strength in wt, H64Q, and H64V Mb where KD(O2) increases from 0.88 to 5.6 to 91 µM (Figure 5A) [4951]. The effects of H-bonding donor observed for both DHP and Mb are only selective for O2 binding and show essentially no effect on CO or NO binding. Overall, the gaseous ligand selectivity of DHP agrees well with the ‘sliding scale rule’ hypothesis. By removing the H-bonding donor, as observed in the H55V mutant, DHP exhibits good selectivity between NO/CO and CO/O2 pairs, each with 3−4 orders separation in binding affinities (Table 3 and Figure 6).

The selective binding of DHP to the gaseous ligands presents a sharp contrast with those of typical peroxidases such as HRP and CCP, which show very poor gaseous ligand selectivity due to the presence of a strong electron-donating imidazolate proximal ligand. DHP is structurally more similar to globins, such as Mb and Hb (Supplementary Figure S2), than peroxidases despite its peroxidase activity. The presence of a neutral proximal imidazole ligand in DHP confers a high degree of selectivity among NO, CO, and O2. Based on this rationale, other enzymes that show peroxidase activity and possess a neutral histidine proximal ligand should also show good discrimination among the three gaseous ligands. Prostaglandin H synthase (also known as a cyclooxygenase) [52] will provide a good test of this hypothesis since it has a neutral proximal histidine heme ligand and is a much more efficient peroxidase than DHP.

Conclusions

We have measured the binding of three gaseous ligands, CO, NO, and O2, to DHP and two of its mutants to further test the ‘sliding scale rule’ hypothesis. Recent work to assess six different H-NOXs including sGC with the same type of protein scaffold in order to gain insights into the functional differences among them has revealed significant differences especially in the multiple-step NO binding, the extent of oxygen binding, and autoxidation [29,34,4446]. The current study on DHP provides insights into a new example obeying the ‘sliding scale rule’ in a hemeprotein with a very different structural fold, a globin exhibiting multiple enzyme functions. The ratio of KD(NO)/KD(CO) and KD(CO)/KD(O2) remains 3‒4 orders regardless of the protein fold. The fact that DHP conforms well to the ‘sliding scale rule’ is consistent with its neutral proximal histidine heme ligand and the results of many experiments, suggesting a large and relatively open-distal pocket in DHP [53,54]. Therefore, the binding of DHP to diatomic gaseous ligands follows single-step binding kinetics with no intermediates, but at the same time the dissociation constant is an order of magnitude larger than those of other typical globins.

Abbreviations

     
  • 6cLS

    six-coordinated low-spin

  •  
  • CCP

    cytochrome c peroxidase

  •  
  • DHP

    dehaloperoxidase–hemoglobin

  •  
  • FHb

    flavohemoglobin

  •  
  • Hb

    hemoglobin

  •  
  • H-NOX

    heme nitric oxide and oxygen-binding protein

  •  
  • HRP

    horseradish peroxidase

  •  
  • Lb

    soybean leghemoglobin

  •  
  • Mb

    myoglobin

  •  
  • MD

    molecular dynamics

  •  
  • Ns H-NOX

    H-NOX from Nostoc sp. PCC7120

  •  
  • sGC

    human soluble guanylyl cyclase

  •  
  • SWMb

    sperm whale myoglobin

  •  
  • Tt H-NOX

    H-NOX from Thermoanaerobacter tengcongensis.

Author Contribution

G.W. performed the experimental work and wrote the manuscript. J.Z. expressed and purified the proteins and revised the manuscript. S.F. and A.-L.T. conceived the study and wrote and revised the manuscript.

Funding

This work was supported by NIH grant [NS094535] to A.-L.T. and NSF grant [CHE-1507947] to S.F.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
Chen
,
Y.P.
,
Woodin
,
S.A.
,
Lincoln
,
D.E.
and
Lovell
,
C.R.
(
1996
)
An unusual dehalogenating peroxidase from the marine terebellid polychaete Amphitrite ornata
.
J. Biol. Chem.
271
,
4609
4612
doi:
2
Han
,
K.
,
Woodin
,
S.A.
,
Lincoln
,
D.E.
,
Fielman
,
K.T.
and
Ely
,
B.
(
2001
)
Amphitrite ornata, a marine worm, contains two dehaloperoxidase genes
.
Mar. Biotechnol.
3
,
287
292
doi:
3
Ma
,
H.
,
Thompson
,
M.K.
,
Gaff
,
J.
and
Franzen
,
S.
(
2010
)
Kinetic analysis of a naturally occurring bioremediation enzyme: dehaloperoxidase-hemoglobin from Amphitrite ornata
.
J. Phys. Chem. B
114
,
13823
13829
doi:
4
Zhao
,
J.
,
Lu
,
C.
and
Franzen
,
S.
(
2015
)
Distinct enzyme–substrate interactions revealed by two dimensional kinetic comparison between dehaloperoxidase-hemoglobin and horseradish peroxidase
.
J. Phys. Chem. B
119
,
12828
12837
doi:
5
Franzen
,
S.
,
Thompson
,
M.K.
and
Ghiladi
,
R.A.
(
2012
)
The dehaloperoxidase paradox
.
Biochim. Biophys. Acta, Proteins Proteomics
1824
,
578
588
doi:
6
D'Antonio
,
J.
,
D'Antonio
,
E.L.
,
Thompson
,
M.K.
,
Bowden
,
E.F.
,
Franzen
,
S.
,
Smirnova
,
T.
et al. 
(
2010
)
Spectroscopic and mechanistic investigations of dehaloperoxidase B from Amphitrite ornata
.
Biochemistry
49
,
6600
6616
doi:
7
D'Antonio
,
E.L.
,
Chen
,
T.K.
,
Turner
,
A.H.
,
Santiago-Capeles
,
L.
and
Bowden
,
E.F.
(
2013
)
Voltammetry of dehaloperoxidase on self-assembled monolayers: reversible adsorptive immobilization of a globin
.
Electrochem. Commun.
26
,
67
70
doi:
8
D'Antonio
,
E.L.
,
Bowden
,
E.F.
and
Franzen
,
S.
(
2012
)
Thin-layer spectroelectrochemistry of the Fe(III)/Fe(II) redox reaction of dehaloperoxidase-hemoglobin
.
J. Electroanal. Chem.
668
,
37
43
doi:
9
Ayala
,
M.
(
2010
) Redox potential of peroxidases. In
Biocatalysis Based on Heme Peroxidases: Peroxidases as Potential Industrial Biocatalysts
(
Torres
,
E.
and
Ayala
,
M.
, eds), pp.
61
77
.
Springer
,
Berlin, Heidelberg
10
D'Antonio
,
J.
and
Ghiladi
,
R.A.
(
2011
)
Reactivity of deoxy- and oxyferrous dehaloperoxidase B from Amphitrite ornata: identification of compound II and its ferrous–hydroperoxide precursor
.
Biochemistry
50
,
5999
6011
doi:
11
Chen
,
Z.
,
de Serrano
,
V.
,
Betts
,
L.
and
Franzen
,
S.
(
2009
)
Distal histidine conformational flexibility in dehaloperoxidase from Amphitrite ornata
.
Acta Crystal. D Biol. Crystallogr.
65
,
34
40
doi:
12
Du
,
J.
,
Huang
,
X.
,
Sun
,
S.
,
Wang
,
C.
,
Lebioda
,
L.
and
Dawson
,
J.H.
(
2011
)
Amphitrite ornata dehaloperoxidase (DHP): investigations of structural factors that influence the mechanism of halophenol dehalogenation using ‘peroxidase-like’ myoglobin mutants and ‘myoglobin-like’ DHP mutants
.
Biochemistry
50
,
8172
8180
doi:
13
Osborne
,
R.L.
,
Raner
,
G.M.
,
Hager
,
L.P.
and
Dawson
,
J.H.
(
2006
)
C. fumago chloroperoxidase is also a dehaloperoxidase: oxidative dehalogenation of halophenols
.
J. Am. Chem. Soc.
128
,
1036
1037
doi:
14
Sumithran
,
S.
,
Sono
,
M.
,
Raner
,
G.M.
and
Dawson
,
J.H.
(
2012
)
Single turnover studies of oxidative halophenol dehalogenation by horseradish peroxidase reveal a mechanism involving two consecutive one electron steps: toward a functional halophenol bioremediation catalyst
.
J. Inorg. Biochem.
117
,
316
321
doi:
15
Franzen
,
S.
,
Belyea
,
J.
,
Gilvey
,
L.B.
,
Davis
,
M.F.
,
Chaudhary
,
C.E.
,
Sit
,
T.L.
et al. 
(
2006
)
Proximal cavity, distal histidine, and substrate hydrogen-bonding mutations modulate the activity of Amphitrite ornata dehaloperoxidase
.
Biochemistry
45
,
9085
9094
doi:
16
Nicoletti
,
F.P.
,
Thompson
,
M.K.
,
Howes
,
B.D.
,
Franzen
,
S.
and
Smulevich
,
G.
(
2010
)
New insights into the role of distal histidine flexibility in ligand stabilization of dehaloperoxidase–hemoglobin from Amphitrite ornata
.
Biochemistry
49
,
1903
1912
doi:
17
Zhao
,
J.
,
de Serrano
,
V.
,
Zhao
,
J.
,
Le
,
P.
and
Franzen
,
S.
(
2013
)
Structural and kinetic study of an internal substrate binding site in dehaloperoxidase-hemoglobin A from Amphitrite ornata
.
Biochemistry
52
,
2427
2439
doi:
18
Du
,
J.
,
Sono
,
M.
and
Dawson
,
J.H.
(
2010
)
Functional switching of Amphitrite ornata dehaloperoxidase from O2-binding globin to peroxidase enzyme facilitated by halophenol substrate and H2O2
.
Biochemistry
49
,
6064
6069
doi:
19
Sun
,
S.
,
Sono
,
M.
,
Du
,
J.
and
Dawson
,
J.H.
(
2014
)
Evidence of the direct involvement of the substrate TCP radical in functional switching from oxyferrous O2 carrier to ferric peroxidase in the dual-function hemoglobin/dehaloperoxidase from Amphitrite ornata
.
Biochemistry
53
,
4956
4969
doi:
20
Zhao
,
J.
,
Zhao
,
J.
and
Franzen
,
S.
(
2013
)
The regulatory implications of hydroquinone for the multifunctional enzyme dehaloperoxidase-hemoglobin from Amphitrite ornata
.
J. Phy. Chem. B
117
,
14615
14624
doi:
21
Barrios
,
D.A.
,
D'Antonio
,
J.
,
McCombs
,
N.L.
,
Zhao
,
J.
,
Franzen
,
S.
,
Schmidt
,
A.C.
et al. 
(
2014
)
Peroxygenase and oxidase activities of dehaloperoxidase-hemoglobin from Amphitrite ornata
.
J. Am. Chem. Soc.
136
,
7914
7925
doi:
22
McCombs
,
N.L.
,
D'Antonio
,
J.
,
Barrios
,
D.A.
,
Carey
,
L.M.
and
Ghiladi
,
R.A.
(
2016
)
Nonmicrobial nitrophenol degradation via peroxygenase activity of dehaloperoxidase-hemoglobin from Amphitrite ornata
.
Biochemistry
55
,
2465
2478
doi:
23
Ghiladi
,
R.A.
,
McCombs
,
N.L.
and
Carey
,
L.M.
(
2016
)
The multiple enzymatic activities of the hemoglobin dehaloperoxidase from
Amphitrite Ornata. 19th Conference on Oxygen-binding and Sensing Proteins
,
Hamburg, Germany
, pp.
58
24
Sun
,
S.
,
Sono
,
M.
,
Wang
,
C.
,
Du
,
J.
,
Lebioda
,
L.
and
Dawson
,
J.H.
(
2014
)
Influence of heme environment structure on dioxygen affinity for the dual function Amphitrite ornata hemoglobin/dehaloperoxidase. Insights into the evolutional structure–function adaptations
.
Arch. Biochem. Biophys.
545
,
108
115
doi:
25
Carver
,
T.E.
,
Brantley
, Jr,
R.E.
,
Singleton
,
E.W.
,
Arduini
,
R.M.
,
Quillin
,
M.L.
,
Phillips
, Jr,
G.N.
et al. 
(
1992
)
A novel site-directed mutant of myoglobin with an unusually high O2 affinity and low autooxidation rate
.
J. Biol. Chem.
267
,
14443
14450
PMID:
[PubMed]
26
Draghi
,
F.
,
Miele
,
A.E.
,
Travaglini-Allocatelli
,
C.
,
Vallone
,
B.
,
Brunori
,
M.
,
Gibson
,
Q.H.
et al. 
(
2002
)
Controlling ligand binding in myoglobin by mutagenesis
.
J. Biol. Chem.
277
,
7509
7519
doi:
27
Quillin
,
M.L.
,
Li
,
T.
,
Olson
,
J.S.
,
Phillips
, Jr,
G.N.
,
Dou
,
Y.
,
Ikeda-Saito
,
M.
et al. 
(
1995
)
Structural and functional effects of apolar mutations of the distal valine in myoglobin
.
J. Mol. Biol.
245
,
416
436
doi:
28
Tsai
,
A.L.
,
Berka
,
V.
,
Martin
,
E.
and
Olson
,
J.S.
(
2012
)
A ‘Sliding-scale rule’ for selectivity between NO, CO and O2 by heme protein sensors
.
Biochemistry
51
,
172
186
doi:
29
Tsai
,
A.L.
,
Martin
,
E.
,
Berka
,
V.
and
Olson
,
J.S.
(
2012
)
How do heme-protein sensors exclude oxygen? Lessons learned from cytochrome c′, Nostoc puntiforme heme nitric oxide/oxygen-binding domain, and soluble guanylyl cyclase.
Antioxid Redox Signal.
17
,
1246
1263
doi:
30
Berry
,
E.A.
and
Trumpower
,
B.L.
(
1987
)
Simultaneous determination of hemes a, b, and c from pyridine hemochrome spectra
.
Anal. Biochem.
161
,
1
15
doi:
31
Capellos
,
C.
and
Bielski
,
B.H.J.
(
1972
)
Kinetic Systems, Mathematical Description of Chemical Kinetics in Solution
.
Wiley Interscience
,
New York
32
Lewis
,
R.S.
and
Deen
,
W.M.
(
1994
)
Kinetics of the reaction of nitric oxide with oxygen in aqueous solutions
.
Chem. Res. Toxicol.
7
,
568
574
doi:
33
Brent
,
R.P.
(
1973
)
Algorithms for Minimization Without Derivatives
.
Prentice-Hall
,
Englewood Cliffs, NJ
34
Wu
,
G.
,
Liu
,
W.
,
Berka
,
V.
and
Tsai
,
A.L.
(
2017
)
Gaseous ligand selectivity of the H-NOX sensor protein from Shewanella oneidensis and comparison to those of other bacterial H-NOXs and soluble guanylyl cyclase
.
Biochimie
140
,
82
92
doi:
35
Yonetani
,
H.
and
Yamamoto
,
H.
(
1973
)
Optical and electron paramagnetic resonance properties of the nitric oxide compounds of cytochrome c peroxidase and horseradish peroxidase
. In
Oxidases and Related Redox Systems
(
King
,
T.E.
,
Mason
,
H.S.
and
Morrison
,
M.
, eds), pp.
279
299
.
University Park Press
,
Baltimore, MD
36
Nienhaus
,
K.
,
Nickel
,
E.
,
Davis
,
M.F.
,
Franzen
,
S.
and
Nienhaus
,
G.U.
(
2008
)
Determinants of substrate internalization in the distal pocket of dehaloperoxidase hemoglobin of Amphitrite ornata
.
Biochemistry
47
,
12985
12994
doi:
37
Zhao
,
J.
,
de Serrano
,
V.
and
Franzen
,
S.
(
2014
)
A model for the flexibility of the distal histidine in dehaloperoxidase-hemoglobin A based on X-ray crystal structures of the carbon monoxide adduct
.
Biochemistry
53
,
2474
2482
doi:
38
Feis
,
A.
,
Lapini
,
A.
,
Catacchio
,
B.
,
Brogioni
,
S.
,
Foggi
,
P.
,
Chiancone
,
E.
et al. 
(
2008
)
Unusually strong H-bonding to the heme ligand and fast geminate recombination dynamics of the carbon monoxide complex of Bacillus subtilis truncated hemoglobin
.
Biochemistry
47
,
902
910
doi:
39
Guallar
,
V.
,
Lu
,
C.
,
Borrelli
,
K.
,
Egawa
,
T.
and
Yeh
,
S.-R.
(
2009
)
Ligand migration in the truncated hemoglobin-II from Mycobacterium tuberculosis: the role of G8 tryptophan
.
J. Biol. Chem.
284
,
3106
3116
doi:
40
Franzen
,
S.
,
Roach
,
M.P.
,
Chen
,
Y.-P.
,
Dyer
,
R.B.
,
Woodruff
,
W.H.
and
Dawson
,
J.H.
(
1998
)
The unusual reactivities of Amphitrite ornata dehaloperoxidase and Notomastus lobatus chloroperoxidase do not arise from a histidine imidazolate proximal heme iron ligand
.
J. Am. Chem. Soc.
120
,
4658
4661
doi:
41
Hoshino
,
M.
,
Ozawa
,
K.
,
Seki
,
H.
and
Ford
,
P.C.
(
1993
)
Photochemistry of nitric oxide adducts of water-soluble iron(III) porphyrin and ferrihemoproteins studied by nanosecond laser photolysis
.
J. Am. Chem. Soc.
115
,
9568
9575
doi:
42
Rose
,
E.J.
,
Venkatasubramanian
,
P.N.
,
Swartz
,
J.C.
,
Jones
,
R.D.
,
Basolo
,
F.
and
Hoffman
,
B.M.
(
1982
)
Carbon monoxide binding kinetics in ‘capped’ porphyrin compounds
.
Proc. Natl Acad. Sci. U.S.A.
79
,
5742
5745
doi:
43
Tsai
,
A.L.
(
1994
)
How does NO activate hemeproteins?
FEBS Lett.
341
,
141
145
doi:
44
Wu
,
G.
,
Liu
,
W.
,
Berka
,
V.
and
Tsai
,
A.L.
(
2013
)
The selectivity of Vibrio cholerae H-NOX for gaseous ligands follows the ‘sliding scale rule’ hypothesis. Ligand interactions with both ferrous and ferric Vc H-NOX
.
Biochemistry
52
,
9432
9446
doi:
45
Wu
,
G.
,
Liu
,
W.
,
Berka
,
V.
and
Tsai
,
A.L.
(
2015
)
H-NOX from Clostridium botulinum, like H-NOX from Thermoanaerobacter tengcongensis, binds oxygen but with a less stable oxyferrous heme intermediate
,
Biochemistry
54
,
7098
7109
doi:
46
Tsai
,
A.L.
,
Berka
,
V.
,
Martin
,
F.
,
Ma
,
X.
,
van den Akker
,
F.
,
Fabian
,
M.
et al. 
(
2010
)
Is Nostoc H-NOX a NO sensor or redox switch?
Biochemistry
49
,
6587
6599
doi:
47
Poulos
,
T.L.
(
2014
)
Heme enzyme structure and function
.
Chem. Rev.
114
,
3919
3962
doi:
48
Zhao
,
J.
,
de Serrano
,
V.
,
Dumarieh
,
R.
,
Thompson
,
M.
,
Ghiladi
,
R.A.
and
Franzen
,
S.
(
2012
)
The role of the distal histidine in H2O2 activation and heme protection in both peroxidase and globin functions
.
J. Phys. Chem. B
116
,
12065
12077
doi:
49
Dou
,
Y.
,
Maillett
,
D.H.
,
Eich
,
R.F.
and
Olson
,
J.S.
(
2002
)
Myoglobin as a model system for designing heme protein based blood substitutes
.
Biophys. Chem.
98
,
127
148
doi:
50
Eich
,
R.F.
(
1997
)
Reactions of Nitric Oxide with Myoglobin
.
Biochemistry and Cell Biology, Rice University
,
Houston
51
Foley
,
E.W.
(
2005
)
Physiologically Relevant Reactions of Myoglobin and Hemoglobin with NO
.
Biochemistry and Cell Biology, Rice University
,
Houston
52
Tsai
,
A.L.
and
Kulmacz
,
R.J.
(
2010
)
Prostaglandin H synthase: resolved and unresolved mechanistic issues
.
Arch. Biochem. Biophys.
493
,
103
124
doi:
53
Zhao
,
J.
,
Moretto
,
J.
,
Le
,
P.
and
Franzen
,
S.
(
2015
)
Measurement of internal substrate binding in dehaloperoxidase–hemoglobin by competition with the heme–fluoride binding equilibrium
.
J. Phys. Chem. B
119
,
2827
2838
doi:
54
de Serrano
,
V.
and
Franzen
,
S.
(
2012
)
Structural evidence for stabilization of inhibitor binding by a protein cavity in the dehaloperoxidase-hemoglobin from Amphitrite ornata
.
Peptide Sci.
98
,
27
35
doi: