Nitrite reduction by CoII and MnII substituted myoglobins

Journal of Inorganic Biochemistry 107 (2012) 47–53

Contents lists available at SciVerse ScienceDirect

Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio

Nitrite reduction by Co II and Mn II substituted myoglobins Towards understanding necessary components of Mb nitrite reductase activity Julie L. Heinecke a, Jun Yi b, Jose Clayston Melo Pereira a, George B. Richter-Addo b,⁎, Peter C. Ford b,⁎⁎ a b

Department of Chemistry and Biochemistry, University of California at Santa Barbara, Santa Barbara, CA 93106-9510, United States Department of Chemistry and Biochemistry, University of Oklahoma, 101 Stephenson Parkway, Norman, OK 73019, United States

a r t i c l e

i n f o

Article history: Received 25 May 2011 Received in revised form 28 September 2011 Accepted 17 October 2011 Available online 25 October 2011 Keywords: Myoglobin Manganese Cobalt Nitrite Reduction Kinetics

a b s t r a c t Nitrite reduction to nitric oxide by heme proteins is drawing increasing attention as a protective mechanism to hypoxic injury in mammalian physiology. Here we probe the nitrite reductase (NiR) activities of manganese(II)- and cobalt(II)-substituted myoglobins, and compare with data obtained previously for the iron(II) analog wt MbII. Both Mn IIMb and Co IIMb displayed NiR activity, and it was shown that the kinetics are first order each in [protein], [nitrite], and [H+], as previously determined for the Fe II analog wt Mb II. The second order rate constants (k2) at pH 7.4 and T = 25 °C, were 0.0066 and 0.015 M− 1 s− 1 for Co IIMb and MnIIMb, respectively, both orders of magnitude slower than the k2 (6 M− 1 s− 1) for wt MbII. The final reaction products for Mn IIMb consisted of a mixture of the nitrosyl MnIIMb(NO) and MnIIIMb, similar to the products from the analogous NiR reaction by wt Mb. In contrast, the products of NiR by Co IIMb were found to be the nitrito complex Co IIIMb(ONO−) plus roughly an equivalent of free NO. The differences can be attributed in part to the stronger coordination of inorganic nitrite to Co IIIMb as reflected in the respective MIIIMb(ONO −) formation constants Knitrite: 2100 M− 1 (Co III) and b~0.4 M− 1 (MnIII). We also report the formation constants (3.7 and 30 M− 1, respectively) for the nitrite complexes of the mutant metmyoglobins H64V Mb III(NO2−) and H64V/ V67R MbIII(ONO −) and a Knitrite revised value (120 M− 1) for the nitrite complex of wt metMb. The respective Knitrite values for the three ferric proteins emphasize the importance of a H-bonding residue, such as His64 in the MbIII distal pocket or the Arg67 in H64V/V67R Mb III, in stabilizing nitrite coordination. Notably, the NiR activities of the corresponding ferrous Mbs follow a similar sequence suggesting that nitrite binding to these centers are analogously affected by the H-bonding residues. © 2011 Elsevier Inc. All rights reserved.

1. Introduction Nitric oxide (NO) in mammals is produced during aerobic respiration from the enzyme nitric oxide synthase and has many biological roles including vasodilation, signaling, and regulation of cellular function [1–4]. It has been recently established that under conditions of low oxygen tension, such as organ ischemia, nitrite-derived NO is responsible for similar signaling responses. Although enzyme-independent pathways exist for the nitrite-to-NO conversion, the reaction is greatly enhanced by various metalloproteins [5,6]. This in turn has led to increased

⁎ Correspondence to: G.B. Richter-Addo, Department of Chemistry and Biochemistry, University of Oklahoma, 101 Stephenson Parkway, Norman, OK 73019, United States. Tel.: +1 405 325 6401; fax: +1 405 325 6111. ⁎⁎ Correspondence to: P.C. Ford, Department of Chemistry and Biochemistry, University of Oklahoma, 101 Stephenson Parkway, Norman, OK 73019, United States. Tel.: + 1 805 893 2443; fax: + 1 805 893 4120. E-mail addresses: [email protected] (G.B. Richter-Addo), [email protected] (P.C. Ford). 0162-0134/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2011.10.006

attention to mammalian proteins that may display nitrite reductase (NiR) activity (Eq. (1)) [7–15]. −

−

þ

NO2 þ e þ 2H →NO þ H2 O

ð1Þ

The potential physiological roles of mammalian NiRs add imperative to understanding the specific protein components and traits that dictate such activity. For example, accessibility of nitrite to the ferrous site of cross-linked hemoglobin-based oxygen carriers has been argued to limit the nitrite reduction rate [16]. In another example, sickle cell hemoglobin (HbS) has higher NiR activity than hemoglobin A (HbA), and the difference correlates with the more favorable Fe II to Fe III half-cell potential of HbS [17]. In these laboratories, we have shown that modification of distal pocket residues regulates NiR activities of ferrous myoglobin (Mb II) as well as the mode of nitrite coordination in ferric myoglobins (Mb III) [18,19]. Thus, while wild type (wt) Mb III binds NO2− in the distal site as the O-nitrito (i.e., Obound nitrite) isomer, the N-nitro isomer is seen in the Mb III mutant H64V, where the H-bonding His64 residue is replaced by a non-polar valine. Furthermore, nitrite reduction by wt Mb II is significantly faster

48

J.L. Heinecke et al. / Journal of Inorganic Biochemistry 107 (2012) 47–53

than by H64V Mb II, indicating that His64 is crucial for the NiR activity [18]. Perhaps an even greater perturbation to the properties of a protein such as Mb would be expected from substituting another metal center for the heme iron. In this context, the present study compares the NiR reactivity of wt Mb II to those of the metal-substituted myoglobins MnIIMb and CoIIMb, where the FeII of wt MbII has been replaced by the metal centers MnII or Co II. Various properties of the protein change as a result of such metal substitution, including the binding mode and affinity of nitrite, the redox potential of the protein, and the d-electron configuration. Synthetic MnIII porphyrin complexes bind nitrite via the O-nitrito mode [20,21], whereas the CoIII porphyrins [22–25] and most of the FeIII and FeII analogs bind nitrite as the N-nitro isomers [25]. Nonetheless, the nitrite complexes of the proteins MnIIIMb and CoIIIMb are in the O-nitrito form [26] and thus in the same structural class as wt Mb III [27], although the strength of nitrite binding would be expected to differ among these species. Described here are studies to assess the formation constants for the nitrite complexes of Mn IIIMb, Co IIIMb and wt Mb III as well as those for the metMb distal pocket mutants H64V Mb III and H64V/ V67R Mb III. We have determined the kinetics for nitrite reduction by the reduced proteins Mn IIMb and Co IIMb, and compare these data with that obtained for wt Mb II. 2. Materials and methods 2.1. Materials Horse heart myoglobin (hh-Mb) was obtained from Sigma. Manganese(III) protoporphyrin IX chloride and cobalt(III) protoporphyrin IX chloride were purchased from Frontier Scientific. Apo-myoglobin (apo-Mb) was prepared using the method of Yonetani [28] and Teale [29]. The apo-Mb was reconstituted with Mn III(PPIX) and Co III (PPIX) to give Mn IIIMb or Co IIIMb, respectively, using the method of Yonetani and Asakura [30]. A 1.2-fold of CoIII(PPIX) was slowly added into a solution of ~2 mM apo-Mb in 10 mM sodium phosphate buffer at pH 6.8. The protein solution was stirred on ice for 30 min and then was centrifuged at 5000 g for 10 min. The supernatant was then loaded on a CM-52 column equilibrated with 10 mM phosphate buffer at pH 6.8. After washing the column with a two-fold volume equivalent of buffer, the protein was eluted with a pH gradient developed by using an equal volume of 10 mM sodium phosphate buffer (pH 6.8 at 4 °C) and 10 mM sodium dibasic phosphate. The fractions displaying a ASoret/A280 ratio of >4.0 were pooled and concentrated for storage. The purity of CoIIIMb was checked by SDS (12%) PAGE. MnIIIMb was prepared similarly using MnIII(PPIX). Preparations of the metMb mutants H64V Mb III and H64V/V67R Mb III have been described previously [18]. Nitric oxide (nitrogen monoxide) (99.5%) was purchased from Praxair and plumbed to a customized Schlenk line using welded stainless steel lines and fittings. The NO was purified of NO2, N2O3 and N2O as described elsewhere [31]. All gas pressures were measured using a mercury monometer connected to the Schlenk line.

solutions were maintained at pH 7.4 (100 mM phosphate buffer) and at 25 °C under deaerated conditions in a quartz cell adapted for sample preparation on a Schlenk vacuum line. High grade argon (4.8), which was first passed through a commercial grade O2 scrubber (OxiClear®) to give dioxygen levels below 50 ppb, was used for all Schlenk line manipulations. Additions to these closed systems were accomplished using gas tight syringes previously purged with high grade argon by injecting through a Thermogreen LB-2 septa (Sigma) sealed electrode adaptor. Solutions of the reduced myoglobin derivatives MnIIMb and CoIIMb were prepared as follows. A concentrated sample of the respective MIIIMb protein (5–6 μM) and protocatechuate 3,4-dioxygenase (0.2 units, Sigma) in 1 mL phosphate buffer solution was deaerated by entraining with argon for 20 min. To this solution, solid sodium dithionite (1.0 mg, 5.7 μmol) was added to the reaction mixture and bubbled with argon for 5 more minutes. The color of the protein solution changed to yellow (Mn) or pink (Co), indicating that reduction occurred. This solution was then passed through a G-25 Sephadex column (to remove the excess dithionite and byproducts) on a Schlenk line using deoxygenated pH 7.4 phosphate buffer (100 mM) as the eluent, and the colored fraction was collected into a gas tight UV/Vis cell equipped with an injection port. This cell was previously deaerated and contained 3,4 dihydroxybenzoic acid (~1.0 mg, ~6.5 μmol) as a substrate for the dioxygenase, which is present to deplete spurious O2. The final solution volume ranged between 1.9 and 2.4 mL, and this was used to calculate the amount of NaNO2 needed to obtain the desired nitrite concentrations. 3. Results and discussion 3.1. Chemical reduction of Mn IIIMb and Co IIIMb It was previously reported from these laboratories that cysteine is a mild and effective reducing agent for the ferriheme proteins wt MbIII and the mutants H64V MbIII and H64V/V67R MbIII [18]. However, cysteine (1 mM) is not effective in reducing MnIIIMb, a difference we attribute to the less favorable reduction potential of Mn IIIMb (E° = −170 and −10 mV vs. NHE [32]) compared to wt metMb (E° = 61 mV) [33]. As a consequence, it was necessary to use dithionite as the reducing agent. Fig. 1 shows the absorption spectrum of MnIIIMb (2.5 μM, split Soret band at λmax 378 and 470 nm [28]) and that of MnIIMb (λmax 440 nm) formed by reduction with excess dithionite (0.1 mM) under anaerobic conditions. Although two distinct reduction potentials have been reported for MnIIIMb [32], the temporal spectral changes appear to be first order in protein with clean isosbestic points (400 and 459 nm) (Supporting Information Fig. S-1). The reduction of

2.2. Procedures for kinetics studies All UV/Vis spectral measurements were recorded using a Shimadzu dual beam UV-2401 PC spectrophotometer and 1.00 cm path length quartz cells. Nitric oxide was detected using a GE Seivers model 280i NOA nitric oxide analyzer. Known volumes of the gases from the solution headspace were injected into the NOA purge vessel and these gases were entrained to the detector using He(g). The NO present in the sample was detected and quantified using the NOA and a calibration curve generated from the reaction of NaNO2 with acidified KI. The nitrite reductase (NiR) reactions were all studied by following changes in the optical spectra. Unless otherwise noted, reaction

Fig. 1. Visible range spectra and extinction coefficients (mM− 1 cm− 1) for MnIIIMb, MnIIMb, and MnIIMb(NO) in pH 7.4 aqueous buffer. The peak at 378 nm for MnIIIMb is not shown.

J.L. Heinecke et al. / Journal of Inorganic Biochemistry 107 (2012) 47–53

49

Mn IIIMb by dithionite was previously attributed to the SO2• − anion radical formed by the reversible dissociation of S2O42 − (Eq. (2)) due to a square root dependence of the rate on dithionite concentration (kobs = kMn(III) [S2O42 −] 1/2) [34]. Under our conditions (2.5 μM MnIIIMb and 0.10 mM S2O42 −), the kobs for MnIIIMb reduction was found to be 7.6 × 10− 3 s− 1. Dividing by [S2O42 −] 1/2 gives a kMn(III) value of 0.76 M− 1/2 s− 1, consistent with the reported value of 0.54 M− 1/2 s− 1. 2−

−

S2 O4 ⇄2SO2 •

ð2Þ

Similar reduction of Co IIIMb (4 μM; E° = 100 mV vs. NHE [35]) with dithionite (0.3 mM) led to loss of the Soret band for Co IIIMb (λmax 424 nm) and appearance of a new band at 404 nm indicating formation of Co IIMb (Fig. 2). The temporal changes were biphasic (SI Fig. S-2), giving kobs values of 4.9 × 10 − 3 and 5.0 × 10 − 4 s − 1, consistent with a previous report [36]. Hambright et al. attributed this biphasic behavior to two forms of Co IIIMb in solution, which after reduction generates a single species [36]. The nature of the two forms of Co IIIMb has not been conclusively established, although contributions from a hemichrome-like species were suggested [37]. The two rate constants, kCo(III) and kCo(III)′ (kobs/[S2O42 −] 1/2), were calculated to be 0.28 and 0.029 M − 1/2 s − 1, roughly consistent with those reported (0.17 and 0.012 M − 1/2 s − 1). 3.2. Nitrite binding to Mn IIIMb and Co IIIMb Titration of nitrite (1 mM to 2.5 M) into a solution of MnIIIMb (3 μM) led to minimal changes in the electronic absorbance spectrum (SI Fig. S-3). Thus, we conclude that the equilibrium constant for formation of the nitrite complex MnIIIMb(ONO−) must be very small (Knitrite b ~0.4 M− 1) (Eq. (3), M= Mn). III

−

III

−

M Mb þ NO2 ⇄M MbðONO Þ

ð3Þ

A likely explanation is that the axial Jahn–Teller distortion for this high-spin d 4 system [38] would favor a five-coordinate structure. Consistent with this explanation is the observation that the crystallographically characterized Mn IIIMb(ONO −) displaying a long Mn\O (nitrite) bond (2.3 Å) was only obtained after excess nitrite was added to Mn IIMb, oxidizing the metal center to Mn III by the NiR activity described below [26]. Nitrite coordination to CoIIIMb proved to be slow, so the temporal spectroscopic changes were recorded to determine the kinetics of the NO2− substitution into the Co(III) coordination sphere. Addition of various NaNO2 concentrations to buffered CoIIIMb solutions (pH 7.4) resulted in a small, unexplained initial shift in the Soret band, followed by larger changes indicating a very slow binding event. The reason for

Fig. 2. Spectra and extinction coefficients (mM− 1 cm− 1) for CoIIIMb, CoIIMb, CoIIMb(NO), and CoIIIMb(ONO−).

Fig. 3. A plot of kobs vs. [NO2−] for the reaction of CoIIIMb with NO2− is linear giving a slope (kon) of 0.023 M− 1 s− 1 and y-intercept (koff) of 1.04 × 10− 5 s− 1. The equilibrium constant (Knitrite) for the formation of CoIIIMb(ONO−) was determined from kon/koff to be 2100 M− 1.

the initial shift is unclear, but loss of CoIIIMb (424 nm) coincided with formation of the new Soret absorption at 429 nm, presumed to be the result of CoIIIMb(ONO −) formation (SI Fig. S-4). This slow change was fit to a single exponential, indicating it was first order in CoIIIMb. The plot of kobs vs. [NO2−] (0.5–24 mM) was linear with a non-zero intercept; the slope (2.30± 0.03) × 10 − 2 M− 1 s− 1 corresponds to kon and the intercept (1.04± 0.30) × 10 − 5 s − 1 is attributed to koff (Fig. 3). The equilibrium constant, Knitrite, for nitrite binding determined from the ratio kon/koff was 2100 ± 600 M − 1, a value much higher than that reported previously for wt metMb (60 M − 1) [39] and that indicated above for the manganese analog MnIIIMb (b0.4 M− 1). The crystal structure of the nitrite complex CoIIIMb(ONO −) shows the nitrite ligand to be O-coordinated which is stabilized by H-bonding with the distal His64 residue [26]. In contrast, synthetic CoIII porphyrins not having such H-bonding interactions display N-nitro coordination [22–25,40]. 3.3. Nitrite binding to wt Mb III and the metMb mutants H64V Mb III and H64V/V67R Mb III To place the nitrite binding events in proper perspective, we also determined Knitrite for the metMb mutants H64V MbIII and H64V/ V67R Mb III and re-determined Knitrite for wt MbIII using a spectroscopic titration technique (Fig. 4, and SI Figs. S-5 and S-6). The respective Knitrite values determined thus are 3.7, 30 and 120 M− 1 for H64V, H64V/V67R and wt Mb III, respectively. These values reinforce the importance of a H-bonding residue such as His64 in the Mb III distal pocket (or the Arg67 in H64V/V67R Mb III) in stabilizing nitrite coordination

Fig. 4. Spectral changes for titration of nitrite (0.3 mM to 0.1 M) into a solution of wt MbIII (2.0 μM, pH 7.4) showing loss of MbIII (500 nm) and formation of MbIII(ONO−) (540 and 574 nm). Knitrite was calculated by plotting the log (Abs 574 nm) vs. log [NO2−] and fitting to a sigmoidal curve (inset).

50

J.L. Heinecke et al. / Journal of Inorganic Biochemistry 107 (2012) 47–53

[18]. The absence of such H-bonding may also explain the observation that the H64V MbIII coordinates NO2− as the N-nitro linkage isomer while H64V/V67R Mb III and wt Mb III, having H-bonding residues in the distal pocket, form O-nitrito complexes. Notably, the Knitrite determined here for wt MbIII using a direct spectrophotometric titration is about 2-fold larger than that obtained previously [39] via the inherently more uncertain kinetics method. 3.4. Reduction of nitrite by Mn IIMb and Co IIMb Solutions of MnIIMb (2 μM) were prepared by dithionite reduction of MnIIIMb followed by passage through a G-25 Sephadex column to remove excess dithionite and its oxidation byproducts. Addition of NaNO2 (1 mM) to the resulting solution led to little spectral change over a 10 min period, indicating that the reaction is much slower than that observed previously for the wt ferrous Mb (k2 = 6 M− 1 s− 1) [41,42]. Fig. 5 illustrates the spectral changes observed for a deaerated solution containing nitrite (3.9 mM) and MnIIMb (2.1 μM). The spectrum observed after completion of the reaction (SI Fig. S-7) analyzed as a 1:1 mixture of MnIIIMb and MnIIMb(NO) as expected if the NO formed in the NiR reaction (Eq. (4)) is trapped by the remaining MnIIMb (Eq. (5)) owing to the large equilibrium constant for MnIIMb(NO) formation [43]. The behavior of the MnIIMb/nitrite system in this regard is quite similar to that seen in solution studies with wt Mb, where the NO is trapped by the ferrous Mb, so the NiR reaction only proceeds to 50% completion. −

II

þ

III

Mn Mb þ NO2 þ 2H →Mn Mb þ NO þ H2 O II

ð4Þ

II

Mn Mb þ NO⇄Mn MbðNOÞ

ð5Þ

The temporal absorbance changes at 440 nm (λmax for Mn IIMb) deviated from exponential behavior (Fig. 5, inset), and for this reason the initial rates method was used to analyze the kinetics of the spectral changes. Plots of the initial rates (M/s) vs. [Mn IIMb] (1.5–3.7 μM), [NO2−] (4.3–8.1 mM), or [H +] (pH range 6.8–7.4) were each linear (Fig. 6) consistent with the rate law presented in Eq. (6). From these data the second order rate constant for nitrite reduction by Mn IIMb at pH 7.4 (kMn = k3Mn [H +]) was calculated to be 0.015 ± 0.002 M − 1 s − 1, which is a factor of 400 smaller than the analogous k2 for wt Mb (6 M − 1 s − 1). Since Mn IIMb II

Mn

−

þ

II

Mn

5

−2 −1

dðMn MbÞ=dtÞ ¼ −k3 ½NO2 ½H ½Mn Mbk3 ¼ 3:8  10 M

s

ð6Þ should be a stronger reductant than wt Mb II (see above), the slower rate for the former clearly indicates that other mechanistic factors

Fig. 5. Spectral changes and kinetic trace (inset) for reduction of nitrite (3.9 mM) by MnIIMb (2.1 μM) showing loss of MnIIMb (440 nm) and formation of MnIIIMb (471 nm) and MnIIMb(NO) (427 nm).

Fig. 6. The linear dependences of the initial rates of NiR reaction by MnIIMb on [MnIIMb] (1.5–3.7 μM MnIIMb, 8.1 mM NO2−, pH 7.4) (A), [nitrite] (4.3–8.1 mM NO2−, 2.7 μM MnIIMb, pH 7.4) (B), and [H+] (pH range 6.8–7.4, 1.5 μM MnIIMb, 4.3 mM NO2−) (C).

must be playing an important role in defining the dynamics of these systems. The NO2−/Mn IIMb kinetics were also studied without removing the excess dithionite from the solutions, a procedure that eliminates concern about spurious contamination with trace oxygen, since dithionite reacts readily with O2 [44]. The concentration of excess dithionite (0.1 mM) was calculated from its absorption band at 314 nm (8000 M − 1 cm − 1). The second rate constant (k′Mn) determined by the initial rates method was 0.019 ± 0.003 M − 1 s − 1 at pH 7.4, close to the kMn value in the absence of dithionite. Thus we conclude that the excess reductant does not have a large impact on the reaction rate. Furthermore, the initial rates were found to be independent of [S2O42 −] below 0.1 mM. However, one key difference is that the final product is not a mixture, but only Mn IIMb(NO), which is formed with clean isosbestic points (SI Fig. S-8). This observation is readily explained by the dithionite reduction of the Mn IIIMb formed (Eq. (4)) back to Mn IIMb, which is trapped by NO. The net reaction

J.L. Heinecke et al. / Journal of Inorganic Biochemistry 107 (2012) 47–53

51

under these conditions is represented by Eq. (7), where the extra electron is provided by the dithionite. II

−

þ

−

II

Mn Mb þ NO2 þ 2H þ e →Mn MbðNOÞ þ H2 O

ð7Þ

Fig. 7 illustrates the temporal spectral changes when nitrite (18 mM) was added to a solution of Co IIMb (2.5 μM). Notably, the final spectrum was not consistent with formation of either Co IIIMb (λmax 424 nm) or Co IIMb(NO) (420 nm) or of a mixture of these two species as the final products. Instead the spectrum indicated formation of the O-nitrito complex Co IIIMb(ONO−) (429 nm) (see also Fig. 2) as the end product. Clean isosbestic points were not the case, with the spectral intersection trailing from 412 nm to 418 nm as the reaction proceeded, indicating an A ➔ B ➔ C type process. There was no clear spectral demonstration that Co IIMb(NO) is formed, although the formation constant from NO and Co IIMb should be large [45]. One complication is that the Soret λmax for Co IIMb(NO) (420 nm) is in the same region as the trailing isosbestic point and the other expected product Co IIIMb has a similar λmax (424 nm). Thus, while these species may be formed transiently, it appears that the system evolves during the course of the experiment to give Co IIIMb(ONO−). If so, then NO must be liberated as the other product as indicated by Eq. (8). II

−

þ

III

−

Co Mb þ 2NO2 þ 2H →Co MbðONO Þ þ NO þ H2 O

ð8Þ

In order to test this premise, a buffered solution of CoIIMb (1.2 μM, 3.2 nmol) was generated in the presence of protocatechuate 3,4-dioxygenase and 3,4 dihydroxybenzoic acid (to maintain an anaerobic environment). Nitrite (18 mM) was then added, and the headspace above the solution was sampled using a gas tight syringe and analyzed using the Seivers NOA. A sample taken shortly after the reaction was initiated (~2 min) showed very little NO in the headspace, as might be expected since excess CoIIMb is present at the beginning of the reaction. However, the gas sample taken at the conclusion of the reaction (90 min) showed that, accounting for the partitioning between the gas and liquid phase, 2.8 ±1.1 nmol (0.9±0.3 equivalents) of free NO were formed (SI Fig. S-9), thus confirming the stoichiometry of Eq. (8). When the analogous experiment was carried out with wt Mb, the amount of free NO present in the gas and solution phases was dramatically smaller (~0.01 equiv.) confirming NO capture by the ferrous protein present. The initial rates method showed that the kinetics for Co IIMb disappearance were linearly dependent on [Co IIMb] (1.5–6.5 μM), [NO2−] (3–20 mM), and [H +] (pH 6.4–7.4) (Fig. 8, Eq. (9)), consistent with the NiR kinetics seen for Mn IIMb and for wt Mb II. At pH 7.4, the second order rate constant for Co IIMb (kCo) was determined to be

Fig. 8. The dependence of the initial rate of the CoIIMb NiR reaction on [CoIIMb] (1.5–6.5 μM CoIIMb, 6 mM NO2−, pH 7.4) (A), [nitrite] (3–20 mM NO2−, 2.5 μM CoIIMb, pH 7.4) (B), and [H+] (pH 6.4–7.4, 2.5 μM CoIIMb, 6 mM NO2−) (C).

0.0066 ± 0.0009 M − 1 s − 1 three orders of magnitude slower than wt Mb II and ~ 2-fold slower than Mn IIMb. II

Co

−

þ

II

Co

5

−2 −1

dðCo MbÞ=dtÞ ¼ −k3 ½NO2 ½H ½Co Mb k3 ¼ 1:7  10 M

Fig. 7. The reaction of CoIIMb (2.5 μM) and nitrite (18 mM) in the absence of dithionite was monitored by the absorbance changes. The reaction proceeds from the CoIIMb (404 nm) to CoIIIMb(ONO−) (429 nm) without defined isosbestic points. Inset: temporal absorption changes at 404 nm (CoIIMb) and at 429 nm (CoIIIMb(ONO−)).

s

ð9Þ

In the presence of excess dithionite (20 μM), addition of nitrite (12 mM) to a solution of Co IIMb (2.5 μM) gave only Co IIMb(NO) [λ in nm (ε M − 1 cm − 1), 420 (1.4 × 10 5), 540 (1.8 × 10 4), 576 (1.8 × 10 4), Fig. 2] suggesting that any Co IIIMb formed is rapidly intercepted by this reductant, and that the resulting Co IIMb is trapped by the NO generated. However, the reaction rate is much faster than in the absence of dithionite, and temporal absorbance changes were quite complex. Further details are presented in the Supporting Information (Figs. S-10 to S-12). Catalyzed dithionite reduction of nitrite has been observed with the iron and cobalt phthalocyanine complexes in alkaline solutions [46,47], so it is likely that a similar process is occurring in the present case as well.

52

J.L. Heinecke et al. / Journal of Inorganic Biochemistry 107 (2012) 47–53

3.5. Comparing the relative rates of nitrite reduction The first order dependence of the initial rates on the three concentration parameters [M IIMb], [NO2−] and [H +], shown here for Mn IIMb and Co IIMb and previously for wt Mb [13], suggests a common sequence of steps such as illustrated by Eqs. (10–12). The first would be the reversible formation of the M II O-nitrito complex (Eq. (10)) followed by irreversible reaction in an H + dependent rate limiting step (Eq. (11)) to give the nitrosyl complex of M IIIMb, which readily dissociates NO (Eq. (12)). A second H + is necessary in the second step(s) in order to balance the reaction, although the actual transformation of coordinated nitrite to NO may be assisted by a proton source within the distal cavity of the protein. This sequence of reactions would be consistent with the rate laws indicated by Eqs. (6) and (9) with k3M = K M k M, where K M is the metal-center dependent equilibrium constant for forming the M II nitrite complex (Eq. (10)) and k M the rate constant for the rate limiting step (Eq. (11)). II

−

II

−

M Mb þ NO2 ⇄M MbðONO Þ

ð10Þ

Acknowledgments

þ

þH II – þ III M MbðONO Þ þ H →?→ M MbðNOÞ þ H2 O III

III

M MbðNOÞ⇄M Mb þ NO

proved to be much less reactive than wt Mb in this reaction. By using the initial rates method, it was shown that the kinetics are first order in [protein], [nitrite], and [H +], the pattern seen for the Fe analog [13]. The k2 values were 0.015, 0.0066, and 6 M − 1 s − 1, respectively, at pH 7.4 for Mn IIMb, Co IIMb, and wt Mb. While the slow rate of nitrite reduction by CoIIMb can be rationalized in part by a less favorable oxidation potential, the same argument cannot be applied to MnIIMb, which is a stronger reductant than wt Mb. It is obvious that other factors are important. For example the reaction may be limited by the stability of the putative MIII nitrosyls, which, unlike Fe IIIMb [51], were not observed for MnIIIMb or CoIIIMb upon addition of NO. Nitrite reduction by h.s. MnIIMb [49] could also be limited by the large reorganizational energy associated with oxidation of MnIIMb to MnIIIMb. We note that the crystal structures of MnIIMb and MnIIIMb show a change in metal displacement out of the porphyrin plane from 0.34 to 0.18 Å (Δ = 0.16 Å) [26]. With regard to CoMb species, it is notable that the final species detected in solutions containing CoMb and NO or nitrite was always CoIIIMb(ONO−), the thermodynamic sink for this system.

ð11Þ ð12Þ

We were not able to observe spectrally the formation of nitrite complexes of MIIMb and therefore can only speculate regarding what are the contributions of these two terms may be to the overall rate. For wt MbII and the mutant myoglobins H64V MbII and H64V/V67R MbII, the metal center is invariant while the differences lie in the distal pocket residues, principally in the ability of these residues to hydrogen-bond to a coordinated ligand. Thus, the Knitrite values measured here for the ferric Mb's (120, 3.7, and 30 M− 1, respectively), should follow a similar pattern as that for nitrite coordination to the ferrous analogs. Consistent with this argument are the previously reported second order rate constants at pH 7.4 (k2) for nitrite reduction by the ferrous proteins, 6, 0.35, and 1.8 M− 1 s− 1, respectively, paralleling the Knitite values for Eq. (3). A somewhat different result was obtained in a recent study of nitrite reduction by the ferrous form of human neuroglobin, where replacement of the Ngb distal histidine by leucine or glutamine led to marked increases in NiR activity [50]. However, in wt Ngb, the distal histidine competes with the nitrite for the 6th coordination site on the Fe (II), and this effect on access to the redox active Fe(II) dominates the NiR activity. A similar correlation between the various M IIIMb Knitite values and the NiR rates would be unlikely, given the nature of the electronic configurations changes between the M III and M II oxidation states. For example, the high value of Knitite for Co IIIMb (as well as the slow rate formation of the Co IIIMb(ONO −) complex) can be attributed to the low spin (l.s.) d 6 electronic configuration of this species [37]. In contrast, the l.s. d 7 electronic configuration of the corresponding Co IIMb [48] would be expected to be labile and to have a lower affinity for nitrite. Similarly, the very low affinity of MnIIIMb for nitrite (Knitrite b 0.4 M − 1) can be attributed to the strong Jahn-Teller distortion of the h.s. d4 metal center [34–38], but it is not clear whether the h.s. d 5 MnIIMb [49] would have a higher or lower affinity for nitrite than the analogous h.s. d6 wt Mb, although, if this is accompanied by a h.s. to l.s. conversion, the higher ligand field stabilization of the d6 configuration would favor the Fe II center. At any rate, the observation that the latter has much greater NiR activity despite the fact that MnIIMb is a stronger reducing agent, points to the strength and mode of NO2− binding to the metal center as being responsible for the differences in activity. In summary, we have determined that the metal-substituted myoglobins Mn IIMb and Co IIMb both displayed NiR activity but

This research was supported by the Oklahoma Center for the Advancement of Science and Technology (HR09-081 to GBR-A), by grants from the Chemistry Division of the National Science Foundation (NSF-CHE-0749524 and NSF-CHE-1058794 to PCF) and by a fellowship from the ConvEne IGERT Program (NSF-DGE 0801627 to JH). Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.jinorgbio.2011.10.006. References [1] [2] [3] [4] [5] [6] [7]

[8] [9] [10]

[11] [12] [13] [14]

[15]

[16] [17] [18] [19] [20] [21]

L.J. Ignarro, Angew. Chem. Int. Ed Engl. 38 (1999) 1882–1892. R.F. Furchgott, Angew. Chem. Int. Ed Engl. 38 (1999) 1870–1880. F. Murad, Angew. Chem. Int. Ed Engl. 38 (1999) 1857–1868. A.J. Gow, B.P. Luchsinger, J.R. Pawloski, D.J. Singel, J.S. Stamler, Proc. Natl. Acad. Sci. USA 96 (1999) 9027–9032. J.L. Zweier, P. Wang, A. Samouilov, P. Kuppusamy, Nat. Med. 1 (1995) 804–809. J. Heinecke, P.C. Ford, Coord. Chem. Rev. 254 (2010) 235–247. M. Feelisch, B.O. Fernandez, N.S. Bryan, M.F. Garcia-Saura, S. Bauer, D.R. Whitlock, P.C. Ford, D.R. Janero, J. Rodriguez, H. Ashrafian, J. Biol. Chem. 283 (2008) 33927–33934. M.P. Doyle, R.A. Pickering, T.M. DeWeert, J.W. Hoekstra, D. Pater, J. Biol. Chem. 256 (1981) 12393–12398. M.T. Gladwin, R. Grubina, M.P. Doyle, Acc. Chem. Res. 42 (2008) 157–167. A.J. Webb, A.B. Milsom, K.S. Rathod, W.L. Chu, S. Qureshi, M.J. Lovell, F.M.J. Lecomte, D. Perrett, C. Raimondo, E. Khoshbin, Z. Ahmed, R. Uppal, N. Benjamin, A.J. Hobbs, A. Ahluwalia, Circ. Res. 103 (2008) 957–964. S. Basu, N.A. Azarova, M.D. Font, S.B. King, N. Hogg, M.T. Gladwin, S. Shiva, D.B. Kim-Shapiro, J. Biol. Chem. 283 (2008) 32590–32597. H.T. Li, H.M. Cui, T.K. Kundu, W. Alzawahra, J.L. Zweier, J. Biol. Chem. 283 (2008) 17855–17863. S. Shiva, Z. Huang, R. Grubina, J. Sun, L.A. Ringwood, P.H. MacArthur, X. Xu, E. Murphy, V.M. Darley-Usmar, M.T. Gladwin, Circ. Res. 100 (2007) 654–661. N2O3 may also form from nitrite in the hydrophobic distal pocket of Hb. (a) B.O. Fernandez, P.C. Ford, J. Am. Chem. Soc. 125 (2003) 10510–10511; (b) P.C. Ford, B.O. Fernandez, M.D. Lim, Chem. Rev. 105 (2005) 2439–2455. A related nitrite dehydratase reaction has been proposed for Hb. (a) S. Basu, R. Grubina, J. Huang, J. Conradie, Z. Huang, A. Jeffers, A. Jiang, X. He, I. Azarov, R. Seibert, A. Mehta, R. Patel, S.B. King, N. Hogg, A. Ghosh, M.T. Gladwin, D.B. Kim-Shapiro, Nat. Chem. Biol. 3 (2007) 785–794; (b) K.H. Hopmann, B. Cardey, M.T. Gladwin, D.B. Kim-Shapiro, A. Ghosh, Chem. Eur. J. 17 (2011) 6348–6358. C. Bonaventura, R. Henkens, A.I. Alayash, A.L. Crumbliss, IUBMB Life 59 (2007) 498–505. R. Grubina, S. Basu, M. Tiso, D.B. Kim-Shapiro, M.T. Gladwin, J. Biol. Chem. 283 (2008) 3628–3638. J. Yi, J. Heinecke, H. Tan, P.C. Ford, G.B. Richter-Addo, J. Am. Chem. Soc. 131 (2009) 18119–18128. N. Xu, J. Yi, G.B. Richter-Addo, Inorg. Chem. 49 (2010) 6253–6266. K.S. Suslick, R.A. Watson, Inorg. Chem. 30 (1991) 912–919. G.G. Martirosyan, A.S. Azizyan, T.S. Kurtikyan, P.C. Ford, Chem. Commun. (2004) 1488–1489.

J.L. Heinecke et al. / Journal of Inorganic Biochemistry 107 (2012) 47–53 [22] H. Adachi, H. Suzuki, Y. Miyazaki, Y. Iimura, M. Hoshino, Inorg. Chem. 41 (2002) 2518–2524. [23] P.G. Jene, J.A. Ibers, Inorg. Chem. 39 (2000) 3823–3827. [24] J. Goodwin, T. Kurtikyan, J. Standard, R. Walsh, B. Zheng, D. Parmley, J. Howard, S. Green, A. Mardyukov, D.E. Przybla, Inorg. Chem. 44 (2005) 2215–2223. [25] G.R.A. Wyllie, W.R. Scheidt, Chem. Rev. 102 (2002) 1067–1089. [26] Z.N. Zahran, L. Chooback, D.M. Copeland, A.H. West, G.B. Richter-Addo, J. Inorg. Biochem. 102 (2008) 216–233. [27] D.M. Copeland, A. Soares, A.H. West, G.B. Richter-Addo, J. Inorg. Biochem. 100 (2006) 1413–1425. [28] T. Yonetani, J. Biol. Chem. 242 (1967) 5008–5013. [29] F.W.J. Teale, Biochim. Biophys. Acta 35 (1959) 543. [30] T. Yonetani, T. Asakura, J. Biol. Chem. 244 (1969) 4580–4588. [31] M.D. Lim, I.M. Lorkovic, P.C. Ford, Methods Enzymol. 396 (2005) 3–17. [32] R. Lin, C.E. Immoos, P.J. Farmer, J. Biol. Inorg. Chem. 5 (2000) 738–747. [33] E.L. Raven, A.G. Mauk, Adv. Inorg. Chem. 51 (2001) 1–49. [34] R. Langley, P. Hambright, K. Alston, Inorg. Chem. 25 (1986) 114–117. [35] C.Z. Li, K. Nishiyama, I. Taniguchi, Electrochim. Acta 45 (2000) 2883–2888. [36] P. Hambright, S. Lemelle, K. Alston, P. Neta, H.H. Newball, S. Di Stefano, Inorg. Chim. Acta 92 (1984) 167–172. [37] L.C. Dickinson, M.C.R. Symons, J. Phys. Chem. 86 (1982) 917–921. [38] L.J. Boucher, Coord. Chem. Rev. 7 (1972) 289–329.

53

[39] A. Wanat, J. Gdula-Argasinska, D. Rutkowska-Zbik, M. Witko, G. Stochel, R. van Eldik, J. Biol. Inorg. Chem. 7 (2002) 165–176. [40] K. Yamamoto, Y. Iitaka, Chem. Lett. (1989) 697–698. [41] K.T. Huang, A. Keszler, N. Patel, R.P. Patel, M.T. Gladwin, D.B. Kim-Shapiro, N. Hogg, J. Biol. Chem. 280 (2005) 31126–31131. [42] Z. Huang, S. Shiva, D.B. Kim-Shapiro, R.P. Patel, L.A. Ringwood, C.E. Irby, K.T. Huang, C. Ho, N. Hogg, A.N. Schechter, M.T. Gladwin, J. Clin. Invest. 115 (2005) 2099–2107. [43] M. Hoshino, Y. Nagashima, H. Seki, M.D. Leo, P.C. Ford, Inorg. Chem. 37 (1998) 2464–2469. [44] J.A. Morello, R.E. Forster, M.R. Craw, H.P. Constantine, J. Appl. Physiol. 19 (1964) 522. [45] L.E. Laverman, P.C. Ford, J. Am. Chem. Soc. 123 (2001) 11614–11622. [46] E.V. Kudrik, S.V. Makarov, A. Zahl, R. van Eldik, Inorg. Chem. 42 (2003) 618–624. [47] E.V. Kudrik, S.V. Makarov, A. Zahl, R. van Eldik, Inorg. Chem. 44 (2005) 6470–6475. [48] J.C.W. Chien, L.C. Dickinson, Proc. Natl. Acad. Sci. USA 69 (1972) 2783–2787. [49] H. Hori, M. Ikeda-Saito, G.H. Reed, T. Yonetani, J. Magn. Reson. 58 (1984) 177–185. [50] M. Tiso, J. Tejero, S. Basu, I. Azarov, X. Wang, V. Simplaceanu, S. Frizzell, T. Jayaraman, L. Geary, C. Shapiro, C. Ho, S. Shiva, D.B. Kim-Shapiro, M.T. Gladwin, J. Biol. Chem. 286 (2011) 18277–18289. [51] M. Hoshino, K. Ozawa, H. Seki, P.C. Ford, J. Am. Chem. Soc. 115 (1993) 9568–9575.