Kinetics of Azanone (HNO) Reactions with Thiols: Effect of pH

HNO (nitroxyl, IUPAC name azanone) is an electrophilic reactive nitrogen species of growing pharmacological and biological significance. Here, we present data on the pH-dependent kinetics of azanone reactions with the low molecular thiols glutathione and N-acetylcysteine, as well as with important serum proteins: bovine serum albumin and human serum albumin. The competition kinetics method used is based on two parallel HNO reactions: with RSH/RS− or with O2. The results provide evidence that the reaction of azanone with the anionic form of thiols (RS−) is favored over reactions with the protonated form (RSH). The data are supported with quantum mechanical calculations. A comprehensive discussion of the HNO reaction with thiolates is provided.

RSNHOH þ RSH ! RSSR þ NH 2 OH ð2Þ The high reactivity of HNO toward thiols and their abundance in biological systems are major factors determining the short lifetime of azanone in vivo [34]. On the other hand, it has been proposed that azanone can be generated in several thiol-related pathways [35,36]. The first is the reaction of thiols with S-nitrosothiols (Reaction 4) [37,38]. Similar routes of HNO generation include RSNO reactions with H 2 S (Reaction 5 and/or Reactions 6-7) [39] or ascorbate anion (Asc − ) (Reactions 8-9) [40].
Recently, it has been proposed that HNO is also produced in the reaction of thiols with nitric oxide [36]. • NO is known to react with thiols, with the formation of N 2 O and corresponding disulfides [43] and/or sulfenic acids [44]. However, both the mechanism [36,43,44] and the kinetics of these processes are elusive [36,[45][46][47]. The formation of HNO has also been linked to the mechanism describing the formation of dinitrosyl-iron complexes (DNIC) from • NO, RS − and Fe 2+ [48][49][50]. DNIC are biologically relevant bioinorganic complexes of • NO, and perhaps the most abundant nitric oxide-derived adducts present in cells producing • NO [51,52]. It has been suggested that they can act as RSNO precursors [49,50,53] and HNO/NO − donors [54,55]. Due to the rapid scavenging of HNO by thiols, the generation of azanone is not expected to affect the DNICdependent RSNO formation. The number of these processes makes it challenging to formulate a proper description of the thiols/ • NO/HNO interactome.
In the absence of scavengers, HNO is known to spontaneously dimerize with a second-order rate constant of 8 × 10 6 M −1 s −1 [10]. The intermediate product of this reaction, hyponitrous acid, dehydrates to final decomposition products, nitrous oxide and water (Reaction 11).
The propensity of HNO to undergo the above reaction requires the use of donor molecules, the decomposition product of which is the HNO molecule. The most often studied and commonly used HNO donor is Angeli's salt, which decomposes at 25˚C with a rate constant of 6.8 × 10 −4 s −1 (t 1/2~1 7 min) in a pH range from 4 to 8.6 [56,57]. The fact that its decomposition rate constant is independent of pH is a unique feature of Angeli's salt compared to other HNO donors. Other frequently used HNO donors are Piloty's acid and its derivatives [57][58][59][60][61]. Unsubstituted Piloty's acid (Nhydroxybenzenesulfonamide) releases azanone favorably under alkaline conditions only, whereas Piloty's acid derivatives, substituted at different positions of the aromatic ring, release azanone across a wide range of pH values [57,60,61]. The rate constant of HNO release at a given pH depends on the ring substituents in Piloty's acid derivatives [60,61].
Similarly to • NO, HNO exhibits unique pharmacological effects that have potential benefits for the treatment of a variety of diseases. Chronologically, the first described biological action of HNO was the inhibition of alcohol dehydrogenase by cyanamide (a pharmacological alcohol deterrent agent), via its catalase-dependent bioactivation into an HNO donor [62][63][64]. More recently, HNO donors have been proposed as agents for the treatment of heart failure [65][66][67][68]. Azanone donors have been shown to induce apoptosis, suppress tumor angiogenesis, and help to achieve analgesia [69][70][71][72][73]. Some of these effects could be connected to HNO reactions, mainly with cysteine residues of key enzymes responsible for the observed pharmacological effects. For instance, HNO generated from cyanamide modifies the cysteine-302 residue in aldehyde dehydrogenase, leading to irreversible inhibition of the enzyme [63]. The mechanism by which azanone affects the heart is a matter of intense research. It has been proposed that HNO donors enhance cardiac contractility, by targeting the regulatory protein phospholamban [74,75]. Keceli et al. found that HNO reacts with Cys-41 and Cys-46 via the formation of the intramolecular disulfide bond, which forces conformational changes in the protein and enhances cardiac function as a result [74].
In a previous study, we investigated the reactivity of HNO toward selected thiols: cysteine (k Cys = (4.5 ± 0.9) × 10 6 M −1 s −1 , pK a = 8.3), glutathione (k GSH = (3.1 ± 0.6) × 10 6 M −1 s −1 , pK a = 8.8), N-acetylcysteine (k NAC = (1.4 ± 0.3) × 10 6 M −1 s −1 , pK a = 9.5) and captopril (k Cap = (6 ± 1) × 10 5 M −1 s −1 , pK a = 9.8). We found that at pH 7.4 the rate constant of the HNO reaction with thiol depends on its -SH group pK a . [3] In the present study, we explored the dependence of the rate constants of the reactions of HNO with selected biologically important thiols on pH. The data show the effect of pH on HNO reactivity toward the low molecular thiols N-acetylcysteine and glutathione and the thiol proteins bovine and human serum albumins.

Materials
Angeli's salt (AS, Sodium Trioxodinitrate, Na 2 N 2 O 3 ) was synthesized according to a published procedure [57]. Stable solutions of Angeli's salt prepared in 1 mM NaOH (pH ∼11) were stored on ice during the experiments and each day a fresh AS stock solution was prepared [56,57]. A boronate probe, coumarin boronic acid (CBA), which enables the detection of peroxynitrite (ONOO -) formed in the reaction of HNO with molecular oxygen, was synthesized according to a published procedure [76]. All thiols (glutathione (GSH), N-acetylcysteine (NAC), bovine serum albumin (BSA) and human serum albumin (HSA)), as well as all other chemicals (of the highest purity available) were purchased from Sigma-Aldrich Corp. By varying the amounts of salts (monobasic dihydrogen phosphate and dibasic monohydrogen phosphate) a range of buffers between pH 6.4 and 8.3 were prepared. All solutions were prepared using deionized water (Millipore Milli-Q system).

Competition Kinetic Method
The competition kinetic method used in this study followed the procedure described previously [3,12]. Angeli's salt, the most common HNO donor, was used. Azanone released from Angeli's salt reacts either with a corresponding thiol (RSH and RS − ) or with molecular oxygen (Scheme 1). The latter, relatively fast reaction (k = (1.8 ± 0.3) × 10 4 M −1 s −1 ) [3] results in the formation of peroxynitrite, which can be easily detected fluorometrically with the use of the fluorogenic probe, coumarin boronic acid (CBA). Across the whole studied pH range, CBA reacts rapidly and directly with ONOO -(k = 7.3 × 10 5 M −1 s −1 , pH 6.6; k = 1 × 10 6 M −1 s −1 , pH 7.4; k = 4.5 × 10 5 M −1 s −1 , pH 8.2), with the formation of blue fluorescent 7-hydroxycoumarin (COH) as the main product [3,76]. The ratio of initial rates of COH formation in the absence and presence of scavenger S can be expressed by the equation where k obs and k O 2 are the second order rate constants of HNO reactions with the scavenger (thiol/thiolate) and molecular oxygen, respectively, and (1), the k obs /k O 2 ratio was determined for each pH value. Figure 1 illustrates the used method to determine the k obs /k O 2 ratio at pH 6.5.

Stopped-flow Measurements
Angeli's salt (6 μM in 1 mM NaOH) was mixed rapidly with a solution containing the coumarin based monoborate probe CBA (50 μM), phosphate buffer (50 mM, pH range 6.4-8.3), metal chelator dtpa (100 μM), 10% CH 3 CN and the corresponding thiol compound at the appropriate concentration. Glutathione and N-acetylcysteine were used in a concentration range from 1 to 3 μM. Human or bovine serum albumin were used in the concentration range from 2 to 6 μM. Both reaction mixtures -the alkaline solution of the HNO donor and the solution of the corresponding thiol in the appropriate phosphate buffer -remained in equilibrium with air. The formation of fluorescent COH was monitored using an Applied Photophysics SX20 stopped-flow spectrophotometer equipped with a fluorescence detector and a thermostatically controlled cell (25°C) with a 10-mm optical pathway. The reaction mixtures were excited at 332 nm and the emitted light intensity was measured at 470 nm (PMT voltage = 850 V, emission/excitation slit = 2.5 nm). The initial rates of the increase in the fluorescence intensity were fitted with a linear function. The data were analyzed using the Origin Pro 2015 program (Ori-ginLab Corporation, Northampton, MA, USA).

pH Determination
The pH of the phosphate buffers and the exact pH of the solutions after mixing were measured using a Seven-Multi TM pH meter (Mettler Toledo GmbH, Schwerzenbach, Switzerland).

Computational Details
Quantum mechanical calculations were performed in the Gaussian G09 suite of programs, Revision E01 [78]. The geometries of the stationary points were fully optimized using the Hartree-Fock (HF) method as well as Density Functional Theory (DFT). The functionals M06-2X [79], B2PLYP and B3LYP were used with Grimme's D3 dispersion correction, B2PLYP-D3 [80][81][82] and B3LYP-D3 [83,84], respectively. A 6-311++G(2df,2p) basis set [85] was used, with the inclusion of a water solvent. Water was represented according to the IEFPCM [86,87] method by a default polarizable continuum solvent model in Gaussian. Structure optimizations were followed by frequency calculations, in order to verify the nature of the stationary points and to obtain corresponding free energy values. We were unable to locate any transition states for the reaction between CH 3 SH and CH 2 O.

Results
To examine the effect of pH on the reactivity of azanone toward thiols, we used the competition kinetics method previously described in the literature [3,12]. Due to the spontaneous dimerization of azanone, donor compounds are required that decompose with the release of the HNO molecule [10]. In our study, Angeli's salt was used as the HNO donor because it decomposes with a constant rate in the pH range from 4 to 8.6 [56,57]. Above pH 8, the rate of decomposition decreases [56,57]. The low concentration of the donor compound (3 μM) in the system resulted in an initial flux of HNO below 0.15 μM/min. Therefore, the steady-state concentration of HNO was very low. In the entire considered pH range, the system remained in equilibrium with air. Hence, the concentration of molecular oxygen was equal to 225 μM [77]. Given all the abovementioned factors, the HNO dimerization process was negligible and was not taken into account. Azanone is a weak acid, with a pK a value of 11.4 [1]. Therefore, released azanone exists in its protonated form in the studied pH range. The rate constant of the HNO reaction with molecular oxygen had been determined previously as equal to (1.8 ± 0.3) × 10 4 M −1 s −1 . We assume that this rate constant does not depend on the pH [3]. The reaction between HNO and O 2 results in the formation of peroxynitrite (ONOO − ) [3], which in aqueous solutions exists in an acid-base equilibrium with its protonated form peroxynitrous acid (ONOOH, pK a = 6.8) [88]. In the absence of scavengers, peroxynitrite undergoes isomerization to HNO 3 (~70%) (Reaction 12) and homolysis to • OH and • NO 2 radicals (~30%) (Reaction 13). All these radical species are highly oxidizing and nitrating agents. Its formation in the presence of the HNO donor could lead to one-electron oxidation of the donor compound, affecting the kinetics and mechanism of its decay [61]. The use of boronate probes in the system helps effectively scavenge peroxynitrite and prevent oxidation of Angeli's salt [61].
The oxidation of boronate compounds by ONOO − is a direct, stoichiometric and rapid reaction (k~10 5 -10 6 M −1 s −1 ), leading to the formation of the corresponding phenols as major products [76,[89][90][91][92][93]. However, at pH higher than 9 boronates undergo an addition reaction with hydroxyl ions (HO -), yielding a product unreactive toward ONOO - [91]. Given the pH-dependence of boronates reactivity toward ONOO − and the lower release of HNO from Angeli's salt in alkaline solutions (pH > 8.6), our studies were performed in a limited pH range (6.4-8.3) [56,57]. The probe used in our study, coumarin boronic acid (CBA), is converted by peroxynitrite to blue fluorescent 7-hydroxycoumarin (COH) as a major product [76]. The high reactivity of CBA toward peroxynitrite within the studied pH range ensures quantitative peroxynitrite scavenging in the presence of low micromolar concentrations of the studied thiols. The formation of COH formation in the presence of thiols was slower than in their absence. Based on Eq. (1), we determined the ratios of the second-order rate constants of the HNO reactions with thiol and molecular oxygen for each pH in the range from 6.4 to 8.3.
Assuming that HNO can react with the thiolate anion RS -(k RS -) as well as with its protonated form RSH (k RSH ), the observed rate constant (k obs ) can be expressed as a function of pH, which depends on thiol pK a , and the rate constants k RSand k RSH : The ratio k obs /k O 2 can be expressed in a similar way: The k obs /k O 2 ratios obtained for different pH were fitted to Eq. (3), which allowed us to estimate the rate constants of the HNO reaction with the corresponding thiol and thiolate, separately. Figure 2A shows the dependence of the k obs /k O 2 ratio on pH for the reaction between HNO and glutathione. It is noticeable that the reactivity of HNO toward thiols is pH-dependent. The pK a value for the dissociation of the -SH group in glutathione was taken from the literature as being equal to 8.8 [94]. The best fitting was obtained assuming k RS -/k O 2 equal to (2.1 ± 0.1) × 10 3 and k RSH /k O 2 = 100 ± 10. These results indicate that thiolate anions are much more reactive toward HNO than their protonated forms (k RS -» k RSH ). A similar observation was made for N-acetylcysteine (Fig. 2B). The pK a value of the -SH group in N-acetylcysteine is equal to 9.5 [94] and the corresponding ratios k RS -/k O 2 and k RSH / k O 2 are equal to (5.5 ± 0.4) × 10 3 and 5 ± 7, respectively. Again, the reaction of HNO with thiolate is faster, hence favored.
We also performed analogical experiments for the two most abundant thiol proteins, bovine (BSA) and human (HSA) serum albumins. The pK a values of these proteins are debatable. In the literature, the pK a value of BSA cysteine -SH group is estimated to be in the range from 7.86 to 8.00 [95], whereas the spectrum of pK a values for HSA is even broader (5.0-8.8) [96][97][98][99][100][101]. Figure 2C shows the dependence of the k obs /k O 2 ratio on pH for the reaction between HNO and BSA. The experimental data are best fitted with Eq. (3) and give a pK a value for the dissociation of the -SH group in BSA equal to 7.9 ± 0.1. The obtained pK a value fits well into the range of values described in the literature. The ratios k RS -/k O 2 and k RSH / k O 2 are equal to 180 ± 20 and 3 ± 4, respectively. Figure  2D shows the dependence of the k obs /k O 2 ratio on pH for the reaction between HNO and HSA. In calculations performed for HSA, we used a value for pK a of 8.1 [101], as has been recently established by three independent approaches. The experimental data, best fitted with Eq. (3) assuming the above-mentioned pK a value for the dissociation of the -SH group in HSA, give the ratios k RS -/k O 2 = 310 ± 20 and k RSH /k O 2 = 14 ± 6. The ratio values obtained for the thiol proteins confirm the observed relationship between HNO reactivity and protonation of the sulfhydryl group in the studied compounds.
To further examine the reaction of HNO with thiolates, we analyzed the correlation between k obs and the thiolate concentration. The values for k obs were determined based on the rate constant of the HNO reaction with molecular oxygen k O 2 = (1.8 ± 0.3) × 10 4 M −1 s −1 [3], whereas the concentration of thiolate was calculated based on the corresponding pK a value of the thiol. The linear relationship between k obs and the thiolate concentration can be expressed by Eq. 4. The correlation is illustrated in Fig. 3.
The variables [RS -] and [S] denote the concentration of thiolate and the total concentrations of the thiol ([S] = [RSH] + [RS − ]), respectively, while k RS− or k RSH are the rate constants of the HNO reaction with the thiolate (RS − ) and thiol (RSH), respectively. Therefore, our approach also allows us to estimate the rate constants of the HNO reaction with the corresponding thiol and thiolate, separately. The rate constants of the HNO reaction with the appropriate thiolates is high and varies in the range k~10 6 -10 7 M −1 s −1 , whereas the rate constants of the HNO reaction with the  (Table 1). Therefore, the same tendency can be observed: the reaction between azanone and thiolate is favored. There is a slight discrepancy between the k RSH values for thiol proteins computed with the aid of each approach. As can be seen in Fig. 3, the thiolate percentage is strongly dependent on the pK a value, which according to the literature varies in the case of BSA and HSA [95][96][97][98][99][100][101].

Discussion
The mechanism of the HNO reaction with thiols is currently understood to involve an initial nucleophilic attack of the thiol on the electrophilic nitrogen of azanone, forming N-hydroxysulfenamide (Reaction 1) [34,35]. Therefore, our finding that the rate constant of the reaction of HNO with thiols depends on its pK a value is not surprising. Using the presented approaches, we were able to estimate the rate constants of the HNO reaction with the corresponding thiol and thiolate separately. Supported by quantum mechanical calculations, the results indicate that azanone is much more reactive toward thiolates (RS − ) than toward protonated forms of thiols (RSH). This leads to the conclusion that it is the thiolate that nucleophilically attacks the HNO double bond. This mechanism is similar to the well-established reaction mechanism of thiols with carbonyl compounds, including CH 2 O, leading to the formation of hemithioacetals. It is commonly accepted that in these reactions the addition of thiols proceeds by the reaction of thiolate anion RS - [102,103]. The mechanism of the reaction between azanone and thiols may be comparable to the formaldehyde (CH 2 O) reaction with thiols, in which formaldehyde acts as an electrophile that reacts with biological nucleophiles, including thiols [104]. By analogy, the first step of the reaction may be a nucleophilic attack by the thiolate on the CH 2 O double bond, leading to the formation of the Shydroxymethyl adduct, an analog of N-hydroxysulfenamide [104].
Quantum mechanical results obtained for the detachment of HNO from MeSNHO -/ MeSNHOH can be compared with our recently published data on the decomposition of Piloty's acid (N-hydroxybenzenesulfonamide, C 6 H 5 SO 2 N-HOH) and its derivatives [61]. The mechanisms in the processes are quite similar. The decomposition mechanism of Piloty's acid and its derivatives include initial deprotonation of oxygen (C 6 H 5 SO 2 NHO − ) and subsequent S-N bond heterolysis, leading to slow release of the productsbenzenesulfinate and HNO [57,61]. We hypothesize that the pK a value of RSNHOH at physiological pH may be high, so the protonation reaction of RSNHOoccurs spontaneously. As a consequence, the stable RSNHOH is formed.

Conclusion
The reaction of azanone with thiol proteins is one of the major factors responsible for its unique pharmacological effects. In the present study, we have demonstrated both that this reaction depends strongly on pH and that HNO is highly reactive toward thiolates (RS − ). These results support the currently proposed reaction mechanism of HNO with thiols, involving an initial nucleophilic attack by the thiol on the electrophilic nitrogen of azanone.

Compliance with Ethical Standards
Conflict of Interest The authors declare no competing interests.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/.