Nitroxyl (HNO) Reacts with Molecular Oxygen and Forms Peroxynitrite at Physiological pH

Background: Nitroxyl (HNO) is a reactive nitrogen species implicated in cardioprotection. Results: Nitroxyl reacts with oxygen to form an oxidizing and nitrating species, peroxynitrite. Conclusion: In the presence of oxygen, HNO donors may be a source of peroxynitrite. Significance: Peroxynitrite formation should be taken into account in the extracellular milieu when exposing cells to HNO donor under aerobic conditions. Nitroxyl (HNO), the protonated one-electron reduction product of NO, remains an enigmatic reactive nitrogen species. Its chemical reactivity and biological activity are still not completely understood. HNO donors show biological effects different from NO donors. Although HNO reactivity with molecular oxygen is described in the literature, the product of this reaction has not yet been unambiguously identified. Here we report that the decomposition of HNO donors under aerobic conditions in aqueous solutions at physiological pH leads to the formation of peroxynitrite (ONOO−) as a major intermediate. We have specifically detected and quantified ONOO− with the aid of boronate probes, e.g. coumarin-7-boronic acid or 4-boronobenzyl derivative of fluorescein methyl ester. In addition to the major phenolic products, peroxynitrite-specific minor products of oxidation of boronate probes were detected under these conditions. Using the competition kinetics method and a set of HNO scavengers, the value of the second order rate constant of the HNO reaction with oxygen (k = 1.8 × 104 m−1 s−1) was determined. The rate constant (k = 2 × 104 m−1 s−1) was also determined using kinetic simulations. The kinetic parameters of the reactions of HNO with selected thiols, including cysteine, dithiothreitol, N-acetylcysteine, captopril, bovine and human serum albumins, and hydrogen sulfide, are reported. Biological and cardiovascular implications of nitroxyl reactions are discussed.

arginine (17)(18)(19). Moreover, it has been shown in vitro that HNO can be formed via the reduction of ⅐ NO by mitochondrial cytochrome c, xanthine oxidase, or ubiquinol (2, 20 -22). Reports indicate that HNO is also presumably formed from nitrosothiol reaction with other thiols or ascorbate (23,24). Although based on kinetic considerations, their biological relevance remains doubtful (23,24).
More recently, new fluorescent probes for quantitative detection of HNO in aqueous solution were reported (25,26). The ground state of HNO is a singlet, but NO Ϫ anion, which is formed upon HNO deprotonation, similarly to the isoelectronic O 2 molecule, has the triplet ground state. HNO is a weak acid (pK a ϭ 11.4) (27), and its deprotonation is a spin-forbidden reaction and an exceptionally slow process. It has been reported that nitroxyl anion (NO Ϫ ) reacts very quickly with molecular oxygen forming peroxynitrite.
In the absence of scavengers, HNO spontaneously dimerizes in aqueous solutions with the second order rate constant of k ϭ 8 ϫ 10 6 M Ϫ1 s Ϫ1 to yield hyponitrous acid, which upon water elimination yields N 2 O (28).
One of the most intriguing aspects of nitroxyl (HNO/NO Ϫ ) chemistry is its reaction with molecular oxygen (42)(43)(44)(45)(46). It is commonly accepted that the reaction between nitroxyl anion (NO Ϫ ) and molecular oxygen leads to the formation of ONOO Ϫ (45,46). Despite numerous studies on the reactivity of HNO with molecular oxygen, the product of this reaction has not been unequivocally identified (42,44). The possibility of peroxynitrite formation from the reaction between HNO and O 2 remained skeptical and understudied (42,44). This is, in part, due to the lack of availability of specific probes that directly react with ONOO Ϫ to form specific diagnostic products. Here, we present unambiguous evidence for ONOO Ϫ formation during the reaction between HNO and molecular oxygen using the recently developed assays for specific detection and quantification of peroxynitrite (47,48).
We have reported previously that boronic acids and their esters react directly and rapidly with ONOO Ϫ (k ϭ 10 6 M Ϫ1 s Ϫ1 ), with the formation of corresponding phenols as major products (85-90%) (49). Although boronates also react with H 2 O 2 , the rate of reaction is rather sluggish (k ϭ ϳ1 M Ϫ1 s Ϫ1 ). An array of pro-fluorescent boronate-based probes has been reported, providing now the opportunity for noninvasive, real time monitoring and imaging of ONOO Ϫ in cell-free and cellular systems (50 -54). In this study two profluorescent probes (CBA and FBBE) and a nonfluorescent probe, phenylboronic acid with a bulky triphenylphosphonium group (p-MitoPhB (OH) 2 ) (47) were used (Fig. 1B). The reaction between ONOO Ϫ and boronate leads to the formation of ONOO Ϫ adduct, followed by the cleavage of the O-O bond, which may proceed via a nonradical, major pathway (heterolytic cleavage of O-O bond) or radical-mediated, minor pathway (homolytic cleavage of the O-O bond). The minor pathway yields unique nitrated products that are highly specific for ONOO Ϫ reaction with boronate. The detailed mechanism of the reaction of boronates with ONOO Ϫ has been recently reported (47,48). We demonstrated that the formation of the products of both major and minor pathways in a specific ratio can be used as a "peroxynitrite fingerprint" and applied this strategy to unambiguously determine the identity of the product of the reaction of HNO with molecular oxygen.

EXPERIMENTAL PROCEDURES
Chemicals-Boronate probes CBA and p-MitoPhB(OH) 2 were synthesized according to the published procedure (50,55). FBBE was synthesized by benzylation of fluorescein methyl ester (FME) with the use of 4-(iodomethyl)phenylboronic acid pinacol ester. 4 In all experiments Angeli's salt was used as a HNO donor, synthesized according to the published procedure (29). The concentration of Angeli's salt stock solutions was determined by measuring the absorbance at 248 nm (⑀ ϭ 8.3 ϫ 10 3 M Ϫ1 s Ϫ1 ) (29) in 1 mM aqueous NaOH solutions. Another HNO donor, 2-bromo-N-hydroxybenzenesulfonamide, was synthesized according to the published procedure (56). Peroxynitrite was prepared according to the published procedure (57), by reacting nitrite with H 2 O 2 . The concentration of ONOO Ϫ was determined by measuring the absorbance at 302 nm (⑀ ϭ 1.7 ϫ 10 3 M Ϫ1 s Ϫ1 ) (57) in 1 mM aqueous NaOH solutions. Stock solution of hydrogen sulfide was prepared by dissolving sodium sulfide in water. HNO scavengers, and all other reagents (of the highest purity available) were obtained from Sigma-Aldrich, whereas DPTA-NONOate was purchased from Cayman Chemicals, and catalase was from Boehringer. Each solution was prepared using deionized water (Millipore Milli-Q system).
UV-visible Absorption Measurements-The UV-visible absorption spectra were collected with Agilent 8453 spectrophotometer equipped with photodiode array detector and thermostated (25°C) cell holder.
Stopped Flow Measurements-Stopped flow kinetics experiments were performed using Applied Photophysics SX 20 stopped flow spectrophotometer equipped with a fluorescence detector. Angeli's salt (3 M in 1 mM NaOH) was rapidly mixed with the solution containing CBA (25 M), phosphate buffer (25 mM, pH 7.4), dtpa (50 M), 5% CH 3 CN, and HNO scavenger (at an appropriate concentration). The reaction mixtures were excited at 332 nm, and the emitted light intensity was measured at 470 nm (PMT voltage ϭ 850 V, emission/excitation slit ϭ 2.5 nm). The thermostated cell (25°C) with a 10-mm optical pathway was used for kinetic measurements.
Fluorescence Measurements-Fluorescence measurements were performed using Varian Cary Eclipse spectrofluorometer equipped with a plate reader accessory. The solutions containing FBBE were excited at 494 nm, and the emitted light intensity was measured at 518 nm (PMT voltage ϭ 580 V, emission slit ϭ 2.5 nm, excitation slit ϭ 5 nm). Reactions were performed at room temperature. Typically, the boronate probe (25 M) was incubated in phosphate buffer (50 mM, pH 7.4), dtpa (100 M), 5% CH 3 CN, and HNO donor and scavenger (at the appropriate concentration), and the extent of FBBE oxidation to FME was monitored over time.
UPLC/MS Analyses-Boronate probes and products of their oxidative conversion were analyzed using Acquity UPLC (Waters) system coupled online with LCT Premier XE (Waters) mass spectrometer with a time-of-flight mass detector. One microliter of sample was injected into the UPLC system equipped with the C 18 column (Waters Aquity BEH C18, 100 ϫ 2.1 mm, 1.7 m) maintained at 40°C and equilibrated with 30% CH 3 CN (containing 0.1% (v/v) TFA) in 0.1% TFA solution. The compounds were separated by a linear increase of CH 3 CN concentration in mobile phase from 30 to 60% using a flow rate of 0.35 ml/min. Under these conditions, the following retention times were observed: CBA, 2.79 min; COH, 3.04 min; coumarin (CH), 3.66 min; 7-nitrocoumarin (CNO 2 ), 3.86 min; p-MitoPhB (OH) 2 Kinetic Simulations-The kinetic simulation was carried out using a freely available software, Kintecus 4.55 (61). The kinetic model used in this study is a modification of the published model (62). The list of the chemical reactions, rate constants, and major modifications used in the simulation is shown in supplemental Table S1.

The Oxidation of CBA Probe in O 2 -saturated Angeli's Salt
Solutions-Angeli's salt slowly decomposed at pH 7.4, releasing HNO. As shown in Fig. 2A, its decay resulted in the disappearance of the absorption band maximum at 235 nm. In the presence of CBA the decomposition of Angeli's salt was accompanied by 7-hydroxycoumarin (COH) formation as reflected by the disappearance of the absorption band of boronate probe at 287 nm and the build-up of absorption of COH at 370 nm (Fig.  2B). The kinetics of CBA decay and COH appearance mirror the dynamics of Angeli's salt decomposition observed in the absence of CBA (Fig. 2C). These data indicate that decomposition of Angeli's salt resulted in the formation of an oxidant capable of rapid conversion of CBA probe into COH. The other HNO donor, 2-bromo-N-hydroxybenzenesulfonamide, also oxidized CBA in the same manner as Angeli's salt (data not shown).
To test the possibility of the formation of ONOO Ϫ via the reaction of HNO with O 2 , it was essential to confirm that Angeli's salt itself does not trigger the oxidation of CBA. As expected, neither COH formation was observed in the absence of Angeli's salt (Fig. 3, line d) nor in the absence of oxygen (Fig.  3, line c). Similar yields of COH were observed when CBA was incubated with Angeli's salt in aerated (line b) or oxygenated (line a) conditions. These data indicate that both HNO and molecular oxygen are essential for oxidative conversion of boronate probe. Only a minimal increase was observed after the Profiling of CBA Products Formed in Aqueous Solutions of Angeli's Salt: UPLC Study-For detailed characterization of oxidant formed in the reaction between HNO and molecular oxygen, the profiles of major and minor oxidation products of CBA were examined. For qualitative and quantitative determination of these products, we used the UPLC analysis. As shown in Fig. 4, in the presence of authentic ONOO Ϫ , CBA was converted to COH as the major product, and coumarin and 7-nitrocoumarin (CNO 2 ) as minor products formed in the radical pathway. This result is consistent with previous reports for simple arylboronates (49) and mitochondria-targeted boronate probes (47). Replacement of the boronate moiety in boronic compounds by the nitro group strongly supports an intermediary role of peroxynitrite (47,48). Mechanism of the reaction of CBA with ONOO Ϫ is shown in Fig. 5. The profile of products formed during Angeli's salt-induced oxidation closely resembles the profile observed for ONOO Ϫ -induced oxidation of CBA (Fig. 4). The addition of glutathione, a known scavenger of HNO, almost completely inhibited oxidation of boronate probe in the solutions containing Angeli's salt, but not when a bolus ONOO Ϫ was used. Although glutathione is known to react with peroxynitrite (k ϭ 1.36 ϫ 10 3 M Ϫ1 s Ϫ1 ) (63), 100 M GSH cannot compete with 100 M CBA probe for ONOO Ϫ (the rate constant of the CBA reaction with peroxynitrite is equal to k ϭ 1.1 ϫ 10 6 M Ϫ1 s Ϫ1 ) (50).
The quantitative analysis of substrate depletion and product formation during the incubation of CBA probe in aerated Angeli's salt solution and in the reaction between CBA and authentic ONOO Ϫ , added as bolus or generated from superoxide and nitric oxide fluxes, is shown in Fig. 6.
Previous studies indicated that neither O 2 . produced by xanthine/xanthine oxidase, nor ⅐ NO (DPTA-NONOate), caused the oxidation of boronate. Only ONOO Ϫ generated in situ from these systems was able to cause an oxidative conversion of the probe (47,49,50,52). The titration profiles are very similar for ONOO Ϫ and for Angeli's salt, indicating that the reaction between HNO and molecular oxygen results in the formation of ONOO Ϫ , which reacts rapidly with CBA (k ϭ 1.1 ϫ 10 6 M Ϫ1 s Ϫ1 ) yielding COH as a major product (50) and the minor products formed via the radical pathway (Fig. 5).
As shown previously (47), the ratios of rate constants for the radical (k r ) and nonradical (k nr ) pathways can be determined from the plot of the sum of the concentrations of minor products versus the concentration of the major, phenolic product. Fig. 6 shows the k r /k nr ratio determined for CBA oxidation.  Clearly, the oxidation of CBA probe in the aerated solution of Angeli's salt and in xanthine/xanthine oxidase/DPTA NONOate system results in the same product formation profiles and equal k r /k nr ratio. The yields of both major (COH) and minor (coumarin and CNO 2 ) products were the same (within the experimental error) for Angeli's salt-and O 2 . / ⅐ NO-induced oxidation of CBA (Table 1). The oxidative transformation of CBA by peroxynitrite added as bolus leads to the higher yield of COH and, as a consequence, a lower k r /k nr ratio. A possible explanation of that phenomenon is the formation of peroxynitrate (O 2 NOO Ϫ ) in the reaction between ONOO Ϫ and its protonated form (ONOOH) at the high local concentration of ONOO Ϫ , when added by bolus addition (64). Peroxynitrate anion is also able to oxidize CBA to COH (data not shown), but not through the radical pathway. As discussed previously, the profile of the products formed from the reaction between peroxynitrite and boronate probes is highly specific for this oxidant, providing a "peroxynitrite fingerprint." Therefore, the similar profiles of the products formed during the peroxynitrite (both bolus and generated in situ from co-generated O 2 . and ⅐ NO fluxes) and Angeli's salt-dependent oxidation of CBA, both qualitatively and quantitatively (Table 1) provide strong evidence that the oxidant formed during the reaction between HNO and molecular oxygen is ONOO Ϫ . Because the minor products of CBA oxidation by ONOO Ϫ have not been previously identified, we used another boronate probe, p-MitoPhB(OH) 2 (47), from which the minor products of reaction with ONOO Ϫ have been well characterized (47). The detailed mechanism of the reaction of this probe with ONOO Ϫ is similar to the previously mentioned CBA oxidation mechanism (Fig. 5) (49). The major pathway leads to the formation of corresponding phenol (p-MitoPhOH), whereas under the conditions used the minor pathway yields two products: MitoPh (dominant) and p-MitoPhNO 2 . Titration of p-MitoPhB(OH) 2 probe with Angeli's salt results in the product profiles shown in Fig. 7, indicating the formation of both phenolic (major) and radical-mediated products (minor). The profiles of the product formation resemble the profiles described previously for authentic peroxynitrite (47). Detection of ONOO Ϫ -specific products using two different boronate probes in aerobic reaction mixtures of Angeli's salt undoubtedly shows that the reaction of HNO with O 2 leads to formation of peroxynitrite. CBA and its corresponding oxidation products were detected using UV absorption detection at 300 Ϯ 5 nm.

Determination of the Rate Constant of the Reaction of HNO
with Oxygen-To determine the second order rate constant of the reaction between HNO and molecular oxygen, the competition kinetics approach was used. In this study, the HNO dimerization was not taken under consideration as in the aerated aqueous solutions, and at low Angeli's salt concentration (0 -10 M) this process is negligible. In fact, a 5-fold increase in oxygen concentration did not enhance the extent of CBA oxi-dation to COH (Fig. 3), indicating that reaction with oxygen outcompetes the self-decomposition of nitroxyl. HNO released from Angeli's salt reacts with molecular oxygen and HNO scavengers, when they are present (Fig. 8). Reaction of HNO with molecular oxygen leads to the formation of peroxynitrite, and there are two pathways of peroxynitrite-derived CBA oxidation (Fig. 5). The nonradical (nr) pathway of that reaction leads to the formation of COH, and the rate of COH accumulation over time is expressed by Equation where k CBA,nr is the rate constant for the nonradical pathway of CBA oxidation. Changes in the concentrations of HNO and ONOO Ϫ over time are expressed by Equations 2 and 3,  Finally, the solution of aforementioned equations leads to Equations 5 and 6.
The initial rate of the COH formation is expressed by Equation 7, whereas in the absence of HNO scavenger it is expressed by relationship Equation 8.
Comparison of the equations expressing initial rates results in Equation 9.
The ratios of the rate constants of the second order reaction of HNO with its scavenger (k S ) and molecular oxygen (k O2 ) were estimated using Equation 9. Two fluorogenic probes, CBA and FBBE, were used, which upon oxidation yield blue fluorescent (COH) and green fluorescent (FME) products, respectively (Fig. 8). In this survey, an excess of boronate probe (25 M CBA or FBBE) was used over Angeli's salt (3 or 10 M) and HNO scavenger (0 -10 M). The oxidation products of those two probes are fluorescent; therefore, changes in the initial rate of the formation of the corresponding phenols were monitored with the use of spectrofluorometer or the stopped flow spectrophotometer. As shown in Fig. 9 (A and B), the addition of glutathione, an efficient HNO scavenger, inhibits the initial rates of oxidation of both probes. The kinetic traces were obtained from at least two independent measurements. A linear function was fitted to data obtained within first 3 min of steady state measurements (Fig. 9, B and E). In the stopped flow experiments, the initial period of the reaction was reduced to 60 s. The results were plotted versus the ratio of the HNO scavenger to oxygen concentration (225 M) and fitted linearly. The slope of the dashed lines in Fig. 9 (C and F) is equal to the value of the ratio of rate constants of the reactions of HNO with glutathione and with molecular oxygen. The values of the second order rate constant published in the literature for the reaction of HNO with molecular oxygen vary. Liochev and Fridovich (31) estimated that this rate constant equaled Ϸ1 ϫ 10 4 M Ϫ1 s Ϫ1 , whereas others reported the rate constant to be ϳ3 ϫ 10 3 M Ϫ1 s Ϫ1 (36). We calculated the values of the k O2 from the k S /k O2 ratios obtained experimen-  tally using the published rate constants for selected scavengers of HNO (CPTIO, PTIO, TEMPOL, superoxide dismutase, and HRP) (36,40,41). From obtaining the mean of values (Table 2), we determined this rate constant to be ϳ(1. Reactivity of HNO toward Thiols-Next, we determined the rate constants for the reactions between HNO and selected thiols. The calculated values of the rate constant of selected thiols are shown in Table 3. With small molecular weight thiols (except H 2 S), there is a correlation between the rate constants and pK a values of SH groups. The higher the value of pK a , the lower the reaction rate constant of HNO with thiol. The rate constants for the reactions of HNO with glutathione and with N-acetylcysteine have been previously reported (2 ϫ 10 6 and 5 ϫ 10 5 M Ϫ1 s Ϫ1 , respectively) (36). Our analysis leads to similar but slightly higher values of the rate constants of HNO scavenging, namely 3.1 Ϯ 0.6 ϫ 10 6 M Ϫ1 s Ϫ1 for glutathione and 1.4 Ϯ 0.3 ϫ 10 6 M Ϫ1 s Ϫ1 for N-acetylcysteine (Table 3).
Kinetic Simulations-The value of the rate constant of HNO reaction with O 2 was also determined by kinetic simulations, which mirrored the kinetic experiments. In computational calculations, we were changing the value of k O2 to find the best fit of the computed values of k S /k O2 ratios to the experimental values. The resulting values of second order rate constant for the reaction of HNO with molecular oxygen are presented in the Table 2, and the averaged value is equal to 2 ϫ 10 4 M Ϫ1 s Ϫ1 . In the reaction with HNO all selected scavengers: CPTIO, PTIO, TEMPOL, superoxide dismutase, and HRP yield nitric oxide, which subsequently reacts with nitroxyl (Reaction 3, k ϭ B and E, initial period of the reaction presented in A or D, respectively. C and F, the effect of glutathione on the initial rate of COH or fluorescein methyl ester formation, respectively. The solutions containing CBA were excited at 332 nm, and the emitted light intensity was measured at 470 nm (PMT voltage ϭ 585 V, emission/excitation slit ϭ 5 nm), whereas the solutions containing FBBE were excited at 494 nm, and the emitted light intensity was measured at 518 nm (PMT voltage ϭ 580 V, emission slit ϭ 2.5 nm, excitation slit ϭ 5 nm). Each point represents the average value of two samples. The error bars represent standard deviations. DECEMBER

REACTION 3
Because this reaction cannot be separated from experimentally observed effect of HNO scavengers, the obtained k O2 rate constant differs from the value obtained via classical kinetic analysis. Similar simulations were performed to determine the rate constants of HNO reaction with thiols (cysteine, glutathione, dithiothreitol, N-acetylcysteine, and captopril), bovine and human serum albumin, and HS Ϫ . In these simulations, the rate constant for the reaction of HNO with O 2 was set to be equal to 2 ϫ 10 4 M Ϫ1 s Ϫ1 . During kinetic simulation, we were changing the value of k S to get the best agreement with experimentally determined k S /k O2 ratios. The calculated rate constants are presented in Table 3. Fig. 10 (A and B) shows the comparison of the experimentally recorded increase of fluorescence corresponding to CBA oxidation to COH and the simulated kinetic traces of COH build-up for two HNO scavengers: TEMPOL and glutathione.

DISCUSSION
Chemical Considerations-Here we have shown that ONOO Ϫ is the major product of the reaction between HNO and molecular oxygen. Previously Miranda (42) has shown that Angeli's salt, like ONOO Ϫ , is able to oxidize dihydrorhodamine in the presence of oxygen. However, the structure of the reactive oxidant formed from the reaction between HNO and molecular oxygen was not characterized. It was also shown that Angeli's salt, unlike ONOO Ϫ , is unable to oxidize phenols to their dimeric and fluorescent products. This led to the conclusion that ONOO Ϫ is not formed in the reaction of HNO with molecular oxygen, and different products of the reaction of HNO with O 2 were proposed (42). However, the dimeric phenols are formed via recombination of phenoxyl radicals. Because phenoxyl radicals are moderate one-electron oxidants, they can be readily reduced by HNO formed in this system. Contrary to Miranda (42), Kirsch and de Groot (43) have shown that ONOO Ϫ and oxidant formed in aerated solutions of Angeli's salt exhibit a similar pattern of the reactivity, suggesting formation of peroxynitrite in the reaction of HNO with O 2 . Several studies examining HNO in vitro demonstrated that Angeli's salt (HNO donor) exerted marked and oxygen-dependent cytotoxicity (13)(14)(15)(16). The present results show that ONOO Ϫ is formed as a major product from HNO reaction with O 2 , and thus, the cytotoxicity of HNO donors can be at least partially explained through formation of peroxynitrite.
The rate constant for the reaction of HNO with molecular oxygen was determined to be ϳ(1.8 Ϯ 0.3) ϫ 10 4 M Ϫ1 s Ϫ1 at pH 7.4, and this value is two times higher than the reported value by Liochev and    determined for the reaction of HNO with O 2 and the kinetic model used, we evaluated the rate constants of the reactions of HNO with selected thiols of possible biological and/or pharmacological relevance. The proposed methodology is easy to conduct and could be used for further HNO reactivity assessment. As previously shown for glutathione, we observed that other thiols also react with HNO significantly faster than does molecular oxygen, with the rate constants 2 orders of magnitude higher. These results indicate that it is very unlikely that HNO would react with O 2 in the intracellular milieu, because of the high content of reduced thiols (both low molecular and proteins) and relatively low concentration of oxygen in most tissues. However, the reaction of HNO with O 2 may be important in well oxygenated biological systems under decreased content of reduced thiols (for example under oxidative stress conditions). In addition, this reaction should be taken into account during in vitro studies, when exposing cells to HNO donor under aerobic conditions.
Biological Considerations-Recently, the potential beneficial actions of Angeli's salt in failing rat hearts were attributed to concomitant vasodilation and inotropic mechanism via soluble guanylyl cyclase-dependent and -independent mechanisms (65). The cardioprotective effect (e.g. enhanced contractile function) of HNO donor (CXL-1020) in patients with heart failure was recently demonstrated (66). CXL-1020 is synthesized from Piloty's acid and decomposes to HNO more slowly than Angeli's salt (66). Published results show that CXL-1020 decomposes to form N 2 O through a mechanism similar to that of Angeli's salt (66). Although CXL-1020 was shown to be safe and well tolerated in patients with heart failure (66), prolonged administration of this drug at slightly higher doses caused inflammatory reactions in dogs (66). The authors indicated that this toxic side effect could derail its use as a human therapeutic (66). Based on the present results, it is abundantly clear that in the presence of oxygen, HNO donors will form ONOO Ϫ as a major intermediate in the extracellular milieu. This finding may be relevant in explaining their toxic side effects.