Para-Substituted O-Benzyl Sulfohydroxamic Acid Derivatives as Redox-Triggered Nitroxyl (HNO) Sources

Nitroxyl shows a unique biological profile compared to the gasotransmitters nitric oxide and hydrogen sulfide. Nitroxyl reacts with thiols as an electrophile, and this redox chemistry mediates much of its biological chemistry. This reactivity necessitates the use of donors to study nitroxyl’s chemistry and biology. The preparation and evaluation of a small library of new redox-triggered nitroxyl sources is described. The condensation of sulfonyl chlorides and properly substituted O-benzyl hydroxylamines produced O-benzyl-substituted sulfohydroxamic acid derivatives with a 27–79% yield and with good purity. These compounds were designed to produce nitroxyl through a 1, 6 elimination upon oxidation or reduction via a Piloty’s acid derivative. Gas chromatographic headspace analysis of nitrous oxide, the dimerization and dehydration product of nitroxyl, provides evidence for nitroxyl formation. The reduction of derivatives containing nitro and azide groups generated nitrous oxide with a 25–92% yield, providing evidence of nitroxyl formation. The oxidation of a boronate-containing derivative produced nitrous oxide with a 23% yield. These results support the proposed mechanism of nitroxyl formation upon reduction/oxidation via a 1, 6 elimination and Piloty’s acid. These compounds hold promise as tools for understanding nitroxyl’s role in redox biology.


Introduction
The one-electron reduction and protonation of nitric oxide (NO), a well-known biological signaling agent characterized as a gasotransmitter, formally produces nitroxyl (HNO) [1]. These structural/electronic differences give HNO a distinct chemistry from NO, as HNO dimerizes to hyponitrous acid (H 2 N 2 O 2 ) that dehydrates to nitrous oxide (N 2 O) [2]. This reactivity necessitates the use of HNO donors and highlights the extreme electrophilic and oxidizing character of HNO [3]. HNO demonstrates different biological properties from NO [4,5], and at least three drugs that chemically release HNO have been used clinically or evaluated in trials for the treatment of cancer (hydroxyurea), alcoholism (cyanamide) and congestive heart failure (Cimlanod), showing the clinical potential of HNO donors [6][7][8]. Much of our understanding of the pharmacology and therapeutic potential of HNO comes from experiments utilizing HNO donors [3,9]. The most common HNO donors include Angeli's salt (AS, Na 2 N 2 O 3 ) and Piloty's acid (PA, PhSO 2 NHOH), which are commercially available solids that rapidly and cleanly release HNO under neutral conditions [3,9].
In addition to dimerization, HNO reacts with thiols, generating a N-hydroxysulfenamide intermediate that can rearrange to a sulfinamide or further react with more thiol to yield a disulfide and hydroxylamine [10]. HNO similarly reacts with hydrogen sulfide (H 2 S), a second recognized gasotransmitter [1], to give short-chain hydrogen polysulfides (H 2 S n ) or S 8 depending on the relative concentrations of HNO and H 2 S [11]. Based on this chemistry, HNO can influence thiol-mediated biochemistry and potentially H 2 S-controlled reactions. For example, HNO inhibits enzymes with active site cysteines, such as aldehyde dehydrogenase (AlDH) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) [12,13]. HNO modifies cardiac myofilament proteins affecting myocardial contractility by increasing calcium cycling and sensitizing myocardial tissue responsiveness to calcium [14,15]. HNO directly activates the chemosensory TRPA1 channel [16]. In bacteria, the addition of HNO to Staphylococcus aureus increases the formation of the persulfide (RSSH) of the predominant low-molecular-weight thiol, bacillithiol (BSH) [17]. This oxidized thiol derivative may control transcription factors responsible for directing the overall sulfur metabolism in these bacteria [18]. Under oxidative conditions, the addition of HNO to Bacillus subtilis synergistically enhances hydrogen peroxide (H 2 O 2 )-mediated cell killing. Angeli's salt (AS), which decomposes to HNO and nitrite (NO 2 − ) at neutral pH, was used as the HNO source in these redox investigations in B. subtilis [3,19].
Redox-triggered HNO donors may possess value in probing biological crosstalk between HNO and other small redox active signaling agents, such as H 2 S or H 2 O 2 . The use of AS as a HNO donor is limited by the co-production of NO 2 − , a relatively fast and pH-insensitive release rate (t 1/2~2 .8 min, pH 4-8) and limitations in the synthesis of new donors [3]. Piloty's acid (PA), another common HNO donor, decomposes to HNO and phenyl sulfinic acid (PhSO 2 H) at a neutral pH [3]. The aryl portion of PA tolerates structural modification while still supporting HNO release, which allows for the installation of designed redox-sensitive elements to initiate a 1, 6-elimination [20]. Scheme 1 shows Piloty's acid derivatives designed to release HNO via a 1, 6-elimination mechanism upon exposure to either reductants or oxidants. These processes convert the azide, nitro or boronate ester groups into either the aniline or phenol derivative that should decompose to Piloty's acid with the release of p-quinone methide or its imine (Scheme 1). hydrogen polysulfides (H2Sn) or S8 depending on the relative concentrations of HNO H2S [11]. Based on this chemistry, HNO can influence thiol-mediated biochemistry potentially H2S-controlled reactions. For example, HNO inhibits enzymes with active cysteines, such as aldehyde dehydrogenase (AlDH) and glyceraldehyde 3-phosph dehydrogenase (GAPDH) [12,13]. HNO modifies cardiac myofilament proteins affec myocardial contractility by increasing calcium cycling and sensitizing myocardial tis responsiveness to calcium [14,15]. HNO directly activates the chemosensory TRP channel [16]. In bacteria, the addition of HNO to Staphylococcus aureus increases formation of the persulfide (RSSH) of the predominant low-molecular-weight th bacillithiol (BSH) [17]. This oxidized thiol derivative may control transcription fac responsible for directing the overall sulfur metabolism in these bacteria [18]. Un oxidative conditions, the addition of HNO to Bacillus subtilis synergistically enhan hydrogen peroxide (H2O2)-mediated cell killing. Angeli's salt (AS), which decompose HNO and nitrite (NO2 − ) at neutral pH, was used as the HNO source in these re investigations in B. subtilis [3,19].
Redox-triggered HNO donors may possess value in probing biological cross between HNO and other small redox active signaling agents, such as H2S or H2O2. use of AS as a HNO donor is limited by the co-production of NO2 − , a relatively fast pH-insensitive release rate (t1/2~2.8 min, pH 4-8) and limitations in the synthesis of n donors [3]. Piloty's acid (PA), another common HNO donor, decomposes to HNO phenyl sulfinic acid (PhSO2H) at a neutral pH [3]. The aryl portion of PA toler structural modification while still supporting HNO release, which allows for installation of designed redox-sensitive elements to initiate a 1, 6-elimination [20]. Sch 1 shows Piloty's acid derivatives designed to release HNO via a 1, 6-elimina mechanism upon exposure to either reductants or oxidants. These processes convert azide, nitro or boronate ester groups into either the aniline or phenol derivative should decompose to Piloty's acid with the release of p-quinone methide or its im (Scheme 1).

Scheme 1. Designed redox-triggered HNO donors.
Similar constructs have found extensive use for the detection or generation of o signaling agents, such as H2S, but, to the best of our knowledge, have not been applie HNO release under redox conditions [21]. A hydrogen-peroxide-based prodrug sys that releases a structurally similar hydroxamic acid as a histone deacetylase inhibitor been described [22]. Such compounds permit HNO formation under specific re conditions and would form the basis of an improved understanding of the role that H plays in redox biochemistry. We describe the preparation and characterization of a sm library of redox-triggered HNO donors and an evaluation of their ability to produce H under specific redox conditions.

Synthesis
Possible redox-triggered HNO donors (1a-d and 2a-c) were prepared with a 27yield by the condensation of the properly substituted O-benzyl hydroxylamine deriva with either p-toluene or methyl sulfonyl chloride under basic conditions (Schem Scheme 1. Designed redox-triggered HNO donors. Similar constructs have found extensive use for the detection or generation of other signaling agents, such as H 2 S, but, to the best of our knowledge, have not been applied to HNO release under redox conditions [21]. A hydrogen-peroxide-based prodrug system that releases a structurally similar hydroxamic acid as a histone deacetylase inhibitor has been described [22]. Such compounds permit HNO formation under specific redox conditions and would form the basis of an improved understanding of the role that HNO plays in redox biochemistry. We describe the preparation and characterization of a small library of redox-triggered HNO donors and an evaluation of their ability to produce HNO under specific redox conditions.

Synthesis
Possible redox-triggered HNO donors (1a-d and 2a-c) were prepared with a 27-79% yield by the condensation of the properly substituted O-benzyl hydroxylamine derivative with either p-toluene or methyl sulfonyl chloride under basic conditions (Scheme 2) [23,24]. The variable yields likely arise from mixtures of O and N-mono alkylated products and N, O-dialkylated products as reported [23,24]. [23,24]. The variable yields likely arise from mixtures of O and N-mono alkylated product and N, O-dialkylated products as reported [23,24]. Scheme 2. Synthesis of potential redox-triggered HNO donors. This sequence followed by extraction generally gave the desired products with excellent purity, as judged by proton and carbon nuclear magnetic resonance (NMR spectroscopy and mass spectrometry (MS). Individual compounds could be furthe purified by recrystallization or silica gel flash chromatography if necessary.
The substituted O-benzyl hydroxylamines required in Scheme 1 were purchased (R = -H or -NO2) or prepared using reported literature procedures for the azide and the pinacol-derived boronate ester [22,25]. Scheme 3 summarizes the preparation of the azide containing hydroxylamine (6) from p-toluidine through a four-step sequence that feature diazotization/azide substitution, bromination, N-hydroxyphthalimde substitution to install the N-O bond and hydrazine deprotection through intermediates 3-5 (Scheme 3 [25].  This sequence followed by extraction generally gave the desired products with excellent purity, as judged by proton and carbon nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS). Individual compounds could be further purified by recrystallization or silica gel flash chromatography if necessary.
The substituted O-benzyl hydroxylamines required in Scheme 1 were purchased (R = -H or -NO 2 ) or prepared using reported literature procedures for the azide and the pinacol-derived boronate ester [22,25]. Scheme 3 summarizes the preparation of the azidecontaining hydroxylamine (6) from p-toluidine through a four-step sequence that features diazotization/azide substitution, bromination, N-hydroxyphthalimde substitution to install the N-O bond and hydrazine deprotection through intermediates 3-5 (Scheme 3) [25].
Molecules 2022, 27, x FOR PEER REVIEW 3 of 11 [23,24]. The variable yields likely arise from mixtures of O and N-mono alkylated products and N, O-dialkylated products as reported [23,24].

Scheme 2. Synthesis of potential redox-triggered HNO donors.
This sequence followed by extraction generally gave the desired products with excellent purity, as judged by proton and carbon nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS). Individual compounds could be further purified by recrystallization or silica gel flash chromatography if necessary.
The substituted O-benzyl hydroxylamines required in Scheme 1 were purchased (R = -H or -NO2) or prepared using reported literature procedures for the azide and the pinacol-derived boronate ester [22,25]. Scheme 3 summarizes the preparation of the azidecontaining hydroxylamine (6) from p-toluidine through a four-step sequence that features diazotization/azide substitution, bromination, N-hydroxyphthalimde substitution to install the N-O bond and hydrazine deprotection through intermediates 3-5 (Scheme 3) [25].  The Mitsonobu coupling of N-hydroxyphthalimide with the commercially available pinacol ester of 4-hydroxymethylphenylboronic acid followed by hydrazine deprotection yields the boronate-ester-substituted hydroxylamine (8) via 7 (Scheme 3) [22]. These intermediates were characterized by NMR spectroscopy and MS and purified by normal phase flash chromatography (see Supplementary Materials).

HNO Donation Ability
Piloty's acid (PhSO 2 NHOH), a common HNO donor, decomposes to HNO and phenyl sulfinic acid (PhSO 2 H) at a neutral pH. Compounds 1b-d and 2b,c were designed to decompose upon either reduction or oxidation as depicted in Scheme 1 to yield a reactive intermediate that should release a Piloty's acid derivative (pCH 3 C 4 H 6 SO 2 NHOH or CH 3 SO 2 NHOH) that fragments to HNO and the corresponding sulfinic acid. The gas chro-matographic (GC) headspace measurement of nitrous oxide (N 2 O), the dimerization and dehydration product of HNO, provides a rapid and simple measure of HNO formation from these transformations [26]. The decomposition of the PA derivative pCH 3 C 4 H 6 SO 2 NHOH, a known HNO donor, in methanol/buffer generates 76% of N 2 O by this measure after 24 h ( Table 1). The sensitivity of N 2 O formation to the addition of glutathione (GSH), which rapidly reacts with HNO blocking N 2 O production [10], provides evidence for HNO's intermediacy (Table 1). Compounds 1a and 2a do not contain reduction or oxidation-sensitive functional groups, preventing the proposed decomposition to a Piloty's acid derivative, and do not produce N 2 O under these conditions (Table 1).
The incubation of 1b-d and 2b-c in a methanol/100 mM PBS buffer showed~1% N 2 O formation, indicating essentially no formation of HNO over 24 h from these compounds, which was as expected, in the absence of any reducing/oxidizing agents. The formation of small amounts of N 2 O could arise from the slow hydrolysis/methanolysis of the benzyl group producing pCH 3 C 4 H 6 SO 2 NHOH or CH 3 SO 2 NHOH or from the presence of small amounts of these sulfohydroxamic acids in these samples, which could form from the condensation of residual hydroxylamine potentially present in the commercial or synthetic O-benzylhydroxylamine derivatives. Table 2 summarizes N 2 O formation at 24 h from the treatment of 1a-d and 2a-c with various reducing and oxidizing agents. The addition of sodium borohydride, a reducing agent capable of nitro-to-amine reduction [27], to compounds 1b and 2b generates N 2 O with a 25 and 51% yield, respectively, providing evidence for initial HNO formation ( Table 2). The lower observed amounts of N 2 O from 1b may reflect the poor solubility of 1b, which did not completely dissolve in 1:1 CH 3 OH:H 2 O (2 mL), possibly due to the presence of two para-substituted aromatic rings, including one with the polar nitro group. Compound 1b dissolved with the addition of another 0.5 mL of CH 3 OH (1.5:1 CH 3 OH:H 2 O, 2.5 mL), and the results reported in Table 2 for 1b were obtained under these conditions. Thin-layer chromatography (TLC) and an MS analysis of the reduction of 1b with sodium borohydride provides evidence for the formation of p-amino benzyl alcohol, the product of water addition to the imine of p-quinone methide. The addition of sodium borohydride/copper (II) sulfate, a mixture known to reduce azides to amines [28], to 1c and 2c produces 89 and 92% N 2 O, respectively ( Table 2). These results support Scheme 1 and suggest a reduction of the nitro and azide groups to the amine followed by decomposition to the Piloty's acid derivative. As expected, the treatment of 1a and 2a with sodium borohydride or sodium borohydride/copper (II) sulfate under these conditions did not generate N 2 O (Table 2).
The incubation of 1c and 2c with GSH (2.5 or 5 equivalents) does not produce N 2 O, but the treatment of 1c and 2c with H 2 S (2.5 equivalents) generated small reproducible amounts (6% and 5%, respectively) of N 2 O. These results suggest H 2 S-mediated azide reduction to the amine as described, followed by HNO formation [29]. The lower amounts of N 2 O observed are likely due to the competition between HNO dimerization and H 2 S addition. Increasing the amounts of H 2 S to five equivalents abolished N 2 O formation, supporting this explanation. Table 2 also shows hydrogen peroxide (H 2 O 2 )-mediated N 2 O release from 1d, a boronate-containing compound designed to release HNO via boronate oxidation to the phenol followed by decomposition. The incubation of 1d with H 2 O 2 resulted in the formation of N 2 O with a 23% yield, presumably indicating the formation of HNO upon oxidation to the phenol. The treatment of 1a and 2a with H 2 O 2 did not produce N 2 O ( Table 2). Table 2 also shows that the amount of N 2 O produced increases from the 1 h to 24 h measurements, likely reflecting the kinetics of these model oxidations and reductions.
Bacterial nitroreductases using NADH as a co-substrate act as competent reducing agents of aromatic nitro groups, with the subsequent release of desired compounds via a 1, 6-elimination [30,31]. The treatment of 1b or 2b with Escherichia coli nitroreductase (Sigma) in the presence of NADH failed to generate N 2 O as initially expected. A previous report indicates that NADH reduces HNO to hydroxylamine, thus blocking N 2 O formation and suggesting that the lack of observed N 2 O in these experiments results from HNO reduction by the NADH co-substrate [32]. Further investigation will be necessary to define the practicality of nitroreductase-triggered HNO donors, but these results suggest that HNO may exert a portion of effects through NADH/NADPH depletion.

Materials and Methods
All materials and solvents used for extraction and purification were purchased from commercial vendors and used as received. 1 H and 13 C NMR spectra were recorded using a Bruker Avance 400 MHz NMR spectrometer. Mass spectra were obtained using a Bruker Amazon SL ion trap. An Agilent Technologies 7890A gas chromatograph equipped with a micro-electron capture detector and a 30 m × 0.32 m (25 µm) HP-MOLSIV capillary column was used for the gas chromatographic analysis of N 2 O. CAUTION: Any experiments preparing alkyl or aryl azides should be performed in a well-ventilated fume hood and behind a blast shield. Sodium azide should not be mixed with strong acids.

Gas Chromatographic N 2 O Analysis
For headspace analysis, substrate (1a-d, 2a-c, 0.04 mmol) was placed in a 10 mL roundbottom flask with a stir bar and sealed with a rubber septum. Solvent (methanol:water or methanol:PBS (100 mM), pH = 7.4; 1:1, 2 mL) was added using a syringe, and headspace aliquots (0.1 mL) from each experiment were injected at 1 and 24 h onto a 7890A Agilent Technologies gas chromatograph equipped with a micro-electron capture detector and a 30 m × 0.32 m (25 µm) HP-MOLSIV capillary column. The oven was operated at 200 • C for the duration of the run (4.5 min). The inlet was held at 250 • C and run in split mode (split ratio 1:1) with a total flow (N 2 as carrier gas) of 4 mL/min and a pressure of 37.9 psi. The µECD was held at 325 • C with a makeup flow (N 2 ) of 5 mL/min. The retention time of nitrous oxide was 3.4 min, and yields were calculated based on a standard curve for nitrous oxide gas (Gasco Precision Calibration Mixtures). To some samples, sodium borohydride (1.1 equivalents), sodium borohydride (1.1 equivalents) and copper (II) sulfate (0.1 equivalents), GSH (5 equivalents) or sodium sulfide (2-5 equivalents) were added. For oxidation of 1d, hydrogen peroxide (5 equivalents) was added to a solution (2 mL) of the substrate in a mixture of 3:2:0.5 acetonitrile, water, PBS (100 mM, pH = 7.4).

Conclusions
Nitroxyl (HNO) demonstrates a unique biological profile compared to NO that deserves more detailed studies that have been afforded to other nitrogen oxides and gasotransmitters. The high chemical reactivity of HNO requires the use of donors and relatively complex detection methods, which complicate such studies and the confirmation of endogenous HNO production. HNO exhibits rich redox reactivity with thiols, hydrogen sulfide and heme proteins, suggesting a potential role in various redox-mediated processes.
We report a small library of derivatives (1a-d and 2a-c) of the HNO donor, Piloty's acid, that liberate HNO upon oxidation or reduction through a 1, 6 elimination mechanism. These compounds were quickly constructed by the condensation of a sulfonyl chloride and an appropriately substituted hydroxylamine derivative. The nitro and azide-containing molecules (1b, c and 2b, c) demonstrated HNO release, as measured by headspace GC for N 2 O, upon chemical treatment with reducing agents. Similarly, a boronate ester derivative (1d) generates HNO upon hydrogen peroxide oxidation. Control experiments with 1a and 2a show that only compounds with redox active groups generate HNO under these conditions, supporting the mechanism. Overall, these results show the ability of these redox-sensitive HNO donors to release HNO upon oxidation/reduction, and could find use in further defining the role of HNO in redox-based biological processes.
Author Contributions: Compound synthesis, purification, characterization and evaluation of nitrous oxide release, Y.L. and Z.X.; evaluation of nitrous oxide release, student mentorship and manuscript preparation, A.M.R.; project conception and direction and manuscript preparation, S.B.K. All authors have read and agreed to the published version of the manuscript.