Vascular Pharmacology 60 (2014) 32–41

Contents lists available at ScienceDirect

Vascular Pharmacology journal homepage: www.elsevier.com/locate/vph

Pharmacological characterization of the vascular effects of aryl isothiocyanates: Is hydrogen sulfide the real player? Alma Martelli a, Lara Testai a, Valentina Citi a, Alice Marino a, Francesca G. Bellagambi b, Silvia Ghimenti b, Maria C. Breschi a, Vincenzo Calderone a,⁎ a b

Dipartimento di Farmacia, Università di Pisa, via Bonanno, 6, I-56126 Pisa, Italy Dipartimento di Chimica e Chimica Industriale, Università di Pisa, via Risorgimento, 35, I-56126 Pisa, Italy

a r t i c l e

i n f o

Article history: Received 5 July 2013 Received in revised form 12 November 2013 Accepted 16 November 2013 Keywords: Hydrogen sulfide H2S-releasing drugs Hypertension Vascular smooth muscle Isothiocyanate Chemical compounds studied in this article: Hydrogen sulfide (PubChem CID: 402) Phenyl isothiocyanate (PubChem CID: 7673) 4-carboxyphenyl isothiocyanate (PubChem CID: 242946) GYY4137 (PubMed CID: 53393943) Diallyl disulfide (PubMed CID: 16590)

a b s t r a c t Hydrogen sulfide (H2S) is an endogenous gasotransmitter, which mediates important physiological effects in the cardiovascular system. Accordingly, an impaired production of endogenous H2S contributes to the pathogenesis of important cardiovascular disorders, such as hypertension. Therefore, exogenous compounds, acting as H2Sreleasing agents, are viewed as promising pharmacotherapeutic agents for cardiovascular diseases. Thus, this paper aimed at evaluating the H2S-releasing properties of some aryl isothiocyanate derivatives and their vascular effects. The release of H2S was determined by amperometry, spectrophotometry and gas/mass chromatography. Moreover, the vascular activity of selected isothiocyanates were tested in rat conductance (aorta) and coronary arteries. Since H2S has been recently reported to act as an activator of vascular Kv7 potassium channels, the possible membrane hyperpolarizing effects of isothiocyanates were tested on human vascular smooth muscle (VSM) cells by spectrofluorescent dyes. Among the tested compounds, phenyl isothiocyanate (PhNCS) and 4-carboxyphenyl isothiocyanate (PhNCS–COOH) exhibited slow-H2S-release, triggered by organic thiols such as L-Cysteine. These compounds were endowed with vasorelaxing effects on conductance and coronary arteries. Moreover, these two isothiocyanates caused membrane hyperpolarization of VSM cells. The vascular effects of isothiocyanates were strongly abolished by the selective Kv7-blocker XE991. In conclusion, the isothiocyanate function can be viewed as a suitable slow H2S-releasing moiety, endowed with vasorelaxing and hypotensive effects, typical of this gasotransmitter. Thus, such a chemical moiety can be employed for the development of novel chemical tools for basic studies and promising cardiovascular drugs. © 2013 Elsevier Inc. All rights reserved.

1. Introduction Hydrogen sulfide (H2S) has been recently discovered as a physiological gasotransmitter endogenously produced in substantial amounts by mammalian tissues. At physiological concentrations, H2S mediates

Abbreviations: AB, assay buffer; ACh, acetylcholine; AngII, angiotensin II; CF, coronary flow; CSE, cystathionine-gamma-lyase; DADS, diallyl disulfide; DiBac4(3), bisoxonol dye bis-(1,3-dibutylbarbituric acid); DMSO, dimethylsulfoxide; GYY4137, morpholin-4-ium4-methoxyphenyl-morpholino-phosphinodithioate; H2S, hydrogen sulfide; HASMC, human aortic smooth muscle cell; HR, heart rate; L-NAME, L-nitroarginine methyl ester; LVDP, left ventricular developed pressure; NA, noradrenaline; NO, nitric oxide; PhNCS, phenylisothiocyanate; PhNCS-CF3, 2-trifluorometyl phenylisothiocyanate; PhNCS-CH3, metyl-phenylisothiocyanate; PhNCS-COOH, 4-carboxyphenylisothiocyanate; PhNCS-iPr, 2-isopropyl phenylisothiocyanate; SBP, systolic blood pressure; SMGS, smooth muscle growth supplement; VSM, vascular smooth muscle. ⁎ Corresponding author. Tel.: +39 0 50 2219589; fax: +39 0 50 2219609. E-mail addresses: [email protected] (A. Martelli), [email protected] (L. Testai), [email protected] (V. Citi), [email protected] (A. Marino), [email protected] (M.C. Breschi), [email protected] (V. Calderone). 1537-1891/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.vph.2013.11.003

important physiological effects in many systems. In the cardiovascular system, H2S production is mainly ensured by the enzyme cystathioninegamma-lyase (CSE), starting from the aminoacid L-Cysteine [1]. H2S plays a key role in regulating cardiovascular homeostasis, acting as a direct relaxing agent in the vascular smooth muscle [2]. It is presently accepted that an impaired production of endogenous H2S contributes to the pathogenesis of important cardiovascular disorders, such as hypertension. In fact, the H2S pathway is pivotally involved in the regulation of blood pressure and exogenous H2S-donors effectively prevent the progression of hypertension [3] and decrease the blood pressure in experimental models of hypertension [4]. Consistently, the genetic deletion of CSE in mice is associated with blunted levels of H2S in blood, heart and aorta, with increased blood pressure values and decreased endothelium-mediated vasorelaxant effects [5]. Impaired H2S biosynthesis has been also observed in the setting of cardiovascular complications associated with experimental models of diabetes mellitus [6]. The patho-physiological roles of endogenous H2S in the cardiovascular system highlight the great usefulness of its pharmacological modulation for pharmacotherapeutic purposes [7,8]. Thus, exogenous compounds, which exhibit the pharmacodynamic profile of H2S-releasing agents, are

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

viewed as useful tools for basic studies and promising drugs for cardiovascular diseases. Of course, the administration of gaseous H2S is greatly limited for the risk of poor posologic control and overdose; the use of other appropriate chemicals, behaving as H2S-releasing agents, is largely preferred. Sodium hydrogen sulfide (NaHS) is the prototypical example of H2S-generating agent. It is a rapid H2S-donor and the most widely used H2S-donor for experimental purposes. However, this salt is not appropriate for clinical uses, since the quick release of H2S may cause adverse effects, such as rapid and excessive lowering of blood pressure. Calcium sulfide (CaS) has been proposed as a possible alternative [9], but it should be noted that the rate and mechanism of H2S release from these two inorganic salts are almost equivalent. Ideal H2S-donors should generate H2S with slower releasing rates [10,11]. This pharmacological feature is exhibited by some natural derivatives, typically present in several plants belonging to the botanical family of Alliaceae. Recent and convincing evidences show that some organic polysulfides of garlic, such as diallyl disulfide (DADS; Fig. 1), act as H2S-releasing compounds with relatively slow mechanism, requiring the presence of reduced glutathione [12]. In addition to these natural organic polysulfides, early examples of original synthetic H2S-releasing agents have been described. Among them, the phosphinodithioate derivative GYY4137 (morpholin-4-ium4-methoxyphenyl-morpholino-phosphinodithioate, Fig. 1) is currently used for pharmacological studies, since it ensures a sustained release of H2S over a prolonged period [13]. Interesting H2S-releasing feature of aminothiol [14] and aryl thioamide [15] derivatives has been also reported. As well, the H2S-donor properties of dithiolethiones and thioamides have been widely used for the synthesis of multitarget drugs [16–18]. In contrast, although an isothiocyanate derivative (4-hydroxyphenyl isothiocyanate) has been reported as an example of potential H2Sreleasing side-chain of multitarget anti-inflammatory agents in a recent patent [19], its real H2S-releasing properties have not been characterized. To date, no specific investigation has been focused on the isothiocyanate functional group as a potential source of H2S, and its properties of H2S-donor and H2S-dependent effects on the vascular smooth muscle are unexplored. Thus, this paper aimed to the evaluation and exploitation of the isothiocyanate moiety, as a suitable H2S-donor moiety, by evaluating the H2S-releasing properties of some aryl isothiocyanate derivatives (PhNCS, PhNCS–COOH, PhNCS–CH3, PhNCS–CF3, PhNCS–iPr; Fig. 1)

Fig. 1. Chemical structures. Chemical structures of DADS, GYY4137 and the tested isothiocyanates.

33

and the comparison with the properties exhibited by reference molecules already reported in literature. Moreover, selected isothiocyanate derivatives were submitted to further experimental protocols, aimed to evaluate the vasorelaxing effects on rat's aorta and coronary arteries and the membrane hyperpolarizing activity on human vascular smooth muscle cells. 2. Methods 2.1. Determination of H2S 2.1.1. Amperometry The characterization of the potential H2S-generating properties of the tested compounds has been carried out by amperometric approaches, through an Apollo-4000 Free Radical Analyzer (WPI) detector and H2S-selective minielectrodes. The experiments were carried out at room temperature. Following the manufacturer's instructions, a “PBS buffer 10 ×” was prepared (NaH2PO4·H2O 1.28 g, Na2HPO4·12H2O 5.97 g, NaCl 43.88 g in 500 mL H2O) and stocked at 4 °C. Immediately before the experiments, the “PBS buffer 10 ×” was diluted in distilled water (1:10), to obtain the assay buffer (AB); pH was adjusted to 7.4. The H2S-selective minielectrode was equilibrated in 10 ml of the AB, until the recovery of a stable baseline. Then, 100 μl of a dimethyl sulfoxide (DMSO) solution of all the tested H2S-releasing compounds was added (final concentration of the tested H2S-donors 1 mM; final concentration of DMSO in the AB 1%). The eventual generation of H2S was observed for 15 min. When required by the experimental protocol, L-Cysteine (0.33, 1, 2 or 4 mM) was added, before the H2S-donors. The correct relationship between the amperometric currents (recorded in pA) and the corresponding concentrations of H2S was determined by opportune calibration curves, which were previously obtained by the use of increasing concentrations of NaHS (1 μM, 3 μM, 5 μM, 10 μM) at pH 4.0. The curves relative to the progressive increase of H2S vs time, following the incubation of the tested compounds, were analysed by the equation   −k·t Ct ¼ Cmax − Cmax ·e where Ct is the instant concentration at time t, and Cmax is the highest concentration achieved in the recording time. The constant k is 0.693/thc, where thc (time for half concentration) is the time required to reach a concentration = ½ Cmax. The values of Cmax and thc were calculated by a computer fitting procedure (software: GraphPad Prism 4.0) and expressed as mean ± standard error; at least 5 different curves were performed for each compound. ANOVA and Student t test were selected as statistical analysis, P b 0.05 was considered representative of significant statistical differences. 2.1.2. Spectrophotometry At the end of the above amperometric approach, an usual spectrophotometric method has been also used, in order to have further indication about the H2S-releasing properties of PhNCS. Briefly, at the end of the amperometric recording, 800 μl of the AB containing PhNCS (in the absence or in the presence of L-Cysteine) were mixed with 50 μl of N,N-dimethyl-phenylen-diamine (Sigma-Aldrich) (20 mM in hydrochloric acid solution 7.2 M) and 50 μl of FeCl3 (Sigma-Aldrich) (30 mM in hydrochloric acid solution 1.2 M). After 20 min, required for the methylene blue formation, the absorbance at 670 nm has been read through a spectrophotometer Novaspec Plus (Amersham Biosciences). The spectrophotometric measurements were converted to the corresponding concentrations of H2S, by opportune calibration curves previously obtained by the use of increasing concentrations of NaHS (1 μM, 3 μM, 5 μM, 10 μM). The values of Cmax were expressed as mean ± standard error, from at least 5 different measurements.

34

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

2.1.3. Gas chromatography/mass spectrometry The confirmation of the H2S formation was also verified by headspace-gas chromatography–mass spectrometry analysis. 100 μl of DMSO solution of PhNCS–COOH was added (final concentration of the tested H2S-donor 1 mM; final concentration of DMSO in the AB 1%), to 10 ml of the AB, in the presence or in the absence of L-Cysteine (4 mM). 5 ml of the above solutions was stored in 6 ml glass screw cap headspace vials (Agilent Technologies) with PTFE/ silicone pre-slit septa. All headspace equilibrations were performed at 25 °C for 30 min. 500 μl of headspace gas of each sample was directly injected into a 6% cyanopropyl phenyl siloxane and 94% dimethylpolysiloxane capillary column (DB-624, 60 m length, 0.25 mm internal diameter, 1.4 μm film thickness, Agilent Technologies) of the gas chromatograph (Trace GC Ultra, Thermo Electron Corporation) coupled to a quadrupole mass spectrometer (Trace DSQ, Thermo Electron Corporation) operated in the positive electron impact ionisation (70 eV). Chromatograms were collected in total ion current (TIC) acquisition mode (18–200 amu). The oven temperature programme was 30 °C for 5 min, 15 °C min−1 to 260 °C, 10 min hold. The temperature of the injector was set at 200 °C. Helium (constant pressure 210 kPa, split flow of 10 ml min−1) was used as carrier gas. 2.2. Evaluation of the functional effects on rat aortic rings All the experimental procedures were carried out following the guidelines of the European Community Council Directive 86–609 and in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki, EU Directive 2010/63/EU for animal experiments). To determine a possible vasodilator mechanism of action, the compounds were tested on isolated thoracic aortic rings of male normotensive Wistar rats (250–350 g). Rats were sacrificed by cervical dislocation under light ether anaesthesia and bled. The aortae were immediately excised and freed of extraneous tissues. When required by the experimental protocol, the endothelial layer was removed by gently rubbing the intimal surface of the vessels with a hypodermic needle. Five mm wide aortic rings were suspended, under a preload of 2 g, in 20 ml organ baths, containing Tyrode solution (composition of saline in mM: NaCl 136.8; KCl 2.95; CaCl.22H2O 1.80; MgSO.47H2O 1.05; NaH2PO.4H2O 0.41; NaHCO3 11.9; Glucose 5.5), thermostated at 37 °C and continuously gassed with Clioxicarb, a mixture of O2 (95%) and CO2 (5%). Changes in tension were recorded by means of an isometric transducer (Grass FTO3), connected with a preamplifier (Buxco Electronics) and with a software of data acquisition (BIOPAC Systems Inc., MP 100). After an equilibration period of 60 min, the endothelial presence or removal was confirmed by the administration of acetylcholine (ACh) (10 μM) to KCl (25 mM)-precontracted vascular rings. As regards the preparations with intact endothelium, a relaxation ≥70% of the KCl-induced contraction was considered representative of a suitable presence of functional endothelium; while the organs, showing a relaxation b70% (i.e. unsatisfactory endothelial functionality), were discarded. In contrast, for the endothelium-removed aortic rings, a relaxation b10% of the KCl-induced contraction was considered representative of an acceptable lack of the endothelial layer, while the organs, showing a relaxation ≥10% (i.e. significant presence of the endothelium), were discarded. Then, the preparation was submitted to the following experimental protocols. 2.2.1. Direct vasorelaxing effects on endothelium-denuded aortic rings 45 min after the confirmation of the endothelium removal, the aortic preparations were re-contracted by KCl 25 mM and when the contraction reached a stable plateau, NaHS, PhNCS or PhNCS–COOH were added cumulatively (10 μM–1 mM). In some experiments, the selective Kv7-blocker XE991 10 μM was added 20 min before KCl.

Preliminary experiments showed that the KCl (25 mM)-induced contractions remained in a stable tonic state for at least 40 min. The vasorelaxing efficacy (E max ) was defined as maximal vasorelaxing response achieved with the highest concentration (1 mM) of the tested compounds, expressed as a percentage (%) of the contractile tone induced by KCl. The parameter of potency was expressed as pEC50, calculated as negative Logarithm of the molar concentration of the tested compounds evoking an effect = 50% of Emax. The parameters of efficacy and potency were expressed as mean ± standard error, from aortae of 6–10 animals.

2.2.2. Direct vasorelaxing effects on endothelium-intact aortic rings 45 min after the confirmation of the endothelium presence/ functionality, the aortic preparations were re-contracted by KCl 25 mM and when the contraction reached a stable plateau, PhNCSCOOH was added cumulatively (10 μM–1 mM). In some experiments, the L-NAME 100 μM (inhibitor of the biosynthesis of endothelial NO) was added 20 min before KCl. Preliminary experiments showed that the KCl (25 mM)-induced contractions remained in a stable tonic state for at least 40 min. The vasorelaxing efficacy (Emax) was defined as maximal vasorelaxing response achieved with the highest concentration (1 mM) of the tested compounds, expressed as a percentage (%) of the contractile tone induced by KCl. The parameter of potency was expressed as pEC50, calculated as negative Logarithm of the molar concentration of the tested compounds evoking an effect = 50% of Emax. The parameters of efficacy and potency were expressed as mean ± standard error, from aortae of 6–10 animals.

2.2.3. Inhibition of NA-mediated vasoconstriction 45 min after the confirmation of the endothelium removal, the vessels were contracted by 60 mM KCl, i.e. a depolarizing stimulus able to evoke an almost full contraction of vascular smooth muscle. After the contraction achieved a stable plateau, the preparations were submitted to a wash-out and to a further equilibration time (45 min). Then, the H2S-releasing compounds (or the corresponding vehicle) were incubated for 20 min at different concentrations (10 μM–1 mM). At the end of the incubation time, 3-fold increasing concentrations of noradrenaline (NA, 1 nM–1 μM) were added cumulatively. After completing the above concentration-response curve for NA, the preparations were again submitted to a wash-out and to a further equilibration time (45 min). Then, 3-fold increasing concentrations of noradrenaline (NA, 1 nM–1 μM) were cumulatively added again, to test the viability of the vessel and the “reversibility” of the H2S-releasing compounds. The concentration-response curves relative to the vasoconstriction induced by NA were analyzed by the Hill equation. The NA efficacy was evaluated as maximal contractile response, expressed as a percentage (%) of the contractile tone previously induced by 60 mM KCl. The parameter of potency was expressed as pEC50, calculated as negative Logarithm of the molar concentration of NA, evoking a half-maximal response. The parameters of efficacy and potency of NA were expressed as mean ± standard error, for 6–10 experiments. Moreover, the relationships between the increasing concentrations of the H2S-releasing compounds and the progressive reduction (expressed as a %) of the vasoconstriction evoked by NA 1 μM, allowed the construction of inhibition curves and the determination of the inhibitory potencies and efficacies of the H2S-donors. In particular, the efficacy of the H2S-donors was evaluated as maximal inhibitory effects (expressed as a %) of the vasoconstriction evoked by NA 1 μM, and the parameter of potency was expressed as pIC50, calculated as negative Logarithm of the molar concentration of the H2S-donor, evoking a half-reduction of the vasoconstriction evoked by NA 1 μM. The parameters of inhibitory efficacy and potency of the H2S-releasing agents were expressed as mean ± standard error, from aortae of 6–10 animals.

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

2.3. Evaluation of the functional effects on the coronary flow The animals were euthanized with an overdose of sodium pentobarbital (100 mg/kg, i.p.) and bled. After the opening of the chest, the hearts were quickly excised and placed in Krebs solution at 4 °C (composition mM: NaHCO3 25.0, NaCl 118.1, KCl 4.8, MgSO4·7H2O 1.2, CaCl2·2H2O 1.6, KH2PO4 1.2, glucose 11.5) equilibrated with 95% O2 5% CO2. Rapidly, the ascending aorta was cannulated and the heart was mounted on a Langendorff apparatus, and perfused with Krebs solution (thermostated at 37 °C and continuously bubbled with a gas mixture of 95% O2 and 5% CO2). Perfusion was carried out at constant pressure (70–80 mmHg). Then, a bolus of heparin (100 UI i.p.) was administered to prevent blood clotting. The above procedure was completed within about 2 min. In order to monitor the functional parameters, a water-filled latex balloon connected to a pressure transducer (Bentley Trantec, mod 800) was introduced into the left ventricle via the mitral valve and the volume was adjusted to achieve a stable left ventricular end-diastolic pressure of 5-10 mmHg during initial equilibration. The heart rate (HR) and left ventricular developed pressure (LVDP) were continuously monitored by a computerised Biopac system (California, USA), in order to discard hearts showing severe arrhythmia or unstable LVDP and HR values, during the equilibration period [20]. Coronary flow (CF) was volumetrically measured at 5 min intervals and expressed as ml/min, normalized by the heart weight. After a 20 min equilibration period, the hearts were subjected to the following experimental protocols. 2.3.1. Effects of H2S-donors on basal CF The hearts were perfused with Krebs solution medicated with NaHS, PhNCS or PhNCS–COOH at cumulatively increasing concentrations (10–30–100–300 μM, 20 min of perfusion for each concentration). 2.3.2. Effects of H2S-donors on angiotensin II-reduced CF Angiotensin II (AngII) 0.1 μM was administered through the perfusion. Once obtained a stable coronary spasm (evaluated as a reduction of the CF), cumulatively increasing concentrations of NaHS, PhNCS or PhNCS–COOH (10–30–100–300 μM, 20 min of perfusion for each concentration) were administered (in the constant presence of AngII 0.1 μM). Preliminary experiments demonstrated that the treatment with AngII 0.1 μM alone caused a rapid decrease of the CF, which reached and maintained a stable level for at least 2 h. 2.3.3. Data analysis Changes in CF, recorded after the pharmacological treatments, were expressed as a % of the CF measured in the last 5 min of the equilibration period. The parameters were evaluated as mean ± standard error, from hearts of 6–10 animals. 2.4. Evaluation of the membrane hyperpolarizing effects on HASMCs The possible hyperpolarizing effects were evaluated on human aortic smooth muscle cell (HASMC, Invitrogen) by spectrofluorimetric methods, as already described [21,22]. Briefly, HASMCs were cultured in Medium 231 (Invitrogen) supplemented with Smooth Muscle Growth Supplement (SMGS, Invitrogen) and 1% of 100 units/ml penicillin and 100 mg/ml streptomycin (Sigma Aldrich) in tissue culture flasks at 37 °C in a humidified atmosphere of 5% CO2. HASMCs were cultured up to about 90% confluence and 24 h before the experiment cells were seeded onto a 96-well black plate, clear bottom pre-coated with gelatine 1% (from porcine skin, Sigma Aldrich), at density of 72 × 103 per well. After 24 h to allow cell attachment, the medium was replaced and cells were incubated for 1 h with an appropriate buffer (HEPES 20 mM, NaCl 120 mM, KCl 2 mM, CaCl2·2H2O 2 mM, MgCl2·6H2O 1 mM, Glucose 5 mM, pH 7.4, at room temperature) containing the

35

bisoxonol dye bis-(1,3-dibutylbarbituric acid) DiBac4(3) at 2.5 μM (Sigma Aldrich). This membrane potential-sensitive dye DiBac4(3) allows us a non-electrophysiological measurement of cell membrane potential [23,24]; in fact, this lipophylic and negatively-charged oxonol dye shuffles between cellular and extracellular fluids in a membrane potential-dependent manner (following the Nernst equation), thus allowing to assess changes in membrane potential by means of spectrofluorimetric recording. In particular, an increase of fluorescence, corresponding to an inward flow of the dye, reflects a membrane depolarization; in contrast, a decrease in fluorescence, due to an outward flow of the dye, is linked to membrane hyperpolarization. The spectrofluorimetric recording is carried out at excitation and emission wavelengths of 488 and 520 nm, respectively (Multiwells reader, Enspire, PerkinElmer). Retigabine (50 μM), a well-known activator of Kv7 channels, was used as reference hyperpolarizing drug. After the assessment of base-line fluorescence, when requested by the experimental protocol, the selective Kv7 inhibitor XE991 10 μM or its vehicle were incubated 20 min before the addition of the test compounds. The concentration 1 mM of NaHS was selected because it caused maximal vasorelaxing effects in the functional assay on rat aortic rings. Preliminary experiments showed that NaHS 1 mM was devoid of significant toxic effects on HASMCs, while higher concentrations caused a reduction in cell viability (data not shown). The relativefluorescence decrease, linked to hyperpolarizing effects, was calculated as: ð Ft −F0 Þ=F0 where F0 is the basal fluorescence before the addition of the tested compounds, and Ft is the fluorescence at time t after their administration. The change in fluorescence where expressed as a % of the maximal one induced by retigabine 50 μM. Data were expressed as mean ± standard error; six different experiments were performed, each carried out in six replicates.

2.5. Statistic analysis Experimental data were analysed by a computer fitting procedure (software: GraphPad Prism 4.0). ANOVA and Student t test were selected as statistical analyses; P b 0.05 was considered representative of significant statistical differences.

2.6. Substances NaHS (Sigma-Aldrich), phenylisothiocyanate (Sigma-Aldrich) and 4-carboxy-phenylisothiocyanate, metyl-phenylisothiocyanate, trifluorometyl-phenylisothiocyanate and isopropyl-phenylisothiocyanate (Fluorochem), were dissolved 100 mM in DMSO (Carlo Erba) and further diluted in Tyrode solution. KCl (Carlo Erba) was dissolved 2.5 M in Tyrode solution. (±) Noradrenaline (+)-bitartrate (Sigma-Aldrich) was dissolved (1 mM) and further diluted in Tyrode solution. L-Cysteine (Sigma-Aldrich) was dissolved 4 mM in PBS and, when required, dissolved in PBS. XE991 hydrochloride (Tocris) was dissolved (2 mM) in bidistilled water and further diluted in the appropriate buffer. Angiotensin II was dissolved 10 mM in Krebs and further diluted in Krebs solution. Retigabine was dissolved 10 mM in DMSO and further diluted in the appropriate buffer cited above. Stock solutions (100 mM in EtOH 95%) of acetylcholine chloride (Sigma-Aldrich) were further diluted in Tyrode solution. All the solutions were freshly prepared immediately before the pharmacological experimental procedures. Previous experiments have demonstrated ineffectiveness of the administration of the vehicles used under these experimental conditions.

36

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

3. Results 3.1. Determination of H2S The addition of NaHS 1 mM at pH 7.4 (in the absence of L-Cysteine) was followed by an immediate formation of highest concentration of H2S (about 200 μM), according to its physicochemical properties [25,26]. Thereafter, the concentration of H2S showed a rapid fall (by about 40%) in the first 2 min after the addition of NaHS, followed by a slower and apparently constant decline in the remaining time (Fig. 2). The H2S release from NaHS was not influenced by the presence of L-Cysteine (data not shown). As already reported [15], the incubation of the reference drugs DADS and GYY4137 led to a negligible formation of H2S in the absence of L-Cysteine. In the presence of the aminoacid (4 mM), the incubation of DADS was followed by a relatively slow release of H2S, with Cmax of about 20 μM and thc of about 1.5 min. In the presence of L-Cysteine 4 mM, the other reference drug GYY4137 exhibited a similar profile, leading to a slow and significant release of H2S, with Cmax of about 10 μM and thc of about 2.5 min (Fig. 3). The values of Cmax and thc of the reference drugs along with the tested isothiocyanates are shown in Table 1. In the absence of L-Cysteine, the incubation of all the isothiocyanates led to the formation of negligible amounts of H2S. In contrast, the presence of increasing concentrations of this aminoacid lead to a wellrelated increase of the H2S release from PhNCS, as clearly indicated by both the amperometric recordings and the matched spectrophotometric measurements (Fig. 3; Table 2). In the presence of the highest concentration of L-Cysteine (4 mM), the incubation of PhNCS led to a slow (thc of about 5 min) release of H2S (up to a Cmax of about 20 μM). The presence of the 4-carboxy group (PhNCS–COOH) caused an improvement of the L-Cysteine-mediated H2S-release (Cmax of 35 μM, with L-Cysteine 4 mM), without affecting the rate of the process (thc of about 5 min) (Fig. 3). The presence of any substituent in position 2 caused a dramatic fall in the H2S-releasing properties; indeed, PhNCS–CH3, PhNCS–CF3 and PhNCS–iPr were almost completely unable to generate H2S, even in the presence of L-Cysteine. Finally, PhNCS–COOH was selected for a further unequivocal demonstration of the generation of H2S by means of gas/mass chromatographic approaches. After the incubation of PhNCS–COOH in the AB, in the presence of L-Cysteine, H2S was clearly detected in the headspace of the vials. In contrast, no significant formation of H2S was revealed in the absence of L-Cysteine (chromatographic recordings and mass spectra are supplied as supporting information). 3.2. Evaluation of the functional effects on rat aortic rings 3.2.1. Direct vasorelaxing effects As shown in Fig. 4, NaHS promoted concentration-dependent vasorelaxing effects on endothelium denuded rat aortic rings pre-

contracted with KCl 25 mM, with a pEC50 value of 3.84 ± 0.02 and full vasorelaxing efficacy of (Emax = 98 ± 1). XE991 significantly antagonized the vasodilator effects of NaHS (Fig. 4). PhNCS produced weak vasorelaxing effects, with lower levels of efficacy (Emax = 18 ± 2). The carboxy derivative PhNCS–COOH showed an almost full vasorelaxing efficacy (Emax = 90 ± 2), with level of potency (pEC50 = 4.17 ± 0.06) higher than that exhibited by NaHS. The Kv7-blocker XE991 almost completely abolished the vasorelaxing effects of both PhNCS and PhNCS–COOH (Fig. 4). The H2S-donor PhNCS–COOH was selected for further investigation on endothelium-intact aortic rings, i.e. in an experimental model more closely resembling the physiological feature of blood vessels. In endothelium-intact vessels, PhNCS–COOH showed almost full vasorelaxing efficacy (Emax = 91 ± 3), with level of potency (pEC50 = 4.68 ± 0.07) higher than that exhibited in endotheliumdenuded aortic rings. L-NAME did not influence the vasorelaxing efficacy of PhNCS–COOH, but caused a significant reduction in the potency (pEC50 = 4.26 ± 0.07) (Fig. 4). 3.2.2. Inhibition of NA-mediated vasoconstriction In the vehicle pre-treated endothelium-denuded vessels, NA produced a strong vasoconstriction, with an Emax of 128 ± 7 and a potency value (pEC50) of 7.76 ± 0.08. The pre-incubation of increasing concentration of NaHS (0.1, 0.3 and 1 mM) caused a progressive inhibition of the potency and efficacy of NA. In the presence of NaHS 1 mM, the catecholamine exhibited an Emax of 44 ± 3 and a potency value (pEC50) of 6.98 ± 0.09 (Fig. 5). The pre-incubation of the aortic rings with PhNCS (0.01–1 mM) or PhNCS–COOH (0.01–1 mM) led to concentrationdependent inhibitory effects stronger than those produced by NaHS. In fact, both the compounds, at the concentration 300 μM almost fully abolished the ability of NA to evoke any vasoconstriction (Fig. 5). The inhibition curves allow the evaluation of a pIC50 value of 3.27 ± 0.07, 4.01 ± 0.01 and 4.17 ± 0.03, for NaHS, PhNCS and PhNCS–COOH, respectively (Fig. 5). Noteworthy, the inhibitory effects of NaHS, PhNCS, PhNCS–COOH, were fully reversible; in fact, their wash-out was followed by a complete recovery of the contractile effects of NA (Fig. 6). 3.3. Evaluation of the functional effects on the CF The basal CF in Langendorff-perfused rat hearts was 10.6 ± 0.44 ml/min/g. The administration of 10, 30 and 100 μM NaHS led to a concentration related increase of CF, up to about 130%. Further increase of NaHS concentration did not cause additional effects (Fig. 7A). A very similar feature was exhibited by both the isothiocyanates: indeed, PhNCS and PhNCS–COOH increased the CF, up to 130 and 140%, respectively (Fig. 7). In the presence of AngII (0.1 μM), the CF was significantly reduced (~70%, Fig. 7). Even in the constant presence of AngII, NaHS and the two isothiocyanates caused a concentration-related improvement of the CF, leading (at the concentration 30 μM) to an almost complete recovery of the basal coronary flow and, thus, abolishing the AngIImediated vasoconstriction. Higher concentrations of NaHS and PhNCS–COOH really led to further increases of CF, up to 130 and 150%, respectively (Fig. 7). 3.4. Evaluation of the membrane hyperpolarizing effects on HASMCs

Fig. 2. H2S-release from NaHS. Data from a single representative experiment, describing the change of H2S concentration with respect to time, after the addition of NaHS 1 mM in the AB, in the absence of L-Cysteine. H2S was recorded by amperometry.

The administration of NaHS caused a significant membrane hyperpolarization of HASMCs. The hyperpolarization induced by NaHS 1 mM was 45 ± 10% of that induced by retigabine, and the Kv7blocker XE991 significantly antagonized such a hyperpolarizing response (Fig. 8). Both PhNCS and PhNCS–COOH hyperpolarized the membranes of vascular smooth muscle cells, in a concentrationdependent fashion. The maximal hyperpolarizing effects of PhNCS 1 mM and PhNCS–COOH 1 mM were 81 ± 6 and 55 ± 15, respectively (expressed as a % of the effect produced by retigabine); however, almost

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

37

Fig. 3. H2S-release from the slow H2S-donors. A: the curves describe the increase of H2S concentration with respect to time, following the incubation of PhNCS-COOH in AB, in the absence (white squares) or in the presence of L-Cysteine 4 mM (black squares). H2S was recorded by amperometry; the vertical bars indicate the SEM. B: the curves describe the increase of H2S concentration with respect to time, following the incubation of PhNCS in AB, in the absence (white squares) or in the presence of L-Cysteine 0.33 mM (inverted white triangles), 1 mM (white triangles), 2 mM (white diamonds) or 4 mM (black squares). H2S was recorded by amperometry; the vertical bars indicate the SEM. C: the histograms describe the H2S concentration, determined by spectrophotometric technique, after the incubation of 1 mM PhNCS in AB, in the absence or in the presence of different concentrations of L-Cysteine (0.33, 1, 2 or 4 mM). The vertical bars indicate the SEM.

maximal effects were evoked even by lower concentrations of the two isothiocyanates. The hyperpolarizing effects of both the isothiocyanates (at the concentration 100 μM, equieffective of NaHS 1 mM) were significantly antagonized by XE991 (Fig. 8).

4. Discussion The emerging role of H2S in regulating the function of vascular smooth muscle is presently stressing the importance of developing suitable H2S-releasing agents as useful experimental tools and potentially innovative cardiovascular drugs. At the best of our knowledge, no experimental study has been specifically focused on the isothiocyanate Table 1 The table summarizes the parameters of Cmax and thc, relative to the H2S-releasing effects exhibited by the synthesized compounds and reference drugs, after their incubation in AB at physiological pH and temperature, in the absence (AB) or in the presence (AB + L-Cys) of L-Cysteine 4 mM. Data are expressed as means ± SEM. AB + L-Cys

AB Compound

Cmax (μM)

thc (min)

Cmax (μM)

thc (min)

DADS GYY4137 PhNCS PhNCS–COOH PhNCS–CH3 PhNCS–CF3 PhNCS–iPr

b3 b3 b3 b3 b3 b3 b3

nc nc nc nc nc nc nc

19.4 10.3 21.9 35.0 b3 b3 b3

1.5 2.5 4.6 5.1 nc nc nc

± ± ± ±

5.5 2.6 2.9 3.2

± ± ± ±

0.3 0.8 0.5 0.2

functional group, as a potentially suitable H2S-donor moiety, and no specific characterization of the H2S-releasing profile of this chemical class has been undertaken, to date. In this paper, the H2S-donor profile of five aryl isothiocyanates (PhNCS, PhNCS–COOH, PhNCS–CH3, PhNCS–CF3 and PhNCS–iPr) has been investigated and compared with well known reference drugs, such as NaHS, DADS and GYY4137. All the compounds were first submitted to an amperometric evaluation, which allowed us to have a “real-time” determination of the H2Srelease and thus to perform a qualitative/quantitative description of the process. As expected, the addition of NaHS led to a rapid generation of H2S. In contrast, DADS and GYY4137 showed a relatively slow and L-Cysteine-dependent release of H 2 S. Indeed, the process of H 2 Srelease from DADS has been already described to require the

Table 2 The table summarizes the parameters of Cmax recorded by amperometric or spectrophotometric measurements, after 15 min incubation of PhNCS in the absence or in the presence of different concentrations of L-Cysteine. Data are expressed as means ± SEM. Cmax of PhNCS (μM) L-Cysteine

0 0.33 1 2 4

(mM)

Amperometry

Spectrophotometry

0.0 9.7 14.2 14.8 21.9

1.5 4.3 13.6 19.4 28.7

± ± ± ± ±

0.0 0.2 0.2 0.4 2.9

± ± ± ± ±

0.3 0.3 0.8 0.5 4.0

38

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

Fig. 4. Vasorelaxing effects. Concentration-response curves, relative to the vasorelaxing effects evoked by NaHS (A), PhNCS (B) or PhNCS–COOH (C), in endothelium-denuded rat aortic rings, pre-contracted with KCl 25 mM. The vasorelaxing effects have been recorded either in the absence or in the presence of the Kv7-blocker XE991 10 μM. The concentrationresponse curve for PhNCS–COOH, recorded in endothelium-intact aortic rings (+End), in the presence or in the absence of L-NAME 100 μM (inhibitor of the NO-biosynthesis) are also shown (D). The vertical bars indicate the SEM.

Fig. 5. Inhibition of NA-induced vasoconstriction. A–C: concentration-response curves to NA, obtained in endothelium-denuded rat aortic rings pre-incubated with NaHS (A), PhNCS (B) or PhNCS-COOH (C). Each H2S-donor was pre-incubated at the concentration 1 mM (white diamonds), 300 μM (white cyrcles), 100 μM (white inverted triangles), 30 μM (white triangles) or 10 μM (white squares). The effects of the corresponding vehicles on the concentration-response curves to NA are also shown (black squares). The effects are expressed as a % of the contractile responses previously induced by the administration of KCl 60 mM. The vertical bars indicate the SEM. D: Concentration-related inhibiting effects evoked by NaHS (black squares), PhNCS (white triangles) or PhNCS–COOH (black triangles), on the vasoconstriction induced by 1 μM NA in endothelium-denuded rat aortic rings. The NA-induced effects are expressed as a %. The vertical bars indicate the SEM.

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

39

Fig. 6. Microdynamometric tracings. Representative microdynamometric tracings obtained in two different aortic rings, demonstrating that the inhibitory effects of the H2S-donors were fully reversible. In the upper panel, one aortic ring was treated with KCl 60 mM (phase A) and the KCl 60 mM-induced contraction was followed by wash-out (WO) and adequate equilibration time to recover the resting tension. Then, the aortic ring was treated with vehicle (20 min); thereafter cumulative increasing concentrations of NA – evoking concentrationdependent vasoconstriction – were added (phase B). The aortic ring was submitted to wash-out (WO) and adequate equilibration time to recover again the resting tension. Then, the aortic ring was treated again with vehicle (20 min); thereafter, cumulative increasing concentrations of NA – evoking concentration-dependent vasoconstriction – were added again (phase C). In the lower panel, the other aortic ring was treated with KCl 60 mM (phase A') and the KCl 60 mM-induced contraction was followed by wash-out (WO) and adequate equilibration time to recover the resting tension. Then, this aortic ring was treated with PhNCS–COOH 300 μM (20 min); thereafter cumulative increasing concentrations of NA – evoking reduced concentration-dependent vasoconstriction – were added (phase B'). The aortic ring was submitted to wash-out (WO) and adequate equilibration time to recover again the resting tension. Then, the aortic ring was treated with vehicle (20 min); thereafter cumulative increasing concentrations of NA – evoking concentration-dependent vasoconstriction – were added again (phase C'). In the lower panel the complete recovery of the vasoconstriction effects of NA is evident.

presence of organic thiols, such as glutathione [12]. In the absence of L-Cysteine, the incubation of all the isothiocyanates led to the formation of negligible amounts of H2S. In the presence of L-Cysteine, the

incubation of PhNCS led to a slow release of H2S. The presence of a 4-carboxy group (PhNCS–COOH) caused an increase of L-Cysteine-mediated H2S-release, without affecting the slow rate of the process. Like

Fig. 7. Effects on coronary flows. Changes of CF (expressed as a % of the basal CF) induced by the perfusion of increasing concentrations of NaHS, PhNCS or PhNCS–COOH in the absence (A) or in the presence (B) of AngII. The vertical bars indicate the SEM.

40

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

Fig. 8. Hyperpolarizing effects. A) Hyperpolarizing effects induced by NaHS 1 mM (black columns), and by increasing concentrations of PhNCS (10 μM–1 mM white columns) or PhNCS–COOH (10 μM–1 mM grey columns) in HASMCs. B) Hyperpolarizing effects induced by NaHS 1 mM (black columns), PhNCS (100 μM, white columns) or PhNCS– COOH (100 μM, grey columns) in HASMCs, in the presence or in the absence of the Kv7blocker XE991 10 μM. Data are calculated as changes in relative fluorescence and then expressed as a % of the maximal effect induced by retigabine 50 μM. The vertical bars indicate the SEM.

DADS and GYY4137, the incubation of these two isothiocyanates (in the presence of L-Cysteine) led to reach a steady-state concentrations of H2S; such a characteristic is likely to be a valuable quality to avoid the some dangerous side-effects typical of rapid H2S-donors [13,27,28]. The generation of H2S, recorded by amperometric approaches, was further confirmed by suitable spectrophotometric and gass/mass chromatographic determination. The presence of substituents in position 2 caused a dramatic fall in the H2S-releasing properties; indeed, the phenylisothiocyanates PhNCS–CH3, PhNCS–CF3 and PhNCS–iPr were almost completely unable to generate H2S, even in the presence of L-Cysteine. Presently, it cannot be clearly defined whether the effect of the substituent in position 2 is due to the steric hindrance or the electronic influence. However, the obligatory role of organic thiols (such as L-Cysteine) in the H2S-releasing process may imply a possible nucleophilic attack by this aminoacid. Noteworthy, the formation of H2S from a possibly labile phenylthiohydantoin (deriving, in turn, by the reaction between a isothiocyanates and sufur aminoacids, such as cystine) was first described by Edman [29], but the mechanism of reaction is yet unknown. Since this work did not aim at defining the chemical mechanisms and the side products of the reaction, future experimental work will be carried out, in order to achieve a more exhaustive interpretation of the chemistry of the process and of the experimental data of the present paper.

Irrespective of the mechanism of reaction, such an organic thioldependency of the H2S-releasing process has been viewed as a particularly advantageous property, because it allows these compounds to behave as “smart” H2S-donors, able to release the gasotransmitter only in a biological environment. The original H2S-donors PhNCS and PhNCS–COOH were used for further pharmacological studies, to evaluate whether they were able to produce the vascular effects typical of endogenous H2S. The rapid H2Sdonor NaHS was tested for comparison, in order to evaluate if the H2S-releasing rate may influence the biological responses. As reported above, H2S is an ubiquitary mediator, which regulates the homeostasis of the cardiovascular function, through different mechanisms of action [30]. In particular, H2S promotes relaxing effects on vascular smooth muscle [2]. Therefore, the possible direct vasorelaxing effects of PhNCS and PhNCS–COOH were tested in endothelium-denuded isolated rat aortic rings. Moreover, the influence of PhNCS and PhNCS–COOH on the vasoconstriction induced by NA was tested in this experimental model. As expected, NaHS promoted concentration-dependent vasorelaxing effects on endothelium denuded rat aortic rings. The pre-incubation of increasing concentration of NaHS caused also a concentration-related inhibition of the vasocontracturant potency and efficacy of NA. According to their behaviour of H2S-donors, also PhNCS and PhNCS– COOH produced vasorelaxing effects on endothelium-denuded rat aortic rings. As well, the two isothiocyanates led to concentration-related inhibition of the vasoconstriction induced by NA, stronger than that produced by NaHS. The H2S-donor PhNCS–COOH was selected for further investigation on endothelium-intact aortic rings, i.e. in an experimental model more closely resembling the physiological feature of blood vessels. In endothelium-intact vessels, PhNCS–COOH showed a vasorelaxing potency higher than that exhibited in endotheliumdenuded aortic rings. L-NAME, inhibitor of the biosynthesis of endothelial NO, caused a significant reduction in the vasorelaxing potency of PhNCS–COOH. Noteworthy, the effects of PhNCS–COOH in endothelium-intact L-NAME treated vessels were almost completely equivalent to those observed in the endothelium-denuded ones. Thus, the H2S-donor PhNCS is able to produce direct and endotheliumindependent vasorelaxation; however, in the presence of the endothelial layer, these vasorelaxing effects are strengthened by a NO-mediated contribution. The vasorelaxing effects of NaHS and of the two isothiocyanates were also observed in the coronary vascular bed. Indeed, all these compounds caused an increase of the basal coronary flow; moreover, they effectively counteracted the coronary vasoconstriction induced by AngII. The mechanism of the vasodilator effects of H 2 S is likely to be due to heterogeneous mechanisms of action: H 2S influences the activity of several protein kinases, such as p38 mitogen-activated protein kinase, extracellular signal-regulated kinase, and Akt [30]. H 2 S-mediated inhibition of phosphodiesterases and consequent rise of intracellular concentration of cGMP has been reported in aortic tissue [31]. Furthermore, many classes of ion channels are targets of gasotransmitters and, in particular, of H 2 S [32,33]. Indeed, the vasorelaxing effects of H2S are due – at least in part – to the activation of ATP-sensitive potassium (KATP ) channels [4,34], but other important vascular ion channels are likely to be influenced by H 2 S. The inhibition of L -type voltage-operated calcium channels is involved in the H2S-mediated vasodilator effects in rat cerebral arteries [35]; as well, H2S-mediated activation of 4-AP sensitive potassium channels has been observed in rat coronary arteries [36]. However, more recent evidence indicates that the activation of vascular Kv7 potassium channel and the hyperpolarization of vascular smooth muscle cells is a prevalent mechanism for the vasodilator effects of H2S [21]. Therefore, the possible activation of Kv7 channels by the two selected isothiocyanates was evaluated by means of pharmacological tools.

A. Martelli et al. / Vascular Pharmacology 60 (2014) 32–41

The vasorelaxing effects produced by PhNCS and PhNCS–COOH in rat aortic rings were strongly antagonized by the selective Kv7blocker XE991. Moreover, PhNCS and PhNCS–COOH, as well as the reference H2S-donor NaHS, caused significant concentration-related hyperpolarization of the membrane potential of HASMCs. Noteworthy, the Kv7-blocker XE991 significantly reduced the hyperpolarizing responses induced by NaHS and the tested isothiocyanates. The above evidence strongly suggests that the vasorelaxation induced by the two isothiocyanates is largely due to H2S-mediated activation of Kv7 channels and hyperpolarization of vascular smooth muscle cells. The vascular effects of the tested isothiocyanates were generally equivalent to or even higher than those produced by NaHS, although the amperometric measurements in vitro indicated that the isothiocyanates lead to L-Cysteine-dependent formation of lower (but more stable) concentrations of authentic H2S. Noteworthy, exogenous L-Cysteine was not added in the functional experiments on aorta, coronary bed and HASMCs. Therefore organic thiols endogenously present in the biological samples can effectively ensure the release of H2S from these isothiocyanates and, mainly, the prolonged presence of relatively low concentration of H2S seems to be highly effective in modulating the activity of blood vessels. Similar considerations emerged from previous comparisons between the cardiovascular effects produced by the slow H2S-donor GYY4137 and those evoked by the rapid H2S-donor NaHS [13]. In conclusion, this work first demonstrated that the isothiocyanate functional group can be viewed as an original and suitable slow H2Sreleasing moiety, endowed with vascular effects, typical of this gasotransmitter. Thus, such a chemical moiety can be employed for the synthesis of novel chemical tools for investigating the biological roles of H2S; moreover, it can be also viewed as a novel and versatile chemotype of H2S-donor, for the development of promising cardiovascular drugs. Role of the funding source This study was funded by “Regional Health Research Program 2009” of Regione Toscana, Italy. Regione Toscana had no role in the conduct of the study, in the collection, analysis and interpretation of data, in the writing of the report and in the decision to submit the paper for publication. Conflict of interest The authors declare no conflict of interest. Acknowledgements This article has been supported by the “Regional Health Research Program 2009” of Regione Toscana, Italy. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.vph.2013.11.003. References [1] Hosoki R, Matsuki N, Kimura H. The possible role of hydrogen sulfide as an endogenous smooth muscle relaxant in synergy with nitric oxide. Biochem Biophys Res Commun 1997;237:527–31. [2] Cheng Y, Ndisang JF, Tang G, Cao K, Wang R. Hydrogen sulfide-induced relaxation of resistance mesenteric artery beds of rats. Am J Physiol Heart Circ Physiol 2004;287: H2316–23. [3] Zhong G, Chen F, Cheng Y, Tang C, Du J. The role of hydrogen sulfide generation in the pathogenesis of hypertension in rats induced by inhibition of nitric oxide synthase. J Hypertens 2003;21:1879–85. [4] Zhao W, Zhang J, Lu Y, Wang R. The vasorelaxant effect of H2S as a novel endogenous gaseous KATP channel opener. EMBO J 2001;20:6008–16. [5] Yang G, Wu L, Jiang B, Yang W, Qi J, Cao K, et al. H2S as a physiologic vasorelaxant: hypertension in mice with deletion of cystathionin gamma-lyase. Science 2008;322:587–90.

41

[6] Brancaleone V, Roviezzo F, Vellecco V, De Gruttola L, Bucci M, Cirino G. Biosynthesis of H2S is impaired in non-obese diabetic (NOD) mice. Br J Pharmacol 2008;155:673–80. [7] Pan L, Liu X, Gong Q, Yang H, Zhu Y. Role of cystathionine γ-lyase/hydrogen sulfide pathway in cardiovascular disease: a novel therapeutic strategy? Antioxid Redox Signal 2012;17:106–18. [8] Martelli A, Testai L, Breschi MC, Blandizzi C, Virdis A, Taddei S, et al. Hydrogen sulfide: novel opportunity for drug discovery. Med Res Rev 2012;32:1093–130. [9] Li YF, Xiao CS, Hui RT. Calcium sulfide (CaS), a donor of hydrogen sulfide (H(2)S): a new antihypertensive drug? Med Hypotheses 2009;73:445–7. [10] Caliendo G, Cirino G, Santagada V, Wallace JL. Synthesis and biological effects of hydrogen sulfide (H2S): developement of H2S-releasing drugs as pharmaceuticals. J Med Chem 2010;53:6275–86. [11] Martelli A, Testai L, Marino A, Breschi MC, Da Settimo F, Calderone V. Hydrogen sulfide: biopharmacological roles in the cardiovascular system and pharmaceutical perspectives. Curr Med Chem 2012;19:3325–36. [12] Benavides GA, Squadrito GL, Mills RW, Patel HD, Isbell TS, Patel RP, et al. Hydrogen sulfide mediates the vasoactivity of garlic. Proc Natl Acad Sci U S A 2007;104:17977–82. [13] Li L, Whiteman M, Guan YY, Neo KL, Cheng Y, Lee SW, et al. Characterization of a novel, water-soluble hydrogen sulfide-releasing molecule (GYY4137): new insights into the biology of hydrogen sulphide. Circulation 2008;117:2351–60. [14] Zhao Y, Wang H, Xian M. L-Cysteine activated hydrogen sulfide (H2S) donors. J Am Chem Soc 2011;12:15–7. [15] Martelli A, Testai L, Citi V, Marino A, Pugliesi I, Barresi E, et al. Arylthioamides as H2S donors: L-cysteine-activated releasing properties and vascular effects in vitro and in vivo. ACS Med Chem Lett 2013;4:904–8. [16] Isenberg JS, Jia Y, Field L, Ridnour LA, Sparatore A, Del Soldato P, et al. Modulation of angiogenesis by dithiolethione-modified NSAIDs and valproic acid. Br J Pharmacol 2007;151:63–72. [17] Sparatore A, Perrino E, Tazzari V, Giustarini D, Rossi R, Rossoni G, et al. Pharmacological profile of a novel H2S-releasing aspirin. Free Radic Biol Med 2009;46:586–92 (Erratum in: Free Radic Biol Med 2009;47:1781). [18] Shukla N, Rossoni G, Hotston M, Sparatore A, Del Soldato P, Tazzari V, et al. Effect of hydrogen sulfide-donating sildenafil (ACS6) on erectile function and oxidative stress in rabbit isolated corpus cavernosum and in hypertensive rats. BJU Int 2009;103:1522–9. [19] Wallace JL, Cirino G, Santagada V, Caliendo G. Hydrogen sulfide derivatives of non-steroidal anti-inflammatory, drugs. 2008;WO2008/009127 A1. [20] Breschi MC, Calderone V, Digiacomo M, Manganaro M, Martelli A, Minutolo F, et al. Spirocyclic benzopyran-based derivatives as new anti-ischemic activators of mitochondrial ATP-sensitive potassium channel. J Med Chem 2008;51(21):6945–54. [21] Martelli A, Testai L, Breschi MC, Lawson K, McKay NG, Miceli F, et al. Vasorelaxation by hydrogen sulphide involves activation of K(v)7 potassium channels. Pharmacol Res 2013;70(1):27–34. [22] Martelli A, Manfroni G, Sabbatini P, Barreca ML, Testai L, Novelli M, et al. 1,4Benzothiazine ATP-sensitive potassium channel openers: modifications at the C-2 and C-6 positions. J Med Chem 2013;56(11):4718–28. [23] Gopalakrishnan M, Buckner SA, Shieh CC, Fey T, Fabiyi A, Whiteaker KL, et al. Br J Pharmacol 2004;143(1):81–90. [24] Yasui S, Mawatari K, Kawano T, Morizumi R, Hamamoto A, Furukawa H, et al. Insulin activates ATP-sensitive potassium channels via phosphatidylinositol 3-kinase in cultured vascular smooth muscle cells. J Vasc Res 2008;45(3):233–43. [25] Dorman DC, Moulin FJ, McManus BE, Mahle KC, James RA, Struve MF. Cytochrome oxidase inhibition induced by acute hydrogen sulfide inhalation: correlation with tissue sulfide concentration in the rat brain, liver, lung, and nasal epithelium. Toxicol Sci 2002;65:18–25. [26] Dombkowski RA, Russel MJ, Olson KR. Hydrogen sulfide as an endogenous regulator of vascular smooth muscle tone in trout. Am J Physiol Regul Integr Comp Physiol 2004;286:R678–85. [27] Baskar R, Li L, Moore PK. Hydrogen sulfide-induces DNA damage and changes in apoptotic gene expression in human lung fibroblast cells. FASEB J 2007;21:247–55. [28] Yang G, Sun X, Wang R. Hydrogen sulfide-induced apoptosis of human aorta smooth muscle cells via the activation of mitogen-activated protein kinases and caspase-3. FASEB J 2004;18:1782–4. [29] Edman P. Preparation of phenyl thiohydantoin from some natural amino acids. Acta Chem Scand 1950;4:277–82. [30] Li L, Rose P, Moore PK. Hydrogen sulfide and cell signaling. Annu Rev Pharmacol Toxicol 2011;51:169–87. [31] Bucci M, Papapetropoulos A, Vellecco V, Zhou Z, Pyriochou A, Roussos C, et al. Hydrogen sulfide is an endogenous inhibitor of phosphodiesterase activity. Arterioscler Thromb Vasc Biol 2010;30(10):1998–2004. [32] Tang G, Wu L, Wang R. Interaction of hydrogen sulfide with ion channels. Clin Exp Pharmacol Physiol 2010;37:753–63. [33] Peers C, Bauer CC, Boyle JP, Scragg JL, Dallas ML. Modulation of ion channels by hydrogen sulfide. Antioxid Redox Signal 2012;17:95–105. [34] Tang G, Wu L, Liang W, Wang R. Direct stimulation of KATP channels by exogenous and endogenous hydrogen sulphide in vascular smooth muscle cells. Mol Pharmacol 2005;68:1757–64. [35] Tian XY, Wong WT, Sayed N, Luo J, Tsang SY, Bian ZX, et al. NaHS relaxes rat cerebral artery in vitro via inhibition of l-type voltage-sensitive Ca2+ channel. Pharmacol Res 2012;65:239–46. [36] C h e a n g W S , W o n g W T , S h e n B , L a u C W , T i a n X Y , T s a n g S Y , e t a l . 4-aminopyridine sensitive K+ channels contributes to NaHS-induced membrane hyperpolarization and relaxation in the rat coronary artery. Vascul Pharmacol 2010;53:94–8.