Abstract

Selenium (Se), an essential trace element, and hydrogen sulfide (H2S), an endogenously produced signalling molecule, affect many physiological and pathological processes. However, the biological effects of their mutual interaction have not yet been investigated. Herein, we have studied the biological and antioxidant effects of the products of the H2S (Na2S)/selenite (Na2SeO3) interaction. As detected by the UV-VIS and EPR spectroscopy, the product(s) of the H2S-Na2SeO3 and H2S-SeCl4 interaction scavenged superoxide-derived radicals and reduced cPTIO radical depending on the molar ratio and the preincubation time of the applied interaction mixture. The results confirmed that the transient species are formed rapidly during the interaction and exhibit a noteworthy biological activity. In contrast to H2S or selenite acting on their own, the H2S/selenite mixture cleaved DNA in a bell-shaped manner. Interestingly, selenite protected DNA from the cleavage induced by the products of H2S/H2O2 interaction. The relaxation effect of H2S on isolated thoracic aorta was eliminated when the H2S/selenite mixture was applied. The mixture inhibited the H2S biphasic effect on rat systolic and pulse blood pressure. The results point to the antioxidant properties of products of the H2S/selenite interaction and their effect to react with DNA and influence cardiovascular homeostasis. The effects of the products may contribute to explain some of the biological effects of H2S and/or selenite, and they may imply that a suitable H2S/selenite supplement might have a beneficial effect in pathological conditions arisen, e.g., from oxidative stress.

1. Introduction

Exogenously added and endogenously produced H2S affects many physiological and pathological processes [13]. Accumulating evidence supports the involvement of H2S in the regulation of cardiovascular homeostasis [4]. It has mostly beneficial effects during oxidative stress by reacting with reactive oxygen and nitrogen species, i.e., hydrogen peroxide (H2O2), superoxide anion radical (O2⋅−), hypochlorite (HOCl), or peroxynitrite (ONOO) [58]. However, several effects of H2S on cells are toxic [911].

Selenium (Se) is a relatively rare but an essential trace element for humans, plants, and microorganisms. Se, which exerts multiple and complex effects on human health, is known as an antioxidant due to its presence in 25 selenoproteins in the form of selenocysteine amino acid. Both beneficial and detrimental effects of Se deficiency and/or supplementation are well known. The biological effects of Se compounds (selenite, selenate, selenocysteine, and selenomethionine) on cardiac oxidative damage, heart disease, cancer prevention, immunity, diabetes, neuroregeneration, or dementia have been reported [1217]. However, the beneficial effect of Se supplementation for men’s health is still a controversial issue [1823]. Selenite, which is a common Se supplement, is considered as a promising anticarcinogen [2426]. It can induce apoptosis in cancer cells through the production of reactive oxygen species (ROS) leading to oxidative stress [27, 28]. However, Se compounds were also found to damage DNA in healthy cells [29] and therefore may not be considered as a suitable protective agent against cancer and/or other chronic diseases. Actually, they can cause or advance some kinds of cancers [30, 31]. The exact mechanisms of the beneficial and toxic effects of Se are not yet fully understood, giving rise to further uncertainty about its potential use in nutrition supplements and/or clinical treatment.

Se and H2S are present in living organisms, and each one has beneficial and/or toxic effects through its interaction mostly with ROS [1, 2, 3033]. However, the biological effects of products of the H2S/selenite interaction are not yet known, namely, their involvement in the production and/or inhibition of ROS, reaction with DNA, or influence on cardiovascular system. Therefore, we have studied the effects of products of the H2S/selenite interaction on O2⋅− and cPTIO radicals, DNA cleavage, tension of isolated aortic rings, and rat blood pressure (BP). We found that the products have significant biological effects that differ from those caused by H2S or by selenite on their own. These results may contribute to the understanding of possible coupled biological effects of H2S and Se.

2. Material and Methods

2.1. Chemicals
2.1.1. Selenium Compounds

Stock solutions of sodium selenite (Na2SeO3, 10 or 40 mmol L−1, Merck K34993707-542 or Sigma 214485), selenium tetrachloride (SeCl4, 10 mmol L−1, Aldrich 323527), and sodium selenate (Na2SeO4, 10 mmol L−1, Sigma S0882) were prepared freshly in deionized H2O, stored at 23°C, and used within 5 h. Na2SeO3 dissociates in solution to yield mostly H2SeO3 at acidic pH, HSeO3 at neutral pH, and SeO32− at alkaline pH. For simplicity, the term SeO32− is employed as representative expression to encompass the total mixture of these different (de)protonation states.

2.1.2. Hydrogen Peroxide

Hydrogen peroxide (H2O2) (14.7 mol L−1; Sigma-Aldrich 85321), according to a particular experiment, was diluted in H2O or in 100 mmol L−1 sodium phosphate buffer, supplemented with 200 or 50 μmol L−1 DTPA, pH 7.4, 37°C before application.

2.1.3. Radicals

5-tert-butoxycarbonyl-5-methyl-1-pyrroline-N-oxide (BMPO, 100 mmol L−1, Dojindo B568-10, Japan) was dissolved in deionized H2O, stored at -80°C, and used after thawing. 2-(4-Carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO, 10 mmol L–1, Cayman 81540 or Sigma C221) was dissolved in deionized H2O and was stored at -20°C for several weeks.

2.1.4. Sulfide

Na2S (100 mmol L−1 stock solution, Dojindo SB01, Japan) was dissolved in argon degassed deionized H2O, stored at −80°C, and used immediately after thawing. Na2S dissociates in aqueous solution and reacts with H+ to yield H2S, HS, and a trace of S2−. We use the term H2S to describe the total mixture of H2S, HS, and S2− forms. The stock concentration was checked by UV-VIS spectroscopy: by the absorbance of 1000x diluted stock solution at 230 nm (, diluted by deionized water) and also by the reduction of 100 μmol L−1 DTNB by 2000x diluted stock solution (1 H2S molecule generates 2 TNB equivalents, , measured in 1 mmol L−1 phosphate buffer), according to Nagy et al. [34].

2.1.5. Buffers

100 mmol L−1 sodium phosphate buffer supplemented with 100 μmol L−1 diethylenetriaminepentaacetic acid (DTPA), pH 6.5, 7.0, 7.4, 8.0, and 9.0, 37°C, was employed for UV-VIS experiments. 50 and 25 mmol L−1 sodium phosphate buffer, supplemented with 100 and 50 μmol L−1 DTPA, pH 7.4, 37°C, was used for electron paramagnetic resonance (EPR) and plasmid DNA (pDNA) cleavage studies, respectively.

2.2. UV-VIS of cPTIO

To obtain 1 mL of the working solution, 10 or 100 μL of stock solution of the compounds studied was added to the appropriate volume (990 or 900 μL, respectively) of 100 mmol L−1 sodium phosphate buffer (at given pH, 37°C) containing the final concentrations of 100 μmol L−1 cPTIO and 100 μmol L−1 DTPA. UV-VIS absorption spectra (900-190 nm) were recorded every 30 s for 20 to 40 min with a Shimadzu 1800 (Kyoto, Japan) spectrometer at 37°C. The cPTIO extinction coefficient of 920 mol−1 L cm−1 at 560 nm was used. The reduction of the cPTIO radical was determined as the decrease of the absorbance at 560 nm (absorption maximum of cPTIO in VIS range) or at 358 nm after subtracting the absorbances at 730 or at 420 nm, respectively [5, 35].

To study the involvement of O2 in the H2S/SeO32−-induced reduction of the cPTIO radical, 10 mmol L−1 Na2S in H2O, 10 mmol L−1 Na2SeO3 in H2O, and 102 μmol L−1 ⋅cPTIO in the 100 mmol L–1 sodium phosphate buffer, supplemented with 100 μmol L−1 DTPA (pH 7.4; 37°C), were deaerated with argon for 10 min at 37°C. The compounds were mixed in a closed UV-cuvette, and the UV-VIS spectra were recorded. The O2 concentration in the deaerated samples was 3-5%, confirmed with an oxygen electrode (OXELP, SYS-ISO2, WPI, USA). In all UV-VIS experiments, H2O was used as a blank.

2.3. EPR of the BMPO Adducts

To study the ability of H2S/SeO32− to scavenge the O2⋅− radical or its derivatives produced in DMSO/KO2 solution, sample preparation and EPR measurements were conducted in accordance with previously reported protocols [5]. The solution (final concentrations) of BMPO (20 mmol L−1), DTPA (100 μmol L−1) in sodium phosphate buffer (50 mmol L−1, pH 7.4) was incubated for 1 min at 37°C; an aliquot of the compound was added, followed by saturated KO2/DMSO solution (10% DMSO/final buffer) 3 s later. The sample was mixed for 5 s and transferred to a standard cavity aqueous EPR flat cell. The first EPR spectrum was recorded 2 min after the addition of KO2/DMSO solution at 37°C. The sets of individual EPR spectra of the BMPO spin adducts were recorded as 15 sequential scans, each 42 s, with a total time of 11 min. Each experiment was repeated at least twice. EPR spectra of the BMPO spin adducts were measured on a Bruker EMX spectrometer, X-band ~9.4 GHz, 335.15 mT central field, 8 mT scan range, 20 mW microwave power, 0.1 or 0.15 mT modulation amplitude, 42 s sweep time, 20.48 ms time constant, and 20.48 ms conversion time at 37°C. Intensities of the BMPO adducts in the EPR spectra were reproducible, when the KO2/DMSO stock solution was stored at 5°C for 1 day or at for 4 h.

2.4. Plasmid DNA Cleavage

pDNA cleavage assay with the use of pBR322 plasmid (New England BioLabs Inc., N3033L) was performed as reported previously [5, 36]. In this assay, all samples contained 0.2 μg pDNA in sodium phosphate buffer (25 mmol L−1 sodium phosphate, 50 μmol L−1 DTPA, pH 7.4, 37°C). After addition of compounds, the resulting mixtures were incubated for 30 min at 37°C. All concentrations listed in the section were final in the samples. After incubation, the reaction mixtures were subjected to 0.6% agarose gel electrophoresis. Samples were electrophoresed in TBE buffer (89 mmol L−1 Tris, 89 mmol L−1 boric acid, and 2 mmol L−1 EDTA) at 5.5 V cm−1 for 2 h; gels were stained with GelRed™ Nucleic Acid Gel Stain and photographed using a UV transilluminator. Integrated densities of all pBR322 forms in each lane were quantified using the TotalLab TL100 image analysis software to estimate pDNA cleavage efficiency (Nonlinear Dynamic Ltd., USA).

2.5. Guide for the Use and Care of Laboratory Animals
2.5.1. Isolated Thoracic Aorta

Procedures were performed in accordance with the Institutional Guidelines of the Ethical Committee on the Ethics of Procedures in Animal, Clinical and other Biomedical Experiments (permit number: EC/CEM/2017/4) of the Institute of Normal and Pathological Physiology, Centre of Experimental Medicine and were approved by the State Veterinary and Food Administration of the Slovak Republic and by an Ethical Committee according to the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes, Directive 2010/63/EU of the European Parliament. The Institute of Normal and Pathological Physiology provided veterinary care.

2.5.2. Rat Blood Pressure

All procedures were approved by the State Veterinary and Food Administration of the Slovak Republic (No.: Ro-1545/15-221) according to the guidelines from Directive 2010/63/EU of the European Parliament. Experiments were carried out according to the guidelines laid down by the animal welfare committee of the Institute of Normal and Pathological Physiology of the Slovak Academy of Sciences and conformed to the principles and regulations, as described in the editorial by Grundy [37].

2.6. Functional Study of Isolated Thoracic Aorta

Normotensive Wistar Kyoto (WKY) rats () were killed by decapitation after a brief anesthetization with CO2, and the thoracic aorta was isolated as described in our previous study [38]. The changes in isometric tension were measured by the electromechanical transducers (FSG-01, MDE, Budapest, Hungary). The resting tension of 1 g was applied to each ring and maintained throughout a 45 to 60 min of equilibration period until stress relaxation no longer occurred. Changes in thoracic aorta tension were followed by noradrenaline (NA; 1 μmol L−1) precontracted arterial rings after a stable plateau was achieved.

2.7. Functional Study of Rat Blood Pressure

Male Wistar rats (; ) were from the Department of Toxicology and Laboratory Animal Breeding at Dobra Voda, Slovak Academy of Sciences, Slovakia. The rats were housed under a 12 h light-12 h dark cycle, at a constant humidity (45-65%) and temperature (20-22°C), with free access to standard laboratory rat chow and drinking water. The Institute of Experimental Pharmacology and Toxicology, Centre of Experimental Medicine, Slovak Academy of Sciences, provided veterinary care. The tranqualizer xylazine (Rometar) was purchased from Zentiva (Czech Republic), and the anesthetic combination of tiletamine+zolazepam (Zoletil 100) was acquired from Virbac (France). All other chemicals were purchased from Sigma-Aldrich. Experiments were carried out as previously described [39]. Rats were anesthetized with Zoletil 100 (tiletamine+zolazepam, 80 mg kg-1, i.p.) and Rometar (xylazine, 5 mg kg-1, i.p.). During the anesthesia, BP, heart rate, and reflex responses to mechanical stimuli were monitored. The animals were under anesthesia during the whole experiment and were euthanized with an overdose of Zoletil via jugular vein at the end of the surgical procedure. All experiments were supervised and performed under the same experimental conditions.

2.8. Blood Pressure Measurement

The right jugular vein was cannulated to administer compounds under anesthesia as described above. The left arteria carotis communis was cannulated for inserting the fiber optic microcatheter pressure transducers (FISO LS 2F Harvard Apparatus, USA). The analog signal was digitalized at 10 kHz, filtered at 1 kHz, and recorded by DEWESoft 6.6.7 (GmbH, Austria). The signal was evaluated 5 s before and 10 min after compound administration. After stabilization of BP (10-20 min), the compounds were administered into the right jugular vein as a bolus of 500 μL kg−1 over 15 s period. The solution of the H2S/SeO32− mixture (10/5 in mmol L−1) was prepared as follows: to 123.5 μL of 100 mmol L−1 phosphate buffer, 100 μmol L−1 DTPA, 14 μL of 1 mol L−1 HCl was added, followed by 62.5 μL of 40 mmol L−1 Na2SeO3 in 0.9% NaCl, and finally 50 μL of 100 mmol L−1 Na2S in H2O was added. The pH of the buffered mixture was 7.4. The mixture was incubated for at 23°C before i.v. administration. Unbuffered H2S/SeO32− mixture (10/5 in mmol L−1) was prepared, when 0.9% NaCl was used instead of phosphate buffer and HCl. The pH of the unbuffered mixture was ~11 measured by a pH paper indicator.

2.9. Statistical Analysis

Unless otherwise stated, data are represented as the . Statistical significance was determined by Student’s -test or one-way ANOVA followed by the multiple comparison test. Differences between means were considered significant at . Data analysis and plot construction were carried out using SigmaPlot 12 (Systat Software GmbH).

3. Results

3.1. H2S Interacts with Na2SeO3 and SeCl4, but Not with Na2SeO4, to Form Initial Reactive Intermediate(s), which Reduce the cPTIO Radical

Since the antioxidant properties of H2S/SeO32− products are unknown, we have used the cPTIO radical to study the reducing properties of the products of H2S/SeO32− interaction. The cPTIO radical is stable in aqueous solution, and its formation and reduction can be monitored by the UV-VIS spectrophotometry at 358 or 560 nm. Even in the presence of up to 100 μmol L−1 H2S or 100-400 μmol L−1 SeO32−, the absorbance (ABS) of the radical at 358 and 560 nm decreases only by <7% after 40 min, indicating that neither H2S nor SeO32− on its own reduces this radical (Figure 1). In contrast, once H2S (25-100 μmol L−1) was added to the cPTIO/SeO32− (100/2.5-400 μmol L−1) mixture, or SeO32− was added to cPTIO/H2S, the absorbances at 358 and 560 nm decreased rapidly over the time (≤30 s), indicating a possible formation of strong reducing agent(s) which fastly and efficiently reduced the cPTIO radical (Figures 1 and 2). Similar results were obtained when SeCl4, but not Na2SeO4, was used instead of SeO32− (Figures S1 and S2a).

The reduction of cPTIO followed a bell-shaped dependence on the concentration of SeO32− at a constant cPTIO/H2S concentration (Figure 2(a)), with a maximum radical scavenging activity at an H2S : SeO32− ratio of roughly 4 : 1. The ability of H2S to reduce cPTIO in the presence of SeO32− increased with the increasing H2S concentration (Figure 2(b)) and followed also a bell-shaped dependence on pH (Figures 3 and S3).

If H2S and SeO32− were preincubated for different periods of time before the addition to cPTIO, it clearly resulted in the highest radical scavenging activity, which was subsequently lost over the time (Figure 4). An H2S/SeO32− (100/100 in μmol L−1) mixture preincubated for ≥1 min prior to cPTIO addition did not reduce cPTIO, demonstrating that later products of the reaction of sulfide with SeO32− could not be responsible for the reduction of the radical and that the relevant active species were formed swiftly, in less than 1 min and have a short lifetime, as recently suggested [40]. Notably, formation of these active early intermediates to reduce cPTIO was prolonged with the increase of the H2S/SeO32− ratio (Figure 4(e)). At H2S/SeO32− concentration of 100/25 in μmol L−1 and 5 min preincubation time, the mixture still possessed around 50% potency to reduce cPTIO (Figure 4(e)). This timing, once more, accounts for a rapidly formed selenosulfide intermediate as beeing the ultimate responsible for this radical scavenging activity, a species also possibly being sensitive to oxidation over prolonged time periods (Figure 4).

While the available evidence is in accordance with the formation of HSSeSH as the main reactive species, there are other chalcogen-based candidates which are good reducing agents, namely, hydrogen selenide (H2Se), hydroselenide anion (HSe), selenide (Se2−), persulfides, and polysulfides (Sx2–) [4143]. Interestingly, O2 does not seem to play a major role in the H2S/SeO32−-induced reduction of the cPTIO radical probably due to slower kinetics of interaction of reactants and/or intermediates with O2 in comparison to the rate of cPTIO reduction [4448]. Under argon flushed conditions, reduction of the cPTIO radical was neither enhanced nor suppressed significantly, hence ruling out any major involvement of H2Se, as this selenium compound is highly sensitive to O2 (Figure S2b).

3.2. EPR of BMPO-OOH: The Initial Products of the H2S/SeO32− Interaction Also Scavenge Derivatives of Superoxide Anion (O2⋅–) Radical

We aimed to ascertain whether the initial products of the H2S/SeO32− interaction are able to scavenge other radicals, i.e., O2⋅− or its derivatives. The interactions with O2⋅− were studied with the EPR spin trap method based on the reaction of this dioxygen radical with BMPO to form the BMPO-OOH adduct [49]. This assay was chosen due to biological and mechanistical reason; O2⋅− is a simple radical reduced by one-electron transfer.

O2⋅− was dissolved in phosphate buffer (pH 7.4, 37°C) and trapped by BMPO. Under these conditions, the relative intensity of the BMPO-OOH adduct decreased slowly over the time and was comparable to the values reported under physiological conditions (Figures 5(a1)–5(a3)) [49]. The addition of SeO32− (25 and 50 μmol L–1) did not significantly interfere with the BMPO-OOH adduct formation, its concentration, and rate of decay (Figures 5(b1)–5(b3), S4b1-b3). In contrast, H2S (25 and 50 μmol L−1) decreased the BMPO-OOH concentration and increased the rate of the decay, and this “antioxidant” activity of H2S was increased further in the presence of SeO32− (more than three times at a 50/25 in μmol L−1 H2S/SeO32− ratio). This indicates that H2S/SeO32− products scavenge O2⋅− and/or its derivatives. The effects of H2S/SeO32− were less pronounced when KO2/DMSO was added 5 min after addition of H2S/SeO32−, suggesting that the reducing intermediate had already decomposed at this stage (Figure 5, S4). Similar results were obtained when H2S was incubated with SeCl4 (Figure S5). This indicates that the production of highly active species, products of the H2S/SeO32− interaction, was time-dependent: they appeared within a few seconds after addition of SeO32− to the H2S solution and their effects diminished after few minutes of interaction.

From our previous studies of O2⋅− reaction with BMPO [5], we assumed that the EPR spectra of the BMPO adducts were superposed on the BMPO-OOH/OH radicals with minor contribution from the BMPO-C radical. Therefore, we simulated the spectra using hyperfine coupling constants for BMPO-OOH, BMPO-OH, and BMPO-C (derived from DMSO) radicals. The means of hyperfine coupling constants used were as follows: BMPO-OOH1 (black) , ; BMPO-OOH2 (red) , ; BMPO-OH1 (green) , ; BMPO-OH2 (yellow) , ; and BMPO-C (blue) , . The constants are similar to those reported by Zhao et al. [49]. The simulation revealed that O2⋅− was trapped in the control and in the presence of SeO32− (Figures 5(a) and 5(b)), since the EPR spectra of BMPO-OOH were only observed. However, BMPO-OH component was present in the samples containing H2S alone or with SeO32− (Figures 5(c)–5(h)). The results may indicate that H2S alone or with SeO32− decomposed BMPO-OOH to BMPO-OH and/or OH was formed as a result of compound presence. Since the first spectrum was recorded after sample preparation, we cannot exclude a possibility of trapping of other radicals with lifetimes shorter than 110 s.

3.3. H2S/SeO32− Cleaves pDNA

It was of interest to know if the products of the H2S/SeO32− interaction have biological effects in vitro. Therefore, we investigated the direct effects of H2S and SeO32− on the cleavage of pDNA in vitro using the Fenton reaction as a positive benchmark control (Figure 6) [5].

Interestingly, neither H2S (1 mmol L−1) nor SeO32− (0-1 mmol L−1) alone significantly cleaved pDNA. In contrast, SeO32− in a concentration-dependent manner caused damage to DNA in the presence of 1 mmol L−1 H2S (Figures 6(a1) and 6(a2)). The observed cleavage of DNA caused by SeO32− and H2S showed a bell-shaped concentration ratio dependence similar to the one observed in the reduction of the cPTIO radical. We can suggest that the intermediate responsibility of the action can be a selenopolysulfide with a H2S : SeO32− ratio of 4 : 1. In the presence of H2O2, damage to pDNA by H2S occurs also without SeO32−, since H2O2 now takes on the role of SeO32− as an oxidant, with the simultaneous formation of the OH radicals (Figures 6(b1) and 6(b2)). Further evidence for the involvement of radicals may come from the fact that dimethylsulfoxide (DMSO), a known OH scavenger [50, 51] frequently employed as solvent in biology, is able to interfere with the damage to DNA (Figures 6(c1) and 6(c2)).

3.4. H2S/SeO32− Modulates Tension of Isolated Thoracic Aorta

As some of our in vitro assays indicated an antioxidant activity of the H2S/SeO32− mixture, and H2S is known to promote relaxation of blood vessels [52], the impact of the H2S/SeO32− mixture on the isolated thoracic aorta was examined. The thoracic aorta was precontracted by noradrenaline (NA) (1 μmol L−1). After a stabile plateau of the contraction was reached (Figure 7(a)), SeO32− (100 μmol L–1) showed only negligible activity, while Na2S (200 μmol L−1) significantly relaxed the aortic rings, in agreement with our previous studies [38]. A simultaneous addition of SeO32− (100 μmol L−1) and H2S (200 μmol L−1) resulted once more in a biphasic activity profile, where a minor relaxation was noticed first, followed by a pronounced contraction (Figure 7). It is supposed that the contraction effect may result also from the antioxidant properties of the mixture, similarly as it has been reported for ascorbate [53].

3.5. H2S/SeO32− Modulates Rat Systolic and Pulse Blood Pressure

Since the products of the H2S/SeO32− interaction modulated the tension of the thoracic aorta, we subsequently studied whether the products influence blood pressure (BP). Intravenous (i.v.) administration of 5 μmol kg−1 SeO32− had only minor effects on BP (Figure 8(g)). The administration of 10 μmol kg−1 of Na2S transiently decreased and increased BP (Figures 8(a) and 8(g)), as observed in our previous study [54]. The stock solution of the H2S/SeO32− mixture (20/10 in mmol L−1) prepared in 0.9% NaCl was colorless and had pH ~ 11. However, when the mixture was prepared in solution with pH 7.4, it had orange color with an absorption maximum at 570 nm (Figure S6), indicating formation of the sulfur-selenium complexes [40]. The i.v. administration of the mixture H2S/SeO32− (10/5 μmol kg−1, pH ~ 7.4), in comparison to H2S alone, inhibited both BP decrease and increase (Figures 8(b) and 8(g)). The effects of the mixture were less pronounced at pH ~ 11, being the effects at this pH similar to those observed for H2S alone (Figures 8(c) and 8(g)).

The studied compounds influenced pulse BP, as an important parameter of cardiovascular system reflecting arterial stiffness [55, 56]. The administration of 5 μmol kg−1 SeO32− intravenously had minor effects on pulse BP (Figure 8(h)). The administration of 10 μmol kg−1 of Na2S transiently increased and later decreased pulse BP (Figures 8(d) and 8(h)) [54]. The i.v. administration of the mixture H2S/SeO32− (10/5 in μmol kg−1, pH ~7.4), with comparison to H2S alone, eliminated pulse BP increase, but did not affect pulse BP decrease (Figures 8(e) and 8(h)). The effects of the mixture were less pronounced at pH ~ 11 and were similar to those observed for H2S alone (Figures 8(f) and 8(h)). The inefficiency of the H2S/SeO32− products at pH ~ 11 may be connected with the minor effect of the mixture on the cPTIO radical reduction at high pH (Figures 3(e) and 3(f)). This may imply that the reduction properties of the mixture could be responsible for their effects on systolic and pulse BP. The results confirm that the products of the H2S/SeO32− interaction depend on pH and influence differently the cardiovascular system. In conclusion, the reactivity and biological activity of the H2S/SeO32− interaction products prepared at pH 7.4 differ from those of Na2S alone.

4. Discussion and Conclusions

Overall, our studies demonstrate that the two suspected and commonly used “antioxidants,” H2S and SeO32−, are not necessarily typical reducing agents, such as ascorbic acid or tocopherol, when employed on their own. Interestingly, these two chalcogen agents, when added together, rapidly activate each other and form a cascade of considerably more reactive, often reducing species, supposing the involvement of inorganic HSSeSH and polysulfides Sx2−, which may account for some of the observed biological actions. The nature of some of these intermediate reactive sulfur and/or selenium species was suggested in a recently published review [40].

The fast and efficient reduction of the cPTIO radical by H2S/SeO32− products (Figures 1 and 4) supports the notion that the initial intermediate(s) formed by the reaction of H2S with SeO32− are responsible for this kind of action. The bell shape and the maximum radical scavenging activity of H2S : SeO32− at a ratio of ~4 : 1 (Figure 2) may indicate the suggested formation of (HSS)2Se [40]. The kinetics and efficiency of the H2S/SeO32− products to reduce cPTIO (Figure 3) point out to complex pH-dependent chemical and radical reactions of the species.

Reactions of H2S or polysulfides with SeO32− and/or SeCl4 were reported [40, 5759]. They point towards a rapid conversion of SeO32− and SeCl4 to an intermediate, probably HSSeSH, and a subsequent, slower reductive elimination of this intermediate to elemental (mixed) chalcogen particles and disulfides [40]. However, to our knowledge, there is no information about the formation and detection of HSSeSH in cells or its cytoprotective or other biological effect. The first synthesis of H-S-Se-S-H (1.3-dithiatriselane) was reported by Hahn and Klünsch in 1994, but the stability and reactivity in aqueous solution were not investigated. Solid HSSeSH has a melting point at −40°C. It was one component of a mixture of H2S2Sen, prepared by the interaction of 2 H2S with Se2Cl2 [60].

The EPR data (Figure 5) once more confirm that SeO32− on its own is not an antioxidant; it becomes activated by reduction, for instance, by H2S, which concurrently is activated by oxidation. The mutual redox activation is fast, and, as in the case of the cPTIO radical scavenging, the pristine mixture of H2S and SeO32− is most active, with a decrease of activity over the time, pointing once more at simple H2Sx or H2SxSe, and notably HSSeSH, but not an SxSey, as being responsible for this kind of activity.

We found that the products of this described H2S/SeO32− interaction have several noteworthy biological effects, involving ROS scavenging, modulation of the redox state, reaction with DNA, tensing isolated aorta, and influencing BP and pulse BP (Figures 68). These effects obviously need to be investigated further and in considerably more detail and were not present or were less pronounced when H2S or selenite acted alone. The properties of the products of the H2S/SeO32− interactions significantly depended on the H2S/SeO32− molar ratio, pH, and preincubation time. The combination of these variables makes the work with H2S/SeO32− very complicated, and these facts should be taken into account at the time of designing in vitro and in vivo experiments. These properties of the products may explain the previously published beneficial and contrasting toxic Se effects, for example, in conditions of oxidative stress and cancer [1214, 18, 20, 22, 23, 30].

H2S is endogenously produced in vivo in most, if not in all, cells, and H2S donors are commonly used in biological experiments, and they are considered to be applied in medicine. Our results suggest that in biological experiments with selenite, in its nutrition supplement and clinical use, effects of the H2S/SeO32− interaction should be considered. While SeO32− is used widely as a nutritional supplement already, one may, in the future, wish to spice it up with some reduced sulfur. Natural spices such as garlic and onions contain suitable sulfide releasing agents, such as diallyltrisulfide (DATS) and diallyltetrasulfide (DATTS), which both occur naturally in garlic, or dipropyltrisulfide and dipropyltetrasulfide, both present in onions [61]. Our results imply that application research of suitable H2S/SeO32− supplements may lead to the beneficial effects in pathological conditions arising, e.g., from ROS overproduction.

Abbreviations

ABS:Absorbance
BMPO:5-tert-butoxycarbonyl-5-methyl-1-pyrroline-N-oxide
BP:Blood pressure
cPTIO:2-(4-Carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide
DTPA:Diethylenetriaminepentaacetic acid
EPR:Electron paramagnetic resonance
H2S:Hydrogen sulfide
H2O2:Hydrogen peroxide
i.p.:Intraperitoneal
i.v.:Intravenous
NA:Noradrenaline
pDNA:Plasmid DNA
SeCl4:Selenium tetrachloride
SeO32−:Selenite
UV/VIS:Ultraviolet-visible.

Data Availability

All findings and conclusions are based on the presented figures in the main text or in the supplementary information. Original source files (UV-VIS spectra, EPR spectra, DNA gels, rat blood pressure records, and aorta relaxation records) can be sent from the corresponding author, Dr. Karol Ondrias, upon request.

Conflicts of Interest

The authors declare no conflict of interest.

Authors’ Contributions

K.O. and M.C. conceived, initiated, and coordinated the study; K.O., M.C, and V.B. designed research; K.O., M.G., A.M., and A.K. performed UV-VIS experiments and analyzed data; V.B., K.O., M.G., and A.M. performed EPR experiments and analyzed data; M.C. and A.M. performed pDNA cleavage experiments and analyzed data; L.T., A.M, K.O., S.C., A.B., and P.B. performed aorta experiments and analyzed data; L.K., A.M., S.C., A.B., and K.O. performed rat in vivo experiments. K.O. wrote the paper; K.O., M.C., V.B., A.K., and E.A-D contributed to analyze data and manuscript writing. Marian Grman and Anton Misak contributed equally to this work. Miroslav Chovanec and Karol Ondrias shared senior authorship.

Acknowledgments

This research was funded by the Slovak Research and Development Agency (grant numbers APVV-15-0371 to A.M., M.G., and K.O.; APVV-17-0384 to M.C.; and APVV-15-0565 to A.M, S.C., A.B., and P.B.), the VEGA Grant Agency of the Slovak Republic (grant numbers 2/0079/19 to M.G.; 1/0026/18 to V.B.; 2/0053/19 to M.C.; and 2/0014/17 to K.O.), and University Science Park for Biomedicine (ITMS 26240220087).

Supplementary Materials

The supplementary information’s file (PDF) contains only additional figures with legends to the main manuscript text. (Supplementary Materials)