The Effects of Novel Triazolopyrimidine Derivatives on H₂S Production in Lung and Vascular Tonus in Aorta

Hydrogen sulfide (H₂S), which is considered the third gasotransmitter, is synthesized endogenously from l-cysteine (l-cys) by cystathionine γ-lyase (CSE), cystathionine β-synthase, and 3-mercaptopyruvate sulfur transferase enzymes in lungs [1]. As a signaling mediator, H₂S is involved in physiological functions in the respiratory tract such as the regulation of proliferation, vascular tone, and pulmonary circulation [2, 3]. It has been reported that the synthesis of H₂S is decreased in acute and chronic inflammatory lung diseases such as asthma, pulmonary hypertension (PAH), and chronic obstructive pulmonary disease [4‒6]. Decreased H₂S levels have been associated with pathological processes such as inflammation, oxidative stress, and fibrosis underlying inflammatory chronic respiratory diseases and acute lung injury [2, 5, 7]. Furthermore, H₂S has antioxidative, antifibrotic, and anti-inflammatory properties; it may be beneficial to restore decreased H₂S levels in inflammatory respiratory system diseases [7‒9].

Due to the therapeutic potential of H₂S in these diseases, donors/hybrid molecules that increase endogenously or release H₂S have been widely investigated in recent years [8, 10, 11]. However, H₂S can inhibit the respiratory chain at high concentrations [12], so increasing the endogenous synthesis of H₂S may be a much safer and more useful method rather than H₂S donors. Therefore, drugs/substances that increase endogenous H₂S formation in the lungs are a potential target in the treatment of respiratory inflammatory diseases [2, 3].

Resveratrol (RVT, 3,4′,5-trihydroxy-trans-stilbene) is a natural polyphenolic compound with anti-inflammatory, antiapoptotic, antioxidant, antifibrotic effects similar to like H₂S [13‒15]. In our previous studies, we have revealed for the first time that RVT increases the endogenous synthesis of H₂S in the aorta and penile tissue and that the vasodilator effects of RVT are mediated by H₂S [16, 17]. However, the effect of RVT on H₂S formation in the lungs was not reported to our knowledge.

In order to discover a new endogenous H₂S synthesis activator, we designed and synthesized an RVT analog with a novel scaffold, Cpd2, based on the in-house chalcone compound (Cpd1) as shown in Figures 1 and 2. In our study, we tested the capacity of both RVT and newly synthesized molecules to increase endogenous H₂S synthesis in mouse lungs. Also, we investigated whether the newly synthesized molecules have relaxant effects on mouse aorta.

Sodium chloride (NaCl), potassium chloride (KCl), magnesium chloride (MgCl₂), potassium dihydrogen phosphate (KH₂PO₄), calcium chloride (CaCl₂), glucose, l-cys, sodium sulfide nonahydrate (Na₂S) and aminooxyacetic acid (AOAA), phenylephrine hydrochloride (Phe), acetylcholine chloride (ACh) was purchased from Sigma-Aldrich Inc., St. Louis, Leupeptin and Aprotinin from Cayman, Pepstatin from Bioshop. BCA Protein Quantitation Kit (K812) and RVT were obtained from BioVision Inc. H₂S biosensor purchased from Unisense (Model H₂S-500LR, Aarhus, Denmark). l-cys, Na₂S, and AOAA were dissolved in distilled water; RVT was dissolved in ethanol. All the reagents and solvents used in synthesis and purification studies were commercially available materials and purchased from Alfa Aesar, Sigma‐Aldrich, and Merck.

In the first step, Cpd1 was obtained from the reaction of 3,4-dimethoxybenzaldehyde and acetophenone in basic conditions (Claisen-Schmidt condensation) (shown in Fig. 2). The applied cyclization method of Cpd1 was modified from the literature [18]. By the heterocyclization of chalcone and aminotriazole, we restricted the cycling of phenyl rings and increased the number of heteroatoms increasing for possible interactions (shown in Fig. 2). At the same time, the distance between the centers of the two phenyl rings in Cpd2 and trans-RVT is similar.

Cpd1 (1 mmol) and 3-aminotriazole (2 mmol) were heated in dimethylformamide (DMF) with MW in basic conditions (Cs₂CO₃ or TEA; 1 mmol) for 2 h at 140°C. After removal of the solvent at low pressure, the residue was purified by column chromatography with 3:2 n-hexane: ethylacetate. Cpd2 was obtained from methanol crystallization with a 22% yield as yellow crystals (shown in Fig. 2).

IR spectra of compounds were recorded on the PerkinElmer 100 FT-IR Spectrometer with ATR attachment. Mass spectra (ESI-MS) were measured on a Thermo MSQ Plus LC/MS. 1H NMR spectra were recorded on a Varian Mercury Plus-AS400 400-MHz for protons without internal standard.

(E)-3-(3,4-dimethoxyphenyl)-1-phenylprop-2-en-1-one (Cpd1)

IR n_max (cm^-1) 1,584, 1,573, MS (ESI) m/z (intensity %): 105.14 (100), 191.20 (70), 269.22 (95) [M + H]⁺¹H NMR (DMSO-d6) d 3.66 (6H, s, OCH₃), 4.20 (2H, s, CH₂), 6.87 (1H, t, J = 9.0 Hz, Ar-H), 6.96 (1H, d, J = 8.0 Hz, Ar-H), 7.01 (1H, t, J = 7.3 Hz, Ar-H), 7.13–7.15 (2H, m, Ar-H), 7.47 (2H, bs, Ar-H), 8.07 (1H, d, J = 7.9 Hz, Ar-H), 9,92 (1H, s, NH), 12,70 (1H, s, NH) ppm.

7-(3,4-Dimethoxyphenyl)-5-phenyl-[1,2,4] triazolo[1,5-a] pyrimidine (Cpd2)

IR n_max (cm^-1) 2,357, 1,133 ¹H NMR (400 MHz, DMSO-d6) δ 8.71 (s, 1H), 8.41–8.36 (m, 2H), 8.12 (s, 1H), 8.03 (dd, J = 8.5, 2.2 Hz, 1H), 7.92 (d, J = 2.2 Hz, 1H), 7.60 (ddd, J = 5.0, 2.8, 1.6 Hz, 3H), 7.22 (d, J = 8.7 Hz, 1H), 3.90 (d, J = 1.3 Hz, 6H). ¹³C NMR (100 MHz, DMSO-d6) δ 160.5, 156.0, 151.7, 148.4, 147.3, 136.3, 131.2, 128.9, 127.9, 123.7, 121.8, 113.2, 111.4, 105.9, 55.9, 55.8, 40.2, 40.0, 39.9, 39.8, 39.7, 39.5, 39.3, 39.1, 38.9. HRMS: calcd. for C19H16N4O2, 332.1273, found, 332.1261.

The present study was approved by the Animal Experiment Local Ethical Committee of Ege University (No: 2019-009) in agreement with European guidelines for animal care. All experiments were conducted with 2-month-old, 25–30 g Swiss albino male mice obtained from Ege University Center for Research on Laboratory Animals (EGEHAYMER) (n = 60). All animals were housed at 20–25°C with 50–60% humidity under a 12 h light-dark cycle with free access to food and water. Animals were sacrificed by cervical dislocation under the ketamine/xylazine (60/5 mg/kg) anesthetization.

The lungs were isolated from mice and approximately 25–30 mg of tissue were homogenized by cryogenic grinding (Retsch, Cryomill) in phosphate-buffered saline (PBS; pH 7.4) including proteases and phosphatase inhibitors. Measurements of total protein concentrations in homogenates were detected by the bicinchoninic acid assay (BCA Kit, BioVision). H₂S measurements were performed in homogenates containing equal amounts of protein (25 μg) in PBS [16].

Time-dependent measurement of endogenous H₂S formation was performed by the Unisense H₂S biosensor (Unisense, Model H₂S-500LR) in conjunction with the Unisense PA2000 amplifier [19]. To stabilize the signals at the initial stage of each experiment, the H₂S biosensor was polarized. All the measurements started after the sensor signal was stabilized. First, we measured the signals of compounds in PBS-containing tubes (pH 7.4, 37°C) to confirm if they interact with the assay. After confirmation of the absence of interaction, we measured the effect of compounds on basal or L cysteine-induced H₂S formation in lung homogenates, without or with the substrate of H₂S synthesis enzymes, respectively.

To measure basal H₂S formation in the lung, homogenates including 25 μg proteins were added into the tube and signals were recorded for 10 min. Then either RVT (100 μm) or one of the new compounds Cpd1 (100 μm), Cpd2 (100 μm) or vehicle was added into the tube and signal was recorded for a further 30 min. To measure l-cys-induced H₂S formation, l-cys (10 mm) and cofactor pyridoxal phosphate (2 mm) were added, and signals were recorded for 40 min. In another set of experiments to test whether the possible stimulatory effect of chemicals is inhibited by an H₂S synthesis inhibitor, AOAA (10 mm) was incubated for 10 min before the addition of compounds as well as l-cys and cofactor, and signals were recorded again for further 40 min.

H₂S production rates were determined by initial steepest slopes dividing concentration to time (µM/min) at the each trace. The H₂S sensor was calibrated before each experiment as well as calculation of maximal H₂S concentration that has been formed through a standard calibration curve which was obtained by measuring signals (millivolt) against cumulative concentrations of Na₂S (5–80 μm) as H₂S donor.

After the aorta was isolated from mice, taken into the Krebs solution (NaCl, 130 mm; NaHCO₃, 14.9 mm; glucose, 5.5 mm; KCl, 4.7 mm; KH₂PO₄, 1.18 mm; MgSO₄7H₂O, 1.17 mm and CaCl₂·2H₂O, 1.6 mm) and washed. The surrounding connective tissue was removed and divided into 2–3 mm-long rings. The individual aortic rings were separated, mounted in a 5 mL organ bath of DMT Ring Myograph for isometric force recording (Danish Myograph Technology, Aarhus, Denmark) coupled to a Power Lab 7 data acquisition system (Chart 7.0 software; ADInstruments, Colorado Springs, CO) and bathed in carboxygenated (95% O₂; 5% CO₂) Krebs solution at 37°C [16]. The tissues were allowed to equilibrate for 45 min under a resting tension of 20 mN. All the relaxation responses were obtained after precontraction to phenylephrine (Phe, 1 μm). Experiments were performed in aortic rings with endothelium as confirmed by relaxation of more than 40% to ACh (1 μm). Cpd1 or Cpd2 was dissolved in ethanol and distilled water in a 1:1 ratio and responses to cumulative concentration of Cpd1 or Cpd2 (10⁻⁸–10⁻⁴ M) were obtained. The concentrations were given as final molar concentrations in the bath solution. The solvent ratio used in the experiment did not exceed 0.01% in the bath.

Data are expressed as % relaxation of the Phe-induced tone. All calculations as well as graphical presentations were done by using the statistical software GraphPad Prism 8 (San Diego, CA, USA). Significance was accepted at p < 0.05. The data were computed as means ± standard error of the mean (SEM), and statistical analysis was performed by using two way ANOVA test or unpaired t test which is appropriate. If there is an interaction between concentrations and treatments, Bonferroni post hoc test multiple comparison was used after ANOVA. “n” refers to the number of animals.

The compounds or solvents had no effects on H₂S signals in PBS at the concentrations applied as expected since there is no S-atom in the chemical structures of the molecules (data not shown). Furthermore, compounds have no effects on H₂S formation in homogenates alone as shown in representative traces of time-dependent measurements of H₂S formation in mice lung homogenates (shown in Fig. 3). These results indicate that RVT or Cpd compounds are not a H₂S donor.

We confirmed the endogenous H₂S formation in lung homogenates by increased signals after the addition of the L-cys together with co-enzyme pyridoxal phosphate (shown in Fig. 3a). We found that incubation of RVT or Cpd2, but not Cpd1, significantly stimulated L-cys-induced endogenous H₂S formation as shown by time-concentration responses curves (shown in Fig. 4), as well as slope of H₂S formation rates (shown in Fig. 4c, p < 0.05). Further this augmented H₂S production and formation rate of H₂S by RVT and Cpd2 is significantly inhibited in the presence of H₂S inhibitor AOAA (shown in Fig. 4a–c, p < 0.001).

The maximum concentration (E_max) of H₂S formation by RVT (E_max = 37.99 ± 2.00 μm) increased compared to control (E_max = 30.87 ± 2.23 μm) (shown in Fig. 4b). The stimulation of H₂S formation reached a higher level in the Cpd2 group (E_max = 43.11 ± 2.17 μm) compared to control (E_max = 34.40 ± 2.04 μm) (shown in Fig. 4b). Although 23% increase by RVT or 25% increase by Cpd2 was statistically significant compared to control, 11% augmentation of H₂S formation by Cpd1 (E_max = 35.53 ± 1.52 μm) did not reach significance compared to the control (E_max = 39.46 ± 3.19 μm) (shown in Fig. 4b). The increased L-cys-induced H₂S formation induced by RVT or Cpd2 was not statistically significant when compared to each other (p > 0,05). In the presence of AOAA, the maximum H₂S levels induced by L-cys stimulated by RVT or Cpd2 were reduced to 6.41 ± 0.14 and 4.87 ± 0.61 μm, respectively (shown in Fig. 4b, p < 0.001).

As we found in our previous experiments that H₂S levels are reduced under oxidative stress in vascular tissues, we tested whether this is also the case in the lung as well. If so, we then aimed to investigate whether Cpd2 could maintain the H₂S levels in the lung under oxidative stress conditions.

We found that superoxide producer pyrogallol (0.1 mm, 5 min) decreased the l-cys-induced H₂S formation in the lungs, as we have shown previously in penile and aorta (shown in Fig. 5a) [16, 17]. Cpd2 restored the decrease in endogenous H₂S formation under oxidative stress time-dependently (shown in Fig. 5a). The maximum H₂S concentration was reduced by the Pyro from 34.15 ± 1.27 μm to 8.40 ± 1.53 μm (shown in Fig. 5b). But Cpd2 restored H₂S levels to 27.59 ± 1.98 μm (shown in Fig. 5b). However, it did not reach the initial levels without oxidative stress in the control.

The reaction rate of H₂S production was decreased by pyro (1.0 ± 0.06) compared to the control (0.50 ± 0.05) (shown in Fig. 5c, p < 0.001). It was significantly restored by 100 μm Cpd2 (0.74 ± 0.04) (p < 0.01). The addition of AOAA strongly inhibited increased H₂S formation and formation rates in the presence of Cpd2, once more this result suggested that the increased H₂S levels by Cpd2 were due to stimulation of endogenous H₂S formation (shown in Fig. 5a–c, p < 0.001). Since Cpd1 did not cause an increase in H₂S formation in healthy lungs, we did not investigate its effect under oxidative stress.

Both compounds Cpd1 (E_max = 73.39 ± 3.93%) and Cpd2 (E_max = 73.46 ± 4.22%) caused a significant concentration-dependent relaxations in the phe-precontracted mouse aorta with intact endothelium (shown in Fig. 6, 7). There was no significant difference in maximum relaxation responses between the two compounds (shown in Fig. 7a–b). Also, the pD2 values of Cpd2 (4.75 ± 0.24) or Cpd1 (4.60 ± 0.06) were not significantly different (shown in Fig. 7a–b).

H₂S is a new endogenous gasotransmitter that plays a regulatory role in physiological and pathophysiological processes in the respiratory system [2, 3]. H₂S relaxes airway smooth muscle cells [20], reduces the production of pro-inflammatory cytokines [8], and also protects against oxidative stress by decreasing levels of reactive oxygen species in the lungs [21]. Decreased H₂S levels in respiratory system diseases such as asthma and PAH have been associated with pathophysiological mechanisms such as inflammation, oxidative stress, apoptosis, and fibrosis [9]. Therefore, H₂S-inducing molecules have a potential to be used in the respiratory diseases.

In our previous study, we reported that RVT increases H₂S levels in the aorta and penile of mice [16, 17]. However, the effect of RVT on H₂S formation in the lungs was not known and for the first time, we have shown that RVT stimulates endogenous H₂S synthesis in the lung (shown in Fig. 4). We also synthesized diphenyl triazolopyrimidine (Cpd2) from its precursor diphenylpropenone (Cpd1) as RVT (diphenylethene structure) analogs and investigated whether they increase endogenous H₂S levels in the lung. Our findings showed that RVT and Cpd2 increased l-cys-induced H₂S production in the lung homogenates of healthy mice, but Cpd1 did not (shown in Fig. 3, 4). Nevertheless, the lack of H₂S-inducing activity of Cpd1 may be due to containing a chalcone structure. The H₂S-inducing activity of Cpd2 may result from its heterocyclic structure, containing nitrogen atoms could be increased the potential for interaction with biological macromolecules. However, the magnitude of the H₂S-inducing activity of Cpd2 was not different than RVT (shown in Fig. 4).

Although several RVT analogs have been reported with different activities, our study is the first report of new synthesized RVT analogs to increase H₂S production in our knowledge. RVT analogs with condensed heterocyclic structures similar to Cpd2 have been reported [22]. Among these studies, 1,2,4-oxadiazole-based structural analogs of RVT have an inhibitory effect on NF-κB activation, ROS production, and LPS-induced pro-inflammatory cytokines release [23]. Besides 1,2,4-thiadiazole [24] and 1,2,3-triazole analogs [25] have anticancer activities. Also, 1,2,4-triazolo[1,5-a] pyrimidine derivatives, which contain the same chemical scaffold in Cpd2, have been reported to have anticancer effects [26, 27].

Previously, we showed that pyrogallol, a superoxide anion generator [28], increased superoxide radicals and other ROS, and decreased H₂S levels, whereas RVT inhibited ROS formation by inducing H₂S formation in mice aorta [16]. In the current study, we found that oxidative stress also decreased endogenous H₂S formation in mice lungs as we have shown before in vascular tissues (shown in Fig. 5) [16]. Although the direct effect of oxidative stress on H₂S formation in the lung is reported by us for the first time as we known, decreased H₂S levels have been accepted as a biomarker in chronic respiratory system diseases where oxidative stress and inflammation are present such as asthma [5, 29], KOAH [30], and PAH [31, 32].

Since H₂S levels were reported to decrease in oxidative stress, we then tested the effects of Cpd2 on H₂S formation under pyrogallol-induced oxidative stress conditions. Cpd2 restored declined H₂S formation under pyrogallol-induced oxidative stress in the lungs (shown in Fig. 5). These results suggest the potential of Cpd2 as a novel therapeutic agent in respiratory diseases. However, further in vivo studies are needed. Supporting this suggestion, we showed that, in PAH in vivo treatment with Na₂S, an H₂S donor improved the augmented right ventricular pressure and cardiac hypertrophy as well as decreased NO and H₂S dependent relaxation responses by increasing H₂S levels [33]. Increased vascular tonus is the main problem in PAH, and cardiac pathologies are the consequences of increased vascular resistance. Thus, in the current study further potential of compounds in vascular relaxation has been tested. We found that both Cpd1 (73.39 ± 3.93%) and Cpd2 (73.46 ± 4.22%) caused relaxations in mice aorta (shown in Fig. 7). Moreover, previously it has been reported that RVT causes relaxation of 74.4% in mice aorta [34]. The relaxation responses were in similar magnitude between RVT and compounds when used at the same concentration (10⁻⁴ M). Due to Cpd2 overlapping well with RVT, it seems that the distance between two phenyl rings may be important to have H₂S-inducing activity (shown in Fig. 2). But it may not be important to induce vascular relaxation since Cpd1, Cpd2, and RVT caused vasorelaxation. Thus, rather than distance, two phenyl rings and an ethylene bridge in the main structure seems to have a role in these vascular effects. Moreover, the mechanism of vascular relaxation was not investigated but since Cpd1 does not induce H₂S, whereas Cpd2 does, the vasorelaxant effect of compounds may not be related to endogenous H₂S formation.

Interestingly, RVT analogs with this similar structure of Cpd2 have been patented for the treatment or prevention of asthma and chronic obstructive pulmonary disease through inhibition of airway contraction and eosinophil peroxidase activity [35]. In light of this report, together with our finding that Cpd2 causes vasorelaxation and induces the formation of H₂S in the lung, which is a target in respiratory disorders, Cpd2 is suggested to have therapeutic potential in respiratory diseases.

Reduced H₂S levels have been linked to pathophysiological processes such as inflammation, oxidative stress, apoptosis, and fibrosis in inflammatory pulmonary diseases. Newly synthesized small molecule Cpd2, as a structural analog of RVT, can induce endogenous H₂S formation in mice lung under both healthy and pathological conditions where oxidative stress is increased. Since H₂S-targeting molecules are potential therapeutic targets for inflammatory respiratory diseases, this study may be a pioneering result to study the effect of Cpd2 potential in respiratory diseases where H₂S levels are decreased. Cpd2 could be studied as a new scaffold for compounds that increase H₂S levels with further studies.

Some equipment in the pharmaceutical research laboratory (FABAL) of Faculty of Pharmacy, Ege University has been used in this study. The authors would like to thank Tubitak #114s448 for being a pioneer project for the current study and also the COST action CA15135 (multi-target paradigm for innovative ligand identification in the drug discovery process MuTaLig) for the support.

This study protocol was reviewed and approved by the Animal Experiment Local Ethical Committee of Ege University No: [2019-009] in agreement with European guidelines for animal care.

The authors have no conflicts of interest to declare.

The authors would like to thank the financial support of the Ege University BAP 21428 project.

G.Y-A. and H.I. designed the experiments. E.N.O. and G.Y-A. drafted the manuscript. E.N.O., U.K., E.A.A., and G.Y-A. performed the study and the statistical analysis. G.Y-A., G.S., and H.I. revised the manuscript. All the authors have read and approved the final manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.