Synthesis, H2S releasing properties, antiviral and antioxidant activities and acute cardiac effects of nucleoside 5'-dithioacetates

Abstract

Hydrogen sulfide (H₂S) is an endogenous gasotransmitter with cardioprotective and antiviral effects. In this work, new cysteine-selective nucleoside-H₂S-donor hybrid molecules were prepared by conjugating nucleoside biomolecules with a thiol-activatable dithioacetyl group. 5'-Dithioacetate derivatives were synthesized from the canonical nucleosides (uridine, adenosine, cytidine, guanosine and thymidine), and the putative 5'-thio metabolites were also produced from uridine and adenosine. According to our measurements made with an H₂S-specific sensor, nucleoside dithioacetates are moderately fast H₂S donors, the guanosine derivative showed the fastest kinetics and the adenosine derivative the slowest. The antioxidant activity of 5'-thionucleosides is significantly higher than that of trolox, but lower than that of ascorbic acid, while intact dithioacetates have no remarkable antioxidant effect. In human Calu cells, the guanosine derivative showed a moderate anti-SARS-CoV-2 effect which was also confirmed by virus yield reduction assay. Dithioacetyl-adenosine and its metabolite showed similar acute cardiac effects as adenosine, however, it is noteworthy that both 5'-thio modified adenosines increased left ventricular ejection fraction or stroke volume, which was not observed with native adenosine.

Keywords: Antioxidant; Antiviral; Dithioacetate; H2S donor; H2S release kinetics; Nucleoside; SARS-CoV-2.

Fig. 1

(A) H₂S donors, previously synthesized by our research group; (B) H₂S donor nucleosides in the literature; (C) this work: synthesis and characterization of cysteine selective H₂S donor nucleosides and their metabolites.

Fig. 2

Synthesis of nucleoside 5′-tosylates.

Fig. 3

(A) Synthesis of the uridine dithioacetate 21 by freshly prepared dithioacetate salt CH₃CSSMgBr; (B) Comparison of the characteristic NMR signals of the ¹H spectra of dithioacetate 21, bromouridine 20 and their mixture.

Fig. 4

(A) Synthesis of nucleoside dithioacetates 22, 24, 27, 29 and 30; (B) synthesis of 5′-thionucleosides 32 and 34.

Fig. 5

H₂S releasing curves measured by H₂S selective sensor (A–F, 10% w/v) and two possible routes of H₂S release from 5′-dithioacetyl nucleosides under physiological conditions (G). (A) Compound 22 (0.31 mM); (B) compound 24 (0.29 mM) (inset: first 60 min); (C) compound 27 (0.28 mM); (D) compound 29 (0.32 mM); (E) compound 30 (0.32 mM); (F) the blue curve shows the H₂S release of compound 24 (0.31 mM), the orange curve shows the H₂S release of NaHS (0.1 mg in 5 mL of distilled water, 0.36 mM) and the green curve shows the H₂S release of GYY4137 (0.5 mg in 5 mL of medium, 0.27 mM); (G) H₂S release (a) by cysteine-mediated transesterification, (b) by aspecific esterase-catalyzed hydrolysis.

Fig. 6

DPPH inhibition ratio of (A) compounds 22, 32, and 22 + extract (20 µL), and (B) compounds 24, 34, and 24 + extract (20 µL) immediately and after 30 min of incubation, compared to the positive controls (0.5 mM ascorbic acid (AA, IR = 97.25% and 97.13%), 0.5 mM GYY4137 (IR = 5.95% and 10.63%), and 0.5 mM NaHS (IR = 56.14% and 90.03%), immediately and after 30-min incubation, respectively). Error bars symbolize the standard deviation from the mean value (n = 3). Results express mean ± SD. IR = inhibition ratio. (C) IC₅₀ values (mM) of synthetic compounds compared to the corresponding unmodified nucleosides and ascorbic acid (AA). (D) ABTS (2,2′-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) radical cation decolorization assay of compounds 22, 24, 32, 34 GYY4137, and NaHS. Values are expressed as Trolox equivalents (a vitamin E analogue). Error bars symbolize the mean ± SEM (n = 9).

Fig. 7

Comparison of the cardiac effects of adenosine (n = 4), compound 24 (n = 4), 34 (n = 7) and 22 (n = 5) on anaesthetized SD rats. Panels (A–D) show M-mode traces of the left ventricle, obtained from PSAX view at baseline and 30 s after drug administration (sweep speed adjusted to 60 0 Hz). Compound 24 and 34 caused similar but smaller effect than the parent molecule adenosine. Ejection fraction (panel H) increased after compound 24 administration but slightly decreased under the influence of compound 22. Heart rate decreased after adenosine, compound 24 and 34 administration, but only slightly changed after compound 22 injection (panel E). ECG PR interval was unchanged in all cases (panel F). A significant QT interval prolongation on the ECG was observed after the administration of all adenosine-based compounds, but not after the uridine-based (panel G). One-way ANOVA with Tukey post-test was used to compare % changes of the parameters in the treatment groups. Asterisks denote the level if significance: *p < 0.05; **p < 0.01; ***p < 0.001. Green arrows denote lead II ECG traces. Panel (I) Effect of adenosine, compound 24, 34 and 22 on ECG and echocardiographic parameters, at the baseline and 30 s after i.v. drug administration (2 mg/kg, i.v.). Adenosine-based compounds significantly decreased heart rate with a marked QT prolongation, while increased SV and EF. All data is presented as mean ± SD. Asterisks denote the level if significance: *p < 0.05; **p < 0.01; ***p < 0.001 (paired t-test, vs. baseline values).