Phenacylselenoesters allow facile selenium transfer and hydrogen selenide generation

Abstract

Hydrogen selenide (H₂Se) is a precursor to several selenium-containing biomolecules and is emerging as an important redox-active species in biology, with yet to be completely characterized roles. Tools that reliably generate H₂Se are key to achieving a better understanding of selenium biology. Here, we report the design, synthesis and evaluation of phenacylselenoesters as sources of H₂Se. These compounds are prepared in two steps from commercial compounds, some are crystalline solids, and all are stable during storage. In the presence of esterase and a thiol in pH 7.4 buffer, these compounds produce H₂Se, with half-lives of 5-20 min. We developed a colorimetric assay for the detection of gaseous H₂Se by trapping it as zinc selenide (ZnSe), which is then converted to lead selenide (PbSe), which serves as a convenient visual indicator for this gas. The major organic products that are formed in nearly quantitative yields are relatively benign ketones and carboxylic acids. We provide evidence for these donors producing a thioselenide, a key intermediate in biological selenium metabolism. Finally, we compared sulfur and selenium transfer, both critical processes in cells. Phenacylthiol is relatively stable to cleavage by a thiol, and requires a sulfurtransferase enzyme to produce a persulfide and H₂S. By contrast, the selenium analogue reacted with a thiol in the absence of this enzyme to produce H₂Se. This result underscores the greater lability of the C-Se bond as compared with a C-S bond, and may have implications in biological selenium transfer. Together, phenacylselenoesters are easy to prepare, stable and generate H₂Se under mild and biocompatible conditions. We anticipate that these will be valuable additions to the growing selenium redox toolbox.

Fig. 1. (A) Selenium containing biomolecules are derived from the reduction of inorganic SeO₃²⁻ by glutathione (GSH) to generate the thioselenide, gluathioselenol GS-SeH. Cleavage of this intermediate by biothiols or reductases gives H₂Se. (B) Major classes of hydrogen selenide donors reported in the literature. (C) Proposed esterase-activated thioselenide generation involves the cleavage of the selenoester bond to produce the donor, which can react with a thiol to produce a thioselenide and a ketone. Under reducing conditions, H₂Se is formed from the thioselenide, or when the thioselenide is exposed to other cysteine-containing proteins, a protein thioselenide results.

Scheme 1. Synthesis of selenium donors 1a, 1b and 1c. Overall yields were 28–58%.

Fig. 2. ORTEP diagram of 1b; H-atoms are not shown for clarity (CCDC number 2326967). Details of the crystallographic refinement parameters are listed in the ESI, Tables S1–S4. Both the C–Se bond distances were found to be 1.94 Å, which is consistent with literature reports.

Fig. 3. (A) Workflow for the detection of hydrogen selenide as PbSe. Absorbance at 400 nm was measured. The reaction of selenite and glutathione (GSH) is expected to produce GS-Se-SG, which reacts with DTT to generate H₂Se; 1 is similarly expected to react with ES and DTT to generate H₂Se. (B) PbSe assay of 1a, 1b and 1c (100 μM) with ES and different thiols (DTT = 200 μM, other thiols = 400 μM) in the presence of 400 μM of Zn(OAc)₂ at 30 min. Addition of ES inhibitor PMSF (1 mM) decreases H₂Se production. Student's two-tailed unpaired parametric t-test was used to determine significance: **p < 0.01, ***p < 0.001. All the results are represented as mean ± SD (n = 3 per group). SD stands for standard deviation. (C) A 96-well plate image during the reaction of 1b with ES and DTT, in the presence of zinc acetate: Ctrl is 1b (100 μM) alone, ES: ES (0.1 U mL⁻¹), DTT indicates 1b + DTT (200 μM), and ES + DTT signifies the incubation of 1b, ES, and DTT together. (D) Time courses for the formation of PbSe during incubation of the compound alone or under the standard reaction conditions. All data are represented as mean ± SD (n = 3 per group). Data for: (i) 1a, (ii) 1b and (iii) 1c.

Fig. 4. (A) Mechanism of cleavage of 1b in the presence of ES and DTT produces acetophenone 2 and H₂Se. (B) Representative HPLC chromatograms of 1b (100 μM) in the presence of ES (0.1 U mL⁻¹), DTT (200 μM) and zinc acetate (400 μM), showing the formation of PhCOOH and acetophenone 2 over 30 min. (C) Curve fitting gave (i) the apparent rate constant for decomposition of 1b, k_1b of 0.15 min⁻¹ and the formation of PhCOOH as 0.29 min⁻¹ and (ii) apparent rate constant for the formation of 2 was found to be 0.16 min⁻¹.

Fig. 5. (A) Representative HPLC chromatograms of 1c (100 μM) in the presence of ES (0.1 U mL⁻¹), DTT (200 μM) and zinc acetate (400 μM) over 60 min. (B) Kinetic representation of HPLC traces used for measuring the rate of (i) decomposition of 1c and formation of PhCOOH. (ii) Formation of propiophenone 3. Standard curve fitting using the equation gave apparent rate constants of loss of 1c = 0.11 min⁻¹, formation of benzoic acid = 0.42 min^—1, and formation of 3 = 0.11 min⁻¹.

Fig. 6. (A) Thioacetate 4 is cleaved by ES to produce the designed 3-MST substrate 5. The thiol is positioned to undergo a sulfur transfer reaction to produce persulfidated 3-MST (3-MST-SS⁻) along with acetophenone as a by-product. The 3-MST-SS⁻ reacts with dithiothreitol (DTT) to generate H₂S. (B) Molecular docking analysis of the active site of h3-MST with phenacylselenol intermediate derived from 1b. The distance between the sulfhydryl group of the active site cysteine residue and the Se is 5.5 Å. The R188 and R197 residues are also 4.5 Å and 7.1 Å from the carbonyl group, respectively. (C) Detection of H₂Se using modified PbSe assay; no significant difference is observed between the background reaction of 1b with ES and DTT, and the reaction in the presence of wt 3-MST. The C238A single mutant also shows a similar response for H₂Se generation. Analysis was carried out after 30 min. All data represented as mean ± SD (n = 3 per group). Student's two-tailed unpaired parametric t-test was carried out to determine significance. ****p < 0.0001 for the comparison between the control reactions vs. the reactions of 1b with ES and DTT alone, or with the addition of either E. coli wt 3-MST or C238A mutant.