Susceptibility of the antioxidant selenoenyzmes thioredoxin reductase and glutathione peroxidase to alkylation-mediated inhibition by anticancer acylfulvenes

Abstract

Selenium, in the form of selenocysteine, is a critical component of some major redox-regulating enzymes, including thioredoxin reductase (TrxR) and glutathione peroxidase (Gpx). TrxR has emerged as an anticancer target for drug development due to its elevated expression level in many aggressive human tumors. Acylfulvenes (AFs) are semisynthetic derivatives of the natural product illudin S and display improved cytotoxic selectivity profiles. AF and illudin S alkylate cellular macromolecules. Compared to AFs, illudin S more readily reacts with thiol-containing small molecules such as cysteine, glutathione, and cysteine-containing peptides. However, a previous study indicates that the reactivity of AFs and illudin S with glutathione reductase, a thiol-containing enzyme, is inversely correlated with the reactivity toward small molecule thiols. In this study, we investigate mechanistic aspects underlying the enzymatic and cellular effects of the AFs and illudin S on thioredoxin reductase. Both AF and HMAF were found to inhibit mammalian TrxR in the low- to submicromolar range, but illudin S was significantly less potent. TrxR inhibition by AFs was shown to be irreversible, concentration- and time-dependent, and mediated by alkylation of C-terminus active site Sec/Cys residues. In contrast, neither AFs nor illudin S inhibits Gpx, demonstrating that enzyme structure-specific small molecule interactions have a significant influence over the inherent reactivity of the Sec residue. In human cancer cells, TrxR activity can be inhibited by low micromolar concentrations of all three drugs. Finally, it was demonstrated that preconditioning cells by the addition of selenite to the cell culture media results in an enhancement in cell sensitivity toward AFs. These data suggest potential strategies for increasing drug activity by combination treatments that promote selenium enzyme activity.

Figure 1

Dose-dependent inhibition of pre-reduced TrxR by illudin S and AFs. TrxR (80 nM) was first incubated with NADPH (100 μM) at 25 °C for 10 min followed by the addition of test compounds and further incubated for 2 h at 25 °C (A, Illudin S, 62.5, 125, 250, 500, 1000, 2500 μM; B, AF, 2.5, 5, 12.5, 25, 50 μM; C, HMAF, 0.125, 0.25, 0.5, 1, 1.5, 2.5 μM). The residual activity was measured with DTNB assay described in the experimental details.

Figure 2

Inhibition of non-reduced TrxR by AF and HMAF. TrxR (80 nM) was incubated with the test compounds for 2 h at 25 °C (AF, 0.25, 0.5, 1.0, 2.5 mM; HMAF, 0.25, 0.5, 1.0, 2.5, 3.6, 5 mM). The residual activity was measured with DTNB assay described in the experimental details.

Figure 3

Time-dependent inhibition of TrxR by AFs. TrxR (80 nM) was first incubated with NADPH (100 μM) at 25 °C for 10 min followed bby the addition of test compounds at concentrations that would completely inhibit TrxR upon 30 min treatment (A, AF 50 μM; B, HMAF 7.5 μM) at 25 °C. To assay the enzyme activity, 100 μL of the incubation solution was taken out at different time intervals and the residual activity was measured with DTNB assay described in the experimental details. Each data point represents an average of two measurements.

Figure 4

Gel-filtration analysis of AF and HMAF-inactivated TrxR. TrxR (80 nM) was first incubated with NADPH (100 μM) at 25 °C for 10 min followed by the addition of test compounds and further incubation for 2 h at 25 °C (A, AF, 3.3, 6.7, 16.7, 33.3, 66.7 μM; B, HMAF, 0.13, 0.27, 0.53, 1.33, 2.67 μM). Unbound compound was removed by Micro Bio-Spin™ P-6 pre-packed size exclusion columns. The residual activity was measured with DTNB assay described in the experimental details.

Figure 5

Free C-terminus redox-active site Sec and Cys of TrxR were detected by biotin-conjugated iodoacetamide. Different concentrations of AFs were added to NADPH (200 μM) pre-reduced TrxR (0.9 μM) and incubated at 25 °C for 2 h. BIAM was added to label active site free Sec at pH 6.5 and free Cys and Sec at pH 8.5. Results are representative of three independent experiments.

Figure 6

LC/MS analysis of AF-modified TrxR active-site peptide (SGGDILQSGCysSecG). A, Average mass for the unmodified peptide (m/z 1142.4) from an untreated sample of enzyme as a blank control; B, average mass for the AF-modified peptide (m/z 1358.6) from a treated sample; C, proposed mechanism of reaction of AF with Sec at TrxR active site.

Figure 7

Inhibition of TrxR activity in HeLa cells by illudin S and AFs. HeLa cells were exposed to individual compound. A, illudin S: 0.04, 0.20, 1.00 μM; B, AF 0.2, 1.0, 4.0 μM; C, HMAF: 0.2, 1.0, 4.0 μM) for 12 h and then cellular TrxR activity was measured. Asterisks represent a significant difference relative to controls: * p < 0.05, ** p < 0.01.

Figure 8

Differential sensitivity of HeLa cells towards illudin S and AFs. HeLa cells were seeded on 96-well plates (1000 cells/well) and cultured in medium with (red/right) or without (blue/left) addition of sodium selenite (1 μM) for three continuous days and then changed with medium containing test compounds (illudin S, 0.2, 0.5, 1, 2 μM; AFs, 10, 20, 50, 100 μM) for 12 h. Cytotoxicity was measured with MTS assay (Promega). Asterisks represent a significant difference between cytotoxicities resulting from no selenite and 1 μM selenite medium: * p < 0.05, ** p < 0.01.

Chart 1

Structure of illudin S and acylfulvene derivatives AF and HMAF.