A proteasome-dependent inhibition of SIRT-1 by the resveratrol analogue 4,4'-dihydroxy- trans- stilbene

Abstract

Background and aim: Resveratrol (RSV), is a stilbene-based compound exerting wide biological properties. Its analogue 4,4'-dihydroxy-trans-stilbene (DHS) has shown improved bioavailability and antiproliferative activity in vitro and in vivo. One of the hypotheses on how resveratrol works is based on SIRT1 activation. Since their strict structural similarities, we have explored a potential interaction between DHS and SIRT1, in comparison with the parental molecule.

Experimental procedure: Timing of incubation and concentrations of DHS have been determined using MTT assay in normal human lung fibroblasts. Untreated, DHS- or RSV-treated cells were harvested and analysed by Western Blotting or RT-PCR, in order to evaluate SIRT1 levels/activity and expression, and by Cellular Thermal shift assay (CETSA) to check potential DHS or RSV-SIRT1 interaction. Transfection experiments have been performed with two SIRT1 mutants, based on the potential binding pockets identified by Molecular Docking analysis.

Results and conclusion: We unexpectedly found that DHS, but not RSV, exerted a time-dependent inhibitory effect on both SIRT1 protein levels and activity, the latter measured as p53 acetylation. At the mRNA level no significant changes were observed, whereas a proteasome-dependent mechanism was highlighted for the reduction of SIRT1 levels by DHS in experiments performed with the proteasome inhibitor MG132. Bioinformatics analysis suggested a higher affinity of RSV in binding all SIRT1 complexes compared to DHS, except comparable results for complex SIRT1-p53. Nevertheless, both CETSA and SIRT1 mutants transfected in cells did not confirm this interaction. In conclusion, DHS reduces SIRT1 protein level, thereby inhibiting its activity through a proteasome-mediated mechanism.

Keywords: 4,4′-dihydroxystilbene; Deacetylation; Resveratrol; Sirtuin-1.

Graphical abstract

Fig. 1

(A) Chemical structure of the studied stilbenes. (B) Effects of 24 h-incubation of DHS, PTS and RSV (30 μM) on the SIRT1 protein levels in LF1 cells. (C) Quantification of SIRT1 protein by densitometric analysis of the Western blot and normalized to the internal loading control (β-actin). Data are the mean ± SD from at least three independent experiments; values are expressed as arbitrary units (a.u.). (*p ≤ 0.05 and **p ≤ 0.01 compared to control cells).

Fig. 2

(A) Cell toxicity determined by the MTT assay after 24 h of treatment with different concentrations of DHS (1, 2.5, 5, 7.5, 10, 15, 30, 60 and 90 μM) in LF1 cells. (B) Effect on DNA synthesis evaluated by BrdU incorporation in LF1 cells seeded on coverslips, and treated with increasing concentration of DHS as indicated. Representative images of BrdU-positive cells are showed in the panel (Scale bar = 10 μm). (C) Analysis of BrdU incorporation in LF1 cells treated with different concentration of DHS as indicated. (D) Analysis of BrdU incorporation of 30 μM DHS or RSV at different time (1, 3, 6, 24 h). All results are expressed as mean ± SD from three independent experiments; values are expressed as percentage of BrdU-stained positive cells (*p ≤ 0.05; **p ≤ 0.01 compared to control cells).

Fig. 3

(A) Effects of 30 μM of DHS in LF1 cells at different incubation times and (B) at different concentrations for 24 h on SIRT1 protein levels. (C and D) Quantification of SIRT1 protein by densitometric analysis of the Western blot and normalization to the internal loading control (β-actin and vinculin, respectively). Data are the mean ± SD from at least three independent experiments; values are expressed as arbitrary units (a.u.). (**p ≤ 0.01 compared with control cells). (E) Time-dependent analysis of SIRT1transcripts by quantitative real-time PCR (RT-PCR) in LF1 cells treated with 30 μM DHS or RSV at the indicated times. (F) Time dependent analysis of SIRT1 protein levels by Western blot in LF1 cells treated with 30 μM DHS or RSV at the indicated time and (G) the relative densitometric analysis. Both RT-PCR and Western Blot were performed on samples derived from the same experiment.

Fig. 4

(A) p53 acetylation levels after treatment with 20 μM Eto (3 h) and in combination 30 μM DHS or RSV (1 h) + 20 μM Eto (3 h) and (C) the relative densitometric analysis. (B) Kinetic experiments on SIRT1, p53 and p21 protein levels in untreated or treated with 20 μM Eto plus 30 μM DHS LF1 cells, and (D) the relative densitometric analysis of the correspondent protein levels. Data are the mean ± SD from at least three independent experiments; values are expressed as arbitrary units (a.u.). (*p ≤ 0.05, **p ≤ 0.01 compared with control cells; • p ≤ 0.05 compared with Eto-treated cells).

Fig. 5

(A) SIRT1 protein levels obtained by Western blotting and (B) the correspondent quantitative analysis. Data are the mean ± SD from at least three independent experiments; values are expressed as arbitrary units (a.u.). (C) p53 acetylation levels in untreated or pre-treated with MG132 (25 μM) samples in the presence of Eto, DHS, DHS + Eto, RSV and RSV + Eto and (D) the relative densitometric analysis. (*p ≤ 0.05, **p ≤ 0.01 compared with control cells).

Fig. 6

SIRT1 – RSV/DHS docking pose. SIRT1 is represented in cartoon, RSV and DHS ligands are in licorice and p53 is in lines. (A) SIRT1, (B) SIRT1-p53, (C) SIRT1-p53AMC, (D) SIRT1-p53W.

Fig. 7

(A and B) DHS or RSV and SIRT1 interaction evaluated by CETSA in untreated LF1 cells or treated with 30 μM DHS or RSV at the indicated times. The levels of non-denaturated protein fractions are normalized to β-actin, and (C and D) the relative densitometric analysis of the correspondent bands. Data are the mean ± SD from at least three independent experiments; values are expressed as arbitrary units (a.u.).

Fig. 8

(A) Full length (Fl) SIRT1 and its mutant Δ5 e Δ6 in the absence or presence of 30 μM DHS for 24 h and (B) the relative densitometric analysis. Data are the mean ± SD from at least three independent experiments; values are expressed as arbitrary units (a.u.).

Fig. S1

Binding pockets as resulted by the in silico docking.