Biological Effects of HDAC Inhibitors Vary with Zinc Binding Group: Differential Effects on Zinc Bioavailability, ROS Production, and R175H p53 Mutant Protein Reactivation

Abstract

The coordination of zinc by histone deacetylase inhibitors (HDACi), altering the bioavailability of zinc to histone deacetylases (HDACs), is key to HDAC enzyme inhibition. However, the ability of zinc binding groups (ZBGs) to alter intracellular free Zn⁺² levels, which may have far-reaching effects, has not been explored. Using two HDACis with different ZBGs, we documented shifts in intracellular free Zn⁺² concentrations that correlate with subsequent ROS production. Next, we assayed refolding and reactivation of the R175H mutant p53 protein in vitro to provide greater biological context as the activity of this mutant depends on cellular zinc concentration. The data presented demonstrates the differential activity of HDACi in promoting R175H response element (RE) binding. After cells are treated with HDACi, there are differences in R175H mutant p53 refolding and reactivation, which may be related to treatments. Collectively, we show that HDACis with distinct ZBGs differentially impact the intracellular free Zn⁺² concentration, ROS levels, and activity of R175H; therefore, HDACis may have significant activity independent of their ability to alter acetylation levels. Our results suggest a framework for reevaluating the role of zinc in the variable or off-target effects of HDACi, suggesting that the ZBGs of HDAC inhibitors may provide bioavailable zinc without the toxicity associated with zinc metallochaperones such as ZMC1.

Keywords: HDAC inhibitors; R175H mutant p53; ROS production; p53 reactivation; zinc bioavailability.

Figure 1

In vitro measurement of R175H mutant p53 response element (RE) binding. (A) R175H mutant p53 binding to a p53 RE increases with an increasing concentration of exogenous ZnCl₂ at or above 25 μM. (B) In 50 μM exogenous ZnCl₂, the addition of 50 μM of the HDACi MS275 significantly increased p53 RE binding, with no increase observed following the addition of 50 μM of the HDACi FK228 or 50 μM of the zinc metallochaperone ZMC1. (C) At 50 μM exogenous ZnCl₂, the addition of 50 μM FK228 did not increase the p53 RE binding, but both the absence of zinc or the addition of DDT resulted in a decrease in p53 RE binding that was abolished when a higher concentration of ZnCl₂ (100 μM) was used. (D) In the absence of exogenous ZnCl₂, 50 μM MS275 increased p53 RE binding over the control with the addition of 50 or 100 μM ZnCl₂, resulting in significantly higher p53 RE binding. Horizontal lines with error bars represent the mean ± SEM. * denotes p < 0.05, ** denotes p < 0.01, *** denotes p < 0.001, **** denotes p < 0.0001.

Figure 2

(A) FluoZin^TM-3 assay measuring intracellular concentrations of free Zn⁺² following treatment with 10 nM ZMC1, 5 μM MS275, or 1 nM FK228 for 15 min. Compared to untreated cells (control), both ZMC1 and MS275 significantly increased the intracellular concentration of free Zn⁺² while FK228 had no effect. (B,C) CellROX assay measuring intracellular ROS production following treatment of cells for 45 min (B) or 24 h (C). Compared to the untreated cells (control, CNTL), ROS production was significantly elevated in the cells treated for either 45 min or 24 h with ZMC1 and MS275, with significantly higher ROS production following ZMC1 treatment compared to MS275 after 24 h. No increase in ROS production was observed when the cells were treated for either 45 min or 24 h with FK228. Horizontal lines with error bars represent the mean ± SEM. * denotes p < 0.05, *** denotes p < 0.001, **** denotes p < 0.0001.

Figure 3

In vivo refolding and reactivation of R175H mutant p53. (A) Cells treated with either 10 nM ZMC1 or 5 μM MS275, but not 1 nM FK228, resulted in the refolding of R175H p53, as identified by anti-p53 pAb1620 binding (green). Nuclei are stained with DAPI, as indicated in the bottom four panels. (B) Quantification of fluorescent intensity, normalized to DAPI, show that ZMC1 and MS275 treatment resulted in a significantly increased fluorescent intensity compared to the control. Treatment with 1 nM FK228 had no effect on the refolding of R175H p53. (C) An increase in p53 RE binding, as measured by luciferase activity, was observed when the cells were treated with either 1 nM FK228 or 5 μM MS275, but not with 10 nM ZMC1. Horizontal lines with error bars represent the mean ± SEM. * denotes p < 0.05, **** denotes p < 0.0001.

Figure 4

Measurement of R175H mutant p53 RE binding following treatment. (A) Compared to the untreated H1299 cells transfected with R175H mutant p53 (control), cells treated with either 10 nM ZMC1, 5 μM MS275, or 1 nM FK228 showed a significant increase in their R175H mutant p53 RE binding. Untransfected H1299 cells, null for p53 expression, served as a negative control. Horizontal lines with error bars represent the mean ± SEM. ** denotes p < 0.01, *** denotes p < 0.001, **** denotes p < 0.0001. (B) Immunoblotting of purified nuclear extracts from the treated and untreated cells showed that treatment with ZMC1 or MS275 resulted in a greater proportion of p53 that was in the folded vs. the misfolded configuration, while treatment with FK228 showed the opposite pattern. Immunoblot original image can be found in Figure S5. (C) An ELISA using an anti-p53 pAb240 antibody performed on nuclear extracts isolated from cells treated with ZMC1 or MS275 confirmed that treatment resulted in a significant reduction in the fraction of misfolded R175H mutant p53.

Figure 5

Detection of post-translational modifications (PTMs) present on R175H mutant p53 following treatment. (A) Immunoblot using anti-phospho-Ser15, anti-phospho-Ser315 p53, and anti-Lys382 antibodies on the cell lysates obtained from the untreated cells vs. cells treated with either 10 nM ZMC1, 5 μM MS275, or 1 nM FK228. Results showed that all PTMs were present on R175H mutant p53 but in varying abundances. Immunoblot original images can be found in Figure S6. (B) Immunoblot (top) and quantification (bottom) of p53 phosphorylation on serine 15 in nuclear (N) and cytosolic (C) extracts, normalized to total p53 by immunoblotting revealed a significant increase in this PTM in nuclear lysates from cells treated with 10 nM ZMC1 (* p < 0.05). Immunoblot original image can be found in Figure S8.