A New Opportunity for "Old" Molecules: Targeting PARP1 Activity through a Non-Enzymatic Mechanism

Abstract

In recent years, new therapies have been developed based on molecules that target molecular mechanisms involved in both the initiation and maintenance of the oncogenic process. Among these molecules are the poly(ADP-ribose) polymerase 1 (PARP1) inhibitors. PARP1 has emerged as a target with great therapeutic potential for some tumor types, drawing attention to this enzyme and resulting in many small molecule inhibitors of its enzymatic activity. Therefore, many PARP inhibitors are currently in clinical trials for the treatment of homologous recombination (HR)-deficient tumors, BRCA-related cancers, taking advantage of synthetic lethality. In addition, several novel cellular functions unrelated to its role in DNA repair have been described, including post-translational modification of transcription factors, or acting through protein-protein interactions as a co-activator or co-repressor of transcription. Previously, we reported that this enzyme may play a key role as a transcriptional co-activator of an important component of cell cycle regulation, the transcription factor E2F1. Here, we show that PARP inhibitors, which interfere with its activity in cell cycle regulation, perform this without affecting its enzymatic function.

Keywords: PARP inhibitors; animal disease models; cancer; neoplasm; poly(ADP-ribose) polymerase-1.

Figure 1

Structure of PARP1 inhibitors and gossypol, an inhibitor of PARP1 protein–protein interactions. Compound 1, 3-aminobenzamide; compound 2, NU1025; compound 3, PJ34; compound 4, TiQ-A; compound 5, ABT-888; compound 6, gossypol. Gossypol is obtained from its natural source (cotton) as a racemic mixture of two atropisomers where the (-)-gossypol isomer (6a) specifically interferes with PARP1 protein–protein interactions.

Figure 2

Effect of PARP1 inhibition on E2F1 transcriptional activity. (A), EdU incorporation assay in MEF treated with the PARP1 inhibitors PJ34 (10 µM), ABT-888 (10 µM), NU1025 (100 µM), TIQ-A (50 µM), 3-AB (5 mM), and gossypol (25 µM). A vertical line separates cells treated with inhibitors dissolved in water and inhibitors dissolved in DMSO (* = p < 0.01). (B), representative images of results presented in panel A. Cell nuclei were counterstained with Hoechst 33258 (bisbenzimide). Scale (20×) = 25 μm. (C), luciferase activity assay in HEK293 transfected with E2F-Luc plasmid treated with PARP1 inhibitors. A vertical line separates cells treated with inhibitors dissolved in water and inhibitors dissolved in DMSO. Cells were synchronized prior to treatment, and results were normalized to its corresponding vehicle (water or DMSO). Shown is a representative experiment of three independent replicates (* = p < 0.01) (D), semiquantitative RT-PCR of E2F1 transcriptional targets. RNA was isolated from Parp1^+/+, Parp1^−/−, and Parp1^−/− transduced with pBABE-hPARP1 and specific PCR products from each gene indicated in the figure were run on an agarose gel. Results are representative of three independent experimental replicates.

Figure 3

Effect of PARP1 inhibitors on primary astrocytes. Postnatal-day-3 astrocytes obtained from Parp1^−/− cRb^−/− HRAS^V12 or Parp1^+/+ cRb^−/− HRAS^V12 mice were treated in vitro with PARP1 inhibitors. (A), morphological changes in cells stained with crystal violet. (B), proliferation rate. Cells were stained with crystal violet, and cell number was determined by spectrophotometry. (C), percent of senescent cells, obtained by quantification of SA-β-galactosidase activity. Shown is a representative experiment of three independent replicates (* = p < 0.001). (D), colony formation. Astrocytes were fixed with methanol-acetic acid (3;1, v/v) and stained with 0.1% (w/v) crystal violet in PBS on day 7 of culture. (E), biochemical analysis of the main DDR checkpoints and transcriptional targets of E2F1. All results are representative of, at least, three independent experiments.

Figure 4

Effect of partial restoration of PARP1 catalytic activity. Primary astrocytes obtained from Parp1^−/− cRb^flox/flox, Parp1^−/− cRb^flox/flox HRAS^V12, Parp1^−/− cRb^−/−, or Parp1^−/− cRb^−/− HRAS^V12 mice were transduced with a copy of hPARP1 with reduced catalytic activity. (A), morphological changes in cells stained with crystal violet. (B), proliferation rate. Cells were stained with crystal violet, and cell number was determined by spectrophotometry. (C), percent of senescent cells, obtained by quantification of SA-β-galactosidase activity. (D), percent of apoptotic cells. Apoptotic nuclei were stained with Hoechst 33258. All results are representative of, at least, three independent experiments.

Figure 5

Effect of PARP1 inhibition in vivo. (A), timeline of the in vivo transformation assay. (B,C), effect of PJ34 on tumoral growth. Parp1^+/+ cRb^−/− HRAS^V12 mice were injected subcutaneously with 3·10⁶ astrocytes and treated with vehicle (control) or PJ34 (10 mg/kg). (D), tumor volume in control (n = 11) or PJ34-treated (n = 8) mice. Representative images or results presented in panel B and C (* = p < 0.05). (E), caspase 3 and p-histone 3 expression in tumors obtained from control or PJ34-treated.