NADP+ is an endogenous PARP inhibitor in DNA damage response and tumor suppression

Abstract

ADP-ribosylation is a unique posttranslational modification catalyzed by poly(ADP-ribose) polymerases (PARPs) using NAD⁺ as ADP-ribose donor. PARPs play an indispensable role in DNA damage repair and small molecule PARP inhibitors have emerged as potent anticancer drugs. However, to date, PARP inhibitor treatment has been restricted to patients with BRCA1/2 mutation-associated breast and ovarian cancer. One of the major challenges to extend the therapeutic potential of PARP inhibitors to other cancer types is the absence of predictive biomarkers. Here, we show that ovarian cancer cells with higher level of NADP⁺, an NAD⁺ derivative, are more sensitive to PARP inhibitors. We demonstrate that NADP⁺ acts as a negative regulator and suppresses ADP-ribosylation both in vitro and in vivo. NADP⁺ impairs ADP-ribosylation-dependent DNA damage repair and sensitizes tumor cell to chemically synthesized PARP inhibitors. Taken together, our study identifies NADP⁺ as an endogenous PARP inhibitor that may have implications in cancer treatment.

Fig. 1

The NADP⁺/NAD⁺ ratio is associated with the cell sensitivity to poly(ADP-ribose) polymerase (PARP) inhibitor. a The sensitivity of 20 ovarian cancer cell lines to olaparib. Twenty ovarian cancer cell lines were treated with olaparib at the indicated doses. Cells were examined by MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assays. Based on the sensitivity to olaparib, the cell were classified into four groups. The data were summarized from three independent experiments. Data are presented as mean ± SD. b Spearman’s correlation analysis was performed between DNA damage-induced PARylation levels and the IC50 of olaparib. DNA damage-induced endogenous PARylation levels were summarized from the western blotting in Supplementary Figure 3. IC50 of olaparib was calculated from the MTT assays. DNA damage-induced endogenous PARylation levels correlated with IC50 of olaparib (left) and PARylation levels in the olaparib-sensitive cells are markedly lower than that in the olaparib-insensitive cells (right). c The expression level of PARP1 is not associated with the sensitivity to olaparib. PARP1 expression levels were summarized from western blotting results in Supplementary Figure 4 by using ImageJ. No correlation was seen between PARP1 expression levels and IC50 of olaparib (left); and no difference was observed in the PARylation levels between the olaparib-sensitive cells and the olaparib-insensitive cells (right). d, e Correlation analysis of NAD⁺ and NADP⁺ concentrations and cell sensitivity to olaparib. NAD⁺ and NADP⁺ concentrations were measured in 20 ovarian cancer cell lines. Both NAD⁺ and NADP⁺ concentrations are associated with the sensitivity to olaparib (left). The NAD⁺ concentrations in the olaparib-sensitive cells are markedly higher than that in the olaparib-insensitive cells (d, right). And the NADP⁺ concentrations in the olaparib-sensitive cells is markedly lower than that in the olaparib-insensitive cells (e, right). f Sensitivity of ovarian cancer cells to olaparib is associated with the NADP⁺/NAD⁺ ratio. The NADP⁺/NAD⁺ ratio was calculated. NADP⁺/NAD⁺ ratio shows significant correlation with sensitivity of ovarian cancer cells to olaparib. p Value of Spearman’s correlation was calculated by R function, cor.test() (b–e, left; f). Statistical significance of the difference between the olaparib-sensitive cells and olaparib-insensitive cells was analyzed using two-tailed unpaired Student’s t tests (b–e, right)

Fig. 2

NADP⁺ is not a donor for poly(ADP-ribose) polymerase (PARP)-mediated ADP-ribosylation. a, b PARP1 cannot use NADP⁺ for PARylation. PARP1-mediated in vitro PARylation assays were performed including recombinant HIS-PARP1, ssDNA, NAD⁺, or NADP⁺. Auto-PARylation of PARP1 was examined by western blotting. 5% of recombinant PARP1 in each sample was extracted before the reaction and was examined as the loading control (a). [³²P]NAD⁺ or [³²P]NADP⁺ was included in the in vitro PARylation assay. Auto-PARylation of PARP1 was examined by autoradiography (b). Coomassie staining of HIS-PARP1 was shown as the loading control. c NADP⁺ cannot serve as the donor for PARP2-mediated PARylation. The in vitro PARylation was performed using recombinant HIS-PARP2, ssDNA, and [³²P]NAD⁺ or [³²P]NADP⁺. Auto-PARylation of PARP2 was examined by autoradiography. Coomassie staining of HIS-PARP2 was shown as the loading control. d NADP⁺ cannot serve as the donor for PARP10-mediated MARylation. Auto-MARylation of PARP10 was examined by autoradiography. Coomassie staining of GST-PARP10 was shown as the loading control

Fig. 3

NADP⁺ suppresses poly(ADP-ribose) polymerase (PARP) activity in vitro. a PARPs can bind to NADP⁺. HIS-PARP1, HIS-PARP2, and GST-PARP10 were immobilized on Ni Sepharose and glutathione S-transferase (GST) beads respectively, followed by incubating with [³²P]NADP⁺. [³²P]NADP⁺ was heat-released from Ni Sepharose or GST beads and examined by thin-layer chromatography. As the negative controls (NC), Ni Sepharose or GST beads without recombinant PARP proteins was incubated with [³²P]NADP⁺. b, c NADP⁺ suppresses PARP1’s activity in vitro. In vitro PAPR1-mediated PARylation assay was performed using different ratio of NADP⁺/NAD⁺. Auto-PARylation of PARP1 was examined by western blotting (b). [³²P]NAD⁺ was used as the donor in PARP1-mediated in vitro PARylation assay. The auto-PARylation of PARP1 was examined by autoradiography. Coomassie staining of His-PARP1 was shown as the loading control (c). d NADP⁺ suppresses PARP2 in vitro. Auto-PARylation of PARP2 was examined by western blotting with the indicated antibody. e NADP⁺ suppresses PARP10 in vitro. Auto-MARylation level of PARP10 was examined by autoradiography

Fig. 4

NADP⁺ negatively regulates DNA damage-induced PARylation in cells. a The schematic diagram shows the tetracycline-inducible expression of NADK. A nuclear localization signal (NLS) and mCherry were added at the N-terminus of NADK. The Tet-on system is from Addgene. b Doxycycline induces the expression of NADK (mCherry) in U2OS cells. c The NADP⁺/NAD⁺ ratio in U2OS cells is significantly increased following doxycycline induction. Three independent experiments were performed. Data are presented as mean ± SD. **p < 0.01. Data were analyzed using two-tailed unpaired Student’s t tests. d High level of NADP⁺ suppresses the DNA damage-induced PARylation. The cells were treated with 10 mM methyl methanesulfonate. Dot blotting was performed to examine PARylation at the indicated time points after DNA damage. The samples were also diluted three times (3×) or 9 times (9×) in the same dot blotting to avoid sample loading variation. Three independent experiments were performed. Data are presented as mean ± SD, **p < 0.01. Data were analyzed using two-tailed unpaired Student’s t tests. e Kinetics of PARylation at DNA lesions. The cells were treated with laser microirradiation. PARylation at the laser stripes was examined by immunofluorescence staining with anti-PAR antibody. Representative cells at different time points are shown. The relative signal strength of PAR was summarized from three independent experiments and at least 15 cells at each time point in each experiment. Data are presented as mean ± SD. Bars: 10 μM. **p < 0.01. Data were analyzed using two-tailed unpaired Student’s t tests

Fig. 5

NADP⁺ suppresses the PARylation-dependent early DNA damage response. a Doxycycline-induced NADK suppresses the recruitment of the PARylation-dependent DNA damage response. Twenty-four hours after induction, the recruitment kinetics of XRCC1, CHFR, PNKP, NBS1, BARD1, and LIG4 were examined with live cell imaging. The relative intensity on the laser stripes were quantified and summarized at the right panel (mean ± SD, from 15 cells at each time point in each experiment). The highest intensity on the laser stripes was calculated as 100% in each cell, and the kinetics of the recruitment were plotted. Three independent experiments were performed. Bars: 10 μM. b High level of NADP⁺ impairs DNA damage repair. The cells were treated with methyl methanesulfonate (10 mM, 30 min). The kinetics of double-strand break and single-strand break repair was examined by neutral and alkaline comet assays, respectively. Representative comet tails at different time points are shown. The tail moments were summarized from at least 50 cells at each time point in each experiment, and three independent experiments were performed. NT non-treated. **p < 0.01. Data were analyzed using two-tailed unpaired Student’s t tests

Fig. 6

High level of NADP⁺ enhances the sensitivity of cancer cells to chemically synthesized PARP inhibitor. a, b Increased in the levels of NADP⁺ and olaparib additively suppresses tumor cell growth. MDAH2774 and OVCAR5 with exogenous expression of NADK were examined for colony-formation assays. After treatment with 2 μM olaparib for 2 weeks, the colonies were stained with crystal violet and counted. The results were summarized from three independent experiments. Data are presented as mean ± SD. p Value was calculated using two-tailed unpaired Student’s t tests. c High level of NADP⁺ sensitizes tumor cells to olaparib. MDAH2774 and OVCAR5 with or without the exogenous expression of NADK were treated with olaparib at the indicated concentrations. MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assays were performed to examine cell viability. The data are summarized from three independent experiments. Data are presented as mean ± SD. *p < 0.05. Data were analyzed using two-tailed unpaired Student’s t tests. d, e High level of NADP⁺ sensitizes xenograft tumors to olaparib. Ovarian cancer xenografts were initiated by subcutaneous inoculation of MDAH2774 cells with or without exogenous expression of NADK in the lower flank of NOD SCID mice as described previously. In all, 5 mg kg⁻¹ olaparib was given intraperitoneally daily. Controls were dosed with vehicle only. Both digital photograph (d) and tumor growth curve (e) shows that exogenous expression of NADK significantly sensitized the MDAH2774 xenograft tumors to olaparib treatment. Data were presented as mean ± SD. **p < 0.01. Data were analyzed using two-tailed unpaired Student’s t tests. f A model for NADP⁺-mediated suppression of PARylation and DNA damage repair