NF-Kb and Poly (ADP-Ribose) Polymerase 1 Form a Positive

Published OnlineFirst December 17, 2018; DOI: 10.1158/1541-7786.MCR-18-0523

Genome Maintenance Molecular Cancer Research NF-kB and Poly (ADP-ribose) Polymerase 1 Form a Positive Feedback Loop that Regulates DNA Repair in Acute Myeloid Leukemia Cells Ding Li1, Yufei Luo1, Xianling Chen2, LingYu Zhang1, Tingting Wang1, Yingting Zhuang3, Yingjuan Fan3, Jianhua Xu1,3,4, Yuanzhong Chen2, and Lixian Wu1,3,4

Abstract

NF-kB mediates acquired resistance in acute myeloid DNA repair. Simultaneous treatment with the NF-kB inhib- leukemia (AML) cells treated with DNA-damaging agents. itor BMS-345541 and the PARP1 inhibitor olaparib resulted Because DNA repair is the major molecular shift that alters in robust killing of AML cells. This dual inhibition signif- sensitivity to DNA-damaging agents, we explored whether icantly suppressed tumor growth and extended survival activation of the NF-kB pathway promotes AML cell survival times in xenograft tumor models. by regulating DNA repair after chemotherapy. Our results showed that RELA, an important subunit of NF-kB, regu- Implications: RELA and PARP1 form a positive feedback loop lated DNA repair by binding to the promoter region of the to regulate DNA damage repair, simultaneous inhibition of PARP1 gene and affecting PARP1 gene transcription. Con- NF-kB and PARP1 increases the antileukemic efﬁcacy of dau- versely, PARP1 knockdown reduced NF-kB activity, indicat- norubicin in vitro and in vivo, broadening the use of PARP1 ing that NF-kB and PARP1 create a positive feedback loop in inhibitors.

inhibit DNA repair as a rational sensitization method to improve Introduction genotoxicity therapy. Acute myeloid leukemia (AML) is a highly aggressive hemato- Constitutive NF-kB pathway activation has been found in logic malignancy characterized by the overproduction of imma- different types of AML (5). DNA-damaging agents activate ture white blood cells (1). Despite important advances in our NF-kB initiating in the nucleus instead of via membrane-bound understanding of the molecular basis of AML, survival outcomes receptors (6–8). NF-kB activation induced by DNA damage is of patients with AML have not improved over the past 20 years (2), necessary for secondary resistance in AML cells (9, 10). Because indicating the need to develop novel therapies that are more efﬁcient repair of DNA damage is required for the survival of effective and less toxic. AML cells (11, 12), we investigated whether NF-kB pathway DNA damage is the common mechanism induced by radio- activation also regulates DNA repair to promote cell survival therapy and chemotherapy in the clinical treatment of cancer (3). after chemotherapy. Although therapy-induced DNA damage is widespread, AML cells PARP1 is a nuclear protein that is mainly known for its ability to utilize endogenous DNA repair mechanisms to enable their facilitate DNA repair by catalyzing the poly ADP-ribosylation survival, recurrence, and resistance (4). In addition, enhancement (PARylation) of itself and other repair-associated proteins of DNA repair pathways in chemotherapeutic-resistant AML cells (13, 14). PARP1 hyperactivation in DNA repair is critical for the is often considered as the major molecular change that alters its resistance to genotoxic agents, which has been conﬁrmed not only sensitivity to DNA-damaging agents. Given the key roles of DNA in cancer cell lines and xenografts, but also in several clinical repair pathways in chemoresistance, it has been proposed to studies (15–17). In addition, PARP1 inhibitors are a new type of anticancer drugs that selectively kills cancer cells with homologous recombination (HR) repair defects (18, 19). However,

1 deficient HR and the rapid emergence of resistance have largely Department of Pharmacology, School of Pharmacy, Fujian Medical University, – Fuzhou, P.R. China. 2Fujian Institute of Hematology, Union Hospital, Fujian limited the clinic usage of PARP1 inhibition (20 22). Thus, there Medical University, Fuzhou, P.R. China. 3Institute of Materia Medica, Fujian is an urgent need to develop novel therapeutic approaches to Medical University, Fuzhou, P.R. China. 4Fuijan Key Laboratory of Natural extend the efficacy of PARP1 inhibitor–based therapies. Medicine Pharmacology, Fujian Medical University, Fuzhou, P.R. China. Because both NF-kB and PARP1 activities are indispensable to Note: Supplementary data for this article are available at Molecular Cancer the establishment of resistance to genotoxic agents, we sought to Research Online (http://mcr.aacrjournals.org/). explore the regulatory link between these two DNA repair com- Corresponding Authors: Lixian Wu, Fujian Medical University, 1 Xueyuan Road, ponents. Previous work identified NF-kB as a component of the Fuzhou 350108, Fujian, China. Phone: 86-1825-9000-966; Fax: 86-591-8356-9307; DNA damage signaling pathway initiated by PARP1 and ATM that E-mail: [email protected]; and Yuanzhong Chen, Fujian Institute of Hematology, leads to IKKg phosphorylation, inhibitor kBa (IkBa) degrada- Union Hospital, Fujian Medical University, Fuzhou 350004, Fujian, China. Phone: tion, and ultimately nuclear translocation of the NF-kB subunit 86-591-8711-3828; Fax: 86-591-8711-3828; E-mail: [email protected] RELA (23–25). Conversely, here we showed that RELA knock- doi: 10.1158/1541-7786.MCR-18-0523 down causes the downregulation of PARP1 expression. Our 2018 American Association for Cancer Research. results demonstrate that NF-kB, as a critical transcription factor

www.aacrjournals.org 761

Li et al.

for the DNA damage response, can regulate PARP1 transcription Confocal microscopy analysis of gH2AX foci and thus form a feedback loop. On the basis of this mechanism, The confocal microscopy assay was performed as described we showed that combining a PARP inhibitor with an NF-kB previously (29). Briefly, after washing with ice-cold PBS, cells were inhibitor resulted in robust synergy in in vitro and in vivo AML fixed, permeabilized, and incubated with gH2AX antibodies models by decreasing DNA repair efficacy and increasing DNA (Thermo Fisher Scientific, catalog no LF-PA0025) and fluores- damage accumulation and apoptosis induction. These findings cence-labeled secondary antibodies. Immunostained cells were suggest that dual inhibition of NF-kB and PARP1 may be an examined using a Leica Laser-Scanning Microscope (TCS SP8) effective approach to increase the efficacy of DNA damage agents with a 63X/1.4 objective. Image quantification was performed such as daunorubicin by blocking DNA repair jointly in AML using ImageJ software (NIH, Bethesda, MD). therapy. Neutral comet assay Cells were processed for neutral comet assay using a Comet Materials and Methods Assay Kit from Trevigen (catalog no ADI-900-166), according the Cell culture manufacturer's protocol. Approximately 50 nucleus images per The AML cell lines KG1a and Kasumi-1 were obtained from slide were captured and processed by a Zeiss Axio Observer Z1 ATCC, where they were characterized by DNA fingerprinting, Microscope (Carl Zeiss). The tail moments from the cells were Mycoplasma detection, and cell vitality detection. These cell lines measured by CASP software. were immediately expanded and frozen. They were cultured in complete Iscove's modified Dulbecco's medium (IMDM; Gibco) qRT-PCR and RPMI1640 medium (Gibco), respectively, each medium Total RNA was extracted using BIOzol Reagent (BioFlux, cat- was supplemented with 10% FBS (Gibco) and 1% penicillin– alog no BSC52M1), and cDNA was synthesized with a PrimeScript streptomycin. The cells were maintained in a humidified incuba- RT Reagent Kit (Takara Bio, catalog no RR047A). qPCR was tor at 37 C with 5% CO2. performed in triplicate on a LightCycler 96 Real-Time PCR System (Roche, G10120-100G) with SYBR Premix Ex Taq (Takara Bio, Reagents catalog no RR420A). Relative expression was normalized to that C The IKKb inhibitor BMS-345541 (BMS) and the PARP1 inhib- of GAPDH by the 2 DD t method. The primers used in this study itor olaparib were obtained from Meilunbio Co., Ltd. BMS and are shown in Supplementary Table S1. olaparib were separately dissolved in dimethyl sulfoxide to achieve a concentration of 20 mmol/L, and then each solution Western blotting was serially diluted to specific concentrations. Daunorubicin was Western blotting was performed according to standard proto- purchased from Zhejiang Hisun Pharmaceutical Co., Ltd. cols using a Chemiluminescence Detection System from Clinx Science Instruments. The following primary antibodies were HR repair assays used: anti-RELA (Proteintech, catalog no 10745-1-AP), anti-PAR HR repair was measured in AML/DR-GFP cells, according to the (Calbiochem, catalog no AM80), and anti-PARP1 (Santa Cruz previous publications as shown in Fig. 2A (26, 27). Specifically, Biotechnology, catalog no sc-8007), as well as goat anti-rabbit 6 105 AML cells were cotransfected with 2 mg of pDR-GFP (Proteintech, catalog no SA100001-2) or anti-mouse secondary (Addgene, catalog no 46085) and 2 mg of pCBASceI (Addgene, antibodies (Proteintech, catalog no SA00001-1). Expression catalog no26477) plasmid DNA using Nucleofector solution and levels were normalized to those of the GAPDH mouse mAB the appropriate Nucleofector program. After 48 hours of trans- (Proteintech, catalog no 60004-1-Ig); these antibodies were used fection, the cells were incubated with daunorubicin for 2 hours, as internal controls. Quantification was performed with ImageJ washed, and incubated with vehicle, BMS, or olaparib for software. 12 hours. The cells were then collected and stained with 7-AAD (ClonTech) for 15 minutes at room temperature. The percentages Luciferase reporter assays of 7-AAD–negative and GFP-positive cells were analyzed by flow All transfections were carried out using Lipo2000 (Invitrogen), cytometry. according to the manufacturer's instructions. A total of 100 ng of PARP1 promoter construct, 50 ng of the expression plasmid Nonhomologous end-joining repair assays pcDNA3.1-RELA, and 20 ng of the pRL-TK Renilla luciferase vector The nonhomologous end-joining (NHEJ) assay was carried (GenePharma) were used for each transfection in 96-well plates. out using the EJ5-GFP reporter assay, according to previous Firefly luciferase and Renilla luciferase assays were performed publications (Fig. 2B; refs. 27, 28). AML cells were treated as using a Dual-Luciferase Reporter Assay System (Promega, catalog described in the HR repair assays, and the percentage of GFP- no E1910). Activity was normalized according to the Firefly/ positive cells in AML cells was analyzed after cotransfection of Renilla ratio. EJ5-GFP (Addgene, catalog no 44026) and I-SceI to measure NHEJ activity. Chromatin immunoprecipitation assay Experiments were performed using a ChIP Assay Kit according Flow cytometric analysis of DNA damage to the manufacturer's protocol (Millipore, catalog no 17-10086). Cells were incubated with 10 mg/mL anti-phospho-Histone Immunoprecipitation was conducted using an anti-NF-kB P65 H2AX (gH2AX; Ser139; BD Biosciences, catalog no 562253) antibody-ChIP Grade (Abcam, catalog no 19870) or normal conjugated to Alexa Fluor 647 for 20 minutes at room temperature rabbit IgG, followed by precipitation using protein A/G coupled and then analyzed using a flow cytometer (BD FACSCanto II). to magna beads. After decrosslinking and protease digestion, DNA Data were acquired from 10,000 gated events. fragments were recovered, purified, and used as templates for PCR

762 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

NF-kB and PARP1 Form a Loop Regulating DNA Repair

(TaKaRa, catalog no RR071A); the PARP1 promoter containing G1004) and Ki67 (1:500 dilution, Abcam, catalog no AB15580) RELA-binding sites was amplified; and the PCR products were using routine methods. Terminal deoxynucleotidyl transferase– separated in 2% agarose gels and visualized with SYBR green. The mediated dUTP nick end labeling (TUNEL) staining was assessed primers used in this study are shown in Supplementary Table S2. using a TUNEL Kit (Roche, catalog no 11684817910) according to the manufacturer's instructions. Images were recorded with a DNA pull-down assay digital microscope camera (Servicebio, catalog no ML-1600). DNA pull-down assay was performed as described previous- Image quantification was performed using ImageJ software. ly (30). Briefly, 5 mg of biotinylated double-stranded DNA probes were incubated with 500 mg of whole-cell extracts of KG1a in the Statistical analyses presence of streptavidin-conjugated agarose beads and rocked at a Values are presented as the mean SD of at least three inde- gentle speed for 2 hours at room temperature; the mixture was pendent experiments. Student t test was used to compare two isolated by centrifugation, and the proteins in the complex were groups of independent samples. One-way ANOVA analysis was dissolved and analyzed by immunoblotting with a RELA anti- used to evaluate the statistical significance of multiple groups. P < body. The DNA probes used in this study are shown in Supple- 0.05 were considered significant (); no significant difference (ns). mentary Table S3. pNFkB-MetLuc2 reporter Results AML cells were transiently transfected with a pNFkB-MetLuc2 RELA regulates DNA repair in AML cells (Clontech, catalog no 631743) construct containing an NF-kB NF-kB is a heterodimeric complex of REL family proteins (p50, promoter element upstream of the secreted Metridia luciferase p52, RELA, c-REL, and RELB) that is maintained in an inactive (MetLuc) gene, according to the AmaxaTM 4D-NucleofectorTM state in the cytoplasm through its interaction with IkB pro- protocol. Twenty-four hours after transfection, the cells were teins (31). When stimulated, NF-kB is detached from IkBs, and treated with daunorubicin for 2 hours, washed, and then incu- binds to the promoter of its target genes in the nucleus to regulate bated in medium containing BMS or olaparib alone or in com- their transcription (32, 33). bination for an additional 12 hours. The cell supernatant was To assess the role of NF-kB in DNA repair in AML cells, we harvested for secreted Metridia luciferase assays. constructed RELA-knockdown KG1a and Kasumi-1 cells using lentiviral shRNA (lentivirus target sequences are shown in Sup- Apoptosis assay plementary Table S4) and then treated cells with daunorubicin, Apoptosis was analyzed with FITC Annexin V/PI Apoptosis which is a DNA intercalating agent that causes DNA strand breaks Detection Kit I (KGA107, KeyGen Biotech). Collected cells were and a subsequent DNA damage repair response. gH2AX was used incubated with 100 mL of Annexin-V/PI staining solution for as an indicator of DNA double-strand breaks (DSB). The ratio of 15 minutes in the dark, according to the manufacturer's instruc- gH2AX increased greatly after daunorubicin treatment for 2 hours tions, and analyzed by flow cytometry. in AML cells but decreased over time after removal of the drug, while RELA knockdown caused more gH2AX accumulation Proliferation assay (Fig. 1A–E; Supplementary Fig. S1A–S1C), indicating that inhi- The viability of the treated cells was evaluated by MTT assay. biting NF-kB may impair DNA repair capacity in AML cells. To Briefly, AML cells were seeded in 96-well plates at 8 103 per well. verify the effect of RELA in DNA repair, we further constructed MTT solution (20 mL) was added to each well, and then the cells RELA-overexpressing AML cells. The results suggested that RELA were incubated at 37C for 4 hours. The absorbance was measured overexpression decreased DNA damage and promoted DNA at 570 nm on a microplate reader (Bio-Rad Laboratories, Inc.). repair, which were exactly opposite the effects of RELA knock- Each sample was analyzed in triplicate. down (Fig. 1F–J; Supplementary Fig. S1D–S1F).

In vivo study RELA regulates the DNA repair pathway at least partly through Animal experiments were carried out at the Experimental modulating PARP1 gene transcription Animal Center of Fujian Medical University (Fuzhou, China). To gain more insight into the mechanisms by which RELA The experiments were conducted in accordance with institutional regulates DSB repair, we measured HR and NHEJ repair ability via guidelines and with prior approval from the Institutional Animal DR-GFP and EJ5-GFP reporter assays, respectively (Fig. 2A and B). Care and Use Committee. A total of 40 male Balb/c athymic nude The ratio of GFP-positive cells in RELA-knockdown AML cells mice aged 5 weeks were purchased from the Slack Laboratory cotransfected with the DR-GFP/I-SceI or EJ5-GFP/I-SceI plasmids Animals Co., Ltd. (Shanghai, China). KG1a cells (5 106)in was significantly decreased (Fig. 2C–D; Supplementary Fig. S2A– medium containing 1:1 Matrigel (BD Biosciences#356234) were S2B). Then, we examined the level of PARylation after DNA injected subcutaneously into the right flank. Two weeks after damage. The results showed that PARylation was also obviously injection, the tumor-forming mice were randomly divided into reduced in RELA-knockdown AML cells (Fig. 2E; Supplementary five groups. Tumor volume was measured and calculated using Fig. S2C). Although there have been some reports that PARP1 the following formula: [(width)2 (height)]/2. Survival times regulates NF-kB activity through PAR activation during DNA were observed and analyzed using the Kaplan–Meier method and damage (24, 25), our results demonstrated beyond our expecta- log-rank test (GraphPad Prism software). tions that NF-kB can conversely regulate PAR activation. Accumulating evidence suggests that PARP1 is involved in Histologic examination several DNA repair pathways, including HR, NHEJ, and BER (34); Paraformaldehyde-fixed and paraffin-embedded tumor sec- moreover, because RELA is a transcription factor, we focused tions were stained with hematoxylin (Servicebio, catalog no on the precise role of RELA in regulating PARP1 transcription.

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 763

Li et al.

KG1α KG1α A B 60 Ctrl Damage Repair shCtrl γH2AX γH2AX γH2AX shRELA#1 5.14% 41.4% 20.1% 40 shRELA#2 shCtrl

γ H2AX (%) 20 ns ns

γH2AX γH2AX γH2AX 8.61% 48.8% 34.7% 0

Vehicle Repair Damage

shRELA#1 α C KG1 γH2AX γH2AX γH2AX RELA 65 kDa 8.00% 47.8% 27.8%

GAPDH 37 kDa Figure 1. shRELA#2

APC-A- γ H2AX RELA regulates DNA repair in AML cells. shCtrl shCtrl and two shRELA KG1a cells were FSC-A shRELA#1 shRELA#2 treated with medium as a control (Ctrl); with 0.5 mmol/L daunorubicin for 2 hours D KG1α as the damage group; and with medium Ctrl Damage Repair for an additional 12 hours after daunorubicin treatment as the repair E shCtrl group. A, Representative images of flow shRELA#1 cytometric analyses of shCtrl and shRELA shRELA#2 shCtrl α γ H2AX DAPI KG1 KG1a cells. B, The ratio of gH2AX in shCtrl 60 and shRELA KG1a cells from three independent experiments was quantified. 40 C, The efficiency of shRNAs was measured in KG1a cells by Western fl shRELA#1 blotting. D, Immuno uorescence 20 micrographs of gH2AX foci in shCtrl and Intensity of γ H2AX ns ns two shRELA KG1a cells. E, The intensity of 0 gH2AX in at least 50 shCtrl and shRELA KG1a cells quantified by ImageJ software. Vehicle Repair shRELA#2 Damage Vector- and RELA-overexpressing KG1a cells were treated as above. F, Representative images of flow cytometric KG1α analyses in vector- and RELA- KG1α G 60 F Vector overexpressing KG1a cells. G, The ratio of Ctrl Damage Repair RELA gH2AX in vector- and RELA- γH2AX γH2AX γH2AX 40 overexpressing KG1 cells from three 4.73% 44.2% 25.2% a independent experiments was quantified. fi Vector 20 H, The ef ciency of RELA overexpression γ H2AX (%) ns was measured in KG1a cells by Western 0 blotting. I, Immunofluorescence γH2AX γH2AX γH2AX micrographs of gH2AX foci in vector- and 6.43% 38.4% 11.0% Vehicle Repair Damage RELA-overexpressing KG1a cells. J, The KG1α intensity of gH2AX in at least 50 vector- RELA H and RELA-overexpressing KG1a cells RELA 65 kDa

APC-A- γ H2AX quantiﬁed by ImageJ software. Mean FSC-A SD of three independent experiments are GAPDH 37 kDa shown ( , P < 0.05).

KG1α I Vector RELA Ctrl Damage Repair J Vector RELA KG1α 60

γ H2AX DAPI 40

20 ns Intensity of γ H2AX 0 RELA Vector

Vehicle Repair Damage

RT-PCR and Western blotting results showed that PARP1 mRNA using a putative luciferase construct containing the first 2 kb of the and protein levels were significantly downregulated in RELA- PARP1 50-flanking region in the pGL3-Basic reporter vector. The knockdown AML cells, but upregulated in RELA-overexpressing results showed that transfection of RELA dramatically enhanced AML cells (Fig. 2F–H). Furthermore, we confirmed these findings the activation of the PARP1 reporter, indicating that PARP1 may

764 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

NF-kB and PARP1 Form a Loop Regulating DNA Repair

pDR-GFP A SceGFP iGFP CDα E KG1 KG1α KG1α shCtrl shRELA#1 I-Scel 5 I-Scel plasmid shCtrl GFP+ GFP+ HR 4.94% 2.41% 4 shRELA

3 PAR

GFP HR Repair cells (%)

+ 2 EJ5-GFP GFP+ GFP+ 1

B GFP 4.23% 2.14% GAPDH puro GFP 0 I-Scel I-Scel shCtrl ATAA GFP

TATT FITC-A-GFP NHEJ Repair HR Repair shRELA#1 NHEJ Repair NHEJ FSC-A

GFP H F G PARP1 116 kDa

KG1 α GAPDH 37 kDa 1.2 KG1α 2.5 KG1α 1.0 Kasumi-1 2.0 Kasumi-1 0.8 PARP1 116 kDa 0.6 1.5 mRNA expression 1.0 0.4 mRNA expression GAPDH 37 kDa

0.2 0.5 Kasumi-1 PARP1 0.0 0.0

Relative Relative RELA RELA shCtrl Vector RELA shCtrl shCtrl Vector Vector Relative PARP1Relative shRELA#1 shRELA#2 shRELA#1shRELA#2 shRELA#1shRELA#2

IJ L 8 KG1α Kasumi-1

6 IgG RELA Input IgG RELA Input

2 promoter-2 kb PARP1

Relative luciferase activity luciferase Relative 0

pcDNA3.1 RELA CDS

K293T M pGL3-Basic Luc KG1α Kasumi-1 Promoter-2kb Luc Probe Ctrl siteI siteII Ctrl siteI siteII site I -Del Luc

site II -Del Luc RELA

site I/II -Del Luc

0 2 4 6 Relative luciferase activity

Figure 2. RELA regulates the DNA repair pathway at least partly through modulating PARP1 gene transcription. A and B, Diagrams illustrating the HR/DR-GFP and NHEJ/ EJ5-GFP reporter assays, which are cited from Lixian Wu, Radiat Res. doi: 10.1667/RR3034.1. GFP-positive KG1a cells in which I-SceI-generated DSBs repaired by HR or NHEJ were detected by flow cytometry (C), with quantification from three independent experiments (D). E, PAR expression detected by Western blotting in shCtrl and shRELA KG1a cells. F and G, Validation of PARP1 transcription by qRT-PCR in RELA-knockdown (I) and RELA-overexpressing (J) AML cells. H, Western blotting validation of PARP1 expression in RELA-knockdown and RELA-overexpressing AML cells. I, 293T cells were transfected with a RELA expression plasmid in the presence of the PARP1-2.0 kb promoter construct. At 48 hours posttransfection, luciferase activities were determined and are expressed relative to the control (pcDNA3.1). J, A computer-based search for the first 2 kb upstream of PARP1 resulted in the identification of two potential RELA-binding sites. K, 293T cells were transfected with a RELA expression plasmid in the presence of the PARP1-2.0 kb promoter construct or the same construct with a mutation in RELA-binding site I or II. At 48 hours posttransfection, luciferase activities were determined and are expressed relative to the control (pGL3-Basic). L, A ChIP assay was performed in AML cells to analyze the interactions of RELA with the PARP1 promoter covering two RELA-binding sites. M, Biotin-labeled 24-mer PARP1 promoter fragments containing site I or II and a negative control oligonucleotide containing no banding site were used to illustrate the specific binding of RELA to this regulatory element in AML cells. Mean SD of three independent experiments are shown (, P < 0.05). be one of the target genes regulated by RELA, which has not been bioinformatic prediction tool Jaspar (http://jaspar.genereg.net/). identified previously (Fig. 2I). We found two potential RELA-binding sites, located at 35/44 To elucidate the underlying molecular basis of this pathway, we bp and 105/114 bp, which were designated sites I and II, identified RELA-binding motifs in the PARP1 promoter using the respectively (Fig. 2J). To investigate the roles of these two putative

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 765

Li et al.

KG1α A CDα KG1 KG1α 80 shCtrl Vehicle Damage Repair shPARP1 γH2AX γH2AX γH2AX 60 shPARP1/RELA 4.19% 37.5% 12.2% 40 H2AX (%) γ shCtrl 20 ns ns 0.0 0.5 1.0 1.5 Relative luciferase activity 0

shCtrl γH2AX γH2AX γH2AX 5.67% 45.0% 24.2% Vehicle Repair shPARP1#1shPARP1#2 Damage

Kasumi-1 E Kasumi-1

shPARP1 80 B shCtrl shPARP1 60 shPARP1/RELA γH2AX γH2AX γH2AX 6.78% 62.1% 41.3% 40 H2AX γ H2AX (%)

γ 20 ns ns 0 APC- A - APC- A 0.0 0.5 1.0 1.5 1.0 0.0 0.5 Relative luciferase activity shPARP1/RELA FSC-A shCtrl Vehicle Repair Damage shPARP1#1shPARP1#2

Figure 3. RELA and PARP1 form a positive feedback loop to regulate DNA damage repair. A and B, NF-kB activity at 12 hours post-daunorubicin in PARP1-knockdown KG1a and Kasumi-1 cells was examined using secreted Metridia luciferase assays. C–E, shCtrl, shPARP1 and shPARP1/RELA KG1a and Kasumi-1 cells were treated as described in the legend of Fig. 1. Representative flow cytometric analyses of DSBs in KG1a cells (C) and quantification from three independent experiments (D). Quantification of gH2AX in Kasumi-1 cells from three independent experiments (E). Mean SD of three independent experiments are shown (, P < 0.05).

RELA-response elements in the RELA-mediated transactivation of single target (Fig. 3C–E). These results indicated that dually the PARP1 promoter, we deleted these two sites from the PARP1-2 targeting this positive feedback loop with simultaneous NF-kB kb promoter reporter plasmid and then cotransfected the plasmid and PARP1 inhibition would generate more pronounced thera- with the RELA expression plasmid into 293T cells. We found that peutic effects than single inhibition. deleting either site I or site II significantly attenuated transcriptional activity (Fig. 2K). A ChIP assay was performed to further Pharmacologic inhibition of both NF-kB and PARP1 blocks investigate whether RELA regulates PARP1 gene expression DNA repair, increasing DNA damage accumulation in AML cells through directly binding to the PARP1 promoter. The analysis To verify the mechanism of positive feedback between NF-kB/ of PCR products showed that RELA was able to bind to the RELA and PARP1, as well as the possibility of clinical translation, promoter region of PARP1, covering two RELA-binding sites we used BMS and olaparib, which are small-molecule inhibitors (Fig. 2L). Moreover, we assessed the potential physiologic recruit- of IKKb and PARP1/2, respectively (35, 36). We found that NF-kB ment of RELA to the PARP1 promoter using DNA pull-down was activated by daunorubicin-induced DNA damage; BMS obvi- assays. As shown in Fig. 2M, the endogenous RELA protein in AML ously inhibited NF-kB activation; and the inhibitory effect of BMS cells was efficiently recruited to binding sites I and II. Together, was potentiated in the presence of olaparib (Fig. 4A and B). We these data suggest that RELA regulates the DNA repair pathway at further investigated the interaction of these pathways in DNA least partly through modulating the transcriptional activity of the damage repair. The results showed that the PARylation and HR PARP1 gene by binding to its promoter. pathways were activated by daunorubicin in KG1a cells, which might be one of mechanisms underlying the resistance of AML RELA and PARP1 form a positive feedback loop to regulate DNA cells to daunorubicin (Fig. 4C–F). Olaparib clearly inhibited damage repair PARylation (Fig. 4C) but slightly increased daunorubicin-induced To test whether PARP1 plays a critical role in NF-kB activation HR activation, which was significantly inhibited by BMS (Fig. 4D after DNA damage in AML cells, as reported in other cell types and E). In addition to HR, BMS also suppressed PARylation and (23–25), we knocked down PAPR1 (lentivirus target sequences NHEJ to some extent (Fig. 4C, D, and F). Similar to the results are shown in Supplementary Table S4) in KG1a and Kasumi-1 obtained in KG1a cells, daunorubicin activated the PARylation cells and then investigated the effect on NF-kB activity by exam- and HR repair pathways (Supplementary Fig. S3A and S3B), and ining the expression of the pNFkB–MetLuc2 reporter after DNA BMS suppressed PARylation, HR, and NHEJ in Kasumi-1 cells damage. Our results showed that PARP1 knockdown indeed (Supplementary Fig. S3A–S3C); however, in contrast to the results reduced NF-kB activity during DNA repair (Fig. 3A and B), which obtained in KG1a cells, olaparib decreased HR in Kasumi-1 cells was consistent with previous reports. Therefore, NF-kB and (Supplementary Fig. S3B). PARP1 interfere with each other, which indicate that these two Next, we monitored the clearance of gH2AX in KG1a cells. components form a closed loop in DNA repair in AML cells. Next, More gH2AX remained when cells were treated with BMS or we knocked down RELA in PARP1-knockdown AML cells and olaparib during the repair process, and most of the gH2AX found that double-knockdown of these two targets caused more remained following treatment with the combination of olaparib serious DNA damage accumulation than knockdown of either and BMS (Fig. 5A–D). Treatment with BMS and/or olaparib

766 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

NF-kB and PARP1 Form a Loop Regulating DNA Repair

A BC KG1α Kasumi-1 KG1α 12 h post-DNR

3 3 Vehicle Damage Ctrl BMS Ola BMS+Ola

2 2 PAR 1 1

0 0 Relative luciferase activity Relative luciferase activity

Ola Ctrl Ola Ctrl BMS BMS GAPDH Vehicle Vehicle BMS+Ola BMS+Ola 12 h post-DNR 12 h post-DNR

DEα 12 h post-DNR F KG1 KG1α HR Repair KG1α NHEJ Repair Vehicle Ctrl BMS Ola BMS+Ola 8 5 GFP+ GFP+ GFP+ GFP+ GFP+ ns 4.15% 5.37% 2.52% 6.42% 2.78% 6 4 ns 3

HR Repair 4 Cells (%) Cells (%)

+ + 2 2 + + + + + 1 GFP GFP GFP GFP GFP GFP GFP 2.67% 3.28% 1.51% 3.68% 1.73% 0 0

Ctrl Ola Ctrl Ola BMS BMS Vehicle Vehicle FITC-A-GFP BMS+Ola BMS+Ola NHEJ Repair 12 h post-DNR 12 h post-DNR FSC-A

Figure 4. Pharmacologic inhibition of both NF-kB and PARP1 blocks DNA repair in AML cells. KG1a cells were treated with 10 mmol/L BMS or/and 20 mmol/L olaparib for 12 hours after 0.5 mmol/L daunorubicin (DNR) washout. Vehicle cells were incubated with 0.1% w/v DMSO alone. Control cells were incubated with 0.1% w/v DMSO for 12 hours after daunorubicin washout. Cells in the damage group were treated with daunorubicin for 2 hours. A and B, NF-kB activity was examined using secreted Metridia luciferase assays. C, PARylation in KG1a cells with PAR expression detected by western blotting. GFP-positive cells among KG1a cells repaired by HR or NHEJ were detected by flow cytometry (D); data were quantified from three independent experiments (E and F). Mean SD of three independent experiments are shown (, P < 0.05). without daunorubicin pretreatment caused negligible DSBs in Double inhibition of NF-kB and PARP1 increases the efficacy AML cells (data not shown). These findings were validated by the of daunorubicin in vivo neutral comet assay based on the tail moment, which is an On the basis of the effects observed in vitro, we next indicator of DNA damage (Fig. 5E and F). Treatment of addressed whether double inhibition of NF-kBandPARP1 Kasumi-1 cells produced similar results (Fig. 5G and H). Collec- would increase the efficacy of daunorubicin in vivo.AKG1a tively, these observations indicated that pharmacologic inhibition xenograft mouse model was established via subcutaneous of both NF-kB and PARP1 blocks DNA repair. injection. Two weeks after tumor cell injection, the tumor- forming mice were randomized into five groups to receive Double inhibition of NF-kB and PARP1 increases the efficacy treatment with vehicle, daunorubicin, BMS, and olaparib of daunorubicin in AML cells alone or in combination every other day until six doses were To determine whether double inhibition of NF-kB and PARP1 administered, as depicted in Fig. 7A. The tumor volumes were influences the tumor response to daunorubicin in vitro, cell measured every day during treatment. We found that the apoptosis and proliferation were measured in AML cells. Because KG1a xenograft tumors grew rapidly without drug treatment. NF-kB plays an important antiapoptotic role, we speculated that Daunorubicin mono-treatment and daunorubicin and BMS or BMS, which is an upstream inhibitor of NF-kB, would induce olaparib combination treatment decreased tumor growth apoptosis and sensitize cells to chemotherapy. The results showed slightly; in contrast, the combination of daunorubicin, ola- that both BMS and olaparib mono-treatment increased apoptosis parib, and BMS significantly decreased the tumor volume induction by daunorubicin, and combination treatment further (Fig. 7A). Survival time analysis showed that daunorubicin increased the apoptosis ratio at 36 hours after the induction of mono-treatment had almost no effect on median survival DSBs (Fig. 6A–D). We then determined whether BMS and ola- times (34 vs. 29.5 days; P>0.05), while combination treatment parib increased the efficacy of daunorubicin in inhibiting prolif- with BMS (46 vs. 34 days) or olaparib (44 vs. 34 days) eration. We treated cells with different concentrations of dauno- increased the efficacy of daunorubicin to some extent, and rubicin, either alone or in combination with BMS and olaparib, the simultaneous combination of all three drugs significantly for 72 hours. Both BMS and olaparib mono-treatment increased extended mouse survival times compared with the combina- the inhibition of proliferation by daunorubicin, and the combi- tion of only two drugs (62 vs. 46/44 days; Fig. 7B). In nation treatment further increased the inhibition ratio (Fig. 6E addition, the treatment caused no obvious systemic toxicity, and F). The cytotoxicity of the single drug in AML cells is shown in as assessed by weight loss (Fig. 7C). We performed TUNEL Supplementary Fig. S4. and Ki67 assays of xenograft tumor tissues to measure the

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 767

Li et al.

A B KG1α 50 KG1α 12 h post-DNR 40 Vehicle Damage Ctrl BMS Ola BMS+Ola 30 γH2AX γH2AX γH2AX γH2AX γH2AX γH2AX 2.11% 39.0% 13.3% 21.6% 19.5% 27.1% 20 r-H2AX% 10 H2AX 0

Ctrl Ola BMS

APC-A- γ Vehicle Damage FSC-A BMS+Ola 12 h post-DNR

C D KG1α KG1α 12 h post-DNR 80 Vehicle Damage Ctrl BMS Ola BMS+Ola 60 40

H2AX 20 γ

Intensity of γ H2AX 0

Ctrl Ola

DAPI BMS Vehicle Damage BMS+Ola 12 h post-DNR

12 h post-DNR KG1α EFKG1α 30

Vehicle Damage Ctrl BMS Ola BMS+Ola 20

10 Tail moment Tail

Ctrl Ola BMS Vehicle Damage BMS+Ola 12 h post-DNR

Kasumi-1 12 h post-DNR Kasumi-1 G H 80 Vehicle Damage Ctrl BMS Ola BMS+Ola 60 γH2AX γH2AX γH2AX γH2AX γH2AX γH2AX 4.34% 55.9% 21.3% 33.9% 30.8% 42.8% 40 H2AX r-H2AX% 20

APC-A- γ Ctrl Ola FSC-A BMS Vehicle Damage BMS+Ola 12 h post-DNR

Figure 5. Pharmacologic inhibition of both NF-kB and PARP1 increases DNA damage accumulation in AML cells. Cells were treated as described in the legend of Fig. 4. DNA damage in KG1a cells was determined by ﬂow cytometry (A and B), confocal microscopy analysis (C and D), and the neutral comet assay (E and F). G and H, DNA damage in Kasumi-1 cells was determined by ﬂow cytometry. Mean SD of three independent experiments are shown (, P < 0.05).

apoptosis and proliferation of AML cells in the xenograft Discussion tumors; the results shown in Fig. 7D–F suggested that com- In this study, we provide evidence that RELA can regulate DNA bination treatment with BMS or olaparib increased the efficacy repair after treatment with a DNA-damaging agent, and how the of daunorubicin in proapoptosis and proliferation inhibition underlying mechanism regulates this pathway was demonstrated to some extent, while the simultaneous combination of all by us. We found that RELA regulates DNA repair by binding to the three drugs significantly increased the efficacy of the combi- promoter region of the PARP1 gene and affecting PARP1 gene nation of two drugs, similar to the results obtained in cells. transcription (Fig. 2). RELA knockdown or NF-kB inhibition Comprehensive statistical analysis of these data illustrated reduced PARylation in daunorubicin-damaged AML cells. Con- that the combination of three drugs caused significant tumor versely, PARP1 knockdown reduced NF-kB activity, indicating regression. Together, double inhibition of NF-kBandPARP1 that NF-kB and PARP1 create a positive feedback loop in DNA significantly increased the efficacy of daunorubicin in vivo.

768 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

NF-kB and PARP1 Form a Loop Regulating DNA Repair

A Ctrl BMS Ola BMS+Ola C KG1α 60 No DNR 36 h post-DNR No DNR 40

KG1 α 20 Apoptosis %

PI Ctrl Ola Ola BMS Ctrl BMS 36 h post-DNR Ola+BMS Ola+BMS FITC-Annexin V

B Ctrl BMS Ola BMS+Ola D Kasumi-1 60 No DNR 36 h post-DNR No DNR 40

20 Kasumi-1 Apoptosis %

PI Ola Ctrl Ola Ctrl BMS BMS 36 h post-DNR Ola+BMS Ola+BMS FITC-Annexin V

KG1α E 100 F 100 Kasumi-1 DNR DNR 80 DNR+5 μmol/L BMS 80 DNR+5 μmol/L BMS DNR+20 μmol/L Ola μ 60 DNR+20 mol/L Ola DNR+BMS+Ola 60 DNR+BMS+Ola 40 40 20 20

Inhibition ratio % 0 Inhibition ratio % 0 0.000 0.025 0.050 0.075 0.100 0.000 0.025 0.050 0.075 0.100 μ DNR Concentration ( mol/L) DNR Concentration (μmol/L)

Figure 6. Double inhibition of NF-kB and PARP1 increases the efficacy of daunorubicin (DNR) in AML cells. A and B, KG1a cells were treated with 5 mmol/L BMS or/and 20 mmol/L olaparib for 36 hours after 0.5 mmol/L daunorubicin damage. The Control cells were incubated with 0.1% w/v DMSO alone for 36 hours after daunorubicin washout; apoptosis was determined through flow cytometric measurement of cells stained with propidium iodide and Annexin V. C and D, Kasumi-1 cells were treated with 5 mmol/L BMS or/and 20 mmol/L olaparib for 36 hours after 0.25 mmol/L daunorubicin damage. Representative images of the flow cytometric analyses and quantification are shown. E and F, KG1a and Kasumi-1 cells were treated with different concentrations of daunorubicin, either alone or in combination with 5 mmol/L BMS and 20 mmol/L olaparib for 72 hours, and cell viability was evaluated via the MTT assay (, P < 0.05). repair. This new loop complements to the DNA damage and therapy. In addition, the dose of NF-kB inhibitors can be reduced repair mechanism of NF-kB upon exposure to a DNA-damaging in combination treatments, thereby minimizing their side effects agent. What we have found in human AML cell lines in this study on other physiologic processes. Moreover, the baseline DNA provided an insight into improving treatment for AML. repair activity varies with tumor type. Therefore, it is vital to NF-kB has a variety of physiologic functions, including roles in understand the intrinsic DNA repair activity of each tumor type immune and inflammatory responses, oncogenesis, and prior to DNA-damaging agent treatment. For instance, the defi- responses to genotoxic stress (37–39), as a result, it might cause cient HR pathway in BRCA1/2-deficient breast or ovarian cancer severe adverse effects like immunodeficiency if it was inhibited creates a distinct vulnerability to PARP inhibitions (41, 42). AML nonselectively (40). Thus, selective inhibition of NF-kB is expect- cells are more dependent than another tumor cells on NF-kB, and ed to effectively reduce resistance to genotoxic treatments with here, we clarified that NF-kB plays a critical role in DNA repair in minimal toxicity. PARP1 inhibitors can selectively inhibit NF-kB AML cells (Figs. 1 and 2A–E). BMS is a potent and specific activation by radiation and chemotherapeutics, and the inhibi- inhibitor of IKKb that binds to an allosteric site on the IKKb tory effect will be amplified by the positive feedback loop between catalytic subunit and inhibits IkBa phosphorylation and degra- NF-kB and PARP1 observed in this study. Therefore, when used in dation, thereby inhibiting NF-kB activation (35). The results from combination with PARP1 inhibitors, NF-kB inhibitors may resen- our current study revealed that BMS inhibits PARylation, HR, and sitize treatment-refractory cancer cells to conventional chemo- NHEJ in daunorubicin-damaged AML cells (Fig. 4C–F) and

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 769

Li et al.

A 3,200 Ctrl DNR D 2,800 DNR+BMS TUNEL Ki67 DNR+Ola 50 μ m E ) DNR+BMS+Ola 3 2,400

2,000

Ctrl 40 1,600

1,200 30 Tumor volume (mm Tumor 800 20 cells (%) 400 10 DNR TUNEL-positive 0 13 14 15 16 17 18 19 20 21 22 23 24 25 26 0 Drug treated Ctrl Days DNR DNR+Ola B Survival proportions DNR+BMS 100 Ctrl DNR+BMS+Ola DNR 80 DNR+BMS DNR+Ola DNR+BMS 60 DNR+BMS+Ola F

40 100 20 Percent survival

0 80 0 10 20 30 40 50 60 70 60 Time (days) DNR+Ola 40 C Ctrl cells (%) DNR 1.2 Ki67-positive DNR+BMS 20 DNR+Ola 1.1 DNR+BMS+Ola 0

1.0 Ctrl DNR

DNR+BMS+Ola DNR+Ola 0.9 DNR+BMS DNR+BMS+Ola

Weight change from day 14 Weight 0.8 14 16 18 20 22 24 26 Days after cells injection

Figure 7. Pharmacologic inhibition of NF-kB and PARP1 increases the efﬁcacy of daunorubicin (DNR) in vivo. A, Growth curves of KG1a-bearing mouse xenografts after six rounds of treatment with PBS, daunorubicin (0.6 mg/kg daunorubicin, i.v., every other day), daunorubicinþBMS (0.6 mg/kg daunorubicin, i.v., 20 mg/kg BMS, i.g., every other day), daunorubicinþ olaparib (0.6 mg/kg daunorubicin, i.v., 25 mg/kg olaparib, i.p., every other day), or daunorubicin þ BMS þ olaparib (0.6 mg/kg daunorubicin, i.v., 20 mg/kg BMS, i.g., and 25 mg/kg olaparib, i.p., every other day) on days 14, 16, 18, 20, 22, and 24 after the tumors were established in the mice. The data are presented as the mean SD of the tumor volumes from six tumors per group. B, Kaplan–Meier curves for KG1a-bearing mouse xenografts after six treatments. C, Relative body weights of each group of recipient mice during the treatment. D, IHC staining (TUNEL and Ki67) of tumor tissues. E and F, Quantitative analysis of the percentage of TUNEL- and Ki67-positive cells in tumor tissues (, P < 0.05).

decreases the repair of daunorubicin-induced DSBs (Fig. 5), caus- and/or NHEJ pathways to allow cell survival (44), the increased ing cell apoptosis (Fig. 6). DSBs will activate HR, which may lead to olaparib resistance. Exploration of the DNA repair mechanisms and pathways is Thus, the approach of combining DNA damage agents, a PARP1 likely to provide favorable strategies for sensitizing tumor cells to inhibitor, and an NF-kB inhibitor may block DNA repair more drug therapy. Intracellular repair mechanisms are quite complex, robustly than any combination of two drugs, and as a result, drug one disordered repair pathway may cause compensatory activa- resistance is less likely to occur. Our results demonstrated that the tion of other repair pathways, leading to incomplete inhibition of combination of three drugs simultaneously inhibits PARylation, DNA repair. Thus, only the simultaneous suppression of multiple HR, and NHEJ (Fig. 4C–F), effectively decreasing DNA repair, pathways can achieve better therapeutic results. Our results increasing DNA damage accumulation (Fig. 5), and inducing showed that olaparib inhibits PARylation, but slightly activates apoptosis, thereby resulting in reduced AML cells proliferation the HR repair pathway in AML cells with daunorubicin damage (Fig. 6), delayed disease progression, and increased survival in an (Fig. 4C–E). Possible reasons for these ﬁndings are as follows: AML xenograft tumor model (Fig. 7). olaparib impairs BER-mediated single-strand break (SSB) repair, This study on DNA repair regulation by PARP1 and NF-kBin and the accumulated SSBs are converted to DSBs during DNA AML cells has signiﬁcantly improved our understanding of the replication (43); because DSBs are primarily repaired via the HR molecular signaling induced by genotoxic agents. This study has

770 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

NF-kB and PARP1 Form a Loop Regulating DNA Repair

also provided a promising drug combination approach for selec- Writing, review, and/or revision of the manuscript: D. Li, L. Wu tively inhibiting NF-kB activation by chemotherapeutics and Administrative, technical, or material support (i.e., reporting or organizing DNA damage in AML cells, and this combination may resensitize data, constructing databases): D. Li, L. Wu Study supervision: Y. Chen, L. Wu treatment-refractory AML cells or leukemia-initiating cells to fi conventional chemotherapy. Patients with AML may bene t from Acknowledgments these cost-effective "old" chemotherapeutic drugs. Thus, system- The authors gratefully acknowledge the support of this project by National atically testing whether the triple combination approach exhibits Natural Science Foundation of China (81872898, 81273541, and 81572662), synergistic therapeutic effects in the clinic may be worthwhile. We Joint Funds for the Innovation of Science and Technology, Fujian Province hope that this innovative approach will lead to more complete (2016Y9057), Joint Research Program of Health and Education of Fujian clinical remissions and improved patient outcomes. Province (WKJ2016-2-33), University Industry Cooperation Project of Fujian Province (2016Y4005), Natural Science Foundation of Fujian Province fi Disclosure of Potential Conflicts of Interest (2018J01842 and 2017J01823), and Startup Fund for Scienti c Research, Fujian Medical University (2016QH014 and 2017XQ2019). No potential conflicts of interest were disclosed.

The costs of publication of this article were defrayed in part by the Authors' Contributions payment of page charges. This article must therefore be hereby marked Conception and design: D. Li, J. Xu, Y. Chen, L. Wu advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate Development of methodology: D. Li, Y. Luo, X. Chen, Y. Zhuang this fact. Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D. Li, Y. Luo, L. Zhang, T. Wang Analysis and interpretation of data (e.g., statistical analysis, biostatistics, Received May 21, 2018; revised September 20, 2018; accepted December 12, computational analysis): D. Li, Y. Luo, Y. Fan 2018; published ﬁrst December 17, 2018.

References 1. Dohner€ H, Estey E, Grimwade D, Amadori S, Appelbaum FR, Buchner€ T, suppressing the myeloproliferative leukemia virus oncogene. Oncotarget et al. Diagnosis and management of AML in adults: 2017 ELN recommen- 2015;6:27490–504. dations from an international expert panel. Blood 2017;129:424–47. 16. Pratz KW, Rudek MA, Gojo I, Litzow MR, Mcdevitt MA, Ji JJ, et al. A phase I 2. Saultz JN, Garzon R. Acute myeloid leukemia: a concise review. J Clin Med study of topotecan, carboplatin and the PARP inhibitor veliparib in acute 2016;5:33. leukemias, aggressive myeloproliferative neoplasms and chronic myelo- 3. Frosina G. DNA repair and resistance of gliomas to chemotherapy and monocytic leukemia. Clin Cancer Res 2016;23:899–907. radiotherapy. Mol Cancer Res 2009;7:989–9. 17. Wahner AH, Menefee ME, Hartmann LC, Long HJ, Northfelt DW, Reid JM, 4. Lu Y, Liu Y, Pang Y, Pacak K, Yang C. Double-barreled gun: combination of et al. A phase I clinical trial of the poly(ADP-ribose) polymerase inhibitor PARP inhibitor with conventional chemotherapy. Pharmacol Thera 2018; veliparib and weekly topotecan in patients with solid tumors. 188:168–75. Clin Cancer Res 2018;24:744–52. 5. Kagoya Y, Yoshimi A, Kataoka K, Nakagawa M, Kumano K, Arai S, et al. 18. Bryant HE, Schultz N, Thomas HD, Parker KM, Flower D, Lopez E, et al. Positive feedback between NFkB and TNF-a promotes leukemia-initiating Speciﬁc killing of BRCA2-deﬁcient tumours with inhibitors of poly(ADP- cell capacity. J Clin Invest 2014;124:528–42. ribose) polymerase. Nature 2005;434:913–7. 6. Mccool KW, Miyamoto S. DNA damage-dependent NFkBactivation: 19. Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB, et al. NEMO turns nuclear signaling inside out. Immunol Rev 2012;246: Targeting the DNA repair defect in BRCA mutant cells as a therapeutic 311–26. strategy. Nature 2005;434:917–21. 7. Napetschnig J, Wu H. Molecular basis of NFkB signaling. Annu Rev Biophys 20. Swisher EM, Sakai W, Karlan BY, Wurz K, Urban N, Taniguchi T. Secondary 2013;42:443–68. BRCA1 mutations in BRCA1-mutated ovarian carcinomas with platinum 8. Wu ZH, Miyamoto S. Many faces of NF-kappaB signaling induced by resistance. Cancer Res 2008;68:2581–6. genotoxic stress. J Mol Med 2007;85:1187–202. 21. Fojo T, Bates S. Mechanisms of resistance to PARP inhibitors–three and 9. Kraft D, Rall M, Volcic M, Metzler E, Groo A, Stahl A, et al. NFkB-dependent counting. Cancer Discov 2013;3:20–3. DNA damage-signaling differentially regulates DNA double-strand break 22. Ricks TK, Chiu HJ, Ison G, Kim G, McKee AE, Kluetz P, et al. Successes and repair mechanisms in immature and mature human hematopoietic cells. challenges of PARP inhibitors in cancer therapy. Front Oncol 2015; Leukemia 2015;29:1543–54. 5:222. 10. Zhou J, Ying QC, Chng WJ. Aberrant nuclear factor-kappa B activity in acute 23. Wu ZH, Shi Y, Tibbetts RS, Miyamoto S. Molecular linkage between the myeloid Leukemia: from molecular pathogenesis to therapeutic target. kinase ATM and NFkB signaling in response to genotoxic stimuli. Science Oncotarget 2015;6:5490–500. 2006;311:1141–6. 11. Nieborowskaskorska M, Sullivan K, Dasgupta Y, Podszywalowbartnicka P, 24. McCool K, Miyamoto S. A PAR-SUMOnious mechanism of NEMO acti- Hoser G, Maifrede S, et al. Gene expression and mutation-guided synthetic vation. Mol Cell 2009;36:349–50. lethality eradicates proliferating and quiescent leukemia cells. J Clin Invest 25. Stilmann M, Hinz M, Arslan SC, Zimmer A, Schreiber V, Scheidereit C. A 2017;127:2392–406. nuclear poly(ADP-ribose)-dependent signalosome confers DNA damage- 12. Jacoby MA, De Jesus Pizarro RE, Shao J, Koboldt DC, Fulton RS, Zhou G, induced IkappaB kinase activation. Mol Cell 2009;36:365–78. et al. The DNA double-strand break response is abnormal in myeloblasts 26. Pierce AJ, Johnson RD, Thompson LH, Jasin M. XRCC3 promotes homol- from patients with therapy-related acute myeloid leukemia. Leukemia ogy-directed repair of DNA damage in mammalian cells. Gen Develop 2014;28:1242–51. 1999;13:2633–8. 13. Langelier MF, Pascal JM. PARP-1 mechanism for coupling DNA damage 27. Wu L, Zhou D. BMS-345541 sensitizes MCF-7 breast cancer cells to ionizing detection to poly(ADP-ribose) synthesis. Curr Opin Struct Biol 2013; radiation by selective inhibition of homologous recombinational repair of 23:134. DNA double-strand breaks. Radiat Res 2013;179:160–70. 14. Jang NY, Kim DH, Cho BJ, Choi EJ, Lee JS, Wu HG, et al. Radiosensitization 28. Bennardo N, Cheng A, Huang N, Stark JM. Alternative-NHEJ Is a mech- with combined use of olaparib and PI-103 in triple-negative breast cancer. anistically distinct pathway of mammalian chromosome break repair. BMC Cancer 2015;15:89. PLoS Genet 2008;4:e1000110. 15. Wang L, Cai W, Zhang W, Chen X, Dong W, Tang D, et al. Inhibition of poly 29. Du Y, Shen J, Hsu JL, Han Z, Hsu MC, Yang CC, et al. Syntaxin (ADP-ribose) polymerase 1 protects against acute myeloid leukemia by 6-mediated Golgi translocation plays an important role in nuclear

www.aacrjournals.org Mol Cancer Res; 17(3) March 2019 771

Li et al.

functions of EGFR through microtubule-dependent trafficking. Oncogene 39. Baeuerle PA, Baltimore D. I kappa B: a specific inhibitor of the NF-kappa B 2014;33:756–70. transcription factor. Science 1988;242:540–6. 30. Wu KK. Analysis of protein-DNA binding by streptavidin-agarose pull- 40. Wang W, Mani AM, Wu ZH. DNA damage-induced nuclear factor-kappa B down. Methods Mol Biol 2006;338:281–90. activation and its roles in cancer progression. J Cancer Metast Treat 2017;3: 31. Hunter JE, Willmore E, Irving JA, Hostomsky Z, Veuger SJ, Durkacz BW. 45–59. NFkB mediates radio-sensitization by the PARP-1 inhibitor, AG-014699. 41. Bhalla KN, Rao R, SharmaP, Gupta SD, Chauhan L, Stecklein S, et al. Treatment Oncogene 2012;31:251–64. with histone deacetylase inhibitors creates `BRCAness' and sensitizes human 32. Hayden MS, Ghosh S. Shared principles in NF-kappaB signaling. Cell 2008; triple negative breast cancer cells to PARP inhibitors and cisplatin [abstract]. 132:344–62. In: Proceedings of the 35th Annual San Antonio Breast Cancer Symposium; 33. Perkins ND, Gilmore TD. Good cop, bad cop: the different faces of NF- 2012 Dec 4–8; San Antonio, TX. Philadelphia (PA): AACR; 2012. Abstract kappaB. Cell Death Differ 2006;13:759–72. nrS3-7. 34. Scott CL, Swisher EM, Kaufmann SH. Poly (ADP-Ribose) polymerase 42. Konstantinopoulos PA, Spentzos D, Karlan BY, Taniguchi T, Fountzilas E, inhibitors: recent advances and future development. J Clin Oncol 2015; Francoeur N, et al. Gene expression profile of BRCAness that correlates with 33:1397–406. responsiveness to chemotherapy and with outcome in patients with 35. Burke JR, Pattoli MA, Gregor KR, Brassil PJ, Macmaster JF, Mcintyre KW, epithelial ovarian cancer. J Clin Oncol 2010;28:3555–61. et al. BMS-345541 is a highly selective inhibitor of I kappa B kinase that 43. Mariano G, Ricciardi MR, Trisciuoglio D, Zampieri M, Ciccarone F, binds at an allosteric site of the enzyme and blocks NF-kappa B-dependent Guastafierro T, et al. PARP inhibitor ABT-888 affects response of transcription in mice. J Biol Chem 2003;278:1450–6. MDA-MB-231 cells to doxorubicin treatment, targeting Snail expres- 36. Gunderson CC, Moore KN. Olaparib: an oral PARP-1 and PARP-2 inhibitor sion. Oncotarget 2015;6:15008–21. with promising activity in ovarian cancer. Future Oncol 2015;11:747–57. 44. Chevanne M, Zampieri M, Caldini R, Rizzo A, Ciccarone F, Catizone A, 37. Sen R, Baltimore D. Inducibility of kappa immunoglobulin enhancer-binding et al. Inhibition of PARP activity by PJ-34 leads to growth impairment protein Nf-kappa B by a posttranslational mechanism. Cell 1986;47:921–8. and cell death associated with aberrant mitotic pattern and nucleolar 38. Sen R, Baltimore D. Multiple nuclear factors interact with the immuno- actin accumulation in M14 melanoma cell line. J Cell Physiol 2010;222: globulin enhancer sequences. Cell 1986;46:705–16. 401–10.

772 Mol Cancer Res; 17(3) March 2019 Molecular Cancer Research

NF-κB and Poly (ADP-ribose) Polymerase 1 Form a Positive Feedback Loop that Regulates DNA Repair in Acute Myeloid Leukemia Cells

Ding Li, Yufei Luo, Xianling Chen, et al.

Mol Cancer Res 2019;17:761-772. Published OnlineFirst December 17, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-18-0523

Supplementary Access the most recent supplemental material at: Material http://mcr.aacrjournals.org/content/suppl/2018/12/15/1541-7786.MCR-18-0523.DC1

Cited articles This article cites 43 articles, 10 of which you can access for free at: http://mcr.aacrjournals.org/content/17/3/761.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/17/3/761.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/17/3/761. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.