Suppression of Sirt1 sensitizes lung cancer cells to WEE1 inhibitor MK-1775-induced DNA damage and apoptosis

INTRODUCTION

DNA replication and cell division are tightly controlled to ensure genomic integrity.1 Before mitosis, cells go through multiple cascades of checkpoints, including G1/S, intra S and G2/M. Dysregulation of normal checkpoint signaling leads to replication stress, mitotic catastrophe, genomic instability and cell death.2 Thus, the proteins governing normal cell cycle checkpoint are considered as potential therapeutic targets for fighting human cancers.3

WEE1 kinase is a well-known G2/M checkpoint, which inhibits CDK1 and CDK2 activity through phosphorylation at their tyrosine 15 sites.4 Loss function of WEE1 kinase induces unscheduled mitosis resulting from premature CDK1 activation.5 In addition, WEE1 depletion has been recently reported to cause DNA replication stress, replication fork stalling and double-strand break (DSB) due to elevated CDK activity during S phase.6 Physiologi- cally, depletion of WEE1 in mouse model causes genomic instability, mitotic catastrophe and tumorigenesis.7 Furthermore, overexpression of WEE1 has been detected in various types of human cancers, including breast cancers, hepatocellular carci- noma, melanoma and lung cancers.8–10 Inhibition of WEE1 either by small molecules or by RNA silence-mediated gene knockdown has been shown to sensitize cancer cells to DNA damage agents such as cisplatin, irradiation and topoisomerase inhibitors.11–13 Besides DNA damage agents, combination inhibi- tion of WEE1 and CHK1 has been also proved to induce strong synergistic antitumor effect.14 Most recently, it has been reported that loss of histone H3K36me3 sensitizes cancer cells to WEE1
inhibitor treatment.15

MK-1775, also named AZD1775, has been developed as a potent and selective small-molecule WEE1 inhibitor.16 MK-1775 is currently in multiple clinical trials as single-agent therapy or in combination with DNA-damaging agents.17,18 Combination of MK-1775 and DNA-damaging agents has been reported to primarily kill p53-deficient or mutant cancer cells, however, some studies recently showed that MK-1775 possess potent anticancer activity regardless of p53 status.15

Acetylation-mediated protein post-translational modification has critical roles in epigenetics, metabolism and DNA damage repair.19–21 The protein deacetylases, which are responsible to remove acetyl groups from the modified proteins, can be categorized into histone deacetylases and sirtuin family members (Sirt).22 Sirtuin family of deacetylases requires NAD as the cofactor to complete its deacetylation reaction and it consists of seven members named Sirt1–7 individually.22 Sirtuin family of deacetylase has been recently reported to regulate multiple aspects of DNA damage and repair.23,24 In the present work, we investigated the effect of sirtuin deacetylase inhibition on MK-1775 treatment for lung cancer cells. We demonstrated that inhibition of Sirt1 strongly sensitized MK-1775-induced antitumor effects. Further, we showed that Sirt1 inhibition greatly impaired homologous recombination (HR) repair through accumulation of NBS1 and Rad51 acetylation. In sum, our work provides a novel strategy to enhance WEE1 inhibitor- induced antitumor efficacy.

RESULTS

Inhibition of Sirt1 sensitizes lung cancer cells to MK-1775 Impairment of WEE1 kinase function induces DNA replication stress and DNA damage.6 Sirtuin family of deacetylases has been recently reported to be involved in regulation of DNA damage response and repair.23 To investigate the role of sirtuin deacetylases on MK-1775 treatment in lung cancers, we treated A549 cells with 5 mM of nicotinamide (a potent inhibitor of the Sirt family deacetylases) and found that nicotinamide treatment markedly sensitized A549 cell to MK-1775 treatment (Figure 1a).

Figure 1. Inhibition of Sirt1 enhances killing of lung cancer cells by MK-1775. (a) Viability curves of A549 cells after exposure to increasing concentration of MK-1775 in presence or absence of 5 mM nicotinamide (NAM). (b) Western blotting analysis of Sirt protein expression after treatment of indicated siRNA. Δ is the nonspecific band. (c) Viability curves of indicated siRNA-treated A549 cells after exposure to increasing concentration of MK-1775. (d) Viability curves of dimethylsulfoxide (DMSO)- or 5 μM Ex527-treated A549 cells after exposure to MK-1775.(e) The colony formation assay of A549, H1299, H157, DMS53 and Calu-1 cells after treatment with 200 nM MK-1775, 5 μM Ex527 or their combination. Data presented as mean ± s.d. of three replicates. *P o0.001 when compared with MK-1775 alone.

To further study which protein in sirtuin family causes resistance to MK-1775 treatment, we treated A549 cells with small interfering RNA (siRNA) targeting all the seven sirtuin members, including Sirt1–7 and the expression of sirtuin proteins were efficiently knocked down after transfection of indicated siRNA (Figure 1b).

Meanwhile, we found that Sirt1-deficient cancer cells are hypersensitive to MK-1775 treatment (Figure 1c). Moreover, treatment of 5 μM Ex527, a selective Sirt1 inhibitor, greatly potentiates antitumor effect of MK-1775 on A549 cells (Figure 1d). Meanwhile, overexpression of Sirt1 reduced sensitivity to MK-1775 treatment (Supplementary Figure S1a).

Next, we evaluated protein levels of WEE1 and Sirt1 among nine non-small-cell lung cancer cell lines and six small cell lung cancer cell lines, and found that human lung cancer cells have different levels of WEE1 expression, but have comparable levels of Sirt1 (Supplementary Figure S1b). Then, we treated H1299, A549, H157, DMS53 and Calu-1 lung cancer cells with 200 nM MK-1775, 5 μM Ex527 and their combination, the colony formation assay showed that Ex527 and MK-1775 synergistically inhibited cell growth in all three lung cancer cell lines (Figure 1e; Supplementary Figure S1d). In consistence with Ex527, combination of Sirt1 siRNA and MK-1775 also induce great synergistic growth inhibition effects on lung cancer cells (Supplementary Figures S1c and d). PKMYT1 kinase has been reported to be involved in MK-1775-dependent antiproliferative effects,25 thus, we tested whether inhibition of Sirt1 could change PKMYT1 expression and affect MK-1775 sensitivity. The western blot analysis revealed that either Sirt1 siRNA or Ex527 did not change PKMYT1 protein level (Supplementary Figure S1e).

Ex527 potentiates MK-1775-induced apoptotic cell death

As a single agent, MK-1775 induces significant apoptosis in various cancers.26 We next used annexin V/propidium iodide double staining to analyze whether Ex527 and MK-1775 synergize to induce apoptosis in A549 cells. As expected, 1 μM of MK-1775 treatment induced slight apoptosis in A549 cells (16.9 ± 1.6%), but combination of 5 μM Ex527 with 1 μM MK-1775 caused greater apoptosis (37.1 ± 3.2%; Figure 2a). These results suggested that Ex527 greatly potentiated MK-1775-induced apoptosis in lung cancer cells. It is also known that MK-1775 induced apoptosis triggers by activation of caspase 9/3 pathway,27 thus we measured caspase 3 and caspase 9 activities using fluorescein isothiocyanate (FITC) staining kit (Abcam, Cambridge, MA, USA) 48 h after drug treatment. In consistent with annexin V staining, combination treatment of Ex527 and MK-1775 sharply increased caspase 3 activation compared with MK-1775 treatment alone (Figure 2b). We also detected greater activation of caspase 9 in combination treatment (Figure 2c). During apoptosis, PARP1 is specifically proteolyzed by caspases to produce a shorter fragment of 89 kDa.28 In consistence with annexin V and caspase assay, we detected a clear cleavage of PARP1 in combination treatment, but not in MK-1775 and Ex527 treatment-alone conditions (Figure 2d). Nevertheless, treatment of MK-1775, Ex527 or their combination did not change Bcl-2 family proteins, including Bcl-2, Mcl-1, Bcl-xL, Bax and Bak (Supplementary Figure S2).

Ex527 enhances MK-1775-induced DNA damage

MK-1775 kills cancer cells mainly through induction of S-phase- specific DNA damage, subsequently the unrepairable DNA damage causes apoptotic cell death.6 To evaluate whether the synergistic anticancer activity induced by combination of Ex527 and MK-1775 results from greater DNA damage, we used γH2AX staining, a marker of DNA damage,1 to evaluate the DNA damage level after treatment. We found that Ex527 alone caused slight γH2AX foci, but much greater level of γH2AX foci was detected along combination treatment in A549 cells (Figure 3a). Western blot analysis of γH2AX also showed that combination treatment induced much higher level of γH2AX than MK-1775 alone in both A549 and H1299 cells (Supplementary Figure S3).

Figure 3. Ex527 enhances MK-1775-induced DNA damage. (a) γ-H2AX staining in A549 cells treated with 200 nM of MK-1775, 5 μM of Ex527 or their combination for 24 h. Data presented as mean ± s.d. of three replicates. *P o0.05; ***P o0.001. (b) BrdU/7-AAD analysis of the cell cycle distribution of A549 cells treated with 200 nM of MK-1775, 5 μM of Ex527 or their combination for 24 h. (c) Quantification of BrdU-positive cells. Data presented as mean ± s.d. of three replicates. (d, e) The histogram of flow cytometry profile (d) and the image of immunofluorescence assay (e) of BrdU incorporation for indicating treatment.

We next examined whether the extensive DNA damage during

S phase affects cell cycle progression. Compared with MK-1775 alone, Ex527 plus MK-1775 treatment had similar G1, S and G2/M population (Figure 3b). In addition, combination treatment did not change the cell percentage in S phase, we detected near 50% of S-phase cells for both MK-1775 alone and MK-1775 plus Ex527 treatment (Figure 3c). Intriguingly, we found that MK-1775 induced decrease of bromodeoxyuridine (BrdU) incorporation fluorescent intensity. Moreover, combination treatment caused greater reduction in BrdU incorporation compared with MK-1775 treatment alone (Figures 3d and e). These results suggested that the greater DNA damage in combination treatment slowed DNA replication in lung cancer cells.

Ex527 impairs HR repair

As the DNA damage caused by DNA replication stress is primarily repaired by HR repair, and HR repair has been commonly considered to contribute resistance to MK-1775 treatment.29 To evaluate the HR repair activity, we measured the number of Rad51 foci formation after treatment. RAD51 is loaded onto resected single-strand DNA to form a nucleo-protein filament to complete HR repair and it is known as the marker for HR repair.30 Results showed that Ex527 significantly reduced the level of Rad51 foci in MK-1775-treated cells (Figure 4a), indicating that the greater DNA damage in the combination treatment might result from the reduction of HR repair activity.

To further study the effect of Sirt1 inhibition on HR repair, we used the DR-GFP HR repair reporter,31 in which the GFP gene is disrupted by I-Sce-I endonuclease recognition site. After transfection of I-Sce-I expression plasmid to induce DSB, the functional GFP gene can be restored only after repair through HR pathway (Figure 4b). In consistence with Rad51 foci, inhibition of Sirt1 either by siRNA or by Ex527 significantly reduced HR repair efficacy (Figure 4c).

Figure 4. Inhibition of Sirt1 reduces HR repair activity. (a) Analysis of Rad51 foci in A549 cells after exposure to indicated treatment for 24 h. Data presented as mean ± s.d. of three replicates. ***P o0.001. (b) Schematic diagram of HR reporter system. HR reporter composed of two defective GFP genes that only can be rescued by HR and resulting in GFP fluorescence. (c) Analysis of HR repair activity in A549 cells treated with Sirt1 siRNA or 5 μM of Ex527. Data presented as mean ± s.d. of three replicates. *P o0.05 when compared with control (Ctrl).

Sirt1 interacts and deacetylases HR repair machinery proteins During HR repair, DSB is first recognized by MRN complex (Mre11- Rad50-NBS1) to initiate DSB end resection, which is facilitated by CtIP and generating 3′-single-stranded DNA overhangs onto which the RAD51 recombinase coating as a filament.31 To explore the mechanism underlying this HR-promoting function of Sirt1, we performed co-immunoprecipitation assay using anti-Sirt1 anti- body. Results showed that Sirt1 was associated with HR repair machinery proteins, including Rad50, Mre11, NBS1 and Rad51, except CtIP (Figure 5a). This association was diminished after silence of Sirt1 (Supplementary Figure S4a), demonstrating HR proteins specifically interacts with Sirt1. In addition, we detected an increased level of acetylated NBS1 and Rad51 after Ex527 treatment (Supplementary Figure S4b). Meanwhile, we failed to detect Mre11 and Rad50 acetylation in our system (Supplementary Figure S4b). These results suggested that Sirt1 supported HR repair through interacting and deacetylating NBS1 and Rad51 HR proteins. To further confirm Sirt1 deacetylation of NBS1 and Rad51, we transfected A549 cells with Flag-NBS1 or Flag-Rad51 and treated with Sirt1 siRNA or Ex527. We observed significant increase of NBS1 and Rad51 acetylation after Sirt1 siRNA or Ex527 treatment (Figures 5b and c). Moreover, we co- transfected wild-type (WT) Sirt1 or its catalytically inactive mutant (H363Y) along with Flag-NBS1 or Flag-Rad51 to examine the effect of overexpression of Sirt1 on NBS1 and Rad51 acetylation.

As expected, acetylation of Flag-NBS1 and Flag-Rad51 was reduced in WT Sirt1-overexpressed cells, but not in H363Y Sirt1- overexpressed cells (Figures 5d and e).To characterize the acetylation sites on NBS1 and Rad51, we used ASEB (http://bioinfo.bjmu.edu.cn/huac/predict_p/), which is a web server designed to predict KAT-specific acetylation sites.32 K208, K233 and K690 were predicted as the potential candidate sites for NBS1, besides, K156 and K338 were predicted to as the candidate sites for Rad51. We next performed mutational analysis and generated diverse acetyl-deficient mutants with lysine (K) changed to arginine (R).22 Then, we transfected A549 cells with WT or mutant Flag-NBS1 and treated with nicotinamide to enrich the acetylation of NBS1, we found that substitution of either K208, K233, K690 or compound mutations caused significantly reduction of Flag-NBS1 acetylation (Figure 5f), sug- gesting all three sites are acetylated. Meanwhile, substitution of K156, K338 or their combination greatly decreased Flag-Rad51 acetylation (Figure 5g), suggesting Rad51 is acetylated on K156 and K338.

We then replaced lysine (K) with glutamine (Q) at acetylation sites to generate Flag-NBS1 QQQ and Flag-Rad51 QQ mutants to mimic their acetylation.22 We transfected A549-DR-GFP cells with WT or their acetylation mimic mutations to examine the effects of NBS1 and Rad51 acetylation on HR repair. Compared with WT NBS1 or WT Rad51 protein, we detected a decrease of HR repair activity in cells transfected with QQQ NBS1 or QQ Rad51 (Supplementary Figures S4c and d). In consistence with impaired HR repair, QQQ NBS1 or QQ Rad51 reduced the clonogenic survival upon MK-1775 treatment (Figures 5h and i). These results suggested that accumulated acetylation of NBS1 and Rad51 resulted from Sirt1 inhibition has a negative role on HR repair and sensitize cancer cells to MK-1775 treatment.

Figure 5. Sirt1 deacetylases NBS1 and Rad51 and controls MK-1775 sensitivity. (a) Co-immunoprecipitation (co-IP) assay was performed using anti-Sirt1 antibody, the HR proteins in the immunoprecipitation complex, including Rad50, Mre11, NBS1, Rad51 and CtIP were analyzed by western blot. (b, c) A549 cells were transfected with Sirt1 siRNA or treated with 5 μM Ex527. Twenty-four hours after treatment, cells were transfected with Flag-NBS1 or Flag-Rad51, following IP with Flag antibody and analysis of NBS1 acetylation (b) or Rad51 acetylation (c) using pan acetyl lysine antibody. (d, e) A549 cells were co-transfected with WT or H363Y HA-Sirt1 and Flag-NBS1 or Flag-Rad51 plasmid, following IP with Flag antibody and analysis of NBS1 acetylation (d) or Rad51 acetylation (e) using pan acetyl lysine antibody. (f, g) A549 cells were transfected with WT or indicated acetylation-deficient mutant of Flag-NBS1 or Flag-Rad51, and treated with 5 mM nicotinamide (NAM), following IP with Flag antibody and analysis of NBS1 acetylation (f) or Rad51 acetylation (g) using pan acetyl lysine antibody. (h, i) A549 cells were stably transfected with WT or acetylation mimic mutant of Flag-NBS1 (h) or Flag-Rad51 (i), cells were then treated with or without 200 nM MK-1775 and colony formation assays were performed. Data presented as mean ± s.d. of three replicates.

Ex527 and MK-1775 synergistically inhibits lung cancer growth in vivo

Then, we used A549 human lung cancer xenografts to study the antitumor effect of MK-1775, Ex527 and their combination. The 30 mg/kg of Ex527 treatment alone did not show obvious antitumor effect. However, it greatly enhanced tumor inhibition effect induced by 50 mg/kg of MK-1775 treatment (Figures 6a and b).

Figure 6. Combination of MK-1775 and Ex527 synergistically represses lung cancer growth in vivo. (a, b) Mice bearing A549 xenografts were treated with vehicle control, MK-1775 (50 mg/kg/day), Ex527 (30 mg/kg/day) or their combination. Tumor growth curve (a) and tumor weight (b) were shown. Data presented as mean ± s.d. of six mice per group. **P o0.01 and ***P o0.001. (c, d) Ki67 (c) and cleaved caspase 3 (d) were analyzed in tumor tissues at the end of experiments by immunohistochemical staining. Data presented as mean ± s.d. Scale bars represent 100 μm. ***P o0.001.

In addition, treatment of MK-1775 and Ex527 greatly reduced Ki67 expression compared with MK-1775 treatment alone (Figure 6c). The combination treatment also resulted in greater activation of apoptosis, as evidenced by increased expression of activated caspase 3 (Figure 6d). Whereas, treatment of MK-1775, Ex527 or combination did not change Sirt1, WEE1, NBS1 and Rad51 protein levels in vivo (Supplementary Figure S5).

Importantly, combination treatment was well tolerated without weight loss compared with MK-1775 or Ex527 single-agent treatment (Figure 7a). There were also no significant elevation of alanine aminotransferase and aspartate aminotransferase levels (Figure 7b), which are considered as sensitive indicators of organ damage or injury. Histopathological analysis further revealed no evidence of toxicity in normal tissue (Figure 7c).

DISCUSSION

Lung cancer is one of the highest-incidence and -mortality cancers, it is difficult to treat due to its poor prognosis and adaptive oncogenic mutations.33 The average 5-year survival rate for patients diagnosed with late stage of lung cancer is o18%.33 Therefore, there is an urgent need to develop new therapies for this disease. MK-1775 (AZD1775), a recently developed small molecule of WEE1 inhibitor, has been demonstrated to exhibit promising antitumor effect on lung cancers.34 Although, human lung cancer cells have diverse levels of WEE1 protein, in consistence with previous report,34 our data showed that the therapeutic efficacy of MK-1775 or MK-1775 plus Sirt1 inhibitor (Ex527) was not correlate with WEE1 protein level. Currently, there are multiple clinical trials are ongoing to evaluate therapeutic efficacy of MK-1775 monotherapy or combined with radiation therapy or chemotherapy.17,18 Here we showed that inhibition of Sirt1 greatly sensitizes MK-1775 treatment in lung cancer cells, and combination treatment of selective Sirt1 inhibitor Ex527 and MK-1775 induced synergistic antitumor effect both in vitro and in vivo. Thus, our study provided a novel strategy for MK-1775- mediated lung cancer treatment.
The WEE1 kinase is known as an important G2/M checkpoint in response to DNA damage.4,27 Functional WEE1 is necessary for cells, particularly for p53-deficient or -mutant cancer cells, to guarantee prompt DNA repair before undergoing mitosis.11,26,27 In line with this G2/M checkpoint role of WEE1, combined treatment with WEE1 inhibitor MK-1775 and chemotherapeutic drugs induces unscheduled mitotic entry with extend DNA damage, subsequently results in cell death or apoptosis.5 On the basis of this phenotype, combination of MK-1775 with DNA-damaging agents, including gemcitabine, cisplatin or carboplatin has been intensively investigated in various categories of cancers.11,18,34 It has been shown that dysregulation of protein acetylation impairs efficient DNA damage response and repair.20,23 In addition, histone deacetylase inhibitor panobinostat has been previously reported to synergistically with MK-1775 to kill leukemia cells.16,35 In the present study, we evaluated the roles of sirtuin family deacetylase on MK-1775 treatment in lung cancer cells. We found that inhibition of Sirt1 greatly potentiated MK-1775-induced cell killing and apoptosis. We further demonstrated Sirt1 deacetylates HR repair protein NBS1 and Rad51, and Sirt1 impairment caused accumulation of NBS1 and Rad51 acetylation, which leads to disturb HR repair function. Besides WEE1 kinase, ATR/Chk1 pathway is also widely known as key G2/M checkpoint regulator. These proteins also regulate CDK activity during S phase, and thereby prevent the induction of DNA damage during normal S phase progression.36 In agree with our results, co-treatment of MK-1775 with Chk1 inhibitor (MK8776) or ATR inhibitor (VE-821) also induced synergistic therapeutic efficacy in leukemia.37 Thus, our data suggested a novel therapeutic strategy to optimize MK-1775 treatment on lung cancers.

Figure 7. Toxicity analysis for indicated treatment in mice bearing A549 xenografts. (a) Body weight of mice for treatment with control (Ctrl), 50 mg/kg MK-1775, 30 mg/kg Ex527 or their combination. (b) Serum analysis, including alanine aminotransferase (ALT) and aspartate aminotransferase (AST) of mice after various treatment. Data presented as mean ± s.d. of three replicates. (c) Hematoxylin and eosin histological assay of liver and kidney organs after various treatments.

MK-1775 recently has been shown to induce replication stress through CDK1 hyperactivation-dependent excessive firing of replication origins, replication fork collapse and subsequent DNA DSBs.5,6 MK-1775 therefore induces DNA damage particularly in S phase, and HR repair is considered to be the primary mechanism to repair the DNA damage caused by MK-1775 treatment.6,29 Here we show that Sirt1 regulates HR repair activity through deacetylation of NBS1 and Rad51. As a result, Ex527 presence impairs repair capacity of the damaged DNA induced by MK-1775 treatment, and co-treatment of Sirt1 inhibitor Ex527 and MK-1775 induces greater DNA damage compared with MK-1775 alone. Although, it is well known that NBS1 and Rad51 are indispensable during HR process, our data suggest that acetylation-mediated post-translational modification on NBS1 and Rad51 negatively regulates HR repair activity.

Recently, it was reported that WEE1 inhibitor MK-1775 caused dNTP depletion particularly in H3K36me3-defective cancers.15 They suggested MK-1775 and hydroxyurea, an inhibitor of ribonucleotide reductase responsible for dNTP production, was a lethal combination and a promising therapeutic strategy in further clinic study. Although combination treatment of Sirt1 inhibitor Ex527 and MK-1775 did not decrease percentage of S phase cells, their combination significantly slowed down BrdU incorporation and synergistically inhibited clonogenic survival rate. The slowing of DNA replication in the combination treatment might contribute to the synergistic antitumor effect.

In summary, we demonstrated that Ex527 cooperatively enhanced MK-1775 therapeutic efficacy in human lung cancer cell lines and A549 xenograft mouse model. Our results suggested that Sirt1 interacted multiple HR repair machinery proteins such as MRN complex and rad51. Inhibition of Sirt1 by Ex527 therefore disturbed HR repair efficacy and allowed accumulation of DNA damage caused by MK-1775 treatment and subsequently induced apoptosis. Our results supported the clinical development of MK-1775 and Ex527 combination for the treatment on human lung cancers.

MATERIALS AND METHODS

Materials

The siRNA library targeting sirtuin family of deacetylase, including Sirt1 − 7 was obtained from Dharmacon (Lafayette, CO, USA) and the sequences are available in Supplementary Table S1. pDR-GFP (#26475) and pCBASceI (#26477) were purchased from Addgene (Cambridge, MA, USA). Flag-NBS1 and Flag-Rad51 were purchased from GeneChem (Shanghai, China). Sirt1 WT and H363Y plasmids were obtained from Addgene (Cambridge, MA, USA). Anti-β-actin (sc-58673), anti-PARP1 (sc-1562), anti-Sirt1 (sc-74504), anti-Sirt2 (sc-20966), anti-Sirt3 (sc-365175), anti-Sirt4 (sc-135053), anti- Rad50 (sc-74460), anti-Rad51 (sc-8349), anti-Flag (sc-807), anti-H2AX (sc- 54606), anti-Mcl-1 (sc-819), anti-Bax (sc-20067) and anti-HA (sc-57592) antibodies were purchased from Santa Cruz (Dallas, TX, USA). Anti-γ-H2AX (05-636) antibody was purchased from Millipore (Billerica, MA, USA). Anti- NBS1antibody (NB110-57272) was purchased from Novus Biologicals (Littleton, CO, USA). Anti-PKMYT1 antibody (ab134108), anti-Bcl-xL (ab32370) and anti-Bak (ab69404) were purchased from Abcam (Cam- bridge, MA, USA). Anti-WEE1 (#13084), anti-Mre11 (#4895), anti-Sirt5 (#8779), anti-Sirt6 (#12486) and anti-Sirt7 (#5360) antibodies were obtained from Cell Signaling Technology (Danvers, MA, USA). Anti-Flag M2 agrose beads was obtained from Sigma (St Louis, MO, USA).

Cell culture and transfection

A549, H1299, H157, DMS53 and Calu-1 cell lines were obtained from American Type Culture Collection and cultured in RPMI 1640 medium containing 10% fetal bovine serum. All cell lines are subjected to mycoplasma testing on a quarterly basis. DNA and siRNA transfection were performed using Lipofectamine 2000 and RNAiMAX reagent (Invitrogen, Carlsbad, CA, USA), respectively, as described previously.11 Briefly, cells were allowed to growth at 60–70% confluence before starting transfection, DNA/liposome or siRNA/liposome complex was prepared in opti-MEM medium and evenly added to the cell plates for 4–6 h before changing the medium. Analysis was carried out at 48–72 h after transfection.

Immunofluorescence assay

Cells grown on the chamber-slides (Lab-Tek, Rochester, NY, USA) were treated with the indicated concentration of MK-1775, Ex527 or their combination for 24 h. After treatment, cells were washed once with phosphate-buffered saline (PBS) and fixed with 4% paraformaldehyde for 10–20 min. Then, 0.5% Triton X-100 in PBS was added for permeabilization, followed by blocking in 10% goat serum. After incubation with anti-γH2AX (1:500, Millipore) or anti-Rad51 (1:300, Millipore) antibodies for 1 h, slides were washed with PBS and incubated with Alexa Fluor 488-conjugated anti-mouse IgG or Alexa Fluor 555-conjugated anti-rabbit IgG (Molecular Probes, Eugene, OR, USA) for 30 min in dark. After washing with PBS, slides were then mounted in Vectashield with 4,6-diamidino-2-phenylindole before analyzed by confocal microscopy.

Immunoprecipitation

Immunoprecipitation assay was performed as described before.38 Briefly, cells were suspended in NP-40 lysis buffer (1% NP-40, 50 mM Tris-HCl, pH
7.6, 120 mM NaCl, 1 mM EDTA, 20 mM NaF, 0.2 mM NaVO3 and 1 mM β-mercaptoethanol) supplemented with protease inhibitor cocktail (EMD Biosciences, San Diego, CA, USA) and lysed by sonication. The cell lysates were collected by centrifugation at 14 000 g for 10 min and then incubated with anti-Sirt1 (sc-74504, Santa Cruz) or anti-acetylated-lysine antibody (#9441, Cell Signaling Technology) for 4 h at 4 °C, followed by adding 40 μl protein A/G agarose beads and incubating overnight. After incubation, the beads were washed three times with lysis buffer for 5 min each. The precipitant complexes were dissolved with 2 × SDS loading buffer and analyzed by western blotting assay.

Immunohistochemistry

The immunohistochemistry was performed as described previously.39 Briefly, paraffin section on the sliders was first deparaffinized with 100% xylene, followed by rehydration with gradient ethanol (100, 95, 70, 30 and 0%). Sliders were then incubated with 3% hydrogen peroxide (H2O2) to inactivate endogenous peroxidase. After washing with water, antigen retrieval was performed by microwave heating in citrate buffer. Then, the staining was conducted using R.T.U. Vectastain Kit (Vector Laboratories,Burlingame, CA, USA) according to the manufacturer’s instructions.

Colony formation assay

About 800–1000 cells were plated in six-well plate and treated with the indicated concentration of MK-1775, Ex527 or their combination. After 7–10 days of treatment, cells were washed once with PBS and then fixed and stained with 0.1% crystal violet in 20% methanol. Surviving colonies were counted and the surviving percentage was calculated and normal- ized with untreated cells. The combination index values were calculated by using CalcuSyn software (Biosoft, Cambridge, UK). If combination index value is higher than 1, it is defined as antagonism; if combination index value is equal to 1, it is defined as additive; if combination index value is less than 1, it is defined as synergy.

Caspase activity assay

Caspase activities were measured using active caspase 3 (ab65613, Abcam) and active caspase 9 (ab65615, Abcam) staining kits according to the manufacturer’s instructions. Briefly, 1 × 106 of A549 cells after indicated treatment were incubated with FITC-DEVD-FMK or FITC-LEHD-FMK for
30 min at 37 °C in dark. After incubation, cells were washed with PBS and analyzed by flow cytometry.

FITC-annexin V and propidium iodide double staining

Staining of annexin V/propidium iodide was used to measure apoptotic cell death induced by indicated drug treatment as described previously.38 Briefly, cells after treatment were washed once with PBS and resuspended in binding buffer (10 mM Hepes (pH 7.4), 150 mM NaCl and 2.5 mM CaCl2). Then, FITC-labeled Annexin V (BD Biosciences, San Jose, CA, USA) was added into cell suspension and incubated at room temperature for 20 min in dark. Cells were then analyzed by flow cytometry after adding 2 μg/ml of propidium iodide.

Analysis of BrdU incorporation

The cell cycle distribution and cell proliferation were determined using FITC BrdU flow kit (BD PharMingen, San Diego, CA, USA) according to the manufacturer’s instructions. Briefly, cells were pulse-labeled with 10 μM BrdU for 40 min in the complete medium at 37 °C. Cells were then fixed
with BD cytofix/cytoperm buffer. After washing, cells were then treated with 300 μg/ml of DNase at 37 °C for 1 h to expose incorporated BrdU. Cells were then stained with FITC-anti-BrdU antibody for 20 min at room temperature in dark. 7-AAD was added to stain the DNA content before analyzing by flow cytometry.

Measurement of HR repair

HR repair activity was measured according to previous report.40 Briefly, A549-DR-GFP cells, which stably carrying HR cassette DR-GFP, was transfected with control or sirt1 siRNA using Lipofectamine 2000. Twenty-four hours after siRNA transfection, cells were transfected with pCBA-I-Sce-I to induce DSB. Fouty-eight hours after I-Sce-I transfection, expression of GFP was analyzed by flow cytometry.

Lung tumor xenografts

Female BALB/c nude mice were purchase from Xi’an Jiaotong University School of Medicine. All the animal experiments were performed according
to the Guide for the Animal Care and Use Committee of Xidian University. About 1 × 107 of A549 cells were subcutaneously implanted into 6-week- old nude mouse flanks. A total of 28 tumor-bearing mice were randomly divided into four groups (I–IV) with 7 mice per group and tumors were allowed to grow to an average volume of 100 mm3 before treatment. Group I injected with 0.1 ml PBS intraperitoneally as control; group II received 50 mg/kg of MK-1775 twice daily by oral gavage; and group III received 30 mg/kg of Ex527 daily by intraperitoneal injection. Group IV received their combination. Tumor volumes were measured every 4 days and calculated by the equation V = (L × W2)/2 (V, volume; L, length; and W, width) as described.15

Site-directed mutagenesis

The site-directed mutagenesis kit (Agilent Technologies, Santa Clara, CA, USA) was used to generate NBS1 and Rad51 substitution mutations according to product instructions. Briefly, the PCR was used to generate the mutagenesis and the WT of NBS1 and Rad51 in GV141 vector was used as template. The primers harboring indicated substitution mutations were used to amplify the plasmids and the amplification products were digested by DpnI to remove the non-mutated WT template and transformed into DH5α. All the mutagenesis was confirmed by sequencing.

Statistical analysis

Data were presented as the mean ± s.d. from at least three replicates. Two- tailed t-test was used to analyze statistical significance between groups. P-value o0.05 was considered significant difference. Statistical analysis was performed Adavosertib with Graphpad5 software (GraphPad Inc., San Diego, CA, USA).