A Novel Recurrent CHEK2 Y390C Mutation Identified in High-Risk

Home , MTOR

ORIGINAL ARTICLE A novel recurrent CHEK2 Y390C mutation identiﬁed in high-risk Chinese breast cancer patients impairs its activity and is associated with increased breast cancer risk

N Wang1,5, H Ding2,5, C Liu1,2,5,XLi2,LWei1,3,JYu2, M Liu2, M Ying1, W Gao1, H Jiang2 and Y Wang1,4

Certain predisposition factors such as BRCA1/2 and CHEK2 mutations cause familial breast cancers that occur early. In China, breast cancers are diagnosed at relatively younger age, and higher percentage of patients are diagnosed before 40 years, than that in Caucasians. However, the prevalence for BRCA1/2 mutations and reported CHEK2 germline mutations is much lower or absent in Chinese population, arguing for the need to study other novel risk alleles among Chinese breast cancer patients. In this study, we searched for CHEK2 mutations in young, high-risk breast cancer patients in China and detected a missense variant Y390C (1169A4G) in 12 of 150 patients (8.0%) and 2 in 250 healthy controls (0.8%, P = 0.0002). Four of the Y390C carriers have family history of breast and/or ovarian cancer. In patients without family history, Y390C carriers tend to develop breast cancer early, before 35 years of age. The codon change at Y390, a highly conserved residue located in CHEK2’s kinase domain, appeared to significantly impair CHEK2 activity. Functional analysis suggested that the CHEK2 Y390C mutation is deleterious as judged by the mutant protein’s inability to inactivate CDC25A or to activate p53 after DNA damage. Cells expressing the CHEK2 Y390C variant showed impaired p21 and Puma expression after DNA damage, and the deregulated cell cycle checkpoint and apoptotic response may help conserve mutations and therefore contribute to tumorigeneisis. Taken together, our results not only identified a novel CHEK2 allele that is associated with cancer families and confers increased breast cancer risk, but also showed that this allele significantly impairs CHEK2 function during DNA damage response. Our results provide further insight on how the function of such an important cancer gene may be impaired by existing mutations to facilitate tumorigenesis. It also offers a new subject for breast cancer monitoring, prevention and management.

Oncogene (2015) 34, 5198–5205; doi:10.1038/onc.2014.443; published online 26 January 2015

INTRODUCTION Li-Fraumeni syndrome patients who were highly prone to develop Breast cancer is the most frequently diagnosed cancer and the malignancies, including breast cancer. Two of the identified second leading cause of cancer death in females worldwide. mutations, 1100delC and I157T, have since been widely studied in Certain hereditary factors confer significant risk for breast cancers many populations, and the prevalence of these two alleles also vary and account for 5–10% of breast cancer cases.1 To date, most of greatly among different populations. They occur mostly in people of European ancestry, and confer 2 and 1.48-fold of increased risk for the known breast cancer susceptibility genes are participants of – the DNA damage response, such as ATM, CHEK2, p53, BRCA1/2, breast cancer, respectively.9 11 However, such CHEK2 variants have PALB1, BRIP1 and RAD51C/D.2–5 Other breast cancer susceptibility not been identified in Asians, South Americans or Spaniards.12,13 genes include STK11 and PTEN, two negative regulators of the Many of the familial breast cancer cases tend to occur early,16 PI3K-mTOR pathway.6 and study of early occurring breast cancer may bring new insights Two of the most prevalent breast cancer susceptibility genes, into breast cancer predisposition. In China, the age of diagnosis of BRCA1 and BRCA2, are associated with very high penetrance. It breast cancer is almost 10 years earlier than in the European union has been proposed that germline mutations of BRCA1 and BRCA2 and United States,14 and percentage wise more breast cancer account for 32–82% of hereditary breast cancer case.7 However, patients are diagnosed before age 40 in Chinese than in the prevalence of BRCA1/2 mutations varies significantly among Caucasians.15 Therefore, it is believed that Chinese breast cancer different populations. For example, BRCA1/2 gene mutations are patients under 40 years of age may have certain genetic responsible for only ~ 10% of familial Chinese breast cancer case.8 predisposition. However, BRCA1/2 mutation rates are low in Another widely studied breast cancer gene is CHEK2, which Chinese, and no germline CHEK2 1100delC or I157T variants have functions upstream of p53 during DNA damage response. It was been identified in Chinese breast cancer patients to date.12,13 In originally observed that CHEK2 was mutated in a subset of this study, we searched for other CHEK2 mutations to find

1Department of Oncology, Changhai Hospital, the Second Military Medical University, Shanghai, China; 2Key Laboratory of Systems Biology, State Key Laboratory of Cell Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China; 3Department of Oncology, the 401 hospital of PLA, Qingdao, China and 4Xinhua Cancer Center, Xinhua Hospital affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China. Correspondence: Professor H Jiang, State Key Laboratory of Cell Biology, Key Laboratory of Systems Biology of Chinese Academy of Sciences, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 320, Yueyang Road, Xuhui District, Shanghai 200031, China. or Professor Y Wang, Xinhua Cancer Center, Xinhua Hospital affiliated to Shanghai Jiao Tong University School of Medicine, 1665 Kongjiang Rd, Shanghai, China. E-mail: [email protected] or [email protected] 5These authors contributed equally to this work. Received 17 July 2014; revised 31 October 2014; accepted 21 November 2014; published online 26 January 2015 Association of CHEK2 Y390C mutation with increased breast cancer risk N Wang et al 5199 potential new alleles that may exist in young Chinese breast RESULTS cancer patients. A novel CHEK2 allele, Y390C, may confer increased breast cancer CHEK2 is the human homolog of Saccharomyces cerevisiae risk Rad53 and S. pombe Cds1.17,18 It encodes a multifunctional kinase 18 In search for potential CHEK2 variants that may account for the crucial for cell cycle regulation, DNA repair and apoptosis. The increased risk for early onset breast cancer in China, we CHEK2 protein contains three distinct functional domains: the sequenced exon 10 of the CHEK2 gene using genomic DNA – SQ/TQ-rich domain (residues 19 69), the forkhead-associated extracted from fresh blood of 150 Chinese breast cancer patients – domain (residues 115 165), and the Serine/threonine kinase domain that included 136 young breast cancer patients (o40 years of – (residues 225 490) (Figure 1a). CHEK2 acts as an important link age), 4 patients with bilateral breast cancer and 10 patients with between ATM/ATR kinases and downstream checkpoint effectors family history of breast and/or ovarian cancers. Blood from 250 during DNA-damage response. The impairment of such cellular healthy women served as controls. Both the disease and control process in CHEK2-deficient patients may corrupt DNA repair, groups were from Shanghai, and the median age of disease and 11 conserve mutations and facilitate tumorigenesis. control groups was 35.25 ± 4.56 and 32.64 ± 4.28 years old, The previously reported CHEK2 1100delC variant is located in respectively. Consistent with previous studies,12,13 the CHEK2 exon 10 of the CHEK2 gene, and expresses a truncated, labile 1100delC variant was not identified in either the patients or protein lacking the kinase domain. In this study, we set out to first controls. Three variants, 1149C4T, 1169A4G and 1176G4T were sequence the exon 10 of CHEK2 in blood samples taken from 150 identified in this study (Table 1). Among these, four patients and high risk, early onset Chinese female breast cancer patients and two controls were positive for CHEK2 1149C4T variation and one 250 healthy controls. Interestingly, we found a novel allele, Y390C, patient and one control were positive for CHEK2 1176G4T which was significantly enriched in high-risk breast cancer variation. Both were silent mutations and were not further studied. patients. We further found that this mutation had a deleterious Another novel allele, the 1169A4G variant (Table 1 and Figure 1b) effect on CHEK2 kinase function, resulting in a defective p53 resulted in an amino-acid change from tyrosine to cysteine at response and reduced apoptosis upon DNA damage. These results amino-acid residue 390, which is in the kinase domain of CHEK2. suggest that the Y390C mutation abolishes CHEK2 function and This variant was found in 12 patients (8.0%) and 2 healthy controls represents a novel breast cancer risk allele, and Y390C carriers (0.8%) and was significantly enriched in breast cancer patients should be monitored for early onset breast cancer. (P = 0.0002). Of note, none of these 12 patients were related to

Figure 1. Novel Y390C allele identiﬁed in exon 10 of CHEK2 gene. (a) Schematic diagram showing the functional domains and sites of CHEK2 mutations. (b) Novel Y390C allele identiﬁed in our study: T → C substitution (1169A4G, red arrow).

Table 1. CHEK2 variants in Chinese young breast cancer patients and controls

Base change Protein change Variant class Frequency in patients (%) Frequency in controls (%) P-valuea OR (95% CI) HWE

1169A4G p.Y390C Missense 12/150 (8.0) 2/250 (0.8) 0.0002 10.78 (2.38–48.88) 0.94 1149C4T p.T383T Silent 4/150 (2.7) 2/250 (0.8) 0.29 3.40 (0.62–18.78) 0.94 1176G4T p.A392A Silent 1/150 (0.7) 1/250 (0.4) 1.00 1.67 (0.10–26.92) 0.97 Abbreviations: CI, conﬁdence interval; HWE, Hardy–Weinberg equilibrium; OR, odds ratio. aShown are P-vaules from two-sided Fisher’s exact test.

© 2015 Macmillan Publishers Limited Oncogene (2015) 5198 – 5205 Association of CHEK2 Y390C mutation with increased breast cancer risk N Wang et al 5200 each other, and all of the Y390C carriers were heterozygous for the damaging drugs owing to the blockage of p53-mediated allele. The heterozygous carriers of Y390C variant were associated apoptotic response.19,20 Therefore, we chose this cell line as the with a 10.78-fold (95% CI: 2.38–48.88) higher risk to develop breast initial study tool in order to understand whether and to what cancer than non-Y390C-carriers (Table 1), and the Hardy–Wein- extent the Y390C mutation may impair CHEK2 function. berg equilibrium for the distribution of the alleles of Y390C was To address the functional implication of the CHEK2 Y390C 0.94. In addition, out of the 10 patients with strong family history mutation, we first established the phenotypes of CHEK2- (at least two first- or second-degree relatives with breast and/or knockdown cells. We used a retroviral RNAi approach in ovarian cancer in the family), 4 were carriers of the Y390C allele. conjunction with Eμ-Myc p19Arf − / − B cell using a method Therefore, in our breast cancer patient cohort, the rate of Y390C established in previous studies.9,20 In CHEK2-knockdown cells, carriers having breast/ovarian cancer family history (4/12 = 33.3%) we observed robust drug resistance to two classic DNA damaging is significantly higher than non-carriers (6/138 = 4.4%, P = 0.004). agents (cisplatin, doxorubicin), and a PARP1 inhibitor AZD2281 For patients without family history, the majority of Y390C carriers (Figure 2a). These data suggested that inactivation of CHEK2 gene also develop cancer before 35 years of age (6/48 = 12.5%) rather caused cellular resistance to genotoxic chemotherapy and cellular than over 35 years old (2/92 = 2.2%, P = 0.02). Taken together, escape from apoptosis after DNA damage. these data suggest that the CHEK2 Y390C allele is strongly After establishing the drug-resistance phenotype in CHEK2- associated with high-risk Chinese breast cancer patients. knockdown cells, we asked whether re-expression of the CHEK2 Of note, in our study, both the patients and control groups were Y390C variant was able to reverse such phenotype, which could recruited at Shanghai, a city with extremely rapid population indicate the functional status of such mutation. Of note, human growth, mostly due to immigration in the past century. Therefore, CHEK2 kinase is highly similar to mouse CHEK2 kinase structurally it is unlikely that our study subjects have less genetic variations. In and biochemically,18 and CHEK2 Y390C in human maps to Y394C addition, query of the 1000 human genome project suggested that in mouse. We therefore cloned CHEK2 wild-type complementary this allele is present in Southern Han Chinese at about 0.5% DNA (cDNA) (CHEK2 WT), Y394C mutation cDNA (CHEK2 Y394C) for (1 out of 199), which is consistent with the rate of 0.8% found in further analysis. First, we infected the Eμ-Myc p19Arf − / − B cell with our study. Moreover, the allele was also detected in British, Finish CHEK2 shRNA virus and selected the infected cells with puromycin and Toscanini population at similar frequency (Table 2). These data to make sure that all surviving cells express CHEK2 shRNA. Next, further indicated that the Y390C allele exists in certain populations we infected the CHEK2 stable knockdown cells with control at low frequency and is significantly enriched in high-risk breast retrovirus or retrovirus that expressed CHEK2 Y394C or CHEK2 WT, cancer patients in China, and it may represent a predisposition along with green fluorescent protein (GFP). Of note, the factor for high-risk, early onset Chinese breast cancer patients. expression levels of wild-type and Y394C CHEK2 were very similar as detected by quantitative polymerase chain reaction and CHEK2 Y390C mutation impairs the efficacy of CHEK2 in response western blot (Figures 2b and c). As a positive control, to DNA damage agents re-expression of wild-type CHEK2 restored drug sensitivity in To evaluate whether this variant is a pathogenic mutation, CHEK2-knockdown cells (Figure 2d). In contrast, neither functional analysis was implemented to detect how the CHEK2 CHEK2Y394C or vector control was able to restore drug sensitivity Y390C mutation influenced the function of CHEK2. In our previous in CHEK2-knockdown cells (Figure 2d). When GFP-sorted cells study, we found that the Eμ-Myc p19Arf − / − mouse B cells present were exposed to cisplatin, doxorubicin and the PARP1 inhibitor a highly sensitive system to address the functionality of the ATM- AZD2281, we again found than cells expressing CHEK2Y394C cDNA CHEK2-p53 signaling axis.19 Functional ablation of any of these and vector control exhibited higher survival rate than cells that re- three proteins results in significant cellular resistance to DNA expressed wild-type CHEK2 (Figure 2e). These data suggested that

Table 2. Frequency of the CHEK2 Y390C allele according to the 1000 human cancer genome

Population Population description Super Allele Allele Genotype Genotype population frequency count frequency count

CTCTC|TT|TC|TT|T

1000GENOMES:phase_1_ALL 0.003 0.997 6 2178 0.005 0.995 6 1086 1000GENOMES:phase_1_AFR African 0.000 1.000 0 492 0.000 1.000 0 246 1000GENOMES:phase_1_AMR American 0.000 1.000 0 362 0.000 1.000 0 181 1000GENOMES:phase_1_ASN Asian 0.002 0.998 1 571 0.003 0.997 1 285 1000GENOMES:phase_1_EUR European 0.007 0.993 5 753 0.013 0.987 5 374 1000GENOMES:phase_1_CEU Utah Residents (CEPH) with Northern and Western EUR 0.000 1.000 0 170 0.000 1.000 0 85 European ancestry 1000GENOMES:phase_1_CHB Han Chinese in Bejing, China EAS 0.000 1.000 0 194 0.000 1.000 0 97 1000GENOMES:phase_1_CHS Southern Han Chinese EAS 0.005 0.995 1 199 0.010 0.990 1 99 1000GENOMES:phase_1_CLM Colombians from Medellin, Colombia AMR 0.000 1.000 0 120 0.000 1.000 0 60 1000GENOMES:phase_1_ASW Americans of African Ancestry in SW USA AFR 0.000 1.000 0 122 0.000 1.000 0 61 1000GENOMES:phase_1_FIN Finnish in Finland EUR 0.005 0.995 1 185 0.011 0.989 1 92 1000GENOMES:phase_1_GBR British in England and Scotland EUR 0.011 0.989 2 176 0.022 0.978 2 87 1000GENOMES:phase_1_IBS Iberian population in Spain EUR 0.000 1.000 0 28 0.000 1.000 0 14 1000GENOMES:phase_1_JPT Japanese in Tokyo, Japan EAS 0.000 1.000 0 178 0.000 1.000 0 89 1000GENOMES:phase_1_LWK Luhya in Webuye, Kenya AFR 0.000 1.000 0 194 0.000 1.000 0 97 1000GENOMES:phase_1_MXL Mexican Ancestry from Los Angeles USA AMR 0.000 1.000 0 132 0.000 1.000 0 66 1000GENOMES:phase_1_PUR Puerto Ricans from Puerto Rico AMR 0.000 1.000 0 110 0.000 1.000 0 55 1000GENOMES:phase_1_TSI Toscani in Italia EUR 0.010 0.990 2 194 0.020 0.980 2 96 1000GENOMES:phase_1_YRI Yoruba in Ibadan, Nigera AFR 0.000 1.000 0 176 0.000 1.000 0 88

Figure 2. CHEK2 Y394C-expressing cells are resistant to cell death caused by DNA damage. Cells were infected with retrovirus that expressed a shRNA targeting CHEK2 along with GFP, such that about 20–30% cells were infected and GFP positive. Following drug treatments, the surviving population was assessed by flow cytometer to determine the percentage of GFP/shRNA-expressing cells. (a) Suppression of CHEK2 in Arf − / − Eμ-Myc p19 B cells conferred cellular resistance to doxorubicin (15 nM), cisplatin (3 μM) and AZD2281 (1 nM). Empty vector was used as a negative control. Shown are mean ± s.d. from three independent experiments. The upper panel shows the knockdown efficiency of shRNA. (b) Re-expression of CHEK2 WT and CHEK2 Y394C in CHEK2-knockdown cells was analyzed by QPCR. Empty vector was used as a negative control. (c) Similar level of CHEK2 WT and CHEK2 Y394C was re-expressed in our experiments. β-actin was used as a loading control. (d) Re-expression of WT Y394C CHEK2 , but not CHEK2 , restored cellular sensitivity to cisplatin (3 μM), doxorubicin (15 nM) and AZD2281 (1 nM) in CHEK2-knockdown cells. Empty vector was used as a negative control. (e) CHEK2-knockdown cells were infected with retrovirus that encoded CHEK2 Y394C or CHEK2 WT, and treated with cisplatin, doxorubici and AZD2281 for 48 h. Percentage of viable cells was measured by PI exclusion using FACS. Cells re-expressing the CHEK2 Y394C cDNA showed a higher survival rate than the CHEK2 WT cDNA. Empty vector was used as a negative control. *Po0.05, **Po0.01, ***Po0.001. a–e were performed in Eμ-Myc p19Arf − / − B cells. the Y390C mutation significantly impaired CHEK2 function during understand the potential relationship between the Y390C muta- DNA damage response. tion and the increased early breast cancer risk, we analyzed CHEK2 downstream events upon DNA damage. Phosphorylation of p53 on Ser20 by CHEK2 activates the p53 CHEK2 Y390C mutation impairs p53 activation, cell cycle arrest and protein24–26 and enhances its ability to upregulate genes involved apoptosis upon DNA damage in DNA repair, cell death and cell cycle control.27,28 Mouse CHEK2- CHEK2 is activated by ATM (ataxia-telangiectasia mutation) in deficient cells exposed to ionizing radiation fail to stabilize and response to DNA damage. It subsequently activates BRCA1, p53 activate p53,29 so do cells that express kinase-defective CHEK2 21 and inactivates CDC25 to facilitate DNA repair and trigger cell mutants.26 Phosphorylation of the serine 20 residue of p53 by cycle checkpoints as well as p53-mediated apoptosis.22 Impair- CHEK2 was shown to be required for its DNA damage-dependent ment of these functions may corrupt DNA repair, enable the stabilization.24,26 Importantly, the Eμ-Myc p19Arf − / − cell line we mutated cells to escape from apoptosis and ultimately contribute used, carrying Arf deletion, retains an intact p53 response upon to tumorigenesis. Previous studies have demonstrated that DNA damage,27 and therefore allowed us to address whether the mutations in CHEK2 gene increased breast cancer risk, especially CHEK2 Y390C mutation affects p53 response. Consistent with among women with a family history of breast cancer.23 To further CHEK2 Y390C being a non-functional allele, we found that

Figure 3. CHEK2 Y394C impairs p53 activation and transcription of p53 downstream effectors. (a) Western blot analysis of p53 Ser20 phosphorylation after DNA damage. CHEK2-knockdown cells re-expressing CHEK2 WT were able to phosphorylate p53 at Ser20 after Y394C doxorubicin exposure (15 nM for 6 h), whereas cells re-expressing CHEK2 did not show detectable p53 Ser20 phosphorylation after DNA damage. Empty vector was used as a negative control. (b, c) QPCR quantiﬁcation of the mRNA levels of p21 (b) and Puma (c) mRNA expression after treatment with cisplatin (3 μM) or doxorubicin (15 nM). Data are shown as mean ± s.d. from three independent experiments. *Po0.05, **Po0.01, ***Po0.001. (d, e) Western blot analysis p53 and CDC25A after DNA damage. MCF7 cells expressing the CHEK2Y390C cDNA showed reduced p53 and CDC25A phosphorylation after doxorubicin treatment. Wild-type CHEK2 was used as a positive control and empty vector as a negative control. a–c were performed in Eμ-Myc p19Arf − / − B cells, d and e were performed in MCF7 cell line.

CHEK2-knockdown cells that re-expressed the CHEK2 mutant These data again indicated that the CHEK2 Y390C is a functionally showed little to no p53 Ser20 phosphorylation upon DNA damage null allele. (Figure 3a). Downstream of p53, expression levels of the cell cycle checkpoint mediator p21 and pro-apoptotic genes Puma28,29 were DISCUSSION all significantly lower in CHEK2Y394C cells when compared with CHEK2 WT cells after DNA damage (Figures 3b and c). In this study, we identified a novel CHEK2 Y390C allele that may To further address the role of CHEK2 Y390C mutation in breast represent a genetic predisposing factor in high risk, early onset cancer, we expressed wt or Y390C human CHEK2 proteins in Chinese breast cancer patients. The CHEK2 kinase is an important human breast cancer cell line MCF7. Consistent with the results in transducer of DNA damage checkpoint signal. It is a central Eμ-Myc p19Arf − / − cell line, MCF7 cells that express the Y390C effector of cell’s response to DNA double-strand breaks and serves – CHEK2 protein failed to efficiently phosphorylate p53 upon DNA as a phosphorylation target of ATM.30 33 Consistent with its critical damage (Figure 3d). In addition, phosphorylation of another roles in the DNA-damage checkpoint, mutations in CHEK2 important kinase substrate for CHEK2, CDC25A, was also contribute to hereditary cancers in human.30 By facilitating DNA significantly lower in Y390C-expressing MCF7 cells (Figure 3e). repair, triggering cell cycle arrest and apoptosis once the DNA

Oncogene (2015) 5198 – 5205 © 2015 Macmillan Publishers Limited Association of CHEK2 Y390C mutation with increased breast cancer risk N Wang et al 5203 damage is beyond repair, CHEK2 may act as a barrier against Approximately 150 000 new breast cancer cases are diagnosed tumorigenesis.34,35 In addition to germline mutations that cause in China annually.41 Given the alarming increase of new breast familial cancer cases, somatic mutations of CHEK2 may also cancer cases and their general early occurrence in China, the contribute to tumorigenesis. A recent study of diffuse large B-cell finding that CHEK2 Y390C mutation confers an increased risk of lymphoma identified several CHEK2 somatic mutations.36 breast cancer in young Chinese breast cancer patients is The CHEK2*1100delC and I157T variants were associated with remarkable. It may bring clinical benefit to monitor CHEK2 breast cancer risk in western countries.37 Both alleles were Y390C carriers in Chinese female and other relevant populations originally identified in cancer-prone Li-Fraumeni syndromes. The for early onset breast cancer. Our findings also supported the 1100delC variant is thought to express a truncated protein that theory that different variants in CHEK2 gene exist among different lacks the kinase domain, therefore a null allele. The other variant, populations. Therefore, different diagnosis and therapeutic I157T, despite being widely investigated for breast cancer risk, has choices for different populations should also be considered for not been shown to significantly affect CHEK2 activity. Wu et al.38 future research. showed that the I157T mutation behaves like wild-type CHEK2. Several other CHEK2 variants such as R406H have been reported in women of European ancestry, but it is not clear whether such MATERIALS AND METHODS variants impact CHEK2 function.39 Common to all these variants, Study approval they were mostly relevant to Caucasians and have not been All the tissue specimens used in this study were obtained with patients’ identified in Chinese hereditary breast cancer patients.12,13 written informed consent and the Ethics Committee of Changhai Hospital In this study, we identified a novel CHEK2 Y390C allele that is granted approval for this measure as well as the research protocol. associated with increased cancer risk in high-risk Chinese breast cancer patients. Importantly, we also showed that it is a Subjects and DNA sequencing 4 fi functionally null allele. The Y390C (1169A G) allele was identi ed One hundred and fifty high-risk Chinese female young breast cancer in 8.0% of our cohort of high-risk Chinese breast cancer patients patients were admitted in the Department of Oncology, Changhai Hospital, and conferred an ~ 10.78-fold higher risk of developing breast from October 2005 to January 2011. Two hundred and fifty healthy women cancer than non-carriers, which is higher than the 1.48 to two fold served as controls. Genomic DNA was extracted from leukocyte pellet by increase breast cancer risk reported for I157T and 1100delC phenol–chloroform extraction following proteinase K digestion. Germline allele.9,10 The CHEK2 Y390C variant was previously reported in a mutations within entire exon 10 coding regions of CHEK2 of two group female patient who was diagnosed with colorectal cancer and samples were detected by PCR, followed by direct DNA sequencing of both 40 strands using an automated ABI 3730 DNA sequencer (Applied Biosystems, endometrial cancer at 58 years and breast cancer at 63-year old. ’ The daughter of this patient was also diagnosed with breast Foster City, CA) following the manufacturer s instructions. Owing to the presentation of exon 10 of CHEK2 gene in multiple homologous copies in cancer, suggesting that this variant might cause hereditary breast the genome, nested PCR procedure was performed to avoid amplification cancer. Importantly, our study provided additional evidence of of pseudogene sequences and three sets of specific primers for CHEK2 four more incidences of familiar breast/ovarian cancers with this exon 10 on chromosome 22 described elsewhere42–44 and one set of genotype. Moreover, we showed that the Y390C mutation has a primers designed by Invitrogen (Carlsbad, CA, USA) were used for PCR. The deleterious effect on the function of CHEK2. Cells expressing this nucleotide sequence was compared with the published CHEK2 sequence in mutant exhibit impaired DNA-damage response and escape GenBank by the National Center of Biological Information (NM_007194.3, from apoptosis after DNA damage. Dysregulated DNA damage http://www.ncbi.nlm.nih.gov). response may result in accumulation of cells with improperly repaired DNA that escapes from apoptosis and cell cycle Cell lines and chemicals 34,35 arrest, and this may explain the increased breast cancer risk Phoenix cell and Eμ-Myc p19Arf − / − mouse B-cell lymphoma cell line have associated with the Y390C genotype. been previously described.19,20 Phoenix cell was cultured in Dulbecco's The Y390 residue is located in the kinase domain as well as the modified Eagle's medium supplemented with 10% fetal bovine serum. T loop of the catalytic domain (residues 371–391) of CHEK2.30 Eμ-Myc p19Arf − / − cell was cultured in B-cell medium (50% Dulbecco's Moreover, this Y390 residue is highly conserved through evolution modified Eagle's medium and 50% Iscove's modified Dulbecco's media, from yeast to human. It may have an important role in maintaining supplemented with 10% fetal bovine serum, L-glutamate, and 5 μM the structure and/or function of CHEK2, which will be an b-mercaptoenthanol). Doxorubicin and cisplatin were purchased from interesting topic for future studies. Interestingly, according to Selleck chemicals. PARP1 inhibitor AZD2281 was product of Selleck Bio Company. the Cancer Genome Atlas data extracted from COSMIC (Catalogue of Somatic Mutations in Cancer), 7% (37 out of 523) of glioblastoma patients carry this Y390C mutation. It is possible shRNA, cDNA constructs, RNA Preparation and RT-PCR that this Y390C allele may also predispose to glioblastoma and is Retroviral pMSCV-IRES-GFP vector was kindly provided by Professor. selectively enriched in glioblastoma patients. However the Michael Hemann at MIT. The cloning procedure was performed as described.19,20 In brief, recombinant viruses were produced in Phoenix somatic status of these patients is not available, therefore it μ Arf − / − remains a topic for future study. Moreover, given that many of the cells transfected with retroviral vector plasmids. E -Myc p19 cells were infected by spin in the presence of polybrene (7 µg/ml). GFP glioblastoma patients were treated with chemotherapy, and presence was detected in BD Accuri C6 Flow Cytometer (BD Biosciences, considering our previous result that temozolomide, the front-line San Jose, CA, USA). GFP-positive cells were sorted on BDFACS Aria II Flow drug for glioblastoma requires an intact ATM-CHEK2-p53 pathway Cytometer (BD Biosciences). Genes’ knockdown or overexpression was 20 to elicit cancer cell death, it is also possible that the Y390C confirmed by QPCR or western blotting. The hairpin targeting mutation is generated by the mutagenic effect of temozolomide, 3′ untranslated region of chek2 cDNA and the target sequence is: and cells that carry this allele lacked CHEK2 activity, became drug 5′-CCAGAAACACATAATCATTAA-3′. The hairpin sequence has been 19 resistant and were selected for thereafter. Taken together, the previously validated. recurrence of this Y390C mutation in glioblastoma patients further After retroviral infection, cells were either directly subjected to GFP- stresses the importance of the Y390 amino-acid residue, and competition assays or GFP-sorted for pure population survival assays. The PCR primer sequences for chek2 wild-type cDNA are: forward 5′-CCGG supports the notion that this mutation is deleterious to CHEK2 AATTCATGAAGAGTCATCATCAGTCCCACAGC-3′, reverse 5′-CCGCTCGAG function. TCACAACACAGCCCCACACCTCGAGCG-3′; primers for point mutation of Our results expanded the case for CHEK2 in breast cancer chek2 Y394C are: forward 5′-CCTTATGTGGTACGCCCACTTGTCTGGCTCCTG and provided further example of how certain existing mutations AGGTTCTTGTCTCC-3′, Reverse 5′-GGAGACAAGAACCTCAGGAGCCAGACAA may inactivate this important DNA damage response gene. GTGGGCGTACCACATAAGG-3′.

© 2015 Macmillan Publishers Limited Oncogene (2015) 5198 – 5205 Association of CHEK2 Y390C mutation with increased breast cancer risk N Wang et al 5204 Total messenger RNA of cell lines was extracted with Trizol reagent 10 Liu C, Wang Y, Wang QS, Wang YJ. The CHEK2 I157T variant and breast cancer (Invitrogen). According to the manufacturer’s instruction, total messenger susceptibility: a systematic review and meta-analysis. Asian Pac J Cancer Prev RNA was transcribed to cDNA with the SuperScriptIII First-Strand Synthesis 2012; 13: 1355–1360. System (Invitrogen) according to the instruction of the manufacturer. QPCR 11 Antoni L, Sodha N, Collins I, Garrett MD. CHK2 kinase: cancer susceptibility for amplifying was performed using the following primers: chek2-forward and cancer therapy - two sides of the same coin? Nat Rev Cancer 2007; 7: 5′-TGACAGTGCTTCCTGTTCACA-3′, reverse 5′-AGCTGGACGAACCCTGA 925–936. TACT-3′. p21-forward 5′-CCTGGTGATGTCCGACCTG-3′, reverse 5′-CCAT 12 Chen W, Yurong S, Liansheng N. Breast cancer low-penetrance allele 1100delC in GAGCGCATCGCAATC-3′; Puma: forward 5′-AGCAGCACTTAGAGTCGCC-3′, the CHEK2 gene: not present in the Chinese familial breast cancer population. Adv reverse 5′-CCTGGGTAAGGGGAGGAGT-3′; Noxa-Forward 5′-CTCAGGAAGA Ther 2008; 25:496–501. TCGGAGACAAAGT-3′, reverse 5′-GAGTTGAGCACACTCGTCCTT-3′. 13 González-Hormazábal P, Castro VG, Blanco R, Gómez F, Peralta O, Waugh E et al. The expression level of each gene was measured by Bio-Rad CFX Absence of CHEK2 1100delC mutation in familial breast cancer cases from a South Connect QPCR Detection System. American population. Breast Cancer Res Treat 2008; 110:543–545. 14 Song Q-K, Li J, Huang R, Fan J-H, Zheng R-S, Zhang B-N et al. Age of diagnosis of Antibodies and western blot analysis breast cancer in china: almost 10 years earlier than in the United States and the European union. Asian Pac J Cancer Prev 2014; 15: 10021–10025. Antibodies against CHEK2 (Millipore, Temecula, CA, USA), P53 (Santa Cruz, 15 Kwong A, Cheung P, Chan S, Lau S. Breast cancer in Chinese women younger than Dallas, TX, USA), phospho-Ser20 P53 (Santa Cruz), CDC25A (ABGENT, San age 40: are they different from their older counterparts? World J Surg 2008; 32: Diego, CA, USA), phospho-CDC25A (ABGENT), β-actin (Sigma, St Louis, MO, 2554–2561. USA) were used for Western blot analysis. Western blotting was performed as described.19 16 Vahteristo P, Bartkova J, Eerola H, Syrjäkoski K, Ojala S, Kilpivaara O et al. A CHEK2 genetic variant contributing to a substantial fraction of familial breast cancer. Am J Hum Genet 2002; 71: 432–438. Cell survival assays and competition assays 17 Brown AL, Lee CH, Schwarz JK, Mitiku N, Piwnica-Worms H, Chung JH. For drug treatment experiments, 1 × 106 B cells/well were seeded in A human Cds1-related kinase that functions downstream of ATM protein 48-well cell culture plates and treated with various drugs at the indicated in the cellular response to DNA damage. Proc Natl Acad Sci USA 1999; 96: concentrations. Cell survival assays were done at 48 h after the initial 3745–3750. treatment, and GFP-competition assays were analyzed at 72 h after drug 18 Bartek J, Falck J, Lukas J. CHK2 kinase-a busy messenger. Nat Rev Mol Cell Biol exposure as previously described.19 2001; 2: 877–886. 19 Jiang H, Reinhardt HC, Bartkova J, Tommiska J, Blomqvist C, Nevanlinna H et al. The combined status of ATM and p53 link tumor development with therapeutic Statistics response. Genes Dev 2009; 23: 1895–1909. Differences of event frequency between two groups were analyzed using a 20 Jiang H, Pritchard JR, Williams RT, Lauffenburger DA, Hemann MT. A mammalian two-sided Fisher's exact test. A P-value o0.05 was considered significant. functional-genetic approach to characterizing cancer therapeutics. Nat Chem Biol 2011; 7:92–100. 21 Baloch AH, Daud S, Raheem N, Luqman M, Ahmad A, Rehman A et al. CONFLICT OF INTEREST Missense mutations (p.H371Y, p.D438Y) in gene CHEK2 are associated with The authors declare no conflict of interest. breast cancer risk in women of Balochistan origin. Mol Biol Rep 2014; 41: 1103–1107. 22 O'Driscoll M, Jeggo PA. The role of double-strand break repair - insights from ACKNOWLEDGEMENTS human genetics. Nat Rev Genet 2006; 7:45–54. ł This study was supported by the major scientific research project (2013CB910404), 23 Cybulski C, Woko orczyk D, Jakubowska A, Huzarski T, Byrski T, Gronwald J et al. Natural Science Foundation of the People’s Republic of China (No. 81372854; Risk of breast cancer in women with a CHEK2 mutation with and without a family 29 – 31371418; 81202096; 81102010), and Shanghai Science and Technology Committee history of breast cancer. J Clin Oncol 2011; :3747 3752. (No. 114119a7500, No. 06DZ19505 and No.13NM1401504). The funders had no role in 24 Shieh SY, Ahn J, Tamai K, Taya Y, Prives C. The human homologs of checkpoint study design, data collection and analysis, decision to publish or preparation of the kinases Chk1 and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage manuscript. We thank Dr Qingpeng Kong of the Kunming Institute of Zoology, inducible sites. Genes Dev 2000; 14:289–300. Chinese Academy of Sciences for his help in analysis of the 1000 human genome 25 Shieh SY, Ahn J, Tamai K, Taya Y, Prives C. Phosphorylation of Ser-20 mediates project. stabilization of human p53 in response to DNA damage. Proc Natl Acad Sci USA 1999; 96: 13777–13782. 26 Chehab NH, Malikzay A, Appel M, Halazonetis TD. Chk2/hCds1 functions REFERENCES as a DNA damage checkpoint in G(1) by stabilizing p53. Genes Dev 2000; 14: 278–288. 1 Lacroix M, Leclercq G. The "portrait" of hereditary breast cancer. Breast Cancer Res 27 Stott FJ, Bates S, James MC, McConnell BB, Starborg M, Brookes S et al. Treat 2005; 89:297–304. The alternative product from the human CDKN2A locus, p14 (ARF), participates 2 Walsh T, King MC. Ten genes for inherited breast cancer. Cancer Cell 2007; 11: 17 103–105. in a regulatory feedback loop with p53 and MDM2. EMBO J 1998; : – 3 Meindl A, Hellebrand H, Wiek C, Erven V, Wappenschmidt B, Niederacher D et al. 5001 5014. fi 408 Germline mutations in breast and ovarian cancer pedigrees establish RAD51C as a 28 Vogelstein B, Lane D, Levine AJ. Sur ng the p53 network. Nature 2000; : – human cancer susceptibility gene. Nat Genet 2010; 42: 410–414. 307 310. 4 Gutiérrez-Enríquez S, Bonache S, de Garibay GR, Osorio A, Santamariña M, 29 Ryan KM, Phillips AC, Vousden KH. Regulation and function of the p53 tumor 13 – Ramón y Cajal T et al. About 1% of the breast and ovarian Spanish families suppressor protein. Curr Opin Cell Biol 2001; :332 337. testing negative for BRCA1 and BRCA2 are carriers of RAD51D pathogenic 30 Cai Z, Chehab NH, Pavletich NP. Structure and activation mechanism of the CHK2 35 – variants. Int J Cancer 2014; 134:2088–2097. DNA damage checkpoint kinase. Mol Cell 2009; :818 829. 3 5 Osorio A, Endt D, Fernández F, Eirich K, de la Hoya M, Schmutzler R et al. 31 Ahn J, Urist M, Prives C. The Chk2 protein kinase. DNA Repair (Amst) 2004; : Predominance of pathogenic missense variants in the RAD51C gene occurring in 1039–1047. breast and ovarian cancer families. Hum Mol Genet 2012; 21: 2889–2898. 32 Bartek J, Lukas J. Chk1 and Chk2 kinases in checkpoint control and cancer. Cancer 6 Apostolou P, Fostira F. Hereditary breast cancer: the era of new Cell 2003; 3:421–429. susceptibility genes. Biomed Res Int 2013; 2013: 747318. 33 Stracker TH, Usui T, Petrini JH. Taking the time to make important decisions: the 7 King MC, Marks JH, Mandell JB. Breast and ovarian cancer risks due to inherited checkpoint effector kinases Chk1 and Chk2 and the DNA damage response. DNA mutations in BRCA1 and BRCA2. Science 2003; 302:643–646. Repair (Amst) 2009; 8: 1047–1054. 8 Liede A, Narod SA. Hereditary breast and ovarian cancer in Asia: genetic epide- 34 Takacova S, Slany R, Bartkova J, Stranecky V, Dolezel P, Luzna P et al. DNA damage miology of BRCA1 and BRCA2. Hum Mutat 2002; 20:413–424. response and inflammatory signaling limit the MLL-ENL-induced leukemogenesis 9 Weischer M, Bojesen SE, Ellervik C, Tybjaerg-Hansen A, Nordestgaard BG. in vivo. Cancer Cell 2012; 21:517–531. CHEK2- 1100delC genotyping for clinical assessment of breast cancer risk: 35 Bartkova J, Horejsí Z, Koed K, Krämer A, Tort F, Zieger K et al. DNA damage metaanalyses of 26,000 patient cases and 27,000 controls. J Clin Oncol 2008; 26: response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 542–548. 2005; 434:864–870.

Oncogene (2015) 5198 – 5205 © 2015 Macmillan Publishers Limited Association of CHEK2 Y390C mutation with increased breast cancer risk N Wang et al 5205 36 de Miranda NF, Peng R, Georgiou K, Wu C, Falk Sörqvist E, Berglund M et al. DNA 41 Cao AY, Hu Z, Shao ZM. Mutation screening of breast cancer repair genes are selectively mutated in diffuse large B cell lymphomas. J Exp Med susceptibility genes in Chinese high-risk families: the results will develop 2013; 210: 1729–1742. the genetic testing strategy in China. Breast Cancer Res Treat 2010; 120: 37 Consortium CBCC, CHEK2*1100delC and susceptibility to breast cancer: a colla- 271–272. borative analysis involving 10,860 breast cancer cases and 9,065 controls from 42 Baysal BE, DeLoia JA, Willett-Brozick JE, Goodman MT, Brady MF, Modugno F et al. 10 studies. Am J Hum Genet 2004; 74: 1175–1182. Analysis of CHEK2 gene for ovarian cancer susceptibility. Gynecol Oncol 2004; 95: 38 Wu X, Webster SR, Chen J. Characterization of tumor-associated Chk2 mutations. 62–69. J Biol Chem 2001; 276: 2971–2974. 43 Sodha N, Houlston RS, Williams R, Yuille MA, Mangion J, Eeles RA. A robust 39 Novak DJ, Chen LQ, Ghadirian P, Hamel N, Zhang P, Rossiny V et al. Identiﬁcation method for detecting CHK2/RAD53 mutations in genomic DNA. Hum Mutat 2002; of a novel CHEK2 variant and assessment of its contribution to the risk of breast 19:173–177. cancer in French Canadian women. BMC Cancer 2008; 8: 239. 44 Vahteristo P, Tamminen A, Karvinen P, Eerola H, Eklund C, Aaltonen LA et al. P53, 40 Brinkman H, Barwell J, Rose S, Tinworth L, Sodha N, Langman C et al. Evidence CHK2, and CHK1 genes in Finnish families with Li-Fraumeni syndrome: further against a major genetic basis for combined breast and colorectal cancer evidence of CHK2 in inherited cancer predisposition. Cancer Res 2001; 61: susceptibility. Clin Genet 2006; 70: 526–529. 5718–5722.