Oncogene (2010) 29, 26–33 & 2010 Macmillan Publishers Limited All rights reserved 0950-9232/10 $32.00 www.nature.com/onc ORIGINAL ARTICLE Functional variant of KLOTHO: a breast cancer risk modifier among BRCA1 mutation carriers of Ashkenazi origin I Wolf1,2,9, Y Laitman3,9, T Rubinek1, L Abramovitz1, I Novikov4, R Beeri5, M Kuro-O6, HP Koeffler7, R Catane1,2, LS Freedman4, E Levy-Lahad5, BY Karlan8, E Friedman2,3 and B Kaufman1 1 The Institute of Oncology, Chaim Sheba Medical Center, Ramat Gan, Israel; 2Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel; 3The Susanne Levy Gertner Oncogenetics Unit, Institute of Human Genetics, Chaim Sheba Medical Center, Ramat Gan, Israel; 4Gertner Institute for Epidemiology and Health Policy Research, Chaim Sheba Medical Center, Ramat Gan, Israel; 5 Medical Genetics Unit, Shaare Zedek Medical Center, Jerusalem, Israel; 6Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX, USA; 7Division of Hematology/Oncology, Cedars-Sinai Medical Center, Los Angeles, CA, USA and 8Women’s Cancer Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, USA Klotho is a transmembrane protein that can be shed and act as a circulating hormone and is a putative tumor suppressor in breast cancer. A functional variant of KLOTHO (KL-VS) contains two amino acid substitutions F352V and C370S and shows reduced activity. Germ-line mutations in BRCA1 and BRCA2 substantially increase lifetime risk of breast and ovarian cancers. Yet, penetrance of deleterious BRCA1 and BRCA2 mutations is incomplete even among carriers of identical mutations. We examined the association between KL-VS and cancer risk among 1115 Ashkenazi Jewish women: 236 noncarriers, 631 BRCA1 (185delAG, 5382insC) carriers and 248 BRCA2 (6174delT) carriers. Among BRCA1 carriers, heterozygosity for the KL-VS allele was associated with increased breast and ovarian cancer risk (hazard ratio 1.40, 95% confidence intervals 1.08–1.83, P ¼ 0.01) and younger age at breast cancer diagnosis (median age 48 vs 43 P ¼ 0.04). KLOTHO and BRCA2 are located on 13q12, and we identified linkage disequilibrium between KL-VS and BRCA2 6174delT mutation. Studies in breast cancer cells showed reduced growth inhibitory activity and reduced secretion of klotho F352V compared with wildtype klotho. These data suggest KL-VS as a breast and ovarian cancer risk modifier among BRCA1 mutation carriers. If validated in additional cohorts, the presence of KL-VS may serve as a predictor of cancer risk among BRCA1 mutation carriers. Oncogene (2010) 29, 26–33; doi:10.1038/onc.2009.301; published online 5 October 2009 Keywords: breast cancer; klotho; BRCA; ovarian cancer; insulin growth factor-1; tumor suppressor Correspondence: Dr I Wolf, Institute of Oncology, Chaim Sheba Medical Center, Tel-Hashomer, Ramat-Gan 52621, Israel. E-mail: wolf-i@inter.net.il 9 These authors contributed equally to this work. Received 8 December 2008; revised 18 August 2009; accepted 24 August 2009; published online 5 October 2009 Introduction The KLOTHO gene (MIM 604824) was first identified as a potent suppressor of aging (Kuro-o et al., 1997). Mice homozygous for a mutated klotho allele exhibit a syndrome resembling human aging, whereas klotho over-expression extended mice lifespan (Kuro-o et al., 1997; Kurosu et al., 2005). Klotho is a 1014-amino acid single pass transmembrane protein (Kuro-o et al., 1997; Matsumura et al., 1998; Shiraki-Iida et al., 1998; Ito et al., 2000). Its extracellular domain is composed of two internal repeats, KL1 and KL2, which can be cleaved, shed into the serum and act as a circulating hormone (Kuro-o et al., 1997; Ohyama et al., 1998; Shiraki-Iida et al., 1998; Imura et al., 2004; Chen et al., 2007). Several activities of klotho have been described to date, including retention of the calcium channel TRPV5 (Cha et al., 2008), activation of FoxO forkhead transcription factors (Yamamoto et al., 2005), binding to fibroblast growth factor receptors (FGFR) 1–4 and acting as a co-receptor for FGF23 (Kurosu et al., 2006; Urakawa et al., 2006). Klotho is also a potent inhibitor of the insulin receptor and the IGF-1 receptor (Kurosu et al., 2006). As the IGF-1 and the insulin pathways have important functions in breast cancer pathogenesis (Wolf et al., 2005; Yee, 2006), we have recently studied klotho expression and activities in human breast cancer (Wolf et al., 2008). High klotho expression was found in normal breast samples, but very low expression occurs in ductal carcinoma in situ and invasive breast cancer. Moreover, over-expression of klotho reduced proliferation and inhibited ligand-dependent activation of the IGF-1 and insulin pathways in breast cancer cells. These data suggest that klotho can behave as a potential tumor suppressor in human breast cancer. A functional variant of klotho, termed KL-VS, contains six sequence variants in complete linkage disequilibrium, two of which result in amino acid substitutions F352V and C370S. The presence of phenylalanine at a position equivalent to position 352 in the human KLOTHO gene is highly conserved among species and its substitution to valine may alter the Klotho modifies BRCA1-associated breast cancer risk I Wolf et al 27 excretion and enzymatic activity of the protein (Arking et al., 2002). Homozygosity for this variant was underrepresented in elderly compared with newborns in various populations (Arking et al., 2002) and was also identified as an independent risk factor for early onset coronary artery disease (Arking et al., 2003). Germ-line mutations in BRCA1 (MIM 113705) and BRCA2 (MIM 600185) genes substantially increase lifetime risk of breast and ovarian cancers. Yet, penetrance of deleterious BRCA1 and BRCA2 mutations is incomplete, age-dependent and vary even among carriers of identical mutations (for example, Ashkenazi Jewish, Iceland population) (Levy-Lahad and Friedman, 2007; Begg et al., 2008). Such observations suggest that other genetic and environmental factors may modify cancer risk in BRCA1 and BRCA2 mutation carriers (Antoniou and Easton, 2006). Several genetic variants have been reported as modifiers of cancer risk among these individuals (Antoniou et al., 2007, 2008). Single nucleotide polymorphisms in RAD51, FGFR2 and MAP3K1 have been linked to increased breast cancer risk among BRCA2 mutation carriers only, and a single nucleotide polymorphism in TNRC9 has been linked to increased breast cancer risk among both BRCA1 and BRCA2 mutation carriers (Antoniou et al., 2007, 2008). As klotho is a potential tumor suppressor in breast cancer, we aimed to study the association between the presence of KLOTHO functional variant and risk of breast and ovarian cancer among Ashkenazi Jewish women, non-carriers and carriers of one of three predominant BRCA1 (185delAG, 5382insC) or BRCA2 (6174delT) mutations. (N ¼ 236), BRCA1 mutation carriers (N ¼ 631) and BRCA2 mutations carriers (N ¼ 248). Similar klotho allele frequencies were noted among non-carriers and BRCA1 mutation carriers regardless of disease status (Table 1). Thus, FF, FV and VV frequencies were 58, 35 and 7% and 65, 29 and 6% among non-carriers and carriers, respectively (P ¼ 0.16). No statistically significant differences were noted when the analysis was conducted according to disease status within each group or between the two groups (data not shown). Compared to this population of Ashkenazi Jewish women, lower frequencies of the FV and VV variants were previously reported for Bohemian or Baltimore Caucasian (FF—74%, FV—25%, VV—1%) (Arking et al., 2002). BRCA1 mutation carriers included 290 unaffected, 246 breast cancer patients and 95 ovarian cancer patients. Adjusted for age at cancer diagnosis, the presence of klotho FV variant in BRCA1 carriers was associated with increased risk of breast and ovarian cancer (Table 2, hazard ratio (HR) 1.4, 95% confidence intervals (CI) 1.08–1.83, P ¼ 0.012). Analysis suggested Table 2 Hazard ratio (HR) and 95% confidence interval (CI) for predicting cancer among BRCA1 mutation carriers by KLOTHO variant, adjusted for age Variable Results Study population included 1115 women, 909 from the Sheba Medical Center, 162 BRCA1 carriers from the Shaare Zedek Medical Center and 44 BRCA2 carriers from Cedars-Sinai Medical Center. Sixteen patients who were diagnosed with both breast and ovarian cancers were censored according to the first cancer diagnosis (14 presented first with breast cancer and two with ovarian cancer). The patients were divided into three subsets (Table 1): non-carriers of BRCA1 or BRCA2 mutations HR P-value 95% CI Overall cancer FF vs FV 1.40 FF vs VV 0.65 Overall difference between the groups 1.08–1.83 0.42–1.02 0.012 0.060 0.003 Breast cancer FF vs FV 1.35 FF vs VV 0.68 Overall difference between the groups 0.99–1.83 0.40–1.16 0.060 0.16 0.035 Ovarian cancer FF vs FV 1.54 FF vs VV 0.63 Overall difference between the groups 0.97–2.45 0.28–1.41 0.068 0.26 0.061 The analysis was conducted on 236 non-carriers and 576 BRCA1 carriers; 55 BRCA1 carriers in which cancer diagnosis preceded genetic testing by more than 5 years or belonged to the same family were excluded. Table 1 Distribution of KLOTHO variants among unaffected and cancer patients by BRCA1 and BRCA2 mutation status BRCA1 carriers Non-BRCA1/2 carriers Median agea (years) FF, N (%) FV, N (%) VV, N (%) Healthy N ¼ 109 BC N ¼ 94 OC N ¼ 33 51 50 57 61 (56) 42 (38) 6 (6) 56 (60) 31 (33) 7 (7) 21 (64) 9 (27) 3 (9) Overall N ¼ 236 138 (58) 82 (35) 16 (7) Healthy N ¼ 290 BC N ¼ 246 OC N ¼ 95 38 45 51 195 (67) 80 (28) 15 (5) 158 (64) 72 (29) 16 (7) 60 (64) 28 (29) 7 (9) BRCA2 carriers Overall N ¼ 631 413 (65) 180 (29) 38 (6) Healthy N ¼ 101 BC N ¼ 116 OC N ¼ 31 43 46 62 16 (16) 64 (63) 21 (21) 9 (8) 82 (71) 25 (21) 1 (3) 24 (77) 6 (20) Overall N ¼ 248 26 (10) 170 (69) 52 (21) Abbreviations: BC, breast cancer; OC, ovarian cancer. Bold letters denotes summary of each group. Po0.0001 for the comparison between the frequencies of the klotho alleles among non-BRCA vs BRCA2 carriers. a Age at cancer diagnosis or at counseling for non-cancer patients. Oncogene Klotho modifies BRCA1-associated breast cancer risk I Wolf et al 28 a reduced risk of cancer among women with the VV alleles (HR 0.65, 95% CI 0.42–1.02, P ¼ 0.060). Analysis according to type of cancer suggested an association between FV status and increased risk of breast cancer (HR 1.35, 95% CI 0.99–1.83, P ¼ 0.060), and ovarian cancer risk (HR 1.54, 95% CI 0.97–2.45, P ¼ 0.068). Analysis of age at breast cancer diagnosis according to FF or FV genotype status among BRCA1 carriers was also conducted, using the Kaplan–Meier method (Figure 1). The median age at diagnosis was 48 years for FF/BRCA1, 43 years for FV/BRCA1 status and 53 for VV/BRCA1 (P ¼ 0.04 for FF/BRCA1 vs FV/BRCA1, P ¼ 0.186 for VV/BRCA1 vs FF/BRCA1 and P ¼ 0.02 for differences between the three groups). KLOTHO genotype status was not associated with significant differences in age at diagnosis of ovarian cancer among BRCA1 carriers (data not shown). The KLOTHO gene is located on chromosome 13q12, 616 kb upstream of the BRCA2 gene. Analysis of 248 BRCA2 6174delT mutation carriers showed significantly higher frequency of the FV and VV phenotypes compared with non-carriers. Thus, FF, FV and VV were detected in 10, 69 and 21% of the BRCA2 mutation carriers, compared with 58, 35 and 7% of the non-carriers (Table 1, Po0.0001 for the comparison between the two groups). Analysis of maintenance of the Hardy–Weinberg equilibrium, showed a clear deviation from equilibrium in the BRCA2 population (data not shown), suggesting linkage disequilibrium between BRCA2 6174delT mutation and the KLOTHO-V allele. We assessed possible linkage disequilibrium in all the BRCA2 carriers using the micro-satellite marker D13S171 located between the BRCA2 and the KLOTHO genes at a distance of about B350 kb from each gene. The analysis was consistent with these two genes being in linkage disequilibrium (data not shown). We then constructed a haplotype of the region between BRCA2*6174delT mutation and KL-VS allele, using nine families with both carriers and non-carriers of BRCA2*6174delT (Table 3). Linkage between BRCA2*6174delT and the KL-VS variant was noted in eight out of the nine families, thus indicating the existence of linkage disequilibrium between the two alleles. Analysis of KLOTHO variants among BRCA2 carriers showed higher frequency of the FF variant among unaffected compared with cancer patients (Table 4, 16% vs 7%, P ¼ 0.03). Yet, only 26 BRCA2 carriers, 16 unaffected and 10 cancer patients showed the FF variant and significantly older age was observed among unaffected compared with affected carriers (median age 49 years vs 43 years, Po0.001). Over-expression of klotho in breast cancer cells inhibits clonal growth (Wolf et al., 2008). To elucidate whether the klotho variant has reduced ability to inhibit growth of breast cancer cells, a series of colony formation assays was conducted in breast cancer cell lines, which contain either wild type (MCF-7, T-47D) or mutated BRCA1 (HCC-1937) (Tomlinson et al., 1998). The cells were transfected with either an empty vector (pEF), wild-type human klotho expression vector (pEFhKL) or human klotho-V expression vector, in Table 3 Allelic distribution of the markers used in the haplotype reconstruction in BRCA2*6174delT families ID Family 1 Family 2 Family 3 Family 4 Family 5 Family 6 Family 7 Family 8 Family 9 4128 6870 6000 6794 5016 5119 5151 6638 6717 6718 6719 7483 7580 8439 4847 8188 8690 8329 8387 8466 8591 8689 6174a D13S171 D13S1695 KL-VS D13S1493 + 243 245 243 245 243 245 231 245 235 243 243 245 231 235 231 243 231 235 231 245 235 243 245 245 235 243 243 245 231 245 235 243 231 235 237 245 235 243 231 243 235 245 231 243 326 332 326 332 326 332 326 328 328 332 324 332 328 328 328 332 328 328 326 328 328 332 326 334 332 336 330 332 312 328 332 334 326 326 326 328 328 332 326 328 326 332 326 328 FV FV VV FF FV FV FV VV FV FV FV FF FV FV FV FV FV FF FV FF FF FF 244 244 244 248 244 252 240 256 244 252 244 252 244 244 244 252 244 244 244 244 244 252 248 260 252 252 248 252 244 248 252 252 252 256 244 252 244 252 244 256 244 252 240 252 + + + + + + + + + + a BRCA2*6174delT mutation status. Table 4 Distribution of KLOTHO variants among BRCA2 mutation carriers by disease status Figure 1 Age at breast cancer presentation among BRCA1 carriers according to KLOTHO genotype. FF, wild-type KLOTHO; VF and VV, patients heterozygotes or homozygotes for the KLOTHO functional variant, respectively. The median age at diagnosis was 48 years for FF/BRCA1, 43 years for FV/ BRCA1 status and 53 for VV/BRCA1 (P ¼ 0.04 for FF/BRCA1 vs FV/BRCA1, P ¼ 0.186 for VV/BRCA1 vs FF/BRCA1 and P ¼ 0.02 for differences between the three groups). Oncogene Median agea (years) FF FV/VV a Healthy N (%) Breast or ovarian cancer N (%) 43 16 (16) 85 (84) 49 10 (7) 137 (93) Age at cancer diagnosis or at counseling for non-cancer patients. Klotho modifies BRCA1-associated breast cancer risk I Wolf et al 29 which phenylalanine at position 352 has been substituted to valine (pEFhKL-V). Transfected cells were cultured in media containing G418 for 2 weeks and stained to determine the number of surviving colonies. Expression of wild-type klotho reduced the number and size of surviving colonies by up to 95% (Figure 2a), while expression of the KL-V variant showed reduced growth inhibitory activity (about 70% activity, P ¼ 0.001). Klotho is a trans-membranal protein, which can be cleaved and shed (Kurosu et al., 2005; Chen et al., 2007). Klotho growth inhibitory activities in breast cancer are probably mediated mainly by its secreted form (Wolf et al., 2008; unpublished data). It has been shown that the KL-V variant secretion is reduced compared with wild-type klotho (Arking et al., 2002). To examine the effect of the V mutation on the secretion of the protein in breast cancer cells, MCF-7 cells were transfected with wild-type klotho (pEFhKL), klotho-V (pEFhKL-V) or an empty vector as control (pEF). Klotho secretion into the medium was assessed and compared with klotho expression in the cells. Although klotho-V expression in the cells was elevated compared with wild-type klotho, the secretion of the klotho-V was significantly reduced (Figure 2b). Albumin immunoblot shows that the differences in klotho levels in the medium were not a result of unequal protein precipitation. Discussion We have recently identified klotho as a potent tumor suppressor gene in breast cancer (Wolf et al., 2008). In this study, we observed that heterozygosity for the KLOTHO-V allele was associated with increased breast and ovarian cancers risk, as well as with younger age at diagnosis of breast cancer, in Ashkenazi Jewish women carriers of BRCA1 mutation. This observation is supported by studies in breast cancer cells, which indicated reduced secretion and growth inhibitory activity of the klotho variant compared with wild-type klotho. The biologic interaction between klotho and BRCA1 has not been elucidated yet. Klotho is a potent inhibitor of the IGF-1 pathway. Recent data indicate an association between mutations in BRCA1 and BRCA2 and the IGF-1 pathway in breast cancer. Thus, IGF-1 levels are increased in cancers from carriers of BRCA1 and BRCA2 mutations, and levels of the IGF-1 receptor are increased in breast cancer from BRCA1 carriers. Figure 2 Reduced growth inhibitory effect and secretion of klotho functional variant in breast cancer cells. (a) Colony formation assay. MCF7, T-47D and HCC-1973 cells were transfected with either an empty vector (pEF), human klotho (pEFKL) or human klotho-V expression vector, in which phenylalanine at position 352 has been substituted to valine (pEFhKL-V) and grown in G418 for 2 weeks. Colonies were stained with crystal violet and photographed. Left panel: quantification of at least three independent experiments. Data are shown as mean±s.d. Asterisk indicates Po0.05 for the difference between pEFKL and pEFhKL-V. (b) Secretion analysis. MCF-7 cells were transfected as in (a), 48 h after transfection medium was replaced with serum-free medium and 6 h later the medium was collected to assess klotho secretion, or cells were washed twice with PBS for the analysis of klotho expression in the cell lysate. Albumin served as a carrier for klotho precipitation in the medium, and thus served also as media loading control. Oncogene Klotho modifies BRCA1-associated breast cancer risk I Wolf et al 30 In addition, intact BRCA1 protein can lower IGF-1 receptor levels in breast cancer cells (Hudelist et al., 2007; Maor et al., 2007). Klotho-V allele shows reduced excretion and enzymatic activity (Figure 2b; Arking et al., 2002). Thus, a possible explanation of our observation is the combination of increased activation of the IGF-1 pathway because of BRCA1 mutation and reduced inhibition by klotho. Klotho over-expression specifically reduces colony formation of breast cancer cells (Wolf et al., 2008). Here, we noted reduced growth inhibition following overexpression of a klotho expression vector that contains the phenylalanine to valine substitute. The reduced growth inhibitory activity might be because of decreased secretion of klotho-V to the medium (Figure 2b). Indeed it has been shown that klotho can be shed from the cells (Chen et al., 2007), and our recent data suggest that most of klotho growth inhibitory activities in breast, as well as other cancers, are mediated by the secreted part of the protein (Wolf et al., 2008; Wolf et al., unpublished data). Thus, a possible explanation for the clinical observations is decreased shedding of the klotho variant, which halts its growth inhibitory activities. No association between the klotho variant and breast cancer was noted among non-carriers of BRCA1 or BRCA2 mutations. Breast cancers among BRCA1 mutation carriers are often HER2- and estrogen receptor and progesterone receptor-negative (triple negative phenotype). In contrast, the majority of breast cancers among non-carriers are estrogen receptorpositive and about 25% are HER2-positive. Klotho does not affect the epidermal growth factor pathway and has not been described to influence the activities of hormone receptors (Wolf et al., 2008). Thus, it cannot be ruled out that the klotho variant is associated with the development of specific subtypes of breast cancer. This can only be tested in a much larger cohort. A trend was noted for a lower cancer risk for VV compared with FF status (Table 2). This observation may be explained by the relatively small number of BRCA1 patients who harbored the VV genotype. Interestingly, a study of Ashkenazi Jews also showed differential effects of heterozygosity compared with homozygosity for the V allele (Arking et al., 2005). Thus, heterozygous individuals were at significantly lower risk for stroke, whereas homozygous individuals had significantly higher risk compared with wild-type individuals. Thus, our results may reflect a real protective effect of homozygosity for the V allele against breast and ovarian cancers. The biological explanation for these opposing effects has not been elucidated yet. We have noted linkage disequilibrium between BRCA2 6174delT mutation and the KLOTHO-V allele. We also noted in BRCA2 carriers significantly lower prevalence of the FF allele among cancer patients compared with unaffected individuals. The analysis is limited by the small number of patients with the FF phenotype in this population, and the age difference between the unaffected individuals and the cancer patients is also a possible confounding factor. Yet, a possible role for the klotho functional variant in cancer Oncogene development among Ashkenazi Jewish carriers of the BRCA2 6174delT mutation cannot be ruled out. The results of this study should be viewed with caution. This is a limited study, performed among Ashkenazi Jews carriers of specific BRCA1 and BRCA2 mutations. The role of klotho functional variant in cancer pathogenesis should therefore be explored in a larger cohort of ethnically diverse individuals. If the association of KLOTHO functional variant and breast cancer risk is confirmed in additional studies, klotho status may serve as an identifier of carriers who are at increased risk of cancer development. These individuals may benefit from more intensive follow-up or prevention measures. Materials and methods Subjects and clinical data Study population included consecutive adult women of Ashkenazi Jewish ancestry who were consulted and tested at the Oncogenetics Unit and the Institute of Oncology of the Sheba Medical Center, between 1 January 2000 to 31 December 2006, because of a family history suggestive of an inherited predisposition to breast or ovarian cancers. Women of Ashkenazi Jewish ancestry who were cancer free and had no family history of cancer served as control. All women were genotyped for the presence of three common mutations: BRCA1 (185delAG, 5383insC) and BRCA2 (6174delT). Two additional cohorts were included in the study; 44 women who were screened at the Women’s Cancer Research Institute at Cedars-Sinai cancer center in Los Angeles and found to be carriers of the BRCA2 (6174delT) mutations and 163 women who were screened at the Shaare Zedek Medical Center, Jerusalem, Israel and found to be carriers of a mutation in BRCA1 (185delAG or 5382insC). All participants were interviewed for demographic data and personal and family history of breast and ovarian cancers. Patients carrying more than one mutation (n ¼ 4) were excluded from the study. Four subsets of patients were therefore included: (i) women with no personal and family history of breast or ovarian cancer, (ii) women with breast or ovarian cancer non-carriers of the common BRCA1 or BRCA2 mutations, (iii) BRCA1 mutation carriers with or without breast or ovarian cancer and (iv) BRCA2 mutation carriers with or without breast or ovarian cancer. The study was approved by the Ethical Committees of the participating institutions and each participant signed a written informed consent. DNA extraction DNA was extracted from peripheral blood samples using the Promega DNA extraction kit (Promega Corp., Madison, WI, USA) according to the manufacturer’s instructions and tested for the three mutations common in the Ashkenazi Jewish population as described earlier (Rohlfs et al., 1997). Genetic analyses for the KL-VS variant Genotyping the KL-VS variant (rs9536314) was carried out using restriction endonuclease assays. Different methods were used at Sheba Medical Center and at Shaare Zedek Medical Center. The success rate of the genotyping was 99%. Restriction assay 1 (Sheba Medical Center). Using a naturally occurring restriction site to distinguish the wild-type Klotho modifies BRCA1-associated breast cancer risk I Wolf et al 31 D13S1493 KL-VS D13S1695 D13S171 6174delT Klotho BRCA2 280kb 270kb 67kb 370kb 1.2Mb Figure 3 The markers and polymorphisms in the BRCA2–KLOTHO region used to reconstruct the haplotype and determine linkage between the two genes. allele from the variant allele. DNA was amplified using the following primers: F: GCCAAAGTCTGGCATCTCTA and R: TTCCATGATGAACTTTTTGAGG. PCRs were preformed in a 25 ml reaction containing 50–100 ng genomic DNA, PCR buffer (Peqlab, Erlangen, Germany), 2.5 mM MgCl2, 250 nM dNTPs, 10 pmol of each primer and 1 U DNA polymerase (Peqlab). Amplification was carried out as follows: an initial denaturation step of 5 min at 94 1C, followed by 35 cycles of 94 1C for 30 s, 56 1C for 1 min, 72 1C for 30 s and a final extension step at 72 1C for 10 min. The resulting PCR products were subjected to restriction endonuclease digestion with 1U MaeIII (Roche Applied Science, Indianapolis, IN, USA) at 55 1C for 16 h and the digested PCR products were separated on a 2.8% agarose gel stained with ethidium bromide and visualized with UV light. PCR product size is 495 bp. There is an additional MaeIII restriction site in the KL-VS allele, resulting in fragments of 178, 50, 267 bp. Restriction assay 2 (Shaare Zedek Medical Center). Introducing a novel restriction site-by-site-directed mutagenesis using the following primers. F: GAAGAATGACCGACCACAG and a mismatched R: ATGAACTTTTTCTCAGATTCTTTAA PCRs were preformed in a 25 ml reaction containing 50–100 ng genomic DNA, PCR buffer (Medox Biotech, Chennai, India), 2 mM MgCl2, 200 nM dNTPs, 40 pmol of each primer, 10% DMSO and 0.25 U of SuperTherm DNA polymerase (Medox Biotech). Amplification was carried out as follows: an initial denaturation step of 5 min at 94 1C, followed by 35 cycles of 94 1C for 30 s, 50 1C for 1 min, 72 1C for 30 s and a final extension step at 72 1C for 10 min. The PCR products were subjected to restriction endonuclease digestion with 10 U DraI (Fermentas, Vilnius, Lithuania) at 37 1C for 16 h, separated on a 2.8% agarose gel stained with ethidium bromide and visualized. PCR product size 164 bp. The wild-type allele is characterized by the artificial addition of DraI restriction site resulting in fragments of 138 and 26 bp. Validation of the genetic analyses Each method was validated by direct sequencing of at least 10% of the samples, and 172 samples were also examined using the 50 nuclease assay. Both sequencing and 50 nuclease assay fully correlated with the restriction analyses. 50 nuclease assay (TaqMan). The KL-VS variant was genotyped by the 50 nuclease assay (TaqMan) on the Roche LightCycler 480 Sequence Detection System (Roche Applied science). PCR primers were forward primer 50 -GAGAAAAA GTTCATCAAAGGAACTGC-30 and reverse primer 50 -CAA TTGGCGGAACTTCATGT-30 . Probes were VIC-50 -CTCTT TCCTTTGGACCCACCTT-30 and FAM-50 -CTCTTTGCTT TGGACCCACCT-30 . The annealing temperature was 60 1C. Linkage disequilibrium between BRCA2 and klotho Marker D13S171, which is located within the APRIN gene, approximately 280 kb upstream to the BRCA2 and 335 kb downstream to KLOTHO was analyzed in all study participants to determine possible linkage disequilibrium between the two genes. Genomic DNA from each subject was PCR amplified using primer sequences and protocols as described (http://www.ncbi.nlm.nih.gov). Haplotype analysis To determine whether there is a linkage between the KLOTHO and BRCA2 genes, haplotype structure of the region spanning the distance between these two genes was constructed using three markers: D13S171, D13S1695 and D13S1493. The primer sequences for all markers were retrieved from the Genome DataBase on-line database (www.gdb.org). Markers D13S171 and D13S1695 are intergenic to BRCA2 and KLOTHO, whereas D13S1493 is located downstream to KLOTHO. The three markers span a region of approximately 1.2 Mb (distance from BRCA2 to D13S1493) (Figure 3). Nine families were selected for haplotype reconstruction, based on availability of carriers and non-carriers of the BRCA2*6174delT mutation. Genomic DNA from each subject was amplified by PCR for each marker. The forward primers of each pair of primers were labeled with FAM. Two microlitres of each PCR product were mixed with 0.5 ml of the TAMRA 500 internal size standard (Applied Biosystems Inc., Foster City, CA, USA), and 12 ml of formamide. Samples were read on the ABI Prism 3100 using the GeneScan Software (Applied Biosystems). The GeneScan raw data were analyzed using the Genotyper software to obtain the allele repeat in base pairs. Alleles obtained from the samples were used to construct the haplotype. Constructs The human klotho expression vector was constructed by subcloning the full-length cDNA isolated from a human kidney cDNA library into the pEF1 expression vector (Invitrogen, Carlsbad, CA, USA). The substitution F352V was introduced by PCR amplification using primers containing the mutation and verified by nucleotide sequencing. Cells and transfections Breast cancer cell lines were obtained from the American Type Culture Collection (Manassas, VA, USA). All transfections used LipofectAMINE 2000 (Invitrogen). Colony assays Two days after transfection with the indicated plasmids, G418 (750 mg/ml) was added to the culture media; and at day 14, the cells were stained using crystal violet. Untransfected cells were treated similarly, and all died within the 2 weeks of culture in the selection media. Quantification of the results Oncogene Klotho modifies BRCA1-associated breast cancer risk I Wolf et al 32 was performed using AlphaImager 2000 (Alpha Innotech, San Leandro, CA, USA). Klotho secretion Secretion assay was conducted as described earlier (Chen et al., 2007). Briefly, MCF-7 cells grown on six-well plates were transfected with 4 mg of pEF-hKL, pEF-V-hKL or control empty plasmid pEF. Forty-eight hours after transfection cells were washed twice with PBS and incubated with serum-free DMEM for 6 h. Media was collected and centrifuged at 3000 g for 5 min to remove detached cells. BSA (10 mg/ml) was added to the conditioned medium as a carrier protein and as a control for precipitation efficiency. The samples were precipitated with 25% TCA, kept at 20 1C for 5 min and on ice for at least 1 h, and then centrifuged for 15 min at 16 000 g. The protein pellet was washed three times with ice-cold acetone and centrifuged for 5 min at 16 000 g, dried at 100 1C for 10 min and dissolved in 2  Laemmli sample buffer at 100 1C for 10 min. Cell lysate protein extraction: after media collection, cells were washed twice with ice-cold PBS and lysed with RIPA buffer (50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 1% NP40, 0.25% Na-deoxycholate, 1 mM EDTA, 1 mM NaF) with a protease inhibitor cocktail (Sigma, St Louis, MO, USA). The cell lysate was centrifuged at 16 000 g for 15 min, and the supernatant was collected for SDS–PAGE. Western blot analysis Proteins precipitated from media or 50 mg protein extracts were loaded on 10% polyacrylamide gels, separated electrophoretically and blotted from the gel onto nitrocellulose membrane (Schleicher & Schuell Bioscience GmbH, Dassel, DE, USA). The membranes were then immunoblotted with anti-klotho KM2076 antibody (1:2000, kindly provided by Kyowa Hakko Kogy, Tokyo, Japan) or anti-albumin antibody (Dako, Glostrup, Denmark). Band intensities were quantified using ImageJ software. Statistical analysis Unaffected women were censored at age of last follow-up or at the age of the relevant prophylactic surgery, whichever came first. Risks for women with breast or ovarian cancer were censored at age of cancer diagnosis. Women with both breast and ovarian cancers were censored according to the first cancer diagnosis. The w2 test was used to compare differences in klotho allele frequencies between the study groups. Unweighted and weighted Cox regression analysis was used to evaluate specific genotype-associated HR with corresponding 95% CI. The weights were the same as used by Antoniou et al. (2005). As the unweighted and weighted methods yielded similar HR estimates, we report here the results of the unweighted analysis. Women diagnosed more than 5 years previous to the genetic test were excluded to avoid survival bias. First-degree relatives of a participant who had already been tested were also excluded from the analysis. The method of Kaplan–Meier was used to estimate the distributions of disease-free survival. Po0.05 was considered statistically significant, and all significance tests were two-tailed. Statistical analyses were done using SPSS software. The survival analysis was performed using STATA 10 SE software. Conflict of interest The authors declare no conflict of interest. Acknowledgements This research was supported by the United States-Israel Binational Science Foundation (BSF) (grant no. 2005150 to IW and HPK); the Chief Scientist Office of the Ministry of Health, Israel (grant no. 4055_3 to IW); the Koschitzky Family Foundation for Breast Cancer Research; the Israel Cancer Association Research Grant; the ‘Talpiut’ Sheba Career Development Award; the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel; and the Breast Cancer Research Foundation (to ELL). HPK is a member of the Molecular Biology Institute and Jonsson Comprehensive Cancer Center at UCLA, and holds the endowed Mark Goodson Chair of Oncology Research at Cedars-Sinai Medical Center/UCLA School of Medicine. References Antoniou AC, Easton DF. (2006). Models of genetic susceptibility to breast cancer. Oncogene 25: 5898–5905. Antoniou AC, Goldgar DE, Andrieu N, Chang-Claude J, Brohet R, Matti A et al. (2005). A weighted cohort approach for analysing factors modifying disease risks in carriers of high-risk susceptibility genes. Genet Epidemiol 29: 1–11. Antoniou AC, Sinilnikova OM, Simard J, Léoné M, Dumont M, Neuhausen SL et al. (2007). RAD51 135G’C modifies breast cancer risk among BRCA2 mutation carriers: results from a combined analysis of 19 studies. Am J Hum Genet 81: 1186–1200. Antoniou AC, Spurdle AB, Sinilnikova OM, Healey S, Pooley KA, Schmutzler RK et al. (2008). Common breast cancer-predisposition alleles are associated with breast cancer risk in BRCA1 and BRCA2 mutation carriers. Am J Hum Genet 82: 937–948. Arking DE, Atzmon G, Arking A, Barzilai N, Dietz HC. (2005). Association between a functional variant of the KLOTHO gene and high-density lipoprotein cholesterol, blood pressure, stroke, and longevity. Circ Res 96: 412–418. Arking DE, Becker DM, Yanek LR, Fallin D, Judge DP, Moy TF et al. (2003). KLOTHO allele status and the risk of early-onset occult coronary artery disease. Am J Hum Genet 72: 1154–1161. Oncogene Arking DE, Krebsova A, Macek Sr M, Macek Jr M, Arking A, Mian IS et al. (2002). Association of human aging with a functional variant of klotho. Proc Natl Acad Sci 99: 856–861. Begg CB, Haile RW, Borg A, Malone KE, Concannon P, Thomas DC et al. (2008). Variation of breast cancer risk among BRCA1/2 carriers. JAMA 299: 194–201. Cha S-K, Ortega B, Kurosu H, Rosenblatt KP, Kuro-O M, Huang C-L. (2008). Removal of sialic acid involving Klotho causes cellsurface retention of TRPV5 channel via binding to galectin-1. Proc Natl Acad Sci 105: 9805–9810. Chen C-D, Podvin S, Gillespie E, Leeman SE, Abraham CR. (2007). Insulin stimulates the cleavage and release of the extracellular domain of Klotho by ADAM10 and ADAM17. Proc Natl Acad Sci 104: 19796–19801. Hudelist G, Wagner T, Rosner M, Fink-Retter A, GschwantlerKaulich D, Czerwenka K et al. (2007). Intratumoral IGF-I protein expression is selectively upregulated in breast cancer patients with BRCA1/2 mutations. Endocr Relat Cancer 14: 1053–1062. Imura A, Iwano A, Tohyama O, Tsuji Y, Nozaki K, Hashimoto N et al. (2004). Secreted Klotho protein in sera and CSF: implication Klotho modifies BRCA1-associated breast cancer risk I Wolf et al 33 for post-translational cleavage in release of Klotho protein from cell membrane. FEBS Lett 565: 143–147. Ito S, Kinoshita S, Shiraishi N, Nakagawa S, Sekine S, Fujimori T et al. (2000). Molecular cloning and expression analyses of mouse [beta]klotho, which encodes a novel Klotho family protein. Mech Dev 98: 115–119. Kuro-o M, Matsumura Y, Aizawa H, Kawaguchi H, Suga T, Utsugi T et al. (1997). Mutation of the mouse klotho gene leads to a syndrome resembling ageing. Nature 390: 45–51. Kurosu H, Ogawa Y, Miyoshi M, Yamamoto M, Nandi A, Rosenblatt KP et al. (2006). Regulation of fibroblast growth factor-23 signaling by klotho. J Biol Chem 281: 6120–6123. Kurosu H, Yamamoto M, Clark JD, Pastor JV, Nandi A, Gurnani P et al. (2005). Suppression of aging in mice by the hormone klotho. Science 309: 1829–1833. Levy-Lahad E, Friedman E. (2007). Cancer risks among BRCA1 and BRCA2 mutation carriers. Br J Cancer 96: 11–15. Maor S, Yosepovich A, Papa MZ, Yarden RI, Mayer D, Friedman E et al. (2007). Elevated insulin-like growth factor-I receptor (IGF-IR) levels in primary breast tumors associated with BRCA1 mutations. Can Lett 257: 236–243. Matsumura Y, Aizawa H, Shiraki-Iida T, Nagai R, Kuro-o M, Nabeshima Y. (1998). Identification of the human klotho gene and its two transcripts encoding membrane and secreted klotho protein. Biochem Biophys Res Commun 242: 626–630. Ohyama Y, Kurabayashi M, Masuda H, Nakamura T, Aihara Y, Kaname T et al. (1998). Molecular cloning of rat klotho cDNA: markedly decreased expression of klotho by acute inflammatory stress. Biochem Biophys Res Commun 251: 920–925. Rohlfs EM, Learning WG, Friedman KJ, Couch FJ, Weber BL, Silverman LM. (1997). Direct detection of mutations in the breast and ovarian cancer susceptibility gene BRCA1 by PCR-mediated site-directed mutagenesis. Clin Chem 43: 24–29. Shiraki-Iida T, Aizawa H, Matsumura Y, Sekine S, Iida A, Anazawa H et al. (1998). Structure of the mouse klotho gene and its two transcripts encoding membrane and secreted protein. FEBS Letters 424: 6–10. Tomlinson GE, Chen TT-L, Stastny VA, Virmani AK, Spillman MA, Tonk V et al. (1998). Characterization of a breast cancer cell line derived from a germ-line BRCA1 mutation carrier. Cancer Res 58: 3237–3242. Urakawa I, Yamazaki Y, Shimada T, Iijima K, Hasegawa H, Okawa K et al. (2006). Klotho converts canonical FGF receptor into a specific receptor for FGF23. Nature 444: 770–774. Wolf I, Levanon-Cohen S, Bose S, Ligumsky H, Sredni B, Kanety H et al. (2008). Klotho: a tumor suppressor and a modulator of the IGF-1 and FGF pathways in human breast cancer. Oncogene: 27: 7094–7105. Wolf I, Sadetzki S, Catane R, Karasik A, Kaufman B. (2005). Diabetes mellitus and breast cancer. Lancet Oncol 6: 103–111. Yamamoto M, Clark JD, Pastor JV, Gurnani P, Nandi A, Kurosu H et al. (2005). Regulation of oxidative stress by the anti-aging hormone klotho. J Biol Chem 280: 38029–38034. Yee D. (2006). Targeting insulin-like growth factor pathways. Br J Cancer 94: 465–468. Oncogene