Breast cancer-associated SNP rs72755295 is a cis-regulatory variation for human EXO1

Abstract

Breast cancer is the most common malignant tumor in women. A previous genome-wide association study reports that rs72755295, a SNP locating at intron of EXO1 (exonuclease 1), is associated with breast cancer. Due to the complete linkage disequilibrium between rs72755295 and rs4149909, a nonsynonymous mutation for EXO1, rs4149909 is supposed to be the causal SNP. Since EXO1 is overexpressed in breast carcinoma samples, we hypothesized that the genetic variations in this locus might confer breast cancer risk by regulating EXO1 expression. To substantiate this, a functional genomics study was performed. The dual luciferase assay indicated that G of rs72755295 presents significantly higher relative enhancer activity than A, thus verifying that this SNP can influence gene expression in breast cell. Through chromosome conformation capture it was disclosed that the enhancer containing rs72755295 can interact with the EXO1 promoter. RNA-seq analysis indicated that EXO1 expression is dependent on the rs72755295 genotype. By chromatin immunoprecipitation, the transcription factor PAX6 (paired box 6) was recognized to bind the region spanning rs72755295. In electrophoretic mobility shift assay, G of rs72755295 displays obviously higher binding affinity with nuclear protein than A. Our results indicated that rs72755295 is a cis-regulatory variation for EXO1 and might confer breast cancer risk besides rs4149909.

Keywords
Breast cancer; EXO1; rs72755295; rs4149909; expression regulation

Introduction

Breast cancer is the most common malignant tumor and one of the most important causes of cancer-related mortality among women worldwide (Sung et al., 2021). The predisposing factors of breast cancer can be divided into many environmental risk factors, including alcohol intake, obesity, endogenous hormone exposure and physical inactivity, and genetic susceptibility (Monninkhof et al., 2007, 2009; Lynch et al., 2011; Rojas and Stuckey, 2016; Huang et al., 2019). To disclose the potential genetic contribution for breast cancer, many genome-wide association studies (GWAS) have been carried out for this disease (see GWAS catalog at https://www.ebi.ac.uk/gwas/ for detail). In one GWAS, a genetic marker in EXO1 (exonuclease 1) intron region, rs72755295, was identified to be associated with breast cancer in Caucasians (Michailidou et al., 2015). Due to the fact that this SNP is in strong linkage disequilibrium (LD) with a missense SNP in EXO1, rs4149909 (p.Asn279Ser), and that EXO1 is suggested to be an oncogene (Liberti and Rasmussen, 2004; Keijzers et al., 2018), the GWAS signal in this locus is proposed to result from rs4149909 (Michailidou et al., 2015). On the other hand, EXO1 expression is observed to be significantly elevated in breast tumor tissue (Muthuswami et al., 2013; Qi et al., 2019; Liu and Zhang, 2021), hinting that the cis-regulatory variations for EXO1 might also contribute to breast cancer risk. However, this issue has hardly been surveyed.

In the current study, we hypothesized that the genetic variations in this locus, i.e., rs72755295, rs4149909 and/or other SNP(s) in LD with them, might have the ability to regulate EXO1 expression. Functional genomics approach was used to investigate this possibility.

Material and Methods

1000 Genomes project data analysis

The genotype of +/-100 kb region surrounding rs72755295 was retrieved for all 26 populations from the 1000 Genomes project public dataset (http://www.internationalgenome.org/). The LD pattern was determined by ldSelect (Carlson et al., 2004) or Genome Variation Server (http://gvs.gs.washington.edu/GVS150/) with r² threshold as 0.8.

Plasmid construct and mutagenesis

PCR primers were designed by primerselect 7.0 (DNASTAR Inc, Madison, WI) and synthesized (Sangon Biotech, Shanghai, China). rs72755295 and rs4149909 surrounding regions (~1.5 kb) were amplified with Q5 High-Fidelity DNA Polymerase (NEB, Ipswich, MA) and primers shown in ^{Table S1}. Thermocycling conditions for routine PCR was as follows: 98 ℃ for 30 s; 35 cycles of 98 ℃ for 10 s, 68 ℃ for 30 s, 72 ℃ for 45 s, and finally 72 ℃ for 2 min. The PCR product and pGL3-promoter vector (Promega, Madison, WI) were digested by MluI and XhoI (NEB), purified by GeneJET Gel Extraction Kit (Thermo Fisher Scientific) and ligated by T4 DNA ligase (NEB) according to the manufacturer’s manual. The recombinant plasmids were transformed into E.coli DH5α competent cells (Takara, Dalian, China), cultured, and then extracted by TIANpure Midi Plasmid Kit (Tiangen Biotech, Beijing, China). After sequencing, the plasmids with corresponding alleles were generated by Q5 Site-Directed Mutagenesis Kit (NEB) and primers in ^{Table S1}. Before transfection, all plasmids were sequenced to rule out artificial mutations and verify the haplotype orientation.

Cell culture, transient transfection and Dual Luciferase Reporter Gene Assay

Human breast cancer cell line MCF-7 was cultured in DMEM (High Glucose, Biological Industries, Cromwell, CT) with 10% fetal bovine serum (Biological Industries) and 1% penicillin-streptomycin solution (Solarbio, Beijing, China) and incubated in 5% CO₂ at 37 ℃. MCF-7 cells were seeded into 24-well plate at density 1.0 × 10⁴ cells/well and transfected after 24 hours. The recombinant plasmid DNA (475 ng) was transfected into MCF-7 cells using Lipofectamine 2000 (Thermo Fisher Scientific) according to the manufacturer’s recommendation. Cells were harvested and lysed by passive lysis buffer (Promega) after 36 hours culture. Co-transfection of Renilla luciferase reporter (pRL-TK, 25 ng, Promega) plasmid was performed as an internal control along with the recombinant plasmid. Luciferase activity was read by the Dual-Luciferase Reporter Assay System (Promega) using GloMax Navigator (Promega) with a costar® 96-well white polystyrene plate according to the manufacturer’s protocol. The relative enhancer activity was expressed as the ratio between firefly and Renilla luciferase and the empty pGL3-promoter vector was utilized as the negative control. Six independent replicates were carried out for each experiment.

Chromosome conformation capture (3C)

3C technology was utilized to detect the long-distance interaction between enhancer and promoter of nearby genes and the ligation frequency was quantified by real-time PCR. Generally, ~10⁸ MCF-7 cells in the logarithmic growth phase were detached with 0.25% Trypsin-0.02% EDTA Solution (Solarbio) and harvested into a 50 mL Conical Sterile Polypropylene Centrifuge Tube (Corning life sciences, Tewksbury, MA). MCF-7 cells were resuspended in 10 ml DMEM medium and cross-linked with 278 μl 37% formaldehyde (1% final concentration) for shaking 10 min at room temperature. The cross-linking reaction was terminated by adding 500 μl 2.5 M glycine (0.125 M final concentration) and incubated for 15 min on ice. After centrifugation, cells were lysed with lysis buffer containing protease inhibitor cocktail (Sigma, St. Louis, MO). The chromatin was digested by BglII (1000 units, NEB) at 37 ℃ for 12 hours with 900 rpm shaking and the digestion products were assessed by 0.8% agrose gel electrophoresis. After ligation by high concentration T4 DNA ligase (10000 units; NEB), the products were treated overnight with 15 μl proteinase K (20 mg/ml; Sigma) at 65 ℃ with 300 rpm shaking. After 30 μl RNase A (10 mg/ml; Takara) treatment at 37 ℃ for 45 min, DNA was isolated by the phenol-chloroform method.

The BAC (bacterial artificial chromosome) RP11-610O24 containing partial 1q43 region was obtained from BACPAC Resources Center (http://bacpac.chori.org/), cultured in LB medium supplemented with chloramphenicol, extracted by the Large-Construct Kit (Qiagen, Valencia, CA), digested by BglII (NEB) as abovementioned, ligated as control and also recovered by the phenol-chloroform method.

The relative amount of 3C product was measured by real-time PCR in a CFX96TM Real-time Detection System (Bio-Rad, Hercules, CA) with unidirectional primers listed in ^{Table S2}. The enrichment for MCF-7 cells relative to BAC was calculated using the comparative Ct method. All 3C PCR products were verified by resequencing.

RNA-seq analysis

The RNA-seq data (sra format) for lymphoblastoid cell lines (LCL; Montgomery et al., 2010; Jadhav et al., 2019) was obtained from the SRA database (https://www.ncbi.nlm.nih.gov/sra/) and converted into fastq format by SRA toolkit (https://github.com/ncbi/sra-tools). After alignment with the EXO1 mRNA sequence by bowtie2 (Langmead and Salzberg, 2012), the expression was calculated by RSEM (Li and Dewey, 2011) with default parameter and reported as FPKM (Fragments Per Kilobase of transcript per Million fragments mapped). The genotypes for LCLs were obtained from the 1000 Genomes or HapMap project.

Chromatin immunoprecipitation (ChIP)

TRANSFAC database (http://www.gene-regulation.com/) was used to predict potential transcription factors (TF). ChIP was performed in MCF-7 cell lines by EZ ChIP Kit (Millipore, Burlington, MA) according to the manufacturer’s instruction. In brief, formaldehyde (1% final concentration) was used to cross-link the proteins to the DNA for 10 min at 25 ℃ in ~10⁷ cells. Glycine (0.125 M final concentration) was added to quench the formaldehyde and terminate the cross-linking reaction. Cells were rinsed twice with 10 mL cold PBS, scraped thoroughly with a cell scraper, transferred into 50 mL tube and centrifuged at 1000 g for 5 min at 4 ℃. Cell pellets were resuspend in ChIP lysis buffer and incubated for 10 min on ice. Cells lysates were sonicated to shear DNA into an average fragment size of 200-1000 bp by Ultrasonic Homogenizer (Scientz Biotechnology, Ningbo, China) and the fragment sizes were analyzed on a 1.5% agarose gel. Chromatin samples were diluted with 10-fold dilution buffer, and precleared with protein A beads for 1 h at 4 ℃. For immunoprecipitation, the sheared chromatin was incubated with related mouse antibodies or normal mouse IgG (Santa Cruz Biotechnology, Santa Cruz, CA) at 4 ℃ overnight, respectively, and precipitated with 5000 g for 1 min at 4 ℃ by protein A beads. The immunoprecipitated protein/chromatin complex was washed as follows: once by low salt, high salt, LiCl wash buffer and twice by TE buffer. After washing, the protein/chromatin complex was resuspended in elution buffer. Cross-linking was reversed and protein was digested by proteinase K (Sigma). DNA was purified using GeneJET Gel Extraction Kit (Thermo Fisher Scientific) and quantified by real-time PCR to assess the enrichment by iQ SYBR green (Bio-Rad) and primer pair ACAGTTGCCAGTAGTAGTCTTTTA and TCTCATATCATCCTAGCCAACAAT.

Electrophoretic mobility shift assay (EMSA)

Nuclear proteins were isolated from human MCF-7 cells using Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime, Shanghai, China) and protein concentration was measured in an Epoch Microplate Spectrophotometer (BioTek, Winooski, VT) with Enhanced BCA Protein Assay Kit (Beyotime). The probes for both alleles of rs72755295 listed in ^{Table S3} were labeled with 3′-Biotin by EMSA probe biotin labeling Kit (Beyotime). In brief, 10 fmols of the biotin-labeled probes were incubated with 5 μg of nuclear proteins for 20 min at room temperature. The DNA-protein complexes were run on a 4.9% native polyacrylamide gel, transferred to a positively charged nylon membrane (Beyotime) and cross-linked by UV-light. For each electrophoresis, the biotin-labeled probes without nuclear protein and probe-protein complex incubating with competitor oligonucleotides (non-labeled probes) were also included as controls. After blocking and incubating with streptavidin-HRP (horseradish peroxidase) conjugate, the membrane was visualized by chemiluminescent EMSA kit (Beyotime) according to the manufacturer’s protocol. The 8-bit images of EMSA signals were captured by a Luminescent Imaging Workstation (Tanon, Shanghai, China).

Statistics

Student’s t-test was performed by SPSS 20.0 (IBM, Armonk, NY) to evaluate the luciferase expression difference among plasmid constructs, the enrichment for 3C and ChIP and EXO1 expression between different genotype groups. The null hypothesis was rejected when P< 0.05.

Result

Genetic variations nearby rs72755295

Within the 200 kb region surrounding rs72755295, there are ~1200 SNPs in each population and two distinct LD patterns could be observed. In all populations from Europe, South Asia and America, only one variation, rs4149909 (~10.4 kb away from rs72755295), shows complete LD (r² =1) with rs72755295 and the minor allele (G for both SNPs) frequency varies from 1% to 5% (see ^{Table S4}), which is consistent with previous observation in Caucasians (Michailidou et al., 2015). All other SNPs present a relatively low LD with rs72755295 (all r² <0.11; result not shown). In contrast, in all populations from East Asia and Africa, these two SNPs are not in polymorphism (see ^{Table S4}).

Function of rs72755295 and rs4149909 in regulating gene expression

To investigate the role of the two SNPs on gene expression regulation, we constructed a luciferase plasmid containing the surrounding region of these two SNPs and generated the plasmid with another allele by mutagenesis. For rs4149909, no significant difference was observed in luciferase activity between the A and G allele (P=0.22; ^{Figure S1}), indicating that rs4149909 does not have the function to alter gene expression. In contrast, the G allele of rs72755295 shows ~29.6% higher relative luciferase activity than A (P=0.0022; Figure 1), which indicates that rs72755295 is a functional site and can regulate gene expression in breast cells. rs72755295 is located in an intron region of EXO1 and not within the promoter of any known gene. Moreover, there are multiple H3K27Ac and H3K4me1 peaks nearby rs72755295 in human mammary epithelial cell (^{Figure S2}), two frequent histone modifications in active enhancer (Calo and Wysocka, 2013). Therefore, it is reasonable to hypothesize that rs72755295 is within an enhancer region and can alter enhancer activity.

Figure 1 -
Relative luciferase activity for different rs72755295 alleles in MCF-7 cell. The x axis represents relative enhancer activity. All data are expressed as mean ± standard deviation (SD). * indicates P<0.01.

Interaction between EXO1 promoter and the enhancer containing rs72755295

Given that rs72755295 is within an enhancer region, remains unclear whether its target gene is EXO1. 3C was utilized to examine whether the enhancer region could physically interact with the EXO1 promoter. In our assay, the constant primer was set in the enhancer containing rs72755295 while the anchoring primers were set in the EXO1 promoter and eleven random regions (^{Table S2}). As shown in Figure 2, a strong ligation frequency was detected in the EXO1 promoter region (corresponding to 10^th point in x-axis, ~24.7 kb away from the enhancer). A one-sample t-test utilized to compare the ligation frequency between EXO1 promoter and other regions in our assay, revealed a significant deviation (P<10^-6), thus suggesting that EXO1 should be the regulation target of this enhancer in breast cells.

Figure 2-
Interaction efficiency between the enhancer containing rs72755295 and surrounding genome regions in 1q43. The x axis indicates the location of restriction fragments in chr1 (relative to human genome build 37) while the y axis shows the relative interaction efficiency. The above arrow shows the schematic EXO1 position and transcript direction. All data is shown as mean±SD.

Association between rs72755295 genotype and EXO1 expression

If rs72755295 can indeed influence EXO1 expression, this SNP should be an expression quantitative trait locus (eQTL) for this gene. To verify this issue, RNA-seq data from LCL, a well-established model for eQTL analysis, were obtained from the literature (Montgomery et al., 2010; Jadhav et al., 2019) and EXO1 expression was calculated. Since A is fixed for rs72755295 in CHS (Southern Han Chinese) and YRI (Yoruba in Ibadan, Nigeria; see ^{Table S4}) populations, only their CEU (Utah Residents with Northern and Western European Ancestry) data were included for analysis (Jadhav et al., 2019). In both CEU datasets, no individuals are homozygous for the G allele at rs72755295 due to the low frequency. Therefore, an independent t-test was utilized to compare the EXO1 expression between the A/A and A/G group. As shown in Figure 3A, the average EXO1 expression is ~136.9% higher in the A/G group than in A/A (P=0.001) for the dataset from literature (Jadhav et al., 2019). A similar result was obtained for the data from another literature (P=0.009; Figure 3B; Montgomery et al., 2010), which is consistent with our luciferase result and confirms that rs72755295 is an eQTL for EXO1.

Figure 3-
Relationship between rs72755295 genotype and EXO1 expression in LCL at CEU population from literature Jadhav et al. (2019) (A) and Montgomery et al. (2010) (B). The expression is displayed as FPKM.

Related TF binding rs72755295

Based on the fact that rs72755295 is located in an enhancer, it seems that it might interact with TF and might influence TF binding affinity. The prediction by Match at TRANSFAC indicated that rs72755295 might be within a binding site of PAX6 (paired box 6) and could alter the binding affinity of this transcription factor. To verify this prediction, ChIP was carried out with related antibody in MCF-7 cell line and real-time PCR was used to evaluate the relative chromatin enrichment. As shown in Figure 4, compared with IgG, the region containing rs72755925 is immunoprecipated by the PAX6 antibody (P=0.00050), thus confirming that PAX6 can bind the rs72755925 surrounding region in MCF-7.

Figure 4 -
Enrichment of the chromatin spanning rs72755295 in MCF-7 cell line. The y axis represents relative enrichment. The result is normalized by input and the data is expressed as mean±SD. * indicates P<0.001.

TF binding affinity difference between rs72755295 alleles

To verify the binding capacity difference between rs72755295 alleles, EMSA was performed with nuclear extract prepared from MCF-7 cells. It can be observed that there is a specific protein-DNA complex band composed of the core sequence containing rs72755295 and nuclear proteins (Figure 5). Moreover, the G allele of rs72755295 shows an apparently higher binding affinity with nuclear protein than the A allele (see Figure 5), which is consistent with our luciferase result.

Figure 5 -
Difference in the binding affinity between MCF-7 nuclear proteins and rs72755295 alleles in EMSA. The top line indicates different alleles. NE denotes nuclear protein, and the arrow points out the position of protein-probe complex.

Discussion

In the present research, population genetics and functional genomics approaches were utilized to explore the potential cis-regulatory variations for EXO1, which might further contribute to breast cancer predisposition. To achieve this goal, the 1000 Genomes project data were recruited and the LD pattern was surveyed in this locus. As a result, only rs4149909 was identified to be in complete LD with rs72755295 in multiple populations. Further luciferase and ChIP assays verified that only rs72755295 can regulate gene expression in breast tissue by altering the binding affinity of the transcript factor PAX6. By 3C, the target gene, EXO1, was disclosed for this enhancer. Our effort provides more insight into the expression regulation of EXO1.

EXO1, locating at chromosome 1q42-43, has 14 exons spanning over ~41.7 kb and yields a ~3 kb mRNA transcript (Tishkoff et al., 1998). EXO1 is a 5’ to 3’ exonuclease protein (Schmutte et al., 1998; Lee and Wilson, 1999) and also with ability of 3’-5’ exonucleolytic degradation of DNA (Genschel et al., 2002), thus playing an essential role in DNA repair, replication and recombination (Qiu et al., 1999; Tran et al., 2004; Mimitou and Symington, 2008; Nimonkar et al., 2011). The link between EXO1 and cancer is intriguing and usually interpreted by two distinct models (Liberti and Rasmussen, 2004). The deficiency of EXO1 activity induced by germline mutation can lead to the inactivation of DNA mismatch repair pathway, hypermutation in genome and further predispose the carriers to develop cancer (Liberti and Rasmussen, 2004; Keijzers et al., 2016). This model has been verified by the observation in human hereditary nonpolyposis colorectal cancer (Wu et al., 2001) and mouse model with EXO1 knockout (Wei et al., 2003; Bardwell et al., 2004; Schaetzlein et al., 2013). Alternatively, a higher EXO1 expression or activity will induce the increase of recombination rate, impaired repair of DNA double-strand breaks, telomere resection and activation of Ras/PI3K signaling pathway, which may also further increase the cancer susceptibility (Liberti and Rasmussen, 2004; Muthuswami et al., 2013). In the case of breast, our results indicate that the risk allele, G of rs72755295 (Michailidou et al., 2015), can cause a higher EXO1 expression. Moreover, an increased expression of EXO1 has been frequently observed in tumor tissues compared with normal breast ones (Kretschmer et al., 2011; Muthuswami et al., 2013; Qi et al., 2019; Saha et al., 2019; Liu and Zhang, 2021). All these results hint that the latter model may play a more important role in the association between rs72755295 and breast cancer. An EXO1 overexpression assay or genome editing on this locus and followed by cell function investigation will shed more light on the effect of rs72755295 in tumorigenesis.

Our eQTL analysis indicated that EXO1 expression is dependent on the genotype of rs72755295. To further validate this issue, we searched the GTEx Portal database (https://gtexportal.org/; GTEx Consortium, 2017) but no association was observed (result not shown). This might be due to the relatively low frequency of rs72755295 G allele, which could decrease the power of statistical testing. In addition, the potential correlation might be influenced by some environmental or physiological effects as suggested (Gagneur et al., 2013).

Our result suggests that the cis-regulation of rs72755295 on EXO1 expression is dependent on PAX6 in breast cells. To further validate this issue, we downloaded RNA-seq data for breast tissues (Wenric et al., 2017), calculated EXO1 and PAX6 expression as described above and performed a correlation analysis. As shown in ^{Figure S3}, there is a significant correlation between EXO1 and PAX6 expression (r=0.528, P=0.0046), which is consistent with our conclusion. It is also useful to compare the correlation between A/A and A/G group. However, due to the small sample size of the A/G group, the comparison might not be with enough power to display the binding affinity difference. Moreover, it is interesting to observe that the knockdown of PAX6 can remarkably inhibit cell viability, DNA synthesis and colony formation in breast cancer cell line and tumorigenesis in xenograft nude mice (Zong et al., 2011). Considering the role of EXO1, it might be proposed that PAX6 plays this role, at least partially, through trans-regulation of EXO1.

Besides breast cancer, this locus is also suggested to be associated with pancreas (Dong et al., 2011), colon (Madi et al., 2018) and keratinocyte (Liyanage et al., 2019) cancer. Interestingly, EXO1 overexpression in tumor cells compared with corresponding normal ones is also observed in multiple human tissues, including liver (Dai et al., 2018; Yang et al., 2020), lung (Zhou et al., 2021), pancreas and colon (Rasmussen et al., 2000). Moreover, a database search through UALCAN (http://ualcan.path.uab.edu/index.html; Chandrashekar et al., 2017) confirms that significant EXO1 overexpression in cancer cell is appearing in almost all tissue types from TCGA (The Cancer Genome Atlas) project (results not shown). Considering this and the ubiquitous spread of EXO1 and PAX6 (see http://biogps.org) in human tissues, it might be proposed that the putative enhancer and rs72755295 might be also involved in the carcinogenesis in abovementioned tumor types, which deserves further investigation.

Platinum salts have been widely utilized in chemotherapy of multiple human cancer types and act through crosslinking with DNA, causing DNA damage and further inducing cancer cell apoptosis (Dasari and Tchounwou, 2014). EXO1 can excise the adducted nucleotide and mediate DNA repair, which might lead to resistance in platinum salts treatment. Therefore, a lower EXO1 expression will be beneficial to cancer patients in platinum salts treatment, which has been validated in ovarian cells (Zhou et al., 2014; He et al., 2020). To validate this issue, we also searched TCGA data through EviCor database (https://www.evicor.org/; Petrov and Alexeyenko, 2022). For breast invasive carcinoma (BRCA) patients with lower EXO1 expression, carboplatin treatment can promote patients' survival (see ^{Figure S4}). In contrast, for BRCA patients with higher EXO1 expression, the same treatment presents a higher risk of death (P=0.0304; see ^{Figure S4}), which is consistent with a previous proposal (Zhou et al., 2014; He et al., 2020). Considering the role of rs72755295 on EXO1 expression regulation, this SNP might contribute to the difference in platinum salts response among cancer patients, which has been preliminarily verified by a recent pharmocogenetics study in advanced colorectal cancer (Madi et al., 2018) and deserves further research.

Acknowledgements

This work was supported by the Fundamental Research Funds for the Central Universities (2018CBLY005 and GK202001004) and National Natural Science Foundation of China (No. 31370129).

References

• Bardwell PD, Woo CJ, Wei K, Li Z, Martin A, Sack SZ, Parris T, Edelmann W and Scharff MD (2004) Altered somatic hypermutation and reduced class-switch recombination in exonuclease 1-mutant mice. Nat Immunol 5:224-229.
• Calo E and Wysocka J (2013) Modification of enhancer chromatin: What, how, and why? Mol Cell 49:825-837.
• Carlson CS, Eberle MA, Rieder MJ, Yi Q, Kruglyak L and Nickerson DA (2004) Selecting a maximally informative set of single-nucleotide polymorphisms for association analyses using linkage disequilibrium. Am J Hum Genet 74:106-120.
• Chandrashekar DS, Bashel B, Balasubramanya SAH, Creighton CJ, Ponce-Rodriguez I, Chakravarthi BVSK and Varambally S (2017) UALCAN: A portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia 19:649-658.
• Dai Y, Tang Z, Yang Z, Zhang L, Deng Q, Zhang X, Yu Y, Liu X and Zhu J (2018) EXO1 overexpression is associated with poor prognosis of hepatocellular carcinoma patients. Cell Cycle 17:2386-2397.
• Dasari S and Tchounwou PB (2014) Cisplatin in cancer therapy: Molecular mechanisms of action. Eur J Pharmacol 740:364-378.
• Dong X, Li Y, Hess KR, Abbruzzese JL and Li D (2011) DNA mismatch repair gene polymorphisms affect survival in pancreatic cancer. Oncologist 16:61-70.
• Gagneur J, Stegle O, Zhu C, Jakob P, Tekkedil MM, Aiyar RS, Schuon A-K, Pe’er D and Steinmetz LM (2013) Genotype-environment interactions reveal causal pathways that mediate genetic effects on phenotype. PLoS Genet 9:e1003803.
• Genschel J, Bazemore LR and Modrich P (2002) Human exonuclease I is required for 5’ and 3’ mismatch repair. J Biol Chem 277:13302-13311.
• GTEx Consortium(2017) Genetic effects on gene expression across human tissues. Nature 550:204-213.
• He D, Li T, Sheng M and Yang B (2020) Exonuclease 1 (Exo1) participates in mammalian non-homologous end joining and contributes to drug resistance in ovarian cancer. Med Sci Monit 26:e918751.
• Huang C, Zhang Y and Zhong S (2019) Alcohol intake and abnormal expression of Brf1 in breast cancer. Oxid Med Cell Longev 2019:4818106.
• Jadhav B, Monajemi R, Gagalova KK, Ho D, Draisma HHM, van de Wiel MA, Franke L, Heijmans BT, van Meurs J, Jansen R et al (2019) RNA-Seq in 296 phased trios provides a high-resolution map of genomic imprinting. BMC Biol 17:50.
• Keijzers G, Bakula D, Petr MA, Madsen NGK, Teklu A, Mkrtchyan G, Osborne B and Scheibye-Knudsen M (2018) Human Exonuclease 1 (EXO1) regulatory functions in DNA replication with putative roles in cancer. Int J Mol Sci 20:74.
• Keijzers G, Liu D and Rasmussen LJ (2016) Exonuclease 1 and its versatile roles in DNA repair. Crit Rev Biochem Mol Biol 51:440-451.
• Kretschmer C, Sterner-Kock A, Siedentopf F, Schoenegg W, Schlag PM and Kemmner W (2011) Identification of early molecular markers for breast cancer. Mol Cancer 10:15.
• Langmead B and Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9:357-359.
• Lee BI and Wilson DM 3rd (1999) The RAD2 domain of human exonuclease 1 exhibits 5’ to 3’ exonuclease and flap structure-specific endonuclease activities. J Biol Chem 274:37763-37769.
• Li B and Dewey CN (2011) RSEM: Accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12:323.
• Liberti SE and Rasmussen LJ (2004) Is hEXO1 a cancer predisposing gene? Mol Cancer Res 2:427-432.
• Liu J and Zhang J (2021) Elevated EXO1 expression is associated with breast carcinogenesis and poor prognosis. Ann Transl Med 9:135.
• Liyanage UE, Law MH, Han X, An J, Ong J-S, Gharahkhani P, Gordon S, Neale RE, Olsen CM, MacGregor S et al (2019) Combined analysis of keratinocyte cancers identifies novel genome-wide loci. Hum Mol Genet 28:3148-3160.
• Lynch BM, Neilson HK and Friedenreich CM (2011) Physical activity and breast cancer prevention. Recent Results Cancer Res 186:13-42.
• Madi A, Fisher D, Maughan TS, Colley JP, Meade AM, Maynard J, Humphreys V, Wasan H, Adams RA, Idziaszczyk S et al (2018) Pharmacogenetic analyses of 2183 patients with advanced colorectal cancer; Potential role for common dihydropyrimidine dehydrogenase variants in toxicity to chemotherapy. Eur J Cancer 102:31-39.
• Michailidou K, Beesley J, Lindstrom S, Canisius S, Dennis J, Lush MJ, Maranian MJ, Bolla MK, Wang Q, Shah M et al (2015) Genome-wide association analysis of more than 120,000 individuals identifies 15 new susceptibility loci for breast cancer. Nat Genet 47:373-380.
• Mimitou EP and Symington LS (2008) Sae2, Exo1 and Sgs1 collaborate in DNA double-strand break processing. Nature 455:770-774.
• Monninkhof EM, Elias SG, Vlems FA, van der Tweel I, Schuit AJ, Voskuil DW and van Leeuwen FE (2007) Physical activity and breast cancer: A systematic review. Epidemiology 18:137-157.
• Monninkhof EM, Velthuis MJ, Peeters PHM, Twisk JWR and Schuit AJ (2009) Effect of exercise on postmenopausal sex hormone levels and role of body fat: A randomized controlled trial. J Clin Oncol 27:4492-4499.
• Montgomery SB, Sammeth M, Gutierrez-Arcelus M, Lach RP, Ingle C, Nisbett J, Guigo R and Dermitzakis ET (2010) Transcriptome genetics using second generation sequencing in a Caucasian population. Nature 464:773-777.
• Muthuswami M, Ramesh V, Banerjee S, Thangaraj SV, Periasamy J, Rao DB, Barnabas GD, Raghavan S and Ganesan K (2013) Breast tumors with elevated expression of 1q candidate genes confer poor clinical outcome and sensitivity to Ras/PI3K inhibition. PLoS One 8:e77553.
• Nimonkar AV, Genschel J, Kinoshita E, Polaczek P, Campbell JL, Wyman C, Modrich P and Kowalczykowski SC (2011) BLM-DNA2-RPA-MRN and EXO1-BLM-RPA-MRN constitute two DNA end resection machineries for human DNA break repair. Genes Dev 25:350-362.
• Petrov I and Alexeyenko A (2022) EviCor: Interactive web platform for exploration of molecular features and response to anti-cancer drugs. J Mol Biol 434:167528.
• Qi L, Zhou B, Chen J, Hu W, Bai R, Ye C, Weng X and Zheng S (2019) Significant prognostic values of differentially expressed-aberrantly methylated hub genes in breast cancer. J Cancer 10:6618-6634.
• Qiu J, Qian Y, Chen V, Guan MX and Shen B (1999) Human exonuclease 1 functionally complements its yeast homologues in DNA recombination, RNA primer removal, and mutation avoidance. J Biol Chem 274:17893-17900.
• Rasmussen LJ, Rasmussen M, Lee B, Rasmussen AK, Wilson DM 3rd, Nielsen FC and Bisgaard HC (2000) Identification of factors interacting with hMSH2 in the fetal liver utilizing the yeast two-hybrid system. In vivo interaction through the C-terminal domains of hEXO1 and hMSH2 and comparative expression analysis. Mutat Res 460:41-52.
• Rojas K and Stuckey A (2016) Breast cancer epidemiology and risk factors. Clin Obstet Gynecol 59:651-672.
• Saha I, Rakshit S, Wlasnowolski M and Plewczynski D (2019) Identification of epigenetic biomarkers with the use of gene expression and DNA methylation for breast cancer subtypes. Paper presented at the IEEE Region 10 Conference.
• Schaetzlein S, Chahwan R, Avdievich E, Roa S, Wei K, Eoff RL, Sellers RS, Clark AB, Kunkel TA, Scharff MD et al (2013) Mammalian Exo1 encodes both structural and catalytic functions that play distinct roles in essential biological processes. Proc Natl Acad Sci U S A 110:E2470-2479.
• Schmutte C, Marinescu RC, Sadoff MM, Guerrette S, Overhauser J and Fishel R (1998) Human exonuclease I interacts with the mismatch repair protein hMSH2. Cancer Res 58:4537-4542.
• Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A and Bray F (2021) Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71:209-249.
• Tishkoff DX, Amin NS, Viars CS, Arden KC and Kolodner RD (1998) Identification of a human gene encoding a homologue of Saccharomyces cerevisiae EXO1, an exonuclease implicated in mismatch repair and recombination. Cancer Res 58:5027-5031.
• Tran PT, Erdeniz N, Symington LS and Liskay RM (2004) EXO1-A multi-tasking eukaryotic nuclease. DNA Repair (Amst) 3:1549-1559.
• Wei K, Clark AB, Wong E, Kane MF, Mazur DJ, Parris T, Kolas NK, Russell R, Hou Jr. H, Kneitz B et al (2003) Inactivation of Exonuclease 1 in mice results in DNA mismatch repair defects, increased cancer susceptibility, and male and female sterility. Genes Dev 17:603-614.
• Wenric S, ElGuendi S, Caberg J-H, Bezzaou W, Fasquelle C, Charloteaux B, Karim L, Hennuy B, Frères P, Collignon J et al (2017) Transcriptome-wide analysis of natural antisense transcripts shows their potential role in breast cancer. Sci Rep 7:17452.
• Wu Y, Berends MJ, Post JG, Mensink RG, Verlind E, Van Der Sluis T, Kempinga C, Sijmons RH, van der Zee AG, Hollema H et al (2001) Germline mutations of EXO1 gene in patients with hereditary nonpolyposis colorectal cancer (HNPCC) and atypical HNPCC forms. Gastroenterology 120:1580-1587.
• Yang G, Dong K, Zhang Z, Zhang E, Liang B, Chen X and Huang Z (2020) EXO1 plays a carcinogenic role in hepatocellular carcinoma and is related to the regulation of FOXP3. J Cancer 11:4917-4932.
• Zhou C-S, Feng M-T, Chen X, Gao Y, Chen L, Li L-D, Li D-H and Cao Y-Q (2021) Exonuclease 1 (EXO1) is a potential prognostic biomarker and correlates with immune infiltrates in lung adenocarcinoma. Onco Targets Ther 14:1033-1048.
• Zhou J, Wang Y, Yin X, He Y, Chen L, Wang W, Liu T and Di W (2014) FOXM1 modulates cisplatin sensitivity by regulating EXO1 in ovarian cancer. PLoS One 9:e96989.
• Zong X, Yang H, Yu Y, Zou D, Ling Z, He X and Meng X (2011) Possible role of Pax-6 in promoting breast cancer cell proliferation and tumorigenesis. BMB Rep 44:595-600.

Supplementary Material

The following online material is available for this article:

Table S1 - Primers used in plasmid construction and mutagenesis.

Table S2 - Primers used in 3C-qPCR.

Table S3 - Probes for rs72755295 in EMSA.

Table S4 - r² between rs72755295 and rs44149909 and minor allele frequency in 1000 Genomes project populations.

Figure S1- Relative enhancer activity for different alleles of rs4149909.

Figure S2 - Histone modification for the region surrouding rs72755295 in breast cell.

Figure S3 - The correlation between PAX6 and EXO1 expression in breast tissues.

Figure S4 - Kaplan-Meier survival curves for BRCA patients grouped by EXO1 expression and treatment.

Edited by

Associate Editor:

Anamaria Aranha Camargo

Publication Dates

• Publication in this collection
  10 Oct 2022
• Date of issue
  2022

History

• Received
  28 Dec 2021
• Accepted
  07 Aug 2022