bioRxiv preprint doi: https://doi.org/10.1101/685271; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 1 Defective base excision repair in the response to DNA damaging agents in triple 2 negative breast cancer 3 4 Kevin J. Lee1, Cortt G. Piett2, Joel F Andrews1, Elise Mann3, Zachary D. Nagel2, and 5 Natalie R. Gassman1,3* 6 7 1Mitchell Cancer Institute, University of South Alabama, 1660 Springhill Avenue, Mobile, 8 AL 36604, USA 9 2Harvard T.H. Chan School of Public Health, Harvard University, 220 Longwood 10 Avenue, Boston, MA 02115, USA 11 3University of South Alabama College of Medicine, 307 N University Blvd, Mobile, AL 12 36688 13 *Corresponding author: [email protected] 14 15 Running title: Defective BER in triple negative breast cancer 16 17 Funding: This work was supported by R21ES028015 from National Institutes of Health, 18 National Institute of Environmental Health Sciences. The authors declare there are not 19 conflicts of interest. Editorial support was provided by the Dean’s Office, University of 20 South Alabama College of Medicine. 21 Competing interests statement: The authors declare there are no conflicts of interest. 22 1 bioRxiv preprint doi: https://doi.org/10.1101/685271; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 23 Abstract 24 DNA repair defects have been increasingly focused on as therapeutic targets. In hormone 25 positive breast cancer, XRCC1-deficient tumors have been identified and proposed as 26 targets for combination therapies that damage DNA and inhibit DNA repair pathways. 27 XRCC1 is a scaffold protein that functions in base excision repair (BER) by mediating 28 essential interactions between DNA glycosylases, AP endonuclease, poly(ADP-ribose) 29 polymerase 1, DNA polymerase β (POL β), and DNA ligases. Loss of XRCC1 confers 30 BER defects and hypersensitivity to DNA damaging agents. BER defects have not been 31 evaluated in triple negative breast cancer (TNBC), for which new therapeutic targets and 32 therapies are needed. To evaluate the potential of XRCC1 as an indicator of BER defects 33 in TNBC, we examined XRCC1 expression and localization in the TCGA database and in 34 TNBC cell lines. High XRCC1 expression was observed for TNBC tumors in the TCGA 35 database and expression of XRCC1 varied between TNBC cell lines. We also observed 36 changes in XRCC1 subcellular localization in TNBCs that alter the ability to repair base 37 lesions and single-strand breaks. Subcellular localization changes were also observed for 38 POL β that did not correlate with XRCC1 localization. Basal levels of DNA damage were 39 also measured in the TNBC cell lines, and damage levels correlated with observed 40 changes in XRCC1 expression, localization, and repair functions. The results confirmed 41 that XRCC1 expression changes may indicate DNA repair capacity changes but 42 emphasize that basal DNA damage levels along with expression and localization are 43 better indicators of DNA repair defects. Given the observed over-expression of XRCC1 in 44 TNBC preclinical models and the TCGA database, XRCC1 expression levels should be 45 considered when evaluating treatment responses of TNBC preclinical model cells. 2 bioRxiv preprint doi: https://doi.org/10.1101/685271; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 46 Keywords: triple negative breast cancer, base excision repair, XRCC1, PARP1, DNA 47 repair, DNA damage, nuclear localization 3 bioRxiv preprint doi: https://doi.org/10.1101/685271; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 48 Introduction 49 Defects in DNA damage response and repair are driving factors in carcinogenesis 50 and key determinants in the response to chemotherapy. Breast cancers may display 51 defects in DNA repair such as mutations in key DNA damage response and repair 52 proteins such as breast cancer-susceptibility (BRCA1/2) and tumor suppressor protein 53 p53 (TP53), and altered expression levels of DNA repair proteins thymine-DNA 54 glycosylase (TDG) and poly(ADP-ribose) polymerase 1 (PARP1) [1-3]. Therapeutic 55 outcomes may be improved by exploiting DNA repair defects present in cancer cells but 56 absent in normal cells, as in the use of PARP-inhibitors (PARPi) in cancers that have 57 BRCA1/2 deficiencies. However, characterization of DNA repair pathways often is lacking 58 in preclinical models and cell lines. Examining DNA repair defects in preclinical models 59 and patients is essential for evaluating the efficacy of therapeutic agents. 60 In breast cancer, DNA repair defects often extend beyond homologous 61 recombination defects and observed changes in the expression of base excision repair 62 (BER) proteins such as TDG and DNA polymerase beta (POL β). These defects imply 63 that BER contributes to genomic instability and may alter therapeutic response [2, 3]. 64 Single nucleotide polymorphisms in BER genes such as POL β, PARP1, and X-ray cross 65 complementing protein 1 (XRCC1) may be linked to increased risk of developing breast 66 cancer [4-7]. XRCC1 facilitates critical protein-protein interactions at the site of DNA 67 damage with DNA glycosylases, AP endonuclease 1 (APE1), PARP1, POL β, and DNA 68 ligase III (LIG3) and functions in the overlapping of BER and single-strand break repair 69 (SSBR) pathways [8, 9]. XRCC1 also participates in double-strand break (DSB) repair 4 bioRxiv preprint doi: https://doi.org/10.1101/685271; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 70 through its interaction with PARP1 in the error-prone alternative nonhomologous end 71 joining (NHEJ) [10-12], and with DNA LIG3 in nucleotide excision repair (NER) [13]. 72 XRCC1 is ubiquitously expressed in normal tissues, but low levels of XRCC1 73 causing impaired BER may occur in terminally differentiated muscle cells and neurons 74 [14, 15]. Variations in XRCC1 expression levels have been observed in breast cancer 75 patient samples [16-18], and breast cancers with low expression levels of XRCC1 have 76 been proposed as targets for PARPi treatment [16-19]. Defects in BER and XRCC1 may 77 sensitize breast cancer cells to PARPi, similar to homologous recombination and 78 BRCA1/2 defects [16, 18, 20, 21]. However, the mechanism of sensitization of XRCC1- 79 deficient cancer cells to PARPi therapy remains unclear. 80 Defects in BER including XRCC1 expression level changes have not been 81 explored in triple negative breast cancer (TNBC). TNBC may be aggressive and un- 82 responsive to current treatments that target estrogen receptor (ER), progesterone 83 receptor (PR), and human epidermal growth factor receptor 2 (HER2). However, TNBCs 84 have a high prevalence of DNA repair defects, though not always associated with 85 mutations in the BRCA1/2 genes [1, 16-18, 22]. 86 It is important to understand DNA repair defects that may promote resistance to 87 available treatment. Characterizing and targeting XRCC1-related DNA repair defects in 88 TNBCs may provide new treatment strategies for this aggressive breast cancer subtype. 89 We examined XRCC1 expression levels and DNA repair functions in four TNBC model 90 cell lines that are used commonly for preclinical screening of therapeutics. These models 91 have higher basal levels of DNA damage that correlate with XRCC1 levels or XRCC1 92 cytosolic localization, and affect their sensitivity to DNA damaging agents. These 5 bioRxiv preprint doi: https://doi.org/10.1101/685271; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. 93 observations lead us to hypothesize that XRCC1 expression and localization may be 94 potential markers to develop treatment options for TNBC. 95 96 Materials and Methods 97 Cell culture 98 MDA-MB-157 (MDA-157), MDA-MB-231(MDA-231) and MDA-MB-468 (MDA- 99 468), MCF10A, and HCC1806 cells were purchased from the American Type Culture 100 Collection (ATCC #’s HTB-24, HTB-26, HTB-132, CRL-10317, and CRL-2335, 101 respectively) within the last 12 months and passaged < 15 times for all experiments. 102 Cells were tested biweekly during experiments using a mycoplasma detection kit to 103 confirm absence of mycoplasma contamination using the Lonza MycoAlert® (Lonza 104 #LT07-318). HCC1806 cells were grown in RPMI 1640 Medium (Life Technologies 105 #11875093) and supplemented with 10% Fetal Bovine Serum (FBS, Atlantic Biologicals 106 Premium Select). MDA-157, MDA-231, MDA-468, and MCF10A cells were grown in 107 DMEM High Glucose + GlutaMAX™ (Life Technologies #10566016) and supplemented 108 with 1% sodium pyruvate (Life Technologies #11360070) and 10% FBS. Cells were 109 maintained in a humidified 37°C incubator with 5% carbon dioxide. 110 Fluorescence multiplex host cell reactivation 111 Fluorescence multiplex host cell reactivation (FM-HCR) assays were performed 112 as described previously [23, 24]. Cells were seeded 48 hours before transfection into 113 T25 tissue culture flasks (Thermofisher,156367) and collected for transfection at 85% 114 confluence. Cells were electroporated using the Neon transfection system 6 bioRxiv preprint doi: https://doi.org/10.1101/685271; this version posted June 27, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder.