Radiation is used in multiple procedures as a therapeutic and diagnostic tool. However, ionizing radiation can induce mutations in the DNA of irradiated cells, which can promote tumorigenesis. As malignant transformation is a process that takes many years, there are intermediate stages of cells that have initiated the process but have not yet evolved into cancer. The study here aimed to investigate the effect of ionizing radiation on normal and partially transformed human mammary epithelial cells. Breast primary epithelial cells were derived from normal breast tissue from two different donors and modified by transduction with the SV40 small and large T antigen and hTERT genes to obtain partially transformed cells and also with HRAS to completely and experimentally transform them. After exposure to different doses of ionizing radiation, oncogenic features were analyzed by means of an anchorage-independent growth assay and 3D cell culture. The addition of radiation exposure resulted in an increase in the number and size of colonies formed in each of the conditions analyzed and in the reduction of the capacity of partially transformed cells to form properly polarized 3D structures. Moreover, partially transformed cells require lower doses of radiation than healthy cells to enhance anchorage-independent growth capacity. Although cells from different donors have a different degree of sensitivity in the response to radiation, a higher sensitivity to the radiation-induced cell transformation process was observed in those cells that had already initiated the oncogenic process, which require higher doses of radiation to complete the transformation process.

Implications:

Individuals carrying accumulation of genetic alterations may have an increased susceptibility to radiation-induced neoplastic transformation.

This article is featured in Highlights of This Issue, p. 837

Breast cancer is the most common cancer in women worldwide and the leading cause of cancer-related death among this population (1). It has been postulated that fractioned development, proliferation, and regression cycles and being a nonessential organ contribute to the increased risk of tumor formation in the mammary gland (2). During the last 15 years, the molecular profiling of patient-derived tumors has allowed the characterization of six subtypes of breast cancer based on gene-expression signatures: Basal-like subtype, Claudin-low, ErbB2-overexpressing, Luminal-B, Luminal-A, and normal subtypes (3, 4). Different subtypes have different prevalence and outcomes: although the basal-like subtype is the least frequent and correlates with the worst outcome, the luminal-A subtype is the most prevalent but with the best prognosis. Therefore, the origin of the transformed cell could determine the tumor subtype (5, 6) and thus the characteristic nature of cells used in in vitro studies should be considered when cellular models for the assessment of experimental neoplastic transformation process are developed.

Transformation of normal cells into malignant cells is a multistep process. Over a few decades, the specific characteristics of tumor cells have been extensively evaluated and identified as the hallmarks of cancer, which include the dysregulation of cellular processes such as cell proliferation and cellular energetics or the acquisition of genome instability and immortality (7). In order to further elucidate the mechanisms involved in the emergence of tumor cells, numerous in vitro transformation models have been considered. A transformation method proposed by Weinberg and colleagues is based on the introduction of three genetic elements, SV40 Large T antigen, hTERT, and HRAS, and has been proved to drive healthy cells to a malignant state in different types of cells such as human embryonic kidney cells, human fibroblasts, and myoepithelial breast cells (8, 9). Strikingly, the use of this transformation model in breast primary epithelial cells (BPEC), which are human mammary epithelial cells grown in a defined medium (WIT), gave rise to experimentally transformed cells that yielded tumors closely resembling human breast adenocarcinomas when injected into the mammary fat pad of immunosuppressed mice. It is important to note that the conditions in which BPECs are propagated have been reported to maintain luminal characteristics better than standard conditions (10). Thus, the induced experimental transformation in BPECs provides a valuable approach to study the tumorigenesis in breast epithelial cells, as adenocarcinomas expressing luminal markers are the most prevalent type of breast cancer (11).

Exposure to mutagenic agents, particularly to physical agents, has been revealed as an important etiologic factor in cancer development. The carcinogenic action of ionizing radiation in humans has been well established from epidemiologic data and the formation of solid tumors is linear in the dose range of 0.15 to 1.5 Gy (12). In the particular case of breast cancer, an increased incidence has been reported after exposure to radiation for diagnostic or therapeutic purposes. For instance, increased breast cancer risks were reported in women who received repeated fluoroscopic examinations for tuberculosis (13) or for spinal curvature (14). Furthermore, elevated breast cancer risk was also reported among women who had multiple mammograms >5 years before diagnosis (14, 15). Of importance in the present context is the identification of gaps in knowledge, as the International Commission on Radiological Protection (ICRP) classifies breast tissue among the most sensitive to radiation (16).

Knowledge of the mechanistic basis of the strong link between ionizing radiation and carcinogenesis is based on studies in healthy cells that present proper DNA-repair machinery (reviewed in Mavragani and colleagues 2017; ref. 17). However, less is known about the effects of radiation in cells that have already acquired some oncogenic characteristics. Given that carcinogenesis has a multistep nature following radiation-induced changes (18), those single cells in a premalignant state could present an increased propensity to reach a fully malignant state after exposure to ionizing radiation. As the breast is one of the most radiosensitive organs and is frequently exposed to radiation for both diagnostic and treatment purposes, it is worth studying the effect of radiation in human mammary cells with some defined genetic changes.

Altogether, we aimed to investigate the radiation effect on preneoplastic BPECs, generated by the transduction with SV40 small + large T antigens (Ttag) and hTERT, in comparison with healthy BPECs. By means of different experimental assays, the radiation-induced transformation process of healthy and preneoplastic BPECs has been monitored before and after exposure to different doses of ionizing radiation. Our results indicate a reduction in the radiation dose required by preneoplastic cells to become fully transformed compared with healthy cells. Therefore, pretumor cells could be especially susceptible to completing a radiation-induced transformation process, leading to the formation of new neoplastic events.

Cell cultures derivation and irradiation conditions

BPECs were obtained following the procedure previously described by Ince and colleagues (10) from disease-free breast tissue collected from two donors (d1 and d2) who underwent reduction mammoplasty at Clínica Pilar Sant Jordi (Barcelona) with standard procedure consent. BPECs were cultured in WIT-P-NC medium (00-0051, Stemgent; CM-0104, Cellaria) supplemented with 100 ng/mL cholera toxin (C8052; Sigma-Aldrich) using BD Primaria culture flasks (BD Biosciences, Becton Dickinson) and in a 37°C and 5% CO2 atmosphere. Cells were detached using 0.25% trypsin in 1 × phosphate-buffered solution (PBS), and the medium was replaced every 48 hours. Cells were counted by means of a Scepter Cell Counter (PHCC00000; Millipore) that allows the distinction between living and dead cells. The accumulation of population doublings (PD) was calculated from the number of viable seeded and recovered cells and determined using the following equation:

formula

When indicated, BEPCs were irradiated by exposing them to 137Cs γ-rays at a dose rate of 5 Gy/minute using an IBL 437C source. The doses used in the experiments to assess radiation-induced transformation of breast cells in the present study were 0, 0.25, 0.5, 1, and 2 Gy. Taking into account that X-ray mammograms are the main screening tool for breast cancer, it is useful to compare the radiation doses used in this study with those received by the breast during mammogram examination. The mean glandular doses (per two-view screening examination) when an X-ray mammogram is conducted are 4.5 mGy and the standard entrance skin dose is 20 mGy.

Transduction and selection of transduced cells

Post-stasis BPECs were transduced with lentiviral particles pRRL-CMV-Ttag-IRES-eGFP (d1: PD22; d2: PD11), pRRL-CMV-TERT-IRES-CherryFP (d1: PD26; d2: PD21), and pLenti CMV/TO RasV12 Puro (d1: PD38; d2: PD82). Transduced BPECs were cultured in WIT-T medium (Stemgent, 00-0047). BPECs that presented a fluorescent signal were enriched by fluorescent activated cell sorting (FACS). Cells were analyzed with a BD FACSAria II SORP (BD Biosciences) using the 488 nm and 561 nm lasers to detect either GFP or mCherry; data were obtained using the BD FACSDiva v6 software (BD Biosciences). Recovered cells were maintained on ice during transport until seeded in previously described conditions. BPECs transduced with the vector carrying antibiotic-resistance cassette were selected with puromycin (P8833, Sigma-Aldrich) at 1.47 μmol/L for 2 to 3 weeks.

RNA collection and quantitative reverse transcription PCR

Total RNA was isolated from cells using TRIzol reagent (15596-026, Thermo Fisher Scientific), treated with DNase (Invitrogene), reisolated with TRIzol, and quantified with a NanoDrop2000 spectrophotometer (Thermo Fisher Scientific). RNA was reverse transcribed into cDNA using iScript (Bio-Rad). Quantitative PCR (qPCR) was performed with SYBR green (Bio-Rad) using a CFX96 thermal cycler (Bio-Rad). The qPCR primer sequences were designed online using the Primer3 software and are listed in Supplementary Table S1. Results were normalized using GAPDH RNA levels. Fold change (FC) was obtained using the 2–ΔΔCq method.

Preparation of whole-cell lysates and Western blotting

Proteins were extracted with RIPA lysis buffer, sonicated, and quantified with the Pierce BCA Protein Assay Kit (23227, Thermo Scientific). After 10 minutes at 70°C, 25 μg of protein lysates were separated on a 10% Bis-Tris gel (NP0302, Thermo Fisher Scientific) and transferred to a PVDF membrane (LC2005, Live Technologies). Primary antibodies used were mouse anti-p16INK4a (1:200; MS-218-P, Neomarkers) and rabbit anti-GAPDH (1:1,000; 9484, Abcam). Goat anti-mouse and anti-rabbit horseradish peroxidase (HRP) conjugate (1:5,000; 12-349 and 12-348, Millipore) were used as secondary antibodies. Chemiluminescent detection of antibodies was performed with ChemiDoc XRS (Bio-Rad) using HRP solution and luminol (WBKLS0500, Immobilon Western kit; Millipore).

Cytogenetic analysis

BPECs were treated with 0.02 μg/mL colcemid for 4 hours, followed by hypotonic shock (0.075 M KCl) and fixation with Carnoy's Solution (1:3; acetic acid:methanol; v/v). Cell suspensions were dropped onto clean slides. Centromeres and telomeres were labeled by means of the PNA-FISH techniques using a Cy3-(CCCTAA)3 PNA-probe for telomeres and an FITC-AAACACTCTTTTTGTAGA PNA probe for centromeres (PE Biosystems). Cytogenetic analysis of metaphase chromosome spreads was performed with pantelomeric and pancentromeric labeling combined with reverse staining with 4′,6-diamino-2-phenylindole (DAPI), which results in a reproducible G band-like pattern that allows individual chromosomes to be accurately identified. The cytogenetics of BPECs from donor 1 was analyzed at PD 3.5 and PD 21 when nontransduced, at PD 44 when transduced with Ttag and hTERT and at PD 60 when transduced with Ttag, hTERT, and HRAS, whereas BPECs from donor 2 were analyzed at PD 8.5 and PD 34 when nontransduced, at PD 35 when transduced with Ttag and hTERT and at PD 118 when transduced with Ttag, hTERT, and HRAS.

Immunodetection of luminal and myoepithelial markers

BPECs were seeded onto glass coverslips, fixed with methanol:acetone (1:1; v/v) at −20°C for 10 minutes and rinsed twice with 1 × PBS. Samples were then blocked for 1 hour with 5% normal goat serum and 0.1% TritonX-100 in 1 × PBS at room temperature. Primary antibodies, which were diluted in blocking solution and incubated overnight at 4°C, were rabbit anti-claudin-4 (1:1,000; 15104, Abcam) and mouse anti-cytokeratin 14 (RCK107; 1:1,000; 9220, Abcam; ref. 19). After a 1 × PBS wash, slides were incubated with secondary antibodies for 2 hours at room temperature. The secondary antibodies used were anti-mouse Cy3 (1:500; 115-165-146, Jackson ImmunoResearch Laboratories) and anti-rabbit Alexa Fluor 488 (1:500; A-11034, Thermo Fisher Scientific). Finally, slides were extensively washed with 1 × PBS for 30 minutes and mounted in Vectashield Mounting Medium (H-1000, Vector Laboratories) with DAPI. Slides were analyzed using a BX60 epifluorescence microscope (Olympus Corporation).

TUNEL assay for late apoptosis detection

For the in situ detection of DNA fragmentation, a TUNEL assay (Terminal dUTP Nick-End Labeling, Roche) was conducted following the manufacturer's instructions. Briefly, cells growing on coverslips were fixed for 40 minutes with 2% paraformaldehyde at room temperature, permeabilized with 0.1% Triton X-100 and 0.1% sodium citrate in 1 × PBS for 5 minutes on ice and incubated for 40 minutes in TUNEL mix solution at 37°C. Once dried, slides were mounted with Vectashield Mounting Medium (Vector Laboratories Inc.) supplemented with DAPI. Slides were analyzed using an Olympus BX60 epifluorescence microscope (Olympus Corporation).

Senescence detection

Senescence was assessed by the detection of senescence-associated β-galactosidase (SA β-gal) activity following the procedure described by Debacq-Chainiaux and colleagues (20). Briefly, cells were fixed before confluence with 2% formaldehyde and 0.2% glutaraldehyde in 1 × PBS and then incubated for 13 hours at 37°C in staining solution that contains X-Gal substrate. Only senescent cells that present β-galactosidase activity are capable of X-Gal degradation producing a blue product. The percentage of senescent cells was obtained after counting 500 cells under an IX71 inverted microscope equipped with a DP20 camera (Olympus Corporation).

Anchorage assay and colony analysis

In order to investigate the ability of cells under the different experimental conditions to grow in suspension, their capacity to form colonies in solid agar was assessed. To this end, a total of 25.000 cells/p35 dish were individualized through 40-μm porous strainers and seeded in a medium solution with 0.3% agar settled on a solidified 0.6% agar layer. Once solidified, fresh medium was added above the cell-containing layer and replaced 3 times a week. Cells were maintained in a humidified atmosphere with 5% CO2 at 37°C. After 3 weeks in culture, medium was removed and replaced with 1 mg/mL 3-(4,5-dimethylthiazolyl-2)-2 (MTT, M2128, Sigma-Aldrich) solution (Supplementary Fig. S1). Only the colonies formed by cells with metabolic activity are able to reduce MTT to formazan, forming dark blue crystals (Supplementary Fig. S1b).

Images of the colonies were acquired with an IX71 inverted microscope (Olympus Corporation). The number of colonies and the area of each MTT-positive colony were analyzed with the ImageJ software (Supplementary Fig. 1c; script described at Supplementary data). The diameter of each colony was calculated from the area values obtained using the formula d = 2(A/π)1/2. The radius (R) of an aggrupation of n cells, each with a known radius (r) and taking into account a random packaging efficiency (P = 0.64), was calculated according to the following formula: R = (n/ρ)1/3r. Colonies with a diameter <65 μm were excluded from the analysis because, according to the settings established for this experiment, they have fewer than 8 cells (Supplementary Fig. S1a). Experiments were performed in duplicate and, after dismissing statistical differences between replicates, all of the values obtained for each sample and condition were merged in a unique graph (Supplementary Fig. S1).

Matrigel-based three-dimensional (3D) cultures and acini immunofluorescence detection

Matrigel-based 3D cultures were established as previously described by Debnath and colleagues (21). Briefly, a bottom layer of 100% Matrigel was placed onto 12-well plates. Once the bottom layer was solidified, BPECs were seeded in appropriate WIT medium containing 4% Matrigel, adding fresh medium twice a week. The dishes were maintained in a humidified atmosphere with 5% CO2 at 37°C. When indicated, cells were exposed to γ-rays and kept in conventional BPEC culture conditions for 8 weeks. After that period, cells were seeded in 3D cultures and maintained in the same conditions as the nonirradiated counterparts. For each experiment, an internal control was introduced as, due to differences in Matrigel batch numbers, density of the medium differs between experiments and so does the number of acini.

In order to perform time-lapse experiments, images were captured every 10 minutes during the first 9 hours of culture using a TCS SP5 confocal microscope (Leica Microsystems) equipped with heated stage and 5% CO2 incubation chamber. The speed of cells forming acini during aggregation process was measured using Imaris (v6.3.1, Bitplane Inc.) and Microsoft Excel 2017 (v15.39, Microsoft). The number and size of acini were quantified at indicated time points using an inverted microscope (IX71, Olympus). After 2 weeks in culture, Matrigel containing acini was gently teared and depicted on a glass slide. Once dried on a heating plate, samples were fixed with methanol:acetone (1:1, v/v) at −20°C for at least 30 minutes to remove GFP and mCherry fluorescence. Afterward, acini were blocked for 2 hours in 5% normal goat serum and 0.1% tritonX-100 in 1 × PBS. Incubations of the samples with primary and secondary antibodies, diluted in blocking solution, were performed overnight at 4°C washing the samples with 1 × PBS in between. Primary antibodies used were rabbit anti-claudin-4 (1:100; 15104, Abcam) and mouse anti-cytokeratin 14 (RCK107; 1:100; 9220, Abcam). Secondary antibodies used were anti-mouse Cy3 (1:500; 115-165-146, Jackson ImmunoResearch Laboratories) and anti-rabbit A488 (1:500; Thermo Fisher Scientific). Slides were extensively washed with 1 × PBS and mounted in Vectashield Mounting Medium (Vector Laboratories) supplemented with DAPI. Acini were analyzed using an LSM 700 confocal microscope (Carl Zeiss) and ZEN 2010 B BP1 confocal software (Carl Zeiss).

Statistical analysis

All statistical analyses were performed using Prism v7.0a (GraphPad Software). Differences between growth slopes were obtained using lineal regression and F-test with 95% confidence. In relation to the anchorage assay, differences in the colonies' number between different samples were analyzed with a two-sided χ2 test with a confidence interval of 95%, whereas the difference in the median diameter of the colonies was analyzed by Kruskal–Wallis nonparametric test; multiple comparisons were corrected by the Benjamini–Krieger–Yekutieli method (P < 0.05). Differences between acini size were tested by two-way ANOVA with Tukey correction for multiple comparisons (P < 0.05). The number of acini and the number of fully organized acini were analyzed by Fisher exact test (P < 0.05).

An in vitro model for mammary cell transformation

To approximate radiation-induced cell transformation in vitro, we generated primary cultures of BPECs from disease-free mammary tissue as described by Ince and colleagues (10). The relevance of using BPECs is that, in contrast to mammary cells propagated under standard conditions, experimentally transformed BPECs are highly tumorigenic when introduced into immunosuppressed mice and give rise to adenocarcinoma-like tumors (10). However, the propagation of BPECs in a defined WIT medium and on Primaria surface did not provide a stress-free environment as BPECs from the two donors suffered STASIS (stress or aberrant signaling-induced senescence) at PD 10–15 (Fig. 1A). This was a p16-dependent senescence-like growth arrest during which cells presented morphologic changes such as an enlarged, flattened, and vacuole-rich cytoplasm (Fig. 1B), as well as expression of the stress marker p16INK4a (Supplementary Fig. 2a). The experimental cell transformation model used for the present study was obtained from populations of post-STASIS cells, which spontaneously silenced the p16INK4a gene through the methylation of its promoter (22). Post-STASIS cells presented normal morphology (Fig. 1B) and did not increase the expression of p16INK4a, even after exposure to ionizing radiation (Supplementary Fig. 2a). Moreover, exposure to high doses of ionizing radiation did not activate BPECs death by apoptosis, showing a similar level of TUNEL-positive cells before (1.1%) and after irradiation (24 hours: 1.1%; 48 hours: 1.7%; P > 0.05; Supplementary Fig. 2b). Instead, both pre- and post-STASIS BPECs presented radiation-induced SA β-gal activity (P < 0.05; Supplementary Fig. 2c).

Figure 1.

Culture establishment and characterization of transduced and nontransduced BPECs. A, Growth curves of nontransduced cells from donors 1 and 2. B, Representative images of BPECs in culture before (i), during (ii), and after (iii) STASIS. Bar, 200 μm. C, Growth curves of nontransduced and transduced BPECs from donor 1 (i) and donor 2 (ii) together with the different transfection points (dotted line indicates the antibiotic selection period after HRAS transduction). D, Proliferation slopes comparing nontransduced (closed dots) and Ttag + hTERT + HRAS (open dots) transduced BPECs from donor 1 (i) and donor 2 (ii). E, Results from cytogenetic analysis of donor 1 and donor 2 prestasis and post-stasis BPECs as well as Ttag + hTERT and Ttag + hTERT + HRAS-transduced BPECs (i) and representative DAPI-reverse metaphase images of normal and altered karyotypes (ii). Arrowheads show centromeric constrictions in the same chromosome. F, Immunofluorescent detection of cytokeratin 14 (red) and claudin-4 (green) and DAPI staining of nuclei (blue). (i) BPECs positive for one or another marker. (ii) BPECs positive for both markers. Bar, 20 μm. G, Percentage of BPECs expressing cytokeratin 14, claudin-4, or both markers. Results from nontransduced, transduced with Ttag and hTERT and transduced with Ttag, hTERT, and HRAS BPECs from both donors are represented in the same chart. The proportion of cells without any signal was less than 5% in all of the analyzed cultures.

Figure 1.

Culture establishment and characterization of transduced and nontransduced BPECs. A, Growth curves of nontransduced cells from donors 1 and 2. B, Representative images of BPECs in culture before (i), during (ii), and after (iii) STASIS. Bar, 200 μm. C, Growth curves of nontransduced and transduced BPECs from donor 1 (i) and donor 2 (ii) together with the different transfection points (dotted line indicates the antibiotic selection period after HRAS transduction). D, Proliferation slopes comparing nontransduced (closed dots) and Ttag + hTERT + HRAS (open dots) transduced BPECs from donor 1 (i) and donor 2 (ii). E, Results from cytogenetic analysis of donor 1 and donor 2 prestasis and post-stasis BPECs as well as Ttag + hTERT and Ttag + hTERT + HRAS-transduced BPECs (i) and representative DAPI-reverse metaphase images of normal and altered karyotypes (ii). Arrowheads show centromeric constrictions in the same chromosome. F, Immunofluorescent detection of cytokeratin 14 (red) and claudin-4 (green) and DAPI staining of nuclei (blue). (i) BPECs positive for one or another marker. (ii) BPECs positive for both markers. Bar, 20 μm. G, Percentage of BPECs expressing cytokeratin 14, claudin-4, or both markers. Results from nontransduced, transduced with Ttag and hTERT and transduced with Ttag, hTERT, and HRAS BPECs from both donors are represented in the same chart. The proportion of cells without any signal was less than 5% in all of the analyzed cultures.

Close modal

It is well known that the transformation of normal cells into a malignant phenotype is a multistep process that involves the acquisition of traits affecting the regulatory circuits that govern normal cell proliferation and homeostasis (7). For this reason, in order to develop an experimental transformation model, BPECs were transduced with lentiviral particles containing the Ttag, hTERT, and HRAS genes along with GFP, mCherry, and the puromycin-resistance gene, respectively. Thus, three different cell culture types for each donor were obtained: nontransduced cells (herein also referred to as healthy cells), cells transduced with Ttag and hTERT (also called partially transformed or preneoplastic cells) and cells transduced with Ttag, hTERT, and HRAS (fully transformed cells according to Ince and colleagues; ref. 10). Properly transduced cells were selected with fluorescent cell sorting (those transduced with Ttag + GFP and hTERT + mCherry; Supplementary Fig. 2e) and antibiotics (those transduced with HRAS + puromycin) and the proper expression of transgenes was confirmed by RT-qPCR. Regarding the expression of HRAS, transduced cells presented a 3-fold and 5-fold increase compared with nontransduced cells from donors 1 and 2, respectively (Supplementary Fig. 2d). The growth kinetics of transduced cells was monitored by means of population doubling accumulation (Fig. 1C), and a higher proliferation rate of fully transformed cells (24.9 hours/PD) compared with nontransduced cells (42.5 hours/PD) was detected (Fig. 1D). Therefore, the transduction of cells with the three selected genetic elements affected normal cell proliferation.

Cytogenetic analyses revealed a progressive increase in the proportion of chromosomal rearrangements with PD and after introducing gene alterations (Fig. 1E). In the two donors analyzed, few cells presented an abnormal karyotype before STASIS (d1: 4%; d2: 15%), but the frequency of cells with abnormal cytogenetics increased after overcoming STASIS (d1: 12%; d2: 39%). Moreover, after transduction with Ttag and ectopic hTERT, an even higher frequency of cells with chromosomal abnormalities, including tetraploidy, was observed in BPECs (d1: 46%; d2: 47%) probably due to the excessive telomere attrition achieved before transduction with hTERT. Likewise, the addition of HRAS favored the tetraploidization of the culture reaching 80% of the population with a tetraploid or near-tetraploid chromosome content. Therefore, structural and numerical chromosomal alterations increased with time in culture and the successive incorporation of gene alterations.

The BPEC culture allows an enrichment of epithelial cells with luminal characteristics, although other epithelial cell types may coexist in the culture (10, 22). To further characterize the different BPEC cultures during the transformation process, cytokeratin 14 (K14) and claudin-4 (Cl4) were immunodetected to distinguish between myoepithelial and luminal cell lineages, respectively (Fig. 1F). During the cell transformation process, the proportion of cells positive for cytokeratin 14, for claudin-4 or for both markers remained constant (on average: 57% K14+, 19% Cl4+, and 24% K14+/Cl4+; Fig. 1G). Therefore, although BPECs showed some chromosomal instability, the experimental transformation process preserved the initial cell-lineage heterogeneity, mimicking the cell-lineage diversity of the mammary gland to some extent.

Healthy cells require a higher radiation dose than partially transformed cells to enhance their anchorage-independent growth capacity

The ability to proliferate in suspension and form colonies is a hallmark of transformed cells (7). For this reason, to assess the contribution of radiation in the acquisition of tumor characteristics, we analyzed the capacity of cells to proliferate in suspension in soft-agar medium (Fig. 2A). As expected, nontransduced BPECs presented a low colony formation capacity as very few (d1: 15; d2: 3) and small [median diameter 68 μm (d1) and 70 μm (d2)] colonies were metabolically active at the end of the experiment, indicating that nontransduced BPECs needed a substrate to adhere to and proliferate properly (Fig. 2B). The transduction of BPECs with Ttag and hTERT genes resulted in an increase in the number (d1: 175; d2: 278) and size (d1: 82 μm; d2: 83 μm) of the colonies (Fig. 2B), suggesting that these two genetic elements conferred on them some characteristics typical of neoplastic transformation. The addition of ectopic HRAS entailed the formation of more colonies than the sole introduction of Ttag and hTERT in BPECs from one donor (d1: 369) but not in the other (d2: 243). Interestingly, both donors' colonies were significantly larger (d1: 97 μm; d2: 114 μm) than those of their nontransduced and partially transformed counterparts (Fig. 2B). These results suggest that the transduction with Ttag, hTERT, and HRAS contributes to the acquisition of the anchorage-independence growth ability and, together with data provided by Ince and colleagues (10), we can conclude that triple-transfected cells can be used as a positive control of neoplastic transformation in the following experiments using ionizing radiation.

Figure 2.

Anchorage assay of nontransduced, transduced with Ttag and hTERT and transduced with Ttag, hTERT, and HRAS BPECs from donors 1 and 2. A, Representative images of MTT-positive colonies. Bar, 5 mm; insert bar, 2.5 mm. B, Dot plot representing the median diameter of each colony (μm) and the total number of colonies formed for each condition. C–H, Anchorage-independent growth 8 weeks after exposure to moderate/low doses of radiation of nontransduced (C,F, and G) and partially transformed BPECs (D,E, and H) from donor 1 (C–E) and donor 2 (F–H). Dot plots represent the diameter of each colony and the total number of colonies analyzed for each condition. Dose, number of colonies formed, and median diameter are displayed for each condition. Gray dots indicate colonies discarded from the analysis due to insufficient size. Gray lines show median colony diameter for each group. Different letters indicate clustering of data in different groups of statistical significance for number (Fisher exact test; P < 0.05) and size of the colonies (Kruskal–Wallis test with multiple comparisons correction, P < 0.05).

Figure 2.

Anchorage assay of nontransduced, transduced with Ttag and hTERT and transduced with Ttag, hTERT, and HRAS BPECs from donors 1 and 2. A, Representative images of MTT-positive colonies. Bar, 5 mm; insert bar, 2.5 mm. B, Dot plot representing the median diameter of each colony (μm) and the total number of colonies formed for each condition. C–H, Anchorage-independent growth 8 weeks after exposure to moderate/low doses of radiation of nontransduced (C,F, and G) and partially transformed BPECs (D,E, and H) from donor 1 (C–E) and donor 2 (F–H). Dot plots represent the diameter of each colony and the total number of colonies analyzed for each condition. Dose, number of colonies formed, and median diameter are displayed for each condition. Gray dots indicate colonies discarded from the analysis due to insufficient size. Gray lines show median colony diameter for each group. Different letters indicate clustering of data in different groups of statistical significance for number (Fisher exact test; P < 0.05) and size of the colonies (Kruskal–Wallis test with multiple comparisons correction, P < 0.05).

Close modal

To study the contribution of ionizing radiation in the acquisition of the malignant phenotype, nontransduced and partially transformed cells were exposed to different doses of γ-rays and tested by anchorage assay at 4 and 8 weeks after irradiation. In parallel with the anchorage assay, the growth of the different cell samples in monolayer cultures was monitored and no significant differences were found between exposed and nonexposed cells, indicating that radiation did not affect the growth of healthy and partially transformed cells in monolayer cultures for the following weeks after irradiation (Supplementary Fig. S3). A significant increase in the number and size of the cell colonies grown in suspension was detected in nontransduced cells from donor 1 exposed to 2 Gy after 8 weeks in culture (265 colonies with a median diameter of 76 μm) compared with their nonirradiated counterparts (34 colonies with a median diameter of 68 μm; Fig. 2C). The number and size of the colonies arising from irradiated healthy cells resembled those arising from nonirradiated partially transformed cells, indicating that a radiation dose of at least 2 Gy is needed to ignite the transformation process in healthy BPECs. In the case of nontransduced cells from donor 2, no changes in the results of the anchorage assay were observed after exposing them to a wide range of radiation doses (from 0.25 to 2 Gy; Fig. 2F and G). It is likely that the genetic background of BPECs from donor 2 makes them more resistant to the radiation-induced neoplastic transformation, meaning that they need higher doses of radiation to exhibit the same attributes that cells from donor 1 did at 2 Gy.

After the introduction of Ttag and hTERT, cells from donor 1 required a lower dose to increase the colony formation efficiency in comparison with nontransduced cells from the same donor. The exposure to 1 Gy followed by 8 weeks in culture was sufficient to increase both the colony number and size (127 colonies with an average diameter of 115 μm; Fig. 2D) of partially transformed cells compared with nonirradiated ones. However, the number of colonies formed when cells are exposed to 2 Gy was not significantly higher than their control counterparts. As donor 1 seems to be more radiosensitive than donor 2, it is likely that partially transformed cells, which already have an altered karyotype, cannot cope with the amount of DNA damage induced by 2 Gy. In the absence of p53, these cells cannot maintain a prolonged cell-cycle arrest and may enter a potentially catastrophic mitosis before DNA lesions are repaired (23). Mitotic catastrophe can occur few after mitotic exit or in a delayed fashion, as exemplified by the “delayed radioinduced cell death” that can occur years after radiotherapy in human tissues (24, 25). Alleged radiosensitivity of donor 1 together with a delayed outcome of the cells after mitotic catastrophe induction helps to understand why the number of colonies formed by partially transformed cells from donor 1 is not significantly higher than their nonirradiated counterparts.

In order to determine the minimum transforming dose in cells from donor 1 transduced with Ttag and hTERT, we performed the anchorage-independent assay after exposing them to 0.25 and 0.5 Gy. Although exposure to 0.25 Gy of γ-rays induced a slight increase in colony number and size at 4 weeks post-irradiation, a radiation dose of 0.5 Gy was sufficient to induce an increase at both 4 (Supplementary Fig. S4) and 8 weeks post-irradiation (Fig. 2E). Similarly, partially transformed BPECs from donor 2 exposed to 2 Gy of γ-rays showed a significant boost in the number and size of metabolically active colonies (Fig. 2H), whereas nontransduced cells from this donor exposed to up to 2 Gy did not. Altogether, the results obtained for both donors suggest that cells transduced with Ttag and hTERT require lower doses of radiation than nontransduced counterparts to enhance anchorage-independent growth capacity.

Moderate radiation exposure impairs polarization only in acini derived from partially transformed cells

3D cell cultures based on Matrigel or other laminin-rich matrixes have been proposed as an important alternative to conventional cell culture methods in order to better emulate physiologic environmental conditions. In the current study, keeping the heterogenic cell population of the BPEC culture in a 3D context allowed the organization and interactions of the different types of breast epithelial cells and their ability to form polarized acinar structures to be studied.

In order to investigate whether the acinar structures of mammary epithelial cells in Matrigel resulted from the aggregation or alternatively from the proliferation of individualized cells, time-lapse images were recorded during the first 24 hours after seeding. According to the images obtained, BPECs randomly aggregated during the first 5 hours to form dense, small acinar structures (Fig. 3A; Supplementary multimedia file S1); from this point onward, the speed of aggregation was gradually reduced (Supplementary Fig. S5). Acinar size was measured 7 and 14 days after seeding, and the total number of acini formed was scored after 2 weeks in culture. These results revealed that, once aggregated, the acinar growth was mostly due to proliferation instead of aggregation. Thus, the acinar structures finally obtained after 14 days in Matrigel-based cell culture resulted from an initial phase of aggregation followed by a cell proliferation phase.

Figure 3.

Matrigel-based 3D BPEC culture. A, Representative images of time lapse acquired during acini formation. Pictures taken every 60 minutes from seeding to the seventh hour of the experiment are shown. The complete video is accessible in Supplementary multimedia file S1. Bar, 100 μm. B and C, Number (i) and size increase (ii) of acini resulting from nontransduced (B) and partially transformed BPECs (C) from donor 1 and donor 2 3D cultures established 8 weeks after exposure to ionizing radiation. For each 3D Matrigel-based culture, 10,000 cells were seeded and maintained for 2 weeks. The diameter was normalized using the acini size 24 hours after seeding. Dots represent mean size increase for each time point and culture; bars, SEM. Different letters indicate clustering of data in different groups of significance for cultures analyzed in the same time point (two-way ANOVA with Tukey correction for multiple comparisons; P < 0.05).

Figure 3.

Matrigel-based 3D BPEC culture. A, Representative images of time lapse acquired during acini formation. Pictures taken every 60 minutes from seeding to the seventh hour of the experiment are shown. The complete video is accessible in Supplementary multimedia file S1. Bar, 100 μm. B and C, Number (i) and size increase (ii) of acini resulting from nontransduced (B) and partially transformed BPECs (C) from donor 1 and donor 2 3D cultures established 8 weeks after exposure to ionizing radiation. For each 3D Matrigel-based culture, 10,000 cells were seeded and maintained for 2 weeks. The diameter was normalized using the acini size 24 hours after seeding. Dots represent mean size increase for each time point and culture; bars, SEM. Different letters indicate clustering of data in different groups of significance for cultures analyzed in the same time point (two-way ANOVA with Tukey correction for multiple comparisons; P < 0.05).

Close modal

Although the number and size of acini obtained from the two donors in 3D culture was different in non-, partially, and fully transformed BPECs, as was their growth kinetics (Supplementary Fig. S6), their response to γ-ray exposure was the same. Compared with their nonirradiated counterparts, irradiated nontransduced cells produced much fewer acini (d1: 462 vs. 1,412; d2: 496 vs. 2,796; Fig. 3B) and showed a lower size growth rate during the time in culture (d1: 18.5 μm vs. 72.4 μm; d2: 11.5 μm vs. 59.2 μm; Fig. 3B). On the contrary, BPECs transduced with Ttag and hTERT and exposed to 2 Gy of radiation gave rise to a higher number of acini (d1: 3,888 vs. 1,658; d2: 1,222 vs. 311; Fig. 3C). Similarly, when they were exposed to 0.5 Gy and 1 Gy, they produced acini with a higher growth rate than their nonirradiated counterparts (d1: 70.7 μm vs. 26.9 μm; d2: 110.1 μm vs. 53.4 μm; Fig. 3C). In order to establish a clear relation between the number and size of acini, the results of these two parameters were represented in two-axis plots to detect deviations from the theoretical compensation between these parameters established according to R = (n/ρ)1/3r equation (Supplementary Fig. S7a). Although the acini of nontransduced cells diminish both in number and in their growth after irradiation (Supplementary Fig. S7b), acini derived from cells transduced with Ttag and hTERT increase in number and size after exposure to high doses of radiation (Supplementary Fig. S7c). Therefore, the results indicate that the response to radiation in terms of acini formation is completely different in nontransduced and partially transformed cells and, more importantly, consistent in the two donors analyzed.

Having established that partially transformed BPECs respond to radiation with the formation of more and larger acini than healthy BPECs, we next examined the organization of the different lineages of mammary epithelial cells inside the acinar structures. With this end, we immunodetected K14 and Cl4 in acini from all of the conditions described so far and studied the 3D organization by means of a confocal fluorescence microscope. Those acini displaying myoepithelial lineage (K14)–positive signal cells surrounding an inner layer of luminal lineage (Cl4)–positive cells were considered to be properly polarized structures; otherwise, the acini were considered incorrectly polarized (Supplementary Fig. S7d). The quantification of polarized and unpolarized acini in the two donors showed that the percentage of properly organized acini was similar in healthy and partially transformed BPECs (d1: 78.6% vs. 80%; d2: 100% vs. 91.7%; Fig. 4). Interestingly, even though it was not statistically significant, cells transduced with Ttag, hTERT, and HRAS presented a reduction in their capacity to form polarized structures (d1: 57.1%; d2: 70%; Fig. 4, black bars). These results reveal that although a partial transformation does not modify the BPECs capacity to form 3D structures organized according to the physiologic distribution in the mammary gland, the induction of an experimental transformation process with Ttag, hTERT, and HRAS likely alters the cell-to-cell communication and distribution inside the acini.

Figure 4.

Analysis of acini polarization. A and B, Percentage of polarized acini formed 8 weeks after exposure to 0, 0.5, 1, and 2 Gy of γ-rays by nontransduced, partially transformed, and fully transformed BPECs from donor 1 (A) and donor 2 (B). Numbers inside bars represent the number of analyzed acini for each condition. Different letters indicate statistically significant differences between cells with same genetic alterations exposed to different doses of radiation (Fisher exact test; P < 0.05).

Figure 4.

Analysis of acini polarization. A and B, Percentage of polarized acini formed 8 weeks after exposure to 0, 0.5, 1, and 2 Gy of γ-rays by nontransduced, partially transformed, and fully transformed BPECs from donor 1 (A) and donor 2 (B). Numbers inside bars represent the number of analyzed acini for each condition. Different letters indicate statistically significant differences between cells with same genetic alterations exposed to different doses of radiation (Fisher exact test; P < 0.05).

Close modal

Finally, exposure to ionizing radiation did not alter the organization capacity of nontransduced BPECs; however, partially transformed cells derived from donor 1 presented a dose-dependent reduction in acini polarization, reaching even lower percentages than nonirradiated fully transformed cells (0 Gy: 80%; 0.5 Gy: 60%; 1 Gy: 50%; 2 Gy: 28.6%; Fig. 4B). The fact that irradiated, partially transformed cells from donor 2 maintained their polarization capacity is coherent with the results obtained in the anchorage assay in which cells from this donor required higher doses than cells from donor 1 to exhibit the same effects. In conclusion, the reduction of polarized acini derived from irradiated cells suggests that partially transformed BPECs tend to behave, at least in one of the donors, like their fully transformed counterparts when they are exposed to moderate doses of radiation.

Due to its characteristics, the mammary gland has an increased risk of tumor formation compared with other tissues; it is one of the most radiosensitive organs and is frequently exposed to radiation. Experiments to investigate radiation-induced malignant transformation were performed in BPECs, whose culture is complex and expensive, but provide a reliable model to maintain a heterogeneous population of mammary epithelial cells, including cells with luminal characteristics, for more than 20 passages. The simultaneous culture of this diversity of cell types mimics, in some way, the microenvironment of the mammary gland and allows the distribution of different cell lineages inside 3D structures formed in Matrigel-based cultures to be studied.

Although the different cell types are maintained, contrary to what was described by Ince and colleagues in 2007 and as already published by us, in our hands the p16-dependent STASIS is not avoided in human mammary epithelial cells cultured in WIT medium (BPECs). After only a few passages, the cells arrested their growth, entered senescence, and only those that had silenced the p16 expression could overcome this arrest (22, 26–28). As a point of fact, p16 gene silencing has been reported in vivo in mammary epithelial cells in more than 40% of the examined healthy patients (29). Additionally, p16 gene silencing is one of the most frequently observed mutations in breast cancer (26, 30–32). Therefore, we consider that post-STASIS cells are a valuable tool to study the tumorigenic effect of radiation.

Ttag and hTERT confer a pretransformed grade to healthy BPECs

The genetic elements introduced in post-STASIS BPECs to obtain an experimental transformation model are known to alter key pathways in the transition from healthy to neoplastic cells. The transduction with Ttag inhibited the pathways of two essential tumor suppressors, p53 and pRb, leading to ineffective cell-cycle checkpoints (33, 34). The introduction of ectopic hTERT counteracted telomeric erosion associated with cell proliferation. Finally, the introduction of ectopic HRAS provided a constitutive mitogenic signal. Although alterations in TP53, RB, and telomerase enzyme genes have been reported in breast cancer events (35–38), HRAS is not frequently overexpressed in breast tumors. However, Ras protein levels can be altered in breast cancer cells due to the overexpression of EGFR or ErbB2 receptor, because both proteins are activators of the Ras pathway. It is worth noting that increased levels of Ras pathway proteins have been observed in 11 of 20 women suffering from breast cancer (39). Therefore, the genetic elements used to transform BPECs are directly or indirectly related to breast cancer.

BPEC transduction with Ttag and hTERT, along with the spontaneous p16INK4a gene silencing, contributed to an increase in the frequency of chromosomal rearrangements, which could entail the enhanced ability of these modified cells to grow in an anchorage-independent manner compared with their healthy counterparts. Despite the above-mentioned differences, both healthy and partially transformed BPECs were able to form polarized acini in Matrigel-based 3D cultures. Altogether, the results obtained pointed out that, even if BPECs transduced with Ttag and hTERT retained some typical features of healthy cells, they started the transformation process and consequently accomplished a pretransformed stage.

Overexpression of HRAS in a pretransformed environment culminates the transformation process

In order to establish a positive control, pretransformed BPECs were transduced with HRAS that, as expected, resulted in an enhanced proliferation rate (40), but also in an increased formation of tetraploid cells. The overexpression of HRAS is often accompanied by cytogenetic rearrangements, but HRAS-driven tetraploidization has not been previously reported. It is possible that inhibition of the p53 and pRb pathways supports a positive selection of tetraploid cells that were already present in the culture before the transduction as a result of telomeric attrition. The increased proliferation rate conferred by ectopic HRAS could have accelerated the positive selection of these tetraploid cells, which are known to have a greater tolerance to genome instability. Multiple studies consider tetraploidization a critical step to achieve tumorigenesis (41, 42); consequently, the fact that the karyotype of BPECs transduced with Ttag, hTERT, and HRAS was, in most cases, tetraploid suggests that the experimental transformation process was fulfilled.

The overexpression of the HRAS oncogene resulted in a nonsignificant but still conspicuous reduction in polarized structures. Accordingly, Matrigel-grown acini derived from different tumor mammary epithelial cells presented neither hollow nor physiologic organization of the different cell lineages (43). Although the presence of a hollow was not evaluated because of cavity collapse during the methodological procedure, the reduced number of polarized acini formed by BPECs transduced with Ttag, hTERT, and HRAS suggests an inaccurate communication or impeded motility of these cells inside the acinus. These results, together with the previous observations of Ince and colleagues regarding the ability of these cells to form tumors in immunosuppressed mice, indicate that BPECs transduced with Ttag, hTERT, and HRAS are certainly a good positive control for cellular transformation.

Pretransformed cells present an increased susceptibility to radiation-induced transformation compared with healthy cells

When BPECs were exposed to ionizing radiation, we observed that the acini growth rate between cells differed depending on the dose received by the seeded cells and their genetic modification. Although the acini of irradiated nontransduced cells reduced their growth rate in 3D cultures, acini derived from BPECs transduced with Ttag and hTERT showed an increased growth rate when exposed to moderate doses of radiation in comparison with their nonirradiated counterparts. Remarkably, partially transformed cells from donor 1 showed a noticeable reduction of polarized acini as dose increased, whereas nontransduced BPECs from the same donor formed correctly polarized acini even after exposure to 2 Gy of γ-rays. These results suggest that, although healthy cells remained organized in 3D cultures after exposure to ionizing radiation, partially transformed cells were more susceptible to radiation and became unable to achieve a proper disposition after exposure to moderate doses of γ-rays. Accordingly, the anchorage assay also proved a reduction of the minimum dose required by BPECs transduced with Ttag and hTERT to enhance their anchorage-independent growth. Although nontransduced cells required ≥2 Gy to show a significant increase in colony size and number, partially transformed cells required only 0.5 to 2 Gy (depending on the donor) to increase their colony formation capacity to an even higher grade than that of fully transformed cells. It is worth noting that 4 weeks in culture was insufficient to induce significant differences in the anchorage assay, indicating that not only the dose but also the time in culture after irradiation are important factors to be considered when assessing the transformation process.

Altogether, the different approaches used in this work suggest that BPECs transduced with Ttag and hTERT are in a pretransformed stage and display an increased susceptibility to radiation-induced DNA damage and neoplastic transformation. According to this, it is possible that the background or age-associated accumulation of genetic alterations affect susceptibility to exposure to radiation. Therefore, it would be interesting to review the protocols of exposure to ionizing radiation for those people who, due to their genetic background or age, may have a greater accumulation of partially transformed cells. Efforts to thoroughly dissect the carcinogenic effect of radiation are crucial to keep revealing all of the inconveniences concealed in the use of this diagnostic and therapeutic tool.

No potential conflicts of interest were disclosed.

Conception and design: J. Repullés, A. Genescà, M. Terradas

Development of methodology: J. Repullés, D. Soler, J.C. Ramírez, M. Terradas

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J. Repullés, T. Anglada, D. Soler, J.C. Ramírez

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J. Repullés, D. Soler, A. Genescà, M. Terradas

Writing, review, and/or revision of the manuscript: J. Repullés, T. Anglada, A. Genescà, M. Terradas

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J. Repullés

Study supervision: A. Genescà, M. Terradas

The authors thank the technical and intellectual support for the 3D experiments from Gemma Fuster and Arantzazu Zubeldia-Plazaola from the Department of Medical Oncology of Hospital Clínic de Barcelona. We would like to acknowledge the help received from Raul Torres and Aida García from the former Viral Vector Facility of CNIC (Madrid) led by Juan Carlos Ramírez for their assistance in the lentivirus design and production. We also thank Núria Barba from the UAB Neurosciences Institute (UAB-INc) for confocal microscopy technical support, the Barcelona Surgical Institute for the collection of mammoplasty samples, the UAB Technical Unit of Radiobiological Protection for sample irradiation and the FACS Unit from CRG/UPF for cell sorting support. This work was supported by EURATOM (Dark.Risk GA 323216), by Consejo de Seguridad Nuclear, and by Generalitat de Catalunya (grant no. 2017-SGR-503).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Ferlay
J
,
Soerjomataram
I
,
Dikshit
R
,
Eser
S
,
Mathers
C
,
Rebelo
M
, et al
Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012
.
Int J cancer
2015
;
136
:
E359
86
.
2.
Thomas
F
,
Nesse
RM
,
Gatenby
R
,
Gidoin
C
,
Renaud
F
,
Roche
B
, et al
Evolutionary ecology of organs: a missing link in cancer development?
Trends in Cancer
2016
;
2
:
409
15
.
3.
Perou
CM
,
Sørlie
T
,
Eisen
MB
,
van de Rijn
M
,
Jeffrey
SS
,
Rees
Ca
, et al
Molecular portraits of human breast tumours
.
Nature
2000
;
406
:
747
52
.
4.
Prat
A
,
Perou
CM
. 
Deconstructing the molecular portraits of breast cancer
.
Mol Oncol
2011
;
5
:
5
23
.
5.
Keller
PJ
,
Arendt
LM
,
Skibinski
A
,
Logvinenko
T
,
Klebba
I
,
Dong
S
, et al
Defining the cellular precursors to human breast cancer
.
Proc Natl Acad Sci U S A
2012
;
109
:
2772
7
.
6.
Bhagirath
D
,
Zhao
X
,
West
WW
,
Qiu
F
,
Band
H
,
Band
V
. 
Cell type of origin as well as genetic alterations contribute to breast cancer phenotypes
.
Oncotarget
2015
;
6
:
9018
30
.
7.
Hanahan
D
,
Weinberg
Ra
. 
Hallmarks of cancer: the next generation
.
Cell
2011
;
144
:
646
74
.
8.
Hahn
WC
,
Counter
CM
,
Lundberg
AS
,
Beijersbergen
RL
,
Brooks
MW
,
Weinberg
RA
. 
Creation of human tumour cells with defined genetic elements
.
Nature
1999
;
400
:
464
8
.
9.
Elenbaas
B
,
Spirio
L
,
Koerner
F
,
Fleming
MD
,
Zimonjic
DB
,
Donaher
JL
, et al
Human breast cancer cells generated by oncogenic transformation of primary mammary epithelial cells
.
Genes Dev
2001
;
15
:
50
65
.
10.
Ince
TA
,
Richardson
AL
,
Bell
GW
,
Saitoh
M
,
Godar
S
,
Karnoub
AE
, et al
Transformation of different human breast epithelial cell types leads to distinct tumor phenotypes
.
Cancer Cell
2007
;
12
:
160
70
.
11.
Eheman
CR
,
Shaw
KM
,
Ryerson
AB
,
Miller
JW
,
Ajani
UA
,
White
MC
. 
The changing incidence of in situ and invasive ductal and lobular breast carcinomas: United States, 1999–2004
.
Cancer Epidemiol Biomarkers Prev
2009
;
18
:
1763
9
.
12.
Shah
DJ
,
Sachs
RK
,
Wilson
DJ
. 
Radiation-induced cancer: a modern view
.
Br J Radiol
2012
;
85
:
e1166
73
.
13.
Howe
GR
,
McLaughlin
J
. 
Breast cancer mortality between 1950 and 1987 after exposure to fractionated moderate-dose-rate ionizing radiation in the Canadian fluoroscopy cohort study and a comparison with breast cancer mortality in the atomic bomb survivors study
.
Radiat Res
1996
;
145
:
694
.
14.
Ronckers
CM
,
Land
CE
,
Miller
JS
,
Stovall
M
,
Lonstein
JE
,
Doody
MM
. 
Cancer mortality among women frequently exposed to radiographic examinations for spinal disorders
.
Radiat Res
2010
;
174
:
83
90
.
15.
Ma
H
,
Hill
CK
,
Bernstein
L
,
Ursin
G
. 
Low-dose medical radiation exposure and breast cancer risk in women under age 50 years overall and by estrogen and progesterone receptor status: results from a case–control and a case–case comparison
.
Breast Cancer Res Treat
2008
;
109
:
77
90
.
16.
The 2007 recommendations of the international commission on radiological protection. ICRP publication 103
.
Ann ICRP
2007
;
37
:
1
332
.
17.
Mavragani
I
,
Nikitaki
Z
,
Souli
M
,
Aziz
A
,
Nowsheen
S
,
Aziz
K
, et al
Complex DNA damage: a route to radiation-induced genomic instability and carcinogenesis
.
Cancers (Basel)
2017
;
9
:
E91
.
18.
Cox
R
. 
The multi-step nature of carcinogenesis and the implications for risk analysis
.
Int J Radiat Biol
1998
;
73
:
373
6
.
19.
Labarge
MA
,
Garbe
JC
,
Stampfer
MR
. 
Processing of human reduction mammoplasty and mastectomy tissues for cell culture
.
J Vis Exp
2013
;
pii:50011
.
20.
Debacq-Chainiaux
F
,
Erusalimsky
JD
,
Campisi
J
,
Toussaint
O
. 
Protocols to detect senescence-associated beta-galactosidase (SA-βgal) activity, a biomarker of senescent cells in culture and in vivo
.
Nat Protoc
2009
;
4
:
1798
806
.
21.
Debnath
J
,
Muthuswamy
SK
,
Brugge
JS
. 
Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures
.
Methods
2003
;
30
:
256
68
.
22.
Feijoo
P
,
Terradas
M
,
Soler
D
,
Domínguez
D
,
Tusell
L
,
Genescà
A
. 
Breast primary epithelial cells that escape p16-dependent stasis enter a telomere-driven crisis state
.
Breast Cancer Res
2016
;
18
:
7
.
23.
Vitale
I
,
Galluzzi
L
,
Castedo
M
,
Kroemer
G
. 
Mitotic catastrophe: a mechanism for avoiding genomic instability
.
Nat Rev Mol Cell Biol
2011
;
12
:
385
92
.
24.
Roninson
IB
,
Broude
EV
,
Chang
B
. 
If not apoptosis, then what? Treatment-induced senescence and mitotic catastrophe in tumor cells
.
Drug Resist Updat
2001
;
4
:
303
13
.
25.
Suzuki
K
,
Ojima
M
,
Kodama
S
,
Watanabe
M
. 
Radiation-induced DNA damage and delayed induced genomic instability
.
Oncogene
2003
;
22
:
6988
93
.
26.
Brenner
AJ
,
Stampfer
MR
,
Aldaz
CM
. 
Increased p16 expression with first senescence arrest in human mammary epithelial cells and extended growth capacity with p16 inactivation
.
Oncogene
1998
;
17
:
199
205
.
27.
Foster
SA
,
Wong
DJ
,
Barrett
MT
,
Galloway
DA
. 
Inactivation of p16 in human mammary epithelial cells by CpG island methylation
.
Mol Cell Biol
1998
;
18
:
1793
801
.
28.
Huschtscha
LI
,
Noble
JR
,
Neumann
AA
,
Moy
EL
,
Barry
P
,
Melki
JR
, et al
Loss of p16INK4 expression by methylation is associated with lifespan extension of human mammary epithelial cells
.
Cancer Res
1998
;
58
:
3508
12
.
29.
Holst
CR
,
Nuovo
GJ
,
Esteller
M
,
Chew
K
,
Baylin
SB
,
Herman
JG
, et al
Methylation of p16 INK4a promoters occurs in vivo in histologically normal human mammary epithelia
.
Cancer
2003
;
63
:
1596
601
.
30.
Geradts
J
,
Kratzke
RA
,
Niehans
GA
,
Lincoln
CE
. 
Immunohistochemical detection of the cyclin-dependent kinase inhibitor 2/multiple tumor suppressor gene 1 (CDKN2/MTS1) product p16INK4A in archival human solid tumors: correlation with retinoblastoma protein expression
.
Cancer Res
1995
;
55
:
6006
11
.
31.
Herman
JG
,
Merlo
A
,
Mao
L
,
Lapidus
RG
,
Issa
JP
,
Davidson
NE
, et al
Inactivation of the CDKN2/p16/MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers
.
Cancer Res
1995
;
55
:
4525
30
.
32.
Brenner
AJ
,
Paladugu
A
,
Wang
H
,
Olopade
OI
,
Dreyling
MH
,
Aldaz
CM
. 
Preferential loss of expression of p16(INK4a) rather than p19(ARF) in breast cancer
.
Clin Cancer Res
1996
;
2
:
1993
8
.
33.
Jaganathan
H
,
Gage
J
,
Leonard
F
,
Srinivasan
S
,
Souza
GR
,
Dave
B
, et al
Three-dimensional in vitro co-culture model of breast tumor using magnetic levitation
.
Sci Rep
2015
;
4
:
6468
.
34.
Kim
HY
,
Ahn
BY
,
Cho
Y
. 
Structural basis for the inactivation of retinoblastoma tumor suppressor by SV40 large T antigen
.
EMBO J
2001
;
20
:
295
304
.
35.
Hollstein
M
,
Sidransky
D
,
Vogelstein
B
,
Harris
C
. 
p53 mutations in human cancers
.
Science
1991
;
253
:
49
53
.
36.
Lilyestrom
W
,
Klein
MG
,
Zhang
R
,
Joachimiak
A
,
Chen
XS
. 
Crystal structure of SV40 large T-antigen bound to p53: interplay between a viral oncoprotein and a cellular tumor suppressor
.
Genes Dev
2006
;
20
:
2373
82
.
37.
Shin
E
,
Jung
WH
,
Koo
JS
. 
Expression of p16 and pRB in invasive breast cancer
.
Int J Clin Exp Pathol
2015
;
8
:
8209
17
.
38.
Kirkpatrick
KL
,
Ogunkolade
W
,
Elkak
AE
,
Bustin
S
,
Jenkins
P
,
Ghilchick
M
, et al
hTERT expression in human breast cancer and non-cancerous breast tissue: correlation with tumour stage and c-Myc expression
.
Breast Cancer Res Treat
2003
;
77
:
277
84
.
39.
von Lintig
FC
,
Dreilinger
AD
,
Varki
NM
,
Wallace
AM
,
Casteel
DE
,
Boss
GR
. 
Ras activation in human breast cancer
.
Breast Cancer Res Treat
2000
;
62
:
51
62
.
40.
Serrano
M
,
Gómez-Lahoz
E
,
DePinho
RA
,
Beach
D
,
Bar-Sagi
D
. 
Inhibition of ras-induced proliferation and cellular transformation by p16INK4
.
Science
1995
;
267
:
249
52
.
41.
Coward
J
,
Harding
A
. 
Size does matter: why polyploid tumor cells are critical drug targets in the war on cancer
.
Front Oncol
2014
;
4
:
123
.
42.
Sheltzer
JM
,
Ko
JH
,
Replogle
JM
,
Habibe Burgos
NC
,
Chung
ES
,
Meehl
CM
, et al
Single-chromosome gains commonly function as tumor suppressors
.
Cancer Cell
2017
;
31
:
240
55
.
43.
Petersen
OW
,
Rønnov-Jessen
L
,
Howlett
AR
,
Bissell
MJ
. 
Interaction with basement membrane serves to rapidly distinguish growth and differentiation pattern of normal and malignant human breast epithelial cells
.
Proc Natl Acad Sci U S A
1992
;
89
:
9064
8
.