Role of Pro-inflammatory Cytokines in Radiation-Induced Genomic Instability in Human Bronchial Epithelial Cells Author(s): Erica Werner, Huichen Wang and Paul W. Doetsch Source: Radiation Research, 184(6):621-629. Published By: Radiation Research Society DOI: http://dx.doi.org/10.1667/RR14045.1 URL: http://www.bioone.org/doi/full/10.1667/RR14045.1

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

RADIATION RESEARCH

184, 621–629 (2015)

0033-7587/15 $15.00 Ó2015 by Radiation Research Society. All rights of reproduction in any form reserved. DOI: 10.1667/RR14045.1

Role of Pro-inflammatory Cytokines in Radiation-Induced Genomic Instability in Human Bronchial Epithelial Cells Erica Werner,a,1 Huichen Wangb and Paul W. Doetscha,c,1 a Department of Biochemistry, Emory University School of Medicine, Atlanta, Georgia; b Department of Physics, Radiation Institute for Science and Engineering (RaISE), Prairie View A&M University, Prairie View, Texas; and c Departments of Radiation Oncology and Department of Hematology and Medical Oncology, Winship Cancer Institute, Emory University School of Medicine, Atlanta, Georgia

these results point to a cell-autonomous mechanism sustaining radiation-induced genomic instability in this model system and suggest that while molecules associated with these mechanisms could be markers for persisting damage, they reflect two different outcomes. Ó 2015 by Radiation Research

Werner, E., Wang, H. and Doetsch, P. W. Role of Proinflammatory Cytokines in Radiation-Induced Genomic Instability in Human Bronchial Epithelial Cells. Radiat. Res. 184, 621–629 (2015).

Society

Inflammatory cytokines have been implicated in the regulation of radiation-induced genomic instability in the hematopoietic system and have also been shown to induce chronic DNA damage responses in radiation-induced senescence. We have previously shown that human bronchial epithelial cells (HBEC3-KT) have increased genomic instability and IL-8 production persisting at day 7 after exposure to high-LET (600 MeV/nucleon 56Fe ions) compared to lowLET (320 keV X rays) radiation. Thus, we investigated whether IL-8 induction is part of a broader pro-inflammatory response produced by the epithelial cells in response to damage, which influences genomic instability measured by increased micronuclei and DNA repair foci frequencies. We found that exposure to radiation induced the release of multiple inflammatory cytokines into the media, including GM-CSF, GROa, IL-1a, IL-8 and the inflammation modulator, IL-1 receptor antagonist (IL-1RA). Our results suggest that this is an IL-1a-driven response, because an identical signature was induced by the addition of recombinant IL-1a to nonirradiated cells and functional interference with recombinant IL-1RA (Anakinra) or anti-IL-1a functionblocking antibody, decreased IL-8 production induced by radiation exposure. However, genomic instability was not influenced by this pathway as addition of recombinant IL-1a to naive or irradiated cells or the presence of IL-1 RA under the same conditions as those that interfered with the function of IL-8, did not affect micronuclei or DNA repair foci frequencies measured at day 7 after exposure. While doseresponse studies revealed that genomic instability and IL-8 production are the consequences of targeted effects, experiments employing a co-culture transwell system revealed the propagation of pro-inflammatory responses but not genomic instability from irradiated to nonirradiated cells. Collectively,

INTRODUCTION

The long-term health and cancer risks to humans from very low doses of radiation, even doses that do not cause damage to normal tissue, are a major concern (1, 2). The risk of cancer is a projected hazard for space travel as astronauts are exposed to heavy-ion particle radiation, which animal studies suggest is more effective in cancer induction and promotion (3). Genomic instability has been proposed to be an initiating event and a driver mechanism in radiation-induced carcinogenesis (4–6). Genomic instability is a transgenerational response to radiation exposure persisting in the progeny of cells that are directly or indirectly targeted and involves the emergence of nonclonal genomic changes, including point mutations, chromosomal missegregation and aberrations, gene amplification and other anomalies [reviewed in ref. (7)]. This response has been reported to occur in vitro in cell lines and in vivo after exposure to subtoxic doses of low- and high-LET radiation (8–11). While the biological consequences of genomic instability are still unclear, it is likely that prolonging the period of genomic changes beyond the acute events of radiationinduced DNA damage followed by repair will increase the yield of nonlethal damage and contribute to the overall risk and long-term biological impact of low-dose exposures (7). Extensive genomic rearrangements increase adaptation and fitness in yeast (12, 13), while in tumors these are associated with biologic heterogeneity (14) and the acquisition of resistance to therapy (15). Extrinsic and intrinsic factors have been reported to regulate persistent genomic instability. Intrinsic factors,

Editor’s note. The online version of this article (DOI: 10.1667/ RR14045.1) contains supplementary information that is available to all authorized users. 1 Address for correspondence: 1510 Clifton Rd., Rm 4013, Rollins O. Wayne Research Center, Emory University School of Medicine, Atlanta, GA 30322; emails: [email protected] and ewerner@ emory.edu. 621

622

WERNER, WANG AND DOETSCH

such as genetic susceptibility [reviewed in ref. (16)] and DNA repair status, modify genomic instability in irradiated cells in vitro and in vivo. For example, XRCC1 deficiency increases while Ku80 deficiency prevents genomic instability (17). Deficiencies in the homologous recombination pathway reduce the susceptibility for sister chromatid exchanges (18). Genetic background has been shown to modulate and, in some cases, define genomic instability outcomes (19, 20). Factors external to the cell, such as those mediating the bystander effect, clastogenic factors and inflammatory molecules (7), can modulate genomic instability. The bystander effect takes place when irradiated cells release factors that induce genomic instability in cells that were not directly targeted by radiation. This effect is engaged within the first several hours after exposure and can be induced by low radiation doses without a clear dependence on radiation dose and therefore, it has been proposed to be independent of the DNA damage and repair status of the irradiated/donor cell (21). Two mechanisms may mediate signaling to bystander cells: 1. The cells targeted by radiation release factors into the media (22–26) or 2. Transfer via gap junctions of effector molecules causing genomic instability in physically contacting nonirradiated cells (27). Soluble factors [reviewed in (28)] capable of inducing genomic instability may persist up to years in the plasma of irradiated individuals (29) and are produced by normal cells and tumors after radiotherapy and can exert long-range effects on other cell types in the body. Chromosomal instability in nonirradiated hematopoietic cells is induced by TNFa and FasL released by macrophages exposed in vivo to ionizing radiation (30) and can be mitigated by treatment with nonsteroidal anti-inflammatory agents (31). TNFa is sufficient to induce genomic instability in certain cell models in vitro (32) and promotes colorectal carcinogenesis in a murine model of chronic colitis (33). Thus a significant modulating factor of genomic instability may be concurrent inflammation, which is a generalized long-term response to relatively low doses of heavy-ion radiation affecting the central nervous system, hematopoietic system and heart (34–36). We previously described increased genomic instability measured by micronuclei formation and increased c-H2AX53BP1-positive foci frequency in CDK4 and telomeraseimmortalized human bronchial epithelial cells (HBEC3-KT) persisting at day 7 after exposure to either low- or high-LET radiation, with high-LET radiation being more effective at a 1 Gy dose (37). High-LET radiation also induced increased IL-8 production at day 7 after exposure. Because proinflammatory molecules contribute to genomic instability as well as to many other radiation-induced outcomes in some model systems [reviewed in (38)], we tested whether IL-8 is produced in the context of a wider pro-inflammatory response and also modulate persisting genomic instability in this cell model. Delineating the mechanisms involved in these processes is relevant for cancer risk estimation

because nontargeted responses carry more impact at low doses (3, 7) and inflammation has been shown to be a bona fide tumor promoter. Importantly, inflammatory responses as well as persistent foci are biological outcomes that can be readily detected in cells exposed to space radiation at lowEarth orbit level as demonstrated in human fibroblasts exposed to space radiation (39). The current study evaluates the contribution of inflammatory responses to genomic instability induced by low- and high-LET radiation at low doses in bronchial epithelial cells, which represent a cell type of a radiosensitive tissue with susceptibility to radiation-induced carcinogenesis. MATERIALS AND METHODS Reagents, Cell Culture and Irradiations All reagents, unless otherwise stated, were obtained from SigmaAldricht (St. Louis, MO). Human recombinant IL-1 receptor antagonist, anti-IL-1a function blocking antibody and IL-1a were obtained from R&D Systemse (Minneapolis, MN). Transwell filters (polycarbonate with 3 lm pores) were obtained from Corningt, Inc. (St. Lowell, MA). HBEC3-KT cells (40), a gift from Dr. Michael Story (UT Southwestern, Dallas, TX), were cultured in keratinocyte serum-free media from Invitrogene (Carlsbad, CA) supplemented with pituitary extract, EGF and antibiotics (penicillin and streptomycin). High-LET irradiations were performed using an alternatinggradient synchrotron (AGS), 600 MeV/nucleon 56Fe ions, at an approximate dose rate of 50 cGy/min at the NASA Space Radiation Laboratory at Brookhaven National Laboratory (Upton, NY). LowLET irradiations were performed using an X-ray machine (XRAD320, Precision X-Ray Inc., North Branford, CT) at 320 keV, 10 mA and at a dose rate of 1 Gy/min. For exposures, 200,000 cells were plated in a T25 flask two days before irradiation in triplicate flasks to ensure continuous proliferation before and after exposure. At day 4, each flask was subcultured at a 1:3 ratio. Typically, at this time we recover in the 1 Gy 56Fe-ion-irradiated flasks about half the number of cells contained in the nonirradiated flasks, which have gone through 3–4 population doublings. The cells were plated at day 6 for the experiments. During day 7 cells were treated with or 10 ng/ml IL-1a, 1 lg/ml IL-1a blocking antibody or 500 ng/ml IL-1RA. Samples were collected for analysis in the morning of day 8. Cytokine Profiling At day 6 after irradiation, 300,000 cells were seeded in 6-well plates. At day 7, the media was replaced with 1 ml keratinocyte serumfree media without supplements containing 0.02% BSA and conditioned for 24 h. For analysis employing the R&D Systems Human Cytokine array kit (panel A), the media of duplicate irradiations were combined to reduce the likelihood of profiling an outlier sample in each condition. Cytokinesis-Block Micronucleus Assay Cells were plated at a density of 20,000/well on glass coverslips at day 6 and treated at day 7 for 18 h with 3 lg/ml cytochalasin B in complete media (41). After fixation, they were stained and scored as previously described (37). Immunofluorescence Microscopy DNA repair foci were detected in cells fixed with 4% paraformaldehyde and permeabilized with 0.2% Tritone X-100. Antibodies used were c-H2AX (Millipore, Billerica, MA) and 53BP1 (Novus

CELL-AUTONOMOUS MECHANISMS DRIVE GENOMIC INSTABILITY (47 CH)

Biologicals, Littleton, CO). For intracellular IL-8 staining with antiIL-8 (BD Biosciences, San Jose, CA), cells were treated with 2 lM monensin for 4 h prior fixation to accumulate the cytokine in the cells. After fixation they were permeabilized using saponin. Cells were imaged in a Zeiss LSM510META confocal microscope (Thornwood, NY) using a 203 Plan-Apo objective (NA ¼ 0.75). Images were processed using contrast/brightness enhancement only. Foci positive for both markers were counted in at least five different fields totaling 50 or more cells in duplicate irradiations. Cytokine ELISA Cells (50,000) were plated in triplicate on day 6 and incubated at day 7 for 24 h in media without supplements and containing 0.02% BSA, with or without 5 ng/ml IL-1a, 500 ng/ml IL-1RA and 1 lg/ml anti-IL-1a blocking antibody. IL-8 was measured in the supernatants employing enzyme-linked immunoassay (R&D Systems). Measurements were normalized using crystal violet staining of the cells remaining in the well after supernatant collection. Relative Biological Effectiveness Calculations Clonogenic cell survival was determined by irradiation of the T25 flask with cells sparsely seeded 12 h before irradiation (42). In cells irradiated with 56Fe ions, killing was proportional to the dose. In response to X- ray exposure, HBEC3-KT cells exhibit a survival curve with a shoulder at low dose (37), where damage is considered to arise from single events. Thus, for purposes of estimating relative biological effectiveness (RBE), linear fitting was applied to the data employing KaleidaGraph (Synergy Software, Reading, PA). D0 was calculated interpolating in the obtained equations and RBE for survival estimated by the ratio between D0. The RBE for other biological end points was calculated as the ratio of the slopes of the linear portions of the curves, which were fitted employing KaleidaGraph. Statistical Analysis The error bars in the figures represent the standard deviation when associated with the average of independent samples irradiated simultaneously or the standard error when associated with the average of multiple independent irradiations and are indicated in the figure legend. The statistical test applied in each case is identified in the figure legend. Excel was used for paired two-tailed Student’s t test assuming equal variance of the samples when analyzing the effect of IL-1RA, IL-1 or the blocking antibody. GraphPad software (San Diego, CA) was used for analysis of variance (ANOVA) employing Tukey’s multiple comparison test as a post-test when comparing radiation types. Foci frequency distribution was analyzed employing a chi-square test for contingency tables in VassarStats (http://vassarstats. net) as previously reported (37).

RESULTS

In a previously reported study, we showed that exposure to 1 Gy 56Fe ions, but not X rays, induced a transient increase in IL-8 production at day 7 postirradiation in the context of increased genomic instability (37). Thus, we examined whether IL-8 production at day 7 is part of a broader pro-inflammatory response, which could include the involvement of cytokines in regulating and determining increased genomic instability. We profiled the response with an antibody array to analyze 36 human cytokines in media conditioned by irradiated cells and detected protein signatures, as proteins are direct indicators and effectors of

623

biological phenotypes. While IL-8 production increased proportionally with dose in response to low- and high-LET irradiation (Supplementary Fig. S1A; http://dx.doi.org/10. 1667/RR14045.1.S1), we compared the cytokine profiles after exposure to 3 Gy low-LET X rays and 1 Gy 56Fe ions, to relate this pro-inflammatory response to persistent genomic instability. At these doses, X rays and 56Fe ions induced similar frequencies of micronuclei formation at day 7 [(37) and Fig. 3]. The results of the array analysis are displayed as a heat map in Fig. 1A and as a plot of average densitometric units of the spots in the membrane in Fig. 1B. HBEC3-KT cells released significant amounts of macrophage migration inhibitory factor (MIF) and Serpin E1 constitutively, which were unaffected by exposure to radiation, and were similar in all samples, indicating comparable conditions for cytokine analysis. Exposure to both types of radiation induced multiple cytokines, including GM-CSF, IL-1a, IL-8 and the response modulator, IL-1 receptor antagonist (IL-1RA). Among the induced cytokines, IL-1a drives pro-inflammatory responses (43), thus we tested whether the addition of recombinant human IL-1a is sufficient to induce a qualitatively similar cytokine profile in nonirradiated cells. As shown in Fig. 1, treatment of nonirradiated cells with recombinant human IL-1a reproduced the cytokine profile induced by radiation as it induced the same type of cytokines. However, IL-1a was a weak inducer of GM-CSF production, a cytokine induced robustly by 56Fe ions. Exposure to radiation or IL-1a also induced the expression of IL-1 receptor antagonist, which binds competitively to the IL-1 receptor, thereby reducing or terminating this response when induced by IL-1a and/or b (44). It is relevant to note that this molecule is currently used as a biological therapy (anakinra) for inflammatory diseases such as rheumatoid arthritis [reviewed in (45)]. Next, we examined whether IL-8 is induced in an IL-1dependent mechanism by adding recombinant IL-1RA or an IL-1a function-blocking antibody and measuring IL-8 production employing a capture ELISA assay (Fig. 2). This quantitative assay reproduced the findings of the antibody array and revealed that IL-8 is produced in a twocomponent fashion as a function of X-ray and 56Fe-ion dose (Supplementary Fig. S1A; http://dx.doi.org/10.1667/ RR14045.1.S1), with an initial linear component and a change in slope for doses above 0.5 Gy for 56Fe ions and 3 Gy for X rays. The RBE of 56Fe ions for IL-8 induction at low doses is 4.1 (Supplementary Fig. S1A; http://dx.doi.org/ 10.1667/RR14045.1.S1). IL-8 levels induced by radiation exposure or by IL-1a addition could be reduced with IL1RA, which competes with IL-1a and b for binding to the receptors or with anti-IL-1a function-blocking antibody. Thus, these results indicate that exposure to radiation induces a pro-inflammatory response dependent on the production of IL-1a and suggest that it can be reduced by interfering with the activity of this cytokine. Given that IL-1a has been reported to induce senescence (46), a phenotype associated with persistent DNA damage,

624

WERNER, WANG AND DOETSCH

FIG. 1. Exposure to radiation induces a pro-inflammatory response reproduced by IL-1a. Panel A: Heat map of the average pixel density obtained from the densitometric analysis of positive spots in a human cytokine antibody array after incubation with media conditioned for 24 h by HBEC3-KT cells 7 days after mock irradiation (nonirradiated), 3 Gy X rays or 1 Gy 56Fe ions, and nonirradiated cells incubated for 24 h with 10 ng/ ml recombinant human IL-1a. Panel B: Graphical representation of the pixel intensities for duplicate determinations of each cytokine induced by radiation or IL-1a. Error bars represent SD of the averages.

FIG. 2. Radiation-induced IL-8 expression at day 7 after exposure is dependent on IL-1a activity. IL-8 levels measured by ELISA in media conditioned for 24 h by cells 7 days after mock irradiation or exposure to 3 Gy X rays or 1 Gy 56Fe ions and nonirradiated cells incubated for 24 h with 10 ng/ml recombinant human IL-1a with or without 500 ng/mL recombinant human IL-1RA or 1 lg/ml anti-IL-1a blocking antibody. Error bars are SE obtained from 3 biological replicates at 2 independent irradiations for X rays and 56Fe ions. Paired t test for IL-1RA and blocking antibody effect, *P , 0.05. Comparison among radiation treatments: one-way ANOVA comparisons of the mean. **P , 0.01, ***P , 0.005. ns ¼ no statistical significance.

and that genomic instability can be propagated by soluble factors in some model systems, we addressed next whether the pro-inflammatory response promotes the persistence of genomic instability in HBEC3-KT cells at day 7 postirradiation. Genomic instability encompasses multiple phenotypes, thus it was measured by two approaches: 1. Micronuclei formation frequency, which reports chromosomal damage (Fig. 3A); and 2. By the c-H2AX and 53BP1 markers, which are co-localized in DNA repair foci and report double-strand breaks (Fig. 3B). Both of these markers are associated with radiation-induced genomic instability and persist in irradiated tissue in vivo (47, 48) and in HBEC3-KT cells in vitro (37). Micronuclei were induced proportionally to radiation dose (Supplementary Fig. S1B; http://dx.doi.org/10.1667/RR14045.1.S1) and in a close-tolinear relationship in the tested dose range. 56Fe ions were more effective than X rays, with an RBE of 3.9 estimated from the ratio of the slopes of the curves after linear fitting (Supplementary Fig. S1B). To test the role of IL-1a in modulating genomic instability outcomes, we exposed the cells to 3 Gy X rays or 1 Gy 56Fe ions to induce similar micronuclei frequencies. However, the presence of IL-1RA did not affect low- or high-LET radiation-induced micronuclei formation frequency (Fig. 3A) or the number of DNA repair foci per cell (Fig. 3B). Consistent with the lack of effect of interfering with IL-1 activity in the biological end points measured here, addition of recombinant IL-1a to nonirradiated, X-rays or 56Fe-irradiated cells did not modify

CELL-AUTONOMOUS MECHANISMS DRIVE GENOMIC INSTABILITY (47 CH)

625

FIG. 3. IL-1a-driven responses do not affect radiation-induced persistent genomic instability. Seven days after mock irradiation (nonirradiated) or exposure to 3 Gy X rays or 1 Gy 56Fe ions and after 24 h treatment with or without 10 ng/ml IL-1a and/or 500 ng/ml IL-1 RA as indicated. Panel A: Micronuclei frequency. Error bars represent SE. n ¼ 2 independent irradiations of 3 biological replicate samples. Panel B: c-H2AX and 53BP1positive foci frequency in 1 of 2 experiments. Comparison among radiation treatments: paired chi-square test, *P , 0.05, **P , 0.01. ns ¼ no statistical significance.

either one of these indicators of genomic instability (Fig. 3A and B). To further exclude a role for minute amounts of IL-1a that are not completely eliminated by the antagonist and to evaluate the possible role of other soluble factors not detected by the array, we employed a transwell co-culture system and examined the capability of irradiated cells to release factors into the shared media and influence the phenotype of nonirradiated cells. At day 6 after exposure, 6 Gy X-ray- or 1 Gy 56Fe-ion-irradiated cells were placed in the top well and nonirradiated cells were placed on the bottom well (Fig. 4A). A X-ray dose of 6 Gy was chosen to produce robust responses and ensure that lack of effect cannot be attributed to insufficient levels of cytokines to induce a biological response. After 36 h of co-culture, the nonirradiated cells in the bottom well were analyzed for cell-associated IL-8 induction by immunofluorescence as a control for the biological effects of IL-1a released by the irradiated cells (Fig. 4B). This induction was reduced by the presence of recombinant IL-1RA in the media. Under these same conditions, we did not detect significant increases in micronuclei frequency (Fig. 4C) or induction of c-H2AX and 53BP1-positive foci (Fig. 4D) in nonirradiated cells co-cultured with X-ray- or 56Fe-ionsirradiated cells. Thus, under co-culture conditions where IL-1a released by irradiated cells induced IL-8 expression in nonirradiated cells, genomic instability was not induced, indicating that this phenotype is not induced by factors released by irradiated cells and acting in trans at day 7 after irradiation. These results are consistent with a cellautonomous mechanism driving genomic instability in this cell model.

DISCUSSION

This study addresses the contribution of inflammatory mediators to genomic instability at day 7 after exposure of human bronchial epithelial cells to low- and high-LET radiation. In line with our previous observations for genomic instability (37), high-LET 56Fe ions induced genomic instability more effectively (RBE of 3.9 when measured for micronuclei frequency) compared to low-LET X rays. Genomic instability at day 7 occurs in the context of IL-8 produced as part of an IL-1a-driven pro-inflammatory response involving the production of multiple cytokines. Although this response was proportional to dose and the RBE for 56Fe ions was 4.1, similar to the RBE for genomic instability, our results suggest that this pro-inflammatory response does not influence the concomitant genomic instability. We have also excluded the possibility that at 7 days after exposure, the genomic instability phenotype is sustained by other secreted factors acting in trans. Thus, these results suggest that the phenotype of genomic instability is cell autonomous in HBEC3-KT cells. Previous studies have documented the induction of IL-8 production by particle radiation in fibroblasts (49), longterm inflammation in the central nervous system (35) and also an acute transcriptional upregulation of multiple, proinflammatory cytokines during the first 2 h after exposure of thyroid epithelial cells to 56Fe ions (50). However, no further mechanistic insights or associations with other cell operational systems have been reported that relate to this response. We found that exposure to both high- and lowLET radiation induces a pro-inflammatory cascade that involves the expression of NFjB transactivated genes: GROa, GM-CSF, IL-1a, IL-1b and IL-8. The cytokine

626

WERNER, WANG AND DOETSCH

FIG. 4. Genomic instability is not affected by signaling molecules secreted by irradiated cells. Panel A: For this experimental setup, 6 days after mock irradiation or exposure to the indicated X-ray dose or 1 Gy 56Fe ion, HBEC3-KT cells were plated on transwell inserts with polycarbonate filters containing 3 lm pores (top) and cocultured for 36 h with nonirradiated cells plated on the lower well (bottom). When indicated, 500 ng/ml IL-1 RA was added to the media. Panel B: Immunofluorescence of intracellular accumulated IL-8 by a 4 h incubation with 2 lM monensin prior fixation at day 7 after 6 Gy X irradiation. Comparison among radiation treatments: one-way ANOVA comparisons of the mean. **P , 0.01, ***P , 0.005. ns ¼ no statistical significance. Panel C: Micronuclei frequency of irradiated (6 Gy X rays, 1 Gy 56Fe ions) versus co-cultured nonirradiated cells (Bottom). Panel D: c-H2AX- and 53BP1-positive foci frequency in nonirradiated cells (Bottom) after co-culture with nonirradiated cells or with 6 Gy X-ray- or 1 Gy 56Fe-ion-irradiated cells in the top well. One of two experiments is shown. Comparison among radiation treatments: paired chi-square test, *P , 0.05, **P , 0.01. ns ¼ no statistical significance.

profile induced by radiation was reproduced qualitatively by exposure of nonirradiated cells to exogenous, recombinant IL-1a. However, some of our results suggest that the response elicited by high-LET radiation might be more complex than that elicited by low-LET radiation; IL-1RA and the anti-IL-1a function-blocking antibody were not as efficient in reducing 56Fe ions compared to X-ray-induced IL-8 production. Additionally, GM-CSF was robustly induced by 56Fe ions, but only weakly by low-LET radiation or recombinant IL-1a. The selective expression of this cytokine in different tissues is controlled by epigenetic modification (51), thus this pattern could be a reflection of the activity of such a mechanism engaged selectively by

high-LET radiation. In non-immune cells, GM-CSF has context-dependent functions with reported anti-tumor as well as tumor-promoting activities (51, 52). In lung tissue, high levels of GM-CSF expression induce cell proliferation and hyperplasia (53), thus this could be one factor contributing to increased frequency of carcinogenesis in the lung in response to 56Fe-ion exposure (54, 55). IL-1a is a cell-associated cytokine present in most tissues, and is released upon cell activation and/or death, functioning as a stress signal to promote repair and restoration of tissue and organism homeostasis (43). In fact, injection of dead cells is sufficient to induce an IL-1-dependent systemic response (56). Such an activity is consistent with increased

CELL-AUTONOMOUS MECHANISMS DRIVE GENOMIC INSTABILITY (47 CH)

IL-8 production proportional to dose of both radiation types and with an RBE for 56Fe ions similar to the RBE for cell survival (see Supplementary Fig. S1, compare panel A with C; http://dx.doi.org/10.1667/RR14045.1.S1). Thus, IL-1a release is a candidate biomarker for reporting radiationinduced compromised cell function or cell death in certain tissues. Although the physiological function of this proinflammatory response mounted by epithelial cells remains to be explored, IL-1a released by inactivated cells can affect the surviving cells by recruiting inflammatory cells to the tissue, inducing senescence responses or cell proliferation and carcinogenesis, depending on the biological context (46). In the irradiated lung, increased IL-1a expression has been detected very early postirradiation (57, 58) and during the latent phase, preceding the development of fibrosis (59). At the organism level, IL-1a has radioprotective effects by acting on the immune system, although it also induces fever and inflammation (60). We report here that the cytokine response did not affect genomic instability induced by radiation in our in vitro cell model, pointing to a potentially different mechanism than that reported in the hematopoietic system (30). This conclusion is supported by the finding that TNFa was not detected in the media conditioned by HBEC3-KT cells (Fig. 1), and that the addition of recombinant IL-1a, which could induce a similar cytokine response, could not induce or modify radiation-induced genomic instability. In addition, interference of IL-1 activity with IL-1 receptor antagonist or an IL-1a function-blocking antibody did not modify genomic instability. Finally, we could not detect any other soluble factors in the media conditioned by irradiated cells that were not identified by the array [such as FasL or IGF-1, e.g., see refs. (61, 62)] that modified genomic instability in a co-culture system (Fig. 4). However, our experimental design does not allow us to exclude a role for short-rangeacting molecules such as ROS/NO or by direct cell-to-cell contact, mechanisms of signaling among irradiated and nonirradiated cells [see introduction and references in a published article by Sridharan et al. (38)]. It also does not exclude a role for molecules released earlier, such as bystander signaling occurring during the first hours after exposure, which could have prolonged consequences in the cell progeny (26). While we have chosen a single time point to perform our experiments and elucidate the mechanisms involved, our previous data suggest that at a 1 Gy dose, IL-8 production becomes undetectable at day 14 after exposure. Thus, in this specific in vitro model system, the cytokine response is transient. In vivo, an increasing number of reported studies demonstrate that exposure to radiation can induce proinflammatory responses, sustained over time in humans and rodents. In humans, most studies come from patients exposed to radiotherapy (63). In rodents, pro-inflammatory responses extended over time have been shown after exposures to low doses of low-LET radiation and heavyion radiation. However, IL-8 has not been examined

627

because rodents lack an IL-8 ortholog (64), but have CXCL1/KC, CXCL2/MIP-2 and CXCL5-6/LIX chemokines as functional homologues. Thus, further studies are needed to understand the relevance and consequences of these mechanisms in lung. Collectively, our results point to a cell-autonomous mechanism driving genomic instability in the context of multiple signal transduction pathways leading to the activation of pro-inflammatory responses. The dependency on radiation dose and the similar RBE for 56Fe ions, overlapping the effects of radiation on cell survival, suggests that both responses, genomic instability and IL-8 producing pathways, are the result of targeted effects. A cell-autonomous program driving genomic instability could explain the comparable effects observed on outcomes such as cytotoxicity and genomic rearrangement spectra after in vitro or in vivo irradiation of kidney epithelial cells (65). Such a mechanism also has important implications since it suggests that there is a core phenotype, which is relevant for the development of biomarkers to identify cells undergoing genomic instability in vivo and has the possibility to reveal candidate molecules to target this population of susceptible cells. SUPPLEMENTARY INFORMATION

Fig. S1. Dose-response curves for IL-8 production, micronucleus frequency and surviving fraction by HBEC3-KT cells at day 7 postirradiation. ACKNOWLEDGMENTS We thank the Emory University Integrated Cellular Imaging Microscopy Core for imaging support and resources. We also thank Peter Guida, Adam Rusek and crew at the NASA Space Radiation Laboratory at Brookhaven National Laboratory for assistance and technical support with particle irradiations. This work was funded by NASA-NSCOR (grant no. NNX11AC3OG). Received: February 5, 2015; accepted: September 9, 2015; published online: November 18, 2015

REFERENCES 1. Brenner DJ, Doll R, Goodhead DT, Hall EJ, Land CE, Little JB, et al. Cancer risks attributable to low doses of ionizing radiation: assessing what we really know. Proc Natl Acad Sci U S A 2003; 100:13761–6. 2. Mullenders L, Atkinson M, Paretzke H, Sabatier L, Bouffler S. Assessing cancer risks of low-dose radiation. Nat Rev Cancer 2009; 9:596–604. 3. Cucinotta FA, Chappell LJ. Non-targeted effects and the dose response for heavy ion tumor induction. Mutat Res 2010; 687:49– 53. 4. Sankaranarayanan K, Chakraborty R. Cancer predisposition, radiosensitivity and the risk of radiation-induced cancers. I. Background. Radiat Res 1995; 143:121–43. 5. Ullrich RL, Ponnaiya B. Radiation-induced instability and its relation to radiation carcinogenesis. Int J Radiat Biol 1998; 74:747–54. 6. Little JB. Radiation carcinogenesis. Carcinogenesis 2000; 21:397– 404.

628

WERNER, WANG AND DOETSCH

7. Kadhim M, Salomaa S, Wright E, Hildebrandt G, Belyakov OV, Prise KM, et al. Non-targeted effects of ionising radiation– implications for low dose risk. Mutat Res 2013; 752:84–98. 8. Chua ML, Somaiah N, Bourne S, Daley F, A’Hern R, Nuta O, et al. Inter-individual and inter-cell type variation in residual DNA damage after in vivo irradiation of human skin. Radiother Oncol 2011; 99:225–30. 9. Watson GE, Pocock DA, Papworth D, Lorimore SA, Wright EG. In vivo chromosomal instability and transmissible aberrations in the progeny of haemopoietic stem cells induced by high- and lowLET radiations. Int J Radiat Biol 2001; 77:409–17. 10. Sudo H, Garbe J, Stampfer MR, Barcellos-Hoff MH, Kronenberg A. Karyotypic instability and centrosome aberrations in the progeny of finite life-span human mammary epithelial cells exposed to sparsely or densely ionizing radiation. Radiat Res 2008; 170:23–32. 11. Smith LE, Nagar S, Kim GJ, Morgan WF. Radiation-induced genomic instability: radiation quality and dose response. Health Phys 2003; 85:23–9. 12. Chen G, Bradford WD, Seidel CW, Li R. Hsp90 stress potentiates rapid cellular adaptation through induction of aneuploidy. Nature 2012; 482:246–50. 13. Gresham D, Desai MM, Tucker CM, Jenq HT, Pai DA, Ward A, et al. The repertoire and dynamics of evolutionary adaptations to controlled nutrient-limited environments in yeast. PLoS Genet 2008; 4:e1000303. 14. de Bruin EC, McGranahan N, Mitter R, Salm M, Wedge DC, Yates L, et al. Spatial and temporal diversity in genomic instability processes defines lung cancer evolution. Science 2014; 346:251–6. 15. Lee AJ, Endesfelder D, Rowan AJ, Walther A, Birkbak NJ, Futreal PA, et al. Chromosomal instability confers intrinsic multidrug resistance. Cancer Res 2011; 71:1858–70. 16. Kadhim MA. Role of genetic background in induced instability. Oncogene 2003; 22:6994–9. 17. Kashino G, Suzuki K, Kodama S, Watanabe M, Prise KM. Increased susceptibility to delayed genetic effects of low dose Xirradiation in DNA repair deficient cells. Int J Radiat Biol 2013; 89:295–300. 18. Nagasawa H, Wilson PF, Chen DJ, Thompson LH, Bedford JS, Little JB. Low doses of alpha particles do not induce sister chromatid exchanges in bystander Chinese hamster cells defective in homologous recombination. DNA Repair (Amst) 2008; 7:515– 22. 19. Lorimore SA, Mukherjee D, Robinson JI, Chrystal JA, Wright EG. Long-lived inflammatory signaling in irradiated bone marrow is genome dependent. Cancer Res 2011; 71:6485–91. 20. Nagasawa H, Huo L, Little JB. Increased bystander mutagenic effect in DNA double-strand break repair-deficient mammalian cells. Int J Radiat Biol 2003; 79:35–41. 21. Kashino G, Suzuki K, Matsuda N, Kodama S, Ono K, Watanabe M, et al. Radiation induced bystander signals are independent of DNA damage and DNA repair capacity of the irradiated cells. Mutat Res 2007; 619:134–8. 22. Lehnert BE, Goodwin EH, Deshpande A. Extracellular factor(s) following exposure to alpha particles can cause sister chromatid exchanges in normal human cells. Cancer Res 1997; 57:2164–71. 23. Suzuki M, Zhou H, Geard CR, Hei TK. Effect of medium on chromatin damage in bystander mammalian cells. Radiat Res 2004; 162:264–9. 24. Yang H, Asaad N, Held KD. Medium-mediated intercellular communication is involved in bystander responses of X-rayirradiated normal human fibroblasts. Oncogene 2005; 24:2096– 103. 25. Bowler DA, Moore SR, Macdonald DA, Smyth SH, Clapham P, Kadhim MA. Bystander-mediated genomic instability after high LET radiation in murine primary haemopoietic stem cells. Mutat Res 2006; 597:50–61.

26. Ponnaiya B, Suzuki M, Tsuruoka C, Uchihori Y, Wei Y, Hei TK. Detection of chromosomal instability in bystander cells after Si490-ion irradiation. Radiat Res 2011; 176:280–90. 27. Azzam EI, de Toledo SM, Little JB. Direct evidence for the participation of gap junction-mediated intercellular communication in the transmission of damage signals from alpha-particle irradiated to nonirradiated cells. Proc Natl Acad Sci U S A 2001; 98:473–8. 28. Mukherjee D, Coates PJ, Lorimore SA, Wright EG. Responses to ionizing radiation mediated by inflammatory mechanisms. J Pathol 2013; 232:289–99. 29. Lindholm C, Acheva A, Salomaa S. Clastogenic plasma factors: a short overview. Radiat Environ Biophys 2010; 49:133–8. 30. Lorimore SA, Chrystal JA, Robinson JI, Coates PJ, Wright EG. Chromosomal instability in unirradiated hemaopoietic cells induced by macrophages exposed in vivo to ionizing radiation. Cancer Res 2008; 68:8122–6. 31. Mukherjee D, Coates PJ, Lorimore SA, Wright EG. The in vivo expression of radiation-induced chromosomal instability has an inflammatory mechanism. Radiat Res 2012; 177:18–24. 32. Yan B, Wang H, Rabbani ZN, Zhao Y, Li W, Yuan Y, et al. Tumor necrosis factor-alpha is a potent endogenous mutagen that promotes cellular transformation. Cancer Res 2006; 66:11565–70. 33. Popivanova BK, Kitamura K, Wu Y, Kondo T, Kagaya T, Kaneko S, et al. Blocking TNF-alpha in mice reduces colorectal carcinogenesis associated with chronic colitis. J Clin Invest 2008; 118:560–70. 34. Janko M, Ontiveros F, Fitzgerald TJ, Deng A, DeCicco M, Rock KL. IL-1 generated subsequent to radiation-induced tissue injury contributes to the pathogenesis of radiodermatitis. Radiat Res 2012; 178:166–72. 35. Rola R, Sarkissian V, Obenaus A, Nelson GA, Otsuka S, Limoli CL, et al. High-LET radiation induces inflammation and persistent changes in markers of hippocampal neurogenesis. Radiat Res 2005; 164:556–60. 36. Tungjai M, Whorton EB, Rithidech KN. Persistence of apoptosis and inflammatory responses in the heart and bone marrow of mice following whole-body exposure to 28Silicon (28Si) ions. Radiat Environ Biophys 2013; 52:339–50. 37. Werner E, Wang H, Doetsch PW. Opposite roles for p38MAPKdriven responses and reactive oxygen species in the persistence and resolution of radiation-induced genomic instability. PLoS One 2014; 9:e108234. 38. Sridharan DM, Asaithamby A, Bailey SM, Costes SV, Doetsch PW, Dynan WS, et al. Understanding cancer development processes after HZE-particle exposure: roles of ROS, DNA damage repair and inflammation. Radiat Res 2015; 183:1–26. 39. Dieriks B, De Vos W, Meesen G, Van Oostveldt K, De Meyer T, Ghardi M, et al. High content analysis of human fibroblast cell cultures after exposure to space radiation. Radiat Res 2009; 172:423–36. 40. Ramirez RD, Sheridan S, Girard L, Sato M, Kim Y, Pollack J, et al. Immortalization of human bronchial epithelial cells in the absence of viral oncoproteins. Cancer Res 2004; 64:9027–34. 41. Fenech M. Cytokinesis-block micronucleus cytome assay. Nat Protoc 2007; 2:1084–104. 42. Hall EJ. Radiobiology for the radiobiologist. 5th ed. Philadelphia: Lippincott Williams & Wilkins; 2000. 43. Rider P, Carmi Y, Voronov E, Apte RN. Interleukin-1alpha. Semin Immunol 2013; 25:430–8. 44. Gehr G, Braun T, Lesslauer W. Cytokines, receptors, and inhibitors. Clin Investig 1992; 70:64–9. 45. Dinarello CA, van der Meer JW. Treating inflammation by blocking interleukin-1 in humans. Semin Immunol 2013; 25:469– 84. 46. Orjalo AV, Bhaumik D, Gengler BK, Scott GK, Campisi J. Cell surface-bound IL-1alpha is an upstream regulator of the senes-

CELL-AUTONOMOUS MECHANISMS DRIVE GENOMIC INSTABILITY (47 CH)

47.

48.

49.

50.

51.

52.

53.

54.

55.

cence-associated IL-6/IL-8 cytokine network. Proc Natl Acad Sci U S A 2009; 106:17031–6. Bhogal N, Kaspler P, Jalali F, Hyrien O, Chen R, Hill RP, et al. Late residual gamma-H2AX foci in murine skin are dose responsive and predict radiosensitivity in vivo. Radiat Res 2010; 173:1–9. Schanz S, Schuler N, Lorat Y, Fan L, Kaestner L, Wennemuth G, et al. Accumulation of DNA damage in complex normal tissues after protracted low-dose radiation. DNA Repair (Amst) 2012; 11:823–32. Narayanan PK, LaRue KE, Goodwin EH, Lehnert BE. Alpha particles induce the production of interleukin-8 by human cells. Radiat Res 1999; 152:57–63. Sanzari JK, Nuth M, Kennedy AR. Induction of cytokine gene expression in human thyroid epithelial cells irradiated with HZE particles (iron ions). Radiat Res 2009; 172:437–43. Urdinguio RG, Fernandez AF, Moncada-Pazos A, Huidobro C, Rodriguez RM, Ferrero C, et al. Immune-dependent and independent antitumor activity of GM-CSF aberrantly expressed by mouse and human colorectal tumors. Cancer Res 2012; 73:395– 405. Mueller MM, Fusenig NE. Constitutive expression of G-CSF and GM-CSF in human skin carcinoma cells with functional consequence for tumor progression. Int J Cancer 1999; 83:780–9. Huffman Reed JA, Rice WR, Zsengeller ZK, Wert SE, Dranoff G, Whitsett JA. GM-CSF enhances lung growth and causes alveolar type II epithelial cell hyperplasia in transgenic mice. Am J Physiol 1997; 273:L715–25. Delgado O, Batten KG, Richardson JA, Xie XJ, Gazdar AF, Kaisani AA, et al. Radiation-enhanced lung cancer progression in a transgenic mouse model of lung cancer is predictive of outcomes in human lung and breast cancer. Clin Cancer Res 2014; 20:1610– 22. Wang X, Farris Iii AB, Wang P, Zhang X, Wang H, Wang Y. Relative effectiveness at 1 gy after acute and fractionated exposures of heavy ions with different linear energy transfer for lung tumorigenesis. Radiat Res 2015; 183:233–9.

629

56. Chen CJ, Kono H, Golenbock D, Reed G, Akira S, Rock KL. Identification of a key pathway required for the sterile inflammatory response triggered by dying cells. Nat Med 2007; 13:851–6. 57. Ao X, Zhao L, Davis MA, Lubman DM, Lawrence TS, Kong FM. Radiation produces differential changes in cytokine profiles in radiation lung fibrosis sensitive and resistant mice. J Hematol Oncol 2009; 2:6. 58. Rube CE, Uthe D, Wilfert F, Ludwig D, Yang K, Konig J, et al. The bronchiolar epithelium as a prominent source of proinflammatory cytokines after lung irradiation. Int J Radiat Oncol Biol Phys 2005; 61:1482–92. 59. Johnston CJ, Piedboeuf B, Rubin P, Williams JP, Baggs R, Finkelstein JN. Early and persistent alterations in the expression of interleukin-1 alpha, interleukin-1 beta and tumor necrosis factor alpha mRNA levels in fibrosis-resistant and sensitive mice after thoracic irradiation. Radiat Res 1996; 145:762–7. 60. Dinarello CA, Neta R. An overview on interleukin-1 as a therapeutic agent. Biotherapy 1989; 1:245–54. 61. Rastogi S, Coates PJ, Lorimore SA, Wright EG. Bystander-type effects mediated by long-lived inflammatory signaling in irradiated bone marrow. Radiat Res 2011; 177:244–50. 62. Cianfarani S, Tedeschi B, Germani D, Prete SP, Rossi P, Vernole P, et al. In vitro effects of growth hormone (GH) and insulin-like growth factor I and II (IGF-I and -II) on chromosome fragility and p53 protein expression in human lymphocytes. Eur J Clin Invest 1998; 28:41–7. 63. Kuo AH, Ancukiewicz M, Kozak KR, Yock TI, Padera TP. Cardiac and inflammatory biomarkers do not correlate with volume of heart or lung receiving radiation. Radiat Oncol 2015; 10:5. 64. Guenet JL. The mouse genome. Genome Res 2005; 15:1729–40. 65. Turker MS, Connolly L, Dan C, Lasarev M, Gauny S, Kwoh E, et al. Comparison of autosomal mutations in mouse kidney epithelial cells exposed to iron ions in situ or in culture. Radiat Res 2009; 172:558–66.