Abstract
Ionizing radiation (IR) activates DNA damage response (DDR) that includes induction of bystander effects (BE) in cells not directly targeted by radiation. How DDR pathways in irradiated cells stimulate BE in non-targeted cells is mostly unknown. We show here that extracellular vesicles from irradiated cells (EV-IR) induce reactive oxygen species (ROS), increase γH2AX and Rad51-foci, and reduce clonogenic survival when taken up by un-irradiated cells. Among the direct effects of IR is Abl nuclear accumulation; interestingly, EV-IR from Abl NLS-mutated cells could not induce BE, and restoration of nuclear Abl rescued that defect. Extending our previous finding that Abl stimulates miR-34c expression in DDR, we found that that nuclear Abl also increased miR-34c levels in EV-IR. Co-expression of miR-34c-minigene and activated Abl led to the production, without irradiation, EV-miR-34c with BE-inducing activity. Furthermore, EV-IR from miR34-knockout cells could not induce ROS and raised γH2AX levels to lesser extent than those from miR34-wild type cells. These results establish a novel role for nuclear Abl and miR-34c in transmitting DNA damage response from directly irradiated cells to un-irradiated bystander cells.
In multicellular organisms, ionizing radiation (IR) damages DNA to activate a wide range of DDR not only in directly irradiated cells but also in neighboring or distant cells not targeted by IR. The non-target or bystander effects (BE) of IR occurs when irradiated cells secrete soluble factors and extracellular vesicles (EV) to propagate the damage signal to naïve, non-irradiated cells1-3. It was reported that DDR master regulators, i.e., ATM and p53, are required for irradiated cells to secrete bystander effectors4,5, however, how other DDR pathways stimulate BE is mostly unknown. IR stimulates nuclear accumulation and activation of Abl tyrosine kinase to regulate transcription, DNA repair and microRNA processing6-10. Because EV can transfer microRNAs between cells11-14, we investigated the role of nuclear Abl and its regulated microRNA, miR-34c, in EV-mediated bystander effects of radiation.
We isolated extracellular vesicles (EV) from media conditioned by immortalized mouse embryo fibroblasts (MEFs) and avoided EV from serum by switching cells into serum-free media before irradiation (extended Fig. E1A)15,16. We found similar size distributions, particle concentrations and total proteins among multiple independent EV preparations from non-irradiated (EV-C) or irradiated (EV-IR) MEFs and HEK293T cells (extended Figs E1B, C; E9D, E; E11A-D; E13A). We also found that naïve non-irradiated MEFs (responders) internalize EV-C and EV-IR to comparable levels (extended Fig. E1D, E). Thus, IR does not grossly alter the production or the uptake of EV.
Inhibition of colony formation is both a direct and a bystander effect of ionizing radiation (Fig. 1A, B)1, as media conditioned by irradiated MEFs (CM-IR) inhibits colony formation when transferred to responder MEFs (Fig. 1A, B). We found that the EV-pellet fraction of CM-IR retained more of this colony-inhibitory activity than the supernatant fraction (Fig. 1A, B, EV-IR). Titration experiments showed EV-IR to inhibit colony formation in a dose-dependent manner, reaching saturation at EV-protein levels that approximated a particle/responder cell ratio of 1000 (Fig. 1E, F), while EV-C did not elicit such dose-response (Fig. 1C, D). While direct irradiation of responder MEFs increased p21Cip1 mRNA and protein (extended Fig. E2), induced cell cycle arrest (extended Fig. E3) and senescence (extended Fig. E4), we found that EV-IR did not induce p21Cip1 (extended Fig. E2), cell-cycle arrest (extended Fig. E3), or senescence (extended Fig. E4). Although immortal, the responder MEFs formed colonies at a low frequency (~5%), and EV-IR inhibited the colony-forming activity of this small fraction of responder cells without inducing cell cycle arrest or senescence in the general population that internalized EV-IR (extended Fig. E1D, E).
EV-IR, but not EV-C, also caused dose-dependent increase in the production of reactive oxygen species (ROS) (Fig. 2A, B), which is another direct and bystander effect of IR17. The anti-oxidant N-Acetyl Cysteine (NAC) neutralized EV-IR-induced ROS increase (Fig. 2A, EV-IR+NAC) and interfered with the colony inhibitory activity of EV-IR (Fig. 2C), suggesting that ROS was a major albeit not the only contributing factor to colony inhibition. Treatment with Proteinase K or RNase A16-18 did not abolish the colony-inhibitory or the ROS-inducing activity of EV-IR (extended Fig. E5), indicating that the bystander inducers were sequestered within EV-IR.
Another bystander effect of radiation is the induction of DNA damage in non-irradiated cells19. We found that addition of EV-IR, but not EV-C, unto responder cells increased γH2AX (Fig. 2D, extended Fig. E6B) and Rad51-foci (Fig. 2E,F, extended Fig. E7). Although NAC interfered with EV-IR-induced ROS, it did not block EV-IR-induced γH2AX increase (Fig. 2D). Because EV-IR caused γH2AX and Rad51 foci to increase in only a subpopulation of responder cells (extended Fig. E6B, Fig. 2F, extended Fig. E7), and because γH2AX and Rad51 foci can occur during DNA replication20,21, we tested and found that the Cdk-inhibitor Roscovitine (Rosc) could prevent EV-IR from increasing γH2AX (extended Fig. E8). Thus, EV-IR might cause replication stress to induce bystander DNA damage
To determine the function of nuclear Abl in DDR, we generated the Abl-µNLS allele in mouse germline by mutating the three nuclear localization signals (NLS) in the Abl protein (Fig. 3A)7,22. Although IR did not induce nuclear accumulation of Abl-µNLS (Fig. 3B), it increased ROS and γH2AX in Abl-µNLS MEFs (Fig. 3C, D) and decreased colony formation (extended Fig. E9A). Thus, nuclear Abl is not essential to the direct effects of radiation in mouse embryo fibroblasts. In contrast, we found that EV from irradiated Abl-µNLS MEFs (µEV-IR) (extended Fig. E9D, E) did not induce ROS (Fig. 3E), increase γH2AX (Fig. 3F), nor inhibit colony formation (extended Fig. E9F) in responder MEFs (Abl-wt). We also found that the levels of γH2AX in directly irradiated MEFs were significantly higher than those in EV-IR-treated responder MEFs (extended Fig. E9B & E9C); this could explain why EV-IR did not induce cell cycle arrest. To rescue the µEV-IR defects, we stably expressed AblWT (or AblµNLS as control) in Abl-µNLS MEFs and showed that IR induced nuclear accumulation of AblWT but not AblµNLS (extended Fig. E10A-C). Expression of AblWT or AblµNLS did not affect the direct effects of IR in Abl-µNLS MEFs (extended Fig. 10D-F); however, AblWT but not AblµNLS restored the ability of µEV-IR to increase ROS and γH2AX in responder MEFs (extended Fig. 11E, F). Together, these results establish that nuclear entry of Abl is dispensable to the direct effects of radiation but it is required for irradiated cells to produce BE-inducing EV-IR.
IR induces the expression of many miRs in directly irradiated cells23, with Abl stimulating pri-miR34b/c processing to pre-miR-34b and pre-miR-34c9. In keeping with those results, we found higher levels of miR-34c in directly irradiated Abl-wt than Abl-µNLS MEFs (Fig. 4A). We also found that EV-IR contained higher levels of miR-34c than EV-C (Fig. 4B), but µEV-IR did not contain higher miR-34c levels than µEV-C (Fig. 4B). In responder cells, miR-34c levels increased only after treatment with EV-IR but not with EV-C, µEV-C or µEV-IR (Fig. 4C). Re-expression of AblWT in Abl-µNLS MEFs restored miR-34c increase in µEV-IR, and in responder cells treated with µEV-IR (extended Fig. E12A-C). These results showed that nuclear Abl stimulated not only the expression but also the inclusion of miR-34c in EV-IR for transfer into responder cells.
To determine if miR-34c is an inducer of BE, we co-expressed miR-34c-minigene with a constitutively activated Abl kinase (AblPPn) (Fig. 4D) in HEK293T cells. The miR-34c-minigene raised miR-34c levels in transfected cells (Fig. 4E) and in EV isolated from the media of those transfected cells (Fig. 4F, extended Fig. E13A). Co-expression with AblPPn further increased the intracellular and the EV levels of miR-34c (Fig. 4E, F). When added to responder MEFs, both EV-miR-34c and EV-miR-34c+AblPPn increased the intracellular levels of miR-34c (Fig. 4G), inhibited colony formation (Fig. 4H, extended Fig. E13B), induced ROS (Fig. 4I, extended Fig. E13C, D) and the extent of those effects correlated with the levels of miR-34c in EV and in EV-treated responder cells (Fig. 4F,G). With γH2AX, only EV-miR-34c+AblPPn significantly increased its levels above background (Fig. 4J, extended Fig. E13E, F), indicating that higher levels of miR34c might be required to cause DNA damage. To determine if miR-34c is necessary to cause BE, we isolated EV-IR from media conditioned by miR34-family (a, b, c) triple knockout MEFs (miR34TKO) and wild-type littermate MEFs (miR34WT)24. We found that EV-IR-miR34TKO failed to induce ROS in responder MEFs (Abl-wt) (Fig. 4K, extended Fig. E13G, H). EV-IR-miR34TKO treated MEFs show γH2AX staining but to significantly lower levels than EV-IR-miR34WT (Fig. 4L, extended Fig. E13I, J). These results showed that miR-34c and related family members (miR-34a, miR-34b) are required for EV-IR-induced ROS and they contribute to γH2AX increase in responder cells.
Previous studies have identified a number of pro-mitogenic genes as miR-34 targets25, 26. We tested three such targets and found in responder cells that EV-IR treatment did not reduce Pdgfra or Pdgfrb but significantly reduced Myc RNA (extended Fig. E14); this Myc reduction might account for the colony inhibitory activity of EV-IR and EV-miR-34c. Our finding that NAC reduced ROS but not γH2AX suggests that these EV-IR effects may involve multiple miR-34c target genes. Computational analyses have predicted hundreds of miR-34c targets that may be involved in the observed increase in ROS or γH2AX. It is possible that miR-34c triggers a cascade of gene expression alterations or this microRNA may collaborate with other EV-delivered factors to increase ROS and γH2AX. Our findings that nuclear Abl is essential for irradiated cells to produce EV-IR with ROS- and γH2AX-inducing activities, but the miR34-family is essential only for the ROS-inducing activity suggest that EV-IR must contain other nuclear-Abl-dependent DNA damage-inducers that remain to be identified.
MATERIALS & METHODS
Cell Lines
Fibroblasts were derived from Abl+/+ (Abl-wt) or littermate Ablµ/µ (Abl-µNLS) mouse embryos22. The Abl-µNLS allele was generated by knock-in mutations to substitute the eleven lysines and arginines in the three nuclear localization signals (NLS) with glutamine7. The Abl-wt and Abl-µNLS mouse embryo fibroblasts (MEFs) were immortalized by serial passages, and these MEFs do not express p53. Primary, non-immortalized, MEFs from miR-34a, b, c- triple knockout mice (miR34TKO) and wild-type littermates (miR34WT)24 were irradiated between passages 3 and 6. MEFs and HEK293T cells (Thermo Fisher Scientific) were cultured in DMEM high glucose media with 10% fetal bovine serum (FBS) and antibiotics.
Irradiation
Cells were exposed to 10 Gy of gamma-irradiation using Mark I model 50 irradiator with Cesium 137 isotope as source (Maker: J.L. Shepherd & Associates).
Isolation of Extracellular Vesicle (EV)
To avoid EV from fetal bovine serum (FBS), cells were switched to FBS-free media with 1% BSA two hours before irradiation. At 24 hours after irradiation, the media were collected for EV isolation by differential ultracentrifugation as previously described15 (extended Fig. E1A). The pelleted EV fractions were washed and re-suspended in PBS and stored in aliquots at −80° C. For isolation of EV from HEK293T cells, supernatant collected after the 2000g spin was filtered through a 0.45-micron filter (Corning) before continuing onto the next steps of ultracentrifugation. Protein content of EV was determined by the Lowry method.
Nanoparticle Track Analysis
Nanosight LM-10HS was used to determine the number and the size distribution of particles in the EV fractions. Each EV preparation was analyzed in triplicates as previously described27.
Extracellular Vesicle Uptake
The EV pellets were incubated with PKH26, a fluorescent membrane-binding dye (Sigma Aldrich, St. Louis) for 5min at room temperature, followed by addition of 1% BSA, and then centrifuged at 100,000g for 70min to isolate PKH26-labeled EV28. Responder MEFs were incubated with PKH26 in PBS (phosphate buffered saline), or PKH26-labeled EV-C or PKH26-labeled EV-IR (25µg each). After 3 or 24 hours, cells were fixed with 4% para-formaldehyde (PFA) for 20min at room temperature and counterstained with Hoechst 33342. Cells were viewed using an Olympus FV1000 Spectral Confocal microscope. No fluorescence was detected in cells incubated with PKH in PBS. Using FIJI (ImageJ), we measured the PKH26 mean gray values from at least 200 cells per treatment and calculated the mean and standard deviations. The number of PKH26-positive cells was counted by eye, and percentages were calculated from PKH26-positive cells over total number of nuclei.
Colony Formation Assay
Responder cells (Abl-wt MEFs) were seeded at 1000 cells per 6-cm plate. Media was changed to 1% BSA without FBS before incubation with EV. After 24 hours, cells were switched back to media with 10% FBS and cultured for 15 days with media refreshed every other day. The colonies were fixed with 100% methanol and stained with 0.05% crystal violet. Excess dye was removed & plates were left to dry over-night. Cluster of more than 50 cells were considered as colonies. Survival fraction was calculated as colonies/cells seeded with the survival fraction in PBS treated plates set to 1. Images of the colonies were acquired using Alpha imager HP System.
Reactive Oxygen Species (ROS) Assay
ROS was measured using the ROS-ID kit (Enzo Life Sciences, Farmingdale) according to manufacturer’s protocol. Live cells were also stained with Cell Tracker Red (CTR) (Molecular Probe) as a control for cell volume. Responder cells were seeded into chamber slides, incubated with EV in serum-free media containing 1% BSA for 24 hours, then stained with CTR and DCFDA. Immediately after dye addition, live cell images were captured using an Olympus FV1000 Spectral Confocal Microscope for CTR (Channel 3) & DCFDA (Channel 1). FIJI (ImageJ) software was used to create masks of cells from channel 3 (CTR), and then the masks were transferred onto channel 1 (DCFDA). The mean gray values were measured within the masks, and the DCFDA/CTR mean gray value ratio was calculated for each mask. Ratios of at least 200 cells were calculated per sample per experiment. The DCFDA/CTR ratios of individual cells from a representative experiment are shown in extended Fig. E6A.
Immunofluorescence
Acid-washed coverslips stored in 100% ethanol were placed in 24-well plates and approximately 20,000 MEFs were seeded per well. After incubation with EV for 24 hours in serum-free media containing 1% BSA, cells were fixed in 4% PFA for 15 min, washed with 0.02% Tween-20 in Tri-buffered saline (TBS) twice (5 min each), permeablized with 1% Triton X-100 in TBS for 15 min and then blocked with 5% BSA for 30 min at room temperature. The coverslips were incubated with primary antibody for 1 hour in 37°C: anti-Abl (8E9) (6µg/ml) from ThermoFisher Scientific, anti-phospho-Ser139-H2AX (1/400) from Cell Signaling, anti-RAD51 (1/50) from Santa Cruz. Coverslips were washed twice with 0.02% Tween-20 in TBS twice (5 min each) and then incubated with ALEXA fluor-488 (Invitrogen)-chicken anti-mouse (1/500) or ALEXA fluor-594 (Invitrogen)-donkey anti-rabbit (1/500) at 37° C for 30 min. Nuclei were stained with Hoechst 33342. Coverslips were mounted with Prolong Gold Antifade Reagent and sealed with nail polish before imaging. Images were captured using an Olympus FV1000 Spectral Confocal Microscope.
Quantification of γ-H2AX
FIJI (ImageJ) software was used to create masks of cells from channel 0 (Hoechst 33342), the masks were then transferred onto channel 1 (anti-γH2AX), and the channel 1 mean gray value within each mask was recorded. At least 200 cells were scored per sample per experiment. The γH2AX mean gray values of individual cells in a representative experiment are shown in extended Fig. E6B.
Senescence-Associated β-Galactosidase Staining
Irradiated (10 Gy) or EV-treated (25µg, 24 hours) MEFs were cultured for additional 5 days in serum-supplemented media and counterstained with Hoechst 33342 and the Senescence Cells Histochemical Staining Kit (Sigma-Aldrich). The number of nuclei was counted in images captured by the KEYENCE BZ-X700 All-in-One Fluorescence Microscope at a magnification of 20X, with 50% transmitted light, and white balance red-blue-green areas of 1.46, 0.98, and 1.24, respectively. The β-galactosidase positive blue cells were counted under Phase contrast microscope. At least 200 cells were scored per sample per experiment to calculate the percentage of positive senescent cells.
Immunoblotting
Cell pellets were lysed in RIPA buffer (25mM Tris-HCl pH 7.6, 10% Glycerol, 1% NP40, 0.5% sodium deoxycholate, 1x Protease inhibitors (Roche), 150mM Sodium chloride, 50mM Sodium fluoride, 10mM Sodium beta-glycerophosphate, 10mM Sodium orthovandate, 10mM sodium pyrophosphate, 1mM PMSF). Proteins were separated using SDS-PAGE & transferred onto Nitrocellulose membranes (Millipore). Membranes were blocked for 1 hour at room temperature, incubated with anti-Abl (8E9) (1/500) & anti-actin (1/2000) from Sigma Aldrich for 1 hour, washed and incubated with secondary antibody (Anti-mouse: HRP-linked) & developed using ECL reagents (Pierce).
Cell Cycle Analysis
Cells were collected at 24 hours post irradiation or EV treatment by trypsinization, followed by centrifugation at 1400 RPM for 6 min. Cell-pellets were fixed in ice cold 70% ethanol overnight and stained with 40 µg/ml propidium iodide (PI) (Sigma) and 100 µg/ml RNaseA (Sigma) for 30 min at 37°C in the dark. PI staining was analyzed using Sony SH800 FACS sorter and software.
Retrovirus Packaging and Infection
AblWT & AblµNLS were stably expressed in Abl-µNLS MEFs by retroviral infection29. BOSC23 cells were transfected with retroviral vector pMSCV expressing AblWT or AblµNLS. Culture media collected at 48 hours after transfection was filtered and added to Abl-µNLS MEFs with polybrene (4µg/ml). Infected cells were then selected for resistance to hygromycin (150 µg/ml).
Transfection
Genetran (Biomiga) was used to transfect HEK239T cells with miR-34c-minigene and pCDNA3-AblPPn plasmid DNA9. Transfected cells and their media (for EV isolation) were collected 24 hours after transfections.
RNA measurements
The SeraMir Exosome RNA amplification kit (System Biosciences) was used to extract RNA from EV pellets. Total cellular RNA was extracted using TRIzol (Life Technologies). Synthesis of cDNA was carried out using ABI reverse Transcription kit (Life Technologies). For measurements of mature miR-34c, stem-loop primer was used for reverse transcription30,9. U6 was used as the reference gene for normalization of miR-34c abundance. GAPDH was used as reference gene for normalization of p21Cip1, Pdgfrb, Pdgfra & Myc abundance. Real-time PCR reactions were carried out using StepOnePlus system. Subtraction of the reference gene CT value from the experimental gene CT value generated the normalized ∆CT. Relative abundance was then calculated as 2-ΔΔCT, where ΔΔCT values were ΔCT of sample subtracted by ΔCT of vehicle-treated or vector transfected cells. For Primer sequences please refer to the Primer List in extended Fig. E14B.
Statistical Analysis
The statistical analyses were performed using Graph-Pad Prism 6. For Clonogenic survival & qRT-PCR measurements: the values have been shown as mean ± SD from three independent experiments were analyzed using ONE WAY ANOVA. For ROS & γH2AX measurements the values have been shown as medians with interquartile range from the indicated number of independent experiments (at least 200 cells analyzed per sample per experiment) were analyzed using Kruskal-Wallis test. For each test ns: not significant; *P≤ 0.05, **P≤ 0.01, *** P≤ 0.001, **** P≤ 0.0001.
AUTHOR CONTRIBUTIONS
SR designed and performed the experiments, analyzed the data, and wrote the paper. A.H, J.C, performed experiments and analyzed the data. JYJW conceived of the idea, designed the experiments, analyzed the data and wrote the paper.
ACKNOWLEDGEMENTS
We thank Chenhui Bian, Neal Shah and Louis Nguyen for excellent technical assistance. We thank Dr. Andrea Ventura, Memorial Sloan Kettering Cancer Centre, New York for the miR34WT and miR34TKO primary MEFs. We thank Dr. Clarke Chen & Dr. Johnny Akers for helping with the Nanoparticle Tracking Analysis.
Footnotes
Financial Support: USA National Cancer Institute Grant (R01CA043054) to JYJW
The authors declare no conflicts of interest