Bystander apoptosis in human cells mediated by irradiated blood plasma

Mutation Research 731 (2012) 107–116

Contents lists available at SciVerse ScienceDirect

Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis journal homepage: www.elsevier.com/locate/molmut Community address: www.elsevier.com/locate/mutres

Bystander apoptosis in human cells mediated by irradiated blood plasma Volodymyr Vinnikov a,∗ , David Lloyd b , Paul Finnon b a b

Grigoriev Institute for Medical Radiology of the National Academy of Medical Science of Ukraine, Ukraine Centre for Radiation, Chemical and Environmental Hazards of the Health Protection Agency of the United Kingdom, United Kingdom

a r t i c l e

i n f o

Article history: Received 17 March 2011 Received in revised form 11 October 2011 Accepted 13 December 2011 Available online 4 January 2012 Keywords: Bystander effect Apoptosis Human blood plasma Clastogenic factors Lymphocytes Ionizing radiation

a b s t r a c t Following exposure to high doses of ionizing radiation, due to an accident or during radiotherapy, bystander signalling poses a potential hazard to unirradiated cells and tissues. This process can be mediated by factors circulating in blood plasma. Thus, we assessed the ability of plasma taken from in vitro irradiated human blood to produce a direct cytotoxic effect, by inducing apoptosis in primary human peripheral blood mononuclear cells (PBM), which mainly comprised G0 -stage lymphocytes. Plasma was collected from healthy donors’ blood irradiated in vitro to 0–40 Gy acute ␥-rays. Reporter PBM were separated from unirradiated blood with Histopaque and held in medium with the test plasma for 24 h at 37 ◦ C. Additionally, plasma from in vitro irradiated and unirradiated blood was tested against PBM collected from blood given 4 Gy. Apoptosis in reporter PBM was measured by the Annexin V test using flow cytometry. Plasma collected from unirradiated and irradiated blood did not produce any apoptotic response above the control level in unirradiated reporter PBM. Surprisingly, plasma from irradiated blood caused a dosedependent reduction of apoptosis in irradiated reporter PBM. The yields of radiation-induced cell death in irradiated reporter PBM (after subtracting the respective values in unirradiated reporter PBM) were 22.2 ± 1.8% in plasma-free cultures, 21.6 ± 1.1% in cultures treated with plasma from unirradiated blood, 20.2 ± 1.4% in cultures with plasma from blood given 2–4 Gy and 16.7 ± 3.2% in cultures with plasma from blood given 6–10 Gy. These results suggested that irradiated blood plasma did not cause a radiation-induced bystander cellkilling effect. Instead, a reduction of apoptosis in irradiated reporter cells cultured with irradiated blood plasma has implications concerning oncogenic risk from mutated cells surviving after high dose in vivo irradiation (e.g. radiotherapy) and requires further study. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Radiation protection must consider all possible risks to human health resulting from radiation exposure. According to a modern paradigm, a substantial proportion of induced health effects may be associated with non-targeted mechanisms of radiation action [1–9]. Amongst these, bystander effects comprise a wide class of responses in cells that have received signals produced by irradiated cells. The final outcomes of this signalling can be damaging or protective for the bystander cell population, but usually only the former (i.e. cytotoxic or clastogenic effects) are considered. As remarked by many authors, more data need to be accumulated in order to assess properly the clinical relevance of this radiobiological phenomenon. Research in this area has been mainly

∗ Corresponding author at: Grigoriev Institute for Medical Radiology of the National Academy of Medical Science of Ukraine, Pushkinskaya St., 82 Kharkiv, 61024, Ukraine. Tel.: +380 57 7041072; fax: +380 57 7040202. E-mail address: [email protected] (V. Vinnikov). 0027-5107/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.mrfmmm.2011.12.006

focused on effects produced by low dose irradiation with few reports having considered the possible production of a bystander signal, or a response to it, by human cells after high doses (e.g. [10–12]), such as those used in radiotherapy or sometimes sustained in accidents. The development of acute radiation damage in tissues exposed to high doses depends essentially on the rate of cell death. One of its forms, apoptosis, is a programmed cell death that plays a dual role, depleting cellular pools and simultaneously rescuing tissues from undesirable proliferation of mutant, potentially malignant, clones. A large body of data has been accumulated indicating that apoptosis can be induced in bystander cells [10,13–17]. However, opposite conclusions have also been drawn [18,19]. Likely reasons for the discrepancies between datasets are that the intensity of bystander signal production and the magnitude of reporter cell response both depend significantly on cell type, genotype and experimental conditions [4,20–22]. Thus, any attempt to improve radiation-risk estimates by including untargeted mechanisms must be based on data obtained with test-systems that model in vivo situations as closely as possible.

108

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116

In the case of in vivo irradiation, due to an accidental partial body overexposure or during radiotherapy, bystander signalling could occur via lymph and blood plasma. The presence of radiationinduced clastogenic factors in blood plasma causing chromosomal damage in bystander lymphocytes had been demonstrated in several experiments [23–28] and similarly on clonogenic survival of human keratinocytes [11,17]. Less is known about the ability of plasma from irradiated blood to produce a direct cytotoxic effect, by inducing apoptosis in primary G0 -stage lymphocytes. If such effect exists, it might be possible to develop a clinically applicable system for testing human blood plasma, e.g. from radiotherapy patients, also with in vitro irradiation of blood taken before treatment starts. However, to achieve this, several fundamental issues must be examined, e.g. a dose response and interdonor variability in producing/responding to plasma-mediated cytotoxic signal. Therefore these questions were specifically addressed in the present work, where we report measurements of apoptosis in peripheral blood leukocytes caused by plasma collected from blood irradiated in vitro to high doses of ␥-rays. 2. Materials and methods 2.1. Blood donors Peripheral blood was taken with informed consent and according to the local institutional ethics procedures from a healthy male (donor I, age 33 years, exsmoker) and 2 healthy females (donors II and III, ages 44 and 53 years, respectively, both non-smokers). Blood was collected into VacutainerTM tubes containing lithium heparin anticoagulant. Blood sampling, irradiation and further plasma separation were done on several occasions, due to a large volume of blood needing to be drawn each time. 2.2. In vitro irradiation Vacutainers of donors’ blood were exposed to 60 Co doses of 2, 4, 6, 10, 20 or 40 Gy. The dose rate was 0.5 Gy min−1 for 2, 4 and 6 Gy, and 1.2 Gy min−1 for 10, 20 and 40 Gy. During the time between sampling and plasma separation (including irradiation) the blood was kept at 37 ◦ C. For each experimental dose point, blood was taken on separate occasions, and each time the irradiated sample was accompanied by a zero dose control sample that was sham treated and transported identically with the matched irradiated sample. 2.3. Plasma separation The separation of plasma was performed 2 h after irradiation. To separate plasma, donors’ whole blood was centrifuged for 15 min at 170 × g; the supernatant was collected by syringe, avoiding disturbance of the cell pellet, transferred into Eppendorf tubes, immediately frozen and stored at −20 ◦ C until use. This permitted a number of plasma samples to be tested later in one round, thus avoiding day-today variations in reporter PBM cultures. Regarding the methodological correctness of such an approach, there are a number of reports (e.g. [24–28]) stating that freezing of plasma does not affect its clastogenic properties. In contrast to some published studies on testing human blood plasma for clastogenicity (e.g. [24–27]), plasma was not ultrafiltered thereby preserving all clastogenic factors, including those with high molecular weight. Plasma collected in this manner contained no cells, which might contaminate reporter cell cultures. This was established with clastogenesis cytogenetic studies (to be reported elsewhere) done in parallel with the same blood samples. Reporter cells’ sex chromosome complements were checked when cultured with plasma from an opposite gender donor. 2.4. Cell culture In all experiments the reporter cells were donors’ peripheral blood mononuclear cells (PBM), which were separated from whole blood using Histopaque 1077 (Sigma, Poole, UK) according to the manufacturer’s standard instructions and washed in Hank’s balanced salt solution (Sigma, Poole, UK). After the cell isolation procedure lymphocytes comprised 60–70% of separated PBM, including about 30–45% of CD3-positive T-lymphocytes, that was estimated by staining cells with antibodies to CD3+ (Becton Dickinson, Oxford, UK) and measuring with a flow cytometer (Becton Dickinson FACS CaliburTM ). For each data point three identical cultures were set up, comprising 2.0 × 106 cells placed into 4 ml of Eagle’s MEM supplemented with 20% heat inactivated foetal calf serum (FCS), l-glutamine, 100 IU/ml penicillin and 100 ␮g/ml streptomycin (all reagents from Invitrogen, Paisley, UK). To these cultures 0.2 ml of plasma assigned

for testing was added (in two experimental series 0.4 ml was used to check for a concentration effect). Before adding the plasma it was thawed at room temperature for 20 min and centrifuged at 11,000 × g for 1 min for additional sedimentation of any possible cellular material. A negative control, with no plasma added was also included in each experimental round. Cultures were coded and incubated at 37 ◦ C with 5% CO2 in air for 24 h. 2.5. Apoptosis measurements by the Annexin V assay After incubation the medium was removed from cultures; cells were washed three times in cold (4 ◦ C) phosphate buffered saline (PBS) pH 7.4 (Sigma, Poole, UK) and stained with a FITC-Annexin V kit (Becton Dickinson, Oxford, UK) according to the manufacturer’s instructions. The cells were then run on a flow cytometer (Becton Dickinson FACS CaliburTM , 488 nm argon-ion laser) equipped with Cell QuestTM software for data acquisition and analysis. Annexin V positive apoptotic cells were identified using FL1 detector histogram plots. Non-apoptotic, early apoptotic and late apoptotic/necrotic cell populations were distinguished by simultaneous staining with propidium iodide (PI) and using FL1/FL3 quadrant dot plots. Fixed gating parameters were used for all samples analysed in order to prevent inter- or intra-individual variation related to flowcytometric parameters within or between experiments. To check for proper Annexin V staining, a positive control was included in each experimental round. These were cultures of PBM separated from blood after exposure to 2 or 4 Gy ␥-rays. A good reproducibility of staining and measuring procedures was judged from very low inter-experiment variations in apoptotic cell outcome in these cultures. Two repeated cell staining/FACS analyses per measurement were performed in each of three replicate cell cultures, which were set up for each data point. 2.6. Overall study design The main aim of these experiments was to assess the potential cytotoxic effect caused by plasma from irradiated versus unirradiated blood in reporter PBM by assaying for apoptosis using the Annexin V method. Three hypotheses were tested: (1) The plasma from irradiated blood is able to cause apoptosis in the unirradiated reporter PBM, and this effect has a quantitative dependence on radiation dose. (2) The interdonor variability may exist in producing plasma-mediated cytotoxic factors and response of bystander leucocytes. (3) The irradiated PBM respond to the action of plasma from irradiated blood in the same manner, as unirradiated reporter cells. Thus, the study consisted of three experiments. 2.6.1. Experiment 1: dose response in vitro Plasma from irradiated blood collected from donor II (2, 4 and 40 Gy) and donor III (6, 10 and 20 Gy) was tested for cytotoxicity against donor I PBM. Each sample of plasma from irradiated blood was matched to the sample of plasma separated from unirradiated blood, and tested concurrently. 2.6.2. Experiment 2: interdonor variability Autologous and homologous plasma samples from irradiated blood, each matched to respective plasma from unirradiated blood, were tested against the PBM of donors I, II and III. Autologous plasma set: - donor I reporter PBM + autologous plasma given 6 Gy; - donor I reporter PBM + autologous plasma given 4 Gy; double volume of plasma (0.4 ml instead of 0.2 ml) was used to check for a possible concentration effect; - donor II reporter PBM + autologous plasma given 4 Gy; - donor III reporter PBM + autologous plasma given 4 Gy. Homologous plasma set: - donor I reporter PBM + donor III plasma given 4 Gy; double volume of plasma (0.4 ml instead of 0.2 ml) was used to check for a possible concentration effect; - donor II reporter PBM + donor I plasma given 6 Gy; - donor III reporter PBM + donor I plasma given 6 Gy; - donor III reporter PBM + donor I plasma given 10 Gy. 2.6.3. Experiment 3: irradiated reporter cells PBM of donors II and III blood irradiated to 4 Gy ␥-rays were used as reporter cells for testing plasma, collected from autologous blood or donor I blood after irradiation to 0, 2, 4, 6 or 10 Gy ␥-rays. The irradiated reporter PBM were obtained from the whole blood held for 2 h post-irradiation (or sham-irradiated) as described in Section 2.2. PBM were separated from the whole blood using Histopaque and placed into culture in exactly the same way as was done with unirradiated reporter cells (Section 2.4). The plasma aliquots were then immediately added to the cultures. Two independent, repeat experiments were performed to test plasma after irradiation to 2 and 6 Gy, and three repeat experiments – with plasma after irradiation to

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116 4 and 10 Gy. Cell survival was compared between cultures with plasma from unirradiated and irradiated blood and plasma-free controls. Each series of irradiated cells was matched to the corresponding control series of unirradiated PBM to evaluate possible changes in radiation-induced apoptotic response. 2.7. Statistical analyses The results of all measurements are presented as mean values and their standard errors (SE) of apoptotic and live cell yields in at least two independent experiments, with two repeated cell staining/FACS analyses per measurement performed in each of three replicate cell cultures, which were set up for each data point. Significance of differences was determined by a Student’s paired t-test, considering the differences to be significant if p ≤ 0.05. In addition, full General Linear Model Analysis of Variance (ANOVA) was carried out, to investigate the interaction and significance of the other experimental factors – experimental repeat, individual donor variability, the addition or exclusion of plasma, and the homologous or autologous nature of the donor combinations.

109

cells (factor: donor number; p = 0.513). Importantly, for plasmafree cultures of directly irradiated PBM the ANOVA results revealed the low level of variation in the level of live cells between series of experiments (p = 0.301) as well as between donors (p = 0.158), but the radiation dose appeared to be a highly significant factor (p < 0.001). In all experimental series each sample of plasma from irradiated blood was rigorously matched to the sample of plasma from unirradiated blood, and plasma-treated cultures were tested concurrently with plasma-free controls. Thus, to minimize the influence of the intra-donor variability, statistical analysis of the data was focused on the difference between treated cultures and their control counterparts for apoptotic or live cell yields.

3. Results 3.1. Controls and their variations

3.2. Dose response for apoptosis induced by irradiated blood plasma

Plasma from the three donors was tested against reporter PBM from the same individuals in various combinations of autologous and homologous, irradiated and unirradiated cells and plasma. Negative and positive controls were included, being respectively, cells cultured without additional plasma and cells collected from blood given 2 or 4 Gy ␥-rays. Cell staining profiles obtained by running intact control cells and irradiated cells through the flow cytometer were clearly different, as the profiles in irradiated cell cultures always contained a distinctive enhanced second peak representing Annexin V (FITC)-positive, apoptotic cells. In this study the inter-donor variations of the spontaneous levels of apoptotic and necrotic cells and cellular sensitivity to radiation appeared to be rather low. Instead, some intra-donor variability occurred throughout the experiments. The background level of live cells in plasma-free cultures of PBM from one donor varied in different series (Fig. 1); the ANOVA test showed that it was a significant factor (p < 0.001). However, on the background of intra-donor variations between series, no considerable difference was observed between donors for the percentage of live cells in plasma-free cultures of unirradiated

Plasma from blood irradiated to 2–40 Gy collected from donors II and III was tested for cytotoxicity against donor I PBM (Table 1). In all six series, with both plasma donors, the reporter PBM survival in cultures with plasma from irradiated blood was similar to that with plasma from unirradiated blood. Each dose point was examined twice; therefore the entire dataset comprised 12 paired comparisons. Its overall testing showed no statistical difference between reporter PBM cultures with irradiated and unirradiated blood plasma for early apoptosis, late apoptosis, necrosis and live cell yield (respectively, t = 1.99, 0.01, 1.57 and 0.21; p > 0.05). The ANOVA results in this case are suggestive of an effect of donor of the plasma (donor number nested within addition of plasma; p = 0.044), however no other factors were identified as significant (p < 0.05). Fig. 2 shows the percentages of total apoptotic (i.e. early plus late apoptosis) and live cells in reporter PBM cultures with irradiated blood plasma compared with the average of unirradiated blood plasma and the plasma-free control. There is clearly no dose response evident.

94

Live cells in culture, %

92 90

mean control value

88 86 84 82 80 78 76 I

II II III

I

I II III

I III III III

I

I II III

I

I III III

Donor of PBM

Round 1

Round 2

Round 3

Round 4

Round 5

Fig. 1. Variability of the background yield of live cells in plasma-free control PBM cultures of three donors (I–III). Data are grouped according to rounds of blood sampling; a round spanned for 2–3 weeks; with 1–2-week interval between rounds. Double dashed line represents the mean value for 20 cultures set up throughout the study.

110

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116

Table 1 Effects of in vitro irradiated and unirradiated blood plasma on cell survival in cultures of unirradiated reporter PBM of donor I. Donor of blood plasma

Percentage of cells in PBM culturesa

Radiation dose to blood plasma

Early apoptosis (FITC+/PI−) II II III III III II

0 Gy 2 Gy 0 Gy 4 Gy 0 Gy 6 Gy 0 Gy 10 Gy 0 Gy 20 Gy 0 Gy 40 Gy

7.51 8.39 8.96 10.88 4.08 4.69 5.03 4.01 7.80 9.29 5.29 5.62 5.84

Plasma-free controlc

0.15 0.14 0.04 0.44 0.31 0.21 0.90 0.39 0.27 0.26 1.10 0.34 0.31

4.92 7.52 6.61 6.49 2.78 2.10 2.80 1.60 6.59 5.49 2.12 2.59 4.63

± ± ± ± ± ± ± ± ± ± ± ± ±

Necrosis (FITC−/PI+)

1.18 1.22 0.61 1.01 1.13 0.76 1.36 0.10 0.81 0.75 0.16 0.04 0.88

0.62 1.29 0.75 0.42 3.03 0.77 0.90 0.69 0.77 0.35 3.47 0.82 2.33

± ± ± ± ± ± ± ± ± ± ± ± ±

0.13 0.12 0.45 0.07 2.91 0.12 0.33 0.17 0.23 0.09 2.24 0.67 0.87

Live 86.95 82.81 83.69 82.22 90.12 92.44 91.28 93.71 84.85 84.88 89.12 90.98 87.20

± ± ± ± ± ± ± ± ± ± ± ± ±

1.20 0.96 1.01 1.38 3.72 0.85 1.92 0.11 1.31 1.10 1.30 0.97 0.76

1.55 −2.60 −1.72 −3.19 1.12 3.44 2.28 4.71 −0.56 −0.53 0.12 1.98

± ± ± ± ± ± ± ± ± ± ± ±

0.26 0.02 0.07 0.44 5.79 2.91 3.99 2.17 0.37 0.16 0.76 1.09

Mean value for two independent experiments. Compared to relevant plasma-free control cultures, which were set up specifically in the same experimental series. Mean plasma-free control combined from all series in this experiment.

3.3. Interdonor variability in plasma-induced cytotoxicity and reporter PBM response

In the combined dataset adding plasma per se produced a little effect on cell survival (ANOVA p = 0.245). The response of reporter cells appeared to be dependent on the autologous or homologous status of the plasma being tested, although ANOVA shows only a suggestion for significance of this factor (p = 0.070). Adding autologous plasma from unirradiated blood slightly reduced the surviving fraction of reporter PBM compared to plasma-free controls (t = 3.09; p < 0.01 for 8 paired comparisons). By contrast, the presence of homologous unirradiated blood plasma and plasma collected from irradiated blood (either autologous or homologous) had no cytotoxic effect on the reporter PBM. A comparison between matched cultures with plasma from unirradiated and irradiated blood showed no difference in the survival of reporter PBM tested against homologous blood plasma. With autologous plasma from irradiated blood a reduction in late apoptosis yield (t = 3.08; p < 0.01) and an elevation in the percentage of live cells (t = 3.66; p < 0.01) was observed compared to the respective means in cultures containing autologous plasma from unirradiated blood. In the combined dataset ANOVA results showed that radiation dose given to blood, from which plasma was separated, appeared to be a significant factor (p = 0.038). In contrast to a minor influence of blood plasma on bystander cell survival, the direct irradiation of PBM in blood produced, as expected, a significant elevation of apoptosis yield. Clear changes in

100 90 80 70 60 50 40 30 20 10 0

y G 40

20

G

y

y

G y 6

G y 4

G y 2

0

co nt ro fr ee

R di i d Radiation dose to bl blood d plasma, l G Gy

Pl a

sm a-

G y

Apoptotic Live

l

Percen ntage of cellss

A possible inter-donor variability in plasma-induced cytotoxicity was examined by testing 4 autologous and 4 homologous plasma from irradiated blood samples, each matched to respective plasma sample from unirradiated blood. Cell survival in these reporter PBM cultures was also compared to the background values in parallel plasma-free cultures of PBM from the same donor (Table 2). The absence of a cytotoxic effect of irradiated blood plasma was confirmed with various homologous and autologous combinations of plasma and reporter PBM from different donors. Doubling the volume of plasma likewise had no effect. There was a slight suggestion of a significant effect of donor of the reporter PBM (ANOVA p = 0.041), but that was not so for the donor of plasma (ANOVA p = 0.320), and also a low variation occurred between different series of the experiment (ANOVA p = 0.142). Thus the overall intra- and inter-donor variations in reporter PBM response were concluded unsystematic and insignificant. Data from Experiments 1 and 2 were combined to assess a mean difference for cell survival between plasma-treated and plasmafree cultures. These averaged data were also compared with the effect caused by direct ␥-irradiation (Fig. 3).

G

c

Late apoptosis (FITC+/PI+)

10

a b

± ± ± ± ± ± ± ± ± ± ± ± ±

Difference with plasma-free cultures for the % of live cellsb

Fig. 2. The yield of apoptotic and live cells in donor I reporter PBM cultures containing plasma of homologous blood irradiated in vitro to various doses of ␥-rays. Error bars indicate standard error of the mean (SE) for two independent experiments. The effect of matched unirradiated plasma samples (0 Gy) and plasma free control was averaged from six series.

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116

111

Table 2 An interdonor variability of the effects produced in unirradiated reporter PBM by in vitro irradiated and unirradiated blood plasma. Donor of reporter PBM

Donor of blood plasma

Dose to blood plasma

Percentage of cells in PBM culturesa

Early apoptosis (FITC+/PI−) Autologous plasma Donor I Donor I Donor II Donor III Homologous plasma Donor I Donor II Donor III Donor III Plasma-free controld a b c d

Late apoptosis (FITC+/PI+)

Difference with plasma-free cultures for the % of live cellsb Necrosis (FITC−/PI+)

Live

Donor Ic Donor I Donor II Donor III

0 Gy 4 Gy 0 Gy 6 Gy 0 Gy 4 Gy 0 Gy 4 Gy

7.87 4.72 5.34 4.41 6.36 8.25 7.19 3.50

± ± ± ± ± ± ± ±

0.62 0.22 0.40 0.55 0.56 0.07 0.41 1.10

3.42 2.99 9.50 2.42 8.61 7.37 8.78 4.64

± ± ± ± ± ± ± ±

0.09 1.12 0.59 0.63 0.64 1.08 1.34 1.62

1.03 0.52 0.44 0.24 0.72 0.29 0.68 0.71

± ± ± ± ± ± ± ±

0.32 0.31 0.01 0.17 0.02 0.14 0.68 0.70

87.69 91.78 84.73 92.94 84.33 84.10 83.37 91.16

± ± ± ± ± ± ± ±

0.39 1.03 1.00 0.25 0.06 1.29 0.99 1.22

−3.26 0.83 −6.23 1.99 −4.32 −4.55 −0.21 7.59

± ± ± ± ± ± ± ±

0.01 1.43 0.60 0.65 3.90 2.67 1.24 1.01

Donor IIIc Donor I Donor I Donor I

0 Gy 4 Gy 0 Gy 6 Gy 0 Gy 6 Gy 0 Gy 10 Gy

8.73 9.38 9.08 6.14 6.91 7.04 6.56 3.47 6.40

± ± ± ± ± ± ± ± ±

0.11 0.56 0.79 0.57 0.12 0.36 0.49 0.41 0.41

4.85 4.05 6.95 7.94 5.50 6.70 1.99 2.84 5.63

± ± ± ± ± ± ± ± ±

1.39 0.44 1.09 0.52 0.74 1.83 2.94 0.50 0.86

0.38 0.19 0.21 0.45 0.29 1.06 0.71 0.05 1.37

± ± ± ± ± ± ± ± ±

0.24 0.04 0.04 0.11 0.11 0.58 2.12 0.04 0.46

86.05 86.39 83.77 85.47 87.31 85.21 90.75 93.66 86.60

± ± ± ± ± ± ± ± ±

1.73 0.47 1.92 0.16 0.72 0.89 0.33 0.94 1.20

0.64 0.98 6.83 8.53 3.74 1.64 0.42 3.32

± ± ± ± ± ± ± ±

0.79 0.09 0.54 1.54 1.50 1.33 0.62 0.01

Mean value and standard errors (SE) for two independent experiments. Compared to relevant plasma-free control cultures of the same donor’s reporter cells, which were set up specifically in the same experimental series. Double volume of plasma (400 ␮l) was added to cultures. Mean plasma-free control combined from all series in this experiment.

cell staining profiles, comprising a noticeable reduction of the proportion of surviving PBM, were observed after direct ␥-irradiation at both dose points for every donor. The mean total apoptotic cell yield was increased above the matched background level by (9.01 ± 2.02)% at 2 Gy and (18.13 ± 3.42)% at 4 Gy (respectively, t = 4.45 and 5.30; p < 0.01 for 6 paired comparisons), so showing a positive dose response at these doses also confirmed by ANOVA (p < 001). Thus, the magnitude of cell-killing effect directly induced by radiation exceeded any fluctuations of survival caused by plasma from either irradiated or unirradiated blood in reporter unirradiated PBM.

3.4. Reaction of irradiated reporter PBM to plasma from irradiated blood In order to address to what extent might plasma from irradiated blood modify an apoptotic response induced directly by radiation, PBM of donors II and III blood given 4 Gy ␥-rays were treated with plasma, collected from autologous blood or donor I blood given 0, 2, 4, 6 or 10 Gy ␥-rays. Cell survival was compared between cultures with plasma from unirradiated and irradiated blood and plasmafree controls. Each series of irradiated cells was matched to the corresponding control series unirradiated PBM to evaluate possible changes in radiation-induced apoptosis response.

Fig. 3. Proportions of live and dead cells amongst donor’s PBM, cultured with unirradiated and irradiated blood plasma compared with plasma-free unirradiated control and plasma-free cultures of cells after direct irradiation in vitro to various doses of ␥-rays. Data represent mean values for all repeating series in Experiments 1 and 2. *Statistically significant difference in the surviving fraction of reporter PBM between cultures with autologous unirradiated plasma and plasma-free controls (p < 0.01). **Statistically significant elevation in the percentage of live cells due to accordant reduction in late apoptosis yield in cultures with autologous irradiated plasma compared to that in cultures with autologous unirradiated plasma (p < 0.01). ***Statistically significant reduction of the proportion of live PBM after direct ␥-irradiation to 2 and 4 Gy compared to plasma-free unirradiated control (p < 0.01).

112

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116

Table 3 The apoptotic response of irradiated (4 Gy ␥-rays) and unirradiated PBM cultured with plasma collected from blood after ␥-irradiation to various doses. Radiation dose to blood plasma

Radiation dose to reporter PBM

Percentage of cells in PBM cultures (mean ± SEc )% Early apoptosis (FITC+/PI−)

Plasma-free control 0 Gy 2–4 Gy 6–10 Gy a b c

0 Gy 4 Gy 0 Gy 4 Gy 0 Gy 4 Gy 0 Gy 4 Gy

7.88 18.40 7.89 15.24 7.52 14.63 4.76 12.23

± ± ± ± ± ± ± ±

0.56 1.48 0.56 1.11 0.32 1.18 0.64b 1.06b

Late apoptosis (FITC+/PI+) 7.38 18.85 7.88 22.08 6.58 19.82 4.91 13.95

± ± ± ± ± ± ± ±

0.60 2.45 0.68 0.76 0.82 0.78 1.26 3.87

Necrosis (FITC−/PI+) 1.21 1.38 0.50 0.53 0.57 0.45 0.20 0.40

± ± ± ± ± ± ± ±

0.78 0.90 0.12 0.14 0.27 0.12 0.11 0.14

All apoptotic cells (FITC+) 15.25 37.25 15.77 37.32 14.10 34.45 9.67 26.18

± ± ± ± ± ± ± ±

0.57 1.81 0.70 0.70 0.93 1.34 1.84a 4.47a

Live 83.54 61.36 83.73 62.15 85.33 65.10 90.12 73.42

± ± ± ± ± ± ± ±

0.89 1.62 0.71 0.68 0.88 1.25 1.94a 3.59a

Statistically significant difference between cultures with irradiated plasma versus unirradiated plasma and/or plasma-free control: (p < 0.05). Statistically significant difference between cultures with irradiated plasma versus unirradiated plasma and/or plasma-free control: (p < 0.01). Averaged values and standard errors (SE) of the mean were calculated for 5 independent measurements in each experimental set.

In the overall dataset the ANOVA results showed an insignificant impact of the homologous or autologous status of plasma (p = 0.183) and a radiation dose to blood, from which plasma was separated (p = 0.186). Also the variations between different series were low (ANOVA p = 0.085). As there was no discernable difference between the effects produced by plasma after exposure within moderate dose range (2 or 4 Gy) and within high dose range (6 or 10 Gy), as well as between autologous or homologous plasma, the data were pooled for overall comparison in Table 3. Generally, the presence of plasma in the culture improved the survival of both unirradiated and irradiated PBM compared to plasma-free controls, and according to ANOVA that was approaching significance (p = 0.052). With plasma collected after radiation exposure this effect was more pronounced, and a positive dose response can be concluded from the mean values of live cell fraction. Adding the plasma irradiated to 6–10 Gy to unirradiated reporter PBM led to a significant increase in the percentage of live cells compared to plasma-free controls and cultures with plasma from unirradiated blood (respectively, t = 3.08; p = 0.015 and t = 3.09; p = 0.015 with 8 degrees of freedom). Also, the reduction of early apoptosis amongst unirradiated PBM was more distinct when adding plasma given 6–10 Gy than 2–4 Gy (t = 3.86; p < 0.01). For the increase of live cell percentages the difference between 2–4 Gy and 6–10 Gy datasets was on the borderline of significance (t = 2.25; p = 0.055). When added to irradiated reporter PBM, blood plasma collected after exposure to 2–4 Gy caused a borderline significant increase in live cell yield compared to that in cultures with plasma from unirradiated blood (t = 2.07; p = 0.072). Plasma irradiated to 6–10 Gy produced a significant increase in live cell percentage in cultures of irradiated reporter PBM compared to the effects of unirradiated blood plasma (t = 3.08; p = 0.015). Adding the irradiated blood plasma reduced the direct radiation-induced cell-killing effect in reporter PBM (Fig. 4). The maximum difference in the proportion of dead cells between irradiated and matched unirradiated reporter PBM occurred in the plasma-free control, and this declined progressively with the increase in radiation dose to the blood from which the test plasma was extracted. When all data for plasma from irradiated blood were combined, the total yields of dead cells in cultures set up from unirradiated and irradiated PBM, respectively, were (12.3 ± 1.3)% and (30.7 ± 3.1)%. The resultant difference in cell survival between irradiated and unirradiated reporter PBM was (18.5 ± 2.1)%, and that was 1.2 times lower, then in plasma-free controls or cultures with unirradiated blood plasma.

4. Discussion 4.1. Lack of direct cytotoxic bystander effect in unirradiated reporter cells treated with irradiated blood plasma Apoptosis in lymphocytes and lymphoblastoid cells is a wellknown consequence of DNA damage after direct irradiation. DNA lesions, later processed into chromosomal aberrations, also can be caused through a bystander mechanism by clastogenic factors released from irradiated cells into blood plasma [23–28]. In several studies employing other cell types, such soluble clastogenic factors have been shown also to cause apoptosis in unirradiated reporter cells [10,13–16]. The mechanism of this untargeted cytotoxic radiation effect includes a cascade of events in reporter cells, such as mobilisation of intracellular calcium, loss of mitochondrial membrane potential, mitochondrial dysfunction and other mitochondria-dependent signalling pathways, leading to an overall increase in reactive oxygen species [10,13,14,17,22,29,30], however the latter does not emerge as a primary, mechanistic cause of bystander apoptosis [15,16]. Recently Belloni et al. [10] presented the results of a conditioned medium transfer experiment, amongst which there were a statistically significant decrement in cell viability, concomitant with the loss of mitochondrial membrane potential, increases in hydrogen peroxide and superoxide anion with depletion of intracellular glutathione, which were observed in bystander lymphocytes incubated for 48 h in medium collected from 24 h cultures of lymphocytes given 0.5 and 3 Gy of X-rays. There are also data showing that an injection of apoptotic lymphocytes exacerbated the degree of apoptotic death in neighbouring cells in mice [31]. Blood plasma collected from radiotherapy patients had a varying degree of toxicity towards human keratinocytes assayed by a clonogenic survival test [11]. Plasma of in vitro irradiated blood from several donors showed a large heterogeneity in responses in pre-irradiated keratinocytes assayed by a clonogenic survival test, and in some cases a triggering of apoptosis was noted [17]. Combining these observations it was tempting to suggest that DNA damaging, clastogenic plasma factors of irradiated blood might induce apoptosis in unirradiated reporter leucocytes. The present study was designed to clarify this after in vitro irradiation. If plasma-medicated cytotoxicity had been detected, then a clinically applicable test might be developed for evaluation of this potential source of risk, e.g. in radiotherapy patients. In the present work some intra-individual (inter-experimental) variability in the spontaneous yield of apoptosis was observed. The background level of apoptosis in plasma-free PBM cultures across the series of experiments varied from 7 to 19%. Such a pronounced inter-experimental variations in the level of apoptosis in cells from

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116

113

45

Proportion of dead cells, %

40 35 30 Δ=22.2±1.8%

Δ=21.6±1.1%

Δ=20.2±1.4%

Δ=16.7±3.2%

25 20 15 10 5 0 PBM 0 Gy

PBM 4 Gy

plasma-free

PBM 0 Gy

PBM 4 Gy

plasma 0 Gy

PBM 0 Gy

PBM 4 Gy

plasma 2-4 Gy

PBM 0 Gy

PBM 4 Gy

plasma 6-10 Gy

Fig. 4. The percentage of dead cells amongst unirradiated and irradiated (4 Gy ␥-rays) reporter PBM, cultured with plasma collected from unirradiated or irradiated blood (2–10 Gy ␥-rays) in compare with plasma-free control. Data on each point represents mean values and their standard errors (SE) for five independent measurements. The difference for the percentage of dead cells between irradiated and unirradiated reporter PBM is presented as (±SE)%.

the same donor has been reported previously [32,33]. A possible cause could be fluctuations of the ratio of T- and B-lymphocytes, and within T-cells – that of CD4 and CD8 subpopulations, which are known to have different susceptibility to spontaneous in vitro apoptosis [32,34]. In order to overcome this, in the present work every PBM culture treated with irradiated blood plasma was rigorously matched to cultures with plasma from unirradiated blood and plasma-free control, and all were tested concurrently. Analysis of the data was focused on the difference between treated cultures and their control counterparts for apoptotic and live cell yields. The overall experimental design of the present work allowed covering a wide range of radiation doses, and blood of every donor was assigned to both moderate (2–6 Gy) and high doses (10–40 Gy). Also doses of 4 and 6 Gy were given to blood of every donor to serve as ‘standards’ for checking possible combinations of autologous and homologous blood plasma. The latter was necessary for potential development of a clinical test-system, in which the interdonor variability and choosing proper ‘standard’ donors of reporter cells and/or plasma would become a crucial point. Another essential variable in such a test-system would be the ratio of reporter cell number to the volume of tested plasma. In human whole blood that is about 10–16 × 106 leucocytes to 1 ml of plasma. In the present experiments that was 2.0 × 106 PBM to 0.2–0.4 ml of plasma, which is of the same order of magnitude. The dilution of tested plasma in culture medium (to 5–10%) would represent the partial body irradiation scenario, such as occurs in radiotherapy. Plasma collected from unirradiated blood or blood irradiated in vitro to 2–40 Gy ␥-rays did not induce early apoptosis or late apoptosis/necrosis in reporter PBM, whereas, as expected, direct irradiation caused significant and dose-dependent apoptotic death. Compared to numerous reports about the presence of distinct cytotoxic or clastogenic bystander effects, including those produced by irradiated blood plasma (summarized in [28]), such a result was somewhat surprising. However, it has been reported that cytotoxic bystander effects were induced by low doses of radiation, and the effect appeared to plateau at higher doses [20,35–37]. Thus in the present work a minimum in vitro dose of 2 Gy ␥-rays may have caused an effective kill of PBM but appeared to be too high to induce excretion of a bystander signal into plasma. Another possible reason for the absence of the cytotoxic effect in our study may be the choice of culture medium. Eagle’s minimum essential medium in Earle’s salts was used whereas TC-199 or RPMI-1640, which are poor in free radical scavengers, were used by others in plasma-born clastogenic factor studies [10,24–28]. The absence of cytotoxicity or clastogenic effects in unirradiated bystander cells was observed in other studies, employing different

end-points [38–40]. In particular, an adoptive transfer of irradiated splenic lymphocytes did not change the apoptotic response of the neighbouring bystander spleen cells in mice [18]. Also there are several examples, where cytogenetic testing failed to demonstrate clastogeneity of human blood plasma after either in vitro or in vivo irradiation [41–43]. In the majority of published reports on the bystander effect in various test-systems it was noted that the magnitude of damage in the unirradiated reporter cells was comparable to that in the irradiated, bystander signal-producing cells [10,14–16]. That was not the case in the present experiments, where the magnitude of cell-killing effect directly induced by radiation exceeded any fluctuations of survival caused by plasma from irradiated or unirradiated blood in unirradiated reporter PBM. Indeed the positive dose response for induced apoptosis in directly irradiated cells in the present study is in a good quantitative agreement with results published by others [32,33,44]. There is growing evidence that the magnitude of bystander effect is very much dependent on the cell type of both irradiated and reporter cells, and experimental conditions play a significant role [20–22,45,46]. It was concluded that a bystander signal produced in a multicellular environment induces complex changes in the treated culture, and that these changes are reflective of a coordinated response to maintain integrity throughout the tissue [4,37,45]. In the work of Seymour and Mothersill [11], radiotherapy patients’ plasma showed very large individual variation in its effect on clonogenic survival of human keratinocytes. With plasma collected before radiotherapy, in some cases a decrease, and in others a promotion of proliferation was found in the reporter cells. Moreover in almost all cases sampled after radiotherapy the plasma produced an opposite effect to that observed before treatment. The dissimilarity of these data with our finding of no effect from healthy donors’ plasma either before or after in vitro irradiation could be attributable to the different measured end-points, clonogenic survival versus apoptosis yield, and particularly the types of reporter cells, actively proliferating keratinocytes versus quiescent PBM, mainly G0 lymphocytes. This is especially so in light of the evidence that bystander effects may predominantly occur in replicating cells [47]. 4.2. Irradiated blood plasma contains a “don’t die” signal for irradiated reporter cells Whilst there was a little effect of irradiated blood plasma on unirradiated PBM, clear changes were found in survival of reporter

114

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116

cells, which had been irradiated before testing. Rather than exacerbating apoptosis, irradiated blood plasma appeared to encourage survival of such reporter cells. However in unirradiated PBM, which have a high survival rate per se, this was barely detectable because adding the plasma just increased the percentage of live cells to the maximum possible level, and this increment was on the borderline of resolution of the technique. But for irradiated reporter PBM, which are destined to exhibit a considerable amount of radiationinduced apoptosis, a “don’t die” signal mediated by plasma from irradiated blood produced an appreciably enhanced survival. Use of irradiated cells as reporters in the bystander experiment is an example of “thinking outside the box”, which was proclaimed as the principle of non-targeted effects investigations [48]. To the best of our knowledge, the present study is the first where irradiated quiescent, non-proliferating PBM were used as reporters for measuring the apoptosis rate changes caused by irradiated blood plasma. Previously there have been only three reports of modulating radiation-induced effects in irradiated cells co-cultured with untargeted cells within a bystander experiment [49–51], one specific comment on the survival of cells having received a high radiation when their nearby cells received a low radiation dose [52], and one report about effects caused in irradiated cells (not lymphocytes) by irradiated blood plasma [17]. The first group found a decrease in chromosomal aberration frequencies in irradiated human lymphocytes, which were mixed in culture with unirradiated cells, compared to that in irradiated cells cultured alone [49]. The second group observed a reduction of cytogenetic damage and apoptosis levels in irradiated cells in the presence of unirradiated cells in the culture, and concluded some kind of protective signalling from the latter towards the former [50]. The third group described similar “rescue effects”, when unirradiated bystander cells assisted irradiated cells (human primary fibroblasts or cancer cell line) through intercellular, mediamediated signal feedback [51]. That included a decrease in the numbers of 53BP1 foci, lower yield of micronuclei and lesser extent of apoptosis (measured by Annexin V-FITC method) in the irradiated cells co-cultured with the bystander cells, compared with the respective values observed in irradiated cells cultured alone. In an experimental system comprising malignant melanoma cells exposed to spatially modulated therapeutic megavoltage photon beams, Mackonis et al. [52] distinguished three ways in which cell survival was influenced by the fate of neighbouring cells. The first of these was the “classical” cytotoxic bystander effect, where cell survival was reduced when communicating with irradiated cells. The second type of effect was an increase in cell survival when nearby cells received a lethal dose. The third was an increase in the survival of cells receiving a high dose of radiation, when nearby cells receive a low dose. Obviously, the last two types of reporter cells’ reactions can be considered as relevant to the “don’t die” signal effect found in the present research. A study most similar to present Experiment 3 (see Section 3.4) was reported by Acheva et al. [17], who tested plasma of healthy donors’ blood given 0.5 Gy ␥-rays in vitro against a keratinocyte cell line pre-irradiated to 0.05 Gy ␥-rays. They observed a large individual variability in the plasma action on the low dose irradiated reporter cells; amongst 9 individuals, whose plasma was tested, there were two cases of cytotoxicity and one case of stimulatory effect produced by plasma from both unirradiated and irradiated blood, and also two cases of cell growth stimulation, caused particularly by plasma from irradiated blood. Stimulatory or protective bystander effects are phenomena known for a long time. Lloyd and Moquet [23] reported a statistically significant trend whereby unirradiated human lymphocytes cultured under mitogenic stimulation in medium containing irradiated blood plasma progressed more swiftly through the first cell

cycle, by comparison with the cycling rate in cultures with plasma given zero dose. Similarly to the present Experiment 3, in that earlier study the extent of the effect of speeding up of the cell cycling appeared to be dependent on the dose given to plasma. A similar response, pointing to a possible mechanism for a “nourishing” effect of irradiated blood plasma, can be seen in a work of Iyer and Lehnert [53], who reported a stimulating action of conditioned medium on bystander cell survival, that was particularly associated with p53 (and relevant apoptosis) pathways. Also, Kishikawa et al. [54] showed that human colon adenocarcinoma cells, dying from lethal doses of incorporated 123 IdUrd, significantly augmented the growth of neighbouring, unlabeled tumour cells in vivo (injected in nude mice) and in vitro (co-culturing). This stimulatory bystander effect induced by an Auger electron emitter was repeated in medium transfer experiment, i.e. the proliferation of reporter cells was much enhanced in the presence of supernatants from 123 IdUrd-labeled cell cultures. Seymour and Mothersill [11] reported that 8 out of 19 cancer patients’ plasma samples collected before radiotherapy caused enhanced clonogenic survival of bystander keratinocyte cultures. Later, plasma of 5 out of 9 individuals sampled midway during radiotherapy and 6 weeks after completing the course markedly promoted proliferation of reporters. More recently the same group observed a stimulation of colony-forming activity in several unirradiated reporter cell lines after treatment with high dilutions of irradiated cell conditioned medium [20]. The response was likened to a hormetic effect, whereby a test-system undergoes stimulation by low doses of a substance that is toxic at high doses. The present study finding of an anti-apoptotic effect could be construed as hormetic, but this would need to be compared against much higher concentrations of the bystander plasma-borne factor(s) in a purified state, obtained by centrifugation with cut-off filters. However, the appropriateness of treating the reporter cells with unphysiologically high concentrations of these substances is questionable. Some data on possible effectors acting during plasma testing for clastogeneity can be found in recently published reviews [27,28]. Abrogation of radiation-induced apoptosis might provide an evolutionary advantage, as it may result in an increase in the radioresistance and overall cell population survival after future irradiation, should it occur. It can be a potential mechanism for adaptive response, which is known to exist in bystander testsystems [48]. It has been shown that soluble factors released from irradiated murine lymphocytes initiate a signalling cascade in unirradiated lymphocytes resulting in enhanced proliferation and radioresistance, and thus an adaptive response to radiation [55]. Similarly, radioresistance induced via a bystander mechanism, was also observed in other cell types [56,57]. Regarding possible molecular machinery involved, it was shown that the anti-apoptotic Bcl-2 protein has the remarkable ability to prevent cell death from several noxious stimuli. Bcl-2 overexpression in one cell type has been reported to protect against cell death in neighbouring non-Bcl-2 overexpressing cell types, probably due to release of a cytoprotective factor by Bcl-2 overexpressing cells (reviewed in [31]). Conversely, significantly increased Bcl2 expression was noted in murine bladder cultures treated with irradiated tissue conditioned medium, but despite this, apoptotic nuclear fragmentation was also significantly increased in bystander cells [37]. Alternatively, the factors producing the present study result could be DNA fragments released from irradiated cells undergoing apoptosis into the intercellular space. Experiments on human lymphocytes showed that extracellular DNA fragments interact with bystander cells and activate signal pathways, which lead to an adaptive response [58,59]. As an explanation for our data, it may be surmised that similar processes may also occur in the irradiated

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116

human lymphocytes cultured in the presence of plasma from irradiated blood. Interestingly, Acheva et al. [17] showed that calcium fluxes can be observed in irradiated reporter cells treated with irradiated blood plasma, irrespectively of whether the plasma produced a toxic or stimulatory effect. Moreover, reactive oxygen species appeared to be involved in adaptive plasma-mediated bystander responses when irradiated reporter cells showed an enhanced clonogenic survival, but not when it was reduced. A possible implication of this phenomenon operating in vivo concerns outcomes of the radiotherapy or accidental irradiation to high doses. Such a support of cell survival for irradiated cells, most of which carry radiation-induced genetic changes, could lead to enhanced survival of potentially malignant mutations. In a “standard” bystander situation DNA damage arising in unirradiated cells in excess of their spontaneous rate of mutagenesis is considered as very undesirable. This would be exacerbated by a “don’t die” signal focused on irradiated cells containing directly radiation-induced lesions as some might proliferate into malignant clones. This unexpected finding is clearly deserving of more extended study. A question still to be considered is whether a “don’t die” signal mediated by irradiated blood plasma causes just a delay or a full abrogation of apoptotic process in a proportion of irradiated cells. This should be continued with studying the effectiveness of survival of these cells when they move from a quiescent (G0 ) state into proliferation, as was done earlier with unirradiated reporter lymphocytes [23]. A further approach should be to simulate a fractionated exposure in vitro and to examine whether this plasma-induced effect can render previously irradiated cells more radioresistant (an adaptive response) or more susceptible to subsequent irradiations. The possibility of a dose effect relationship needs to be examined with reporter cells irradiated to a range of doses.

5. Conclusions Apoptosis measurements using the Annexin V test showed that plasma collected from unirradiated blood or blood irradiated in vitro to 2–40 Gy ␥-rays did not induce early apoptosis or late apoptosis/necrosis in reporter quiescent PBM, whereas direct radiation caused significant and dose-dependent apoptotic death. The magnitude of cell-killing effect directly induced by radiation exceeded any fluctuations of survival caused by plasma from irradiated or unirradiated blood in reporter unirradiated PBM. No additional cytotoxicity was detected when either unirradiated or irradiated plasmas were tested against irradiated PBM. Instead, surprisingly, plasma from irradiated blood caused a dose-dependent reduction of apoptosis in the irradiated reporter cells. This has implications concerning oncogenic risk of surviving cells after radiotherapy or high dose accidental irradiation in vivo.

Conflict of interest statement The authors declare no actual or potential conflict of interest including any financial, personal or other relationships with other people or organizations that could inappropriately influence their work. The authors alone are responsible for the content and writing of the paper.

Role of the funding source For this work V.A.V. was supported by EURATOM Research and Training Programme, grant 027289 (F16R). The EURATOM did not influence or participate in study design; in the collection, analysis,

115

and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Acknowledgements We wish to thank Jayne Moquet and Pat Hone (the UK Health Protection Agency) for valuable technical assistance, Elizabeth Ainsbury (the UK Health Protection Agency) for discussion of statistical issues of the paper, and Richard Doull (the UK Medical Research Council) for the irradiations of blood samples. References [1] C. Mothersill, C. Seymour, Radiation-induced bystander effects, carcinogenesis and models, Oncogene 22 (2003) 7028–7033. [2] K.M. Prise, M. Folkard, B.D. Michael, Bystander responses induced by low LET radiation, Oncogene 22 (2003) 7043–7049. [3] A.L. Brooks, Evidence for ‘bystander effects’ in vivo, Hum. Exp. Toxicol. 23 (2004) 67–70. [4] P.J. Coates, S.A. Lorimore, E.G. Wright, Damaging and protective cell signalling in the untargeted effects of ionizing radiation, Mutat. Res. 568 (2004) 5–20. [5] C. Mothersill, M.J. Moriarty, C.B. Seymour, Radiotherapy and the potential exploitation of bystander effects, Int. J. Radiat. Oncol. Biol. Phys. 58 (2004) 575–579. [6] W.F. Morgan, Non-targeted and delayed effects of exposure to ionizing radiation: II. Radiation-induced genomic instability and bystander effects in vivo, clastogenic factors and transgenerational effects, Radiat. Res. 159 (2003) 581–596. [7] E.G. Wright, P.J. Coates, Untargeted effects of ionizing radiation: implications for radiation pathology, Mutat. Res. 597 (2006) 119–132. [8] W.F. Morgan, M.B. Sowa, Non-targeted bystander effects induced by ionizing radiation, Mutat. Res. 616 (2007) 159–164. [9] K. Baverstock, A.V. Karotki, Towards a unifying theory of late stochastic effects of ionizing radiation, Mutat. Res. 718 (2010) 1–9. [10] P. Belloni, P. Latini, F. Palitti, Radiation-induced bystander effect in healthy G0 human lymphocytes: biological and clinical significance, Mutat. Res. 713 (2011) 32–38. [11] C. Seymour, C. Mothersill, Development of an in vivo assay for detection of nontargeted radiation effects, Dose–Response 4 (2006) 277–282. [12] O. Howe, J. O’Sullivan, B. Nolan, J. Vaughan, S. Gorman, C. Clarke, B. McClean, F.M. Lyng, Do radiation-induced bystander effects correlate to the intrinsic radiosensitivity of individuals and have clinical significance? Radiat. Res. 171 (2009) 521–529. [13] F.M. Lyng, C.B. Seymour, C. Mothersill, Early events in the apoptotic cascade initiated in cells treated with medium from the progeny of irradiated cells, Radiat. Prot. Dosim. 99 (2002) 169–172. [14] F.M. Lyng, P. Maguire, N. Klimurray, C. Mothersill, C. Shao, M. Folkard, K.M. Prise, Apoptosis is initiated in human keratinocytes exposed to signalling factors from microbeam irradiated cells, Int. J. Radiat. Biol. 82 (2006) 393–399. [15] M. Konopacka, J. Rzeszowska-Wolny, The bystander effect-induced formation of micronucleated cells is inhibited by antioxidants, but the parallel induction of apoptosis and loss of viability are not affected, Mutat. Res. 593 (2006) 32–38. [16] M. Grifalconi, L. Celotti, M. Mognato, Bystander response in human lymphoblastoid TK6 cells, Mutat. Res. 625 (2007) 102–111. [17] A. Acheva, R. Georgieva, I. Rupova, R. Boteva, F. Lyng, Bystander responses in low dose irradiated cells treated with plasma from gamma irradiated blood, J. Phys.: Conf. Ser. 101 (2008) 012005. [18] B.J. Blyth, E.I. Azzam, R.W. Howell, R.J. Ormsby, A.H. Staudacher, P.J. Sykes, An adoptive transfer method to detect low-dose radiation-induced bystander effects in vivo, Radiat. Res. 173 (2010) 125–137. [19] A.H. Staudacher, B.J. Blyth, M.D. Lawrence, R.J. Ormsby, E. Bezak, P.J. Sykes, If bystander effects for apoptosis occur in spleen after low-dose irradiation in vivo then the magnitude of the effect falls within the range of normal homeostatic apoptosis, Radiat. Res. 174 (2010) 727–731. [20] L.A. Ryan, R.W. Smith, C.B. Seymour, C.E. Mothersill, Dilution of irradiated cell conditioned medium and the bystander effect, Radiat. Res. 169 (2008) 188–196. [21] A.M. Vines, F.M. Lyng, B. McClean, C. Seymour, C.E. Mothersill, Bystander signal production and response are independent processes, which are cell line dependent, Int. J. Radiat. Biol. 84 (2008) 83–90. [22] S. Rajendran, S.H. Harrison, R.A. Thomas, J.D. Tucker, The role of mitochondria in the radiation-induced bystander effect in human lymphoblastoid cells, Radiat. Res. 175 (2011) 159–171. [23] D.C. Lloyd, J.E. Moquet, The clastogenic effect of irradiated human plasma, Int. J. Radiat. Biol. 47 (1985) 433–444. [24] I. Emerit, N. Oganesian, R. Arutyunian, A. Pogossian, T. Sarkisian, L. Cernjavski, A. Levy, J. Feingold, Oxidative stress-related clastogenic factors in plasma from Chernobyl liquidators: protective effects of antioxidant plant phenols, vitamins and oligoelements, Mutat. Res. 377 (1997) 239–246. [25] F. Gemignani, M. Ballardin, F. Maggiani, A.M. Rossi, A. Antonelli, R. Barale, Chromosome aberrations in lymphocytes and clastogenic factors in plasma detected

116

[26]

[27] [28] [29]

[30]

[31]

[32]

[33]

[34] [35]

[36] [37]

[38] [39]

[40]

[41] [42]

[43]

V. Vinnikov et al. / Mutation Research 731 (2012) 107–116 in Belarus children 10 years after Chernobyl accident, Mutat. Res. 446 (1999) 245–253. M. Ballardin, F. Gemignani, L. Bodei, G. Mariani, M. Ferdeghini, A.M. Rossi, L. Migliore, R. Barale, Formation of micronuclei and of clastogenic factor(s) in patients receiving therapeutic doses of iodine-131, Mutat. Res. 514 (2002) 77–85. I. Emerit, Clastogenic factors as potential biomarkers of increased superoxide production, Biomarker Insights 2 (2007) 429–438. C. Lindholm, A. Acheva, S. Salomaa, Clastogenic plasma factors: a short overview, Radiat. Environ. Biophys. 49 (2010) 133–138. S. Chen, Y. Zhao, W. Han, G. Zhao, L. Zhu, J. Wang, L. Bao, E. Jiang, A. Xu, T.K. Hei, Z. Yu, L. Wu, Mitochondria-dependent signalling pathway are involved in the early process of radiation-induced bystander effects, Br. J. Cancer 98 (2008) 1839–1844. S. Gorman, M. Tosetto, F. Lyng, O. Howe, K. Sheahan, D. O’Donoghue, J. Hyland, H. Mulcahy, J. O’Sullivan, Radiation and chemotherapy bystander effects induce early genomic instability events: telomere shortening and bridge formation coupled with mitochondrial dysfunction, Mutat. Res. 669 (2009) 131–138. S.J. Schwulst, C.G. Davis, C.M. Coopersmith, R.S. Hotchkiss, Adoptive transfer of dying cells causes bystander-induced apoptosis, Biochem. Biophys. Res. Commun. 353 (2007) 780–785. A. Schmitz, J. Bayer, N. Dechamps, G. Thomas, Intrinsic susceptibility to radiation-induced apoptosis of human lymphocyte subpopulations, Int. J. Radiat. Oncol. Biol. Phys. 57 (2003) 769–778. R. Lee, Chromosome aberrations in human lymphocytes irradiated with heavy ions, PhD thesis, Gesellschaft für Schwerionenforschung (GSI) and Technischen Universität Darmstadt, Darmstadt, 2006. H. Weng, K. Morimoto, Differential responses to mutagens among human lymphocyte subpopulations, Mutat. Res. 672 (2009) 1–9. H. Yang, N. Asaad, K.D. Held, Medium-mediated intercellular communication is involved in bystander responses of X-ray-irradiated normal human fibroblasts, Oncogene 24 (2005) 2096–2103. M. Ojima, N. Ban, M. Kai, DNA double-strand breaks induced by very low X-ray doses are largely due to bystander effects, Radiat. Res. 170 (2008) 365–371. A.M. Vines, F.M. Lyng, B. McClean, C. Seymour, C.E. Mothersill, Bystander effect induced changes in apoptosis related proteins and terminal differentiation in in vitro murine bladder cultures, Int. J. Radiat. Biol. 85 (2009) 48–56. T. Groesser, B. Cooper, B. Rydberg, Lack of bystander effects from high-LET radiation for early cytogenetic end points, Radiat. Res. 170 (2008) 794–802. C. Fournier, P. Barberet, T. Pouthier, S. Ritter, B. Fischer, K.O. Voss, T. Funayama, N. Hamada, Y. Kobayashi, G. Taucher-Scholz, No evidence for DNA and early cytogenetic damage in bystander cells after heavy-ion microirradiation at two facilities, Radiat. Res. 171 (2009) 530–540. M.B. Sowa, W. Goetz, J.E. Baulch, D.N. Pyles, J. Dziegielewski, S. Yovino, A.R. Snyder, S.M. de Toledo, E.I. Azzam, W.F. Morgan, Lack of evidence for low-LET radiation induced bystander response in normal human fibroblasts and colon carcinoma cells, Int. J. Radiat. Biol. 86 (2010) 102–113. A.C. Stevenson, G. Wiernik, Effects on lymphocytes in vitro of plasma from irradiated patients, Radiat. Res. 58 (1974) 277–285. A. Léonard, E.D. Léonard, G.B. Gerber, M.C. Crutzen-Fayt, F. Richard, J.G. Gueulette, N.B. Akhmatullina, No evidence for radiation-induced clastogenic factors after in vitro or in vivo exposure of human blood, Mutat. Res. 420 (1998) 33–36. N.B. Ahmatullina, J.G. Cherednichenko, K.A. Iskandarova, S.T. Ryskulova, Study of clastogenic effect of irradiated plasma, in: Proceedings of the IVth Congress on Radiation Research, vol. 2, Moscow, 2001, p. 473.

[44] N.V. Orlova, S.G. Smirnova, I.A. Zamulaeva, V.G. Andreev, N.I. Ryabchenko, A.S. Saenko, Apoptosis in peripheral blood lymphocytes of healthy donors and patients with laryngeal cancer after ␥-irradiation in vitro, Radiat. Biol. Radioecol. 41 (2001) 366–372. [45] O.V. Belyakov, M. Folkard, C. Mothersill, K.M. Prise, B.D. Michael, Bystanderinduced apoptosis and premature differentiation in primary urothelial explants after charged particle microbeam irradiation, Radiat. Prot. Dosim. 99 (2002) 249–251. [46] A. Facoetti, L. Mariotti, F. Ballarini, A. Bertolotti, R. Nano, F. Pasi, E. Ranza, A. Ottolenghi, Experimental and theoretical analysis of cytokine release for the study of radiation-induced bystander effect, Int. J. Radiat. Biol. 85 (2009) 690–699. [47] S. Burdak-Rothkamm, K. Rothkamm, K.M. Prise, ATM acts downstream of ATR in the DNA damage response signalling of bystander cells, Cancer Res. 68 (2008) 7059–7065. [48] M.A. Kadhim, S.R. Moore, E.H. Goodwin, Interrelationships amongst radiationinduced genomic instability, bystander effects, and the adaptive response, Mutat. Res. 568 (2004) 21–32. [49] O. Shemetun, M. Pilinskaya, O. Talan, Cytogenetic investigation of radiationinduced bystander effect in human peripheral blood lymphocytes, in: Proceedings of the 35th Annual Meeting of the European Radiation Research Society ERR’2006, Kyiv, 2006, p. 82. [50] M. Widel, A. Szurko, W. Przybyszewski, J. Lanuszewska, Non-irradiated bystander fibroblasts attenuate damage to irradiated cancer cells, Radioprotection 43 (2008) 194. [51] S. Chen, Y. Zhao, W. Han, S.K. Chiu, L. Zhu, L. Wu, K.N. Yu, Rescue effects in radiobiology: unirradiated bystander cells assist irradiated cells through intercellular signal feedback, Mutat. Res. 706 (2010) 59–64. [52] E.C. Mackonis, N. Suchowerska, M. Zhang, M. Ebert, D.R. McKenzie, M. Jackson, Cellular response to modulated radiation fields, Phys. Med. Biol. 52 (2007) 5469–5482. [53] R. Iyer, B.E. Lehnert, Factors underlying the cell growth-related bystander responses to ␣-particles, Cancer Res. 60 (2000) 1290–1298. [54] H. Kishikawa, K. Wang, S.J. Adelstein, A.I. Kassis, Inhibitory and stimulatory bystander effects are differentially induced by Iodine-125 and Iodine-123, Radiat. Res. 165 (2006) 688–694. [55] B. Shankar, R. Pandey, K. Sainis, Radiation-induced bystander effects and adaptive response in murine lymphocytes, Int. J. Radiat. Biol. 82 (2006) 537–548. [56] H. Matsumoto, S. Hayashi, M. Hatashita, H. Shioura, T. Ohtsubo, R. Kitai, T. Ohnishi, O. Yukawa, Y. Furusawa, E. Kano, Induction of radioresistance to accelerated carbon-ion beams in recipient cells by nitric oxide excreted from irradiated donor cells of human glioblastoma, Int. J. Radiat. Biol. 76 (2000) 1649–1657. [57] R. Iyer, B.E. Lehnert, Low dose, low-LET ionizing radiation-induced radioadaptation and associated early responses in unirradiated cells, Mutat. Res. 503 (2002) 1–9. [58] A.V. Ermakov, M.S. Kon’kova, S.V. Kostiuk, E.S. Ershova, N.A. Egolina, N.N. Veiko, Extracellular DNA fragments from culture medium of low-dose irradiated human lymphocyte trigger instigating of the oxidative stress and the adaptive response in non-irradiated bystander lymphocytes, Radiat. Biol. Radioecol. 48 (2008) 553–564. [59] A.V. Ermakov, M.S. Konkova, S.V. Kostyuk, N.A. Egolina, L.V. Efremova, N.N. Veiko, Oxidative stress as a significant factor for development of an adaptive response in irradiated and nonirradiated human lymphocytes after inducing the bystander effect by low-dose X-radiation, Mutat. Res. 669 (2009) 155–161.