Induction and inhibition of the pan-nuclear gamma-H2AX response in resting human peripheral blood lymphocytes after X-ray irradiation

Human peripheral blood lymphocytes (HPBLs) are one of the most sensitive cells to ionizing radiation (IR) in the human body, and IR-induced DNA damage and functional impairment of HPBLs are the adverse consequences of IR accidents and major side effects of radiotherapy. Phosphorylated H2AX (γH2AX) is a sensitive marker for DNA double-strand breaks, but the role and regulation of the pan-nuclear γH2AX response in HPBLs after IR remain unclear. We herein demonstrated that the pan-nuclear γH2AX signals were increased in a time- and dose-dependent manner, colocalized with >94% of TUNEL apoptotic staining, and displayed a typical apoptotic pattern in resting HPBLs after low LET X-ray IR. In addition, the X-irradiation-induced pan-nuclear p-ATM and p-DNA-PKcs responses also occurred in resting HPBLs, and were colocalized with 92–95% of TUNEL staining and 97–98% of the pan-nuclear γH2AX signals, respectively, with a maximum at 6 h post irradiation, but disappeared at 24 h post irradiation. Moreover, ATM/DNA-PKcs inhibitor KU55933, p53 inhibitor PFT-μ and pan-caspase inhibitor ZVAD-fmk significantly decreased X-irradiation-induced pan-nuclear γH2AX signals and TUNEL staining, protected HPBLs from apoptosis, but decreased the proliferative response to mitogen in X-irradiated HPBLs. Notably, whereas both KU55933 and PFT-μ increased the IR-induced chromosome breaks and mis-repair events through inhibiting the formation of p-ATM, p-DNA-PKcs and γH2AX foci in X-irradiated HPBLs, the ZVAD-fmk did not increase the IR-induced chromosomal instability. Taken together, our data indicate that pan-nuclear γH2AX response represents an apoptotic signal that is triggered by the transient pan-nuclear ATM and DNA-PKcs activation, and mediated by p53 and pan-caspases in X-irradiated HPBLs, and that caspase inhibitors are better than ATM/DNA-PKcs inhibitors and p53 inhibitors to block pan-nuclear γH2AX response/apoptosis and protect HPBLs from IR.


INTRODUCTION
In coping with ionizing radiation (IR)-induced double strand breaks (DSBs), mammalian cells promptly initiate the DNA damage response (DDR) to ensure efficient and accurate repair of damaged DNA and cell survival. 1 It is commonly believed that a key step in the response to DSBs is histone variant H2AX phosphorylation at serine 139 (S139) in the megabase chromatin region flanking the break to provide a docking site for DNA repair factors, which form visible γH2AX immunofluorescent foci in mammals. 2,3 Ataxia telangiectasia mutated (ATM) and DNA-dependent protein kinase catalytic subunit (DNA-PKcs) are two primary kinases that phosphorylate H2AX at S139 in response to IR-induced DSBs. [4][5][6] Moreover, ATM also phosphorylates and activates the tumor suppressor p53 to induce cell cycle arrest, DNA repair or apoptosis by transactivating downstream target genes, 7,8 and DNA-PKcs is dispensable for p53 activation. [9][10][11][12][13] In addition to its role in transcription regulation, p53 also functions in a transcriptionally independent manner in DNA repair 14 and apoptosis. 15,16 The activation of the caspase family of cysteine proteases has been shown as an extremely important step in the execution of p53mediated transcription-dependent and -independent apoptosis. 17 Apart from IR-induced formation of γH2AX foci, the induction of pan-nuclear γH2AX signals has been demonstrated as a cellular reaction to IR and as a consequence of various regulation mechanisms. [18][19][20] Cook et al. 18 have reported that H2AX phosphorylation at tyrosine 142 (Y142) induces an apoptotic response instead of a damage repair response to genotoxic stress by preventing the binding of the damage repair factors to phospho-serine 139 of γH2AX and promoting the effective recruitment of pro-apoptotic factors to H2AX. Moreover, H2AX phosphorylation at Y142 produces visible whole-nuclear γH2AX immunofluorescence staining instead of foci in TUNEL-positive mouse embryonic fibroblasts exposed to 100 Gy of irradiation. 18 However, the relationship between pan-nuclear γH2AX response and radiation quality remains unclear. Markova et al. 19 observed the γ-irradiation-induced pan-nuclear γH2AX response in HPBLs, but it was unknown whether the pan-nuclear γH2AX response was related to apoptosis. In addition, it was demonstrated that localized heavy ion irradiation-induced clustered DNA damage in sub-nuclear regions induced the transient nuclear-wide H2AX phosphorylation in undamaged chromatin in mammalian cells, including human fibroblasts, the U2-OS tumor cells, mouse embryonic fibroblast cells and hamster ovary cells; however, none of the neighboring un-irradiated cells showed a nuclear-wide γH2AX response, and targeted cytoplasmic irradiation with particles also did not induce the nuclear-wide γH2AX response. 20 It has been proposed that the particle-induced pan-nuclear γH2AX response is caused by ATM and DNA-PK activity distant to sites of DNA damage and is not related to either apoptosis or DSB repair. 20 The nuclear-wide kinase activity of ATM and DNA-PK may be elicited by dense IR-induced small DNA fragments 21 or by diffusion of active kinases from DNA DSBs. In addition, pan-nuclear γH2AX signals were induced by other stressors via various regulation mechanisms. It has been reported that tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), staurosporine, etoposide and ultraviolet irradiation induce the pan-nuclear γH2AX response associated with apoptosis in human cancer cells, 22 mouse fibroblasts, Chinese hamster ovary cells, human fibroblasts 23-25 and resting human T cells. 26 The pan-nuclear γH2AX response can be caused by replication-mediated DSBs in cancer cells treated with camptothecin, 27 indenoisoquinolines 28 or gemcitabine, 29 which are not related to apoptotic DNA fragmentation. Moreover, small DNA molecules, which mimic DNA DSBs (called D bait) and disorganize the DNA repair system in cancer cells, 30,31 and adeno-associated virus replication in infected host cells, which could be explained as a reaction to the singlestranded DNA part of the virus that compromises DNA repair, also can induce the pan-nuclear γH2AX response. 32,33 The changes in chromatin structure after hypotonic treatment are sufficient to elicit extensive pan-nuclear γH2AX formation in the absence of DNA strand breaks, which are not related to apoptosis. 34 Human peripheral blood lymphocytes (HPBLs) are extremely sensitive to IR and represent the most common target cells for biological dosimetric, 35 radiosensitive 36 and immune function studies. 37,38 γH2AX foci have been shown to be the most sensitive dosimeter and radiosensitive indicator. 39 HPBLs are non-dividing cells in the G0 phase of the cell cycle, and can proliferate to generate the immune response if stimulated. Therefore, HPBLs are able to efficiently repair DSBs to prevent apoptosis and mutations. Low LET γ-ray irradiation induces the activation of DDR proteins through phosphorylation of ATM at Ser 1981, DNA-PKcs at tyrosine 2609 and H2AX at Ser 139 to form DNA repair foci in HPBLs. 19,40,41 If the DNA damage is not repaired after IR, HPBLs undergo p53-dependent apoptosis through the caspase cascade, [42][43][44][45][46][47] and subsequently, functional inhibition of HPBLs may result in immune suppression in individuals exposed to acute or chronic IR at high doses and in radio/chemotherapy patients. 48,49 Therefore, it is of interest to understand the relationships among pan-nuclear γH2AX signals, apoptosis, DSB repair, chromosomal aberration and proliferative function in resting HPBLs exposed to low LET X-irradiation, and to explore the pharmacological approaches to protect HPBLs against IRinduced pan-nuclear γH2AX response/apoptosis. In this study, we demonstrated that X-ray-induced pan-nuclear γH2AX response represents an apoptotic signal in resting HPBLs, and is mediated by transient pan-nuclear ATM and DNA-PKcs activation and dependent on p53 and caspase activation. Our studies suggest that caspase inhibitors are better than ATM/DNA-PKcs inhibitors and p53 inhibitors to block pan-nuclear γH2AX response/ apoptosis and protect HPBLs from IR.

RESULTS
The kinetic patterns of the pan-nuclear responses are different from that of the DDR foci in resting HPBLs after X-irradiation As shown in Figure 1a, we detected the nuclear-wide γH2AX, p-ATM and p-DNA-PKcs responses and the γH2AX, p-ATM and p-DNA-PKcs foci in resting HPBLs after X-irradiation. Importantly, the kinetic patterns of the pan-nuclear responses are different from that of the DDR foci ( Figure 1b). For the γH2AX signals, although the percentages of pan-nuclear γH2AX-positive cells increased steadily after X-irradiation, the frequencies of the γH2AX foci peaked at 30 min after X-irradiation, decreased rapidly during 6 h post irradiation and was reduced to approximately 8% of the peak value at 24 h post irradiation (Figure 1b  suggest that the nuclear-wide γH2AX, p-ATM and p-DNA-PKcs responses are not related to DSB repair in resting HPBLs after X-irradiation. X-irradiation-induced pan-nuclear γH2AX, p-ATM and p-DNA-PKcs signals are colocalized with TUNEL apoptotic staining in resting HPBLs It was reported that pan-nuclear formation of γH2AX in undamaged chromatin induced by high-LET localized ion irradiation was not related to apoptosis in human fibroblasts, tumor cells and hamster ovary cells. 20 However, we found that the kinetics of the increase of the pan-nuclear γH2AX response was similar to that of apoptotic induction at 0.5, 6 and 24 h post irradiation (0.5 or 5.0 Gy) in HPBLs. Moreover, the percentage of TUNELpositive cells was similar to the percentage of pan-nuclear γH2AX-positive cells at each time point after the same dose radiation (Figure 2a Figure 2c), indicating that pan-nuclear γH2AX staining was associated with X-irradiation-induced apoptosis in resting HPBLs.
Furthermore, pan-nuclear p-ATM and p-DNA-PKcs signals were also colocalized with pan-nuclear γH2AX or TUNEL staining at 6 h post irradiation in resting HPBLs. It was found that 93-95% of the TUNEL-positive cells also exhibited pan-nuclear p-ATM or p-DNA-PKcs staining at 6 h post irradiation (5 Gy) (Figures 2b and c). We further found that 97-98% of cells with pan-nuclear γH2AX staining showed pan-nuclear p-ATM and p-DNA-PKcs staining (Figures 2b and c). Notably, HPBLs with pan-nuclear γH2AX staining did not exhibit pan-nuclear p-ATM and p-DNA-PKcs staining at 24 h post irradiation (Figure 2c), suggesting that ATM and DNA-PKcs were only activated at early stages after irradiation.
According to the morphological signs of apoptosis progression such as apoptotic rings, pan-nuclear staining and apoptotic bodies, we observed three types of morphological characterization of the apoptotic γH2AX response: γH2AX ring versus TUNEL ring and very weak TUNEL pan-staining of the nucleus; pannuclear γH2AX versus strong TUNEL pan-staining of the nucleus; and the γH2AX pan-staining within apoptotic bodies. Notably, no DDR foci were observed in the pan-nuclear responses.
X-irradiation-induced pan-nuclear γH2AX responses and apoptosis in resting HPBLs are mediated by ATM and DNA-PKcs We next investigated the role of ATM and DNA-PKcs in nuclearwide H2AX phosphorylation and apoptosis. It was found that pretreatment with KU55933, a potent inhibitor of ATM/DNA-PKcs, resulted in significant inhibition of X-irradiation-induced γH2AX, p-ATM and p-DNA-PKcs focus formation at 30 min post irradiation (0.5 or 5 Gy) in resting HPBLs (Figure 3a), suggesting that KU55933 inhibited X-irradiation-induced DSB repair. In addition, KU55933 significantly inhibited the pan-nuclear p-ATM and p-DNA-PKcs signals at 6 h post irradiation (5 Gy) (Figure 3b, left panel) in HPBLs. The inhibitory effects of KU55933 on p-ATM and p-DNA-PKcs expression were also demonstrated by western blotting analysis ( Figure 3b, right panel). We found that KU55933 also inhibited X-irradiation-induced p53 phosphorylation (Figure 3b, right panel). Correspondingly, KU55933 inhibited the pan-nuclear γH2AX signals and TUNEL apoptosis at 6 and 24 h post irradiation (0.5 or 5 Gy) in resting HPBLs ( Figure 3c). Together, our results suggest that KU55933 decreases X-irradiation-induced pannuclear γH2AX responses and apoptosis by inhibiting transient pan-nuclear p-ATM and p-DNA-PKcs responses and p53 phosphorylation in resting HPBLs.
X-irradiation-induced pan-nuclear γH2AX response and apoptosis in resting HPBLs were also mediated by p53 and pan-caspases It has been shown that apoptosis of HPBLs is p53-and caspase-dependent. 26,[43][44][45][46][47] We next investigated the roles of p53 and caspases in nuclear-wide H2AX phosphorylation and apoptosis. It was found that pretreatment with the p53 inhibitor PFT-μ and pan-caspase inhibitor ZVAD-fmk significantly reduced X-irradiation-induced p53 phosphorylation and cleaved-caspase-3 expression in resting HPBLs (Figure 4a, left panel). Similar to the inhibitory effects of KU55933 on X-irradiation-induced pan-nuclear γH2AX signals and TUNEL staining, pretreatment with PFT-μ or ZVAD-fmk also resulted in the significant suppression of X-irradiation-induced pan-nuclear γH2AX signals and apoptosis at 24 h post irradiation (0.5 or 5 Gy) in resting HPBLs (Figure 4a, right panel), suggesting that p53 and caspases are involved in the regulation of the nuclear-wide γH2AX response. Moreover, PFT-μ inhibited X-irradiation-induced pan-nuclear p-ATM and p-DNA-PKcs responses at 6 h post irradiation (5 Gy) and X-irradiation-induced p-ATM and p-DNA-PKcs focus formation at 4 h post irradiation, although it did not decrease X-irradiation-induced p-ATM and p-DNA-PKcs focus formation at 30 min post irradiation (Figure 4b), suggesting that p53 is involved in mediating the pan-nuclear p-ATM and p-DNA-PKcs responses, and has an important role in enhancing DSB repair in HPBLs after X-irradiation.
Effects of KU55933, PFT-μ and ZVAD-fmk on X-irradiation-induced chromosomal aberrations and proliferative response in resting HPBLs Although KU55933, PFT-μ and ZVAD-fmk protect HPBLs from X-irradiation-induced apoptosis, it is not clear whether they affect X-irradiation-induced genomic instability and protect HPBLs function. Table 1 shows that X-irradiation significantly induced chromosomal aberrations including breaks, dicentrics plus rings in resting HPBLs and increased the percentage of aberrant cells. KU55933 and PFT-μ, but not ZVAD-fmk, further significantly enhanced X-irradiation-induced chromosomal aberrations.
We investigated the impact of KU55933, PFT-μ and ZVAD-fmk on the proliferation of X-irradiated HPBLs. We found that the inhibitor (KU55933, PFT-μ or ZVAD-fmk) combined with X-irradiation resulted in a more significant decrease of the percentage of Ki67-positive cells than the inhibitor or X-irradiation alone in resting HPBLs, suggesting that KU55933, PFT-μ and ZVAD-fmk aggravated the inhibitory effect of X-irradiation on the proliferative response in resting HPBLs (Figures 5a-d).

DISCUSSION
Unlike the kinetics of localized ion-induced transient pan-nuclear γH2AX response with a peak at~1 h post irradiation, we demonstrated herein that X-irradiation-induced pan-nuclear γH2AX response and apoptosis were increased in a dose-and time-dependent manner in resting HPBLs exposed to X-irradiation. This phenomenon is similar to the kinetics of the anticancer agent etoposide-induced pan-nuclear γH2AX response and apoptosis in resting human T cells. 26 More importantly, we provide direct evidence that the X-irradiation-induced pan-nuclear γH2AX response is presented in the apoptotic HPBLs. The apoptotic γH2AX pan-nuclear response is a novel feature of the low LET IRinduced apoptosis in HPBLs, and could serve as a biomarker to distinguish apoptotic cells from the DNA-damaged cells with a focal pattern and to monitor the extent of radiation injury and side effects of radiotherapy.
In contrast, X-irradiation-induced pan-nuclear p-ATM and p-DNA-PKcs responses in resting HPBLs were transient with a maximum response at 6 h post irradiation and complete disappearance at 24 h post irradiation. The transient X-irradiation-pan-nuclear γH2AX response in X-irradiated HPBLs D Ding et al induced pan-nuclear p-ATM and p-DNA-PKcs responses are also presented in the apoptotic HPBLs. Moreover, ATM/DNA-PKcs inhibitor KU55933 not only inhibited X-irradiation-induced p-ATM and p-DNA-PKcs foci and the pan-nuclear p-ATM and p-DNA-PKcs responses but also suppressed the pan-nuclear γH2AX response and apoptosis in resting HPBLs. DNA-PK can be strongly activated during apoptotic DNA fragmentation 24 and ATM also can be activated by small double-stranded DNA fragments. 50 Our results suggest that the pan-nuclear γH2AX response is triggered by transient pan-nuclear p-ATM and p-DNA-PKcs signals in resting HPBLs, and that the pan-nuclear γH2AX response has a key role in X-ray-induced apoptosis in resting HPBLs.  pan-nuclear γH2AX response in X-irradiated HPBLs D Ding et al γH2AX was recently shown to be a crucial component in modulating apoptosis. 18,51 Lu et al. 51 showed that H2AX S139 phosphorylation by ultraviolet A-activated c-Jun N-terminal kinase (JNK) is required for DNA ladder formation mediated by caspase-3/ caspase-activated DNase (CAD) in apoptotic cells. Cook et al. 18 reported that phosphorylation of tyrosine 142 of H2AX promoted the apoptotic response and inhibited the DDR to DNA damage.
When H2AX tyrosine 142 was mutated to alanine, only γH2AX foci, but not pan-nuclear γH2AX and TUNEL staining, occurred after irradiation in H2AX − / − mouse embryonic fibroblast cells transfected with mutant H2AX (Y142F). 18 These observations support our results that X-irradiation-induced pan-nuclear γH2AX is involved in X-irradiation-induced apoptosis in resting HPBLs. Our results also suggest that there are no H2AX mutations at S139 and  Figure 3. Inhibition of X-irradiation-induced pan-nuclear γH2AX response and apoptosis in resting HPBLs by KU55933. (a) HPBLs were treated with 10 μM KU55933 for 2 h before X-irradiation (0.5 or 5 Gy). γH2AX, p-ATM and p-DNA-PKcs foci were measured in 500-1000 HPBLs from each sample. (b) HPBLs were treated with 10 μM KU55933 for 2 h before X-irradiation (5 Gy). Pan-nuclear p-ATM and p-DNA-PKcs cells (left panel) were counted, and western blotting assays (right panel) were performed to examine the expression of p-ATM/ATM, p-DNA-PKcs/DNA-PKcs and p-p53/p53 proteins at the indicated time points after irradiation. (c) HPBLs were treated with 10 μM KU55933 for 2 h before X-irradiation (0.5 or 5 Gy). Pan-nuclear γH2AX and TUNEL-positive cells were measured, and their representative images were captured at 24 h after X-irradiation; γH2AX was labeled in red and TUNEL in green, and nuclei were stained in blue with DAPI. A total of 2000-5000 HPBLs from each sample were examined to determine the percentage of the pan-nuclear γH2AX, p-ATM, p-DNA-PKcs and TUNEL-positive cells. × 100 magnification, × 1.0 zoom. The values shown in panel (a), panel (b) and panel (c) represent the means ± S.D. obtained from three to four donors. *Po 0.05, **P o0.01 and ***P o0.001 compared with corresponding non-irradiated HPBLs; ## Po0.01 and ### P o0.001 compared with corresponding HPBLs exposed to 0.5 Gy X-ray irradiation with/without inhibitor treatment; § P o0.05, § § Po0.01 and § § § P o0.01 compared with corresponding HPBLs without inhibitor treatment.
pan-nuclear γH2AX response in X-irradiated HPBLs D Ding et al Y142 sites in HPBLs, and that pan-nuclear γH2AX staining can be used to detect the apoptosis of HPBLs exposed to low LET IR. In addition, no γH2AX, p-ATM and p-DNA-PKcs foci were observed in the X-ray-induced pan-nuclear response and TUNEL-positive HPBLs. One possible explanation is that the condensed chromatin in the late stages of apoptosis blocks the expression of DNA repair proteins. 19 All together, we proposed that γH2AX regulates apoptosis in a pan-nuclear response pattern and is involved in the DSB repair response in a focal pattern in resting HPBLs.
It has been reported that IR-induced HPBLs apoptosis is mediated by caspase cascade and is p53-dependent, [42][43][44][45][46][47] which raises a question of whether X-irradiation-induced pan-nuclear γH2AX response is also dependent on p53 and caspases. In the present study, we demonstrated that p53 inhibitor PFT-μ and caspase inhibitor ZVAD-fmk inhibited the X-irradiation-induced nuclear-wide γH2AX response and apoptosis. A possible explanation is that the pan-nuclear γH2AX response is involved in the initiation and execution of apoptosis in HPBLs, and that p53 and caspases mediate the pan-nuclear γH2AX response by priming and executing apoptosis, respectively. Coincident with the inhibition of X-irradiation-induced pan-nuclear γH2AX response, we found that p53 inhibitor PFT-μ decreased the X-irradiation-induced pan-nuclear p-ATM and p-DNA-PKcs responses, suggesting that p53 may mediate the pan-nuclear γH2AX response through activating the pan-nuclear p-ATM and p-DNA-PKcs responses. It has been found that wild-type p53 protein is able to rejoin DNA with DSBs to facilitate precise ligation through direct physical interaction of p53 with DNA-PK of NHEJ, [52][53][54] and that p53 phosphorylation by ATM is involved in the initial steps of DNA damage signaling. 55 In the present study, we also demonstrated that p53 inhibition by PFT-μ results in a decrease of DSB repair and an increase of chromosome breaks by inhibiting the X-irradiation-induced p-ATM and p-DNA-PKcs focus formation. Similar to ATM/DNA-PKcs inhibitor KU55933, PFT-μ can inhibit the pan-nuclear γH2AX response, the formation of p-ATM and p-DNA-PKcs foci, and pan-nuclear p-ATM and p-DNA-PKcs responses. However, we also found that caspase inhibitor ZVADfmk decreased the nuclear-wide γH2AX response and apoptosis but did not affect the X-ray-induced chromosome breaks, suggesting that the nuclear-wide γH2AX response is involved in caspase-mediated apoptotic execution, but caspases are not involved in DSB repair, as caspases are activated at the execution phase of apoptosis when DSB repair process is finished.
It is well established that the IR-induced massive cell death in radiosensitive tissues is not due to irreversible damage of cells but rather to activation of apoptosis, therefore blockage of apoptosis has become a pharmacological approach to radioprotection of normal tissues. 56 Similarly, inhibition of DSB repair has been well known as an efficient approach to facilitate the apoptosis and sensitize tumor cells to IR and DSB-inducing chemotherapeutic agents such as etoposide, doxorubicin and camptothecin. 57 Surprisingly, we found that ATM/DNA-PKcs inhibitor KU55933 protected resting HPBLs from the X-ray-induced pan-nuclear γH2AX response and apoptosis, although it inhibited the X-rayinduced DSBs repair and enhanced chromosomal breaks. Our results are consistent with the literature reports that inhibition of ATM and DNA-PKcs phosphorylation by KU55933 can result in the suppression of H2AX phosphorylation and etoposide-induced apoptosis in resting HPBLs. 26,58 It is believed that HPBLs in a resting status tolerate a limited un-repair and inaccurate repair, and that there are opposite actions of KU55933 on normal resting cells and proliferating cancer cells. Owing to dual functions of p53 in mediating the DSB repair and apoptosis in resting HPBLs, p53 inhibitor PFT-μ, like ATM/DNA-PKcs inhibitor KU55933, protected resting HPBLs from the X-ray-induced pan-nuclear γH2AX response and apoptosis, and concurrently induced a decrease of DSB repair and an increase of chromosome breaks and frequencies of chromosomal aberrations. Our observations are consistent with the findings that PFT-α decreases the repair efficiency of X-ray-induced DNA lesions, leading to increased frequencies of chromosomal aberrations and reduced apoptosis in HPBLs. 46 Unlike the dual role of KU55933 and PFT-μ, caspase Table 1 Figure 5. KU55933, PFT-μ and ZVAD-fmk inhibited the proliferative response to mitogen in resting HPBLs exposed to X-ray irradiation. HPBLs were treated with 10 μM KU55933 for 2 h (a), 20 μM PFT-μ for 5 h (b) or 100 μM ZVAD-fmk for 1 h (c) followed by irradiation with 5 Gy X-ray, and were cultured with PHA and colcemid for 48 h. HPBLs were then collected and transformed HPBLs were detected using the Ki67 antibody. A total of 1000 HPBLs from each sample were examined to determine the proliferation ratio. Data are presented as the means ± S.D. obtained from three donors. *Po0.05 and **Po 0.01 compared with corresponding non-irradiated HPBLs with/without inhibitor treatment; § P o0.05 and § § P o0.01 compared with corresponding HPBLs without inhibitor treatment. (d) Representative images of HPBLs labeled with Ki67. The cells were treated with KU55933 and irradiated with X-ray as described above. Ki67 was labeled in red and nuclei were stained in blue with DAPI. × 40 magnification, × 1.0 zoom. pan-nuclear γH2AX response in X-irradiated HPBLs D Ding et al inhibitor ZVAD-fmk protected HPBLs from X-ray-induced nuclearwide γH2AX response and apoptosis, and did not worsen the X-ray-induced chromosomal instability. More importantly, we found that the inhibition of X-ray-induced nuclear-wide γH2AX response and apoptosis by KU55933, PFT-μ and ZVAD-fmk aggravated the inhibitory effects of X-irradiation on the proliferative response of HPBLs to mitogens, suggesting that the inhibition of the nuclear-wide γH2AX response and apoptosis does not protect the function of HPBLs. It is possible that the inhibition of DSB repair by KU55933 and PFT-μ does not result in correction of chromosome aberrations in HPBLs, and that the increased genomic instability leads to further suppression of the proliferative response of HPBLs exposed to X-irradiation, which limits the pharmacological use of DSB repair inhibitors and p53 inhibitors in radioprotection of normal tissues. Nevertheless, the aggravated effect of ZVAD-fmk on IR-induced proliferative inhibition may be related to suppression of the expression of lymphocyte-stimulated factor receptors, such as IL-2. 59 Lawrence et al. 59 found that ZVADfmk inhibits the proliferative response of human T lymphocytes induced by mitogens and IL-2 via inhibiting the expression of the IL-2 receptor, and that suppression of human T-cell proliferation by ZVAD-fmk is independent of its caspase inhibition properties, as ZVAD-fmk inhibits caspase processing during apoptosis but not during T-cell activation. Because caspase inhibitors are capable of providing reversible inhibitory effects on proliferative response in HPBLs by temporarily blocking the expression of proliferative response receptors, we proposed that inhibition of caspases would be a promising pharmacological approach for radioprotection of HPBLs.
In summary, we showed herein that X-irradiation induced not only the pan-nuclear γH2AX response but also the transient pan-nuclear p-ATM and p-DNA-PKcs responses in addition to the formation of nuclear DDR foci containing activated ATM, DNA-PKcs and γH2AX in resting HPBLs. The pan-nuclear γH2AX response and the transient pan-nuclear p-ATM and p-DNA-PKcs responses were directly related to X-ray-induced apoptosis. Moreover, the pan-nuclear γH2AX response and apoptosis were mediated not only by transient pan-nuclear p-ATM and p-DNA-PKcs signals but also by p53 and pan-caspase activities in resting HPBLs exposed to X-irradiation. ATM/DNA-PKcs inhibitor and p53 inhibitor attenuate the X-ray-induced pan-nuclear γH2AX response and apoptosis, but worsen the X-ray-induced functional impairment and genomic instability, suggesting that these inhibitors cannot be used as radioprotective agents for HPBLs. The pancaspase inhibitor attenuates X-ray-induced pan-nuclear γH2AX response and apoptosis with no effect on X-ray-induced genomic instability and holds developmental promise.

Peripheral blood sample collection and lymphocyte isolation and culture
Human peripheral blood samples were obtained from informed healthy male and female volunteers (22-25 years old), and the research was approved by Fudan University School of Public Health Institutional Review Board. For each experiment, blood was collected in sterile heparinized Vacutainers from three to six different donors. Lymphocytes were isolated from blood using density gradient centrifugation in Ficoll-Paque (Cedarlane Co., Burlington, ON, Canada) according to the manufacturer's instructions. Isolated HPBLs at a density of 1 × 10 6 cell/ml were cultured in RPMI 1640 medium (Gibco, Invitrogen Technologies, Carlsbad, CA, USA) supplemented with 15% fetal calf serum (Invitrogen Technologies), 5% (v/v) plasma, 100 U/ml penicillin and 100 μg/ml streptomycin at 5% CO 2 and 37°C in a humidified incubator. The viability of the isolated HPBLs was above 95% as measured with the trypan blue exclusion assay.

Irradiation and treatment
An X-Rad 320 biological irradiator (Precision X-Ray, Inc., North Branford, CT, USA) was used for the X-ray irradiation, and the dose rate was 1.83 Gy/min. The HPBLs were pretreated with 10 μM KU55933 (Selleckchem, Houston, TX, USA) for 2 h, 20 μM PFT-μ (Selleckchem) for 5 h or 100 μM ZVAD-fmk (Selleckchem) for 1 h, and then irradiated in a flask at room temperature at doses of 0.5, 2 or 5 Gy. After irradiation, the cells were incubated in a CO 2 incubator at 37°C and harvested after various repair periods for immunofluorescence staining, apoptosis assay and western blotting analysis. For the lymphocyte proliferative assay, phytohemagglutinin (Shanghai Jikong Biotechnology, Shanghai, China) was added to the culture medium after IR, and the cells were further incubated for 48 h.

Immunofluorescence analysis
The suspension of irradiated and sham-irradiated HPBLs was immediately placed on ice for 10 min and washed in ice-cold PBS, and the cells were spun onto the triple chambered slides precoated with 50 mg/l poly-Llysine (Sigma, St. Louis, MO, USA) using a cytospin centrifuge (Iris sample processing, Beckman Coulter Inc., Miami, FL, USA), fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.5% Triton X-100 for 15 min and followed by blocking with 10% (v/v) fetal calf serum in PBS at 37°C for 1 h. The cells were then incubated overnight at 4°C with anti-γH2AX (S139) antibody (Cell Signaling Technology, Danvers, MA, USA), anti-phospho-ATM (S1981) antibody (Abcam, Hong Kong, China) and antiphospho-DNA-PKcs (T2609) antibody (Abcam) or anti-Ki67 antibody (Cell Signaling Technology), followed by secondary antibody labeled with Alexa Fluor 488-conjugated donkey anti-rabbit (Molecular probe, Life Technologies, Grand Island, NY, USA), Alexa Fluor 555-conjugated donkey antimouse or Alexa Fluor 647-conjugated donkey anti-rabbit (Molecular probe) at room temperature for 1 h. Fluorescence images were acquired by an Olympus BX51 fluorescent microscope (Olympus Optical Co., Tokyo, Japan) equipped with a COHU CCD camera (Audio Video Supply, San Diego, CA, USA) and VideoTesT-FISH software (VideoTesT, Saint Petersburg, Russia). Images used for comparisons between different treatments were acquired and processed with the same instrument settings. The normalized levels of pan-nuclear staining or foci of γ-H2AX, p-ATM (S1981) and p-DNA-PKcs (T2609) were the ratios of the number of pan-nuclear staining or foci to the number of cells scored. In addition, colocalization of TUNEL staining with pan-nuclear γ-H2AX, p-ATM or p-DNA-PKcs staining, and colocalization of pan-nuclear γ-H2AX staining with pan-nuclear p-ATM or p-DNA-PKcs staining were examined. The colocalization ratios were calculated as follows: number of colocalized pan-nuclear staining/number of TUNEL staining or pan-nuclear γ-H2AX staining.

Apoptosis detection
After pretreatment with inhibitors of ATM/DNA-PKcs, p53 and pancaspases or/and X-irradiation, the cell monolayers were spun down on slides, and subjected to terminal deoxynucleotidyl transferase-mediated dUTP (2′-deoxyuridine 5′-triphosphate) biotin nick end labeling (TUNEL) staining using a TUNEL Apoptosis Detection Kit (Vazyme Biotech, Nanjing, Jiangsu Province, China) for in situ detection of apoptosis according to the manufacturer's instructions. For combination with immunofluorescence analysis, fluorescein isothiocyanate-12-dUTP labeling mix was added and incubated for 1 h after incubation with the primary antibody, followed by incubation with the secondary antibody. The TUNEL-stained cells were counter-stained with mounting medium containing 4′,6-diamidino-2phenylindole (DAPI) (Santa Cruz Biotechnology, Santa Cruz, CA, USA); the fluorescein isothiocyanate-labeled TUNEL-positive cells were imaged using an Olympus BX51 fluorescent microscope, and the TUNEL-positive cells were counted to calculate the TUNEL indices.

Chromosomal aberration analysis
After pretreatment with inhibitors of ATM/DNA-PKcs, p53 and pancaspases, HPBLs were irradiated by X-rays. After a 2-h repair period, the cells were incubated with colcemid and phytohemagglutinin, and the levels of X-ray-induced chromosomal aberrations in HPBLs were determined at 48 h after irradiation. Chromosome preparations were made using standard techniques after the incubation of the cells in hypotonic solution (0.075 M KCl) and fixation with methanol/acetic acid (3 : 1). For each sample, 100 metaphases per donor were scored. The chromosomal aberrations were classified according to Savage. 60 pan-nuclear γH2AX response in X-irradiated HPBLs D Ding et al

Statistical analysis
Data are presented as the means ± standard deviation. Statistical analysis of the data was performed using SPSS 17.0 software. Independent-samples T test was applied to analyze the differences. A difference with P o0.05 was considered to be statistically significant.