Human skin fibroblast GM0639 cell line was obtained from the Radiobiological Laboratory of the National Research Center for Environment and Health (GSF), Germany. Human bronchial epithelial 16HBE cell line was obtained from the Shanghai Cell Collection Center of the Chinese Academy of Sciences. These cells were maintained as a monolayer in low glucose DMEM culture medium supplemented with 10% fetal calf serum(Hyclone, Logan, UT, USA), 2 mmol/L glutamine, penicillin(100 U/ml) and streptomycin(100 U/ml). Cells were kept at 37℃ in an atmosphere of 5% carbon dioxide and 95% air, and subcultured twice a week to remain in exponential growth. Cells were washed twice with, and subsequently cultured in, serum-free OPTI-MEM I medium (Gibco BRL, Grand Island, NY) before X-ray irradiation.

Male and female Sprague-Dawley rats(8-12 weeks old, 220-250 g) were allowed to acclimate for 7 days before X-ray irradiation. All animal procedures were approved by the Institutional Animal Care and Use Committee of the Soochow University. Rats were housed in the animal facility of the Key Laboratory of Jiangsu Radiation Medicine and Protection under standard temperature, and light and dark cycles.

As a form of radiation when charged electrons or ions of sufficient energy hit a material with high speed, X-ray was generated in high voltage electron tubes of X-ray generators(Primus High-Energy Sie-mens), which emitted high energy photons at a rate of 200 cGy/min. For cell irradiation, a locator with a source-cell distance of 100 mm, irradiation field of 40 cm×40 cm was employed to expose cell cultures to X-ray at doses ranging from 0 to 8 Gy. For animal experiment, rats were irradiated using X-ray generator emitting at a fixed dose rate of 200 cGy/min, with an irradiation field of 40 cm×40 cm, centered 100 mm above the animals. At various time points after X-ray radiation, 0.5 ml blood was collected from 6 rats(3 male and 3 female) by tail bleeding, and serum hmgb1 levels were determined using the Shino-Test Corporation ELISA kit.

At 24 h after X-ray radiation, cellular HMGB1 was immunostained with HMGB1-specific polyclonal antibodies, and images were acquired using a fluorescent microscope as previously described^[29]. Alternatively, the subcellular localization of HMGB1 was examined by a cell fractionation/Western blotting technique as previously described^[29]. Cell fractionation is based on differential lysis of plasma and nuclear membranes by nonionic detergent(NP-40). Briefly, after selective lysis of the plasma membrane in low salt buffer(10 mmol/L HEPES, pH 7.9; 10 mmol/L KCl; 0.1 mmol/L EDTA; 0.1 mmol/L EGTA; 1 mmol/L DTT; 0.5 mmol/L PMSF, 1% NP-40), the intact nuclei was collected by a quick centrifugation step(7000 g, 1 min, 4℃), leaving the cytoplasmic fraction in the supernatant. The nuclei pellet was resuspended in NP-40 high salt buffer(20 mmol/L HEPES, pH 7.9; 0.4 mol/L NaCl; 1 mmol/L EDTA; 1 mmol/L EGTA; 1 mmol/L DTT; 1 mmol/L PMSF, 1% NP-40), and briefly sonicated to generate the nuclear fraction. After fractionation, the protein content of different fractions was determined by a Bradford method, and each fraction was assayed for levels of various protein by Western blotting analysis using primary antibodies specific for HMGB1, a cytoplasmic protein(β-actin, Santa Cruz Biotechnology), and a nuclear protein (Lamin B1, BD Biosciences).

Immediately after irradiation, cells grown on covered slide chambers(Lab-Tek, Nunc, Napterville, IL, USA) were washed with PBS and fixed with 2% paraformaldehyde in PBS for 15 min. After three washes of PBS with 10 min each, the cells were treated with 0.2% Triton X-100 solution in PBS for 5 min, and stained with mouse monoclonal antibody for Serine 139 phospho-H2AX from Millipore (Cat. 05636, 1: 200) overnight at 4℃. After extensive washings, FITC-labeled rabbit anti-mouse antibodies were added, and fluorescent images of cells were captured using a fluorescence microscope.

Cell viability was measured by the reduction of yellow tetrazolium salt[MTT, 3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide) by dehydrogenase of metabolically active cells, to generate reducing purple formazan that can be solubilized and quantified by spectrophotometry. The MTT substrate was prepared in a physiologically balanced solution, added to cell culture at a final concentration of 0.2 mg/ml, and incubated for 2 hours. The quantity of the resultant purple formazan was measured by recording changes in absorbance at 570 nm using a plate reading spectrophotometer.

Data are expressed as mean ± SEM of two independent experiments in triplicates. One-way analyses of variance (ANOVA) followed by the Tukey′s test for multiple comparisons were used to compare between different groups. A P value less than 0.05 was considered statistically significant.

To investigate the impact of X-ray irradiation on possible HMGB1 release, we first determined its effect on HMGB1 cytoplasmic translocation-an essential step for subsequent HMGB1 release. Quiescent tumor cells constitutively expressed HMGB1 and maintained an intracellular "pool" of HMGB1 predominantly in the nucleus(Fig. 1, left panels). At 24 h post X-ray irradiation(8.0 Gy), large amount of HMGB1 staining was also noticed in numerous cytoplasmic vesicles (Fig. 1, right panels), suggesting that radiation stimulated tumor cells to actively translocate nuclear HMGB1 into the cytoplasmic vesicles before releasing into the extracellular milieu.

Figure 1.  Radiation induced cytoplasmic HMGB1 translocation in tumor cells. Human skin fibroblast(GM0639) and bronchial epithelial(16HBE) cells were subjected to 6 MeV X-ray radiation at a dose of 8 Gy for 24 h, and assayed for HMGB1 cytoplasmic translocation by immunohistochemistry using HMGB1-specific antibodies. Note that HMGB1 was predominantly localized in the nuclear region of un-treated cells("control"), but found in both cytoplasmic and nuclear regions of X-ray radiated cells ("6 MeV-X").

To confirm the cytoplasmic HMGB1 translocation, whole cell lysates were fractionated and the levels of HMGB1 in the cytoplasmic and nuclear fractions were determined by Western blotting analysis. The relative levels of HMGB1(with reference to Lamin B1) in the nuclear fractions were significantly reduced in both GM0639 and 16HBE tumor cells after X-ray irradiation(Fig. 2A). In parallel, the relative levels of HMGB1(with reference to β-actin) in the cytoplasmic fraction were significantly elevated after irradiation (Fig. 2B), confirming that X-ray irradiation induced significant HMGB1 cytoplasmic translocation in these tumor cells.

Figure 2.  Radiation inversely altered nuclear and cytoplasmic HMGB1 levels. Following X-ray radiation, cytoplasmic and nuclear fractions were isolated, and assayed for levels of HMGB1 along with a nuclear(Lamin B1) or cytoplasmic(β-actin) marker by Western blotting analysis. Equal loading of samples was confirmed by Western blotting analysis of respective fractions with cytoplasmic (β-actin) or nuclear (Lamin B1) protein markers.

To determine whether X-ray radiation induces HMGB1 release, extracellular levels of HMGB1 in cell-conditioned culture medium were determined by Western blotting analysis. The levels of HMGB1 in the culture medium conditioned by the quiescent tumor cells were relatively low. Following X-ray irradiation, extracellular HMGB1 levels were elevated in a dose- and time-dependent fashion (Fig. 3). At a dose as low as 4-8 Gy, X-ray radiation induced HMGB1 release as early as 6 h post stimulation(Fig. 3).

Figure 3.  Radiation induced a dose- and time-dependent HMGB1 release. Human skin fibroblast(GM0639) and/or bronchial epithelial (16HBE) were exposed to X-ray radiation at various doses for different time periods, and extracellular levels of HMGB1 were determined by Western blotting analysis. Note that proteins were recovered from equal volume of cell-conditioned medium, and sample loading was normalized by equal volume of cell-conditioned medium.

In addition to active secretion, HMGB1 could also be passively released by injured cells. It is known that ionizing radiation can cause double-stranded breaks of chromosomal DNA, which activates histone γ-H2AX phosphorylation, and results in the recruitment of DNA repair proteins to form the γ-H2AX foci, a biomarker for chromosomal DNA damage. To test the impact of radiation on DNA damage, we examined the effect of X-ray irradiation on the formation of γ-H2AX foci in both tumor cell lines. As indicated in Fig. 4A, X-ray irradiation, at a dose as low as 4 Gy, induced marked DNA damage as judged by the formation of γ-H2AX foci(Fig. 4A). Consistently, the cell viability was significantly reduced by X-ray irradiation in both GM0639 and 16HBE tumor cells (Fig. 4B), suggesting that X-ray irradiation induced HMGB1 release partly through passive leakage from these dying cells.

Figure 4.  Radiation caused DNA damage and loss of cell viability. Human skin fibroblast (GM0639) and bronchial epithelial (16HBE) cells exposed to X-ray at a dose of 4 Gy, and cells were stained with γ-H2AX-specific antibodies to detect DNA damage. In parallel, the cell viability was determined by MTT assay, and expressed as a % of controls in the absence of X-ray radiation. *, P < 0.05 versus untreated control at respective time points.

To examine whether X-ray irradiation induces HMGB1 release in vivo, we subjected male and female rats to X-ray irradiation at different doses, and measured circulating levels of HMGB1 by ELISA. In agreement with the in vitro findings, X-ray irradiation induced systemic HMGB1 accumulation in a dose- and time-dependent fashion (Fig. 5). At a dose as low as 6 Gy, X-ray irradiation induced significant HMGB1 accumulation in the circulation as early as 6 h post stimulation (Fig. 5).

Figure 5.  Radiation elevated circulating HMGB1 levels in vivo. Male Sprague-Dawley rats were exposed to X-ray radiation at various doses and for different time periods, and blood samples were collected to measure serum HMGB1 levels by ELISA. *, P < 0.05 versus untreated controls(no radiation, Panel A; or immediately prior to X-ray radiation, Panel B).

As a form of radiation, X-rays emit high energy photons that can donate energy to cellular molecules, kicking out atomic electrons from the inner orbit to produce unstable and highly reactive free radicals. These radicals quickly react with nearby molecules, resulting in breakage of chemical bonds or oxidation (addition of oxygen atoms) of the affected molecules. In the present study, we demonstrated that X-ray radiation induced DNA damage as manifested by the formation of γ-H2AX foci in the nuclei, and induced cytoplasmic HMGB1 translocation and release in human skin fibroblast(GM0639) and bronchial epithelial(16HBE) cell lines. Similar findings have been demonstrated using the breast(MCF-7, data not shown), lung(NCI-H1703), prostate(DU-145 and PC-3) and colorectal(HCT 15 and SW480)^[30] and glioblastoma (T98G and U251MG)^[31] tumor cell lines.

As aforementioned, similar cytoplasmic HMGB1 translocation occurs in innate immune cells following stimulation with cytokines(IFN-γ)^[11] or hydrogen peroxide^{[14, 32]}. Because cytoplasmic HMGB1 translocation was not closely associated with the occurrence of cell death^{[11, 14, 32]}, it was believed that cytoplasmic HMGB1 translocation might be an active process regulated by chemical modifications of HMGB1 NLS. Notably, ionizing radiation also induces water radiolysis to produce radicals, which can be converted into hydrogen peroxide (H₂O₂). Because hydrogen peroxide can induce active HMGB1 secretion or passive leakage^{[14, 32]}, we propose that X-ray induces HMGB1 release through multiple mechanisms that are dependent on both active cytoplasmic translocation and passive leakage.

In animals, X-ray radiation causes injury initially at the skin, but goes beyond the surface and continues to damage inner tissues in the body. Although many irradiated cells could repair DNA and protein damage, some cells would die of necrosis or apoptosis. It is thus possible that radiation may induce HMGB1 secretion and leakage to amplify an inflammatory response. Indeed, HMGB1 carries three redox-sensitive cysteine residues(C23, C45 and C106), and can exist in three isoforms termed "HMGB1" (all thiol form), "disulfide HMGB1"(partially oxidized), and oxidized HMGB1^[33-34]. The "all-thiol" HMGB1 binds to other chemokines (e.g., CXCL12) to facilitate leukocyte recruitment via the CXCR4 receptor^[35] or other signaling molecules^[36-38] to the sites of injury^[39-40]. In a sharp contrast, the "disulfide" HMGB1 can activate immune cells to produce cytokines/chemokines via TLR4 or other receptors such as RAGE^[41], TLR2, TLR4^[42-44], TLR9^{[10, 41]}, cluster of differentiation 24 (CD24)/Siglec-10^[45], Mac-1^[38], thrombomodulin^[46], or single transmembrane domain proteins(e.g., syndecans)^[47]. Once fully oxidized, HMGB1 is devoid of either chemokine or cytokine activities^[33-34]. Thus, extracellular HMGB1 could serve as a proinflammatory signal to recruit and activate innate immune cells to sustain a potentially injurious inflammatory response to radiation.

It has been well established that excessive HMGB1 release adversely contributes to the pathogenesis of infection- and injury-elicited inflammatory diseases, because HMGB1-neutralizing antibodies are protective in animal models of sepsis^{[9, 48-51]}, ischemia/reperfusion^{[22, 52-53]}, trauma^[54-55], chemical toxemia^{[26, 56-57]}, atherosclerosis^[58], gastric ulcer^[59] and hyperoxia^[60]. Even during lethal infection^{[9, 48-51]}, tissue injury is accompanied by massive HMGB1 release that further amplifies the cytokine storm to precipitate organ dysfunction. In fact, HMGB1 itself could trigger caspase-1-dependent programmed cell death, pyroptosis, which is characterized by rapid plasma membrane rupture, and release of proinflammatory intracellular contents (including HMGB1)^[61]. Although it is difficult to distinguish between microbial infection-induced sepsis from injury-elicited systemic inflammatory response syndrome^[62-63], it might be important to develop strategies to specifically attenuate radiation-mediated inflammatory responses.

Recently, many herbal components including the green tea polyphenolic catechins^[64], tanshinones^[29], carbenoxolone^[65] have been shown to inhibit HMGB1 release through multiple mechanisms. These divergent mechanisms include the stimulation of autophagic degradation of cytoplasmic HMGB1^[66], the enhancement of endocytosis of exogenous HMGB1 into cytoplasmic vesicles^[67], or inhibition of key signaling molecules(e.g., PKR) involved in the regulation of HMGB1 release^[68]. It may be important to assess whether these herbal HMGB1 inhibitors can similarly prevent radiation-induced excessive inflammation without compromising the efficacy of radiation in the treatment of malignant cancers.