Background
For decades an anti-inflammatory and analgetic effect of low-dose X-irradiation (LD-RT) has been well established in the treatment of a plethora of benign diseases and chronic degenerative disorders [
1,
2] with empirically identified single doses < 1 Gy to be most effective in the clinical setting [
3‐
5]. Although the knowledge of the underlying cellular and molecular mechanisms is still at an early stage, a modulation of endothelial cell (EC) activity has already been proven to comprise a key element in the therapeutic effects of LD-RT [
6]. For instance, a hampered adhesion of peripheral blood mononuclear (monocytes, lymphocytes) and polymorphonuclear leukocytes (granulocytes) to EC was shown to result from an elevated secretion of the anti-inflammatory cytokine transforming growth factor β1 (TGF-β1), elevated levels of X-chromosome linked inhibitor of apoptosis protein (XIAP) and transcription factor nuclear factor-κB (NF-κB) DNA-binding and transcriptional activity [
7‐
10]. Moreover, a hampered adhesion and consequently reduced immune cell infiltration [
11] is further supported by a lowered expression of the adhesion molecule E-selectin with a local minimum following 0.3-0.5 Gy exposure [
9,
12]. Strikingly, the mechanisms explored so far display comparable non-linear dose effect relationships, a common hallmark of bystander and non (DNA)-targeted effects of low-dose irradiation [
13] and the phenomenon of low-dose hypersensitivity reported for cellular survival [
14]. These mechanisms are supposed to originate from an overlap of multiple molecular processes that may be initiated at various dose thresholds [
15].
Reactive oxygen species (ROS) like superoxide (O
2•-), Hydroxyl (OH
•) and Hydroperoxyl (HO
2•) ions formed as natural by-product of the mitochondrial electron transport chain and by NADPH oxidase activity display important roles in the regulation of cell signalling, cellular homeostasis, cell death and mutagenesis of DNA [
16,
17]. These regulatory effects include reversible oxidation of serine/threonine phosphatases and kinases, e.g. mitogen-activated protein kinase (MAPK), metalloproteases and activation of transcription factors like NF-κB and activating protein 1 (AP1) [
18]. Moreover, following environmental stress, including ionising radiation and heat exposure, ROS levels increase dramatically resulting in significant damage to cellular structures and induction of DNA double-strand breaks (DSBs) [
19].
A putative interrelationship between DNA damage repair and a discontinuous dose response relationship following low-dose irradiation was recently suggested [
20]. By assessing serine 139 phosphorylated histone γH2AX foci induction, a marker of radiation-induced DSBs [
21], a biphasic behaviour of γH2AX foci induction with a low-dose hypersensitivity in whole blood and less pronounced for isolated T-lymphocytes after X-irradiation was reported in line with a delayed repair with 40% of initial γH2AX foci persisting 24 hours post-irradiation [
20]. A mechanistic involvement of ROS in the modulation of these non-linear dose response effects, however, remains to be established. Thus, in the present study we analysed radiation effects with a particular focus on low-dose (0.3 -1 Gy) irradiation of EA.hy926 EC with respect to γH2AX foci induction, ROS production and SOD activity.
Discussion
Although considerable progress has been achieved in the understanding of immune modulatory effects of ionising radiation, the underlying molecular mechanisms are presently not fully resolved. In line with that, high dose exposure with single doses exceeding 2 Gy displays a pronounced pro-inflammatory effect [
29] whereas irradiation with single doses below 1 Gy experimentally and clinically reveal anti-inflammatory properties [
2,
4,
6]. This may implicate the involvement of complex mechanisms of DNA damage response and immune modulation differentially operating at different dose levels. In that context, EC may comprise ideal targets for modulatory properties of low-dose and high-dose irradiation exposure due to their crucial role in the regulation of the local inflammatory process both by their ability to recruit leukocytes to the site of local inflammation and by expressing a variety of cytokines/chemokines essential for the inflammatory cascade [
30]. Although recent data imply an involvement of a variety of molecular mechanisms in the anti-inflammatory characteristics of EC following low-dose irradiation [
3], the impact of ROS production to give rise or contribute to these effects in EC remains elusive.
Here we show a linear dose response relationship for γH2AX foci induction at 1 h and 4 h after irradiation irrespective of stimulation of the cells in a pro-inflammatory manner by adding TNF-α. This may indicate that induction of DSBs and early DNA damage repair at low-dose irradiation, at least in EA.hy926 EC, may not be altered in an inflammatory surrounding characteristic for benign diseases treated clinically with LD-RT. Moreover, DNA damage repair response is considered to be linear with dose [
31,
32], which is in agreement with our data on the linearity of γH2AX induction at early times (1 h, 4 h) after irradiation. At later times (24 h), however, we observed a discontinuous dose-response relationship of residual γH2AX foci along with a non-linear detection of ROS with elevated levels following a 0.5 Gy exposure. This further confirms a close relationship between intracellular ROS and the induction of histone γH2AX foci as a marker of DNA damage [
21]. In line with that, cellular ROS production is tightly regulated by coordinated activities of pro-oxidant and anti-oxidant defence mechanisms. To further elucidate mechanisms that may contribute to the non-linear induction of γH2AX foci, we thus analysed the activity of the detoxifying enzyme SOD that dismutates O
2•- into H
2O
2 with the latter to be degraded into H
2O and O
2 by catalase and glutathione peroxidase activity [
33]. Data on the expression of SOD following low-dose irradiation, however, are controversial at present. Similar to our findings, they include a reduction in SOD activity in spleens of healthy BALB/C mice following total body irradiation with a dose of 0.4 Gy [
34]. By contrast, they further comprise reports on increased mRNA expression following irradiation with a dose of 0.2 Gy or 0.5 Gy in splenic tissue of BALB/c or C57BL/6NJcl mice suffering from hepatopathy or cold brain injury [
35,
36]. These results pinpoint to a cell type and environment related regulation of anti-oxidative defence mechanisms that should be addressed in continuative investigations on the role of SOD in low-dose irradiation responses.
Notably, Kang et al. recently demonstrated that ROS induction after treatment of osteosarcoma and mammary epithelial cells with the radiation mimetic neocarzinostatin is, at least in part, mediated by γH2AX overexpression or DNA damage triggered γH2AX accumulation. Moreover, ROS induction by H2AX was abrogated by treatment with NAC, knockdown of the NADP(H) oxidase Nox1 and by a dominant negative Ras-related C3 botulinum toxin substrate 1 (Rac1) mutant (Rac1N17) indicating an involvement of the Nox1 and Rac1 GTPase pathway [
37]. These findings thus point to a more complex and reciprocal regulation of γH2AX and ROS production that may further contribute to a discontinuous appearance of γH2AX foci in EA.hy926 ECs.
In this study we focused on the human endothelial cell line EA.hy926 which has been established by fusion of primary HUVEC with the adenocarcinoma epithelial cell line A549 [
22]. As we can’t exclude that the cancerous fusion partner A549 may influence some properties of EA.hy926 cells as shown for apoptosis induction [
38], we performed exemplary experiments on SOD expression and activity in primary HUVEC, showing a similar dose response relationship (Additional file
2: Figure S2). A comparability is further supported by studies indicating similarities between EA.hy926 ECs and HUVEC in terms of adhesion properties and surface marker expression if stimulated with TNF-α [
39]. Thus, we consider that the EA.hy926 line may comprise a valuable system to investigate the role of SOD and DNA damage response following low-dose exposure.
A discontinuous regulation of ROS production following X-irradiation in a comparable dose range between 0.3 and 0.6 Gy is also reported in stimulated murine RAW 264.7 macrophages when they mount an oxidative burst [
40]. However, as compared to elevated levels at a dose of 0.5 Gy in our investigation, a significant reduction of ROS production was observed in these macrophages. This may further indicate that the variety of regulatory effects observed after low-dose X-ray exposure may reflect different functional consequences that are specific for a given cell type (ECs vs. macrophages) or cellular environment.
Applying DNA binding and transcriptional activity assays, we recently reported on a biphasic activity of the transcription factor NF-κB in stimulated EA.hy926 ECs at 24 h after irradiation with locally elevated values following a 0.5 Gy exposure [
8]. Moreover, NF-κB activation has been shown to be regulated by ROS (H
2O
2) by both the classical (canonical) and by alternative pathways including atypical inhibitor κBα (IκBα) phosphorylation independently of IκB kinase (IKK) [
41]. Although experimentally not proven at present, it is tempting to speculate that elevated levels of ROS at a dose of 0.5 Gy may further contribute to an increased NF-κB activity and as a consequence to increased secretion of the cytokine TGF-β1 and the anti-adhesive efficacy of LD-RT [
7,
9]. This assumption is further supported by a very recent report indicating that ROS comprises a regulator of adhesion molecules very late antigen-4 (VLA-4) and vascular cell adhesion molecule-1 (VCAM-1) mediated monocyte/macrophage adhesion to EC following irradiation with a dose of 0.5 Gy [
42].
In conclusion our data implicate a non-linear regulation of SOD activity and ROS production in EC following irradiation with doses < 1 Gy that may contribute to a discontinuous dose-response relationship of phospho-histone H2AX detection and a putative discontinuous behaviour of DNA damage response. A mechanistic involvement of DNA damage repair mechanisms in the modulation of these non-linear dose response effects remains to be established. However, one may assume that a discontinuous detection of residual γH2AX foci in our investigation is related to the phenomenon of low-dose hyper-radiosensitivity (HRS) and induced radioresistance (IRR), which have been reported for cellular survival at doses below 0.3 Gy and in the dose range of 0.3 Gy to 0.6 Gy, respectively [
14]. In this regard, accumulating evidences exist on a reduced non-homologous end joining (NHEJ) repair response associated with HRS and persistent RAD51 foci, an essential component of the homologous recombination (HR) pathway at late time points after low-dose exposure [
15]. This may indicate that a deregulation of both repair pathways may contribute to the non-linear induction of DSBs. Moreover, future investigations will further address a putative involvement of accumulation of DSBs at stalled replication forks [
43] to contribute to the detection of residual γH2AX following a low-dose exposure especially in S-phase cells.
Competing interests
The authors report no declaration of interest.
Authors’ contributions
ML, SR and SH performed IF, ROS and SOD analyses and prepared the manuscript. CF participated in the design of the study and revised the manuscript critically. CR and FR supervised the analysis, and contributed substantially in preparing the manuscript. All authors read and approved the final manuscript.