The molecular basis for the cellular and tissue responses to IR including DNA repair, cell cycle regulation and tissue or organism responses is considered (1
). The ATM pathway is known to play major roles in DDR to radiation, thus individuals with defects in ATM are extremely radiosensitive (33
). TGFβ signaling also appears to be important for cellular response to radiation (35
), including tissue injury, growth inhibition, fibrosis and apoptosis (32
). The current work has focused on understanding possible interactions between the ATM and TGFβ signaling pathways following different radiation quality exposures in order to vary the contribution of complex DSBs. Smad proteins are key TGFβ type I receptor substrates which can transduce extracellular signals from TGFβ to the nucleus and aid in the transcriptional regulation of certain genes.
Smad proteins have also been studied by several groups (27
) and demonstrated to increase in expression following IR. However, little work to date has been done looking at the relationship between Smad protein kinetics and the kinetics for various DNA DSB repair proteins at DSB sites. Following localized small intestinal irradiation, a progressively increased Smad3 mRNA level was observed but not inhibitory Smad7 (43
). Others have observed persistent increasing of Smad7 expression as a result of IR exposure (44
). Phosphorylation of Smad2 and Smad3 and translocation from cytoplasm to nucleus in the response to IR have been detected in several studies involving rat tissue as well as cell lines (27
). Down-regulation of Smad 1/5/8 activation was observed after 10 Gy of IR (46
). Moreover, tumor suppressor p53 is known as a direct sensor to IR, and the fact that Smad2/3 and p53 physically interact implies that p53 activation might serve as a bridge connecting TGFβ signaling and IR responses (49
). The studies above demonstrated the response of Smad proteins to IR through the activation of TGFβ, while the direct effect of Smad proteins to DNA damage induced by IR such as DNA repair was not been revealed.
Uniquely, we observe co-localization of phosphorylated Smad2 (S465/S467) and Smad7, but not Smad3, with DSB repair proteins (e.g. γH2AX, 53BP1, pATF2 and RAD51) in both human epithelial and fibroblast cells following both γ-ray and high LET particle irradiation at moderate doses. High LET particles resulted in foci tracks containing both pSmad2 and Smad7, and the disappearance of these foci was delayed when compared with γ-rays. Smad7 foci formed promptly after radiation (detected as early as 1 h after IR), while pSmad2 foci were not detectable until 4 h after exposure. The time course of resolution of pSmad2 and Smad7 foci was similar to that of γH2AX and 53BP1 foci. Co-localization of Smad7 with the HRR protein-RAD51 was observed in G2 cells, although Smad7 foci were detected in other phases of the cell cycle as well. In contrast, pSmad2 foci were observed primarily in G1 cells. The observation that pSmad2 is mainly observed in G1 cells might be explained by pSmad2 primarily participating in the NHEJ pathway or playing a role in the G1/S checkpoint in addition to its transcription factor role. The late appearance of pSmad2 needs to be further investigated. The relatively early appearance of Smad7 foci perhaps indicates a possible direct binding to DSB breaks, whereas pSmad2, which is unable to directly bind DNA, is likely to be indirectly localized to DSB sites through interactions with other repair molecules at a later stage, perhaps as a result of chromatin remodeling during repair or an additional role in transcription activation.
ATM signaling is essential for optimal cellular and tissue response to IR. To further characterize the potential role for pSmad2 in DDR, we investigated how ATM kinase activity is related to pSmad2 kinetics. It is well established that ATF2 (Ser490/498) is a phosphorylation target of ATM involved in DDR following IR (12
). To confirm the ATM dependence of pATF2 foci, we used both an ATM kinase inhibitor and AT cells and monitored the formation of radiation-induced pATF2 foci. pATF2 foci were not observed in cells treated with ATM kinase inhibitor nor in the AT cells supporting the ATM dependence of ATF2 activation at the Ser490/498 site. We then used the same strategy to investigate the dependence of pSmad2 foci on ATM kinase activity. Similar to pATF2, complete abrogation of pSmad2 focus formation was observed both by addition of the ATM kinase inhibitor as well as in the ATM mutant cells. Thus, this work reveals for the first time that Smad2 is a phosphorylation target of ATM in response to IR-induced DNA damage.
We further tested the TGFβ dependence of pSmad2 foci using a TGFβR1 inhibitor and noted that pSmad2 foci formation was not diminished with this treatment. Together these data indicate that a fraction of Smad2 phosphorylation is ATM dependent. Furthermore, since pSmad2 and pSmad3 can form heteromeric complexes upon TGFβ stimulation (19
), we also investigated how kinetics of pSmad3 foci compared with pSmad2 following IR. However, unlike pSmad2, which showed a delay in co-localization with DSB proteins, neither total Smad3 nor pSmad3 was observed to form foci and co-localize with other DSB proteins at DSBs. Smad3 co-localization to DSB proteins had been observed previously using DNA damage reagents or at very high doses (>10 Gy) (50
). Additionally, unpublished data from our group and others (27
) have noted that activation of ATM following IR is independent of TGFβRI signaling, and these results would support pSmad2 also having a TGFβ signaling-independent role following radiation-induced damage. Smad7 foci formed promptly after radiation, however, were not inhibited upon ATM inhibitor treatment or diminished in AT cells. However, when cells were pretreated with TGFβR1 inhibitor, Smad7 foci were not visible after radiation, indicating a TGFβ dependence unlike pSmad2 foci.
A schematic model based on our new findings combined with previous understanding of the roles of Smad2 and Smad7 in the DDR, and the crosstalk that occurs between the TGFβ/Smad and ATM response pathways is presented in . It has been shown that p53 physically interacts with Smad2 (49
). Phosphorylated Smad2 is a target of ubiquitin Smurf2 after translocation into nucleus, and based on other’s studies, Smurf2 foci can be observed to co-localize with DSB proteins and stay elevated up to months after high LET radiation (52
). Smad2 could act as a potential tumor suppressor (53
), with loss of the protein resulting in an increase in the risk of cell malignancy. Understanding the mechanisms for Smad2 foci formation for different radiation qualities and doses after IR will aid in understanding the role of Smad2 in radiation effects.
Proposed model for the role of Smad2 and Smad7 in response to IR-induced DSBs and relationship to ATM and TGFβ pathways.
Overall, our study reveals for the first time that two Smad proteins, Smad7 and pSmad2, localize to IR-induced DSBs, albeit with differential kinetics (). We identified that IR-induced phosphorylation of Smad2 is ATM-dependent; whereas Smad7 focus formation is TGFβ1 receptor dependent following radiation exposure. Smad2 and Smad7, as a potential tumor suppressor and oncogene, respectively, are involved in the DNA damage signaling pathway. Finally, our studies revealing a delayed disappearance of pSmad2 and Smad7 foci after high LET particle exposure may also indicate an increased biological effectiveness and carcinogenic risk for high LET radiation (54