Author Correction: DNA damage interactions on both nanometer and micrometer scale determine overall cellular damage.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Proton pencil minibeam irradiation of an in-vivo mouse ear model spares healthy tissue dependent on beam size.

Proton radiotherapy using minibeams of sub-millimeter dimensions reduces side effects in comparison to conventional proton therapy due to spatial fractionation. Since the proton minibeams widen with depth, the homogeneous irradiation of a tumor can be ensured by adjusting the beam distances to tumor size and depth to maintain tumor control as in conventional proton therapy. The inherent advantages of protons in comparison to photons like a limited range that prevents a dosage of distal tissues are maintained by proton minibeams and can even be exploited for interlacing from different beam directions. A first animal study was conducted to systematically investigate and quantify the tissue-sparing effects of proton pencil minibeams as a function of beam size and dose distributions, using beam widths between σ = 95, 199, 306, 411, 561 and 883 µm (standard deviation) at a defined center-to-center beam distance (ctc) of 1.8 mm. The average dose of 60 Gy was distributed in 4x4 minibeams using 20 MeV protons (LET ~ 2.7 keV/µm). The induced radiation toxicities were measured by visible skin reactions and ear swelling for 90 days after irradiation. The largest applied beam size to ctc ratio (σ/ctc = 0.49) is similar to a homogeneous irradiation and leads to a significant 3-fold ear thickness increase compared to the control group. Erythema and desquamation was also increased significantly 3-4 weeks after irradiation. With decreasing beam sizes and thus decreasing σ/ctc, the maximum skin reactions are strongly reduced until no ear swelling or other visible skin reactions should occur for σ/ctc < 0.032 (extrapolated from data). These results demonstrate that proton pencil minibeam radiotherapy has better tissue-sparing for smaller σ/ctc, corresponding to larger peak-to-valley dose ratios PVDR, with the best effect for σ/ctc < 0.032. However, even quite large σ/ctc (e.g. σ/ctc = 0.23 or 0.31, i.e. PVDR = 10 or 2.7) show less acute side effects than a homogeneous dose distribution. This suggests that proton minibeam therapy spares healthy tissue not only in the skin but even for dose distributions appearing in deeper layers close to the tumor enhancing its benefits for clinical proton therapy.

Beam size limit for pencil minibeam radiotherapy determined from side effects in an in-vivo mouse ear model.

Side effects caused by radiation are a limiting factor to the amount of dose that can be applied to a tumor volume. A novel method to reduce side effects in radiotherapy is the use of spatial fractionation, in which a pattern of sub-millimeter beams (minibeams) is applied to spare healthy tissue. In order to determine the skin reactions in dependence of single beam sizes, which are relevant for spatially fractionated radiotherapy approaches, single pencil beams of submillimeter to 6 millimeter size were applied in BALB/c mice ears at a Small Animal Radiation Research Platform (SARRP) with a plateau dose of 60 Gy. Radiation toxicities in the ears were observed for 25 days after irradiation. Severe radiation responses were found for beams ≥ 3 mm diameter. The larger the beam diameter the stronger the observed reactions. No ear swelling and barely reddening or desquamation were found for the smallest beam sizes (0.5 and 1 mm). The findings were confirmed by histological sections. Submillimeter beams are preferred in minibeam therapy to obtain optimized tissue sparing. The gradual increase of radiation toxicity with beam size shows that also larger beams are capable of healthy tissue sparing in spatial fractionation.

Local inhibition of rRNA transcription without nucleolar segregation after targeted ion irradiation of the nucleolus.

Nucleoli have attracted interest for their role as cellular stress sensors and as potential targets for cancer treatment. The effect of DNA double-strand breaks (DSBs) in nucleoli on rRNA transcription and nucleolar organisation appears to depend on the agent used to introduce DSBs, DSB frequency and the presence (or not) of DSBs outside the nucleoli. To address the controversy, we targeted nucleoli with carbon ions at the ion microbeam SNAKE. Localized ion irradiation with 1-100 carbon ions per point (about 0.3-30 Gy per nucleus) did not lead to overall reduced ribonucleotide incorporation in the targeted nucleolus or other nucleoli of the same cell. However, both 5-ethynyluridine incorporation and Parp1 protein levels were locally decreased at the damaged nucleolar chromatin regions marked by γH2AX, suggesting localized inhibition of rRNA transcription. This locally restricted transcriptional inhibition was not accompanied by nucleolar segregation, a structural reorganisation observed after inhibition of rRNA transcription by treatment with actinomycin D or UV irradiation. The presented data indicate that even multiple complex DSBs do not lead to a pan-nucleolar response if they affect only a subnucleolar region.

DNA damage interactions on both nanometer and micrometer scale determine overall cellular damage.

DNA double strand breaks (DSB) play a pivotal role for cellular damage, which is a hazard encountered in toxicology and radiation protection, but also exploited e.g. in eradicating tumors in radiation therapy. It is still debated whether and in how far clustering of such DNA lesions leads to an enhanced severity of induced damage. Here we investigate - using focused spots of ionizing radiation as damaging agent - the spatial extension of DNA lesion patterns causing cell inactivation. We find that clustering of DNA damage on both the nm and µm scale leads to enhanced inactivation compared to more homogeneous lesion distributions. A biophysical model interprets these observations in terms of enhanced DSB production and DSB interaction, respectively. We decompose the overall effects quantitatively into contributions from these lesion formation processes, concluding that both processes coexist and need to be considered for determining the resulting damage on the cellular level.

Optimization of beam arrangements in proton minibeam radiotherapy by cell survival simulations.

PURPOSE: Proton minibeam radiotherapy using submillimeter beam dimensions allows to enhance tissue sparing in the entrance channel by spatial fractionation additionally to advantageous proton depth dose distribution. In the entrance channel, spatial fractionation leads to reduced side effects compared to conventional proton therapy. The submillimeter sized beams widen with depth due to small angle scattering and enable therefore, in contrary to x-ray microbeam radiation therapy (MRT), the homogeneous irradiation of a tumor. Proton minibeams can either be applied as planar minibeams or pencil shaped with an additional possibility to vary between a quadratic and a hexagonal arrangement for pencil minibeams. The purpose of this work is to deduce interbeam distances to achieve a homogeneous dose distribution for different tumor depths and tumor thicknesses. Furthermore, we aim for a better understanding of the sparing effect on the basis of surviving cells calculated by the linear-quadratic model. METHODS: Two-dimensional dose distributions are calculated for proton minibeams of different shapes and arrangements. For a tumor in 10-15 cm depth, treatment plans are calculated with initial beam size of σ0  = 0.2 mm in a water phantom. Proton minibeam depth dose distributions are finally converted into cell survival using a linear-quadratic model. RESULTS: Inter proton beam distances are maximized under the constraint of dose homogeneity in the tumor for tumor depths ranging from 4 to 15 cm and thickness ranging from 0.5 to 10 cm. Cell survival calculations for a 5 cm thick tumor covered by 10 cm healthy tissue show less cell death by up to 85%, especially in the superficial layers, while keeping the cell death in the tumor as in conventional therapy. In the entrance channel, the pencil minibeams result in higher cell survival in comparison to the planar minibeams while all proton minibeam irradiations show higher cell survival than conventional broadbeam irradiation. CONCLUSION: The deduced constraints for interbeam distances simplify treatment planning for proton minibeam radiotherapy applications in future studies. The cell survival results indicate that proton minibeam radiotherapy reduces side effects but keeps tumor control as in conventional proton therapy. It makes proton minibeam, especially pencil minibeam radiotherapy a potentially attractive new approach for radiation therapy.

Live cell imaging of mitochondria following targeted irradiation in situ reveals rapid and highly localized loss of membrane potential.

The reliance of all cell types on the mitochondrial function for survival makes mitochondria an interesting target when trying to understand their role in the cellular response to ionizing radiation. By harnessing highly focused carbon ions and protons using microbeams, we have performed in situ live cell imaging of the targeted irradiation of individual mitochondria stained with Tetramethyl rhodamine ethyl ester (TMRE), a cationic fluorophore which accumulates electrophoretically in polarized mitochondria. Targeted irradiation with both carbon ions and protons down to beam spots of <1 µm induced a near instant loss of mitochondrial TMRE fluorescence signal in the targeted area. The loss of TMRE after targeted irradiation represents a radiation induced change in mitochondrial membrane potential. This is the first time such mitochondrial responses have been documented in situ after targeted microbeam irradiation. The methods developed and the results obtained have the ability to shed new light on not just mitochondria's response to radiation but to further elucidate a putative mechanism of radiation induced depolarization and mitochondrial response.

Chromatin organization revealed by nanostructure of irradiation induced γH2AX, 53BP1 and Rad51 foci.

The spatial distribution of DSB repair factors γH2AX, 53BP1 and Rad51 in ionizing radiation induced foci (IRIF) in HeLa cells using super resolution STED nanoscopy after low and high linear energy transfer (LET) irradiation was investigated. 53BP1 and γH2AX form IRIF with same mean size of (540 ± 40) nm after high LET irradiation while the size after low LET irradiation is significantly smaller. The IRIF of both repair factors show nanostructures with partial anti-correlation. These structures are related to domains formed within the chromatin territories marked by γH2AX while 53BP1 is mainly situated in the perichromatin region. The nanostructures have a mean size of (129 ± 6) nm and are found to be irrespective of the applied LET and the labelled damage marker. In contrast, Rad51 shows no nanostructure and a mean size of (143 ± 13) nm independent of LET. Although Rad51 is surrounded by 53BP1 it strongly anti-correlates meaning an exclusion of 53BP1 next to DSB when decision for homologous DSB repair happened.

Depletion of Histone Demethylase Jarid1A Resulting in Histone Hyperacetylation and Radiation Sensitivity Does Not Affect DNA Double-Strand Break Repair.

Histone demethylases have recently gained interest as potential targets in cancer treatment and several histone demethylases have been implicated in the DNA damage response. We investigated the effects of siRNA-mediated depletion of histone demethylase Jarid1A (KDM5A, RBP2), which demethylates transcription activating tri- and dimethylated lysine 4 at histone H3 (H3K4me3/me2), on growth characteristics and cellular response to radiation in several cancer cell lines. In unirradiated cells Jarid1A depletion lead to histone hyperacetylation while not affecting cell growth. In irradiated cells, depletion of Jarid1A significantly increased cellular radiosensitivity. Unexpectedly, the hyperacetylation phenotype did not lead to disturbed accumulation of DNA damage response and repair factors 53BP1, BRCA1, or Rad51 at damage sites, nor did it influence resolution of radiation-induced foci or rejoining of reporter constructs. We conclude that the radiation sensitivity observed following depletion of Jarid1A is not caused by a deficiency in repair of DNA double-strand breaks.

Superresolution light microscopy shows nanostructure of carbon ion radiation-induced DNA double-strand break repair foci.

Carbon ion radiation is a promising new form of radiotherapy for cancer, but the central question about the biologic effects of charged particle radiation is yet incompletely understood. Key to this question is the understanding of the interaction of ions with DNA in the cell's nucleus. Induction and repair of DNA lesions including double-strand breaks (DSBs) are decisive for the cell. Several DSB repair markers have been used to investigate these processes microscopically, but the limited resolution of conventional microscopy is insufficient to provide structural insights. We have applied superresolution microscopy to overcome these limitations and analyze the fine structure of DSB repair foci. We found that the conventionally detected foci of the widely used DSB marker γH2AX (Ø 700-1000 nm) were composed of elongated subfoci with a size of ∼100 nm consisting of even smaller subfocus elements (Ø 40-60 nm). The structural organization of the subfoci suggests that they could represent the local chromatin structure of elementary DSB repair units at the DSB damage sites. Subfocus clusters may indicate induction of densely spaced DSBs, which are thought to be associated with the high biologic effectiveness of carbon ions. Superresolution microscopy might emerge as a powerful tool to improve our knowledge of interactions of ionizing radiation with cells.-Lopez Perez, R., Best, G., Nicolay, N. H., Greubel, C., Rossberger, S., Reindl, J., Dollinger, G., Weber, K.-J., Cremer, C., Huber, P. E. Superresolution light microscopy shows nanostructure of carbon ion radiation-induced DNA double-strand break repair foci.

A New Nanobody-Based Biosensor to Study Endogenous PARP1 In Vitro and in Live Human Cells.

Poly(ADP-ribose) polymerase 1 (PARP1) is a key player in DNA repair, genomic stability and cell survival and it emerges as a highly relevant target for cancer therapies. To deepen our understanding of PARP biology and mechanisms of action of PARP1-targeting anti-cancer compounds, we generated a novel PARP1-affinity reagent, active both in vitro and in live cells. This PARP1-biosensor is based on a PARP1-specific single-domain antibody fragment (~ 15 kDa), termed nanobody, which recognizes the N-terminus of human PARP1 with nanomolar affinity. In proteomic approaches, immobilized PARP1 nanobody facilitates quantitative immunoprecipitation of functional, endogenous PARP1 from cellular lysates. For cellular studies, we engineered an intracellularly functional PARP1 chromobody by combining the nanobody coding sequence with a fluorescent protein sequence. By following the chromobody signal, we were for the first time able to monitor the recruitment of endogenous PARP1 to DNA damage sites in live cells. Moreover, tracing of the sub-nuclear translocation of the chromobody signal upon treatment of human cells with chemical substances enables real-time profiling of active compounds in high content imaging. Due to its ability to perform as a biosensor at the endogenous level of the PARP1 enzyme, the novel PARP1 nanobody is a unique and versatile tool for basic and applied studies of PARP1 biology and DNA repair.

Proton Minibeam Radiation Therapy Reduces Side Effects in an In Vivo Mouse Ear Model.

PURPOSE: Proton minibeam radiation therapy is a novel approach to minimize normal tissue damage in the entrance channel by spatial fractionation while keeping tumor control through a homogeneous tumor dose using beam widening with an increasing track length. In the present study, the dose distributions for homogeneous broad beam and minibeam irradiation sessions were simulated. Also, in an animal study, acute normal tissue side effects of proton minibeam irradiation were compared with homogeneous irradiation in a tumor-free mouse ear model to account for the complex effects on the immune system and vasculature in an in vivo normal tissue model. METHODS AND MATERIALS: At the ion microprobe SNAKE, 20-MeV protons were administered to the central part (7.2 × 7.2 mm(2)) of the ear of BALB/c mice, using either a homogeneous field with a dose of 60 Gy or 16 minibeams with a nominal 6000 Gy (4 × 4 minibeams, size 0.18 × 0.18 mm(2), with a distance of 1.8 mm). The same average dose was used over the irradiated area. RESULTS: No ear swelling or other skin reactions were observed at any point after minibeam irradiation. In contrast, significant ear swelling (up to fourfold), erythema, and desquamation developed in homogeneously irradiated ears 3 to 4 weeks after irradiation. Hair loss and the disappearance of sebaceous glands were only detected in the homogeneously irradiated fields. CONCLUSIONS: These results show that proton minibeam radiation therapy results in reduced adverse effects compared with conventional homogeneous broad-beam irradiation and, therefore, might have the potential to decrease the incidence of side effects resulting from clinical proton and/or heavy ion therapy.

Nanoscopic exclusion between Rad51 and 53BP1 after ion irradiation in human HeLa cells.

Many proteins involved in detection, signalling and repair of DNA double-strand breaks (DSB) accumulate in large number in the vicinity of DSB sites, forming so called foci. Emerging evidence suggests that these foci are sub-divided in structural or functional domains. We use stimulated emission depletion (STED) microscopy to investigate localization of mediator protein 53BP1 and recombination factor Rad51 after irradiation of cells with low linear energy transfer (LET) protons or high LET carbon ions. With a resolution better than 100 nm, STED microscopy and image analysis using a newly developed analyzing algorithm, the reduced product of the differences from the mean, allowed us to demonstrate that with both irradiation types Rad51 occupies spherical regions of about 200 nm diameter. These foci locate within larger 53BP1 accumulations in regions of local 53BP1 depletion, similar to what has been described for the localization of Brca1, CtIP and RPA. Furthermore, localization relative to 53BP1 and size of Rad51 foci was not different after irradiation with low and high LET radiation. As expected, 53BP1 foci induced by low LET irradiation mostly contained one Rad51 focal structure, while after high LET irradiation, most foci contained >1 Rad51 accumulation.

The influence of the channel size on the reduction of side effects in microchannel proton therapy.

The potential of proton microchannel radiotherapy to reduce radiation effects in the healthy tissue but to keep tumor control the same as in conventional proton therapy is further elucidated. The microchannels spread on their way to the tumor tissue resulting in different fractions of the healthy tissue covered with doses larger than the tumor dose, while the tumor gets homogeneously irradiated. The aim of this study was to evaluate the effect of increasing channel width on potential side effects in the normal tissue. A rectangular 180 × 180 µm(2) and two Gaussian-type dose distributions of σ = 260 µm and σ = 520 µm with an interchannel distance of 1.8 mm have been applied by 20-MeV protons to a 3D human skin model in order to simulate the widened channels and to compare the irradiation effects at different endpoints to those of a homogeneous proton irradiation. The number of protons applied was kept constant at all irradiation modes resulting in the same average dose of 2 Gy. All kinds of proton microchannel irradiation lead to higher cell viability and produce significantly less genetic damage than homogeneous proton irradiation, but the reduction is lower for the wider channel sizes. Our findings point toward the application of microchannel irradiation for clinical proton or heavy ion therapy to further reduce damage of normal tissues while maintaining tumor control via a homogeneous dose distribution inside the tumor.

Live cell imaging at the Munich ion microbeam SNAKE - a status report.

Ion microbeams are important tools in radiobiological research. Still, the worldwide number of ion microbeam facilities where biological experiments can be performed is limited. Even fewer facilities combine ion microirradiation with live-cell imaging to allow microscopic observation of cellular response reactions starting very fast after irradiation and continuing for many hours. At SNAKE, the ion microbeam facility at the Munich 14 MV tandem accelerator, a large variety of biological experiments are performed on a regular basis. Here, recent developments and ongoing research projects at the ion microbeam SNAKE are presented with specific emphasis on live-cell imaging experiments. An overview of the technical details of the setup is given, including examples of suitable biological samples. By ion beam focusing to submicrometer beam spot size and single ion detection it is possible to target subcellular structures with defined numbers of ions. Focusing of high numbers of ions to single spots allows studying the influence of high local damage density on recruitment of damage response proteins.

Reduced side effects by proton microchannel radiotherapy: study in a human skin model.

The application of a microchannel proton irradiation was compared to homogeneous irradiation in a three-dimensional human skin model. The goal is to minimize the risk of normal tissue damage by microchannel irradiation, while preserving local tumor control through a homogeneous irradiation of the tumor that is achieved because of beam widening with increasing track length. 20 MeV protons were administered to the skin models in 10- or 50-µm-wide irradiation channels on a quadratic raster with distances of 500 µm between each channel (center to center) applying an average dose of 2 Gy. For comparison, other samples were irradiated homogeneously at the same average dose. Normal tissue viability was significantly enhanced after microchannel proton irradiation compared to homogeneous irradiation. Levels of inflammatory parameters, such as Interleukin-6, TGF-Beta, and Pro-MMP1, were significantly lower in the supernatant of the human skin tissue after microchannel irradiation than after homogeneous irradiation. The genetic damage as determined by the measurement of micronuclei in keratinocytes also differed significantly. This difference was quantified via dose modification factors (DMF) describing the effect of each irradiation mode relative to homogeneous X-ray irradiation, so that the DMF of 1.21 ± 0.20 after homogeneous proton irradiation was reduced to 0.23 ± 0.11 and 0.40 ± 0.12 after microchannel irradiation using 10- and 50-µm-wide channels, respectively. Our data indicate that proton microchannel irradiation maintains cell viability while significantly reducing inflammatory responses and genetic damage compared to homogeneous irradiation, and thus might improve protection of normal tissue after irradiation.

Recruitment kinetics of DNA repair proteins Mdc1 and Rad52 but not 53BP1 depend on damage complexity.

The recruitment kinetics of double-strand break (DSB) signaling and repair proteins Mdc1, 53BP1 and Rad52 into radiation-induced foci was studied by live-cell fluorescence microscopy after ion microirradiation. To investigate the influence of damage density and complexity on recruitment kinetics, which cannot be done by UV laser irradiation used in former studies, we utilized 43 MeV carbon ions with high linear energy transfer per ion (LET = 370 keV/µm) to create a large fraction of clustered DSBs, thus forming complex DNA damage, and 20 MeV protons with low LET (LET = 2.6 keV/µm) to create mainly isolated DSBs. Kinetics for all three proteins was characterized by a time lag period T(0) after irradiation, during which no foci are formed. Subsequently, the proteins accumulate into foci with characteristic mean recruitment times τ(1). Mdc1 accumulates faster (T(0) = 17 ± 2 s, τ(1) = 98 ± 11 s) than 53BP1 (T(0) = 77 ± 7 s, τ(1) = 310 ± 60 s) after high LET irradiation. However, recruitment of Mdc1 slows down (T(0) = 73 ± 16 s, τ(1) = 1050 ± 270 s) after low LET irradiation. The recruitment kinetics of Rad52 is slower than that of Mdc1, but exhibits the same dependence on LET. In contrast, the mean recruitment time τ(1) of 53BP1 remains almost constant when varying LET. Comparison to literature data on Mdc1 recruitment after UV laser irradiation shows that this rather resembles recruitment after high than low LET ionizing radiation. So this work shows that damage quality has a large influence on repair processes and has to be considered when comparing different studies.

Double-strand break-induced transcriptional silencing is associated with loss of tri-methylation at H3K4.

Epigenetic alterations induced by ionizing radiation may contribute to radiation carcinogenesis. To detect relative accumulations or losses of constitutive post-translational histone modifications in chromatin regions surrounding DNA double-strand breaks (DSB), we developed a method based on ion microirradiation and correlation of the signal intensities after immunofluorescence detection of the histone modification in question and the DSB marker γ-H2AX. We observed after ionizing irradiation markers for transcriptional silencing, such as accumulation of H3K27me3 and loss of active RNA polymerase II, at chromatin regions labeled by γ-H2AX. Confocal microscopy of whole nuclei and of ultrathin nuclear sections revealed that the histone modification H3K4me3, which labels transcriptionally active regions, is underrepresented in γ-H2AX foci. While some exclusion of H3K4me3 is already evident at the earliest time amenable to this kind of analysis, the anti-correlation apparently increases with time after irradiation, suggesting an active removal process. Focal accumulation of the H3K4me3 demethylase, JARID1A, was observed at damaged regions inflicted by laser irradiation, suggesting involvement of this enzyme in the DNA damage response. Since no accumulation of the repressive mark H3K9me2 was found at damaged sites, we suggest that DSB-induced transcriptional silencing resembles polycomb-mediated silencing rather than heterochromatic silencing.

Survival of tumor cells after proton irradiation with ultra-high dose rates.

BACKGROUND: Laser acceleration of protons and heavy ions may in the future be used in radiation therapy. Laser-driven particle beams are pulsed and ultra high dose rates of >109 Gy s⁻¹ may be achieved. Here we compare the radiobiological effects of pulsed and continuous proton beams. METHODS: The ion microbeam SNAKE at the Munich tandem accelerator was used to directly compare a pulsed and a continuous 20 MeV proton beam, which delivered a dose of 3 Gy to a HeLa cell monolayer within < 1 ns or 100 ms, respectively. Investigated endpoints were G2 phase cell cycle arrest, apoptosis, and colony formation. RESULTS: At 10 h after pulsed irradiation, the fraction of G2 cells was significantly lower than after irradiation with the continuous beam, while all other endpoints including colony formation were not significantly different. We determined the relative biological effectiveness (RBE) for pulsed and continuous proton beams relative to x-irradiation as 0.91 ± 0.26 and 0.86 ± 0.33 (mean and SD), respectively. CONCLUSIONS: At the dose rates investigated here, which are expected to correspond to those in radiation therapy using laser-driven particles, the RBE of the pulsed and the (conventional) continuous irradiation mode do not differ significantly.

Spatial dynamics of DNA damage response protein foci along the ion trajectory of high-LET particles.

High-linear energy transfer (LET) ion irradiation of cell nuclei induces complex and severe DNA lesions, and foci of repair proteins are formed densely along the ion trajectory. To efficiently discriminate the densely distributed/overlapping foci along the ion trajectory, a focus recognition algorithm called FociPicker3D based on a local fraction thresholding technique was developed. We analyzed high-resolution 3D immunofluorescence microscopic focus images and obtained the kinetics and spatial development of γ-H2AX, 53BP1 and phospho-NBS1 foci in BJ1-hTERT cells irradiated with 55 MeV carbon ions and compared the results with the dynamics of double-strand break (DSB) distributions simulated using the PARTRAC model. Clusters consisting of several foci were observed along the ion trajectory after irradiation. The spatial dynamics of the protein foci supports that the foci clusters are not formed by neighboring foci but instead originate from the DSB cluster damage induced by high-LET radiations.