Biological factors influencing the RBE of neutrons: implications for their past, present and future use in radiotherapy

Sex-Specific Cognitive Deficits Following Space Radiation Exposure

The radiation fields in space define tangible risks to the health of astronauts, and significant work in rodent models has clearly shown a variety of exposure paradigms to compromise central nervous system (CNS) functionality. Despite our current knowledge, sex differences regarding the risks of space radiation exposure on cognitive function remain poorly understood, which is potentially problematic given that 30% of astronauts are women. While work from us and others have demonstrated pronounced cognitive decrements in male mice exposed to charged particle irradiation, here we show that female mice exhibit significant resistance to adverse neurocognitive effects of space radiation. The present findings indicate that male mice exposed to low doses (≤30 cGy) of energetic (400 MeV/n) helium ions (⁴He) show significantly higher levels of neuroinflammation and more extensive cognitive deficits than females. Twelve weeks following ⁴He ion exposure, irradiated male mice demonstrated significant deficits in object and place recognition memory accompanied by activation of microglia, marked upregulation of hippocampal Toll-like receptor 4 (TLR4), and increased expression of the pro-inflammatory marker high mobility group box 1 protein (HMGB1). Additionally, we determined that exposure to ⁴He ions caused a significant decline in the number of dendritic branch points and total dendritic length along with the hippocampus neurons in female mice. Interestingly, only male mice showed a significant decline of dendritic spine density following irradiation. These data indicate that fundamental differences in inflammatory cascades between male and female mice may drive divergent CNS radiation responses that differentially impact the structural plasticity of neurons and neurocognitive outcomes following cosmic radiation exposure.

Impact of charge collection efficiency and electronic noise on the performance of solid-state 3D microdetectors

Microdosimetry has been traditionally performed through gaseous proportional counters, although in recent years different solid-state microdosimeters have been proposed and constructed for this task. In this paper, we analyze the response of solid-state devices of micrometric size with no intrinsic gain developed by CNM-CSIC (Spain). There are two major aspects of the operation of these devices that affect the reconstruction of the probability distributions and momenta of stochastic quantities related to microdosimetry. For micrometric volumes, the drift and diffusion of the charge carriers gives rise to a partial charge collection efficiency in the peripheral region of the depleted volume. This effect produces a perturbation of the reconstructed pulse height (i.e. imparted energy) distributions with respect to the actual microdosimetric distributions. The relevance of this deviation depends on the size, geometry and operating conditions of the device. On the other hand, the electronic noise from the single-event readout set-up poses a limit on the minimum detectable lineal energy when the microdosimeter size is reduced. This article addresses these issues to provide a framework on the physical constraints for the design and operation of solid-state microdosimeters.

Neutron radiation therapy for advanced thyroid cancers

PURPOSE

The aim of this study was to review institutional outcomes for advanced thyroid cancers treated with fast neutron radiation therapy (FNRT) and photon radiation therapy (RT).

METHODS AND MATERIALS

In all, 62 consecutive patients were analyzed. Fifty-nine had stage IV disease. Twenty-three were treated with FNRT and 39 with photon RT. Median follow-up was 14 months. The primary endpoint was overall survival (OS).

RESULTS

There was no significant difference in median OS between FNRT and photon RT (26 vs 16 months; P = .49). Patients with well-differentiated histologies had superior median OS with photon RT (17 vs 69 months; P = .04). There was a nonsignificant trend toward improved OS with FNRT for medullary and anaplastic histologies.

CONCLUSIONS

Outcomes in this study are in line with historical results. There is an apparent detriment in OS with FNRT for well-differentiated histologies and a trend toward improved OS with medullary and anaplastic histologies that warrants further investigation.

Induction and disappearance of γH2AX foci and formation of micronuclei after exposure of human lymphocytes to ⁶⁰Co γ-rays and p(66)+ Be(40) neutrons

PURPOSE

To investigate both the formation of micronuclei (MN) and the induction and subsequent loss of phosphorylated histone H2AX foci (γH2AX foci) after in vitro exposure of human lymphocytes to either (60)Co γ-rays or p(66)+ Be(40) neutrons.

MATERIALS AND METHODS

MN dose response (DR) curves were obtained by exposing isolated lymphocytes of 10 different donors to doses ranging from 0-4 Gy γ-rays or 0-2 Gy neutrons. Also, γH2AX foci DR curves were obtained following exposure to doses ranging from 0-0.5 Gy of either γ-rays or neutrons. Foci kinetics for lymphocytes for a single donor exposed to 0.5 Gy γ-rays or neutrons were studied up to 24 hours post-irradiation.

RESULTS

Micronuclei yields following neutron exposure were consistently higher compared to that from (60)Co γ-rays. All MN yields were over-dispersed compared to a Poisson distribution. Over-dispersion was higher after neutron irradiation for all doses > 0.1 Gy. Up to 4 hours post-irradiation lower yields of neutron-induced γH2AX foci were observed. Between 4 and 24 hours the numbers of foci from neutrons were consistently higher than that from γ-rays. The half-live of foci disappearance is only marginally longer for neutrons compared to that from γ-rays. Foci formations were more likely to be over-dispersed for neutron irradiations.

CONCLUSION

Although neutrons are more effective to induce MN, the absolute number of induced γH2AX foci are less at first compared to γ-rays. With time neutron-induced foci are more persistent. These findings are helpful for using γH2AX foci in biodosimetry and to understand the repair of neutron-induced cellular damage.

Detrimental effect of fast neutrons on cultured immature rat hippocampal cells: relative biological effectiveness of in vitro cell death indices

This in vitro study compared the detrimental effect and relative biological effectiveness (RBE) of high-linear energy transfer (LET) fast neutrons on rat immature hippocampal cultured cells with those of low-LET γ rays. Immature hippocampal cells were exposed to fast neutrons or γ rays. Cytotoxicity and cell viability were analyzed using a lactate dehydrogenase (LDH)-release assay and a 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) assay, respectively. The cytotoxicity and cell viability with fast neutrons or γ rays varied in a dose-dependent pattern. In the LDH release and MTT assay indices, the RBEs of fast neutrons were approximately 2.35 and 2.42, respectively. Fast neutrons markedly induced apoptotic changes in immature hippocampal cells with increased expression of active caspase-3 and cleaved poly(ADP-ribose) polymerase. Increased cytotoxicity and decreased cell viability in immature hippocampal cells were seen in a dose-dependent pattern after fast-neutron and γ irradiation. Fast neutrons have a higher RBE for cell death indices than γ rays.

Dose escalation in prostate cancer using intensity modulated neutron radiotherapy

BACKGROUND AND PURPOSE

Initial promising results of 3D conformal neutron radiotherapy (3D-CNRT) were subsequently limited by high normal tissue toxicities. It is now possible to deliver intensity modulated neutron radiotherapy (IMNRT). The present work compares photon IMRT, 3D-CNRT and IMNRT for three prostate patients to quantify the benefits of IMNRT.

MATERIALS AND METHODS

We compare updated 3D-CNRT plans, IMNRT plans, and conventional IMRT plans by translating neutron DVHs into effective photon DVHs using the dose dependent radiobiological effectiveness (RBE) for each structure. RBE curves are parameterized for a range of normal tissue and prostate tumor values. Generalized equivalent uniform dose (gEUD) and gEUD in 2Gy fractions (gEUD(2)) is calculated for each structure, plan, and parameterization. Rectal sparing and dose to prostate-GTV are compared for 3D-CNRT, IMNRT, and IMRT as a function of normal tissue and prostate RBE.

RESULTS

The closer the RBE values of prostate tumor and normal tissue, the greater the advantage of IMNRT over 3D-CNRT. The rectal sparing achieved using IMNRT ranged from ∼5% to 13% depending upon the choice of RBE for rectum and the α/β value of prostate tumor. IMNRT may provide a theoretical dose advantage over photon IMRT if the α/β value of prostate is 1.5 and the RBEs of prostate and rectum differ by more than 5%. For higher values of prostate α/β any advantages of IMNRT over IMRT could require that the RBEs of prostate and rectum differ by as much as 20%.

CONCLUSIONS

IMNRT provides a clear normal tissue sparing advantage over 3D-CNRT. The advantage increases when the RBEs of the target structure and the normal tissue are similar. This RBE translation method could help identify clinical sites where the dose sparing advantages of IMNRT would allow for the exploitation of the radiobiological advantages of high-LET neutron radiotherapy.

The tumor microenvironment in non-small-cell lung cancer

The tumor microenvironment (TME) of NSCLC is heterogeneous with variable blood flow through leaky immature vessels resulting in regions of acidosis and hypoxia. Hypoxia has been documented in NSCLC directly by polarographic needle electrodes and indirectly by assessing tissue and plasma hypoxia markers. In general, elevated expression of these markers portends poorer outcomes in NSCLC. Impaired vascularity and hypoxia can lead to increased metastasis and treatment resistance. Compounds that directly target hypoxic cells such as tirapazamine have been tested in clinical trials for NSCLC with mixed results. Preclinical data, however, suggest other ways of exploiting the abnormal TME in NSCLC for therapeutic gain. The inhibition of hypoxia-inducible factor-1alpha or vascular endothelial growth factor may increase local control after radiation. Inhibitors of the epidermal growth factor receptor (EGFR)/phosphatidylinositol 3-kinase (PI3K)/Akt pathway, such as erlotinib or PI-103, may "normalize" tumor vessels, allowing for increased chemotherapy delivery or improved oxygenation and radiation response. To select patients who may respond to these therapies and to evaluate the effects of these agents, a noninvasive means of imaging the TME is critical. Presently, there are several promising modalities to image hypoxia and the tumor vasculature; these include dynamic perfusion imaging and positron emission tomography scanning with radiolabled nitroimidazoles.

Dose-response and relative biological effectiveness of fast neutrons: induction of apoptosis and inhibition of neurogenesis in the hippocampus of adult mice

PURPOSE

Our study compared the effects of high linear energy transfer (LET) fast neutrons on the induction of apoptosis and reduction of neurogenesis in the hippocampus of adult ICR mice with those of low-LET (60)Co gamma-rays, to evaluate the relative biological effectiveness (RBE) of fast neutrons in the adult hippocampal dentate gyrus (DG).

MATERIALS AND METHODS

The mice were exposed to 35 MeV fast neutrons or (60)Co gamma-rays. We evaluated acutely the incidence of apoptosis and expression of Ki-67 (a protein marker for cell proliferation originally defined by the monoclonal antibody Kiel-67) and doublecortin (DCX: an immature progenitor neuron marker) in the hippocampus after a single whole-body irradiation.

RESULTS

The number of terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick end-labelling (TUNEL)-positive apoptotic nuclei in the DG increased and both Ki-67- and DCX-positive cells declined in a dose-dependent pattern, with fast neutrons or gamma-rays. In the hippocampus, which showed an apoptosis frequency between 2 and 8 per DG, the RBE of fast neutrons was approximately 1.9. Additionally, the inhibitory effects of fast neutrons on the expression frequencies of Ki-67 (4-8) and DCX (8-32) were approximately 3.2 and 2.5 times, respectively, the effects of gamma-rays at the same dose.

CONCLUSIONS

Increased apoptotic cell death and decreased neurogenesis in the hippocampal DG were seen in a dose-dependent pattern after exposure to fast neutrons and gamma-rays. In addition, the different rate of hippocampal neurogenesis between different radiation qualities may be an index of RBE.

The radiosensitizing effect of Ku70/80 knockdown in MCF10A cells irradiated with X-rays and p(66)+Be(40) neutrons

Background

A better understanding of the underlying mechanisms of DNA repair after low- and high-LET radiations represents a research priority aimed at improving the outcome of clinical radiotherapy. To date however, our knowledge regarding the importance of DNA DSB repair proteins and mechanisms in the response of human cells to high-LET radiation, is far from being complete.

Methods

We investigated the radiosensitizing effect after interfering with the DNA repair capacity in a human mammary epithelial cell line (MCF10A) by lentiviral-mediated RNA interference (RNAi) of the Ku70 protein, a key-element of the nonhomologous end-joining (NHEJ) pathway. Following irradiation of control and Ku-deficient cell lines with either 6 MV X-rays or p(66)+Be(40) neutrons, cellular radiosensitivity testing was performed using a crystal violet cell proliferation assay. Chromosomal radiosensitivity was evaluated using the micronucleus (MN) assay.

Results

RNAi of Ku70 caused downregulation of both the Ku70 and the Ku80 proteins. This downregulation sensitized cells to both X-rays and neutrons. Comparable dose modifying factors (DMFs) for X-rays and neutrons of 1.62 and 1.52 respectively were obtained with the cell proliferation assay, which points to the similar involvement of the Ku heterodimer in the cellular response to both types of radiation beams. After using the MN assay to evaluate chromosomal radiosensitivity, the obtained DMFs for X-ray doses of 2 and 4 Gy were 2.95 and 2.66 respectively. After neutron irradiation, the DMFs for doses of 1 and 2 Gy were 3.36 and 2.82 respectively. The fact that DMFs are in the same range for X-rays and neutrons confirms a similar importance of the NHEJ pathway and the Ku heterodimer for repairing DNA damage induced by both X-rays and p(66)+Be(40) neutrons.

Conclusions

Interfering with the NHEJ pathway enhanced the radiosensitivity of human MCF10A cells to low-LET X-rays and high-LET neutrons, pointing to the importance of the Ku heterodimer for repairing damage induced by both types of radiation. Further research using other high-LET radiation sources is however needed to unravel the involvement of DNA double strand break repair pathways and proteins in the cellular response of human cells to high-LET radiation.

Lack of adaptive response of gamma radiation for protection against neutron-induced teratogenesis

BACKGROUND

Although there are some reports on neutron teratology, there is little information on the adaptive response of gamma radiation for protection against neutron-induced teratogenesis. This study examined whether or not a low dose of gamma radiation can induce an adaptive response in mouse fetuses exposed to a subsequent dose of neutrons in vivo.

METHODS

Pregnant ICR mice were exposed to a priming dose of 0.3 Gy (0.9 Gy/min) of gamma rays on day 10.5 of gestation and challenged with 0.8 Gy (0.94 Gy/minute) of neutrons 24 h later. The mice were sacrificed on day 18.5 of gestation. The fetuses were examined for mortality, growth retardation, and other morphologic abnormalities.

RESULTS

The tail length in the 0.3 Gy of gamma rays + 0.8 Gy of neutrons group was significantly shorter than in the 0.8 Gy of neutrons group. Although there was no significant difference compared with the 0.8 Gy of neutrons group, the number of live fetuses in the 0.3 Gy of gamma rays +0.8 Gy of neutrons group was lower. There was no evidence of primed exposure-related reductions in the malformed fetuses. Although there was no significant difference compared with the unprimed group, the number of malformed offspring in the primed group was higher. Furthermore, the incidence of kinked tail and adactyly was significantly higher in the primed mice than in the unprimed mice.

CONCLUSIONS

Overall, this study shows that exposure to 0.3 Gy of gamma rays failed to induce an adaptive response of fetogenesis to a neutron challenge dose.

Relative biological effectiveness of fast neutrons in a multiorgan assay for apoptosis in mouse

This study compared the effects of high linear energy transfer (LET) fast neutrons on the induction of apoptosis in several tissue types (hair follicle, intestine crypt, testis) of ICR mouse exposed to low LET 60Co gamma-rays. The changes that occurred from 0 to 24 h after exposing the mice to either 2 Gy of gamma-rays (2 Gy/min) or 0.8 Gy of neutrons (94 mGy/min, 35 MeV) were examined. The maximum frequency of apoptosis was observed at 8 or 12 h after irradiation. The mice that had received 0-8 Gy of gamma-rays or 0-1.6 Gy of neutrons were examined 8 h after irradiation. The best-fitting dose-response curves were linear-quadratic, and there was a significant relationship between the number of apoptotic cells and the dose. The stained products in the TUNEL-positive cells or bodies correlated with the typical morphologic characteristics of apoptosis observed by optical microscopy. In the follicles showing an apoptosis frequency between 2 and 14 per hair follicle, the relative biological effectiveness (RBE) of the neutrons in the small and large follicles was 2.09 +/- 0.31 and 2.15 +/- 0.18, respectively. In the intestine crypts showing an apoptosis frequency between 1 and 3 per crypt, the RBE of the neutrons was 4.03 +/- 0.06 and 3.87 +/- 0.04 in the base and total crypts, respectively. The RBE of the neutrons in the seminiferous tubule showing an apoptosis frequency between 0.5 and 2 per tubule was 5.18 +/- 0.06. The results determined the time-response relations and the RBE for fast neutron-induced apoptosis in several organs at the same time. The differences in RBE observed between the high and low LET radiation and it is believed that the difference in the DSB repair capacity in hair follicle, intestine crypt, and seminiferous tubule cells plays a role in determining the RBE of the high-LET radiation for the induced apoptotic cell formation.

Relative biological effectiveness of fast neutrons for apoptosis in mouse hair follicles

This study compared the effects of high linear energy transfer (LET) fast neutrons on the induction of apoptosis in the hair follicles of ICR mice with those of low LET ⁶⁰Co γ-rays. The changes that occurred from 0 to 24 h after exposing the mice to either 2 Gy of γ-rays (2 Gy/min) or 0.8 Gy of neutrons (94 mGy/min, 35 MeV) were examined. The maximum frequency was found at 12 h (γ-rays) or 8 h (neutrons) after irradiation. The mice that received 0-8 Gy of γ-rays or 0-1.6 Gy of neutrons were examined 8 h after irradiation. The dose-response curves were analyzed using the best-fit curve model. The dose-response curves were linear-quadratic, and a significant relationship was found between the frequency of apoptotic cells and the dose. The morphological findings in the irradiated groups were typical apoptotic fragments in the matrix region of the hair follicle, but the spontaneous existence of apoptotic fragments was rarely observed in the control group. In the presence of an apoptosis frequency between 2 and 14 per follicle, the relative biological effectiveness values of neutrons in small and large follicles were 2.09 ± 0.30 and 2.15 ± 0.18, respectively.

Gene therapy vectors containing CArG elements from the Egr1 gene are activated by neutron irradiation, cisplatin and doxorubicin

Combining gene therapy with radiotherapy and chemotherapy holds potential to increase the efficacy of cancer treatment, while minimizing side effects. We tested the responsiveness of synthetic gene promoters containing CArG elements from the Early Growth Response 1 (Egr1) gene after neutron irradiation, doxorubicin and cisplatin. Human MCF-7 breast adenocarcinoma and U373-MG glioblastoma cells were transfected with plasmids containing CArG promoters controlling the expression of the green fluorescent protein (GFP). Exposing the cells to neutrons, doxorubicin or cisplatin resulted in a significant induction of transgene expression. Therapeutic advantage was demonstrated by replacing the reporter with the herpes simplex virus thymidine kinase (HSVtk), able to convert the prodrug ganciclovir (GCV) into a cytotoxin. A 1.3 Gy neutron dose caused 49% growth inhibition in MCF-7 cells, which increased to 63% in irradiated CArG-HSVtk-transfectants treated with GCV. Exposure to 0.5 microM cisplatin or 0.01 microM doxorubicin induced a growth inhibition of 25-30% in MCF-7 cells. In the presence of GCV, this value increased to 65-70% in cells transfected with the CArG promoter constructs driving the expression of HSVtk. These data indicate that combining CArG-mediated HSVtk/GCV suicide gene therapy with radio- and chemotherapy can enhance antitumor toxicity, and validates future in vivo investigations.

Detection of oxidative stress induced by low- and high-linear energy transfer radiation in cultured human epithelial cells

A standardized dichlorofluorescin (DCF) fluorometric assay capable of measuring radiation-induced oxidative stress was used to determine the effectiveness of protons and high-mass, high-atomic number (Z) and high-energy (HZE) particles to produce oxidative stress in vitro. Protons were found to be about equally as effective as X rays in the generation of oxidative stress in cultured cells. However, 56Fe-ion beams with energies of 1 GeV/nucleon and 5 GeV/nucleon were less effective than X rays or gamma rays in inducing dichlorofluorescin (DCFH) oxidation. The relatively lower slope values for the dose responses of HZE-particle radiation-induced DCFH oxidation indicate that the sensitivity of the DCF fluorometric assay is probably dependent on the linear energy transfer (LET) of the radiation beam.

Mutagenic effect of low energy neutrons on human chromosome 11

PURPOSE

The shape of the dose-effect curve for neutrons, i.e. the question as to whether the curve is linear or supralinear in the low-dose region, is still not clear. Therefore, the mutagenic effect of very low doses of low-energy neutrons was determined.

MATERIALS AND METHODS

Human-hamster hybrid A(L) cells contain human chromosome 11, which expresses the membrane protein CD59. This membrane protein can be detected immunologically and quantified by flow cytometry. The A(L) cells were irradiated with neutrons of 0.565, 2.5 or 14.8 MeV and the results were compared with those after 200 kVp X-rays. Before irradiation, cells spontaneously mutated in the CD59 gene were removed by magnetic cell sorting (MACS).

RESULTS

The relative biological effectiveness (RBE) for CD59 mutation induction was 19.8 (+/-2.7) for 0.565 MeV, 10.2 (+/-1.9) for 2.5 MeV, and 10.2 (+/-1.6) for 14.8 MeV neutrons. Linear mutation responses were obtained with all radiations except for 14.8 MeV neutrons where a supralinear curve may be a better fit. The deletion spectrum of mutated cell clones showed 29 Mbp deletions on average after irradiation with 0.069 Gy of 0.565 MeV neutrons. This scale of deletions is similar to that after 3 Gy 100 kV X-rays (=34 Mbp). For 50% cell survival, the RBE of the neutrons was 11 compared with 200 kV X-rays.

CONCLUSIONS

Neutrons of low energies (0.565 or 2.5 MeV) produce a linear dose-response for mutation in the tested dose range of 0.015-0.15 Gy. The neutron curve of 14.8 MeV can be approximated by a curvilinear or linear function.

Energy-dependent RBE of neutrons to induce micronuclei in root-tip cells of Allium cepa onion irradiated as dry dormant seeds and seedlings

The relative biological effectiveness (RBE) of various energy neutrons produced from a Schenkel-type accelerator at the Research Institute for Radiation Biology and Medicine, Hiroshima University (HIRRAC), compared with 60Co gamma-ray radiation was determined. The neutron radiations and gamma-ray radiation produced good linear changes in the frequency of micronuclei induced in the root-tip cells of Allium cepa onion irradiated as dry dormant seeds (seed assay) and seedlings (seedling assay) with varying radiation doses. Therefore the RBE for radiation-induced micronuclei can be calculated as the ratio of the slopes of the fitted linear dose response for the neutron radiations and the 60Co gamma-ray radiation. The RBE values by seed assay and seedling assay decreased to 174 +/- 7, from 216 +/- 9, and to 31.4 +/- 1.0, from 45.3 +/- 1.3 (one standard error), respectively, when neutron energies increased to 1.0 MeV, from 0.2 MeV, in the present study. Furthermore, the ratio of the micronucleus induction rates of seed assay to seedling assay by gamma-ray radiation was much lower than that by neutron radiations.

Micronucleus response of human glioblastoma and neuroblastoma cells toward low-LET photon and high-LET p(66)/Be neutron irradiation

The identification of photon resistant tumors that are sensitive to neutrons is still an unresolved problem, and no radiobiological criteria have been developed that could help the selection of patients for neutron therapy. The micronucleus (MN) assay has been evaluated for this purpose in a panel of human glioblastoma and neuroblastoma cell lines spanning a wide range of photon sensitivities defined by mean inactivation doses ([Latin capital letter D with macron above][gamma]) of 1.25-3.21 Gy. We show that the relative biologic effectiveness (RBE) of the p(66)/Be neutrons is significantly correlated with inherent photon sensitivity (r = 0.89, p < 0.01), indicating that the panel of cell lines used is suitable to study the differential biologic response to neutrons and photons. We find that p(66)/Be neutrons are 1.43 to 5.29 times more effective per unit dose in inducing micronuclei than 60Co [gamma]-rays. Surprisingly, cells that are inherently photon resistant tend to show a higher yield of micronuclei following exposure to either photons or neutrons, but no significant correlation could be demonstrated. However, RBE values based on micronucleus yield were found to strongly correlate with RBE values derived from cell survival data (r = 0.91, p < 0.01). It is concluded that although micronucleus yield does not reflect intrinsic sensitivity to either photons or neutrons, the strong correlation between RBE calculated from micronucleus formation and RBE derived from cell survival demonstrates that the micronucleus endpoint has a potential for detecting photon resistant cells that show increased sensitivity to neutrons.

Relative biological effectiveness of fission neutrons for producing micronuclei in the root-tip cells of onion seedlings after irradiation as dry seeds

The relative biological effectiveness (RBE) of mixed neutron and gamma-ray radiation emitted at a 252Cf source at the Research Institute for Radiation Biology and Medicine, Hiroshima University, compared with 60Co gamma-ray radiation was determined. The tissue-absorbed dose contribution of the accompanying gamma radiation was about 35.7% to the total tissue-absorbed dose from the 252Cf mixed radiation. The 252Cf mixed radiation and 60Co gamma rays produced approximate linear changes in the frequency of micronuclei induced in root-tip cells of Allium cepa L. onion seedlings after irradiation as dry dormant seeds with varying absorbed doses in onion seeds. Therefore, the RBE for radiation-induced micronuclei was calculated as the ratio of the slopes for the 252Cf mixed radiation and the 60Co gamma rays. The deduced RBE value of 252Cf mixed radiation to 60Co gamma rays to induce micronuclei in dry dormant onion seed cells was about 90.5 +/- 3.6 (+/- 1sigma); the RBE of neutrons from the 252Cf mixed radiation was about 150 +/- 6 (+/- 1sigma). Furthermore, the sensitivity ratio of the induction rate of micronuclei in dry dormant seeds to that in seedlings by neutrons from 252Cf mixed radiation was significantly different from that by 60Co gamma rays. From these results, we concluded that the repair efficiency of DNA damage induced by neutrons may be different from that by gamma rays.