Moderate radioprotective role of zeolite in rats

Background/Aim. Exposure of living organisms to ?-radiation results in the overproduction of free radicals. The aim of the study was to test if the subacute administration of micronized zeolite (MZC) accomplishes radioprotective role based on the evaluation of the status of oxidative stress (OS) in the brain and 8-hydroxyguanosine (8-OH-dG) in the plasma of the rats exposed to the single ?-ray irradiation of 2 and/or 10 Gray (Gy). Methods. Wistar rats were on a four-week normal or 5% MZC supplemented diet and afterward exposed to the single ?-ray irradiation of 2 and 10 Gy. Groups of rats were: a) on a normal diet (the control group, and 2Gy and 10Gy groups); b) on 5% MZC supplemented diet (the control group ? MZC, MZC + 2Gy, and MZC + 10Gy groups). We measured malondialdehyde (MDA), glutathione (GSH) total, and activity of total super-oxide dismutase (tSOD) and manganese superoxide dismutase (MnSOD) in vulnerable brain regions (cerebellum, hippocampus, and cerebral cortex) and 8-OH-dG in plasma. Results. Lower MDA was found in the MZC+2Gy and MZC+10Gy groups compared to the 2Gy and 10Gy groups. Activity od total SOD was higher in the MZC+10Gy group than in the 10Gy one. GSH was the highest in the 10Gy group. Compared to the control group, 8-OH-dG was extremely higher in groups radiated with 10 Gy regardless of a diet, but slightly lower in the MZC+2Gy and 2Gy groups. Conclusion. Subacute MZC pretreatment accomplished partial radioprotective effect in irradiated rats compared to non-irradiated rats, based on suppressed SOD activity at 2 Gy, and reduced brain MDA when exposed to 2 Gy and 10 Gy.


Introduction
Exposure of living organisms to γ-radiation results in the overproduction of free radicals through water radiolysis (1) . Generated reactive oxygen and nitrogen species develop oxidative stress (OS) and/or nitrosative stress, reflected in deteriorated cell morphology and physiology, including oxidation/nitration of proteins, lipids, and deoxyribonucleic acid (DNA) (2)(3)(4) . Insufficient cell antioxidative defense and adaptive mechanisms against FRs mostly ended with energy devastation and apoptosis (3) .
Zeolite, a natural clinoptilolite, is a strong, non-selective adsorbents, ion/exchangers, catalysts, detergents or anti-diarrheic with wide range of uses for the treatment of stomach poisoning, poisoning by some harmful agents, alkaloids, mycotoxins, some strains of bacteria, dyspepsia or flatulence (endogenous gasses generated either by-products of digesting particular food or of incomplete digestion in the stomach or small intestine, such as oxygen, ammonium, nitrogen oxide), in humans and animals (5,6) . No selective adsorption and ion/exchange by zeolite limits its prolonged use, because it may decrease the overall bioavailability of nutrients, including essential metals, etc. and so can endanger health. Zeolite does not pass into the systemic circulation after oral intake and remains within the gastrointestinal tract (7) .
The extent of zeolite´s ion/exchange and adsorption (binding) capacity depends on the size of its surface area; thus, we used micronized zeolite (MZC) in our study (7) . The brain is particularly vulnerable to oxidative injury compared to other organs because it spends 15-20 % of the entire body energy (8) . If oxidative stress is going to happen in the body affected by some stimulus (xenobiotics, inflammation, etc.) it may be expected that lipid peroxidation would be developed more in the brain more than in other organs, since the brain has a higher lipid content compared to other body organs and that omega-three polyunsaturated fatty acids are susceptible to oxidation (9)) . Also, the content of transition metals (especially iron and copper) is pretty high in the brain tissue, which additionally contributes to OS development and generation of reactive oxygen species (ROS) concretely, hydroxyl radical (HO • ), through Fenton-like reactions (8,10) . Vulnerable brain regions such as the pyramidal neurons of CA1 and CA3 sectors of the hippocampus, the third layer of the cerebral cortex, and striatum are particularly vulnerable to free radicals toxicity (11) .
Gamma rays induce metabolic oxidative stress and prolonged cell injury by oxidative damage of biomolecules, including, DNA, chromatin materials, lipids, and proteins. 8-Hydroxyguanosine (8-OH-dG) is a well-known biomarker of DNA oxidation (2,3) . Looking for the systemic effect of the applied γ-radiation of rats in terms of OS, we measured the content of 8-OH-dG in plasma.
In this study, we used rats as an animal models to define if the pretreatment of a fourweek diet supplemented with 5% MZC could change radiation responses to applied single γ-ray irradiation of 2 and/or 10 Gray (Gy), based on the evaluation of the tested end-points referring to OS status in the cerebellum, hippocampus and cortex, and plasma 8-OH-dG, after 5 postirradiation days, in rats.

Experimental animals
Adult male Wistar rats (weights of 220 -250 g) were kept under standardized housing conditions (temperature 23±2°C, lighting 12:12 light: dark, light on from 8:00 to 20:00 h) with free access to tap water and a custom pellet rat diet. Suspension of MZC was administered daily by gavage. The Ethical Committee for Experimental Animals, "Vinča" Institute approved this experimental protocol (No. 6/12), which follows the "Guide for the Care and Use of Laboratory Animals"

Experimental design
Wistar rats on normal diet were randomly subdivided into three groups (n=6): control (not treated) group, and 2Gy and 10Gy groups (rats subjected to the single doses radiation of 2 or 10 Gy, respectively); and accordingly, rats on 5 % MZC supplemented diet covered three groups (n=6): MZC, MZC+2Gy, and MZC+10Gy groups. The MZC amount was calculated concerning the quantity of ingested food and rat body mass. The suspension of 0.85-1g of MZC/day (corresponds to 5% of 17-20 g of custom pellet/day) was administered orally, by gavage, during four-weeks (13) . Rats from the control group gained 133.85 ± 24.7 g body weight, and from the MZC group, 126.28 ± 31.42 g during the four week"s diet. No statistically significant differences between them were observed. was between 6th to 10th days (14) . However, Mason et al. established (1989) total-body irradiation sublethal dose of 4 Gy for rats and 5 Gy for mice (ranking radiosensitivity of the organs as follows: lung > hematopoietic system> gastrointestinal tract) within a reasonable time (15) .
According to the abovementioned, we applied a total body irradiation of 0.167Gy/min for 12 minutes, that corresponds to a total non-lethal dose of 2Gy, i.e. 200 rad, and for 60 minutes, that corresponds to a sublethal dose of 10Gy, i.e. 1000 rad, in order to study chosen end-points within the appropriate period of postirradiation time of 5 days (16,17) ..
After the applied treatments, animals continued with the same diet (normal or 5% MZC supplemented diet) for the next 5 five postirradiation days, when they were sacrificed by decapitation (previously anesthetized with an injection of 50 mg of sodium pentobarbital/kg). The brains were removed immediately and stored at -80 °C until analyses were performed.
We measured oxidative stress parameters, including malondialdehyde (MDA), total and manganese superoxide dismutase activities (tSOD and MnSOD), and glutathione (GSH) in the cerebellum, hippocampus, and cerebral cortex and 8-OH-dG in plasma of rats.

Preparation of brain regions for measurement of oxidative/nitrosative stress
The cerebellum, hippocampus, and cerebral cortex were dissected from each frozen brain, and a crude mitochondrial fraction was prepared from each region (18) . Slices of brain structures were transferred separately into a saline solution (0.9 % w/v). Aliquots (1 mL) were placed into a glass tube homogenizer (Tehnica Zelezniki Manufacturing, Slovenia).
Homogenization was performed twice with a Teflon pestle at 800 rpm (1000 g) for 15 min at 4 °C. The supernatant was centrifuged at 2500 g for 30 min at 4 °C. The resulting precipitate was suspended in 1.5 mL of deionized water. The subcellular membranes were constantly mixed in the hypotonic solution for one hour, using a Pasteur pipette. Then, homogenates were centrifuged at 2000 g for 15 min at 4 °C, and the resulting supernatant was used for analysis.
The Lowry method was used to measure protein concentrations in the homogenates of the tested brain regions of rats (19) .

Total superoxide dismutase assay
The activity of SOD (EC 1.15.1.1.; SOD) was measured spectrophotometrically. The principle of the method is related to the sequestration of superoxide anion radical (O 2 •-) by SOD, which disables spontaneous epinephrine auto-oxidation (recorded at 480 nm). The kinetic of sample enzyme activity was followed in a carbonate buffer (50 mM, pH = 10.2, containing 0.1 mM EDTA), after the addition of 10 mM epinephrine (20) . The results were expressed as U tSOD per mg of protein.

Manganese superoxide dismutase measurement
The principle of MnSOD activity measurement (which applies to tSOD, as well) assumes the addition of cyanide anions to block CuZnSOD activity. Samples were prepared in a carbonate buffer (50 mM, pH = 10.2) with the addition of 8 mM KCN, containing 0.1 mM EDTA, and 10 mM of epinephrine (20) . The results were expressed as U MnSOD per mg of proteins.

Malondialdehyde measurement
Lipid peroxidation was measured as the quantity of MDA produced. Upon reaction with thiobarbituric acid, MDA forms a fluorescent red-complex at a ratio of 2:1, whose absorbance is measured at 532 nm (21) .

Glutathione measurement
The amount of GSH present within the tissues was determined by using 5,5-dithiobis-2nitrobenzoic acid (DTNB, 36.9 mg in 10 ml of 100 % methanol) in Tris-HCl buffer (0.4 M, pH=8.9). The intensity of produced yellow-colored p-nitrophenol anion (corresponds to GSH concentration) was spectrophotometrically measured at 412 nm. Brain tissue was prepared in 10 % sulfosalicylic acid for GSH determination (22) .

8-Hydroxyguanosine assay
8-OH-dG was measured in the plasma of rats by using commercial HT 8-OH-dG ELISA Kit II (R&D Systems, Inc. 614 McKinley Place NE Minneapolis, USA).

Statistical analysis
Kruskal-Wallis, posthoc Dunn's tests and Spearman's nonparametric correlations were used for the statistical data analysis using GraphPad Prism, version 5.01. Differences were considered statistically significant at p<0.05. Values were presented graphically as average ± STDEV using GraphPad Prism.

MDA
Decrease of MDA was obtained in the hippocampus (p<0.05) and the cortex (p<0.0001) of rats in the MZC group and in cortex (p<0.0001) of rats in MZC+2Gy and MZC+10Gy groups compared to the C group; and in the cerebellum and the hippocampus (p<0.05) and the cortex (p<0.0001) of rats in the MZC+2Gy group compared to the 2Gy group; and in the cerebellum and the hippocampus (p<0.001) and the cortex (p<0.0001) of rats in the MZC+10Gy group compared to the 10Gy group ( Figure 1).

tSOD
The increase of tSOD was documented in the cortex of rats in 2Gy (p<0.0001), MZC+2Gy (p<0.05) and MZC+10Gy (p<0.0001) groups compared to the C group. In the radiated groups (2Gy, 10Gy), lower tSOD activity was documented in the cortex of rats in the MZC+2Gy group (p<0.05) and higher in the MZC+10Gy group (p<0.0001), respectively.
Also, tSOD activity was lower in the cortex of rats in the 10Gy group (p<0.0001) compared to the 2 Gy group ( Figure 2).

MnSOD
Reduced MnSOD activity (p<0.05) was observed in all examined the tested brain regions of rats in the 10Gy group and the cortex of rats in the MZC+2Gy group (p<0.05) compared to the C group. Comparing to the 2Gy group, a decrease of MnSOD activity was observed in the cortex of rats in the MZC+2Gy group (p<0.0001) and 10Gy [in the cerebellum (p<0.05) and the cortex (p<0.0001)] groups. Contrary, MnSOD activity was higher in the cortex of rats (p<0.05) in the 2Gy group compared to the C group and in the MZC+10Gy group [in the hippocampus (p<0.05) and the cortex (p<0.0001)] compared to the 10Gy group ( Figure 3).

GSH
Higher Also, the GSH level was profoundly higher in the cortex (p<0.0001) of rats in the 10Gy group than in the 2Gy group (Figure 4).

8-OH-dG
Significantly higher values of 8-OH-dG were obtained in the plasma of rats radiated with 10 Gy, regardless of the diet type (MZC+10Gy, 10Gy) (p<0.001), while the decrease was documented in MZC+2Gy and 2Gy (p<0.05) compared to C ( Figure 5).

Figure 5
Spearman's nonparametric correlations data analysis for the tested brain regions and OS parameters is tabular presented ( Table 1). Table 1 Discussion We showed that four-week diet supplemented with 5% MZC per se resulted in a significant systemic shift of red-ox homeostasis towards a reductive potential in the tested brain regions susceptible to OS, based on decreased MDA (in the hippocampus and cortex) and elevated GSH (in the cerebellum, hippocampus, and cortex). The diet supplemented with did not realize protection against DNA oxidation in plasma. Regarding the radioprotective effect of the applied diet supplemented with zeolite in rats, partial results were achieved in the tested brain regions, referring to suppressed SOD activity at 2 Gy and reduced brain MDA levels at 2 and 10 Gy. The zeolite supplemented diet achieved no radioprotection of DNA against oxidation (Figures 1-5).
Assumingly, decline of OS in the brain tested regions occurred due to reduced bioavailability of some nutrients by zeolite, not because of boosted antioxidant defense system: a) zeolite binds metals and gases (by adsorption or through ion/exchange reactions) from food and remains within the alimentary tract after oral intake; b) transition metals may induce ROS overproduction via Fenton reactions, thus, zeolite significantly reduced the yield of this reaction; c) some essential metals are constituents of many antioxidative metalloenzymes, so, zeolite may decrease the availability of the certain metals to be incorporated in the enzymes and make them functional; d) some gasses (nitrogen monoxide and oxygen) liberated from food within the intestine, may contribute to the formation of ROS or reactive nitrogen species (4) .
An irradiated body surface and the applied dose of γ-ray irradiation are equally important and influential factors for the induced postirradiated effects. Short term outcomes (such as an acute radiation syndrome) depend on the exposure dose, while low doses are found to be associated with possible late somatic and long-term genetic effects, unlike large doses of radiation with immediate somatic effects on the body (23) . The human body can probably absorb up to 200 rads (2Gy) acutely without a fatality. Also, the human population can be exposed to 1-10 Gy of γ-irradiation during radiation therapy treatment or radiation accidents or nuclear/radiological terrorism (23,24) . Upon absorption of ionizing radiation, many chemico-biological changes occur in the living cells, including direct structural disruption or indirectly, through interaction with products of water radiolysis (3) .
Often, 30 postirradiated days are taken to determine lethality in mice or rats. Hence, we selected oxidative damage in the brain and plasma DNA as an end-points provoked by the irradiation, imposed prerequisite was to have rats alive, ; therefore, 5-postirradiated days were concluded to be the appropriate period (23) .
Reduced lipid peroxidation was documented by lower brain MDA levels in the tested brain regions in irradiated MZC pretreated rats, compared to irradiated rats on a normal diet (Figure 1). Our previous results have indicated that MZC treatment decreased levels of O 2 •and nitrates in the brain (25) . One-electron reduction reactions between free radicals and unsaturated fatty acids resulted in cell membrane degradation and elevated production of MDA. Though, O 2 •can initiate lipid peroxidation in its protonated form.
However, lipid peroxidation can be triggered more easily by HO • , which is the most potent free radical [easily generated by homolytic cleavage of hydrogen peroxide (H 2 O 2 ) (the product of SOD catalyzed reactions) or generated through Fenton-like reactions that occur between transition metals (in their reduced form) and H 2 O 2 ] (4) . Accordingly to Hill and Switzer (1984), certain brain regions, such as the cortex, striatum, and hippocampus, are highly enriched with nonheme iron, which is catalytically involved in the production of reactive oxygen species (26) . Water radiolysis occurs within cells during radiation (1) . In such circumstances (radiation disease), the overproduction of H 2 O 2 occurs. In Fenton-like reactions, H 2 O 2 easily reacts with low valent transition metals, such as iron-Fe +2 , copper-Cu +1 , manganese-Mn +2 , etc. to form OH • , which spontaneously triggers free radicals` chain reactions with all kind of biomolecules in body. Assumingly, lowered lipid peroxidation in the group of rats on a 5% MZC diet is related to the lower bioavailability of transition metals.
The 5% MZC 4-week diet, per se, did not affect the activity of MnSOD and tSOD (i.e., CuZn-SOD). Herein, we showed that the activity of SOD isoforms is not affected by zeolite subacute intake, but lipid peroxidation supplementation, in rats.
We showed that activities of t-SOD were significantly elevated only in the cortex (not in the cerebellum and the hippocampus) of the rats on normal diet at both doses of γ-ray irradiation, while the activity of MnSOD was significantly elevated only in the cortex at the dose of 2 Gy, and lowered in all three tested brain regions at 10 Gy, compared to the controls. Our results are similar to those of the study of Lee et al. (2013), who observed no changes for SOD and catalase, but GSH in mouse spleen induced by doses of γ-ray irradiation from 0.02 to 0.2 Gy (30) . The disparities in the content of SOD-isoenzymes across the tested brain regions may explain their unequal responses associated with oxidative stress against applied treatments in rats. Also, brain regions are not equally susceptible to a variety of neuronal injuries associated with oxidative stress, for the same reason (Table 1).
Different distribution of the SOD isoforms within the brain and spinal cord tissues and cells is confirmed by confocal laser scanning microscopy and digital photoimaging according to the study of (31) . As reported by Lindenau et al., Cu/Zn-SOD is predominantly localized in astrocytes of the CNS and the motor neurons of the spinal cord (much more than in brain neurons). In lower amounts, Cu/Zn-SOD is present in the nucleus sparing the nucleolus, neuronal perikarya, and in structures of the neuropil.
Mitochondrial MnSOD is more abundant in the brain and the spinal cord neurons than in astroglial cells. The higher susceptibility of the cortex to oxidative stress in rats observed in our study (the activity of SOD isoforms) is in accordance with the study of Melov et al. (1998), who demonstrated that transgenic MnSOD knockout mice easily develop neuronal phenotype and a spongiform degeneration of the cortex and specific brain stem nuclei (32) .
Additionally, the cerebral cortex and striatum are more prone to oxidative damage due to a higher oxygen consumption rate in those regions (33) . The phenomenon of selective neuronal vulnerability is related to the difference in susceptibility of neuronal populations in the CNS to different kinds of stressors (including oxidative and nitrosative stress) that induce neurodegeneration. This phenomenon is not limited to cross-regional differences in the brain, as within a single brain region -such as the hippocampus or the entorhinal cortex -it also manifests in internal, sub-regional differences in relative susceptibility to OS (as it was confirmed by correlation analysis, Table 1). While most brain neurons can tolerate OS, some of them (small pyramidal neurons and the third layer of the cerebral cortex and striatum, the hippocampal CA1 region and cerebellar granule cell layer) are particularly vulnerable to OS (11,34,35) . The susceptibility of the cerebellar neurons to OS and NS might play an important role in the significant loss of these neurons in the aging process (36) .
Activities of both SOD isoforms were significantly higher in rats on the normal diet after radiation by 2 Gy than in those on zeolite supplemented diet and the opposite for 10 Gy. The type of initiated antioxidant defense mechanistic pathway depends on the dose of γ-irradiation, which was confirmed by our results for 2Gy and 10Gy groups (37) .
Additionally, O 2 •can act both as an initiator and a terminator of FRs' chain reaction mediated (3) .
For 2Gy radiation, only in the cortex, the activity of tSOD and MnSOD was higher in rats on the normal diet from the control (p<0.001 and p<0.05 respectively) and those on zeolite supplemented diet (p<0.05 and p<0.001, respectively) (Figures 2 and 3). These results follow the Pathak et al. (2007) study, showing that lower doses of γ-radiation induce the activity of SOD in various organs in rats (38) . According to our results, the dose of 2 Gy of radiation increases the activity of tSOD (in the cortex statistically significant) what implies increased production of H 2 O 2 as well, which may explain higher MDA levels ( Figure 2).
Contrary to that, for 10Gy radiation, the activity of MnSOD were lower in rats on normal diet compared the control (in all tested brain regions, p<0.05) and those on zeolite supplemented diet (in hippocampus, p<0.05 and in cortex, p<0.001), while the activity of tSOD did not differ from the controls and was lower from rats on zeolite supplemented diet only in cortex (p<0.001) (Figures 2 and 3).
The in vitro study of Wu and Navrotsky showed that zeolite binds metals in the following descending order Mn>Zn>magnesium (Mg)>Fe>Cu (39) . Also, the in vitro study of Jacobs and Waite confirmed the strong zeolite affinity for Mn (40) . Our results indicate that zeolite indirectly affects the activity of SOD by binding essential metals (Mn, Cu, and Zn) that are co-factors and constituents of metalloenzymes.
Brain GSH is a confirmed neuromodulator, neurotransmitter, and neurohormone. The sulfhydryl group of cysteine of the tripeptide GSH is responsible for its antioxidant (donor of reducing equivalents) and metal-binding abilities. We showed significantly elevated brain GSH levels after exposure to γ-ray irradiation, at both doses ( Figure 4) (27,41) . These results adhere to the literature, showing that lower doses of radiation cause an increase of GSH levels in many organs, including some adaptive responses (the regeneration pathways) and induction of GSH synthesis-related proteins via the de novo synthesis (42) According to Kawakita et al. (2003) it is known that changed cellular redox signalisation in leads to phosphorylation of various serine/threonine MAP Kinases (Mitogen-Activated Protein Kinases), that further activate different redox-sensitive transcription factors like Nuclear Factor-κB (NF-κB) and Activating protein-1 (AP-1), resulting in the gene expression of various antioxidant defense proteins (enzymatic and no-enzymatic) to overcome the effect of OS-mediated cellular damage (43) . Yamaoka et al. (1998) reported that persistent radiation increases SOD in the rat liver and spleen up to 8-12 weeks after exposure (44) . Additionally, increased induction of GSH by low-dose γ-rays appears to activate immune function, according to Kojima et al. (42,45) .
Seven decades ago, Patt et al. (1949) reported that cysteine (the key amino acid in tripeptide GSH) administered to rats before 800 R of X-rays, significantly increased survival (46) . However, Teshima et al. (2000) found that increase of intracellular GSH induced by low-dose gamma-radiation occurs because of higher expression of mRNA for γ-glutamylcysteine synthetase (γ-GCS), a rate-limiting enzyme of the de novo GSH synthesis pathway, than that of glutathione reductase (42,47) The study of Lee et al. (2013) confirmed that the elevation of GSH at low-dose γ-ray irradiation (0.02 and 0.2 Gy) is accompanied with elevated expression of glutamate-cysteine ligase modifier (not catalytic) subunit, emphasizing that no changes in the expression of thioredoxin occurred in de novo GSH synthesis (30) .
We showed that10 Gy dose of radiation causes oxidative DNA damage in plasma no matter what diet rats were on ( Figure 5) (4,48,49) . Prooxidants such as HO • , excited oxygen, photosensitizers, or ONOO -, produce 8-OH-dG in the reaction with DNA. The study of Floyd RA, Carney JM. (1992) underlined that iron and reactive oxygen-free radical intermediates are involved in oxidative damage of proteins and DNA (50) . Very high plasma 8-OH-dG concentrations obtained in rats subjected to 10 Gy (regardless of diet) are in accordance to Cuttler & Pollycove study (2003), affirming that damaging or lethal cellular effects are observed following high radiation doses, while cellular stimulatory effects happened following low-dose-short-term exposures in the range 0.01 -0.50 Gy (51) .
Based on our results, the applied 5% MZC diet in rats appeared to have no radioprotective effect against oxidative damage of DNA in plasma ( Figure 5).
The correlation analysis confirmed differences across the hippocampus, cerebellum, and cortex responses to applied treatments in rats ( Table 1). Correlation of the data related to lipid peroxidation (MDA) and GSH changes, within the MZC+10 Gy group, shows that both Hipp/Cer and Hipp/Cx, and Cx/Cer similarly responded, respectively. Also, changes of tSOD activity upon radiation of 10 Gy were similar in Hipp/Cer, while the 2 Gy radiation caused similar changes in MDA in Cx/Cer, but the MnSOD activities in Cx/Cer.
Anatomical and physiological characteristics of the tested brain regions (such as localization of antioxidant enzymes, an abundance of transition metals, richness with polyunsaturated free fatty acids, etc.) dictate a profile of oxidative stress response to zeolite diet and against gamma radiation, in rats (31,(33)(34)(35) .

Conclusions
Gamma-ray irradiation of 2 and 10 Gy changes brain red-ox homeostasis and causes oxidative alternations of plasma DNA at higher doses, in rats. Subacute MZC pretreatment accomplished partial radioprotective effect in rats based on reduced brain MDA and activity of SOD, compared to the rats on a normal diet. The cortex appears to be the most susceptible to oxidative stress induced by γ-ray irradiation. Labeling: *-compared to control, #-compared to 10Gy group and &-compared to