J. Radiat. Res., 50, 241–252 (2009) Regular Paper

Microarray Analysis of Differentially Expressed in the Kidneys and Testes of Mice after Long-term Irradiation with Low-dose-rate γ-rays

Keiko TAKI1, Bing WANG1, Tetsuo NAKAJIMA1, Jianyu WU1, Tetsuya ONO2, Yoshihiko UEHARA2, Tsuneya MATSUMOTO3, Yoichi OGHISO3, Kimio TANAKA3, Kazuaki ICHINOHE3, Shingo NAKAMURA3, Satoshi TANAKA3, Junji MAGAE4, Ayana KAKIMOTO1 and Mitsuru NENOI1*

Kidney/Testis/Low-dose-rate radiation/Microarray/Mitochondrial oxidative phosphorylation. Measuring global expression using cDNA or oligonucleotide microarrays is an effective approach to understanding the complex mechanisms of the effects of radiation. However, few studies have been carried out that investigate in vivo after prolonged exposure to low-dose-rate radia- tion. In this study, C57BL/6J mice were continuously irradiated with γ-rays for 485 days at dose-rates of 0.032–13 μGy/min. Gene expression profiles in the kidney and testis from irradiated and unirradiated mice were analyzed, and differentially expressed genes were identified. A combination of pathway analysis and hierarchical clustering of differentially expressed genes revealed that expression of genes involved in mitochondrial oxidative phosphorylation was elevated in the kidney after irradiation at the dose-rates of 0.65 μGy/min and 13 μGy/min. Expression of cell cycle-associated genes was not profoundly modulated in the kidney, in contrast to the response to acute irradiation, suggesting a threshold in the dose-rate for modulation of the expression of cell cycle-related genes in vivo following exposure to radiation. We demonstrated that changes to the gene expression profile in the testis were largely different from those in the kidney. The categories “DNA metabolism”, “response to DNA damage” and “DNA replication” overlapped significantly with the clusters of genes whose expression decreased with an increase in the dose-rate to the testis. These observations provide a fundamental insight into the organ- specific responses to low-dose-rate radiation.

radiation on human health. A number of these studies have INTRODUCTION suggested there were changes in the immune system1) DNA repair rate2) and aberration frequency.3) However, There is increasing concern about the biological effects of the effects of low-dose-rate radiation are still controversial low-dose-rate radiation. Epidemiological studies in naturally because many confounding factors, including lifestyle, smok- high-background-radiation areas (HBRAs) have provided ing habits, and dietary habits, may have affected the results. opportunities to directly observe the effects of low-dose-rate Animal studies under controlled conditions are therefore important in order to compensate for the uncertainties of epidemiological studies. Tanaka et al.4) have investigated the *Corresponding author: Phone: +81-43-206-3082, life shortening of mice after long-term (about 400 days)

Fax: +81-43-255-6497, irradiation with γ-rays at extremely low-dose-rates in the range E-mail: [email protected] μ 1Radiation Effect Mechanisms Research Group, National Institute of of 0.038-16 Gy/min, and reported that the lifespan of female Radiological Sciences, 9-1, Anagawa-4-chome, Inage-ku, Chiba 263-8555 mice irradiated with 0.83 μGy/min and mice of both sexes Japan; 2Graduate School of Medicine, Tohoku University, 2-1 Seiryou- irradiated with 16 μGy/min were significantly shortened. 3 machi, Aoba-ku, Sendai-shi 980-8575 Japan; Institute for Environmental In order to generalize animal data to humans, a study of Sciences, 1-7, Ienomae, Obuchi, Rokkasho-mura, Kamikita-gun, Aomori the mechanism of radiation effects is required. Determina- 039-3212 Japan; 4Nuclear Technology Research Laboratory, Radiation safety Research Center, Central Research Institute of Electric Power tion of global gene expression patterns using cDNA or Industry, 2-11-1, Iwadokita, Komae-shi, Tokyo 277-0861 Japan. oligonucleotide microarrays is an effective approach to doi:10.1269/jrr.09011 understanding the complex mechanisms of radiation effects.

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp 242 K. Taki et al.

Alteration of gene expression profiles in vivo after whole for 485 days. The kidney and testis have been classified as body irradiation of mice has been investigated by Amundson relatively radio-sensitive organs,13) and the likelihood of et al.5) where they reported that only a few genes were com- detecting effects of such low-dose-rate radiation was thought monly regulated in the thymus, spleen and liver following to be high in these organs. Our study is expected to provide exposure to 0.2–2 Gy of γ–rays. On the basis of this obser- insight into the mechanisms underlying the biological vation, they concluded that gene regulation was highly response to low-dose-rate radiation and normal tissue injuries. tissue-specific following γ-irradiation. Markedly different gene expression responses between the kidney and the brain MATERIALS AND METHODS were also reported by Zhao et al.6) who observed altered expression of genes related to transcription/translation and Mice and irradiation oxidation/reduction in the kidney after 10 Gy of whole-body Male specific pathogen free (SPF) C57BL/6J mice, 7 to 8 irradiation.6) It has been also reported that genes associated weeks of age, were obtained from an animal breeding facil- with inflammation, proliferation, metabolism, migration/ ity (CLEA Japan, Tokyo). Mice were not littermate. We used invasion, and cytoskeleton/microtubules were modulated in male mice because the data from female mice are considered the kidney 10 – 30 weeks after irradiation with 16 Gy of γ– to be perturbed by an estradiol cycle. Irradiation with 137Cs rays.7) These studies provided basic insight into the patho- γ-rays was carried out in a facility at the Institute for genesis of late injuries of the kidney. In the brain, differential Environmental Sciences for 485 days (22 h/day) at dose- modulation of genes associated with ribosomal , rates of 0.032 μGy/min, 0.65 μGy/min, and 13 μGy/min. electron transport, and intracellular signaling involving G The total dose after irradiation at these dose-rates was 21 proteins, TGF-β, and Wnt were observed after irradiation mGy, 420 mGy, and 8000 mGy, respectively. Dosimetry and with 20 Gy of X-rays.8) DNA replication/repair, prolifera- maintenance of the mice were carried out in the same tion/apoptosis, cell cycle, and RNA processing genes were manner as described in Tanaka et al..4) We examined 12 reported to be altered in bone marrow following whole-body mice (3 per each dose-rate condition, including the control). exposure of mice to 6.5 Gy of γ–rays.9) Other results indicate The number of mice per dose-rate condition (n = 3) was that the response of the liver to the internal exposure to α– chosen because this is the minimum sample size needed for particles is characterized by up-regulation of genes related statistical analysis. Each mouse was subjected to dissection, to transcription and down-regulation of genes associated during which it was observed that the gross appearance was with signal transduction.10) The irradiated liver is also normal without organ hypertrophy, neoplasia, and/or hair suggested to be in an inflammatory state characterized by loss. All experiments were conducted according to Japanese up-regulation of positive acute phase proteins.10) Although legal regulations and followed the Guidelines for Animal the mouse strains and radiation doses used were not Experiments of the Institute for Environmental Sciences. identical, these studies suggest that the genes regulated by radiation differ between tissues. Preparation of RNA and hybridization The dose-rate effect has been assessed in vitro by Sokolov Mice were euthanized immediately after the termination et al.11) using human normal fibroblasts irradiated with 1 Gy of irradiation, and the kidney and testis were removed and of γ-rays at the dose-rate of 1 Gy/min or 0.045 Gy/min. They stored at –80°C in RNAlater (Ambion; Aplied Biosystems, observed that approximately one-third of genes were differ- Foster City, CA). Total RNA was extracted using an RNeasy entially expressed between these dose-rate conditions. Mini Kit and RNase-Free DNase Set (Qiagen, Valencia, Amundson et al.12) also investigated the dose-rate effect CA). The quality of total RNA samples was assessed using using human myeloid leukemia cells irradiated with γ–rays an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa in the dose-rate range of 0.0028–2.9 Gy/min. It was reported Clara, CA). From each sample, 150 ng of the total RNA was that there exist two classes of low-dose-rate responders; one labeled using an Illumina RNA Amplification Kit (Ambion; was genes induced in a dose-rate dependent fashion, while Aplied Biosystems, Foster City, CA). A total of 1.5 μg of the other was genes with dose-rate independent induction. biotin-labeled cRNA was hybridized for 18 h to the microarray Apoptosis-related genes were predominantly included in the (Sentrix Mouse-6 v1.0 Expression BeadChips, Illumina, San dose-rate dependent gene group, and the majority of genes Diego, CA). RNA samples from each mouse were not in the dose-rate independent gene group were involved in pooled but instead hybridized to the array separately. The cell cycle regulation. hybridized, biotinylated cRNA was detected with Streptavidin- In contrast to the accumulating data on global gene Cy3 (GE, Fairfield, CT) and quantitated using a BeadStation expression in vitro after exposure to low-dose-rate radiation, 500GX-WG Systems scanner (Illumina). Sentrix Mouse-6 few studies have been carried out in vivo. In this study, we v1.0 Expression BeadChips are constructed from small analyzed alteration of gene expression profiles in the kidney beads coated with 46,000 species of oligonucleotide probes. and testis of C57BL/6J mice continuously irradiated with γ- The beads are so densely arrayed on the surface of the rays at the similar dose-rates to those used by Tanaka et al.4) silicon substrate that the signal from one gene is measured

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp Gene Expression after Low-dose-rate Irradiation 243

30 times, on average, across the array. ter analysis in GeneSpring 7.3.1 was used to identify genes with similar expression patterns to reveal the common func- Data Analysis tions of the modulated genes. False positive genes, which are The microarray data were analyzed using GeneSpring expected to be included, were assumed not to have a large 7.3.1 (Agilent Technologies, Santa Clara, CA). Gene expres- sion data were normalized in two ways: “per chip normal- ization” and “per gene normalization”. For “per chip normalization,” all expression data on an array were normal- ized to the 50th percentile of all values on that array. For “per gene normalization,” the data for a given gene were normalized to the median expression level of that gene across all samples. Quality control of the data was performed as schematically represented in Fig. 1A. A whole sample showing an anomalous distribution of signal intensity was excluded from the analysis. Figure 1B shows the distri- bution of the normalized signal intensity obtained from the kidney gene analysis. It can be seen that one of the samples in the 0.032 μGy/min group, designated by an asterisk (*), exhibited a distinct distribution of normalized signal inten- sities compared to those of other samples for an unknown reason. This sample was simply excluded from this study. Next, data containing large statistical errors and/or inter- individual differences were excluded from the analysis. For this purpose, we used the “detection score,” which is a sta- tistical measure computed based on the Z-value of a gene relative to the Z-value of the negative controls.14) In our anal- ysis, the genes whose detection score was less than 0.99 either for 2 samples out of 3 in the unirradiated control group, or for 6 samples out of 8 in the irradiated group, were excluded. In addition, genes were excluded for which the stan- dard deviation within each of the groups (unirradiated, and irradiated at 0.032 μGy/min, 0.65 μGy/min, and 13 μGy/min) exceeded approximately 10% of the mean value. As a result, 4061 and 7893 genes, out of 46000 transcripts contained on the array, remained in the analyses of kidneys and testes, respectively. Figure 1C shows scatter plots of signal intensity in two of the kidney samples from the unirradiated group before quality control (a) and after removal of outlier data (b). It can be seen that quality control was appropriately con- ducted, and that fluctuations due to statistical errors and/or inter-individual differences of the remaining genes were reduced, with the relative ratios generally being less than 1.3. In the present study, we defined genes whose expression level was changed by more than 1.6-fold after irradiation as Fig. 1. Quality control of microarray data. A: Schematic repre- “profoundly modulated genes”, and their function was ana- sentation of overall procedure for quality control of the microarray lyzed by using the Pathway analysis tool of GeneSpring data. B: Distributions of normalized signal intensity from 12 arrays, 7.3.1. including 3 replicates from each of 4 irradiation conditions. An In addition to analyzing the profoundly modulated genes, asterisk (*) indicates an apparently distinct distribution of signal intensities compared to those of other samples. This sample was we also took another approach to analyze 4061 and 7893 simply excluded from subsequent analyses. C: Scatter plots of sig- genes and examined genes that had a statistically significant nal intensities (raw data) between 2 of the 3 control mice are shown change in expression level after irradiation. The “significantly for 46000 genes (a), and for the 4061 genes that remained after modulated genes” were selected from the quality-controlled removal of outlier data (b). Lines representing a 1.6-fold difference genes described above by using Welch’s ANOVA (analysis are drawn to show that fluctuations were reduced with a relative of variance) with a p-value cut-off of 0.05. Hierarchical clus- ratio of mostly less than 1.3 after removal of outlier data.

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp 244 K. Taki et al. affect on the analysis of the common function of radiation- 0.032 μGy/min, and down-regulation of 3010027A04Rik after modulated gene groups. 13 μGy/min were confirmed. It should be noted that Hspa8 were significantly (t < 0.05) responsive to radiation at a dose- Validation of the microarray data rate as low as 20-times above the typical level of natural back- Real-time PCR was performed using the same RNA ground radiation. As shown in the Venn diagram in Fig. 2C, samples used in microarray analyses. RNA was reverse- five genes were identified at two different dose-rate conditions: transcribed using SuperScript II Reverse Transcriptase and A630033E08Rik (0.032 μGy/min and 0.65 μGy/min); and an oligo dT primer (Invitrogen Japan K.K., Tokyo). PCR Lamp2, Ndufb9, Esd, and 2410003B16Rik (0.65 μGy/min and was performed using the Mx3000p real-time PCR system 13 μGy/min). Lamp2 is a lysosome-associated membrane pro- (Stratagene, La Jolla, CA) with Brilliant SYBR Green tein involved in lysosome biogenesis and autophagy, Ndufb9 QPCR Master Mix (Stratagene). Heat activation was carried is a subunit of a complex, located in the mitochondrial out at 95°C for 10 min, followed by PCR cycles of denatur- inner membrane, that forms a part of the mitochondrial respi- ation at 95°C for 30 sec, primer annealing at 60°C for 1 min, ratory chain, while Esd is a formylglutathione hydrolase and primer extension at 72°C for 1 min. The primers used involved in glutathione-dependent formaldehyde oxidation II, were: for Ndufb9, 5’-GCATCCCTCTGAGAAAGCAA-3’/ and 2410003B16Rik is a cytoskeletal associated protein. These 5’-CATCAGGTGATGTTTCCTCCTG-3’; for Lamp2, 5’- genes are considered to be highly sensitive to low-dose-rate TCTCAAGCGCCATCATACTG-3’/ 5’-TCCGAATGAAAGT- radiation. After irradiation at dose-rates of 0.032 μGy/min, TCCAAGG-3’, for Commd1, 5’-TGGCGAAGATGAGAGG- 0.65 μGy/min, 13 μGy/min, 1, 18 and 14 genes were up- ACTT-3’/ 5’-ACTTGCCATCGACTCTCCAG-3’; for Hspa8, regulated, and 4, 3 and 15 genes were down-regulated, respec- 5’-TCATTACCAAGCTGTACCAGAGTGC-3’/ 5’-CTTAA- tively. All primary microarray data are available at the site of TCCACCTCTTCAATGGTGGG-3’; for 3010027A04Rik, GEO (http://www.ncbi.nlm.nih.gov/geo) (data No. GSE14290). 5’-GACGTCGACGACAAGTACGA-3’/ 5’-GTCATGCTG- By examining the physiological role of each of the pro- GCAACAATACG-3’; for GAPDH (the internal control), 5’- foundly modulated genes, four genes (Uqcrb, Ndufb9, AACTTTGGCATTGTGGAAGG-3’/ 5’-TCAGCTCTGGG- Ndufv2, and Atp5k) out of 50 were found to be involved in ATGACCTTG-3’. The primers for Vdac3 were purchased a common pathway of mitochondrial oxidative phosphory- from SuperArray Bioscience (Frederick, MD). lation. Northern blot analysis was carried out as described previ- ously.16) Briefly, approximately 1 μg of the total RNA was Hierarchical cluster analysis of genes exhibiting signif- electrophoresed on a standard 1.2% formaldehyde agarose gel icantly modulated expression in kidneys after continu- and blotted onto a nylon membrane GeneScreen (PerkinElmer ous low-dose-rate irradiation Japan, Co., Ltd., Yokohama, Japan). Membranes were hybrid- Modulation of multiple genes involved in a common bio- ized with randomly-primed probes specific to the Ndufv2 logical process may cause an obvious phenotype, even if the and GAPDH cDNA. The hybridization signal intensity was modulation of each gene is small. Therefore, we examined measured by a BAS1800 Bio-Imaging Analyzer (FUJI genes that had a statistically significant change in expression FILM Corporation, Tokyo). after low-dose-rate radiation, regardless of their fold-change. These genes, defined as “significantly modulated genes” in RESULTS this study (see Materials and Methods), were selected using Welch’s ANOVA with a p-value cut-off of 0.05. The resulting Pathway analysis of genes exhibiting profoundly modu- 621 genes were classified into 16 clusters based on expres- lated expression in kidneys after continuous low-dose- sion patterns after hierarchical cluster analysis. As shown in rate irradiation Figs. 3A and 3B, the genes belonging to the cluster C1 were As the fluctuations in the signal intensity due to statistical the most abundant and exhibited increased expression in a errors and inter-individual differences were generally less than dose-rate-dependent manner. In order to assess the biological 1.3, we here defined those genes in which the expression level interpretation of the gene clusters, we searched for Gene was changed by more than 1.6-fold after irradiation as “pro- Ontology categories that contained a large number of genes foundly modulated genes” (see Materials and Methods). A from each of the gene clusters. As shown in Fig. 3C, it was total of 50 genes were identified as “profoundly modulated” found that there was a statistically significant overlap (Fisher’s (Table 1). Validation of the microarray data was performed by Exact Test, p < 0.05) between cluster C1 and the Gene checking the expression level of some of these genes using Ontology categories “cytoplasm”, “”, “mito- real-time PCR or northern hybridization (Fig. 2A). Although chondrion organization and biogenesis”, “energy pathways”, the fold-change after irradiation was generally small compared “organelle organization and biogenesis”, “cell organization to the microarray results, up-regulation of Ndufb9, Ndufv2 and and biogenesis”, “cytoplasm organization and biogenesis”, Lamp2 after 0.65 μGy/min and 13 μGy/min, up-regulation of and “transferase”. It is noteworthy that the Gene Ontology Commd1 after 13 μGy/min, down-regulation of Hspa8 after categories of “mitochondrion”, “mitochondrion organization

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp Gene Expression after Low-dose-rate Irradiation 245

Table 1. Kidney genes whose expression was profoundly modulated after continuous low-dose-rate irradiation. gene name description fold-change (A) A630033E08Rik hypothetical KRAB box containing protein 1.87 Hspa8 heat shock protein 8 (Hspa8) mRNA. 0.62 Gls glutaminase (Gls) mRNA. 0.62 Nr1d1 nuclear receptor subfamily 1 group D member 1 (Nr1d1) mRNA. 0.59 Ckap5 RIKEN cDNA 4930432B04 gene (4930432B04Rik) mRNA. 0.58 (B) LOC231081 similar to spermidine/spermine N1-acetyltransferase mRNA. 2.1 LOC381649 similar to Peptidyl-prolyl cis-trans isomerase A (PPIase) (Rotamase) (Cyclophilin A) (Cyclosporin A-binding protein) (SP18) mRNA. 1.99 1110054N06Rik RIKEN cDNA 1110054N06 gene (1110054N06Rik) mRNA. 1.98 LOC270186 similar to H3 histone family 3B mRNA. 1.96 Lamp2 Mouse lysosomal membrane glycoprotein type B (lgp-B) mRNA 3prime end. 1.94 LOC382096 similar to Peptidyl-prolyl cis-trans isomerase A (PPIase) (Rotamase) (Cyclophilin A) (Cyclosporin A-binding protein) (SP18) mRNA. 1.93 1200016E24Rik 1.84 LOC236932 similar to 40S ribosomal protein S6 (Phosphoprotein NP33) mRNA. 1.81 A630033E08Rik hypothetical KRAB box containing protein 1.79 Rpl6 ribosomal protein L6 (Rpl6) mRNA. 1.78 Uqcrb ubiquinol-cytochrome c reductase binding protein (Uqcrb) mRNA. 1.77 Gsr glutathione reductase 1 (Gsr) mRNA. 1.71 Ndufb9 NADH dehydrogenase (ubiquinone) 1 beta subcomplex 9 (Ndufb9) mRNA. 1.7 Cul3 cullin 3 (Cul3) mRNA. 1.67 LOC277856 similar to adenine nucleotide translocase mRNA. 1.66 Esd Esterase D/formylglutathione hydrolase (Esd), mRNA 1.66 BC021891 cDNA sequence BC021891 (BC021891) mRNA. 1.63 4921533J23RIK unnamed protein product; 4921533J23RIK PROTEIN (1110017C22RIK PROTEIN) (2810441K11RIK PROTEIN) (THYMUS 1.62 ATROPHY-RELATED PROTEIN). Dhrs8 dehydrogenase/reductase (SDR family) member 8 (Dhrs8) mRNA. 0.6 D7Rp2e DNA segment Chr 7 Roswell Park 2 complex expressed (D7Rp2e) mRNA. 0.6 2410003B16Rik RIKEN cDNA 2410003B16 gene (2410003B16Rik) mRNA. 0.6 (C) Ndufb9 NADH dehydrogenase (ubiquinone) 1 beta subcomplex 9 (Ndufb9) mRNA. 2.19 Lamp2 Mouse lysosomal membrane glycoprotein type B (lgp-B) mRNA 3prime end. 1.93 Klhdc2 kelch domain containing 2 (Klhdc2) mRNA. 1.9 Commd1 COMM domain containing 1 (Commd1) mRNA. 1.83 LOC383826 similar to 60S ribosomal protein L7a (Surfeit locus protein 3) (PLA-X polypeptide) mRNA. 1.81 Esd Esterase D/formylglutathione hydrolase (Esd), mRNA 1.76 Tpt1 tumor protein translationally-controlled 1 (Tpt1) mRNA. 1.75 Atp5k ATP synthase H+ transporting mitochondrial F1F0 complex subunit e 1.71 LOC383964 similar to ribosomal protein L21 mRNA. 1.69 1110013G13Rik RIKEN cDNA 1110013G13 gene (1110013G13Rik) mRNA. 1.66 Enpp2 ectonucleotide pyrophosphatase/phosphodiesterase 2 (Enpp2) mRNA. 1.66 Ndufv2 NADH dehydrogenase (ubiquinone) flavoprotein 2 (Ndufv2) mRNA. 1.65 Pik3r3 phosphatidylinositol 3 kinase regulatory subunit polypeptide 3 (p55) (Pik3r3) mRNA. 1.61 LOC383793 similar to ribosomal protein L27a; ribosomal protein L29 homolog (yeast) mRNA. 1.61 Itgb1 integrin beta 1 (fibronectin receptor beta) (Itgb1) mRNA. 0.62 Ddx27 DEAD (Asp-Glu-Ala-Asp) box polypeptide 27 (Ddx27) mRNA. 0.62 2900016G23Rik RIKEN cDNA 2900016G23 gene mRNA. 0.61 2410003B16Rik RIKEN cDNA 2410003B16 gene mRNA. 0.6 Mll5 0.6 Ece1 endothelin converting 1 (Ece1) mRNA. 0.58 Ase1 CD3E antigen, epsilon polypeptide associated protein (Cd3eap), mRNA 0.58 Cbx3 adult male kidney cDNA RIKEN full-length enriched library clone:0610042G12 product:chromobox homolog 3 (Drosophila 0.58 HP1 gamma) full insert sequence. Gprin1 G protein-regulated inducer of neurite outgrowth 1 (Gprin1) mRNA. 0.57 Prkaa2 protein kinase AMP-activated alpha 2 catalytic subunit (Prkaa2) mRNA. 0.57 Nfic nuclear factor I/C (Nfic) mRNA. 0.56 Pdpk1 3-phosphoinositide dependent protein kinase-1 (Pdpk1) mRNA. 0.55 LOC381747 similar to splicing factor YT521-B mRNA. 0.53 Ppp1r15b protein phosphatase 1 regulatory (inhibitor) subunit 15b (Ppp1r15b) mRNA. 0.52 3010027A04Rik Ankyrin repeat domain 11, mRNA (cDNA clone MGC:60721 IMAGE:30009418) 0.45 (A) genes whose expression was modulated more than 1.6-fold after irradiation with the dose-rate of 0.032 μGy/mim. (B) genes whose expression was modulated more than 1.6-fold after irradiation with the dose-rate of 0.65 μGy/min. (C) genes whose expression was modulated more than 1.6-fold after irradiation with the dose-rate of 13 μGy/min.

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp 246 K. Taki et al.

Fig. 2. Genes profoundly modulated after low-dose-rate irradiation (fold-change > 1.6). A: Validation of microarray data for the kidney. Up-regulation of Ndufb9, Lamp2 and Commd1, and down-regulation of Hspa8 and 3010027A04Rik was confirmed by the real-time PCR. Up-regulation of Ndufv2 was confirmed by a northern blot analysis. B: Up-regulation of Vdac3 in the testis was confirmed by a real-time PCR. C: The number of profoundly modulated kidney genes is represented by the Venn diagram. The genes identified in two different dose- rate conditions are boxed. Genes that are commonly involved in the mitochondrial oxidative phosphorylation pathway are indicated in boldface. D: The number of profoundly modulated testis genes is represented in the Venn diagram. The genes identified in two different dose-rate conditions are boxed. and biogenesis”, and “energy pathways” were found. Thus, analysis using the testes from the same mice used for the in accord with the analysis of profoundly modulated genes, kidney study. A total of 110 genes were identified to be pro- it suggests that genes associated with mitochondrial oxida- foundly modulated (Table 2). Validation of the microarray tive phosphorylation are modulated by low-dose-rate irradi- data was performed by checking modulation of Vdac3 ation, particularly at 0.65 μGy/min and 13 μGy/min. It is expression after 0.65 μGy/min and 13 μGy/min by real-time interesting to note that the clusters C10 and C13 are also PCR (Fig. 2B). As shown in the Venn diagram in Fig. 2D, significantly similar to the Gene Ontology categories of seven genes were identified in two different conditions: “mitochondrion” and “mitochondrion organization and Jam2 (0.032 μGy/min and 13 μGy/min); 1700012M14Rik, biogenesis”. The genes in these clusters were commonly Vdac3, 1110005A23Rik, Hook1, LOC380730, Speer6-ps1 down-regulated after irradiation at the dose-rate of (0.65 μGy/min and 13 μGy/min). Jam2 is a junction adhe- 0.65 μGy/min. A particular mitochondrial response may sion molecule that is localized to the tight junctions between have occurred in a limited dose-rate range that includes high endothelial cells, 1700012M14Rik is an ankyrin repeat 0.65 μGy/min. domain-containing protein that presumably plays a role in spermatogenesis, Vdac3 is one of the mitochondrial Comparison of response to low-dose-rate radiation membrane channels involved in translocation of adenine between kidneys and testes nucleotides through the outer membrane, 1110005A23Rik is In order to assess the tissue-specificity of alteration of a cytokine-induced protein that plays a role in regulation of gene expression profiles, we performed another microarray growth and nucleic acids metabolism, and Hook1 is a

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp Gene Expression after Low-dose-rate Irradiation 247

Fig. 3. Hierarchical cluster analysis of genes showing a statistically significant change in expression in the kidney after irradia- tion. A: The expression levels of the 621 significantly modulated genes are indicated by color. Green, yellow, and red represent low, medium, and high expression, respectively. B: Based on the expression pattern, genes were classified into 16 clusters. Expres- sion profiles of genes in each of the 16 clusters are shown. C: Gene Ontology categories that overlapped with gene clusters are shown, with p values from Fisher’s exact test. Only clusters C1, C10 and C13 had a statistically significant overlap with any Gene Ontology categories. member of the hook family of coiled-coil proteins, which increase in dose-rates. The Gene Ontology categories that interacts with several members of the Rab GTPase family showed a statistically significant overlap with these clusters involved in endocytosis. After irradiation at dose-rates of are listed in Fig. 5C. Mitochondrion-related categories were 0.032 μGy/min, 0.65 μGy/min, 13 μGy/min, 1, 31 and 41 not found to overlap cluster C1 or C2. Only one Gene genes were up-regulated, and 25, 10 and 9 genes were down- Ontology category, “organelle organization and biogenesis,” regulated, respectively. showed a statistically significant overlap with clusters C1 When the lists of profoundly modulated genes were com- and C2 and with cluster C1 in the kidney (Fig. 3C). Taken pared between the kidney and testis, we found only two together, alteration of the gene expression profile after low- genes that were modulated in both these organs, as shown dose-rate irradiation was different between kidneys and in Fig. 4. In addition, the direction of the modulation of testes. It is interesting to note that the Gene Ontology cate- Hspa8 was opposite in the two tissues. Specifically, there gories “DNA metabolism”, “response to DNA damage” and was down-regulation in the kidney but up-regulation in the “DNA replication” were significantly similar to the clusters testis. It should be concluded that gene modulation after C9 and C10 in Fig. 5, which contain genes whose expression low-dose-rate irradiation is not similar between the kidney was decreased in the testes with an increase in the dose-rate. and testis. This observation suggests that cells in testes respond to low- The hierarchical cluster analysis was performed again as dose-rate radiation by shifting down general DNA metabo- shown in Fig. 5A and 5B. It appears that the expression of lism as well as DNA repair activity. genes belonging to the clusters C1 and C2 increased with

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp 248 K. Taki et al.

Table 2. Testis genes whose expression was profoundly modulated after continuous low-dose-rate irradiation. gene name description fold-change (A) B230396O12Rik RIKEN cDNA B230396O12 gene (B230396O12Rik), mRNA. 1.85 Jam2 junction adhesion molecule 2 (Jam2), mRNA. 0.62 AK032979.1 12 days embryo male wolffian duct includes surrounding region cDNA, RIKEN full-length enriched library, clone: 6720483O07 prod- 0.62 uct: S-phase kinase-associated protein 2 (p45), full insert sequence. Casc1 cancer susceptibility candidate 1 (Casc1), mRNA. 0.62 2600005N12Rik RIKEN cDNA 2600005N12 gene (2600005N12Rik), mRNA. 0.62 Zfhx1b zinc finger homeobox 1b (Zfhx1b), mRNA. 0.61 2610018L09Rik 0.61 Pgcp plasma glutamate carboxypeptidase (Pgcp), mRNA. 0.61 Smpdl3b sphingomyelin phosphodiesterase, acid-like 3B (Smpdl3b), mRNA. 0.61 2610018I03Rik RIKEN cDNA 2610018I03 gene (2610018I03Rik), mRNA. 0.61 2410015N17Rik RIKEN cDNA 2410015N17 gene (2410015N17Rik), mRNA. 0.60 Lrig3 0.60 Ppp1cc protein phosphatase 1, catalytic subunit, gamma isoform (Ppp1cc), mRNA. 0.60 Pet112l PET112-like (yeast) (Pet112l), mRNA. 0.60 Lass2 longevity assurance homolog 2 (S. cerevisiae) (Lass2), mRNA. 0.60 Slc22a7 solute carrier family 22 (organic anion transporter), member 7 (Slc22a7), mRNA. 0.59 Gtl3 gene trap locus 3 (Gtl3), mRNA. 0.59 Nfatc2ip nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 2 interacting protein (Nfatc2ip), mRNA. 0.59 Cpeb1 cytoplasmic polyadenylation element binding protein 1 (Cpeb1), mRNA. 0.58 C730015A04Rik 0.58 Commd10 COMM domain containing 10 (Commd10), mRNA. 0.58 Klk27 kallikrein 27 (Klk27), mRNA. 0.57 2310009E04Rik RIKEN cDNA 2310009E04 gene (2310009E04Rik), mRNA. 0.57 Pdk1 0.55 Ephx2 epoxide hydrolase 2, cytoplasmic (Ephx2), mRNA. 0.55 Rapsn receptor-associated protein of the synapse (Rapsn), mRNA. 0.48 (B) Trip12 2.20 Polr2k Polymerase (RNA) II (DNA directed) polypeptide K 2.05 Taf9 TAF9 RNA polymerase II, TATA box binding protein (TBP)-associated factor (Taf9), mRNA. 2.01 1700012M14Rik RIKEN cDNA 1700012M14 gene (1700012M14Rik), mRNA. 1.91 Vdac3 voltage-dependent anion channel 3 (Vdac3), mRNA. 1.84 1110005A23Rik RIKEN cDNA 1110005A23 gene 1.79 Hook1 hook homolog 1 (Drosophila) (Hook1), mRNA. 1.78 Pttg1 pituitary tumor-transforming 1 (Pttg1), mRNA. 1.77 Cklfsf2b chemokine-like factor super family 2B (Cklfsf2b), mRNA. 1.76 Ube2j1 ubiquitin-conjugating enzyme E2, J1 (Ube2j1), mRNA. 1.74 LOC380730 similar to KIAA0563 protein (LOC380730), mRNA. 1.73 H3f3a H3 histone, family 3A (H3f3a), mRNA. 1.73 Ppp4r1 protein phosphatase 4, regulatory subunit 1 (Ppp4r1), mRNA. 1.72 AA959742 expressed sequence AA959742 (AA959742), mRNA. 1.72 Hspa8 Heat shock protein 8 1.70 Ubxd3 UBX domain containing 3 (Ubxd3), mRNA. 1.69 2610205H19Rik RIKEN cDNA 2610205H19 gene (2610205H19Rik), mRNA. 1.67 Speer6-ps1 spermatogenesis associated glutamate (E)-rich protein 6, pseudogene 1 (Speer6-ps1). 1.67 Aif1 allograft inflammatory factor 1 (Aif1), mRNA. 1.67 Cklfsf2a chemokine-like factor super family 2A (Cklfsf2a), mRNA. 1.66 LOC385662 similar to hypothetical protein MGC26717 (LOC385662), mRNA. 1.66 Pdcd5 programmed cell death 5 (Pdcd5), mRNA. 1.66 Psme2b protease (prosome, macropain) 28 subunit, beta, b (Psme2b), mRNA. 1.64 LOC383802 similar to ribosomal protein L27a; ribosomal protein L29 homolog (yeast) (LOC383802), mRNA. 1.64 Polr2g polymerase (RNA) II (DNA directed) polypeptide G (Polr2g), mRNA. 1.63 Mycbpap Mycbp associated protein (Mycbpap), mRNA. 1.62 Usp1 ubiquitin specific protease 1 (Usp1), mRNA. 1.62 C730025P13Rik RIKEN cDNA C730025P13 gene 1.62 1810035L17Rik 1.61 LOC386564 similar to Protein translation factor SUI1 homolog (LOC386564), mRNA. 1.61 Aqp11 aquaporin 11 (Aqp11), mRNA. 1.61 Copg2 coatomer protein complex, subunit gamma 2 (Copg2), mRNA. 0.62 Gpatc4 G patch domain containing 4 0.62 (continued)

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp Gene Expression after Low-dose-rate Irradiation 249

(continued) gene name description fold-change Bckdk branched chain ketoacid dehydrogenase kinase (Bckdk), mRNA. 0.61 9530068E07Rik 0.60 4833412N02Rik RIKEN cDNA 4833412N02 gene (4833412N02Rik), mRNA. 0.60 5730559C18Rik RIKEN cDNA 5730559C18 gene 0.60 LOC385822 similar to mitochondrial inner membrane translocase component Tim17b (LOC385822), mRNA. 0.60 Upf1 UPF1 regulator of nonsense transcripts homolog (yeast) 0.59 Alg1 Asparagine-linked glycosylation 1 homolog (yeast,beta-1,4-mannosyltransferase) 0.56 Gpr39 0.48 (C) Sas; AW742554; tetraspanin 31 1.94 2700085A14Rik Aldh1a1 Aldehyde dehydrogenase family 1, subfamily A1 1.92 Helz Helicase with zinc finger domain 1.84 Vdac3 voltage-dependent anion channel 3 (Vdac3), mRNA. 1.82 Pcyox1 Prenylcysteine oxidase 1 1.82 Idh3g isocitrate dehydrogenase 3 (NAD+), gamma (Idh3g), mRNA. 1.81 Hook1 hook homolog 1 (Drosophila) (Hook1), mRNA. 1.80 Pja1 praja1, RING-H2 motif containing (Pja1), mRNA. 1.79 Nsbp1 nucleosome binding protein 1 (Nsbp1), mRNA. 1.79 Sepm selenoprotein M (Sepm), mRNA. 1.79 Pgk1 Phosphoglycerate kinase1 1.77 1700012M14Rik RIKEN cDNA 1700012M14 gene (1700012M14Rik), mRNA 1.73 Sc4mo1 sterol-C4-methyl oxidase-like (Sc4mol), mRNA 1.72 Sox8 SRY-box containing gene 8 (Sox8), mRNA 1.72 Fasn fatty acid synthase (Fasn), mRNA 1.70 Heterogeneous nuclear ribonucleoprotein U-like 2 1.70 Ppap2b phosphatidic acid phosphatase type 2B (Ppap2b), mRNA. 1.68 Lynx1 Ly6/neurotoxin 1 (Lynx1), mRNA. 1.68 2810489O06Rik RIKEN cDNA 2810489O06 gene (2810489O06Rik), mRNA. 1.68 Prmt6 Protein arginine N-methyltransferase 6 1.67 Rpl5 ribosomal protein L5 (Rpl5), mRNA. 1.67 Gna14 guanine nucleotide binding protein, alpha 14 (Gna14), mRNA. 1.67 2900073H19Rik RIKEN cDNA 2900073H19 gene (2900073H19Rik), mRNA. 1.67 1810022K09Rik 1.66 Hmgcs1 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 1 (Hmgcs1), mRNA. 1.66 1200016E24Rik 1.66 Cxx1a CAAX box 1 homolog A (human) 1.65 1110005A23Rik RIKEN cDNA 1110005A23 gene 1.65 Speer6-ps1 spermatogenesis associated glutamate (E)-rich protein 6, pseudogene 1 (Speer6-ps1). 1.65 Tmod3 tropomodulin 3 (Tmod3), mRNA. 1.65 a Nonagouti 1.64 Cmas cytidine monophospho-N-acetylneuraminic acid synthetase (Cmas), mRNA. 1.64 Acadsb acyl-Coenzyme A dehydrogenase, short/branched chain (Acadsb), mRNA. 1.64 Ifitm2 interferon induced 2 (Ifitm2), mRNA. 1.63 LOC380730 similar to KIAA0563 protein (LOC380730), mRNA. 1.63 3110018K12Rik RIKEN cDNA 3110018K12 gene (3110018K12Rik), mRNA. 1.63 Jam2 junction adhesion molecule 2 (Jam2), mRNA. 1.62 Cyb561 cytochrome b-561 (Cyb561), mRNA. 1.62 Ephb4 Eph receptor B4 1.61 Ftl1 ferritin light chain 1 (Ftl1), mRNA. 1.61 Cyp17a1 cytochrome P450, family 17, subfamily a, polypeptide 1 (Cyp17a1), mRNA. 1.60 Tmc7 transmembrane channel-like gene family 7 (Tmc7), mRNA. 0.62 2900090M10Rik RIKEN cDNA 2900090M10 gene (2900090M10Rik), mRNA. 0.62 Rad23a RAD23a homolog (S. cerevisiae) (Rad23a), mRNA. 0.62 Aup1 ancient ubiquitous protein (Aup1), mRNA. 0.58 Mad2l1bp MAD2L1 binding protein (Mad2l1bp), mRNA. 0.56 B930008I02Rik RIKEN cDNA B930008I02 gene (B930008I02Rik), mRNA. 0.56 Ggn gametogenetin (Ggn), mRNA. 0.55 1500002O10Rik RIKEN cDNA 1500002O10 gene (1500002O10Rik), mRNA. 0.55 4932416A15 hypothetical protein 4932416A15 (4932416A15), mRNA. 0.55 (A) genes whose expression was modulated more than 1.6-fold after irradiation with the dose-rate of 0.032 μGy/mim. (B) genes whose expression was modulated more than 1.6-fold after irradiation with the dose-rate of 0.65 μGy/min. (C) genes whose expression was modulated more than 1.6-fold after irradiation with the dose-rate of 13 μGy/min.

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp 250 K. Taki et al.

and 13 μGy/min. Zhao et al.6) observed that the expression of a large number of genes regulating intracellular oxidation/ reduction status was changed in the kidney after acute whole-body irradiation of mice with 10 Gy of γ-rays. Thus, it is likely that the redox condition in the kidney is modulated after irradiation with γ-rays over a broad range of dose-rates. In addition, specific induction of genes involved in oxidative phosphorylation and the ATP synthesis process was reported after continuous exposure of S. cerevisiae to 32P-emitted β– Fig. 4. Genes profoundly modulated in both of the kidney and μ 23) testis after irradiation. The numbers below the gene name represent particles at dose-rates of 1.6–330 Gy/min. In light of this the fold-change in gene expression in the kidney (left side) and tes- apparent agreement between our results and those of the tis (right side). yeast study, the mitochondrial respiratory chain may be responsive to low-dose-rate radiation in a variety of eukary- otes. Due to elevated mitochondrial respiratory activity, the DISCUSSION kidneys of mice irradiated at these dose-rates may undergo oxidative stresses. It is noteworthy that intracellular redox In the present study, the gene expression profiles of the status has been suggested to also be modulated in the liver. kidneys and testes of mice were analyzed by a oligonucleotide The level of the protein rhodanese, which is thought to be microarray after long-term irradiation with low-dose-rate γ- involved in antioxidative regulation, was increased in the rays. The data suggest the expression of genes involved in liver after continuous irradiation (485 days) of mice at dose mitochondrial oxidative phosphorylation is up-regulated in rates of 0.65 μGy/min or 13 μGy/min.22) the kidney after irradiation at the dose-rates of 0.65 μGy/min The kidney is a radiosensitive organ and its tolerance of

Fig. 5. Hierarchical cluster analysis of genes whose expression was significantly modulated in the testis after irradiation. A: Expression levels of the 2056 significantly modulated genes are indicated by color. Based on the expression pattern, genes were classified into 16 clusters. The cluster 12 contained no genes. B: Expression profiles of genes in each of the 16 clusters are shown together. C: The Gene Ontology categories that significantly overlapped with the gene clusters C1, C2, C9 and C10 are shown.

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp Gene Expression after Low-dose-rate Irradiation 251 radiation is often a limiting factor in clinical radiotherapy. Tubular epithelial cell damage and cell loss after exposure ACKNOWLEDGEMENTS to high doses of radiation, as well as glomerulosclerosis after low-dose irradiation, are the predominant histological This study was supported, in part, by the Budget for features of radiation-induced nephropathy.24) Accumulating Nuclear Research of the Ministry of Education, Culture, data suggest that the pathogenesis of radiation nephropathy Sports, Science and Technology of Japan, based on screen- involves dynamic interaction between multiple cell types in ing by, and in consultation with, the Atomic Energy the kidney.25) Reactive oxygen species (ROS) and angiotensin Commission of Japan. This study was performed, in part, II have been suggested to mediate this interaction.26–28) As under contract with Aomori Prefectural Government, Japan. pointed out above, redox status in the kidney is likely to be modulated following exposure to radiation regardless of the REFERENCES dose-rate. If ROS play a critical role in the pathogenesis of radiation nephropathy, our results may suggest that there is 1. Ghiassi-Nejad, M., Zakeri, F., Assaei, R. G. and Kariminia, A. (2004) Long-term immune and cytogenetic effects of high a risk of nephropathy after exposure to low-dose-rate radia- level natural radiation on Ramsar inhabitants in Iran. J. Environ. tion. 6) Radioact. 74: 107–116. Zhao et al. reported that several cell cycle-related genes 2. Masoomi, J. R., Mohammadi, S., Amini, M. and Ghiassi- are up-regulated in the kidney after acute whole-body irra- Nejad, M. (2006) High background radiation areas of Ramsar diation with 10 Gy of γ-rays. However, we observed no cell in Iran: evaluation of DNA damage by alkaline single cell gel cycle-related genes that were profoundly modulated even electrophoresis (SCGE). J. Environ. Radioact. 86: 176–186. after irradiation with a total dose of 8 Gy (13 μGy/min), and 3. Hayata, I., Wang, C., Zhang, W., Chen, D., Minamihisamatsu, we did not observe any gene clusters of significantly modu- M., Morishima, H., Yuan, Y., Wei, L. and Sugahara, T. (2000) lated genes in the kidney that overlapped significantly with Chromosome translocation in residents of the high back- the Gene Ontology category “cell cycle” (Fig. 3C). This ground radiation areas in southern China. J. Radiat. Res. 41: result suggests that modulation of the cell cycle-related 69–74. 4. Tanaka, S., Tanaka III, I. B., Sasagawa, S., Ichinohe, K., genes in vivo following exposure to radiation may have a Takabatake, T., Matsushita, S., Matsumoto, T., Otsu, H. and dose-rate threshold. Sato, F. (2003) No lengthening of life span in mice continu- We have demonstrated that alteration of the gene expres- ously exposed to gamma-rays at very low dose rates. Radiat. sion profile in the testis was largely different from that of the Res. 160: 376–379. kidney. The examination of profoundly altered genes 5. Amundson, S. A., Bittner, M., Meltzer, P., Trent, J. and Fornace, showed that 110 genes were altered in testes compared to 50 A. J. Jr. (2001) Induction of gene expression as a monitor of genes in kidneys. There were only two genes commonly exposure to ionizing radiation. Radiat. Res. 156: 657–661. modulated in both these organs, and the direction of modu- 6. Zhao, W., Chuang, E. Y., Mishra, M., Awwad, R., Bisht, K., lation for one of these genes (Hspa8) was opposite in the Sun, L., Nguyen, P., Pennington, J. D., Wang, T. J. C., two organs. A difference in gene expression modulation Bradbury, C. M., Huang, L., Chen, Z., Bar-Sela, G., Robbins, between the kidney and testis was also shown by the M. E. C. and Gius, D. (2006) Distinct effects of ionizing radi- ation on in vivo murine kidney and brain normal tissue gene hierarchical clustering analysis. This result agrees with the 5) 6) expression. Clin. Cancer Res. 12: 3823–3830. observations made by Amundson et al. and Zhao et al. 7. Kruse, J. J. C.M., te Poele, J. A. M., Velds, A., Kerkhoven, that cellular responses to radiation are highly tissue-specific. R. M., Boersma, L. J., Russell, N. S. and Stewart, F. A. (2004) In the testis, cells have been suggested to respond to low- Identification of differentially expressed genes in mouse dose-rate radiation by up-regulating genes associated with kidney after irradiation using microarray analysis. Radiat. responses to heat/temperature. Genes related to DNA repair Res. 161: 28–38. have also been found to be repressed in the testis. It seems 8. Mahmoud-Ahmed, A. S., Atkinson, S. and Wong, C. S., as if cells in testes damaged by low-dose-rate radiation will (2006) Early gene expression profile in mouse brain after not repair the damage with high efficiency. However, it exposure to ionizing radiation. Radiat. Res. 165: 142–154. should be noted that RNA was extracted from the whole 9. Dai, J. M., Sun, D. C., Lin, R. X., Yang, J., Lou, S. and Wang, organ, which consists of a heterogeneous cell population. S. Q. (2006) Microarray analysis of differentially expressed genes in mouse bone marrow tissues after ionizing radiation. Therefore, any observation of modulation of gene expression Int. J. Radiat. Biol. 82: 511–521. would reflect changes in the major cell population. Evalua- 10. Roudkenar, M. H., Li, L., Baba, T., Kuwahara, Y., Nakagawa, tion of gene expression in a small number of parenchymal H., Wang, L., Kasaoka, S., Ohkubo, Y., Ono, K. and cells, which are responsible for genetic effects, remains to be Fukumoto, M. (2008) Gene expression profiles in mouse liver performed. cells after exposure to different types of radiation. J. Radiat. Res. 49: 29–40. 11. Sokolov, M., Panyutin, I. G. and Neumann, R. (2006) Genome- wide gene expression changes in normal human fibroblasts in

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp 252 K. Taki et al.

response to low-LET gamma-radiation and high-LET-like Y. (1998) Genotype in the 24-kDa subunit gene (NDUFV2) 125IUdR exposures. Radiat. Prot. Dosimetry 122: 195–201. of mitochondrial complex I and susceptibility to Parkinson 12. Amundson, S. A., Lee, R. A., Koch-Paiz, C. A., Bittner, M. disease. Genomics 49: 52–58. L., Meltzer, P., Trent, J. M. and Fornace, A. J. Jr. (2003) 21. Dunnick, J., Blackshear, P., Kissling, G., Cunningham, M., Differential responses of stress genes to low dose-rate γ irra- Parker, J. and Nyska, A. (2006)Critical pathways in heart diation. Mol. Cancer Res. 1: 445–452. function: bis(2-chloroethoxy)methane-induced heart gene 13. Fajardo, L. F. (1982) Pathology of Radiation Injur, pp.3–5. transcript change in F344 rats. Toxicol. Pathol. 34: 348–356. Masson Publishing USA, Inc., New York. 22. Nakajima, T., Taki, K., Wang, B., Ono, T., Matsumoto, T., 14. Lenk, G. M., Tromp, G., Weinsheimer, S., Gatalica, Z., Oghiso, Y., Tanaka, K., Ichinohe, K., Nakamura, S., Tanaka, Berguer, R. and Kuivaniemi, H. (2007) Whole genome S. and Nenoi, M. (2008) Induction of Rhodanese, a detoxifi- expression profiling reveals a significant role for immune cation enzyme, in livers from mice after long-term irradiation function in human abdominal aortic aneurysms. BMC with low-dose-rate gamma-rays. J. Radiat. Res. 49: 661–666. Genomics 8: 237–248. 23. Mercier, G., Berthault, N., Mary, J., Peyre, J., Antoniadis, A., 15. Daino, K., Ichimura, S. and Nenoi, M. (2002) Early induction Comet, J.-P., Cornuejols, A., Froidevaux, C. and Dutreix, M. of CDKN1A (p21) and GADD45 mRNA by a low dose of (2004) Biological detection of low radiation doses by combin- ionizing radiation is due to their dose-dependent post-tran- ing results of two microarray analysis methods. Nucleic Acids scriptional regulation. Radiat. Res. 157: 478–482. Res. 32: e12. 16. Haut, S., Brivet, M., Touati, G., Rustin, P., Lebon, S., Garcia- 24. Robbins, M. E. C., Wooldridge, M. J. A., Jaenke, R. S., Cazorla, A., Saudubray, J. M., Boutron, A., Legrand, A. and Whitehouse, E., Golding, S. J., Rezvani, M. and Hopewell, J. Slama, A. (2003) A deletion in the human QP-C gene causes W.(1991) A morphological study of radiation nephropathy in a complex III deficiency resulting in hypoglycaemia and lactic the pig. Radiat. Res. 126: 317–327. acidosis. Hum. Genet. 113: 118–122. 25. Jaenke, R. S., Robbins, M. E. C., Bywaters, T., Whitehouse, 17. Malaney, S., Heng, H. H. Q., Tsui, L. C., Shi, X. M. and E., Rezvani, M. and Hopewell, J. W. (1993) Capillary endot- Robinson, B. H. (1996) Localization of the human gene helium. Target site of renal radiation injury. Lab. Invest. 68: encoding the 13.3-kDa subunit of mitochondrial complex III 396–405. (UQCRB) to 8q22 by in situ hybridization. Cytogenet. Cell 26. Robbins, M. E. C., Zhao, W., Davis, C. S., Toyokuni, S. and Genet. 73: 297–299. Bonsib, S. M. (2002) Radiation-induced kidney injury: a role 18. Lin, X., Wells, D. E., Kimberling, W. J. and Kumar, S. (1999) for chronic oxidative stress? Micron 33: 133–141. Human NDUFB9 gene: genomic organization and a possible 27. Robbins, M. E. C. and Zhao, W. (2004) Chronic oxidative candidate gene associated with deafness disorder mapped to stress and radiation-induced late normal tissue injury: a chromosome 8q13. Hum. Hered. 49: 75–80. review. Int. J. Radiat. Biol. 80: 251–259. 19. Bénit, P., Beugnot, R., Chretien, D., Giurgea, I., De Lonlay- 28. Moulder, J. E., Fish, B. L., Cohen, E. P. and Bonsib, S. M. Debeney, P. , Issartel, J.-P., Corral-Debrinski, M., Kerscher, (1996) Angiotensin II receptor antagonists in the prevention of S., Rustin, P., Rötig, A. and Munnich, A. (2003) Mutant radiation nephropathy. Radiat. Res. 146: 106–110. NDUFV2 subunit of mitochondrial complex I causes early onset hypertrophic cardiomyopathy and encephalopathy. Received on January 26, 2009 Hum. Mutat. 21: 582–586. Accepted on March 16, 2009 20. Hattori, N., Yoshino, H., Tanaka, M., Suzuki, H. and Mizuno, J-STAGE Advance Publication Date: April 25, 2009

J. Radiat. Res., Vol. 50, No. 3 (2009); http://jrr.jstage.jst.go.jp