Dental Materials Journal 25（2）：205－213, 2006

Global Expression Analyses of Mouse Fibroblast L929 Cells Exposed to IC50 MMA by DNA Microarray and Confirmation of Four Detoxification ’ Expression by Real-time PCR

Atsuko ISHIKAWA1, Satoshi JINNO2, Tomoo SUZUKI2, Tatsuhide HAYASHI3, Tatsushi KAWAI3, Tatsuya MIZUNO1, Takashi MORI1 and Masami HATTORI1 1The Second Department of Prosthodontics, School of Dentistry, Aichi-gakuin University, 2-11 Suemori-dori, Chikusaku, Nagoyashi, Aichi-ken 464-8651, Japan 2Department of Periodontology, School of Dentistry, Aichi-gakuin University, 2-11 Suemori-dori, Chikusaku, Nagoyashi, Aichi-ken 464-8651, Japan 3 Department of Dental Material Science, School of Dentistry, Aichi-gakuin University, 1-100 Kusumoto-cho, Chikusaku, Nagoyashi, Aichi-ken 464-0045, Japan Corresponding author, Atsuko Ishikawa E-mail:[email protected]

Received November 9, 2005/Accepted January 17, 2006

―――――――――――――――――――――――――――――――――――――――――――――――――――――― Methyl methacrylate（MMA）is the main component of methyl methacrylic resin, which is widely used in dentistry. Previ- ous studies have investigated whether MMA has any adverse effects on growth and gene expression in mouse fibroblast L929 cells. The present study was designed to further understand the effects of MMA by focusing on cDNA microarray data after L929 cells were exposed to MMA. MMA was found to inhibit cell growth and induce detoxification response genes in L929 cells. One of the most highly up-regulated genes was S-transferase, alpha 1（Ya）（Gsta1）, which has recently been shown to participate in Nrf2 regulation and is considered to be related to detoxification response. Molecular biological data obtained in the present study may therefore provide useful insights into the effects of MMA on living tissue.

Key words : MMA, cDNA microarray, Gene expression ――――――――――――――――――――――――――――――――――――――――――――――――――――――

INTRODUCTION MATERIALS AND METHODS Methyl methacrylic resin is an important material Experimental procedure for dental and medical purposes; it is relatively easy The experimental procedure employed in this study is to handle, and has flexible formability and excellent summarized in Fig. 1. reparability. On the other hand, residual monomer after polymerization is sometimes more than 20％ Cell viability test and the cytotoxicity thereof may cause inflammatory Mouse fibroblast cell line L929（ATCC＃ CCL-1）was symptoms or allergies1,2）. Due to these cytotoxicity maintained in Minimum Essential Medium-Earle concerns, the adverse effects of methyl methacrylate （Invitrogen, Japan）supplemented with 10％ fetal bo-

（MMA）present in dental materials have been thor- vine serum（Equitech Bio Inc.）at 37℃ and 5％ CO2. oughly tested in vitro3－15）. L929 cells were seeded in 96-well plates at 2×104 Recently, cDNA microarray-based methods have cells/cm2 and used for experiments after 24 hours of been introduced for high-throughput monitoring of incubation. Culture medium was removed from each gene expression. This technology has revolutionized well, and monomer solution was added at the start gene expression studies by providing a means to si- of the experiment. Monomer solutions were prepared multaneously measure mRNA levels in thousands of by diluting MMA （Wako Pure Chemical Industries genes in both simple and complex biological sam- Ltd., Japan）with culture medium to the concentra- ples16－19）. tion specified for each experiment, and then applied The present study was designed to further inves- to the cells. As a negative control, cells were grown tigate the pleiotropic effects of MMA by systemati- only with culture medium.

cally identifying the genes differentially expressed To determine MMA IC50（concentration causing a after exposure of the mouse fibroblast cell line L929, 50％ decrease in cell number）－ the cell culture to be which is routinely used for testing the cytotoxic used for cDNA microarray experiment, cell viability properties of dental materials due to their reproduci- tests were performed. Each well was preliminarily ble growth rates and biological responses20,21）. Re- exposed to monomer solutions after 24 hours, and sults of this study may provide insight into the mo- concentration was adjusted to 10, 20, 50, 100, 150, or lecular biological effects of MMA. 200 mmol/l. Cell number in each well was counted 206 CHANGES IN GENE EXPRESSION BY MMA

using a hemocytometer after 24 hours of incubation. Using the dose-response curve of relative cell viabil-

ity, IC50 was determined. The MMA exposure period for cDNA microarray was determined by time-course experiments. After 24 hours of preliminary incubation, each well was ex- posed to monomer solution（10 mmol/l）and continu- ous incubation was carried out for 2, 24, 48, or 72 hours. To precisely estimate the logarithmic growth phase, a growth curve was plotted.

DNA microarray analysis A set of 38,467 sequence-verified mouse cDNA clones （Microarrays Inc., Nashville, TN） was used. L929 cells harvested after 24 hours of preliminary incuba- tion were cultured with （＋MMA） or without （－MMA）10 mmol/l MMA for 12 hours, and total RNA was extracted from the cells. Cultured cells were washed twice with PBS and lysed with TRIzol ReagentTM （Invitrogen, Carlsbad, CA）. Cell lysates were cryopreserved at －80℃ and sent to Genomic Inc., Alliance Biosystems, Japan, where DNA microarray analysis, including RNA purification, la- beling, hybridization, data acquisition and the analy- sis itself, was performed. cDNA targets were syn- thesized from total RNA samples in the presence of either Cy3-dUTP or Cy5-dUTP and applied to the cDNA microarray. RNA from the -MMA sample was used as a reference（Cy3）, and RNA from the ＋MMA sample constituted the experimental （Cy5） sample. Microarray scanning and data normalization were performed using GenePix Pro 4.0 （Molecular Devices Corporation, Sunnyvale, CA）. Gene expres- sion values（GEVs: median ratio in Genepix pro 4.0） were calculated for each hybridized spot. Normaliza- tion of raw data was performed using the LOWESS method, and data were filtered to eliminate the low signal data. Data were expressed as fold change in expression. cDNA microarray experiments were re- peated twice, and both fold expression change values for each gene were used for analysis.

Real-time PCR analysis To verify the expression of four selected genes （Gsta1, Gsta2, Gsta4, Nqo1）detected by microarray analysis, real-time PCR was performed. Total cellu- lar RNAs extracted from control and MMA-exposed cells were reverse transcribed to cDNA. Each PCR mix （50 μl） contained between 10 and 100 ng of cDNA, forward and reverse oligonucleopeptide prim- ers, and probe in Taqman master mix. Assays for Gsta2, Gsta4 and Nqo1, as well as PCR experiments, were performed using assay-on-demand primers and probe kits. Custom-made primers and probes were used for Gsta1 because there was no assay-on- demand kit available for this gene（Table 1）. cDNA Fig. 1 Experimental schema. amplification was performed using a sequence detec- ISHIKAWA et al. 207

Table 1 Gene specific primer and probe sequences used for real-time PCR ABI Assay on demand primer & probe Gene name Assay ID Lot Number Gsta2 Mm008333353_mH 249563 Gsta4 Mm0049403_m1 194677 Nqo1 Mm00500821_m1 215862 GAPDH TaqMan Rodent GAPDH Cont

TaqMan probe

Gene name Probe Seqeonce（5’-3’） Primer Seqence（5’-3’） Gsta1 TTCCTTGCTTCTTGAATTTGTT Forward: CCCCTTTCCCTCTGCTGAAG TTGCATCCAT Reverce: TGAGCTTCACTGAATCTTGAAAG Abbreviations: Gsta2, glutathione-S-transferase, alpha type2; Gsta4, glutathione-S- transferase, alpha type 4; Nqo1, NAD（P）H dehydragenase, quinone 1; Gsta1, glutathione-S- transferase, alpha type1; GAPDH, glyceraldehyde 3-phosphate dehydrogenase. tor （Prism Model 7700 Sequence Detector, Applied Biosystems, Japan）, and the thermal cycling condi- tions consisted of initial denaturation at 50℃ for 2 minutes and 95℃ for 10 minutes, followed by 40 cy- cles of 95℃ for 15 seconds and 60℃ for 1 minute. Each assay was performed in triplicate. Relative value of each gene was calculated by the standard curve method. Correlations were verified between relative real-time PCR values and DNA microarray fold expression change values.

RESULTS Fig. 2 Dose-response curve of relative cell viability, where Effects of MMA on L929 cell growth the concentration of MMA that decreases the

We first considered whether MMA exposure had any number of cells by 50％, IC50, is calculated to be 10 effect on L929 cell growth. MMA exposure was mmol/l. found to result in concentration-dependent inhibition of L929 cell growth when compared with the control. tration of 10 mmol/l and the treatment time of 12

IC50 was calculated by the dose-response curve of hours were selected to investigate the effects of relative cell viability and found to be approximately MMA exposure on global gene expression in L929 10 mmol/l（Fig. 2）. This concentration was therefore cells by cDNA microarray. employed for microarray analysis. To determine the MMA exposure time for cells Gene expression changes induced by MMA in L929 in the microarray experiment, a time-course experi- cells ment was performed. As shown in Fig. 3A, a signifi- Of 38,467 genes screened, expression levels of 331 cant difference in cell number between control and genes were found to be affected by more than two- MMA-exposed cells（p＜0.01）was seen after 24 hours fold, and 53 genes were identified as significant by t- of incubation. The logarithmic growth phase ap- test（p＜0.05）in L929 cells treated with 10 mmol/l peared to begin between 24 and 72 hours. It should MMA for 12 hours － with an overlap of 44 genes in be highlighted that MMA-exposed cells underwent the Venn diagram（Fig. 4, Table 2）. morphological changes into an elongated shape, as One of the highly up-regulated genes was Gsta1 well as to bipolar or multipolar cells with long proc- （Glutathione S-transferase, alpha 1（Ya））, which was esses and beaded varicosity（Fig. 3B）. On the other up-regulated by 27.44-fold（median）. The expression hand, the control cells retained the original of Gsta1 was regulated by transcription factor Nrf2, fibroblastic morphology and tended to grow in clus- which is related to detoxification response. Other ters. These drastic changes were observed from 2 genes related to the same detoxification response, hours to 48 hours. From these results, it would such as Gsta2 （21.87-fold）, Gsta4 （5.45-fold）, Nqo1 seem that to detect the genes triggered by MMA, the （5.06-fold）, and HO-1 （10.93-fold）, were also up- ideal exposure period was 12 hours. regulated（Table 2）. Nrf2 expression was not veri- Based on the above findings, the MMA concen- fied due to the absence of this gene in the presently 208 CHANGES IN GENE EXPRESSION BY MMA

Fig. 3 （A）Count of cells, and（B）Examination under a light microscope to assess MMA exposure-induced morphological changes in mouse fibroblast cell line L929, after cells were incubated with MEM-Earle （－MMA）or 10 mmol/l 0f MMA（＋MMA）for 2, 24, 48, and 72 h.

Gene Expression primers and probe kits shown in Table 1. As shown in Fig. 5, comparison revealed that relative values for Gsta1, Gsta2, Gsta4, and Nqo1 were 9.95-, 9.22-, 5.55- and 3.48-fold higher, re- spectively, in MMA-exposed cells than in the con- trols. To assess the correlation between microarray GEVs and real-time PCR relative values, a correla- tion curve was produced where data were found to exhibit strong correlations（correlation coefficient γ ＝0.96）. These results indicated that a quantitative correlation existed between real-time PCR and microarray data.

DISCUSSION Fig. 4 Overlap of 44 genes between two-fold changed Residual MMA monomer is reported to possess genes and t-test result. bioorganic toxicity1,2）. Further, Dillingham et al.8） reported that the hemolytic activity of acrylates and methacrylates is related to lipophilicity（inversely re- used DNA microarray. lated to water solubility）, and that the mechanism of Real-time PCR analysis was performed for these action of the esters is membrane mediated and rela- genes in order to verify the identity and support the tively non-specific. Fujisawa et al.13） hypothesized reliability of the microarray data, namely of the four that methacrylate monomers are generally incorpo- up-regulated genes（Gsta1, Gsta2, Gsta4, Nqo1）and rated into the surface of membrane lipids that are one housekeeping control gene, GAPDH （Table 1）. solubilized by the monomers. In other words, injury For real-time PCR, aliquots of the same total RNA is caused by initial contact with methacrylate mono- samples used for the cDNA microarray experiment mers, and then lipid solubilization occurs on the sur- were used as substrates together with the Taqman face of the cell membrane13）. These notions are based ISHIKAWA et al. 209

response to toxicity; such data would contribute to the elucidation of the molecular effects of MMA. Not surprisingly, most of the highly up- and down-regulated genes after MMA exposure were de- toxification response related. For example, Gsta1 （ code NM_008181）, one of the major detoxification genes induced by Nrf2 activation22）, was up-regulated 27.44-fold. Furthermore, expression of related functional genes, including Gsta2 （Gene ontology code NM_008182, gene expression value 21.44）, Gsta4（NM_010357, 5.45）, Nqo1（NM_008706, 5.06）, and HO-1 （NM_010442, 10.93）, were also al- tered. These genes are promoted by an Nrf2- mediated expression system. Nrf2 is a transcription factor, and non-specific foreign chemical compounds － or anything that exhibits electrophilicity － activate the Nrf2 system. Thus, Nrf2-related reactions act as xenobiotic metabolism in MMA-exposed cells, and genes such as Gsts23,24） or Nqo125） are involved in the metabolism of foreign chemical compounds（Fig. 6）. HO-1, which possesses antioxidative activity, was also up-regulated. HO-1 reduces the oxidative stress re- sulting from contact with foreign chemical com- pounds, and thus HO-1 may act as a catalyst chang- ing heme to biliverdin, which has high antioxidative activity26）. Fig. 5 Quantitative real-time PCR analysis of selected genes. Total RNA isolated from MEM-Earle We initially expected a high expression of some treated control（－） or from MMA treated cells genes（Table 3）. However, the predicted high expres- 27） 28－31） （＋） was used as a template for real-time PCR sion levels of genes such as Cycp450 , HSP , Nf- 32,33） 34,35） using the Taqman primer and probes described in kb , SOD , catalase, Glutaredoxin and Thio- 36） Table 1. GAPDH was used as the housekeeping redoxin , were not observed. Thus, they were gene, and Gsta1, Gsta2, Gsta4, and Nqo1 were rep- thought not to participate in the reaction of L929 resentative of the up-regulated genes, according to cells to MMA exposure. the microarray experiment. Gene expression value Consequently, transcription factor Nrf2 induced （fold change of gene expression）observed in the the detoxification response and antioxidative activity, microarray experiment was compared with the but sequentially promoted Gsts were good markers relative value（real-time PCR）for each gene. Data for detecting cytotoxic damage to cells that had been showed good correlations （correlation coefficient caused by the pleiotropic effects of MMA. Moreover, r＝0.96） between microarray and real-time PCR Gsts may be good markers for exposure to other res- for both fold change and relative values. ins. For that purpose, further investigation regarding the molecular biological functions of Gsts is impor- on the physicochemical properties of MMA, but the tant, and identification of other markers, which may actual biological action in the cytoplasm is still un- include the genes screened in this experiment, is nec- clear. essary. Recent molecular biological methods have pro- vided new insight on cellular response to foreign sub- ACKNOWLEDGEMENTS stances. These approaches have also been applied to the study of biomaterials used in dentistry, and a This study was partly supported by a Grant-in-Aid few specific genes have been investigated in pioneer- for the “High-Tech Research Center” Project for Pri- ing studies. However, comprehensive studies on gene vate Universities matching fund subsidy from MEXT expression in cells after contact with dental materials （Ministry of Education, Culture, Sports, Science and have not yet been reported. We used cDNA Technology）, 2003-2007. The authors would like to microarray to investigate the effects of MMA expo- thank Professor Tanaka of the First Department of sure on changes in global gene expression in L929 Prosthodontics, and Emeritus Professor Kawaguchi cells. It would be interesting to determine the roles, for discussion and comments. if any, that the highly up-regulated genes identified in the present microarray experiment have in 210 CHANGES IN GENE EXPRESSION BY MMA

Table 2 Selected 44 genes that showed more than two-fold changes and which were identified as significant by t-test ID Gene Symbol Description Median XXXX XXXX XXXX 147.29 XXXX XXXX XXXX 144.23 XXXX XXXX XXXX 103.97 NM_008181 Gsta1 Glutathione S-transferase, alpha 1（Ya） 27.44 NM_008182 Gsta2 Glutathione S-transferase, alpha 2（Yc2） 21.87 NM_023422 Hist1h2bc Histone 1, H2bc 21.35 NM_178201 Hist1h2bn Histone 1, H2bn 13.23 NM_201409 Dub1a Deubiquitinating enzyme 1a 12.57 NM_178195 Hist1h2bf Histone 1, H2bf 12.43 NM_178199 Hist1h2bl;H2b- synonym: H2b-291a; Mus musculus histone 1, H2bl（Hist1h2bl）, mRNA. 11.50 NM_178200 Hist1h2bm Histone 1, H2bm 11.28 NM_010442 Hmox1 Heme oxygenase（decycling）1 10.93 NM_178202 Hist1h2bp Histone 1, H2bp 10.84 NM_148949 Tdpoz1 TD and POZ domain containing 1 10.49 NM_010089 Dub2 Deubiquitinating enzyme 2 10.18 NM_178198 Hist1h2bj Histone 1, H2bj 10.11 NM_175664 Hist1h2bb Histone 1, H2bb 9.54 NM_175665 Hist1h2bk Histone 1, H2bk 9.32 NM_027950 1700012B18Rik RIKEN cDNA 1700012B18 gene 8.87 NM_007887 Dub1 Deubiquitinating enzyme 1 8.54 NM_178197 Hist1h2bh Histone 1, H2bh 8.17 NM_011171 Procr Protein C receptor, endothelial 6.41 NM_008129 Gclm Glutamate-cysteine ligase , modifier subunit 5.69 NM_173047 Cbr3 Carbonyl reductase 3 5.62 NM_025404 Arfl4 ADP-ribosylation factor 4-like 5.48 NM_010357 Gsta4 Glutathione S-transferase, alpha 4 5.45 NM_011456 Serpinb9e Serine（or cysteine）proteinase inhibitor, clade B, member 9e 5.14 NM_009378 Thbd Thrombomodulin 5.14 NM_008706 Nqo1 NAD（P）H dehydrogenase, quinone 1 5.06 NM_175663 Hist1h2ba Mus musculus histone 1, H2ba 4.61 XM_355248 Pappa2 Pappalysin 2 4.57 NM_011990 Slc7a11 Solute carrier family 7（cationic amino acid transporter, y＋ system）, member 11 4.50 NM_019687 Slc22a4 Solute carrier family 22（organic cation transporter）, member 4 4.40 XM_485965 9530051K01Rik RIKEN cDNA 9530051K01 gene 4.24 NM_013566 Itgb7 Integrin beta 7 4.17 NM_029688 Npn3 Neoplastic progression 3 3.99 NM_016873 Wisp2 WNT1 inducible signaling pathway protein 2 3.90 NM_007436 Aldh3a1 Aldehyde dehydrogenase family 3, subfamily A1 3.82 NM_020559 Alas1 Aminolevulinic acid synthase 1 3.62 NM_026353 4930570C03Rik RIKEN cDNA 4930570C03 gene 3.58 NM_133659 Erg Avian erythroblastosis virus E-26（v-ets）oncogene related 3.30 NM_145450 BC022687 CDNA sequence BC022687 3.27 NM_013463 Gla Galactosidase, alpha 2.66 NM_031881 Nedd4l Neural precursor cell expressed, developmentally down-regulated gene 4-like 2.10 Information about genefunction can be obtained by using GenBank＃ in the table at http://genome-www5.stanford.edu/cgi-bin/SMD/sorce/sorceSearch/ ISHIKAWA et al. 211

Table 3 Selected genes showing variable functions expressed at the cell injured in 12,321 genes oxidative metabolism / drug metabolism

Cytocrome P450

ID Gene Symbol Description fold expression NM_019823 Cyp2d22 Cytochrome P450, family 2, subfamily d, polypeptide 22 1.10 NM_145499 Cyp2c70 Cytochrome P450, family 2, subfamily c, polypeptide 70 －1.93 NM_009997 Cyp2a4 Cytochrome P450, family 2, subfamily a, polypeptide 5 1.54 celluer chaperone

Heat shock protein

ID Gene Symbol Description fold expression NM_022310 Hspa5 Heat shock 70kD protein 5（glucose-regulated protein） －1.25 NM_133804 Hspa5bp1 Heat shock 70kDa protein 5 binding protein 1 －1.23 NM_008303 Hspe1 Heat shock protein 1（chaperonin 10） －1.16 NM_010477 Hspd1 Heat shock protein 1（chaperonin） －1.15 NM_015765 Hspa14 Heat shock 70kDa protein 14 －1.26 anti ROS（reactive oxygen species ）reaction

Nf-kb

ID Gene Symbol Description fold expression NM_008689 Nfkb1 Ubiquitin-conjugating enzyme E2D 3（UBC4/5 homolog, yeast） －1.31 NM_019408 Nfkb2 RIKEN cDNA 4833438C02 gene 1.82

SOD

ID Gene Symbol Description fold expression NM_011435 Sod3 Superoxide dismutase 3, extracellular －1.25 NM_011434 Sod1 Superoxide dismutase 1, soluble 1.72 NM_013671 Sod2 Superoxide dismutase 2, mitochondrial 1.19

Catalase

ID Gene Symbol Description fold expression NM_009804 Cat Catalase 2.60

Glutaredoxin

ID Gene Symbol Description fold expression NM_023505 Glrx2 Glutaredoxin 2（thioltransferase） －1.46 NM_053108 Glrx1 Glutaredoxin 1（thioltransferase） －1.43 AK005853 Glrx2; Grx2 unnamed protein product; glutaredoxin 2（thioltransferase） －1.46

Thioredoxin

ID Gene Symbol Description fold expression NM_019913 Txn2 Thioredoxin 2 1.18 212 CHANGES IN GENE EXPRESSION BY MMA

Fig. 6 Detoxification and antioxidative response of Nrf2 mediation. Gsts and Nqo1 are expressed by the expression system of Nrf2 mediation. Nrf2 is one of the transcription factors, and the expression system is activated by non-specific foreign chemical compounds － that being xenobiotic metabolism. The expressed genes as Gsts or Nqo1 involve the metabolism of foreign chemical compounds. Further, Nrf2 is also a general regulator of the defense genes against oxidative stress, which include HO-1. Thus, HO-1 catalyzes heme to biliverdin, which has high antioxidative activity.

Further evaluation of quantitative mathematical model REFERENCES for predicting acute toxicity of acrylate and methacrylate esters. J Dent Res 1972; 51: 1632-1638. 1） Giunta JL, Grauer I, Zablotsky N. Allergic contact 8） Dillingham EO, Lawrence WH, Austian J. Acrylate stomatitis caused by acrylic resin. J Prosthet Dent and methacrylate esters: Relationship of hemolytic ac- 1979; 42: 188-190. tivity and in vivo toxicity. J Biomed Mater Res 1983; 2） Austin AT, Basker RM. The level of residual monomer 17: 945-957. in acrylic denture base materials. Br Dent J 1980; 149: 9） Wataha JC, Hanks CT, Strawn SE, Rat JC. 281-286. Cytotoxicity of components of resins and other dental 3） Sadamori S, Yamagami T, Nikawa H, Furuebisu M, restorative materials. J Oral Rehab 1994; 21: 453-462. Shigeto N, Hamada T. Studies on residual monomer in 10） Hanks CT, Strawn SE, Fat JC. Cytotoxicity effects of dental acrylic resin. J Hiroshima Univ Dent Soc 1987; resin components of resins and other dental restorative 19: 471-474. materials. J Oral Rehab 1991; 70: 1450-1455. 4） Imai Y, Watanabe M, Lee HE, Kojima K, Kadoma Y. 11） Takeda S, Hashimoto Y, Miura Y, Kimura Y, Cytotoxicity of monomers used in dental resins. Re- Nakamura M. Cytotoxicity test of dental monomers ports of the Research Institute of Dental Engineering using serum-free cell culture （in vitro）. J J Dent 1988; 22: 87-90. Mater 1993; 12（5）: 613-619. 5） Thonemnn B, Shmalz G, Hiller KA, Schweikl H. Re- 12） Kojima H, Hanamura A, Sato A, Konishi H. Compari- sponses of L9292 mouse fibroblasts, primary and im- son of cytotoxicity of surfactants on a range of mam- mortalized bovine dental papilla-derived cell lines to malian cells cultured under various conditions. Alter- dental resin components. Dental Materials 2002; 18: natives to Animal Testing and Experimentation 1995; 318-323. 3: 111-120. 6） Lawrence WH, Bass GE, Purcell WP, Autian J. Use of 13） Fujisawa S, Kadoma K, Komoda Y. H and C NMR mathematical models in the study of structure-toxicity studies of the interaction of eugenol, phenol, and relationships of dental compounds. Part I: Esters of triethyleneglycol dimethacrylate with phospholipid acrylic and methacrylic acids. J Dent Res 1972; 51: liposomes as a kodel system for odontblast mem- 526-535. branes. J Dent Res 1998; 67: 1438-1441. 7） Bass GE, Lawrence WH, Purcell WP, Autian J. ISHIKAWA et al. 213

14） Hashimoto Y, Nakamura M. Cytocompatibility and 26） Ishii T, Itoh K, Takahashi S, Sato H, Yamagiwa T, et viscoelastic properties of phthalate ester-free tissue al. Transcription factor Nrf2 coordinately regulates a conditioners. Dent Mater J 2004; 23（3）: 412-418. group of oxidative stress-inducible genes in macro- 15） Fujisawa S, Atsumi T. Cytotoxicities of a 4-META/ phage. J Biol Chem 2000; 275: 16023-16029. MMA-TBBO resin against human pulp fibroblasts. 27） Omura T, Ishimura Y, Fujii K. Cytocrome P-450. Dent Mater J 2004; 23（2）: 106-108. Kodansha Tokyo and VCH Publishers, 1993. 16） Brown PO, Botstein D. Exploring the new world of the 28） Boston RS, Viitanen PV, Vierling E. Molecular chaper- genome with DNA microarrays. Net Genet 1999; 20（1）: ones and protein folding in plants. Plant Mol Biol 33-37. 1996; 32: 191-222. 17） Lipshutz RJ, Fodor SP, Gingeras TR, Lockhart DJ. 29） Morimoto RI. Regulation of the heat shock High-density synthetic oligonucleotide arrays. Net transcriptional response: Cross talk between a family Genet 1999; 21（1）: 20-24. of heat shock factors, molecular chaperones, and nega- 18） Clarke PA, Poele R, Wooster R, Workman P. Gene ex- tive regulation. Genes Dev 1998; 12: 3788-3796. pression microarray analysis in cancer biology, phar- 30） Marchler G. Modulation of Drosophila heat shock macology, and drug development progress and poten- transcription factor activity by the molecular chaper- tial. Biochem Pharmacol 2001; 62: 1311-1336. one DROJ1. EMBO J 2001; 20: 499-509. 19） Cuzin M. DNA chips: a new tool for genetic analysis 31） Hashimoto Y, Ueda A, Nakamura M. Evidence that and diagnostics. Transfus Clin Biol 2001;（ 8 3）: 291-296. HSP70 gene expression may be useful for assessing 20） International Organization for Standardization. ISO that cytocompatibility of dental biomaterials. Dent 10993-5 Biological evaluation of medical devices － Part Mater J 2004; 23（2）: 184-189. 5. A test for cytotoxicity: in vitro methods, Geneva 32） Ghosh S, Karin M. Missing pieces in the NF-B puzzle. 1992. Cell 2002; 109（Suppl）: 81-96. 21） International Organization for Standardization. ISO 33） Hayakawa M, Miyashita H, Sakamoto I. Evidence that 7405 Dentistry-Preclinical evaluation of bio- reactive oxygen species do not mediate NF-kappaB ac- compatibility of medical devices used in dentistry － tivation. EMBO J 2003; 22: 3356-3366. Test methods for dental materials, Geneva 1997. 34） Kinouchi H, Okamoto K, Sasaki Y. Attenuation of 22） Itou K, Yamamoto M. The molecular mechanism of focal cerebral ischemic injury in transgenic mice over electrophilic reaction. J Biochem 2004; 76（4）: 339-348. expressing CuZu superoxide dismutase. Proc Natl Acad 23） Jakoby WB, Ziegler DM. The enzymes of detoxifica- Sci USA 1991; 88: 11158-11162. tion. J Biol Chem 1990; 265: 20715-20718. 35） Sugawara T, Noshita N, Lewen A, Gasche Y. 24） Ishikawa T, Li ZS, Lu YP, Rea PA. The GS-X pump Overexpression of copper/zinc superoxide dismutase in in plant, yeast, and animal cells: Structure, function, transgenic rats protects vulnerable neurons against and gene expression. Biosci Rep 1997; 17: 189-207. ischemic damage by blocking the mitochondrial path- 25） Ross D, Kepa JK, Winski SL, Beall HD, Anwar A, way of caspase activation. J Neurosci 2002; 22: 209-217. Siegel D. NAD（P）H: quinone oxidoreductase 1 36） Salmeen A, Andersen JN, Myers MP, Meng TC. Redox （NQO1）: chemoprotection, bioactivation, gene regula- regulation of protein tyrosine phosphatase 1B involves tion and genetic polymorphisms. Chem Biol Interact a sulphenyl-amide intermediate. Nature 2004; 423: 769- 2000; 129: 77-97. 777.