Detection of Mitomycin C-DNA Adducts in Vivo by 32P-Postlabeling: Time Course for Formation and Removal of Adducts and Biochemical Modulation1

[CANCER RESEARCH 58. 453-461. February 1, 1998] Detection of Mitomycin C-DNA Adducts in Vivo by 32P-Postlabeling: Time Course for Formation and Removal of Adducts and Biochemical Modulation1

Amy J. Warren, Alexander E. Maccubbin, and Joshua W. Hamilton2

Department of Chernistn; Dartmouth College, Hanover, /Wir Hampshire 03755-3564 Â¡A.J. W.. J. W. H.j: Department of Pharmacology unti Tiuicologv. Dartmoiilti Medical Sellimi. Hanmer. New Hampshire 03755-3835 Â¡A.J. W.. J. W. H.Â¡:Nnrris Cotton Cancer Center, Dartmouth-Hitchcock Medical Center, Lebanon. New Hampshire 03756-0001 [J. W. H.I; aiul Department of Experimental Therapeutics. Roswell Park Cancer institute, Buffalo. Neu1 York 14263 Â¡A.E. M.J

ABSTRACT ally and hifunctionally activated G-MMC monoadducts. and G-MMC-G interstrand and intrastrand cross-links at CpG and GpG Mitomycin C (MMC) is a DNA cross-linking agent that has been used sites, respectively (1-3). Although a great deal is now known about in cancer chemotherapy for over 20 years, yet little is known either qualitatively or quantitatively about MMC-induced DNA adduct forma the chemistry of MMC adduction to DNA in vitro and in cell culture, very little is known about MMC effects in intact animals and humans. tion and repair in vivo. As an initial means of investigating this, we used a recently developed 32P-postlabeling assay to examine the formation and In addition, although MMC has been used clinically in cancer chem loss of MMC-DNA adducts in the tissues of a simple in vivo model test otherapy for over 20 years, there is still little known about the system, the chick embryo, following treatment with a chemotherapeutic pharmacokinetics or pharmacodynamics of MMC-DNA adduction in dose of MMC. As early as 15 min after MMC treatment, four adducts humans. It is generally believed that the interstrand cross-link is could be detected in the liver which were tentatively identified as the principally responsible for the antitumor activity of MMC. because (CpG) N2G-MMC-N2G interstrand cross-link, the bifunctionally activated these lesions are lethal if unrepaired (4, 5). However, correlations MMC-N2G monoadduct, and two isomers (a and /{i of the monofunction- between specific MMC-DNA adducts and specific biological end ally activated MMC-N2G monoadduct. The (GpG) N2G-MMC-N2G in- points, such as alterations in DNA replication, RNA transcription, and trastrand cross-link appears to be a poor substrate for nuclease PI and/or cytotoxicity. have yet to be determined. Our laboratory has previously T4 kinase and was not Ã©valuableby this assay. Levels of all four detectable provided evidence that MMC and other genotoxic agents appear to adducts increased substantially within the first 2 h after MMC treatment, preferentially "target" their effects to specific genes in vivo (6-8). We reached maximal levels by 6 h, and decreased progressively thereafter through 24 h, although low levels of certain adducts persisted beyond 24 h. were, therefore, interested in whether the potent effects of MMC on Lung and kidney had comparable levels of total MMC adducts, which gene expression could be correlated with specific types of MMC- were approximately 60% those of the liver, and there were no significant induced DNA lesions in vivo. differences in the proportion of specific adducts among the three tissues. A highly sensitive method is required to detect and quantify MMC- The interstrand cross-link represented -13-14% of the total MMC ad DNA lesions at the low levels that occur following physiologically ducts, which is approximately 5-fold greater than the proportion of CpG relevant doses. 32P-Postlabeling is generally the most sensitive sites in the genome. In addition, the interstrand cross-link was selectively method for the detection of DNA adducts and is capable of detecting decreased after 16 h relative to the three monoadducts, suggesting pref one adduct per IO7 to 10"' normal nucleotides (i.e.. as little as one erential repair. The effect of modulating different components of the adduct per cell: reviewed in Refs. 9 and 10). The <2P-postlabeling Phase I and Phase II drug metabolism on MMC adduct formation, using either glutethimide, 3,4,3',4'-tetrachlorobiphenyl, dexamethasone, buthi- assay has now been used by over 100 laboratories to detect a wide onine sulfoximine, ethacrynic acid, or JV-acetylcysteine pretreatments, was variety of specific DNA adducts resulting from chemical exposures examined to characterize the possible pathways of MMC metabolism and related to occupational settings, environmental contamination, diet, and clinical drug use ( 10). Recently, a '2P-postlabeling procedure was adduct formation in vivo. Surprisingly, none of these pretreatments had a significant effect on individual or total adducts with the exception of developed for detecting the major MMC monofunctional monoadduct. dexamethasone, which caused an almost 2-fold proportional increase in all A/2-(2"/3,7"-diaminomitosen-l"-a-yl)-2'-deoxyguanylic acid, which four adducts in the liver. also had the potential for detecting other MMC-DNA adducts (11). Using this methodology, we examined whether specific MMC-DNA adducts could be detected in various tissues following in vivo treat INTRODUCTION ment of a simple whole-animal model system, the chick embryo, with MMC3 is a bifunctional cross-linking agent that requires sequential pharmacologically relevant but nonovertly toxic doses of MMC. Sub chemical or enzymatic reductions to form covalent adducts with DNA sequently, the time course for the formation and removal of specific principally at the N2 position of guanine (G), forming monofunction- MMC-DNA adducts was examined to determine whether there was a temporal relationship between specific adduct levels and previously observed MMC-induced alterations in gene expression. Finally, we Received 7/16/97: accepted 11/25/97. The costs of publication of this article were defrayed in part by the payment of page examined the effects of various biochemical modulators of Phase I charges. This article must therefore be hereby marked advertisement in accordance with and II activation and detoxification pathways to determine whether 18 U.S.C. Section 1734 solely to indicate this fact. ' This work was supported by Grant CA49002 from the National Cancer Institute. NIH alterations in specific pathways would alter MMC metabolism or (to J. W. H.I. grants from the Norris Cotton Cancer Center and the Hitchcock Foundation, MMC-DNA adduct formation in vivo. and Grant CH-526 (to A. E. M.) from the American Cancer Society. J. W. H. was also partially supported by the Norris Cotton Cancer Center, and A. J. W. was partially supported by the Dartmouth College Chemistry Department. Support for the Dartmouth College Molecular Biology Core Facility was provided by the Norris Cotton Cancer MATERIALS AND METHODS Center Core Grant CA23I08. 2 To whom requests for reprints should be addressed, at Department of Pharmacology Preparation of MMC-modified DNA Standards. A duplex 23-hp oligo- and Toxicology. Dartmouth Medical School. 7650 Remsen. Hanover. NH 03755-3835. Phone: (603)650-1316: Fax: (603)650-1129; E-mail: [email protected]. nucleotide containing a single CpG site (referred to as 7/X. lop strand sequence ' The abbreviations used are: MMC. mitomycin C; SDT. sodium dithionite: TCB. ATAAATACGTATTTATTTATAAA) was reacted with MMC in the presence 3.4.3'.4'-telrachlorobiphonyl: BSO. buthionine sulfoximine: DI, days of incubation (days of SDT (6.0 mM Na,S,O4:1.5 mM MMC: 1.5 niM nucleotides) at 4Â°Cunder of embryonic development): NAC. /V-acelylcysteine: EiA. ethacrynic acid: DTD. DT- argon essentially as described previously (12). The interstrund cross-link was diaphorase/NAD(P)H-quinone oxidoreductase: HPLC. high-performance liquid chroma- tographv: GSH. glulathione: NPI. nuclease PI: PAP. prostatic acid phosphatase: SVPD. subsequently purified from the unreacted duplex using a high temperature, snake venom phosphodiesterase I. high resolution Sephacryl S-HMÃ•chromatography method, and characterized as 453

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1998 American Association for Cancer Research. "P-POSTLABELINCÃ¬OF MMC-DNA ADOUCIS IN VIVO described (12). Salmon sperm DNA (43% GC content) was reacted with MMC assays, cytosolic and microsomal fractions were prepared, and the cytosol was under similar reducing conditions, and free drug was removed by ethanol assayed for DTD enzyme activity as described (18). Values were normalized precipitation. An 87-bp DNA fragment (59.8% GC content) generated by PCR for total protein content as determined by the BCA assay (Pierce, Rockford. from the hormone response unit region of the PEPCK gene promoter was also IL). reacted with MMC using xanthine oxidase/NADH as described previously (13) and repurified by ethanol precipitation (12). RESULTS Treatment of Chick Kmbryos and DNA Isolation. Fertile White Leghorn chicken eggs (Truslow Farms. Inc.. Chestertown. MD) were incubated as The general scheme for postlabeling MMC-DNA adducts is shown described previously (8. 14). Chick embryos at the appropriate DI were in Fig. 1. The 3'-phosphodiester bond at an adducted guanine is NP1 administered a single dose of 30 /xmol/kg MMC ( 100 /ng/egg at 14 DI and 125 resistant, making this an ideal method for adduci enrichment (9, 10). Hg at 15 DI; Ref. 8) in 100 Â¡Ã\ofwater. For experiments using biochemical modulators, chick embryos were administered either glutethimide (0.5 mmol/ kg; Ref. 15). dexamethasone-21-phosphate (2.5 /Â¿mol/kg; Ref. 7), TCB (2.5 /uniol/kg: Ref. 14), or BSO ( 1.5 mmol/kg: Ref. 16) 24 h prior to treatment with C G' T MMC (15 nmol/kg). NAC (1.5 mmol/kg) and EA (200 /Â¿mol/kg) were administered 2.5 and l h prior to MMC (15 /j.mol/kg). respectively. At the appropriate time point, livers were removed; one lobe from each chick embryo skkkkk liver was used for postlabeling. whereas the other lobe was reserved for DTD and GSH assays. NP1/PAP 32P-Postlabeling Assay for MMC-DNA Adducts. For postlabeling. the l X liver lobe was homogenized by repealed passage through an 18-gauge needle, and DNA was isolated using a genomic isolation kit as described by the manufacturer (Qiagen). DNA was resuspended in water and stored at â€”¿70Â°C. Typical DNA yields were 200-400 /xg of DNA per lobe of lis'er with an A-,a, nm:A2nâ€žâ€ž¿,â€žratioof 1.7-1.8. Samples were routinely monitored by HPLC (as described below) to assess RNA contamination. A 2.5-/ng aliquot of DNA was dried down in a Speed Vac (Savant) and digested with NP1 and PAP (80 mM HPLC sodium acetate (pH 5.0). 150 Â¿IMZnCl2, 4 /ng of NP1. and 200 milliunits of 1) , KINASE PAP in a total of IO /il; l h at 37Â°C]as described (11). The digestion was 2) APYRASE terminated by adjusting the pH to 11.5 with 2.5 ;u,lof 0.5 M Tris base, and an aliquot was subjected to HPLC analysis for determination of total nucleotide X contenÃ.The remainder of the sample was used for '2P-posllabeling essentially G'T as described previously (II). Briefly, the specific activity of each lot of [y-12P]ATP was calculated by radiolabeling deoxyadenosine 3'-monophos- phate followed by TLC in 0.3 M ammonium sulfate buffered with 10 mM OH sodium phosphate (pH 7.5). Following 5'-end labeling with |y-12P]ATP and 1) C-18 COLUMN T4 kinase and subsequent treatment with apyrase (20 milliunits/jj.1 in 20 mM bicine). the <2P-labeled adducts were purified using a C1Ksolid-phase extrac 2) SVPD tion column (SepPak Vac; Waters; Ref. 11). The 12P-labeled adducts were dried and subsequently further digested with SVPD as described (II). Radio- labeled adducts were separated by two-dimensional TLC on polyethylenei- mine-cellulose plates (Machery-Nagel) using l M NaH2PO4. pH 6.5 (Dl), and 1.3 lithium formate/2.1 Murea. pH 3.5 (11). Adduci signals were visuali/.ed by autoradiography. and these regions were cut from TLC plates and quantified by liquid scintillation. Adduci levels were calculated based on the radioactivity of each spot, the specific activity of the ATP used in the kinase reaction, and the nucleotide content based on the HPLC analysis (11). Values obtained for the interstrand cross-link were divided by 2 iti account for "P-endlabeling of both I OH groups of this adduci fragment.4 Statistical analyses of data were per formed by ANOVA and Student's l test using the InStat software program (version 2.0: Graphpad Software). HPLC Analysis. An aliquot of each NPI/PAP-digested sample was ana lyzed for total nucleoside content by HPLC ((Hewlett Packard 1090 or 1050; C|B column. Rainin Microsorb-MV. 5 ^im, 4.6 mm X 25 cm; flow rate. 1 ml/min). The following gradient program was used to develop the chromato- gram: l(X)'7r buffer A (50 min KH,PO4. pH 4.O. 2.5% methanol) for 3 min to 100% buffer B (50 mM KH,PO4, pH 4.0, 20% methanol) over 9 min; isocratic with a 10-min hold at 100% buffer B for 10 min; gradient from 100% buffer DB B to 60% buffer B over 4.2 min; then gradient to 100% buffer A over 0.8 min. Fig. I. Scheme for 12P-postlabeling of MMC-DNA adducts. Analysis of a hypothetical followed by a 10-min wash/equilibration in 100% buffer A. A standard sequence is shown diagrammatically. MMC-treated DNA is digested with NPl and PAP nucleoside mix was used to calculate the total nucleoside content of each to either nucleosides or to MMC-modified dinucleolides for monoadducts or letranucle- sample (11). otidcs for the interstrand cross-link resulting from the NPl resistance of the 3'-phosphodi- Analysis of DTD and GSH Levels. Each liver lobe was homogenized in ester bond at the site of an adduci (adduci is indicated by an X, reviewed in Refs. 9-11). 0.25 M sucrose; one-half was reserved for GSH determinalion, whereas the An aliquot is then removed for HPLC analysis lo quantify tolal nucleotides and to analyze for RNA contaminalion and completeness of en/ymatic digestion. MMC-modified di- and olher half was used for DTD activity. The cytosolic fraction was isolated and telranucleotides are selectively 5'-end labeled wilh T4 kinase. and Ihe excess [y--'-P|ATP assayed for GSH contenÃusing a fluorimetrie assay as described (17). For DTD is converted lo 12P, by apyrase and removed by a Ct!i solid phase extraction column. The radiolabeled adducted-nucleolides are further digested to nucleosides with SVPD and are separated by two-dimensional (Dl and D2) high resolution TLC (see "Materials and 4 A. E. Maccubhin. unpublished results. Methods").

454

of these latter two adducts has not been performed with purified standards. However, alterations in the relative proportion of the dif ferent spots was shown to be dependent on the method of activation of MMC. further supporting these assignments. Under monofunc tional activation conditions (Fig. 4). spots 3 and 4 were observed I mainly, whereas under bifunctional activation conditions (Fig. 3), spots 1 and 2 were observed mainly, consistent with their previous characterization (1. 11. 13. 19-21). 32P-Postlabeling of an oligonu cleotide containing a single GpG site under conditions favoring for mation of the N2G-MMC-N2G intrustrund cross-link yielded ambig-

Fig. 2. 32P-Postlaheling of the MMC-DNA interstrand cross-link produced /'// vitro. A 23-bp duplex oligonucleotide containing a single central CpG N2G-MMC-N2G inlerslrand cross-link was synthesized and purified to homogeneity (12) and subjected to Ã•2P- postlabeling (see Fig. 1 and "Materials and Methods"). Two adducts were observed that are referred to as spots 1 and 5 in ilk- text. Spot 1 corresponds to the interstrand cross-link. Spot 5 is unidentified but is hU'h either another adduci produced from the in vitro SDT reaction conditions and/or a breakdown product of the cross-link.

NP1 treatment results in unmodified nucleosides (N), monoadducted nucleotides (GxpN, where X indicates an adduci), and cross-linked nucleotides (GxpN-GxpN). Because of the requirement for a 3'- phosphate for efficient 5'-end labeling by T4 kinase, the adducted Fig. 3. 3"P-Postlabeling of salmon sperm DNA treated with MMC' under bifunctional reducing conditions in vitro. Purified salmon sperm DNA was treated with MMC under nucleotides were preferentially radiolabeled. Subsequent treatment bifunctional reducing conditions in vitro (see "Materials and Methods") and subjected to with SVPD yields S'-end labeled adducted nucleosides (i2pGx, 12P-postlabeling (see Fig. 1 and "Materials and Methods"). Five adducts were delected 32pGx-xG32p; reviewed in Refs. 9 and 10). To allow identification of and are referred to as spots I through 5. Spot 1 represents the interslrand i ss-link. and spot 5 is an artifact of the in vitro reaction (see Fig. 2). Spol 3 represent the major specific MMC-DNA adducts. various authentic standards were syn monofunctional monoadduct (Â«-isomeri as described previously Sp( s 2 and 4 thesized. The monofunctioiuil monoadduct, N2-(2"ÃŸ,7"-diaminomito- represent the bifunctiona! monoadduct and the ÃŸ-isomerof the nionofunctior al monoad- sen-l"-a-yl)-2'-deoxyguanylic acid, which is the major MMC mono duct, respectively. Spots 3 and 4 are not visible on this exposure and re uired long autoradiograph times for detection in these samples (data not shown: see Table 1). Their adduct (a isomer), was synthesized using reaction conditions (H2/ positions are shown by the arrows. Pt-,O) that favor monofunctional activation conditions (19). 32P- Postlabeling of this standard had been shown previously to produce one adduci spot (referred to as "spot 3"; Ref. 11). 32P-Postlabeling of the highly purified and previously characterized (12) MMC-DNA interstrand cross-link, N~-(2"ÃŸ,7"-diamino-10"-deoxyguanyl-N2-yl- mitosen-l"-a-yl)-2'-deoxyguanylic acid, generated in vitro produced two spots (Fig. 2, referred to as "spot 1" and "spot 5"). Greater than 97% of the total radioactivity was present in spot 1, which represents the (CpG) N2-G-MMC-N~-G cross-link. Spot 5 was never observed in in vivo-treated samples and. therefore, is not believed to be biologi cally relevant. Spot 5 varied in intensity, depending on how long the cross-link was stored prior to analysis, and most likely represents a major degradation product or side product, perhaps involving oxida- tive damage produced from the SDT reducing conditions. Salmon sperm DNA was reacted in vitro with MMC under optimal bifunctional reducing conditions and 32P-postlabeled. which produced six spots (Fig. 3). In addition to spots 1, 3, and 5, two other spots ("spot 4" and "spot 2") were obtained, which most likely represent the Fig. 4. *2P-Postlabeling of an 87-bp synthetic duplex treated with MMC under ÃŸ-isomerof the monofunctional monoadduct. N2-(2"ÃŸ.7"-diaminomi- monofunctional alkylating conditions in vitro. An 87-hp duplex DNA fragment InO'/r GC tosen-l"-ÃŸ-yl)-2'-deoxyguanylic acid, and the bifunctional monoad content) was treated with xanlhine oxidase/NADH under argon as described in "Materials and Methods" and subjected to l2P-postlabeling (see Fig. I and "Materials and Methods"). duct, N2-(10"-decarbamoyl-2",7"-diaminomitosen-l"-a-yl)-2'-deox- Adducts 1-5 were detected as described in Figs. 2 and 3 above, and an additional yguanylic acid, respectively (13, 20). although absolute confirmation unidentified adduci (spot -) was also detected in these samples. 455

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1998 American Association for Cancer Research. "P-POSTLABEUNQ OF MMC-DNA ADDUCTS IN VIVO uous results (data not shown), suggesting that this adduci is not a good substrate in the kinase reaction. Thus, a different method may be necessary to assess this adduci by 32P-posllabeling. Validation of the quantitative nature of the assay has been described previously ( 11). In addition, radiolabeled ['H]porfiromycin was used as a indicator of MMC binding, and similar total adducts/nucleotides were obtained by the two methods (data not shown). Several of the same MMC-treated samples were independently '2P-posllabeled and analyzed in two dif ferent laboratories (A. E. M.. Roswell Park, and J. W. H., Dartmouth) to examine interlaboratory variability, and comparable results were obtained (data not shown). We then examined whether MMC-DNA adducts could be detected in vivo at physiologically relevant doses. The chick embryo was chosen as a simple and useful in vivo model system for these studies. Embryos were treated with a single dose of 30 /nmol/kg MMC (~LD5), which had previously been shown to cause little or no overt cytotoxicity over a 96-h period, but which caused significant suppres sion of DNA and RNA synthesis, and which also caused profound changes in expression of specific genes (S).5 DNA was isolated from the liver at various times after MMC treatment and analyzed by '"P-postlubeling. Representative autoradiographs of both solvent- and MMC-treated embryos are shown in Fig. 5. Four spots were detected corresponding to the interstrand cross-link (spot 1). the bifunctional monoadduct (spot 2), and the two isomers of the monofunctional B monoadduct (spots 3 and 4; Fig. 5ÃŸ).Unidentified spots .vand v were also present in chick embryos treated with solvent alone (Fig. 5/4), indicating that these most likely represent endogenous adducts (9. 10). Table 1 compares the adduci levels of various DNA samples that were treated in vitro and in vivo with MMC. The most notable difference between in vitro chemical activation of MMC with SDT as compared to biochemical activation in vivo is the proportion of total bifunctional adducts to monofunctional adducts, indicating that monofunctional activation is a major contributor to MMC adduction in vivo. The amount of total MMC binding was approximately 100-fold lower in vivo than in vitro under these conditions. It is also interesting to note I that the a isomer of the monoadduct was the predominant MMC adduci in rat hepatoma H4IIE cells in culture, whereas in chick embryo hepatocytes in vivo, this adduci represented only one-fifth of the total MMC adducls (Table 1). Â« The time course for the formation and removal of specific MMC- DNA adducts was examined over a 48-h period following a single dose of 30 /xmol/kg (Fig. 6). There was a close correlation between the time course for total MMC adducts as determined by postlabeling (Fig. 6) and the time course reported previously for radiolabeled Fig. 5. 32P-Postlabeling of chick embryo liver following treatment with MMC in vivo. porfiromycin binding in 14 DI chick embryo liver (8). It should be Fourteen Dl chick embryos were treated with solvent alone (A} or with a single dose of noted that the largest variability in postlabeling results occurred at the MMC (30 fimol/kg: ÃŸ)for 4 h. and liver DNA was isolated and subjected to 32P- time points with the highest adduci levels and ihe highesl number of postlabeling (see Fig. I and "Materials and Methods"). Adducts 1-4 were observed in replicate samples: therefore, this represenls primarily in vivo interem- MMC-treated embryos (B). which represent the interstrand cross-link, the a-isomer of the monofunctional monoadduct. the bifunctional monoadduct. and the ÃŸ-isomer of the bryo variability rather than assay variability. As early as 15 min after monofunctional monoadduct. respectively (see Figs. 2 and 3). Two additional unidentified MMC administration, all four MMC adducts were detected. Adduci adducts, labeled as spots .v and \. were observed, which were also present in untreated embryos (A} and most likely represent endogenous adducts. levels dramatically increased within the first 2 h after treatment and then more gradually increased to maximal levels by 6 h. Interestingly, the bifunciional monoadduci and ihe ÃŸisomer of the monofunctional to 12 h after MMC ireatment (Fig. 6) and also appeared lo be monoadduct were still at approximately one-half maximal levels by preferenlially losl from liver DNA belween 12 and 48 h (Fig. 6C). 24 h. Elevated levels of ihe ÃŸisomer of ihe monofunctional mono To mimic the repeated dosing that is more typical of human cancer adduct remained ihrough 48 h. The inlerstrand cross-link represented chemotherapy, chick embryos were administered a second dose of approximately 13-14% of the total MMC adducts formed in vivo up MMC 24 h after the first dose and then assayed 24 h after the second dose. Appropriate solvenl controls were included at each time of 5 R. M. CarÃ³n and J. W. Hamilton Developmental])- specific effects of the DNA ireatment because the embryo is also rapidly developing over this crosslinking agent mitomycin C on phosphoenol pyriu ate carboxy kinase gene expression 48-h time period (see scheme in Fig. 7). The 15 DI embryos treated in vivo: correlation with changes in chromalin structure within the promoter region of Ihe gene, manuscript in preparation. with MMC for 24 h had total adduci levels that were 1.7-fold greater 456

Table 1 Summary of maximal MMC-DNA adduci levels in various DNAx reacted with MMC in vitro and in vivo as detected bv '<2P-po.\tlabt'liui>

spermDNA"40138026.961.46.75.087-bp DNAfragment"60158016.049.518.615.9Chickembryoliver'14114.314.536.225.124.2H4IIEcellline'400.64ND''ND100ND

% GCconlenlTotal nucleotidel%adducls (/Â¿mol/mol cross-link%interstrand monoadduct%bifunctional a%monofunctional monoadduct monofunctional monoadduct ÃŸSalmon " Bifunctional activation conditions using SDT under argon. '' Fourteen Dl chick embryo liver, treated with 30 fimol/kg MMC in mo: maximum adduci levels shown above were observed at 6 h (Â»i= 5). ' Rat hepatoma cell line H41IE. treated with 0.1 n\i MMC in 95% air. 5% CO2: maximum adduci levels shown above were observed at 2 h (n = 2). '' ND. not detected. than 14 DI embryos treated for 24 h at body weight-equivalent doses, ments had no significant effect on MMC adduci levels, suggesting that suggesting a greater activation of MMC at the older age (Fig. 7; the dexamethasone effect was not principally due to changes in GSH P < 0.005). Treatment with MMC for 24 h at 14 DI, followed by a per se. second 24 h treatment with MMC at 15 DI. produced a 2.1-fold greater adduci level than a single 24-h treatment at 14 DI (P < 0.005: Fig. 7). The levels of adducts in the two dose treatment were exactly DISCUSSION additive of what was observed for 14 DI embryos treated with a single Previous methods used to detect MMC-DNA adducts have included dose for 48 h and 15 DI embryos treated with a single dose for 24 h, UV-vis absorbance combined with HPLC (1, 26). binding of radio- suggesting that the first dose did not affect the adduci formation of the labeled MMC or porfiromycin (8, 27, 28), and DNA alkaline elution second dose. The relative proportions of specific adducts in all of (29). However, each of these methods is limited by either lack of these treatments were approximately the same (Fig. 7). MMC adduci sensitivity or specificity or limitation to only one or a few types of levels were also compared in the liver, lung, and kidney of 14 DI MMC-DNA adducts. Previous attempts to obtain antibodies against embryos following a single 8-h MMC treatment. Total and individual specific MMC-DNA adducts have been unsuccessful, most likely due MMC adduci levels were comparable in lung and kidney but were 60 to the nondistorting nature of the MMC interstrand cross-link and and 58% those of the liver, respectively, although the relative pro monoadducts in helical DNA that makes these complexes relatively portions of specific MMC adducts were comparable in all three tissues nonantigenic (12).7 Thus, no good method existed for the detection of (Fig. 8). These results are consistent with the higher metabolic capac the low levels of specific MMC-DNA adducts that occur in cell ity and blood flow of the liver as compared to lung and kidney but culture or in vivo experiments at physiologically relevant doses. suggest that there are not large quantitative or qualitative differences Previous attempts to use 32P-postlabeling to detect MMC-DNA ad in MMC metabolism or DNA adduci formation among these three ducts have met with limited success, primarily because the specific major organs in vivo. MMC-DNA adducts had not yet been well characteri/.ed at the time of The effects of various biochemical modulators of specific Phase I those studies (30-32). As a result, there were problems with the assay, and II activation and detoxification pathways were examined to de such as incomplete enzymatic digestion, poor resolution by TLC, and termine the possible role of these pathways on MMC adduci forma inability to characterize and definitively identify the adduci spots. tion in vivo (Fig. 9). Glutethimide, which is a phÃ©nobarbitalanalogue, Recently, a new '"P-postlabeling procedure was described that was was used to principally induce the major phenobarbital-inducible initially optimized to detecl ihe major MMC monofunctional (a chicken cytochrome P450 genes, i.e., CYP2H1 and CYP2H2 (15, 22). isomer) monoadduct (11). This technique has been extended to ex TCB was used to principally induce the major dioxin-inducible amine other MMC-DNA adducts in vitro and in vivo, including the ÃŸ chicken P450 genes, i.e., CYPÃŒA4andCYP1A5 (14, 23). Dexametha- sone was used to principally induce several glucocorticoid-inducible isomer of the monofunctional monoadduct. the bifunctional monoad duct, and the interstrand cross-link. Unfortunately, the intrastrand chicken P450 CYP3A isoforms (24). BSD and EA were used to cross-link does not appear to be a substrate for the first two enzymatic decrease and NAC was used to increase glutathione levels, respec tively (16).6 Each of these agents has been shown previously to be reactions of this postlabeling procedure; thus, a different methodology may be required to examine this other MMC cross-link. effective at modulating these pathways in the chick embryo liver, The four major MMC-DNA adducts detectable by this assay were although they each cause other pleiotropic effects on Phase I and II observed both at the 15 and 30 fimol/kg MMC doses in chick embryo metabolism (reviewed in Ref. 25), including changes in DTD and liver, lung, and kidney in vivo, demonstrating the sensitivily and GSH levels, which were also measured following each of these specificity of this method (maximum total adducts/nucleotide. ~14 pretreatments (Fig. 10). MMC was administered at a lower dose of 15 Â¿Â¿mol/molat30 Â¿Â¿mol/kg.6h). These adducts were identified as the ju.mol/kg in these experiments to avoid possible combined toxicity interstrand cross-link (spot 1). the bifunctional monoadduct (spot 2), with some of these agents. None of these pretreatments had any and two isomers of the monofunctional monoadduct (a and j3. spots significant effect on total or individual MMC adduci levels (Fig. 9) 3 and 4, respectively), based on use of authentic standards and/or with the exception of dexamethasone. which caused a proportional modifications in the reducing reaction that favor formation of specific increase in all four adducts. resulting in a 1.7-fold increase in total adducts as described previously (1. II. 13. 19-21). A new spot was MMC adducts (P < 0.05; Fig. 9). Dexamethasone was also shown to also observed (spot 5). which was only present in in vitro samples decrease GSH levels by approximately 2-fold (Fig. 10). However. under chemical reduction and which appeared to increase in intensity BSO and EA also suppressed liver GSH levels to a comparable extent with increasing storage of samples prior to analysis. The nature of this as dexamethasone, and glutethimide increased GSH levels to a com spot is not known but is assumed to represent either an artifact of the parable extent as NAC (approximately 1.3-fold), whereas these treat- SDT reduction conditions and/or a degradation product of the inter strand cross-link. Interestingly, a higher mutation frequency has been *"J-M. Yuann and J. W. Hamilton. Effects of modulating ascorbate and glutathione levels on chromium-DNA binding and formation of chromium (V) in liver and kidney of chromium (VD-lreated OOS rats in rii'Â»,manuscript in preparation. 7 A. J. Warren and J. W. Hamilton, unpublished results

457

lism in intact animals based solely on cell culture studies. The levels of lolal and individual adducts were approximately comparable in lung and kidney and were approximately 60% that in the liver, which is consistent with the higher blood flow and metabolic capacity of the liver. The lack of significant qualitalive or quantitative differences in specific adduci levels suggest thai there are no major differences in MMC metabolism or adduci dislribulion among these major organs. The rapid increase in adduci levels in vivo during the first 2 h after MMC treatment is consistenl wilh ihe rapid first-pass kinetics of MMC in vivo (33). The time course for formation and removal of ihe MMC inlerslrand 16 24 32 40 48 cross-link correlated closely with time course for effects observed previously of MMC on inducible gene expression in chick embryo liver in vivo (8), suggesting thai ihis adduci may be principally 25 responsible for ihese effects. Similarly, the chromium(VI)-induced interslrand cross-link had been shown to be mosl closely correlated I 20 with effects of chromium(VI) on inducible gene expression in this system (6, 7). The cross-link was presenl at about 13-14% of tolal adducls during Ihe firsl 16 h after MMC exposure, which is in approximately 5-fold excess of its predicted level based on ihe total Bo 10

A 14 DI 15 DI 16 DI

MMC 16 24 32 40 48 MMC24 I MMC 25 SOI+MMC24 i MMC MMC â€¢¿o3 20- MMC24+24 TU co MMC ~m 15- MMC48 B

10- 15-1 B O (O o 5- O 10-

16 24 32 40 48 time after MMC (hr) 5- -o=3 Fig. 6. Time course for formation and removal of MMC-DNA adducts in chick embryo liver in VMYÃ•.I4DI chick embryos were treated with a single dose of MMC (30 fimol/kg), and liver DNA was isolated and subjected to ^2P-postlabeling as shown in Fig. 5. Individual adducts were quantified as described in "Materials and Methods" (see also Fig. MMC24 SOI+MMC24 MMC24+24 MMC48 5 for identities of adducts). Each data point represents the mean of values from three to eight individual embryos; bars, SE. A, individual MMC adducts: â€¢¿.spotI (inlerstrand cross-link); â€¢¿spot2 (bifunctional monoadduct); A. spot 3 (monofunctional monoadduct, o isomer); and 4. spot 4 (monofunctional monoadduct, ÃŸisomer). B. total MMC adducts. C. interslrand cross-link as a percentage of total MMC adducts. observed in SDT-treated plasmid DNA in the absence of MMC as compared to untreated DNA, indicating that these reaction conditions may be inducing other forms of DNA damage (11). All four adducts were detected in chick embryo liver in vivo as early as 15 min after treatment with MMC, and maximal levels of MMC24 SOI+MMC24 MMC24+24 MMC48 adducts were observed at 6 h. It is worth noting that all four adducts were present at roughly comparable levels in liver and other tissues in embryo treatment vivo, whereas only the a isomer monoadduct was detected in MMC- Fig. 7. Effect of various treatment regimens on formation of MMC-DNA adducts in treated rat hepatoma H4IIE cells in culture. This may be due to the chick embryo liver in vivo. Chick embryos were treated with MMC (30 Â¿imol/kg)or solvent at 14 DI and 15 DI and assayed 24 h later ( 16 DI) as shown in the scheme in A. large differences in total MMC adduci levels in these two systems Liver DNA was isolated and subjected to '-P-postlabeling as shown in Fig. 5. Data under these conditions. Alternatively, these results may indicate that represent the means (or range) of values from two to four individual embryos; bars, SD. activation and MMC adduci formation is qualitatively different in B, total MMC adducts; C individual MMC adducts: columns from light to dark (left to right) represent spot I (interstrand cross-link), spot 2 (bifunctional monoadduct). spot 3 culture and in whole animals, even within the same cell type, which (monofunctional monoadduct. a isomer), and spot 4 (monofunctional monoadduct. ÃŸ may make it difficult to draw conclusions regarding MMC metabo isomer), respectively. 458

-fcroco-p*.eÃŸi that this lesion is also preferentially repaired in chick embryo liver in adducts/nt(^mol/mol)D vivo, with approximately 75% of the total cross-links being removed within 18 h of attaining maximal levels. Similarly, chromium(VI)- induced DNA-protein and DNA interstrand cross-links have been shown to be rapidly and specifically removed from chick embryo liver DNA in vivo (42). Thus, preferential repair of cross-links may repre sent a survival strategy for normal dividing tissues, and this may be particularly true in developing systems such as the chick embryo. In contrast, it is interesting to note that there were significantly elevated levels of spot 4 (tentatively identified as the ÃŸisomer of the mono- lung kidney functional monoadduct) relative to the other adducts by 48 h after 2.0-, MMC treatment. These results suggest that this adduci may not be easily detected by DNA damage recognition and repair enzymes; therefore, this adduci may represent a promutagenic lesion. The biotransfonnation of MMC in vivo appears to be very complex. A variety of enzymes have been implicated in the reduction of mitomycin antibiotics to reactive species, including NADPH cytochrome c reducÃase(43-45), DTD (45, 46), xanthine oxidase (43, 44), xanthine dehydrogenase (47-49), cytochrome P-450 (50). and NADH cytochrome B5 reducÃase(51). It had been postulated previously that DTD is one of the principal pathways of MMC activation in vivo, liver kidney based primarily on in vitro and cell culture experiments using the embryonic tissue

Fig. 8. Comparison of MMC-DNA adducts in chick embryo liver, lung, and kidney in 10.0n vivo. 14 DI chick embryos were treated with a single dose of MMC (15 fimol/kg), and liver, lung, and kidney DNA was isolated and subjected to 12P-postlabeling as shown in Fig. 5. Data represent the means (or range) of values from two to four individual embryos; bars, SD. A, total MMC adducts; B. individual MMC adducts; columns from light to dark I 7* (left to right) represent spot 1 (interstrand cross-link), spot 2 (bifunctional monoadduct), spot 3 (monofunctional monoadduct, a isomer), and spot 4 (monofunctional monoadduct. ? 5.0-I ÃŸisomer), respectively. T o T â€¢¿o2.5- T3 genomic CpG sites available and assuming a statistical distribution of n) adducts. CpG dinucleotides are also the sites of 5-methylcytosine 0.0- methylation and are significantly underrepresented in the eukaryotic O o o O o o o o genome as a whole but are also clustered at much higher-than- 5 predicted levels within certain promoters, where they likely play a key o I co role in gene regulation (34). It has been shown that methylated CpG O O co o sites are preferentially cross-linked by MMC in vitro (35). It is, (fi m therefore, possible that the effects of MMC on inducible genes may in large part be due to the high cross-linking efficiency of MMC at CpG sites that are clustered within the promoters of these inducible genes, thereby preferentially "targeting" MMC effects to this class of genes. Such a mechanism could account for the strong and selective effects of MMC on gene expression, even at very low doses in vivo. The MMC interstrand cross-link also appears to be preferentially lost from the genome in vivo, because the proportion of cross-link to total adducts decreased significantly between 16 and 48 h after MMC treatment. Although the chick embryo liver is rapidly dividing during this period, this loss cannot be explained by simple dilution resulting O OO from DNA replication and tissue growth, because this could only account for about 25% of the loss of this adduci (36, 37) and because CO o Ãœ co the other adducts are not lost at the same rate over this time period. It m Ã“ has generally been assumed that interstrand cross-links are more 3 difficult for the cell to repair than are monoadducts (38, 39). Cross embryo treatment links are considered lethal to cells if left unrepaired, and the prefer Fig. 9. Effects of various biochemical modulators on formation of MMC-DNA adducts ential killing of rapidly dividing cells by cross-linking agents is in 14 DI chick embryo liver in vivo. 14 DI chick embryos were treated with dexametha- believed to be primarily a result of formation of this lesion (4, 5). In sone (Oc.v. 2.5 tÂ¿mol/kg).TCB (2.5 /Â¿mol/kg).glutethimide (Glut. 0.5 mmol/kg), or BSO support of this, formation of the MMC cross-link has been reported (1.5 mmol/kg) for 24 h, or NAC (1.5 mmol/kg) for 2.5 h, or EA (200 Â¿imol/kg)for 1 h, prior to treatment with MMC (15 Â¿imol/kg)for 8 h. Liver DNA was isolated and subjected recently to severely compromise phage survival in a bacterial mu- to 12P-postlabeling as shown in Fig. 5. Data represent the means (or range) of values from tagenesis assay (40). However, recent studies in cultured mammalian two to four individual embryos; bars, SD. A, total MMC adducts: B, individual MMC adducts: columns from light to dark (left to right} represent spot I (interstrand cross-link), cells have demonstrated that the MMC cross-link is repaired within spot 2 (bifunctional monoadduct). spot 3 (monofunctional monoadduct. a isomer), and 24 h after drug treatment (41). and the results presented here indicate spot 4 (monofunctional monoadduct. ÃŸisomer), respectively. 459

duu- MMC adduction is not principally due lo changes in GSH. Dexam

250-â€¢f'.i ethasone is a synthelic glucocorticoid hormone that induces specific 200-Ã®lÂ«Ã¨ CYP3A isoforms (24), as well as phosphoenolpyruvate carboxykinase (64), lyrosine aminolransferase, and a number of other liver-specific â€”¿|I-rJL'-OO 150-11Oc, genes (25). It will be interesting to examine these other pathways in

100RI more detail as a means of understanding MMC metabolism in vivo. In summary, we have developed a 32P-postlabeling method for MMC adduci deteclion and have demonstrated that this assay can be a highly nA1 Oi i i* iOP cd lO1iâ€¢f 3 sensitive and useful tool for examining the biological effects of MMC +Q + in vivo. Such studies will aid in understanding the pharmacokinetics P Ã– and pharmacodynamics of MMC as well as addressing mechanislic aspects related to its metabolism and formation and repair of MMC-

n,C\J DNA adducls in vivo. The doses of MMC used in ihis sludy are HSCD4-0B:T roughly comparable to those used in human chemotherapy, and pre CO(ÃŸuu/ÃŸrf) â€”¿iTTTTTT liminary results indicate that this assay can detecl ihese same adducls in cancer patients treated with MMC. Tumor samples from patients Tr-i1T| treated with a single dose of either 5 or 10 mg/m2 of MMC exhibited all four adducts in approximately equal proportions to those seen in the chick embryo tissues in this study.7 Studies to accurately deter mine specific adduct levels in detailed time course sludies in mice and oÅ“o human patients are presently under way. The overall goal of these studies is to use this approach lo develop more effeclive uses of cancer chemotherapy agents in the treatment of human malignancies.

embryo treatment

Fig. 10. Effect of various biochemical modulators on DTD activity and GSH content ACKNOWLEDGMENTS in chick embryo liver in vivti. Chick embryos were treated as described in Fig. 9. and DTD activity (A) and GSH content (ÃŸ)were assayed as described in "Materials and Methods" We dedicate this paper to the memory of the late Karen E. Wetterhahn, in separate lobes of the same liver samples used for 12P-postlabeling. Data represent the Ph.D.. who was an invaluable scientist, mentor, colleague, and friend, and who means (or range) of values from two to four individual embryos; bars, SD. provided reagents, laboratory resources, and valuable advice on this project. We also acknowledge Noreen Ersing for technical help. We thank Dr. Kenneth Tew for valuable advice on the GSH and glutathione S-transferase studies. relatively nonspecific enzyme inhibitor, dicumarol (45, 52-54). How ever, more recent studies have suggested that DTD does not play a REFERENCES major role in MMC activation. For example, a screening of 15 1. Tomasz. M.. Lipman. R., Chowdary. D.. Pawlak. J.. Verdine, G. L., and Nakanishi, different human tumor cell lines, which differed widely in their MMC K. Isolation and structure of a covalent cross-link adduct between mitomycin C and sensitivity, demonstrated no correlation between DTD levels and DNA. Science (Washington DC), 235: 1204-1208, 1987. sensitivity to MMC (55), and NIH 3T3 cells stably transfected with 2. Tomasz, M., and Lipman, R. Reassignment of the guanine binding mode of reduced the human DT-diaphorase cDNA and exhibiting up to 15-fold ele mitomycin C. Biochemistry, 25: 4337-4344. 1986. 3. Bizanek. R., McGuinness. B. F.. Nakanishi, K., and Tomasz, M. Isolation and vated DTD activity showed no difference in sensitivity to MMC over structure of an intrastrand cross-link adduct of mitomycin C and DNA. Biochemistry, control cells (56). In the present study, we demonstrated that pretreat 31: 3084-3091. 1992. 4. Doll, C. D., Weiss. R. B., and Issell, B. F. Mitomycin: ten years after approval for ment with TCB. which elevated DTD levels by approximately 1.5- marketing. J. Clin. Oncol., 3: 276-286, 1984. fold, had no effect on MMC adduci levels in vivo. Conversely, 5. Iyer, V. N., and Szybalski, W. Mitomycins and porfiromycin: chemical mecha nism of activation and cross-linking of DNA. Science (Washington DC), 145: dexamethasone plus MMC decreased DTD activity but generated the 55-58, 1964. highest MMC adduci levels. Thus, although DTD may contribule lo 6. Hamilton, J. W., and Wetterhahn, K. E. Differential effects of chromium(VI) on MMC metabolism, it does not appear to be the predominant activation constitutive and inducible gene expression in chick embryo liver in vivo and corre lation with chromium(VI)-induced DNA damage. Mol. Carcinog., 2: 274-286, 1989. pathway in vivo. In fact, our current results, in conjunclion wilh 7. McCaffrey. J., Wolf, C. M., and Hamilton, J. W. Effects of the genotoxic carcinogen previous studies to date, suggest that MMC reductive metabolism chromium(VI) on basal and hormone-inducible phosphoenolpyruvate carboxykinase involves multiple parallel pathways //( vivo, with no one palhway gene expression in viva: correlation with glucocorticoid- and developmentally-regu- lated expression. Mol. Carcinog.. 10: 189-198, 1994. predominating Â¡nMMC activation or adduci formation. None of the 8. CarÃ³n, R. M., and Hamilton. J. W. Preferential effects of the chemotherapeutic DNA pretreatments we used, with the exception of dexamelhasone, altered crosslinking agent mitomycin C on inducible gene expression in vivo. Environ. Mol. Mutagen.. 25: 4-11, 1995. MMC adduci formation, although these agents have been shown 9. Randerath. K., and Randerath, E. 32P-Postlabeling methods for DNA adduct detec previously to have significant effects on specific Phase I and Phase II tion: overview and critical evaluation. Drug Metab. Rev., 26: 67-85, 1994. enzyme systems. 10. Keith, G-. and Dirheimer, G. Postlabeling: a sensitive method for studying DNA adducts and their role in carcinogenesis. Curr. Opin. Biotechnol., 6: 3-11. 1995. GSH has been shown to be able to bind to reductively activated 11. Maccubbin. A. E.. Mudipalli, A.. Nadadur. S. S., Ersing, N., and Gurtoo, H. L. MMC in vitro to form both GSH-MMC and GSH-MMC-DNA ad- Mutations induced in a shuttle vector plasmid exposed to monofunctionally activated ducts (57, 58). A number of cell culture studies have also indicated mitomycin C. Environ. Mol. Mutagen., 29: 143-151, 1997. 12. Warren, A. J., and Hamilton, J. W. Synthesis and structural characterization of the that modulation of GSH or glutathione 5-transferase can influence cell N2G-mitomycin C-N2G interstrand crosslink in a model synthetic 23 base pair sensitivity to MMC (59-62). In this study, dexamethasone increased oligonucleotide DNA duplex. Chem. Res. Toxicol., 9: 1063-1071, 1996. MMC adduci levels by approximately 1.7-fold and simultaneously 13. Tomasz, M.. Chawla, A. K., and Lipman, R. Mechanism of monofunctional and bifunctional alkylation of DNA by mitomycin C. Biochemistry, 27: 3182-3187, 1988. decreased GSH levels by approximately 2-fold. However, the use of 14. Hamilton. J. W., Denison. M. S., and Bloom, S. E. Development of basal and induced BSO or EA to decrease GSH levels by comparable amounts, or aryl hydrocarbon (benzo[i/]pyrene) hydroxyla.se activity in the chicken embryo in mo. Proc. Nati. Acad. Sci. USA, 80: 3372-3376, 1983. glutethimide or NAC to increase GSH levels ( 16, 63), had no effect on 15. Hamilton, J. W.. Bernent, W. J., Sinclair, P. R., Sinclair, J. F.. and Wetterhahn, K. E. MMC adduci levels, indicating thai the effect of dexamethasone on Expression of 5-aminolaevulinate synthase and cytochrome P-450 Â¡nchicken embryo 460

hepatocytes in vivo and in cell culture: effect of porphyrinogenic drugs and haem. located mitomycin C-DNA adduci in Escherii-hia coli. Proc. Am. Assoc. Cancer Res., Biochem. J., 255: 267-275, 1988. 38: 182, 1997. 16. Misra, M., Alcedo, J. A., and Wetterhahn. K. E. Two pathways for chromium(VI)- 41. Maccubbin. A. E.. Schinieri. P.. Woodman. W., Ersing, N., and Gurtoo. H. L. induced DNA damage in 14 day chick embryos: Cr-DNA binding in liver and Formalion and repair of DNA adducls Â¡ncells Ireated with mitomycin C. Proc. Am. 8-oxo-2'-deoxyguanosine in red blood cells. Carcinogenesis (Lond.), 75: 2911-2917, Assoc. Cancer Res., 38: 449. 1997. 1994. 42. Hamilton. J. W.. and Wetterhahn. K. E Chromium! VI (-induced DNA damage in chick 17. Hissin. P. J., and Hilf. R. A fluorometric method for determination of oxidized and embryo liver and blood cells in vivo. Carcinogenesis (Lond.). 7: 2085-2088, 1986. reduced glutathione Â¡ntissues. Anal. Biochem., 74: 214-226. 1976. 43. Komiyama. T.. Oki. T.. and Inui. T. Activation of mitomycin C and quinone drug 18. Benson, A. M., Hunkeler, M. J., and Talalay. P. Increase in NAD(P)H quinone metabolism by NADPH-cytochrome P-450 reducÃase. J. Pharmacol. Dyn., 2: 407- reducÃaseby dietary oxidants: possible role in protection against Carcinogenesis and 410. 1979. toxicity. Proc. Nati. Acad. Sci. USA, 77: 5216-5220. 1980. 44. Pan, S., Andrews, P. A.. Glover. C. J.. and Bachur. N. R. Reductive activalion of 19. Tomasz, M., Lipman, R., Snyder, J. K., and Nakanishi, J. Full structure of a mitomycin C and mitomycin C metabolites catalyzed by NADPH-cytochrome P-450 mitomycin C dinucleoside phosphate adduci. Use of differential FT-IR spectroscopy reducÃaseand xanthine oxidase. J. Biol. Chem.. 259: 959-966. 1984. in microscale structural studies. J. Am. Chem. Soc.. Â¡05:2059-2063, 1983. 45. Keyes, S. R., Fracasso, P. M.. Heimbrook. D. C.. Rockwell, S., Sugar, S. Ci., and 20. Kumar. S., Lipman, R.. and Tomasz. M. Recognition of specific DNA sequences by Sartorelli, A. C. Role of NADPHxylochrome c reducÃaseand DT-diaphorase in the mitomycin C for alkylation. Biochemistry. 31: 1399-1407. 1992. biotransformation of mitomycin C. Cancer Res., 44: 5638-5643, 1984. 21. Borowy-Borowski. H., Lipman. R.. Chowdary. D.. and Tomasz. M. Duplex oligode- 46. Begleiter, A., and Leith. M. K. Role of NAD(P)H:(quinone acceptor) oxidoreductase oxyribonucleotides cross-linked to mitomycin C at a single site: synthesis, properties, (DT-diaphorase) in activation of mitomycin C under acidic conditions. Mol. Phar and cross-link reversibility. Biochemistry, 29: 2992-2999, 1990. macol., 44: 210-215, 1993. 22. Hamilton. J. W., Bernent, W. J., Sinclair, P. R., Sinclair, J. F.. Alcedo. J. A., and 47. Gustafson, D. L.. and Pritsos. C. A. Bioactivation of mitomycin C by xanthine Wetterhahn, K. E. Inhibition of protein synthesis increases the transcription of the dehydrogenase from EMT6 mouse mammary carcinoma tumors. J. Nail. Cancer lust.. phenobarbital-inducible CYP2HI and CYP2H2 genes Â¡nchick embryo hepatocytes. 84: 1180-1185. 1992. Arch. Biochem. Biophys., 298: 96-104, 1992. 48. Gustafson, D. L.. and Pritsos, C. A. Enhancement of xanthine dehydrogenase medi 23. Gilday, D., Gannon, M., Yutzey, K., Bader, D., and Rifkind, A. B. Molecular cloning ated mitomycin C metabolism by dicumarol. Cancer Res.. 52: 6936-6939. 1992. and expression of two novel avian cytochrome P450 1A enzymes induced by 49. Gustafson. D. L., and Pritsos. C. A. Kinetics and mechanism of mitomycin C 2,3,7,8-letrachlorodibenzo-p-dioxin. J. Biol. Chem.. 271: 33054-33059. 1996. bioactivalion by xanthine dehydrogenase under aerobic and hypoxic conditions. 24. Sinclair, J., and Sinclair, P. Avian cytochrome P-450. In: Â¡.Schenkman and H. Greim Cancer Res., 53: 5470-5474. 1993. (eds.), Cytochrome P450. Heidelberg: Springer-Verlag. 1993. 50. Goeptar, A. R.. Groot. E. J.. Scheerens. H.. Commandeur. J. N. M., and Vermeiden. 25. Parkinson, A. Biotransformation of xenobiotics. In: C. D. Klaassen, M. O. Amdur. N. P. E. Cytoloxicity of mitomycin C and Adriamycin in freshly isolated ral hepa and J. Doull (eds.), Casarett and Doull's Toxicology: The Basic Science of Poisons, tocytes: Ihe role of cytochrome P450. Cancer Res.. 54: 2411-2418. 1994. pp. 139-162. New York: McGraw-Hill. 1996. 51. Hodnick. W. F.. and Sartorelli. A. C. Reductive activation of mitomycin C by 26. Tomasz, M.. Lipman, R., McGuinness, B. F.. and Nakanishi. K. Isolation and NADH:cytochrome fc, reducÃase.Cancer Res.. 53: 4907-4912, 1993. characterization of a major adduci between mitomycin C and DNA. J. Am. Chem. 52. Siegel, D.. Beali, H.. Senekowiisch, C., Kasai. M.. Arai. H.. Gibson. N. W., and Ross. Soc., 110: 5892-5896. 1988. D. Bioreduclive activation of mitomycin C by DT-diaphorase. Biochemistry, 31: 27. Tomasz, M., Hughes, C. S., Chowdary, D., Keyes, S. R., Lipman, R., Sartorelli, A. C., 7879-7885. 1992. and Rockwell, S. Isolalion, identificalion, and assay of pHJ-porfiromycin adducts of 53. Siegel. D., Gibson. N. W., Preusch, P. C.. and Ross, D. Metabolism of mitomycin C EMT6 mouse mammary tumor cell DNA: effecls of hypoxia and dicumerol on adduci by DT-diaphorase: role in mitomycin C-induced DNA damage and cytotoxicity in pallerns. Cancer Commun.. 3: 213-223, 1991. human colon carcinoma cells. Cancer Res.. 50: 7483-7489. 1990. 28. Bizanek, R., Chowdary, D., Arai. H.. Kasai, M.. Hughes. C. S., Sartorelli, A. C.. 54. Prakash. A. S., Beali. H.. Ross. D.. and Gibson. N. W. Sequence-selective alkylation Rockwell. S.. and Tomasz, M. Adducts of mitomycin C and DNA in EMT6 mouse and cross-linking induced by mitomycin C upon activalion by DT-diaphorase. Bio mammary tumor cells: effects of hypoxia and dicumarol on adduci pallerns. Cancer chemistry. 32: 5518-5525. 1993. Res., 53: 5127-5134, 1993. 55. Robertson. N.. Slralford. I. J.. Houlbrook. S.. Carmichael. J.. and Adams, G. E. The 29. Dorr, R. T.. Bowden, G. T., Alberts, D. S.. and Liddil, J. D. Interactions of mitomycin sensitivity of human tumour cells to quinone bioreduclive drugs: what role for C wilh mammalian DNA delecled by alkaline elulion. Cancer Res.. 45: 3510-3516. DT-diaphorase? Biochem. Pharmacol.. 44: 409-412, 1992. 1985. 56. Powis, G., Gasdaska, P. Y.. Gallegos, A.. Sherrill. K.. and Goodman. D. Over- 30. Kalo. S.. Yamashila, K.. Kim, T., Tajiri. T.. Onda. M., and Shigeak. S. Modificalion expression of DT-diaphorase in transfected NIH 3T3 cells does not lead to increases of DNA by mitomycin C in cancer palients detected by '"P-postlabeling analysis. in anticancer quinone drug sensitivity: a questionable role for the en/yme as a large! MutÃ¢t.Res., 202: 85-91, 1988. for bioreductively activated anticancer drugs. Anticancer Res., /5: 1141-1145. 1995. 31. Pan, S., Yu, F., and Hipsher, C. Enzymatic and pH modulation of mitomycin 57. Sharma. M.. and Tomasz. M. Conjugation of glutathione and other thiols with C-induced DNA damage in mitomycin C-resistanl HCT 116 human colon cancer bioreductively activated mitomycin C. Effect of thiols on the reductive activation rale. cells. Mol. Pharmacol., 43: 870-877, 1993. Chem. Res. Toxicol., 7: 390-400, 1994. 32. Reddy. M. V., and Randeralh, K. 12P analysis of DNA adducls in somatic and 58. Sharma. M.. He. Q.. and Tomasz. M. Effects of glutathione on alkylation and reproduclive tissues of rats Ireated with the anlicancer anlibiolic. mitomycin C. Mutai. cross-linking of DNA by mitomycin. Isolation of a ternary glutathione-milomycin- Res., Â¡79:75-88. 1987. DNA adduci. Chem. Res. Toxicol.. 7: 401-407, 1994. 33. den Hartigh, J., McVie, J. G., van On, W. J.. and Pinedo, H. M. Pharmacokinelics of 59. Singh. S. V.. Xu. B. H.. Maurya. A. K.. and Mian. A. M. Modulation of mitomycin mitomycin C in humans. Cancer Res.. 43: 5017-5021. 1983. C resistance by gluiathione transferase inhibitor ethacrynic acid. Biochim. Biophys. 34. Bestor, T. H. Methylalion pallerns in Ihe vertebrate genome. J. NIH Res.. 5: 57-60, Acta, 1137: 257-263. 1992. 1993. 60. Xu. B. H.. and Singh. S. V. Effect of buthioninc sulfoximine and ethacrynic acid on 35. Johnson, W. S.. He. Q. Y.. and Tomasz, M. Seleclive recognilion of the M^CpG cytotoxic aclivily of mitomycin C analogues BMY 25282 and BMY 25067. Cancer dinucleotide sequence in DNA by mitomycin C for alkylation and cross-linking. Res., 52: 6666-6670. 1992. Bioorg. Med. Chem., 3: 851-860. 1995. 61. Perry. R. R.. Kang. Y.. and Greaves, B. Biochemical characterization of a mitomycin C 36. Romanoff, A. Biochemistry of the Avian Embryo. New York: John Wiley and Sons. resistant colon cancer cell line variant. Biochem. Pharmacol., 46: 1999-2(X)5. 1993. 1967. 62. Perry. R. R.. Greaves, B. R.. Rasberry. U.. and Barranco, S. C. Effect of treatment 37. Hamilton. J. W.. and Bloom, S. E. Correlation between induction of xenobiotic duration and glutathione depletion on mitomycin C cytotoxicity in vilro. Cancer Res.. melabolism and DNA damage from chemical carcinogens in Ihe chick embryo in 52: 4608-4612, 1992. vivo. Carcinogenesis (Lond.), 7: 1101-1106. 1986. 63. Tew, K. D. Glutalhione-associaled enzymes in anlicancer drug resistance. Cancer 38. Strauss, B. S. DNA repair: Ihe nick in time. Nature (Lond.), 366: 301. 1993. Res.. 54: 4313-4320, 1994. 39. Hanawalt, P. C. Transcriplion-coupled repair and human disease. Science (Washing- 64. McCaffrey, J.. and Hamilton. J. W. Developmenlal regulalion of basal and honmme- ion DC), 266: 1957-1958, 1994. inducible phosphoenolpyruvale carboxykinase gene expression in chick embryo liver 40. Ramos, L. A., Lipman, R., Tomasz, M., and Basu. A. K. Effect of a site-specifically in vivÂ».Arch. Biochem. Biophys.. 309: 10-17, 1994.

461

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1998 American Association for Cancer Research. Detection of Mitomycin C-DNA Adducts in Vivo by 32 P-Postlabeling: Time Course for Formation and Removal of Adducts and Biochemical Modulation

Amy J. Warren, Alexander E. Maccubbin and Joshua W. Hamilton

Cancer Res 1998;58:453-461.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/58/3/453

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/58/3/453. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.