- Email: [email protected]
In vitro binding of β-propiolactone to calf thymus DNA and mouse liver DNA to form 1-(2-carboxyethyl)adenine
Chem.-Biol. Interactions, o Elsevier/North-Holland
18 (1977) 327-336 Scientific Publishers, Ltd.
327
IN VITRO BINDING OF /3-PROPIOLACTONE TO CALF THYMUS DNA AND MOUSE LIVER DNA TO FORM l-(2CARBOXYETHYL)ADENINE*
URSZULA MATB, JEROME J. SOLOMON and ALVIN SEGAL** Laboratory of Organic Chemistry and Carcinogenesis, Institute of Environmental New York University School of Medicine, New York, N. Y. 10016 (U.S.A.)
Medicine,
(Received May 2nd, 1977) (Revision received June 2Oth, 1977) (Accepted June 30th, 1977)
SUMMARY
In vitro reaction of p-propiolactone (BPL) with calf thymus DNA and mouse liver DNA followed by acid (HCl) hydrolyses of the BPL-reacted DNA’s resulted in the isolation of a new compound, 1-(2-carboxyethyl)adenine (l-CEA). The structure was assigned on the basis of ultraviolet spectra at acidic, alkaline and neutral pH and electron impact and chemical ionization mass spectra as well as chemical synthesis of l-CEA from BPL and 2’-adenosine-5’-monophosphoric acid. The only other compound previously isolated from the in vitro and in vivo reactions of BPL and DNA was 7-(2-carboxyethyl)guanine (7-CEG) which we also identified as a product of our in vitro reaction. Under the conditions used the main product of alkylation was l-CEA and the ratios of the concentrations of l-CEA to 7-CEG was approx 3 : 1. The possible.effect of the formation of l-CEA on the structure of DNA and its role in chemical carcinogenesis is discussed.
INTRODUCTION
BPL is a direct-acting carcinogen in mouse skin [l] and in a variety of rodent species and tissues [2]. The lactone is an initiator of tumorigenesis in mouse skin [ 31 and is mutagenic in Neurosporu crussu [ 41 and has also been reported to transform Wistar rat embryo cells in culture [5]. On a molecular *This research was supported by USPHS Grants CA 16992 and CA 13343 from the National Cancer Institute and Grant ES 00260 from the National Institute of Environmental Health Sciences. **To whom requests for reprints should be sent. Abbreviations: BPL, p-propiolactone; l-CEA, 1-(2-carboxyethyl)adensine; 7-CEG, 7-(2carboxyethyl)guanine; PB, phosphate buffer 0.2 M, pH 7.2.
328 level BPL reacts with derivatives of guanine and in vitro with RNA [6,7], in vitro with DNA [8,9] and in vivo with DNA and RNA to form 7-CEG [3,10]. We have shown that BPL reacts in vitro with histones, non-histone chromosomal proteins and DNA in mouse skin chromatin [9], that the binding of BPL is selective to lysine rich histones Hl and Hl” [9] and that in vitro BPL acylates the c-amino group of L-lysine in calf thymus histones [ll]. BPL also appears capable of causing the formation of a complex between proteins and DNA [12]. The chemical nature of this complex is not known. BPL has been reported to cause in vitro intermolecular linking as well as fragmentation of DNA [13]. Recently Hennings et al. [14] studied the repair of BPL-induced damage to mouse skin cells in culture. Their results suggested that BPL reacted with the purines, adenine and guanine in nuclear DNA and although this is the first reported suggestion based on experimental evidence that BPL reacts with adenine in DNA no BPL-adenine derivative was isolated. We report in this paper that the in vitro reaction of BPL with either calf thymus DNA or mouse liver DNA results in reaction with adenine to form l-CEA (Fig. 1) as well as with guanine to form 7-CEG (Fig. 1).
EXPERIMENTAL
Materials Calf thymus DNA was obtained from the Sigma Chemical Co., St. Louis, MO. Mouse liver DNA was isolated from the livers of Q ICR/Ha mice obtained from ARS/Sprague-Dawley, Madison, Wise. by the method of Marmur [15]. The DNA was further deproteinized by treatment with pronase [ 161. The mouse liver DNA thus obtained had a Tm of 74°C in dilute saline citrate [9] with a hyperchromicity of 33%. BPL (99% purity) was purchased from BDH Chemicals Ltd., Poole, England. 2’-Deoxyadenosine-5’-monophosphoric acid, 2’-deoxycytidine-5’-monophosphoric acid, 2’-deoxyguanosine-5’-monophosphoric acid, thymidine-5’-monophosphoric acid, adenine, guanine, l-methyladenine and Sephadex G-10 were purchased from Sigma Chemical Co., St. Louis, MO. Whatman 3MM paper sheets from chromatography were obtained from Fisher Scientific Co., Springfield, N.J. Cellulose plates (20 X 20 cm., 0.10 mm thick) for analytical and preparative thin-layer chromatography was purchased from Brinkmann Instruments Inc., Westbury, N.Y. N,O-Bis(trimethylsilyl)trifluoroacetamine (“BSTFA”) and pyridine, silylation grade, were from Pierce Chemical Co., Rockford, Ill. The solvent system used for thin-layerchromatography on cellulose plates was isopropanol-H, Ocont. HCl, 6.5 : 1.84 : 1.66 (v/v/v) [17]. The solvent systems used for paper chromatography were: System A, ammonium acetate, 1 M, pH 7.5-100% ethanol, 3.5 : 6.5 (v/v); System B, isopropanol-conc. ammonia-HzO, 7 : 1 : 2; System C, isopropanol-HzO, 7.3 and System D, methanol--cone. HCl-H20, 7 : 2 : 1.
329 Instrumentation
UV spectra were recorded in 1 cm cells in HzO, in HCl at pH 1 and NaOH at pH 13 in a Beckman Model 25 spectrophotometer. Spectra recorded at pH 7.2 were taken in PB [6]. Electron impact (EI) and chemical ionization (CI) mass spectra were acquired using a DuPont 21-492 High Resolution Mass Spectrometer equipped with a dual EI/CI ionization source. The ionizing voltage was 70 eV. Samples were introduced as solids in glass capillary tubes via the direct probe. The source temperature was maintained at 260°C. The probe temperature was gradually increased until spectra were obtained and this temperature is noted for the mass spectra of compounds in the Results section. CI spectra were obtained using Matheson instrument grade (99.5%) isobutane maintained at -0.5 torr. Paper chromatography
Paper chromatograms were run on paper sheets, 17 X 55 cm for 18 h using a descending method. Compounds were identified on the paper chromatograms by observing the chromatograms under UV light at 254 nm. Compounds were isolated from the chromatograms by elution with water and then concentrated by lyophilization. Column
chromatography
Column chromatography was performed on a column of Sephadex G-10 (1.5 X 85 cm) and HZ0 or ammonium formate, 0.05 M, pH 6.8 were used as elm&s [18]. Flow rate was 45 ml/h and 2 ml fractions were collected in a Gilson automatic fraction collector (Gilson Medical Electronics, Middleton, Wise.). Fractions were monitored at 265 nm. Thin-layer
chromatography
This was performed on 20 X 20 cm cellulose plates with the solvent system indicated above. The solvent system was allowed to run to the top of the plate which was about 18.5 cm above the point at which compounds were spotted. METHODS
Reactions
of BPL with calf thymus DNA and mouse liver DNA
These reactions were according to the method of Segal et al. [9]. DNA was dissolved in PB (1 mg/ml) and 100 pl(ll5 mg, 1.6 X 10s3 M) of BPL/mg of DNA was added to the DNA solution at 0-5°C. The mixture was stirred at O--5% for 20 h. The reaction mixture was then dialyzed at 0-5”C, for 48 h against repeated changes of HzO. Purification
and isolation of BPL-reacted
This was performed [ 191. The BPL-reacted
DNA
according to the method of Lawley and Thatcher DNA was precipitated from the aqueous solution by
330 the addition of 2 volumes of 2-ethoxyethanol containing 0.1 volume of 2.5 M sodium acetate. The supernatant was decanted and the precipitated DNA was redissolved in PB and the procedure repeated twice more yielding a white fibrous BPL-reacted DNA preparation. Hydrolysis of l?PL-reacted DNA Method 1. To a solution of 1 mg of BPL-reacted DNA [or control (unreacted) DNA] in 1 ml Hz0 was added sufficient cont. HCl to make a solution of 0.1 N. The acid solution was heated to 100°C for 1 h [20], and then lyophilized to dryness. Method 2. The solution of DNA at 0.1 N HCl was heated at 37°C for 16 h [19]. The hydrolyzed DNA after all hydrolyses was lyophilized to dryness and dissolved in a small quantity of Hz0 prior to application to paper chromatograms. Preparation of compound 7-CEG (Fig. 1) ‘I-CEG was prepared from the reaction of BPL with dGuo-5’-P by the method of Colburn et al. [7]. The 7-CEG thus obtained was a white amorphous powder with UV spectra at pH 1, 7.2 and 13 identical to that reported by Colburn et al. [7]. Preparation of compound 1 -CEA dAdo-5’-P (100 mg, 2.88 X low4 M) was dissolved in 6.0 ml of Hz0 and the pH adjusted to 7.2. BPL (50 (~1,57 mg) was added at 30 min intervals, 6 times, the pH maintained between 7.0 and 7.4 by dropwise addition of NaOH. The temperature was maintained at 37°C. Total BPL added was 300 ~1 (345 mg, 4.8 X 10e3 M). After addition of the last sample of BPL the reaction mixture was kept at 37°C for an additional 1.5 h. The reaction mixture was then extracted with 3 X 1.5 ml portions of ethyl ether, made up to pH 1 with cont. HCl and heated to 100°C for 1 h. Portions of the reaction mixture not used for hydrolysis were stored at -20°C. The hydrolysate was applied to paper chromatograms and eluted with solvent system A. Synthetic l-CEA with an RF identical to l-CEA from DNA was identified and eluted from the paper chromatogram.
N
/H
0
Haglj$H 1 I-CEA Fig. 1. Structures ,m nlT-1
7-CEG of l-(2-carboxyethyl)adenine
(l-CEA)
and 7-(2-carboxyethyl)
guanine
331 Preparation of the trimethylsilyl derivatives of compound l-CEA for mass spectrometry Compound l-CEA (50 pg) (isolated from calf thymus or mouse liver DNA) was mixed with “BSTFA” (50 ~1) and pyridine for silylation (25 ~1) in a closed vial and heated at 55°C for 10 min. A 10 ~1 portion of the solution was injected into a glass capillary tube which was inserted into the mass spectrometer probe. RESULTS
The HCl hydrolysates (Method 1) of BPL-reacted calf thymus DNA and mouse liver DNA were each chromatographed on cellulose plates. The major bands were eluted with water and applied to paper chromatograms. When a band containing compound ‘I-CEG was applied to a paper chromatogram and eluded with solvent system A, a new UV absorbing compound was observed which was not present in paper chromatograms of control DNA hydrolysates. The new compound was designated compound l-CEA (Fig. 1) and gave single spots on paper chromatograms with solvent systems A, B, C and D. RF values obtained were, system A: l-CEA, 0.45; 7-CEG, 0.41; guanine, 0.57; adenine, 0.67; system B: l-CEA, 0.36; ‘I-CEG, 0.19; guanine, 0.23; adenine, 0.44; system C: l-CEA, 0.25; 7-CEG. 0.21; guanine, 0.42.; adenine, 0.65; system D: l-CEA, 0.42; 7-CEG, 6.27; adenine, 0.27. The UV spectra of compound l-CEA from calf thymus and mouse liver DNA were, PH 1, ha 260, hi, 232; PH 7-2 (PB), A,, 268, kin 240, PH 13, x max 271, h min 242 nm. EI and CI mass spectrometry was performed on l-CEA. The probe temperatures were varied between the minimum required to obtain a spectrum (250°C) and 400°C and for both EI and CI mass spectra the various peaks varied slightly with the probe temperatures. The EI mass spectrum of l-CEA gave peaks at m/e 189 (7), 135 (loo), 108 (31), 81 (13) and 72 (55) but the postulated molecular ion peak at m/e 207, the MW of l-CEA was not observed. At this stage of the experimentation we were uncertain as to whether the highest mass peak at m/e 189 in the EI mass spectra of l-CEA represented fragmentation of the molecular ion (m/e 207) or a pyrolytic decomposition to a neutral species of MW 189 prior to volatilization and ionization. The fact that adenine was detected in the capillary tube after EI mass spectrometry, by paper chromatography and UV spectra, strongly suggested that pyrolytic dissociation was playing an important role in the EI spectra of l-CEA that we were obtaining. The EI mass spectra have been obtained for ethyl and isopropyl derivatives of adenine by Lawley et al. [21]. They obtained molecular ion peaks for the ethyl and isopropyladenines and large peaks at m/e 135, 108 and 81. The authors postulated that the principal fragmentation process of the molecular ion in the higher adenines was a rearrangement involving a proton transfer from the alkyl moiety to adenine resulting in the net loss of alkyl-H (ethylene and isopropylene in the cases cited) and formation of adenine’ at m/e 135 and its
332 fragmentation products at 108 and 81 due to successive losses of two molecules of HCN. The EI spectra of adenine derivatives due to loss of HCN obtained by Lawley et al. [21] was essentially identical to relative peak intensity at m/e 135, 108 and 81 observed in our EI spectra of compound l-CEA and an authentic sample of adenine. Evidence for the pyrolytic decomposition of compound l-CEA rather than fragmentation of the molecular ion under the EI conditions used was obtained from CI mass spectrometry. Peaks were obtained at m/e 190 (189 + H’) (49), 136 (135 + H+) (84) and 73 (72 + H’) (100). In addition small addition ions were obtained at m/e 176 [135 + 41’ (C,H;)] , 178 [135 + 43’ (C,H,+)] , 230 (189 + 41’) and 232 (189 + 43’). The isobutane CI spectra of authentic adenine (MW 135) was obtained and exhibited a base peak at m/e 136 (135 + H’) and small addition ions at 176 (135 + 41+) and 178 (135 + 43+). We thus conclude that the peaks at 135 and 189 are neutral molecules formed by pyrolysis prior to ionization, as addition ions were observed which can only be formed by the addition of positively charged ions to neutral molecules [22]. The dissociation of 9-(2carboxyethyl)adenine to adenine and acrylic acid in the presence of water of hydration and at elevated temperatures reported by Lira and Huffman [23] was interpreted as a reverse Michael reaction as the dissociation did not occur in the absence of water of hydration. That heat induced dissociation of the carboxyethyl group (acrylic acid) from adenine in our experiments but not the ethyl and isopropyl groups from adenine in the experiments of Lawley et al. [21] may be due to the stability of the product, the a&unsaturated acrylic acid CH, =CH-COOH. The EI spectra of the trimethylsilyl derivative of l-CEA was run without heating the probe. Peaks were obtained at m/e 423 (1.8), 408 (2.3), 2’79 (62), 264 (100) and 73 (42). Tests for phosphorus [24] and deoxyribose [25] in l-CEA were negative. Reaction of BPL with dAdo-5’-P resulted in the isolation of a compound from paper chromatograms with solvent system A, which was designated synthetic l-CEA. Synthetic l-CEA had identical RF vdues on paper chromatograms with solvent systems A, B and C, identical I_JV spectra at pH 1, 7.2 and 13 and similar EI and CI spectra except for small variations in the relative intensities of the peaks obtained. NO compound identical to compound l-CEA was produced when BPL was reacted with either dCyd-5’-P, dGuo-5’-P or dThd-5’-P under identical conditions as the reaction between BPL and dAdo-5’-P. When BPL-reacted DNA was hydrolyzed at 0.1 N HCl for 1 h at 100°C (Method 1) the ratio of l-CEA to 7-CEGwas 3 : 1 assuming that the extinction coefficient of l-CEA was similar to l-methyl and other l-alkyladenines [26] (see Discussion section). When the BPL-reacted DNA was hydrolyzed at 0.1 N HCl for 16 h at 37°C (Method 2) the ratio of l-CEA to 7-CEG was 1 : 3, i.e., the ratio of the alkylated purines released was reversed. Compound l-CEA was eluted from the Sephadex G-10 column in fractions 45-60 and 7-CEG in fractions 68-77 using either water or ammonium formate (0.05 M, pH 6.8) as eluant.
333 DISCUSSION
A new alkylated base; designated compound l-CEA (Fig. 1) was isolated from the HCl hydrolysis products of BPL-reacted calf thymus or mouse liver DNA and assigned the structure 1-(2-carboxyethyl)adenine on the basis of UV, EI and CI mass spectra and chemical synthesis from BPL and dAdo-5’-P. Also isolated was the known 7-(2-carboxyethyl)guanine(7-CEG, Fig. 1) [6,7] which was identified by comparison to 7-CEG synthesized by us from BPL and dGuo-5’-P by the method of Colburn et al. [7]. The UV spectra of l-CEA was essentially identical to an authentic sample of 1-methyladenine and a listing of 1-alkyladenines at acid, alkaline and neutral pH values [27] and differed from the UV spectra of an authentic sample of 3-methyl-adenine and from UV spectra of monoalkylated C-2, N-3, N6, C-7, C-8 and N-9 substituted adenines [27,28]. A molecular ion was not observed in the CI and EI mass spectra due to thermal decomposition. Strong support for a molecular weight of 207 for compound l-CEA (C,H,N,02) was obtained from the EI spectra of the more volatile trimethylsilyl derivative of l-CEA. A large molecular ion peak was recorded at m/e 279 (l-CEA-H + trimethylsilyl), the trimethylsilyl ester of l-CEA. The large peaks formed at m/e 264 and 73 are characteristic of the EI mass spectra of trimethylsilyl derivatives which easily lose methyl [M-15+ m/e 2641 and a positively charged trimethylsilyl group [(CH,),Si+ m/e 731 [29,30] . Small peaks at m/e 423 and 408 (loss of methyl) suggest that the protons at N6 and N-9 of l-CEA have some lability and that a small amount of trimethylsilyl derivative can also be formed at positions N6 and N-9. The structure of the compound giving rise to a peak at m/e 189 in the El mass spectrum and 190 in the CI mass spectrum of l-CEA represents the loss of water from l-CEA prior to volatilization with the possible formation of a structure such as l-CEA Lactam (Fig. 2). Meot-Ner and Field [31] report that several amino acids form lactams, by losing water prior to ionization and protonated molecular ion formation (CI). When DNA was hydrolyzed at 100°C (Method 1) more l-CEA was isolated than 7-CEG while when DNA was hydrolyzed at 37°C (Method 2) the reverse was true. This is in agreement with the work of Frei and Lawley [18] who found that hydrolysis of alkylated DNA in HCl at lower temperatures released 3 and 7 methyl-adenines and guanines, while a temperature of 100°C was required to release 1-methyladenine, suggesting that the 1-alkyl derivatives of adenine are more stable to depurination than alkylation at the other common positions of adenine and guanine. Additional strong support for the structure of compound l-CEA comes from its synthesis from BPL and dAdo-5’-P. Synthetic l-CEA was found to be identical to l-CEA obtained from DNA in all parameters measured (see Results section). Compound l-CEA is a new compound and this is the first BPL-derivative of adenine isolated following in vitro or in vivo reaction of BPL with DNA. The results tend to confirm the conclusion of Hennings et al. [14] that BPL reacts with adenine and guanine in nuclear DNA in
334 0
I-CEA LACTAM Fig. 2. Structure of l-CEA Lactam, postulated as a possible structure of the dehydration product of l-CEA giving a peak at m/e 189 in the electron impact and 190 (189 + H+) in the chemical ionization mass spectra.
epidermal cells in culture. There is a high probability that bases alkylated in vivo by direct-acting carcinogens are similar to those alkylated in vitro [2]. One of the main reasons for acid hydrolysis of BPL-reacted DNA at 37°C (Method 2) was to try to determine if 6-(2carboxyethoxy)guanine was formed by reaction of BPL and DNA. It was reported that 6_methoxyguanine, a compound of importance in both mutagenesis and carcinogenesis [2], is released and is stable at pH 1 and 37°C for 16 h. This stability was confirmed by us using an authentic sample of 6-methoxyguanine. However we were not able to detect a BPL-6-alkoxy derivative of guanine. 6-Alkoxyderivatives of guanine are relatively easy to detect on paper chromatograms as they give an intense blue fluorescence when exposed to UV light at 254 nm. It should be noted that our methods of hydrolyses of BPL-reacted DNA at a highly acidic pH could easily dissociate or destroy more labile derivatives of BPL and bases or phosphate in DNA. A less destructive method of identifying labile products of BPL would employ the use of enzymatic hydrolyses [32,33] and it may be that more sites of BPL alkylation or acylation exist in DNA. BPL is a powerful carcinogen in several species of rodents, and cancers appear to result wherever BPL reacts directly with the rodent tissue. How alkylation of the l-position of adenine could contribute to this carcinogenicity is not known. It is certain that formation of l-(2-carboxyethyl)adenine in DNA would interrupt Watson-Crick base-pairing between adenine and thymine by blocking one of the positions involved in the hydrogenbonded association between the two bases. Whether the presence of l-CEA in DNA results in inactivation of DNA, produces mutations or is subject to the various forms of repair processes, error free and/or error prone (post replication repair) [ 2,34-361 is unknown at present. ACKNOWLEDGEMENTS
We wish to thank Professor P.D. Lawley of the Chester Beatty Research Institute, Buckinghamshire, England for generous gifts of 3-methyladenine and 6-methoxyguanine and Dr. Gisella Witz of our department for help in spectral determinations.
335 REFERENCES 1 2
8 9
10 11
12 13 14
15 16 17 18
19 20 21 22 23 24
F.S.C. Roe and O.M. Glendenning, The carcinogenicity of P-propiolactone for mouse skin, Br. J. Cancer, 10 (1956) 357. P.D. Lawley, Carcinogenesis by alkylating agents, in C.E. Searle (Ed.), Chemical Carcinogens, ACS Monograph 173, American Chemical Society, Washington, D.C., 1976, pp. 83-244. R.K. Boutwell and N.H. Colburn, The binding of P-propiolactone to mouse skin DNA in vivo; its correlations with tumor-initiating activity, Cancer Res., 26 (1966) 1701. H.H. Smith and A.M. Srb, Induction of mutations with P-propiolactone, Science, 114 (1951) 490. A.K. Powell, The effects of propiolactone on rat fibrocytes cultivated in vitro, Br. J. Cancer, 20 (1966) 402. J.J. Roberts and G.P. Warwick, The reaction of p-propiolactone with guanosine, deoxyguanylic acid and RNA, Biochem. Pharmacol., 12 (1963) 1441. N.H. Colburn, R.G. Richardson and R.K. Boutwell, Studies of the reaction of P-propiolactone with deoxyguanosine and related compounds, Biochem. Pharmacol., 14 (1965) 1113. J.J. Roberta and G.P. Warwick, The reaction of some compounds of biological interest with nucleic acids, Biochem. J., 87 (1963) 14. A. Segal, M. Schroeder, P. Barnett and B.L. Van Duuren, Studies on the effects in vitro of p-propiolactone and p-propiolact[14C]one on whole mouse skin chromatin, Biochem. Pharmacol., 23 (1974) 937. N.H. Colburn and R.K. Boutwell, The in vivo binding of p-propiolactone to mouse skin DNA, RNA and protein, Cancer Res., 28 (1968) 642. A. Segal and S.J. Garte, In vitro acylation of the e-amino group of L-lysine in calf thymus histones by the carcinogen, p-propiolactone, Chem.-Biol. Interact. 15 (1976) 319. W.C. Nietert, L.M. Kellicutt and H. Kubinski, DNA-protein complexes produced by a carcinogen, p-propiolactone, Cancer Res., 34 (1974) 859. K. Kubinski and E.H. Szybalski, Intermolecular linking and fragmentation of DNA by p-propiolactone, a monoalkylating carcinogen, Chem.-Biol. Interact. 10 (1975) 41. H. Hennings, D.M. Michael, S.D.A. Eaton and D.L. Morgan, Purinespecific repair of p-propiolactone-induced DNA damage in mouse skin cells, J. Invest. Dermatol., 62 (1974) 480. J. Marmur, A procedure for the isolation of deoxyribonucleic acid from microorganisms, J. Mol. Biol., 3 (1961) 208. K. Marashige and H. Ozaki, Properties of isolated chromatin from sea urchin embryo, Dev. Biol., 16 (1967) 474. K. Fink and W. Adams, Paper chromatographic data for purines, pyrimidines and derivatives in a variety of solvents, J. Chromatogr., 22 (1966) 118. J.V. Frei and P.D. Lawley, Methylation of DNA in various organs of C57 Bl mice by a carcinogenic dose of N-methyl-N-nitrosourea and stability of some methylation products up to 18 h, Chem.-Biol. Interact., 10 (1975) 413. P.D. Lawley and C.J. Thatcher, Methylation of deoxyribonucleic acid in cultured mammalian cells by N-methyl-N’nitro-N-nitrosoguanidine, Biochem. J., 116 (1970) 693. S.C. Maitre and J.V. Frei, Organ specific effects of DNA methylation by alkylating agents in the imbred Swiss mouse, Chem.-Biol. Interact., 10 (1975) 285. P.D. Lawley, D.J. Orr and M. Jarman, Isolation and identification of products from alkylation of nucleic acids: ethyl and isopropyl purines, Biochem. J., 145 (1975) 73. B. Munson, Chemical ionization mass spectrometry, Anal. Chem., 43 (1971) 28A. E.P. Lira and C.W. Huffman, Some Michael-type reactions with adenine, J. Org. Chem., 31 (1966) 2188. P.S. Chen, Jr., T.Y. Toribara and H. Warner, Microdetermination of small phosphorus,
336 Anal. Chem., 28 (1956) 1756. J.G. Buchanin, Detection of deoxyribonucleosides on paper chromatogram, Nature, 168 (1951) 1091. 26 D.D. Dunn and R.H. Hall, Purines, pyrimidines, nucleosides, and nucleotides: Physical constants and spectral properties, in G.D. Fasman (Ed.), Handbook of Biochemistry and Molecular Biology, 3rd ed., Nucleic Acids, Vol. 1, CRC Press, Cleveland, Ohio, 1975, pp. 65-215. 27 B. Singer, UV Spectral Characteristics and Acidic Dissociation Constants of 280 Alkyl Bases, Nucleosides and Nucleotides, in G.D. Fasman (Ed.), Handbook of Biochemistry and Molecular Biology, 3rd ed.; Nucleic Acids, Vol. 1, CRC Press, Cleveland, Ohio, 1975, pp. 409-449. 28 C.V.Z. Smith, R.K. Robins and R.L. Tolman, Clarification of the ring closure of certain 4,5-diaminopyridines to 6-substituted 8-alkylpurines, J. Chem. Sot., Perkin Trans., 1 (1973) 1855. 29 H. Budzikiewicz, C. Djerassi, D.H. Williams, Mass Spectrometry of Organic Compounds, Holden-Day, San Francisco, Calif., 1967, p. 473. 30 F.W. McLafferty, Interpretation of Mass Spectra, 2nd ed., W.A. Benjamin, Reading, Mass., 1973, p. 196. 31 M. Meot-Ner and F.H. Field, Chemical ionization of mass spectrometry. XX. Energy effects and virtual ion temperature in the decomposition kinetics of amino acid derivatives. J. Am. Chem. Sot., 95 (1973) 7207. 32 D.H. Swensen, P.B. Farmer and P.D. Lawley, Identification of the methyl phosphotriester of thymidylyl(d-5”)tNymidine as a product from reaction of DNA with the carcinogen N-methyl-N-nitrosourea, Chem.-Biol. Interact., 15 (1976) 91. 33 B. Singer, All oxygens in nucleic acids react with carcinogenic ethylating agents, Nature, 264 (1976) 333. 34 J.E. Cleaver, Excision repair: our knowledge based on human (Xeroderma Pigmentosum) and cattle cells, in R.F. Beers Jr., R.M. Herriott and R.C. Tilghman (Eds.), Molecular and Cellular Repair Processes, 5th International Symposium on Molecular Biology, The Johns Hopkins University Press, Baltimore, Md., 1972, pp. 195-211. 35 A. Kornberg, DNA Synthesis, W.H. Freeman, San Francisco, Calif., 1974,pp. 297-327. 36 P. Kleihues and H.K. Cooper, Repair excision of alkylated bases from DNA in vivo, Oncology, 33 (1976) 86.
25
18 (1977) 327-336 Scientific Publishers, Ltd.
327
IN VITRO BINDING OF /3-PROPIOLACTONE TO CALF THYMUS DNA AND MOUSE LIVER DNA TO FORM l-(2CARBOXYETHYL)ADENINE*
URSZULA MATB, JEROME J. SOLOMON and ALVIN SEGAL** Laboratory of Organic Chemistry and Carcinogenesis, Institute of Environmental New York University School of Medicine, New York, N. Y. 10016 (U.S.A.)
Medicine,
(Received May 2nd, 1977) (Revision received June 2Oth, 1977) (Accepted June 30th, 1977)
SUMMARY
In vitro reaction of p-propiolactone (BPL) with calf thymus DNA and mouse liver DNA followed by acid (HCl) hydrolyses of the BPL-reacted DNA’s resulted in the isolation of a new compound, 1-(2-carboxyethyl)adenine (l-CEA). The structure was assigned on the basis of ultraviolet spectra at acidic, alkaline and neutral pH and electron impact and chemical ionization mass spectra as well as chemical synthesis of l-CEA from BPL and 2’-adenosine-5’-monophosphoric acid. The only other compound previously isolated from the in vitro and in vivo reactions of BPL and DNA was 7-(2-carboxyethyl)guanine (7-CEG) which we also identified as a product of our in vitro reaction. Under the conditions used the main product of alkylation was l-CEA and the ratios of the concentrations of l-CEA to 7-CEG was approx 3 : 1. The possible.effect of the formation of l-CEA on the structure of DNA and its role in chemical carcinogenesis is discussed.
INTRODUCTION
BPL is a direct-acting carcinogen in mouse skin [l] and in a variety of rodent species and tissues [2]. The lactone is an initiator of tumorigenesis in mouse skin [ 31 and is mutagenic in Neurosporu crussu [ 41 and has also been reported to transform Wistar rat embryo cells in culture [5]. On a molecular *This research was supported by USPHS Grants CA 16992 and CA 13343 from the National Cancer Institute and Grant ES 00260 from the National Institute of Environmental Health Sciences. **To whom requests for reprints should be sent. Abbreviations: BPL, p-propiolactone; l-CEA, 1-(2-carboxyethyl)adensine; 7-CEG, 7-(2carboxyethyl)guanine; PB, phosphate buffer 0.2 M, pH 7.2.
328 level BPL reacts with derivatives of guanine and in vitro with RNA [6,7], in vitro with DNA [8,9] and in vivo with DNA and RNA to form 7-CEG [3,10]. We have shown that BPL reacts in vitro with histones, non-histone chromosomal proteins and DNA in mouse skin chromatin [9], that the binding of BPL is selective to lysine rich histones Hl and Hl” [9] and that in vitro BPL acylates the c-amino group of L-lysine in calf thymus histones [ll]. BPL also appears capable of causing the formation of a complex between proteins and DNA [12]. The chemical nature of this complex is not known. BPL has been reported to cause in vitro intermolecular linking as well as fragmentation of DNA [13]. Recently Hennings et al. [14] studied the repair of BPL-induced damage to mouse skin cells in culture. Their results suggested that BPL reacted with the purines, adenine and guanine in nuclear DNA and although this is the first reported suggestion based on experimental evidence that BPL reacts with adenine in DNA no BPL-adenine derivative was isolated. We report in this paper that the in vitro reaction of BPL with either calf thymus DNA or mouse liver DNA results in reaction with adenine to form l-CEA (Fig. 1) as well as with guanine to form 7-CEG (Fig. 1).
EXPERIMENTAL
Materials Calf thymus DNA was obtained from the Sigma Chemical Co., St. Louis, MO. Mouse liver DNA was isolated from the livers of Q ICR/Ha mice obtained from ARS/Sprague-Dawley, Madison, Wise. by the method of Marmur [15]. The DNA was further deproteinized by treatment with pronase [ 161. The mouse liver DNA thus obtained had a Tm of 74°C in dilute saline citrate [9] with a hyperchromicity of 33%. BPL (99% purity) was purchased from BDH Chemicals Ltd., Poole, England. 2’-Deoxyadenosine-5’-monophosphoric acid, 2’-deoxycytidine-5’-monophosphoric acid, 2’-deoxyguanosine-5’-monophosphoric acid, thymidine-5’-monophosphoric acid, adenine, guanine, l-methyladenine and Sephadex G-10 were purchased from Sigma Chemical Co., St. Louis, MO. Whatman 3MM paper sheets from chromatography were obtained from Fisher Scientific Co., Springfield, N.J. Cellulose plates (20 X 20 cm., 0.10 mm thick) for analytical and preparative thin-layer chromatography was purchased from Brinkmann Instruments Inc., Westbury, N.Y. N,O-Bis(trimethylsilyl)trifluoroacetamine (“BSTFA”) and pyridine, silylation grade, were from Pierce Chemical Co., Rockford, Ill. The solvent system used for thin-layerchromatography on cellulose plates was isopropanol-H, Ocont. HCl, 6.5 : 1.84 : 1.66 (v/v/v) [17]. The solvent systems used for paper chromatography were: System A, ammonium acetate, 1 M, pH 7.5-100% ethanol, 3.5 : 6.5 (v/v); System B, isopropanol-conc. ammonia-HzO, 7 : 1 : 2; System C, isopropanol-HzO, 7.3 and System D, methanol--cone. HCl-H20, 7 : 2 : 1.
329 Instrumentation
UV spectra were recorded in 1 cm cells in HzO, in HCl at pH 1 and NaOH at pH 13 in a Beckman Model 25 spectrophotometer. Spectra recorded at pH 7.2 were taken in PB [6]. Electron impact (EI) and chemical ionization (CI) mass spectra were acquired using a DuPont 21-492 High Resolution Mass Spectrometer equipped with a dual EI/CI ionization source. The ionizing voltage was 70 eV. Samples were introduced as solids in glass capillary tubes via the direct probe. The source temperature was maintained at 260°C. The probe temperature was gradually increased until spectra were obtained and this temperature is noted for the mass spectra of compounds in the Results section. CI spectra were obtained using Matheson instrument grade (99.5%) isobutane maintained at -0.5 torr. Paper chromatography
Paper chromatograms were run on paper sheets, 17 X 55 cm for 18 h using a descending method. Compounds were identified on the paper chromatograms by observing the chromatograms under UV light at 254 nm. Compounds were isolated from the chromatograms by elution with water and then concentrated by lyophilization. Column
chromatography
Column chromatography was performed on a column of Sephadex G-10 (1.5 X 85 cm) and HZ0 or ammonium formate, 0.05 M, pH 6.8 were used as elm&s [18]. Flow rate was 45 ml/h and 2 ml fractions were collected in a Gilson automatic fraction collector (Gilson Medical Electronics, Middleton, Wise.). Fractions were monitored at 265 nm. Thin-layer
chromatography
This was performed on 20 X 20 cm cellulose plates with the solvent system indicated above. The solvent system was allowed to run to the top of the plate which was about 18.5 cm above the point at which compounds were spotted. METHODS
Reactions
of BPL with calf thymus DNA and mouse liver DNA
These reactions were according to the method of Segal et al. [9]. DNA was dissolved in PB (1 mg/ml) and 100 pl(ll5 mg, 1.6 X 10s3 M) of BPL/mg of DNA was added to the DNA solution at 0-5°C. The mixture was stirred at O--5% for 20 h. The reaction mixture was then dialyzed at 0-5”C, for 48 h against repeated changes of HzO. Purification
and isolation of BPL-reacted
This was performed [ 191. The BPL-reacted
DNA
according to the method of Lawley and Thatcher DNA was precipitated from the aqueous solution by
330 the addition of 2 volumes of 2-ethoxyethanol containing 0.1 volume of 2.5 M sodium acetate. The supernatant was decanted and the precipitated DNA was redissolved in PB and the procedure repeated twice more yielding a white fibrous BPL-reacted DNA preparation. Hydrolysis of l?PL-reacted DNA Method 1. To a solution of 1 mg of BPL-reacted DNA [or control (unreacted) DNA] in 1 ml Hz0 was added sufficient cont. HCl to make a solution of 0.1 N. The acid solution was heated to 100°C for 1 h [20], and then lyophilized to dryness. Method 2. The solution of DNA at 0.1 N HCl was heated at 37°C for 16 h [19]. The hydrolyzed DNA after all hydrolyses was lyophilized to dryness and dissolved in a small quantity of Hz0 prior to application to paper chromatograms. Preparation of compound 7-CEG (Fig. 1) ‘I-CEG was prepared from the reaction of BPL with dGuo-5’-P by the method of Colburn et al. [7]. The 7-CEG thus obtained was a white amorphous powder with UV spectra at pH 1, 7.2 and 13 identical to that reported by Colburn et al. [7]. Preparation of compound 1 -CEA dAdo-5’-P (100 mg, 2.88 X low4 M) was dissolved in 6.0 ml of Hz0 and the pH adjusted to 7.2. BPL (50 (~1,57 mg) was added at 30 min intervals, 6 times, the pH maintained between 7.0 and 7.4 by dropwise addition of NaOH. The temperature was maintained at 37°C. Total BPL added was 300 ~1 (345 mg, 4.8 X 10e3 M). After addition of the last sample of BPL the reaction mixture was kept at 37°C for an additional 1.5 h. The reaction mixture was then extracted with 3 X 1.5 ml portions of ethyl ether, made up to pH 1 with cont. HCl and heated to 100°C for 1 h. Portions of the reaction mixture not used for hydrolysis were stored at -20°C. The hydrolysate was applied to paper chromatograms and eluted with solvent system A. Synthetic l-CEA with an RF identical to l-CEA from DNA was identified and eluted from the paper chromatogram.
N
/H
0
Haglj$H 1 I-CEA Fig. 1. Structures ,m nlT-1
7-CEG of l-(2-carboxyethyl)adenine
(l-CEA)
and 7-(2-carboxyethyl)
guanine
331 Preparation of the trimethylsilyl derivatives of compound l-CEA for mass spectrometry Compound l-CEA (50 pg) (isolated from calf thymus or mouse liver DNA) was mixed with “BSTFA” (50 ~1) and pyridine for silylation (25 ~1) in a closed vial and heated at 55°C for 10 min. A 10 ~1 portion of the solution was injected into a glass capillary tube which was inserted into the mass spectrometer probe. RESULTS
The HCl hydrolysates (Method 1) of BPL-reacted calf thymus DNA and mouse liver DNA were each chromatographed on cellulose plates. The major bands were eluted with water and applied to paper chromatograms. When a band containing compound ‘I-CEG was applied to a paper chromatogram and eluded with solvent system A, a new UV absorbing compound was observed which was not present in paper chromatograms of control DNA hydrolysates. The new compound was designated compound l-CEA (Fig. 1) and gave single spots on paper chromatograms with solvent systems A, B, C and D. RF values obtained were, system A: l-CEA, 0.45; 7-CEG, 0.41; guanine, 0.57; adenine, 0.67; system B: l-CEA, 0.36; ‘I-CEG, 0.19; guanine, 0.23; adenine, 0.44; system C: l-CEA, 0.25; 7-CEG. 0.21; guanine, 0.42.; adenine, 0.65; system D: l-CEA, 0.42; 7-CEG, 6.27; adenine, 0.27. The UV spectra of compound l-CEA from calf thymus and mouse liver DNA were, PH 1, ha 260, hi, 232; PH 7-2 (PB), A,, 268, kin 240, PH 13, x max 271, h min 242 nm. EI and CI mass spectrometry was performed on l-CEA. The probe temperatures were varied between the minimum required to obtain a spectrum (250°C) and 400°C and for both EI and CI mass spectra the various peaks varied slightly with the probe temperatures. The EI mass spectrum of l-CEA gave peaks at m/e 189 (7), 135 (loo), 108 (31), 81 (13) and 72 (55) but the postulated molecular ion peak at m/e 207, the MW of l-CEA was not observed. At this stage of the experimentation we were uncertain as to whether the highest mass peak at m/e 189 in the EI mass spectra of l-CEA represented fragmentation of the molecular ion (m/e 207) or a pyrolytic decomposition to a neutral species of MW 189 prior to volatilization and ionization. The fact that adenine was detected in the capillary tube after EI mass spectrometry, by paper chromatography and UV spectra, strongly suggested that pyrolytic dissociation was playing an important role in the EI spectra of l-CEA that we were obtaining. The EI mass spectra have been obtained for ethyl and isopropyl derivatives of adenine by Lawley et al. [21]. They obtained molecular ion peaks for the ethyl and isopropyladenines and large peaks at m/e 135, 108 and 81. The authors postulated that the principal fragmentation process of the molecular ion in the higher adenines was a rearrangement involving a proton transfer from the alkyl moiety to adenine resulting in the net loss of alkyl-H (ethylene and isopropylene in the cases cited) and formation of adenine’ at m/e 135 and its
332 fragmentation products at 108 and 81 due to successive losses of two molecules of HCN. The EI spectra of adenine derivatives due to loss of HCN obtained by Lawley et al. [21] was essentially identical to relative peak intensity at m/e 135, 108 and 81 observed in our EI spectra of compound l-CEA and an authentic sample of adenine. Evidence for the pyrolytic decomposition of compound l-CEA rather than fragmentation of the molecular ion under the EI conditions used was obtained from CI mass spectrometry. Peaks were obtained at m/e 190 (189 + H’) (49), 136 (135 + H+) (84) and 73 (72 + H’) (100). In addition small addition ions were obtained at m/e 176 [135 + 41’ (C,H;)] , 178 [135 + 43’ (C,H,+)] , 230 (189 + 41’) and 232 (189 + 43’). The isobutane CI spectra of authentic adenine (MW 135) was obtained and exhibited a base peak at m/e 136 (135 + H’) and small addition ions at 176 (135 + 41+) and 178 (135 + 43+). We thus conclude that the peaks at 135 and 189 are neutral molecules formed by pyrolysis prior to ionization, as addition ions were observed which can only be formed by the addition of positively charged ions to neutral molecules [22]. The dissociation of 9-(2carboxyethyl)adenine to adenine and acrylic acid in the presence of water of hydration and at elevated temperatures reported by Lira and Huffman [23] was interpreted as a reverse Michael reaction as the dissociation did not occur in the absence of water of hydration. That heat induced dissociation of the carboxyethyl group (acrylic acid) from adenine in our experiments but not the ethyl and isopropyl groups from adenine in the experiments of Lawley et al. [21] may be due to the stability of the product, the a&unsaturated acrylic acid CH, =CH-COOH. The EI spectra of the trimethylsilyl derivative of l-CEA was run without heating the probe. Peaks were obtained at m/e 423 (1.8), 408 (2.3), 2’79 (62), 264 (100) and 73 (42). Tests for phosphorus [24] and deoxyribose [25] in l-CEA were negative. Reaction of BPL with dAdo-5’-P resulted in the isolation of a compound from paper chromatograms with solvent system A, which was designated synthetic l-CEA. Synthetic l-CEA had identical RF vdues on paper chromatograms with solvent systems A, B and C, identical I_JV spectra at pH 1, 7.2 and 13 and similar EI and CI spectra except for small variations in the relative intensities of the peaks obtained. NO compound identical to compound l-CEA was produced when BPL was reacted with either dCyd-5’-P, dGuo-5’-P or dThd-5’-P under identical conditions as the reaction between BPL and dAdo-5’-P. When BPL-reacted DNA was hydrolyzed at 0.1 N HCl for 1 h at 100°C (Method 1) the ratio of l-CEA to 7-CEGwas 3 : 1 assuming that the extinction coefficient of l-CEA was similar to l-methyl and other l-alkyladenines [26] (see Discussion section). When the BPL-reacted DNA was hydrolyzed at 0.1 N HCl for 16 h at 37°C (Method 2) the ratio of l-CEA to 7-CEG was 1 : 3, i.e., the ratio of the alkylated purines released was reversed. Compound l-CEA was eluted from the Sephadex G-10 column in fractions 45-60 and 7-CEG in fractions 68-77 using either water or ammonium formate (0.05 M, pH 6.8) as eluant.
333 DISCUSSION
A new alkylated base; designated compound l-CEA (Fig. 1) was isolated from the HCl hydrolysis products of BPL-reacted calf thymus or mouse liver DNA and assigned the structure 1-(2-carboxyethyl)adenine on the basis of UV, EI and CI mass spectra and chemical synthesis from BPL and dAdo-5’-P. Also isolated was the known 7-(2-carboxyethyl)guanine(7-CEG, Fig. 1) [6,7] which was identified by comparison to 7-CEG synthesized by us from BPL and dGuo-5’-P by the method of Colburn et al. [7]. The UV spectra of l-CEA was essentially identical to an authentic sample of 1-methyladenine and a listing of 1-alkyladenines at acid, alkaline and neutral pH values [27] and differed from the UV spectra of an authentic sample of 3-methyl-adenine and from UV spectra of monoalkylated C-2, N-3, N6, C-7, C-8 and N-9 substituted adenines [27,28]. A molecular ion was not observed in the CI and EI mass spectra due to thermal decomposition. Strong support for a molecular weight of 207 for compound l-CEA (C,H,N,02) was obtained from the EI spectra of the more volatile trimethylsilyl derivative of l-CEA. A large molecular ion peak was recorded at m/e 279 (l-CEA-H + trimethylsilyl), the trimethylsilyl ester of l-CEA. The large peaks formed at m/e 264 and 73 are characteristic of the EI mass spectra of trimethylsilyl derivatives which easily lose methyl [M-15+ m/e 2641 and a positively charged trimethylsilyl group [(CH,),Si+ m/e 731 [29,30] . Small peaks at m/e 423 and 408 (loss of methyl) suggest that the protons at N6 and N-9 of l-CEA have some lability and that a small amount of trimethylsilyl derivative can also be formed at positions N6 and N-9. The structure of the compound giving rise to a peak at m/e 189 in the El mass spectrum and 190 in the CI mass spectrum of l-CEA represents the loss of water from l-CEA prior to volatilization with the possible formation of a structure such as l-CEA Lactam (Fig. 2). Meot-Ner and Field [31] report that several amino acids form lactams, by losing water prior to ionization and protonated molecular ion formation (CI). When DNA was hydrolyzed at 100°C (Method 1) more l-CEA was isolated than 7-CEG while when DNA was hydrolyzed at 37°C (Method 2) the reverse was true. This is in agreement with the work of Frei and Lawley [18] who found that hydrolysis of alkylated DNA in HCl at lower temperatures released 3 and 7 methyl-adenines and guanines, while a temperature of 100°C was required to release 1-methyladenine, suggesting that the 1-alkyl derivatives of adenine are more stable to depurination than alkylation at the other common positions of adenine and guanine. Additional strong support for the structure of compound l-CEA comes from its synthesis from BPL and dAdo-5’-P. Synthetic l-CEA was found to be identical to l-CEA obtained from DNA in all parameters measured (see Results section). Compound l-CEA is a new compound and this is the first BPL-derivative of adenine isolated following in vitro or in vivo reaction of BPL with DNA. The results tend to confirm the conclusion of Hennings et al. [14] that BPL reacts with adenine and guanine in nuclear DNA in
334 0
I-CEA LACTAM Fig. 2. Structure of l-CEA Lactam, postulated as a possible structure of the dehydration product of l-CEA giving a peak at m/e 189 in the electron impact and 190 (189 + H+) in the chemical ionization mass spectra.
epidermal cells in culture. There is a high probability that bases alkylated in vivo by direct-acting carcinogens are similar to those alkylated in vitro [2]. One of the main reasons for acid hydrolysis of BPL-reacted DNA at 37°C (Method 2) was to try to determine if 6-(2carboxyethoxy)guanine was formed by reaction of BPL and DNA. It was reported that 6_methoxyguanine, a compound of importance in both mutagenesis and carcinogenesis [2], is released and is stable at pH 1 and 37°C for 16 h. This stability was confirmed by us using an authentic sample of 6-methoxyguanine. However we were not able to detect a BPL-6-alkoxy derivative of guanine. 6-Alkoxyderivatives of guanine are relatively easy to detect on paper chromatograms as they give an intense blue fluorescence when exposed to UV light at 254 nm. It should be noted that our methods of hydrolyses of BPL-reacted DNA at a highly acidic pH could easily dissociate or destroy more labile derivatives of BPL and bases or phosphate in DNA. A less destructive method of identifying labile products of BPL would employ the use of enzymatic hydrolyses [32,33] and it may be that more sites of BPL alkylation or acylation exist in DNA. BPL is a powerful carcinogen in several species of rodents, and cancers appear to result wherever BPL reacts directly with the rodent tissue. How alkylation of the l-position of adenine could contribute to this carcinogenicity is not known. It is certain that formation of l-(2-carboxyethyl)adenine in DNA would interrupt Watson-Crick base-pairing between adenine and thymine by blocking one of the positions involved in the hydrogenbonded association between the two bases. Whether the presence of l-CEA in DNA results in inactivation of DNA, produces mutations or is subject to the various forms of repair processes, error free and/or error prone (post replication repair) [ 2,34-361 is unknown at present. ACKNOWLEDGEMENTS
We wish to thank Professor P.D. Lawley of the Chester Beatty Research Institute, Buckinghamshire, England for generous gifts of 3-methyladenine and 6-methoxyguanine and Dr. Gisella Witz of our department for help in spectral determinations.
335 REFERENCES 1 2
8 9
10 11
12 13 14
15 16 17 18
19 20 21 22 23 24
F.S.C. Roe and O.M. Glendenning, The carcinogenicity of P-propiolactone for mouse skin, Br. J. Cancer, 10 (1956) 357. P.D. Lawley, Carcinogenesis by alkylating agents, in C.E. Searle (Ed.), Chemical Carcinogens, ACS Monograph 173, American Chemical Society, Washington, D.C., 1976, pp. 83-244. R.K. Boutwell and N.H. Colburn, The binding of P-propiolactone to mouse skin DNA in vivo; its correlations with tumor-initiating activity, Cancer Res., 26 (1966) 1701. H.H. Smith and A.M. Srb, Induction of mutations with P-propiolactone, Science, 114 (1951) 490. A.K. Powell, The effects of propiolactone on rat fibrocytes cultivated in vitro, Br. J. Cancer, 20 (1966) 402. J.J. Roberts and G.P. Warwick, The reaction of p-propiolactone with guanosine, deoxyguanylic acid and RNA, Biochem. Pharmacol., 12 (1963) 1441. N.H. Colburn, R.G. Richardson and R.K. Boutwell, Studies of the reaction of P-propiolactone with deoxyguanosine and related compounds, Biochem. Pharmacol., 14 (1965) 1113. J.J. Roberta and G.P. Warwick, The reaction of some compounds of biological interest with nucleic acids, Biochem. J., 87 (1963) 14. A. Segal, M. Schroeder, P. Barnett and B.L. Van Duuren, Studies on the effects in vitro of p-propiolactone and p-propiolact[14C]one on whole mouse skin chromatin, Biochem. Pharmacol., 23 (1974) 937. N.H. Colburn and R.K. Boutwell, The in vivo binding of p-propiolactone to mouse skin DNA, RNA and protein, Cancer Res., 28 (1968) 642. A. Segal and S.J. Garte, In vitro acylation of the e-amino group of L-lysine in calf thymus histones by the carcinogen, p-propiolactone, Chem.-Biol. Interact. 15 (1976) 319. W.C. Nietert, L.M. Kellicutt and H. Kubinski, DNA-protein complexes produced by a carcinogen, p-propiolactone, Cancer Res., 34 (1974) 859. K. Kubinski and E.H. Szybalski, Intermolecular linking and fragmentation of DNA by p-propiolactone, a monoalkylating carcinogen, Chem.-Biol. Interact. 10 (1975) 41. H. Hennings, D.M. Michael, S.D.A. Eaton and D.L. Morgan, Purinespecific repair of p-propiolactone-induced DNA damage in mouse skin cells, J. Invest. Dermatol., 62 (1974) 480. J. Marmur, A procedure for the isolation of deoxyribonucleic acid from microorganisms, J. Mol. Biol., 3 (1961) 208. K. Marashige and H. Ozaki, Properties of isolated chromatin from sea urchin embryo, Dev. Biol., 16 (1967) 474. K. Fink and W. Adams, Paper chromatographic data for purines, pyrimidines and derivatives in a variety of solvents, J. Chromatogr., 22 (1966) 118. J.V. Frei and P.D. Lawley, Methylation of DNA in various organs of C57 Bl mice by a carcinogenic dose of N-methyl-N-nitrosourea and stability of some methylation products up to 18 h, Chem.-Biol. Interact., 10 (1975) 413. P.D. Lawley and C.J. Thatcher, Methylation of deoxyribonucleic acid in cultured mammalian cells by N-methyl-N’nitro-N-nitrosoguanidine, Biochem. J., 116 (1970) 693. S.C. Maitre and J.V. Frei, Organ specific effects of DNA methylation by alkylating agents in the imbred Swiss mouse, Chem.-Biol. Interact., 10 (1975) 285. P.D. Lawley, D.J. Orr and M. Jarman, Isolation and identification of products from alkylation of nucleic acids: ethyl and isopropyl purines, Biochem. J., 145 (1975) 73. B. Munson, Chemical ionization mass spectrometry, Anal. Chem., 43 (1971) 28A. E.P. Lira and C.W. Huffman, Some Michael-type reactions with adenine, J. Org. Chem., 31 (1966) 2188. P.S. Chen, Jr., T.Y. Toribara and H. Warner, Microdetermination of small phosphorus,
336 Anal. Chem., 28 (1956) 1756. J.G. Buchanin, Detection of deoxyribonucleosides on paper chromatogram, Nature, 168 (1951) 1091. 26 D.D. Dunn and R.H. Hall, Purines, pyrimidines, nucleosides, and nucleotides: Physical constants and spectral properties, in G.D. Fasman (Ed.), Handbook of Biochemistry and Molecular Biology, 3rd ed., Nucleic Acids, Vol. 1, CRC Press, Cleveland, Ohio, 1975, pp. 65-215. 27 B. Singer, UV Spectral Characteristics and Acidic Dissociation Constants of 280 Alkyl Bases, Nucleosides and Nucleotides, in G.D. Fasman (Ed.), Handbook of Biochemistry and Molecular Biology, 3rd ed.; Nucleic Acids, Vol. 1, CRC Press, Cleveland, Ohio, 1975, pp. 409-449. 28 C.V.Z. Smith, R.K. Robins and R.L. Tolman, Clarification of the ring closure of certain 4,5-diaminopyridines to 6-substituted 8-alkylpurines, J. Chem. Sot., Perkin Trans., 1 (1973) 1855. 29 H. Budzikiewicz, C. Djerassi, D.H. Williams, Mass Spectrometry of Organic Compounds, Holden-Day, San Francisco, Calif., 1967, p. 473. 30 F.W. McLafferty, Interpretation of Mass Spectra, 2nd ed., W.A. Benjamin, Reading, Mass., 1973, p. 196. 31 M. Meot-Ner and F.H. Field, Chemical ionization of mass spectrometry. XX. Energy effects and virtual ion temperature in the decomposition kinetics of amino acid derivatives. J. Am. Chem. Sot., 95 (1973) 7207. 32 D.H. Swensen, P.B. Farmer and P.D. Lawley, Identification of the methyl phosphotriester of thymidylyl(d-5”)tNymidine as a product from reaction of DNA with the carcinogen N-methyl-N-nitrosourea, Chem.-Biol. Interact., 15 (1976) 91. 33 B. Singer, All oxygens in nucleic acids react with carcinogenic ethylating agents, Nature, 264 (1976) 333. 34 J.E. Cleaver, Excision repair: our knowledge based on human (Xeroderma Pigmentosum) and cattle cells, in R.F. Beers Jr., R.M. Herriott and R.C. Tilghman (Eds.), Molecular and Cellular Repair Processes, 5th International Symposium on Molecular Biology, The Johns Hopkins University Press, Baltimore, Md., 1972, pp. 195-211. 35 A. Kornberg, DNA Synthesis, W.H. Freeman, San Francisco, Calif., 1974,pp. 297-327. 36 P. Kleihues and H.K. Cooper, Repair excision of alkylated bases from DNA in vivo, Oncology, 33 (1976) 86.
25