Chem. Res. Toxicol. 1993,6, 261-268
26 1
Characterization of 2’-Deoxycytidine and 2’-Deoxyuridine Adducts Formed in Reactions with Acrolein and 2-Bromoacrolein Ahmed Chenna and Charles R. Iden” Department of Pharmacological Sciences, School of Medicine] Health Sciences Center, State University of New York a t Stony Brook, Stony Brook, New York 11794-8651 Received December 18,1992
The products of the reaction of the mutagenic aldehydes, acrolein and 2-bromoacrolein, with 2’-deoxycytidine and 2‘-deoxyuridine have been determined. These products, formed a t physiological conditions, were isolated by reverse-phase HPLC and characterized by UV, 1H NMR, fast atom bombardment MS, electrospray MS, and chemical transformation. The reaction of 2/-deoxycytidine with acrolein and 2-bromoacrolein produced the exocyclic compounds 3 42‘deoxyribosy1)-7,8,9-trihydro-7-hydroxypyimido[3,4-c] pyrimidin-2-one and 3 42’-deoxyribosy1)7,8,9-trihydro-7-hydroxy-8-bromopyrimido[3,4-c] pyrimidin-2-one, respectively. In addition to the chiral centers of deoxyribose, one new chiral center was formed from C-1 of acrolein and two new chiral centers were formed from C-1 and C-2 of 2-bromoacrolein, creating a mixture of diastereomers for each product. These compounds are not stable in basic solution and undergo ring opening and hydrolytic deamination, resulting in 2’-deoxyuridineadducts. The N3-alkylated 2’-deoxyuridines were also synthesized by permitting 2’-deoxyuridine to react with 2-bromoacrolein and acrolein. An unstable intermediate, N3-(2”-bromo-3”-oxopropyl)-2’-deoxyuridine, was also isolated and characterized from the reaction with 2-bromoacrolein. The reaction of 2’-deoxyuridine with acrolein gave N3-(3”-0xopropyl)-2’-deoxyuridine as the major product, which was reduced to its corresponding alcohol with NaBH4. Reactions of 2’-deoxycytidine with 2-bromoacrolein and acrolein proceed most rapidly a t acidic or neutral pH; however, 2’deoxyuridine reacts most rapidly a t neutral or basic pH.
Introduction Acrolein is ubiquitous in the human environment, and a major source is the result of incomplete combustion of organic matter (1). Studies have shown that acrolein is mutagenic toward Salmonella typhimurium (2,3) and can interact directly with DNA (4-8).Severalinvestigators have characterized adducts of acrolein with deoxyguanosine (4, 9), adenine and cytosine derivatives (10, Il), deoxycytidine 5‘-monophosphate (12),and thymidine (13). Marinelli et al. (14)incorporated the model acroleindeoxyguanosine adduct 1,W-(1,3-propano)-2‘-deoxyguanosine into oligodeoxynucleotides, and the mutagenic effects on Escherichia coli have been examined (15). Halogen substitution at C-2 of acrolein should increase the rate of reaction with nucleobases in DNA; this was confirmed by Rosen et al. (16), who found that 2-bromoacrolein and several chloroacroleins have a higher mutagenic potency than acrolein. 2-Bromoacrolein is a genotoxicmetabolite of tris(2,3-dibromopropyl)phosphate (Tris-BP)’ (17)and the nematocide 1,2-dibromo-3-chloropropane (18). Gordon et al. (19)have claimed that TrisBP is activated by conversion to reactive electrophiles, such as 2-bromoacrolein, by cytochrome P450.In recent articles,several groups investigated the chemical structure of the products from the reaction of 2-bromoacroleinwith 2’-deoxyguanosine (20,21) and thymidine (13)and also with DNA to produce an unstable thymidine adduct (22). The goal of the present study was to determine the chemical structures of all major adducts formed in the
* Author to whom correspondence should be addressed. I Abbreviations: Tris-BP, tris(2,3-dibromopropyl) phosphate; FAB, fast atom bombardment.
reaction of both 2‘-deoxycytidineand 2‘-deoxyuridine with 2-bromoacrolein and acrolein under physiological conditions, and to investigate the hydrolytic deamination of 2’-deoxycytidine adducts in neutral and basic conditions.
Experimental Procedures Chemicals. Caution: Acrolein and 2-bromoacrolein are mutagenic and should be handled carefully. 2’-Deoxycytidine free base was purchased from Sigma Chemical Co. (St. Louis, MO). 2’-Deoxyuridine,acrolein (97%,inhibitedwith -3% water and 200 ppm hydroquinone), and sodium borohydride were purchased from Aldrich Chemical Co. (Milwaukee,WI). Sodium bicarbonate, potassium phosphate dibasic, potassium phosphate monobasic, sodium hydroxide, sodium acetate, and acetonitrile were from Fisher Scientific Co. (Fair Lawn, NJ) and were used as received unless otherwise stated. Silica gel 60 (Merck) for flash chromatography was purchased from Krackler Scientific Co. (Albany, NY). 2-Bromoacrolein was prepared as described previously (13). Instrumentation. Ultraviolet spectra were recorded on a Hewlett-Packard 8452A diode array spectrophotometer using 1-cm cuvettes. The spectra were recorded after applying a background correction function obtained with a cuvette filled only with solvent. lH NMR spectra were obtained on a General Electric QE-300 spectrometer using either tetramethylsilane or the residual absorption of MezSO-dc (2.49 ppm) as the internal reference. Fast atom bombardment (FAB) mass spectra were obtained on a Kratos MS890/DS90 instrument. Glycerine or thioglycerine was used as a matrix. Electrospray mass spectra were obtained on a VG Instruments Trio-2000 mass spectrometer. Loop injections of a freshly collected HPLC fraction were made into solvent (49.5:49.5:1 acetonitrile/water/acetic acid) flowing into the electrospray source (5 pL/min). Analytical and preparative HPLC were conducted using a Waters HPLC system
0893-228x/93/2706-0261$04.0010 0 1993 American Chemical Society
Chenna and Iden
262 Chem. Res. Toxicol., Vol. 6, No. 3, 1993 including a Waters U6K injector, a 990 photodiode array detector, a 600E solvent delivery system, and pBondapak Cia (3.9- X 300mm column, 10-pm)or CiaResolve Radial Pak cartridge column for analytical analysis, and pBondapak CISPrepPak cartridge (25- X 100-mm)column for preparative analysis (HPLC columns were obtained from Millipore Corp., Milford, MA). Solvent systems were programmed as follows. System 1: For the pBondapak CIS(3.9- X 300-mm, 10-pm)column, alinear gradient of acetonitrile/water was programmed from 0 % to 12% organic over 25 min and then to 50% over the next 15 min at a flow rate of 1 mL/min. System 2: For the C18ResolveRadial Pak cartridge column, system 1 was used at a flow rate of 2 mL/min. System 3: For the Waters pBondapak C18PrepPak cartridge (25- X 100mm) column, a linear gradient of acetonitrileiwater was programmed from 0% to 6 % organic over 30 min and held at 6 % isocratic for next 20 min. The flow rate was 9 mL/min. Chemical Syntheses. (A) Formation of 3-(2‘-Deoxyribosyl)-7,8,9-trihydro-7-hydroxy-8-bromopyrimido[ 3,4-c]pyrimidin-2-one ( 2 ) . 2’-Deoxycytidine (400 mg, 1.76 mmol) and 2-bromoacrolein (288 mg, 2.13 mmol) were allowed to react in 24 mL of 0.1 M sodium acetate buffer (pH 4.4) at 37 “C. After 38 h the solvent was evaporated under reduced pressure to leave a yellow viscous oil (0.851g). Silicagel thin-layer chromatography (20230 methanolimethylene chloride) indicated that this oil contained two components, the startingmaterialand an unknown substance. Flash chromatography on silica gel (15:85methanol/ methylene chloride) permitted their separation, the unknown being obtained as a yellow gum (441 mg, 70%) which failed to crystallize from methanol. TLC of this material on silica gel showed a single spot; however, TLC on a C18reverse-phase plate (10:90 acetonitrileiwater) separated the product into two distinct substances with very similar Ri values. HPLC analysis of the reaction mixture (system 2) gave retention times of 10.8 min for 2’-deoxycytidine and 15.4 min for the mixture of products. Individual stereoisomers could not be separated by HPLC. The 1H NMR of the product was difficult to interpret, but demonstrated the presence of more than one stereoisomer. This reaction was repeated at pH 7.4 adding 10 equiv of 2-bromoacrolein. After 24 h, TLC and HPLC showed only traces of starting material. Spectroscopic data for the product were identical to those of the product of the reaction at pH 4.4. UV, A,, 288 nm (pH 2.6 and 4.2), 228 and 274 nm (pH 10.8); FAB/MS (positive ion) m/z 362 (15.0) MH+,364 (14.4) MH+81Br isotope, 246 (100.0) BHz+, 248 (98.9) BHz+ 81Brisotope. (B) Hydrolysis of 2 with Sodium Bicarbonate. 2’Deoxycytidine (200 mg, 0.88 mmol) and 2-bromoacrolein (1.12 g, 8.4 mmol) were incubated in 30 mL of 0.05 M potassium phosphate buffer (pH 7.4) at 37 “C to form 2. After 24 h, the pH of the reaction solution was acidic (pH = 3). The solution was divided equally into two aliquots, A and B. The pH of aliquot A was adjusted to 7.0 and that of aliquot B to 8.5 by adding saturated aqueous sodium bicarbonate. Both flasks were incubated at 37 “C for 24 h, and the progress of the reactions was monitored by HPLC (system 2). SolutionA showed the presence of two new products, 4 and 5, in the first 3 h (retention times 20.0 and 13.5 min, respectively). After 5 h, 5 became dominant and 4 disappeared, while a third product, 6, appeared at 15.5 min. Each of these three materials was detected in reaction B after 1h. After 5 h, the product at 15.5min was dominant. Purification of small quantities of 5 and 6 was accomplished by preparative HPLC (system 3),but product 4 was unstable and was not isolated. Product 4: UV, A,, 262 nm (pH 7.0). Product 5: UV, A, 262 nm (pH 7.0); FABiMS (positive ion) m/z 623 (22.6) (2M + Na)+, 323 (14.9) (M + Na)+, 301 (27.2) MH+, 207 (3.7) (BH + Na)+; ‘H NMR (MezSO-do, 300 MHz) 6 9.504 (s, 35% H, H-3’7, 7.908 (m, 1 H, H-6), 6.800 (d, 1 H, J = 6.6 Hz, exch, HO-2”), 6.169 (m, 1 H, H-l’), 5.775 (m, 1 H, H-5), 5.257 ( s , 1 H, exch, HO-3’), 5.026 (9, 1 H, exch, H O W , 4.228 (s, 1H, H-4’), 4.009 (m, 2 H, H-1”), 3.788 (s, 1 H, H-39, 3.561 (s, 2 H, H-5’), 3.316 (m, 1 H, H-Y’), 2.111 (m, 2 H, H-2’). Product 6: UV, Amax 262 nm (pH 7.0); FABiMS (positive ion) m/z 623 (7.4) (2M + Na)+,323 (75.3) (M + Na)+,301 (22.0) MH+,
207 (81.3) (BH + Na)+; ‘H NMR (MezSO-dG,300 MHz) 6 7.954 ( d , l H , J=8.1Hz,H-6),6.143(t,lH, J = 6 . 6 H ~ , H - l ’ ) , 5 . 7 9 8 (d, 1 H, J = 8.1 Hz, H-5), 5.473 (s, 1 H, exch, HO-3”), 5.271 (s, 1H, exch, HO-3’), 5.050 (s, 1H, exch, HO-5’),4.744 (s, 2 H, H-l”), 4.228 (s,1H, H-4’), 4.178 ( ~ , H, 2 H-3”),3.800 ( ~ ,H, 1 H-3’),3.564 (s, 2 H, H-59, 2.109 (m, 2 H, H-2’). (C) Reaction of %’-Deoxyuridinewith 2-Bromoacrolein. 2’-Deoxyuridine (150 mg, 0.672 mmol) and 2-bromoacrolein (900 mg, 6.7 mmol) were incubated in 150 mL of 0.05 M potassium phosphate buffer (pH 7.4) at 37 “C for 5 days. HPLC (system 1)showed the formation of three products, which proved to be identical to 4-6 isolated in the previous experiment. In this case, however, 4 was present for 5 h, while 5 was detected after 1hand remained for 5 days (62%). The third compound, 6, appeared after 24 h and remained a minor product for the duration of the experiment. Product 4 is unstable and was analyzed by electrospray mass spectrometry immediately after HPLC elution. Small amounts of the other products were isolated by preparative HPLC and analyzed. The reaction was repeated at pH 9, and the same products were detected. At pH 9 the rate of the reaction is faster than at pH 7.4. However, at pH 4.4 no products were obtained, even after 4 days at 37 “C. Product 4 from t h i s reaction showed t h e following 262 nm (pH 7.0); electrospray MS characteristics: UV, A, (positive ion) m/z 405 (81.8) (M + HzO + Na)+81Brisotope; 403 (94.4) (M + HzO+ Na)+;383 (61.9) (M + H 2 0+ H)+81Brisotope; 381 (57.0) (M + HzO + H)+; 365 (100) MH+ 81Br isotope; 363 (94.7) MH+; 249 (29.8) BHz+81Brisotope; 247 (28.3) BHz+. Product 5: UV, A, 262 nm (pH 7.0); FABiMS (positive ion) m/z 333 (13.3) (M MeOH + H)+,301 (31.3) MH+,207 (6.3) (BH + Na)+;‘H NMR (MezSO-d6,300 MHz) 6 9.496 (s,30% H,H-3”), 7.901 (m, 1H, H-6),6.800 (d, 1H, J = 6.6 Hz, exch, HO-2”),6.167 (m, 1 H, H-1’), 5.772 (m, 1 H, H-5), 5.258 (s, 1 H, exch, HO-3’), 5.026 (s, 1 H, exch, HO-5’), 4.219 (s, 1 H, H-47, 4.002 (m, 2 H, H-l”), 3.784 (s, 1 H, H-37, 3.559 (s, 2 H, H-5’1, 3.316 (m, 1 H, H-2’9, 2.099 (m, 2 H, H-2’). Product 6: UV, A, 262 nm (pH 7.0); FABiMS (positive ion) m/z 301 (70.0) MH+,207 (12.2) (BH + Na)+; ‘H NMR (MezSOde, 300 MHz) 6 7.957 (d, 1 H, J = 8.1 Hz, H-6), 6.143 (t, 1 H, J = 6.6 Hz, H-l’), 5.800 (d, 1 H, J = 8.1 Hz, H-5), 5.600-5.000 (br, 3 H, exch, HO-3”, HO-3’, HO-5’1, 4.744 (s, 2 H, H-l”), 4.227 (9, 1 H, H-4’), 4.180 (s, 2 H, H-3”), 3.788 (s, 1 H, H-3’), 3.563 (s, 2 H, H-59, 2.101 (m, 2 H, H-2’). (D) Formation of 3-(2’-Deoxyribosyl)-7,8,9-trihydro-7hydroxypyrimido[3,4-c]pyrimidin-2-one(8). 2’-Deoxycytidine (200 mg, 0.881 mmol) and acrolein (49.35 mg, 0.881 mmol) were incubated in 24 mL of 0.1 M sodium acetate buffer (pH 4.4) at 37 “C for 44 h. Volatile components were removed by evaporation under vacuum, yielding an off-white viscous oil. TLC on silica gel (30:70 methanolimethylene chloride) showed the formation of a new product which was less polar than the starting material. Flash chromatography on silica gel (15:85 methanol/ methylene chloride) permitted its separation as a colorless gum (158 mg, 63%) which was crystallized from methanol at -22°C as white crystals. When the reaction was conducted at pH 7.4, the same product was obtained in lower yield. mp 151-152 “C [lit. mp 152-153 “C ( l o ) ]UV, ; A,, 284 nm (pH 4.4 and 7.3),230 and 272 nm (pH 10.3); FAB/MS (positive ion) mlz 284 (13.5) MH+, 163 (100) (BHz)+;lH NMR (MezSO-&, 300 MHz) 6 7.364 (d, 1 H, J = 8.1 Hz, H-4), 6.227 (t, 1H, J = 6.6 Hz, H-l’), 5.582 (d, 1 H, J = 8.1 Hz, H-5), 5.542 (s, 1 H, exch, HO-7), 5.292 (9, 1 H, exch, HO-3’), 5.022 (s, 1 H, exch, HO-5’), 4.883 (m, 1 H, NCHOH),4,271(s,l H, H-4‘), 3.799 (m, 2 H, H-37, 3.589 (m, 3 H, H-5’, H-5’9, 2.072 (m, 2 H, H-8), 1.958 (m, 1 H, H-2’1, 1.575 (m, 1 H, H-2”). (E) Preparation of Nj-(3”-Hydroxypropyl)-2‘-deoxyuridine (10) from the Hydrolysis of 8. Compound 8 (156 mg, 0.549 mmol) was dissolved in 2 mL of distilled water; 200 p L of 2 N sodium hydroxide was added, and the mixture was stirred for 5 min at room temperature. Sodium borohydride (20 mg, 0.62 mmol) was then added, and the temperature was raised to 80 “C. After 8 h TLC and HPLC (system 1) indicated the
+
Chem. Res. Toxicol., Vol. 6, No. 3, 1993 263
dC and dU Adducts of Acrolein and 2-Bromoacrolein
Scheme I. Products of the Reaction of %-Bromoacroleinwith %‘-Deoxycytidine and %’-Deoxyuridine OH
81
4fH 0
and 01
pH7.4
OH
OH OH
1
3
2 NaHC03 pn7and8.5
0
txo
0 OH
7
0
OH
6 formation of a single product with a retention time of 18.14 min on HPLC. The reaction was terminated at this point by adding a few drops of 0.1 N hydrochloric acid, and the mixture was taken to dryness under vacuum to give a yellow solid containing sodium chloride. Flash chromatography (10:90 methanol/chloroform) on silica gel yielded a pure product as a yellow oil (88 mg, 56%) which failed to crystalize from methanol. UV, A, 262 nm (pH 7.0); FAB/MS (positive ion) m/z 595 (14.1) (2M + Na)+,309 (85.5) (M + Na)+, 287 (85.0) MH+, 171 (100.0) BH2+; FAB/MS (negative ion) m/z 321 (30.0) (M + C1)-, 285 (9.8) (M - H)-, 168 (100.0) (B - H)-. 1H NMR (Me2SO-ds, 300 MHz) 6 7.828 (d, 1 H, J = 8.1 Hz, H-6), 6.076 (t, 1 H, J = 6.6 Hz, H-l’), 5.666 (d, 1 H, J = 8.1 Hz, H-5), 5.40-5.00 (br, 2 H, exch, HO-3’, HO-5’), 4.396 (s, 1 H, exch, HO-3’9, 4.150 (d, 1 H, J = 2.7 Hz, H-4’),3.700 (m, 3 H, H-3’, H-3”),3.464 (t,2 H, J = 3.9 Hz, H-59, 3.275 (t, 2 H, J = 6.3 Hz, H-1”), 2.009 (m, 2 H, H-2’), 1.552 (m, 2 H, H-2”). (F) Preparation of Nj-(3~~-Oxopropyl)-2‘-deoxyuridine (11). 2’-Deoxyuridine (100 mg, 0.44 mmol) and acrolein (230 mg, 4.4 mmol) were incubated in 20 mL of 0.05 M potassium phosphate buffer (pH 7.4) at 37O C for 18 h. A control reaction was also conducted without 2’-deoxyuridine. HPLC analysis (system 1)revealed the formation of two products, one of which appeared in the control reaction, together with some minor products and traces of startingmaterial. The substance common to both reactions (retention time of 13.8 min) is believed to be the acrolein dimer (23,24),while the product eluting at 15.6 min is the desired 2’-deoxyuridine adduct (11) (57%). A small quantity was purified by preparative HPLC (system 3). This reaction was repeated at pH 9.2 and gave the same products although at a faster rate. At pH 4.4 no products were found. UV, A,, 262 nm (pH 7.0); FAB/MS (positive ion) m/z 285 (10.1) MH+, 169 (11.3) BH2+;lH NMR (MenSO-ds,300 MHz) 6 9.408 (9, 1 H, H-3”), 7.807 (d, 1 H, J = 8.1 Hz, H-6), 6.061 (t, 1 H, J = 6.6 Hz, H-1’), 5.659 (d, 1 H, J = 8.1 Hz, H-5), 5.182 (9, 1H, exch, HO-3’), 4.955 ( 8 , 1H, exch, HO-5’), 4.124 (s, 1H, H-47,
3.960 (t, 1 H, J = 6.9 Hz, H-39, 3.697 (m, 2 H, H-l”), 3.464 (s, 2 H, H-59, 2.556 (t, 2 H, J = 6.9 Hz, H-2”), 2.022 (m, 2 H, H-2’). (G) Preparation of Nj-(3~~-Hydroxypropyl)-2‘-deoxyuridine (10) by the Reduction of 11. The above reaction was repeated, but at the end of 18 h sodium borohydride (38 mg, 1 mmol) was added. The mixture was stirred a t room temperature for 2 h and then neutralized by adding 0.1 N hydrochloric acid. HPLC (system 1)showed the formation of one product (81%) having a retention time of 18.5 min. This was purified by preparative HPLC (system 3) to give a yellow oil which failed to crystallize from methanol. Coinjection of this material (HPLC system 1) with product from the reaction of 8 with sodium hydroxide and sodium borohydride showed them to be identical. UV, A, 262 nm (pH 7.0); FAB/MS (positive ion) m/z 573 (8.1)(2M + H)+,309 (3.1) (M + Na)+,287 (87.6)MH+,171 (100.0) BH2+;1H NMR (MezSO-d6,300 MHz) 6 7.780 (d, 1 H, J = 8.1 Hz, H-6), 6.078 (t, 1 H, J = 6.6 Hz, H-1’), 5.638 (d, 1 H, J = 8.1 Hz, H-5), 5.167 (s, 1H, exch, HO-3’), 4.942 ( 8 , 1H, exch, HO-5’), 4.365 (s, 1H, exch, HO-3”),4.119 (d, 1H , J = 2.7 Hz, H-4’),3.696 (m, 3 H, H-3’, H-3”), 3.459 (t, 2 H, J = 3.9 Hz, H-5’), 3.256 (t, 2 H, J = 6.3 Hz, H-l”), 2.017 (m, 2 H, H-2’), 1.546 (m, 2 H, H-2”).
Results Reaction of %’-Deoxycytidineand %‘-Deoxyuridine with 2-Bromoacrolein. The reaction of 2’-deoxycytidine (1) with 2-bromoacrolein (Scheme I) in potassium phosphate buffer (pH 4.4) gave a single major product in 70% yield which was purified by flash chromatography. The UV spectrum of this substance recorded at pH 2.3 and 4.2 shows A,, at 288 nm, whereas at pH 10.8, Amax of 228 and 274 nm are present (Figure 1). These changes in UV spectral character as a function of pH are characteristic of 3,N4-substitutionon a cytidine residue (25)and support structures 2 or 3 for this compound.
264 Chem. Res. Toxicol., Vol. 6, No. 3, 1993
Chenna and Iden
,216
,054
0
200
250
300
350
400
Wavelength
Figure 1. UV spectrum of the exocyclic 2’-deoxycytidine adduct 2 at (A) pH 2.3 or 4.2 and (B) pH 10.8. leel
ml
Figure 2. FAB mass spectrum of 3-(2’-deoxyribosyl)-7,8,9trihydro-7-hydroxy-8-bromopyrimido[3,4-c]pyrimidin-2-one (2).
The positive ion FAB mass spectrum of this compound (Figure 2) has an MH+ at m/z 362 (15.0) and a corresponding peak at mlz 364 (14.4)containing the 81Brisotope. BH2+ peaks at mlz 246 (100.0) and 248 (98.9) are formed by cleavage of the glycosidic bond. This important fragmentation shows that Br isotopes are present in the modified base moiety. In this reaction two new chiral centers are formed at C-7 and C-8 derived from C-2 and C-1 of 2-broqbacrolein. Because of these chiral centers and those of the sugar ring, the product exists as a mixture of diastereomers which were not separable by HPLC. Consequently, the lH NMR is complex and difficult to interpret. However, the spectrum of this mixture does not show any exchangeable protons in the region between 6.2 and 7.5 ppm, indicating there is no NH2 group in the compound. Taken together, the spectral data indicate that the structure of the compound is either 2 or 3; however, the information did not allow the position of the hydroxy group to be uniquely determined. This product is not stable under neutral or basic conditions. When the mixture was incubated with dilute sodium bicarbonate at pH 7.0 at 37 “C for 24 h, ita pH became acidic, suggesting the release of HBr. The reaction was monitored by HPLC, and after 1h, two new products, 4 and 5, were detected having retention times of 20.0 and 13.5 min, respectively. Compound 5 became dominant (85%) after 5 h. A third product (6) with a retention time of 15.5 min was detected after 24 h. When the reaction was conducted at pH 8.5 with an excess of a saturated solution of sodium bicarbonate, the same three products were observed. By measuring the relative concentrations of the products as a function of time by HPLC, we conclude
that the initial product is 4 which is subsequently converted to 5 and then to 6. Product 4 was not isolated from this reaction due to the chemical instability and relatively low concentration. However, the UV spectrum showed a Xmax at 262 nm, distinctly different from that of 2’-deoxycytidine. Products 5 and 6 were purified by preparative HPLC and analyzed by UV, FAB/MS, and ‘H NMR. The UV spectrum of 5 was recorded at pH 7.0 and showed Amax at 262 nm, a value similar to that observed for 3-substitution on uridine (25). FAB mass spectrometry identified the protonated molecular ion MH+ at mlz 301 (27.2), an (M + Na)+ peak at m/z 323 (14.9), a (BH + Na)+ peak at mlz 207 (3.7), and a natriated dimer at mlz 623 (22.6). The ‘HNMR of this product when compared to the ‘H NMR of 2‘-deoxyuridine shows three new proton resonances. A singlet appears at 9.504 ppm which represents about 35 % of an aldehydic proton. The aldehydic proton is not represented completely due to partial formation of a hydrate from the water in MezSO-d6. A multiplet appears at 3.316 ppm which is assigned to the H-2” proton, and an exchangeable doublet at 6.800 ppm corresponds to the 2”-hydroxy. In addition, the spectrum does not show any proton shifts for an NH2 at N4 of 2’-deoxycytidine. The spectral data for this compound are consistent with the proposed structure, W-(2”-hydroxy-3”-oxopropyl)-2’deoxyuridine ( 5 ) . The UV spectrum of 6 has a A,, of 262 nm at pH 7.0, characteristic of 3-substitution on a uridine residue (25). The FAB mass spectrum is similar to that for 5, having peaks at mlz 623 (7.4) (2M + Na)+, mlz 323 (75.3) (M + Na)+, mlz 301 (22.0) MH+, and mlz 207 (81.3) (BH + Na)+. The ‘H NMR analysis was performed in Me2SOdg, and through a deuterium exchange experiment, it was deduced, by comparison with the lH NMR spectrum of 2’-deoxyuridine, that three new protons are present. A peak at 5.473 ppm shows one exchangeable proton, and it is assigned to the 3”-hydroxy group. Singlets at 4.744 and 4.178 ppm eachrepresent two protons, H-1”and H-3”, respectively. There is no evidence of any absorption for aldehydic or aminic protons. The structure assigned is W -(3”-liydroxy-2”-oxopropyl)-2’-deoxyuridine(6). In order to confirm the structures of 5 and 6, these compounds were synthesized by permitting 2‘-deoxyuridine to react with 2-bromoacrolein in a 0.05 M potassium phosphate buffer (pH 7.4). This reaction was monitored by HPLC as a function of time, and from the HPLC data, it appears that the formation of 5 is preceded by the formation of an unstable intermediate 4 (Scheme I). Although this intermediate was the first product formed in any appreciable amount, it is completely transformed to 5 after 5 h. Figure 3 shows the relative concentrations of 4-6 and 2’-deoxyuridine(7) as a function of time. These data also show that the reaction with 2’-deoxyuridine ceased after 2 h. Product 6 could only be detected after 5 h and reached only 13% after 5 days. The sum of the products (5 + 6) stayed roughly constant from 5 h to 5 days, indicating that 6 is derived from 5. Product formation in this reaction was also monitored as a function of pH. By increasing the pH of the system to 9.2, the formation of 6 increased to 50% within a period of 6 days. However, at pH 4.4, essentially no products were observed. The unstable intermediate (4) was isolated by HPLC and analyzed immediately by electrospray MS. The spectrum (Figure 4) shows a protonated molecular ion at
dC a n d dU Adducts of Acrolein a n d 2-Bromoacrolein
cs
N
v
60
Q)
u C
0
?
401
$ 1
/ I
/ 1
,
0.1
1
io
I 100
Time ( h )
Figure 3. Time course of the reaction of 2-bromoacrolein with 2’-deoxyuridine: (A)2’-deoxyuridine (7); (0) N3-(2”-bromo-3”oxopropyl)-2’-deoxyuridine(4); ( 0)W-(2”-hydr0xy-3”-oxopr0pyl)-2’-deoxyuridine (5); ( +) W-(2”-oxo-3”-hydroxypropyl)-2’deoxyuridine (6).
I I .3
it
3ep.m
mlz
Figure 4. Electrospray mass spectrum of N3-(2”-bromo-3”oxopropyl)-2’-deoxyuridine(4), isolated from the reaction of 2-bromoacrolein and 2’-deoxyuridine.
mlz 363 and also shows formation of the hydrate of the aldehyde in the acetonitrilelwaterlacetic acid solvent [mlz 381 (MH + H,O)+; mlz 403 (M Na + HzO)+]. Bromine isotope peaks are clearly evident. When acetonitrile was replaced by methanol, formation of the hemiacetal was observed. This intermediate was assigned the structure (2”-bromo-3”-oxopropyl)-2’-deoxyuridine.This compound has an HPLC retention time and UV spectrum identical to those of the first product formed in the hydrolysis of 2. Indeed, when 4, isolated from this reaction, was coinjected with the hydrolysis products of 2, this substance coeluted with the initial hydrolysis product. Products 5 and 6 were isolated by preparative HPLC and characterized spectroscopically. The spectral data were identical whether they came from the reaction of 2’-deoxyuridine with 2-bromoacrolein or from the hydrolysis of 2 with NaHC03. In addition, 5 and 6 were coinjected on an HPLC with products which were generated from the reaction of compound 2 with NaHC03, and they coeluted. Thus, the reaction of the exocyclic product 2 with sodium bicarbonate and the reaction of 2’-deoxyuridine with 2-bromoacrolein yield the same product, W (2”- bromo-3”- oxopropyl)- 2’-deoxyuridine (4), an intermediate which is converted to N3-(2”-hydroxy-3“-0~0propyl)-2’-deoxyuridine (5) and then to ~-(2”-oxo-3”hydroxypropyl)-2’-deoxyuridine(6). Reaction of %’-Deoxycytidineand %’-Deoxy uridine with Acrolein. The reaction of 2’-deoxycytidine with acrolein (Scheme 11) in 0.1 M potassium phosphate buffer
+
Chem. Res. Toxicol., Vol. 6, No. 3, 1993 265
(pH 4.3) yields one major product (635% ) which was isolated by flash chromatography and characterized by UV, FAB/ MS, IH NMR, and chemical modification. One new chiral center is formed at C-7, derived from the C-1 of acrolein, and the product exists as a mixture of diastereomers. At pH 7.4 the same compound is obtained in lower yield. The UV spectrum was recorded at several pH values, and results are characteristic of 3,N4-substitution on cytidine (25). The FAB mass spectrum shows an MH+ peak at mlz 284 (13.5) and a BHz+ peak of mlz 133 (100). The 300-MHz ‘H NMR spectrum (MezSO-d6)was compared to that for 2’-deoxycytidine (l),and three new peaks were observed. In addition, two other peaks were nearly identical to peaks in the 2’-deoxycytidine spectrum. The following assignments were made: an exchangeable singlet at 5.542 ppm for one proton (HO-7 or HO-9), a multiplet at 4.883 ppm for H-7 or H-9, a multiplet at 3.799 ppm for one proton (H-9) which overlaps the chemical shift of H-3’, another multiplet at 3.589 ppm for one proton (H-9)which overlaps the chemical shift of H-5’, and a multiplet at 2.072 ppm for two protons (H-8). The latter do not appear in the spectrum of 2’-deoxycytidine spectrum. The relative positions of the hydroxy group, whether at C-7 or (2-9, cannot be determined from the spectral data. In order to determine the location of the hydroxy group, the product was hydrolyzed with sodium hydroxide to open the ring and then reduced with sodium borohydride to give the stable alcohol. This reaction was terminated after 8 h, and HPLC and TLC analysis indicated only one major product. This substance was desalted by flash chromatography to produce a yellow oil. The ultraviolet spectrum of this product a t pH 7.0 has a ,A, at 262 nm, characteristic of 3-subs4itution on uridine (25). FABIMS indicated a dimeric species (2M + Na)+at mlz 595 (14.1), a natriated molecular ion at mlz 309 (85.5), a protonated molecular ion at mlz 287 (85), and a BHz+ peak at mlz 171 (100) corresponding to the loss of deoxyribose from the adduct. The negative ion FAB mass spectrum shows a chlorine adduct, (M + C1)-, at mlz 321 (301, a peak at mlz 285 (9.8) for (M - H)-, and a (B - H)- peak at mlz 168 (100). The IH NMR of this product was acquired in MezSO-ds, and a deuterium exchange experiment was performed. Compared to the lH NMR spectrum of 2’-deoxyuridine, four new peaks are observed, and the following assignments were made: an exchangeable singlet at 4.396 ppm for HO3”, a multiplet for two protons at 3.700 ppm for H-3” overlapping the H-3’ peak, a triplet for two protons at 3.289 ppm for H-1”, and a multiplet at 1.558 ppm for two protons for H-2”. No aldehydic or aminic proton absorptions appeared in the spectrum. This product is assigned the structure W-(3”-hydroxypropy1)-2’-deoxyuridine (10). Compound 10 was synthesized by an alternative route (Scheme 11). The method involves the reaction of 2’deoxyuridine with acrolein to produce aldehyde 11 which was isolated by HPLC, characterized spectroscopically, and converted t o the corresponding alcohol 10. 2’Deoxyuridine was permitted to react with acrolein in 0.05 M potassium phosphate buffer (pH 7.4) at 37 “C for a period of 18 h. The product was purified by preparative HPLC and was analyzed spectroscopically. Its UV spectrum at pH 7.0 shows a A,, at 262 nm, characteristic of 2’-deoxyuridine adducts. The FAB mass spectrum shows an MH+ ion at mlz 285 (10.1) and aBHz+atmlz 169 (11.3). These data are consistent with the addition of one molecule
266 Chem. Res. Toxicol., Vol. 6, No. 3, 1993
Chenna and Zden
Scheme 11. Products of the Reaction of Acrolein with 2‘-Deoxycytidine and %‘-Deoxyuridine
tio
no
0
9” and pH7 4
on
Ho
OH
v OH
9
8
1
w on
@OH
Oi
‘ i OH
7
11
of acrolein to 2’-deoxyuridine, The ‘H NMR spectrum was acquired in MenSO-de, and through a DzO exchange experiment, it was concluded (by comparison with the lH NMR of 2’-deoxyuridine) that three new protons were present: a singlet at 9.542 ppm for an aldehydic proton H-3”, a multiplet at 3.697 ppm for two protons H-1”, and a triplet at 2.556 ppm for two protons H-2”. The physical data obtained for this product indicate that its structure is that of W-(3”-oxopropyl)-2’-deoxyuridine(11). It is unstable in aqueous solution a t 37 “C and over a period of 5 weeks is converted to multiple products which were not identified. However,the product could be stored intact in aqueous solution for 5 weeks at -22 “C. As in the other cases, the reaction at pH 9.2 is faster than at neutral pH, whereas at pH 4.4 no product was found after 5 days. Compound 11was reduced to compound 10 with sodium borohydride at pH 7.0 in 2 h at room temperature. Only one major product was formed, and this displayed the same physical characteristics as the product from the hydrolysis and reduction of 8. The structure of this product is assigned as W-(3”-hydroxypropyl)-2‘-deoxyuridine (lo), and these results confirmed the position of the hydroxy group in 8 as being at C-7 and not at C-9.
Discussion Products of the reaction of 2’-deoxycytidine and 2’deoxyuridine with 2-bromoacroleinand acrolein have been characterized. The reaction of 2‘-deoxycytidine with 2-bromoacroleinand acrolein generates exocyclicproducts in which a new ring is fused to positions on the bases that normally are involved in Watson-Crick hydrogen bonding. However, the reaction of 2’-deoxyuridine with 2-bromoacrolein and acrolein forms alkylation products at N3. Results from the reaction of 2‘-deoxycytidine with 2-bromoacrolein and acrolein collectively support a mechanism in which the most basic endocyclic nitrogen atom of 2’deoxycytidine (N3, pK, = 4.17) (26,27) attacks at C-3 of 2-bromoacrolein or acrolein in a Michael addition fashion to form an W-(3”-oxopropyl) adduct. Subsequently, this intermediate cyclizes to the exocyclic amine N4 to form products. Evidence for this comes from the demonstrated conversion of product 2 to 4, 5, and 6, the conversion of 8 to 10, and the lack of alternative products resulting from
I OH 10
the attack of the exocyclic amine N4 at C-3 of 2-bromoacrolein or acrolein. An alternative mechanism was proposed by Sodum et al. ( l o ) ,who suggested that compound 9 might be formed by the attack of the exocyclic amino group N4 on C-3 of acrolein in a Michael addition fashion followed by cyclization at N3. However, we found no evidence for this since only product 10 was formed after basic hydrolysis. Smith et al. (12) reported that the reaction of 2’deoxycytidine 5’-monophosphate with acrolein at pH 7 gave an exocyclic 3,iV4-propanodeoxycytidine5‘-monophosphate having the hydroxy group at (2-7; the position of the hydroxy was determined by two-dimensional NMR analysis. They also suggested a mechanism encompassing attack of the exocyclic amino group of 2’-deoxycytidine 5’-monophosphate to C-1 of acrolein followed by ring closure at N3. However, reactions of bases (nucleophiles) with cr,fl-unsaturated aldehydes such as acrolein and 2-bromoacrolein tend to be dominated by an attack at the C-3 position due to the strong electrophilic character of C-3. Alkylation of 2‘-deoxycytidine by electrophiles such as RX gives W-alkyl adducts as the dominant product (28). The basicity of the endocyclic nitrogen is greater than that of the exocyclic N4,and thus, nucleophilic attack by N3 appears to dominate. Hydrolysis of deoxycytidine adducts 2 and 8 under basic conditions produced 2‘-deoxyuridine adducts 4-6 and 10. Presumably, 2 and 8 open to yield imines which then are deaminated. Deamination of deoxycytidine by selected mutagens is well-known. Nitrous acid, a classicaloxidative deamination reagent (29), converts deoxycytidine to deoxyuridine in DNA, which results in a predictable basepairing change (30). The mutations of bisulfite (31, 32) are likely to result from hydrolytic deamination of deoxycytidine. Solomon et al. (33,341 reported that the mutagenic three-carbon epoxides, propylene oxide and acrylonitrile epoxide, could react in vitro at N3 of deoxycytidine and subsequently undergo rapid hydrolytic deamination at pH 7.0 to form 3-hydroxyalkyldeoxyidine lesions in DNA. A mechanism was proposed in which a 2”-hydroxy moiety catalyzed the deamination which was similar to a mechanism proposed by Fujii et al. (35) for the deamination of 9-substituted adenines. We have found
dC and d U Adducts of Acrolein and 2-Bromoacrolein
that hydrolytic deamination proceeds rapidly without the 2”-hydroxy as in the transformation of 2 to 4. However, since 4 exists as the hydrate in aqueous solution, a catalytic contribution from a 3"-hydroxy cannot be excluded in the mechanism of deamination. Other well-established mechanisms for imine hydrolysis also may play an important role in this conversion ( 3 6 , 3 7 ) . We are not aware of any previous studies in which acrolein or 2-bromoacrolein adducts of 2’-deoxyuridine have been characterized chemically. However, acrolein, when permitted to react with 2-deoxyuridine under physiological conditions, formed W - ( 3 r J - ~ ~ ~ p r ~ p y 1 ) - 2 ‘ deoxyuridine (11) by Michael addition of the W-amide group of 2’-deoxyuridine to C-3 of acrolein. This product was reduced to give W-(3Jr-hydroxypropyl)-2’-deoxyuridine (10). The reaction of 2-bromoacrolein with 2’deoxyuridine gave three adducts, 4-6. Our data suggest that 4 is formed initially and is the kinetically favored product, while products 5 and 6 are formed from 4 and are favored thermodynamically. Michael addition leads to the formation of 4. Attack of hydroxide ion at the carbonyl leads to epoxide formation, displacing bromine, and then base-catalyzed opening of the epoxide ring results in formation of the hydroxy aldehyde 5 which tautomerizes, giving the final product 6 (13). Exocyclic adducts of acrolein and 2-bromoacrolein with 2‘-deoxycytidine formed under physiological conditions have been characterized at the deoxynucleoside level and were found to be unstable in basic conditions. Stable end products which proved to be 2’-deoxyuridine derivatives were identified in the case of 2-bromoacrolein, and these substances may be the ultimate adducts formed in reactions with double-stranded DNA. Synthetic standards which can be used for identification have been prepared.
Acknowledgment. The authors would like to thank Dr. Francis Johnson for many valuable discussions during the course of this work and Mr. Terry Lam for his assistance in the initial phases of this project. We also appreciate the efforts of Mr. Robert Rieger and Dr. Dasari M. Reddy in obtaining mass spectral data. This research was supported, in part, by Grants ES 04068 and CA 47995 from the National Institutes of Health.
References IARC Monograph on the Evaluation of the Carcinogenic Risk of Chemicals t o Humans; Allyl Compounds, Aldehydes, Epoxides and Peroxides (1985)International Agency for Research on Cancer, Lyon, France, Vol. 36, pp 133-161. Lutz,D.,Eder, E., Neudecker,T.,and Henschler,D. (1982)Structuremutagenicity relationshipin a$-unsaturated carbonyliccompounds and their corresponding allylic alcohols. Mutat. Res. 93, 305-315. Beauchamp,R. O., Jr., Andjelkovich,D.A.,Kligerman,A. D., Morgan, K. T.,and Heck, H. d’A. (1985)A Critical Review of the Literature on Acrolein Toxicity. CRC Crit. Reu. Toxicol. 14, 309-380. Chung, F.-L., Young, R., and Hecht, S. (1984) Formation of cyclic 1,N2-propanodeoxyguanosine adducts in DNA upon reaction with acrolein or crotonaldehyde. Cancer Res. 44, 990-995. Foiles, P. G., Akerkar, S. A., Miglietta, L. M., and Chung, F.-L. (1990) Formation of cyclic deoxyguanosine adducts in Chinese hamster ovary cells by acrolein and crotonaldehyde. Carcinogenesis 11, 2059-2061. Wilson, V. L., Foiles, P. G., Chung, F.-L., Povey, A. C., Frank, A. A., and Harris, C. C. (1991) Detection of acrolein and crotonaldehyde DNA adducts in cultured human cells and canine peripheral blood lymphocytes by ‘?P-postlabelingand nucleotide chromatography. Carcinogenesis 12, 1483-1490. Smith, R. A., Williamson, D. S., Cerny, R. L., and Cohen, S. M. (1990) Detection of l,Nfi-propanodeoxyguanosinein acrolein-modified polydeoxyadenylic acid and DNA by :jZPpostlabeling. Cancer Res. 50, 3005-3012.
Chem. Res. Toxicol., Vol. 6, No. 3, 1993 267 McDiarmid, M. A., Iype, P. T., Kolodner, K., Jacobson-Kram, D., and Strickland, P. T. (1991) Evidence for acrolein-modified DNA in peripheral blood leukocytes of cancer patients treated with cyclophosphamide. Mutat. Res. 248,93-99. Chung, F.-L., Roy, K. R., and Hecht, S. S. (1988)A study of reaction of or$-unsaturated carbonyl compounds with deoxyguanosine. J. Org. Chem. 53, 14-17. Sodum, R. S., and Shapiro, R. (1988) Reaction of acrolein with cytosine and adenine derivatives. Bioorg. Chem. 16, 272-282. Shapiro, R., Sodum, R. S., Everett, D. W., and Kundu, S. K. (1986) Reactions of nucleosides with glyoxal and acrolein. In The Role of Cyclic Nucleic Acid Adducts in Carcinogenesis and Mutagenesis (Singer, B., and Bartsch, H., Eds.) pp 165-173, International Agency for Research on Cancer, Lyon, France. Smith,R. A., Williamson,D. S.,andCohen,S.M. (1989)Identification of 3,N4-propanodeoxycytidine5’-monophosphate formed by the reaction of acrolein with deoxycytidine 5’-monophosphate. Chem. Res. Toxicol. 2, 267-271. Chenna, A., Rieger, R. A., and Iden, C. R. (1992) Characterization of thymidine adducts formed by acrolein and 2-bromoacrolein. Carcinogenesis 13, 2361-2365. Marinelli, E. R., Johnson, F., Iden, C. R., and Yu, P.-L. (1990) Synthesis of 1JV-(1,3-propano)-2’-deoxyguanosine and incorporation into oligodeoxynucleotides: A model for exocyclic acrolein-DNA adducts. Chem. Res. Toricol. 3, 49-58. Benamira, M., Singh, U., and Marnett, L. J. Frameshift mutagenesis by a propanodeoxyguanosine adduct site-specifically located in the mutable region of Salmonella typhimurium TA98 contained on an M13 vector. Proceedings of the 83rd Annual Meeting of the American Association for Cancer Research, San Diego, CA, May 20-23, 1992. Rosen, J. D., Segall, Y., and Casida, J. E. (1980)Mutagenic potency of haloacroleins and related compounds. Mutat. Res. 78,113-119. Nelson, S. D., Omichinski, J. G., Iyer, L., Gordon, W. P., Soderlund, E. J., and Dybing, E. (1984) Activation mechanism of tris(2,3dibromopropy1)phosphate to the potent mutagen, 2-bromoacrolein. Biochem. Biophys. Res. Commun. 121, 213-219. Omichinski, J. G., Soderlund, E. J., Dybing, E., Pearson, P. G., and Nelson, S. D. (1988) Detection and mechanism of formation of the potent direct-acting mutagen 2-bromoacrolein from 1,a-dibromo3-chloropropane. Toxicol. Appl. Pharmacol. 92, 286-294. Gordon, W. P., Soderlund, E. J., Holme, J. A., Nelson, S. D., Iyer, L., Rivedal, E., and Dybing, E. (1985) The genotoxicity of 2-bromoacrolein and 2,3-dibromopropanal. Carcinogenesis 6, 705-709. Meermann, J. H. N., Pearson, P. G., Meier, G. P., and Nelson, S. D. (1988) Formation of six cyclic 1,NZ-hydroxybromopropanodeoxyguanosine isomers upon reaction of 2-bromoacrolein with 2‘deoxyguanosine. J. Org. Chem. 53, 30-35. Meerman, J. H. N., Smith, T. R., Pearson, P. G., Meier, G. P., and Nelson, S.D. (1989)Formationof cyclic l,N?-propanodeoxyguanosine and thymidine adducts in the reaction of the mutagen 2-bromoacrolein with calf thymus DNA. Cancer Res. 49, 6174-6179. van Beerendonk, G. J. M., Nivard, M. J. M., Vogel, E. W., Nelson, S. D., and Meerman, J. H. N. (1992)Formation of thymidine adducts and cross-linkingpotential of 2-bromoacrolein,a reactive metabolite of tris(2,3-dibromopropyl)phosphate.Mutagenesis 7, 19-24. Sherlin, S. M., Berlin, A. Y., Serebrennikova,T. A., and Rabinovich, F. E. (1938) Polymerization of acrolein and acrylic acid and the structure of their dimers. J . Gen. Chem. (USSR), 8, 22-34. Alder, K., andRuden, E. (1941) PolymerizationprocessesX: Dimeric acrolein. Chem. Ber. 74, 920-926. Singer, B. (1975)UV spectral characteristicsand acidic dissociation constants of 280 alkyl bases, nucleosides, and nucleotides. In CRC Handbook of Biochemistry and Molecular Biology, Nucleic Acids (Fasman, G.D., Ed.) Vol. I, 3rd ed., pp 409-447, CRC Press, Boca Raton, FL. Clauwaet, J., and Stock, J. (1968) Interactions of polynucleotides and their components. I. Dissociation constants of the bases and their derivatives. 2.Naturforsch. B. 23, 25-30. Dunn, D. B. and Hall, R. H. (1975)Purines, pyrimidines, nucleosides, and nucleotides: Physical constants and spectral properties. In CRC Handbook of Biochemistry and Molecular Biology, Nucleic Acids (Fasman, G.D., Ed.) Vol. I, 3rd ed., pp 409-447, CRC Press, Boca Raton, FL. Singer, B. (1975) The chemical effects of nucleic acid alkylation and their relation to mutagenesis and carcinogenesis. Prog. Nucleic Acid Res. Mol. Biol. 15, 219-284. Shester, H. (1960)Thereaction ofnitrousacid withdeoxyribonucleic acid. Biochem. Biophys. Res. Commun. 2, 320-323. Singer, B., and Kusmierek, J. T. (1982) Chemical niutagenesis. Annu. Rev. Biochem. 52, 655-693.
268 Chem. Res. Toxicol., Vol. 6, No. 3, 1993 (31) Hayatsu, H. (1976) Bisulfite modification of nucleic acids and their constituents. Prog. Nucleic Acid Res. Mol. Biol. 16, 76-125. (32) Shapiro, R. (1982) Genetic effects of bisulfate: Implications for environmental protection. In Induced Mutagenesis Molecular Mechanisms and Theirlmplications for EnuironmentalProtection (Lawrence, C. W., Prakash, L., and Sherman, F., Eds.) pp 35-60, Plenum Press, New York. (33) Solomon,J. J., and Segal, A. (1989) DNA adducts of propylene oxide and acrylonitrile epoxide: Hydrolytic deamination of 3-alkyl-dCyd to 3-alkyl-dUrd. Enuiron. Health Perspect. 81, 19-22. (34) Solomon, J. J., Decker-Samuelian, K., Li, F., Mukai, F., and Segal, A. (1990)On the rapid hydrolytic deamination of epoxide-generated 3-hydroxyalkyl cytosine to 3-hydroxyalkyl uracil. Proceedings of
Chenna and Iden the 38th ASMS Conference on Mass Spectrometry and Allied Topics, Tucson, AZ, June 3-8, 1990. (35) Fujii, T., Saito, T., and Terahara, N. (1986) Purines. XXVII. Hydrolytic deamination uersus Dimroth rearrangement in the 9-substituted adenine ring: Effect of an w-hydroxyalkyl group at the 1-position. Chem. Pharm. Bull. 34, 1094-1107. (36) Hine, J., Craig, J. C., Jr., Underwood, J. G., 11,and Via, F. A. (1970) Kinetics and mechanism of the hydrolysis of n-isobutylidenemethylamine in aqueous solution. J. Am. Chem. SOC. 92, 51945199. (37) Cordes,E. H., and Jencks, W. P. (1963)The mechanismof hydrolysis of Schiff bases derived from aliphatic amines. J. Am. Chem. SOC. 85, 2843-2848.
