Formation of acyclic and cyclic guanine adducts in DNA reacted with

Division of Chemical Carcinogenesis, American Health Foundation, Valhalla,New York 10595. Received July 13, 1989. This paper describes the reactionof ...
0 downloads 0 Views 719KB Size
Chem. Res. Toxicol. 1989,2,423-428

423

Formation of Acyclic and Cyclic Guanine Adducts in DNA Reacted with a-Acetoxy-N-nitrosopyrrolidine’ Mingyao Wang,* Fung-Lung Chung, and Stephen S. Hecht Division of Chemical Carcinogenesis, American Health Foundation, Valhalla, New York 10595 Received July 13, 1989

This paper describes the reaction of a-acetoxy-N-nitrosopyrrolidine with DNA to produce six adducts: two new acyclic adducts, 7-(4-oxobutyl)guanine (6) and 7-(3-~arboxypropyl)guanine (7), and four cyclic adducts-the exocyclic 7,gguanine adducts 5, 11, and 12 and an exocyclic 1,P-guanine adduct 13-which we have previously characterized. The initial purpose of this study was to carry out an independent synthesis to verify the structure of adduct 5, which is formed in liver DNA of rats treated with N-nitrosopyrrolidine. This was accomplished by the reaction of 2‘,3‘,5‘-triacetylguanosinewith 4-iodobutyraldehyde. This reaction also produced 7-(4-oxobutyl)guanine (6), which underwent air oxidation to 7. The new adducts were characterized by their proton NMR, W, and mass spectral properties, by chemical transformations, and by independent syntheses. The six adduct standards were used to develop HPLC systems for their analysis as products of the reaction of a-acetoxy-N-nitrosopyrrolidmewith DNA. Studies on their rates of formation and stability in DNA were carried out. The major products were 7-(4-oxobutyl)guanine (6) and the exocyclic 7,Ei-guanine adduct 5, which apparently were both formed mainly by reaction with DNA of 4-oxobutanediazohydroxide(4). Their concentrations were maximal after 6 h and subsequently decreased due to depurination. Little evidence was obtained for cyclization of 6 to 5, a t the base level or in DNA. The concentrations of adducts 11-13, which were formed by reaction with DNA of crotonaldehyde (lo), increased gradually over the 36-h time period studied. The results of this study provide strong support for the proposed mechanism of DNA adduction by a-hydroxylation of the hepatocarcinogen Nnitrosopyrrolidine.

Introduction In this paper, we describe the reaction of a-acetoxyNPYR2 (1) with DNA to form six adducts (5-7 and 11-13 of Figure 1). We are interested in NPYR because it is found in the dietary and respiratory environments, is carcinogenic in laboratory animals, and is a representative cyclic nitrosamine (1-4). It is the structural parent of commonly occurring environmental nitrosamines such as N’-nitrosonornicotine and N-nitrosoproline. a-Hydroxylation is a major pathway of metabolic activation of NPYR (5-8). The initial product of this pathway, a-hydroxyNPYR (3), is unstable and spontaneously undergoes ring opening to a putative intermediate, the aldehydic diazohydroxide 4. Further reactions of 4 give 2-hydroxytetrahydrofuran (8) and crotonaldehyde (lo), which are metabolites of NPYR (5-8). The reactions with DNA of products formed by a-hydroxylation of NPYR are of interest because of its carcinogenicity. Since 3 and 4 are unstable, we have used a-acetoxyNPYR to generate them in situ and determine their reaction products with DNA (9, 10). Previously, we have characterized compounds 5 and 11-13 as products of this reaction. Two of these adducts, 5 and 13, have been detected in hepatic DNA of NPYR-treated rats and are the only fully characterized DNA adducts of cyclic nitrosamines which have been found in vivo in laboratory animals (10, 12). The former was detected by HPLC with fluorescence detection whereas the latter was detected as the corresponding nul Paper 127 in the series ‘A Study of Chemical Carcinogenesis”. * Abbreviations: a-acetoxyNPYR, a-acetoxy-N-nitrosopyrrolidine; NPYR, N-nitrosopyrrolidine; TAG, 2’,3’,5’-triacetylguanosine.

cleotide by 32P-postlabeling(10, 12). The initial purpose of this study was to obtain greater quantities of 5 than were previously available. This would allow accurate quantitation of its levels in vivo. We also wished to confirm its structure by an independent synthesis. To accomplish this, we investigated the reaction of 4-iodobutyraldehyde with 2’,3’,5’-triacetylguanosine (TAG). As discussed below, 5 was obtained in this reaction. However, two new adducts, 6 and 7, were also characterized. With these new standards, we reinvestigated the reaction of a-acetoxyNPYR with DNA and determined the relative rates of formation and stabilities in DNA of the six adducts illustrated in Figure 1.

Experimental Section Apparatus. HPLC was carried out with a Waters Associates system (Millipore, Waters Division, Milford, MA) equipped with a Model 990 photodiode m a y detector or a Perkin-ElmerModel 650-10s fluorescence detector (Perkin-Elmer Corp., Norwalk, CT). The following solvent elution systems were used: System 1 was a 9.4 mm i.d. X 50 cm Partisill0 ODS-3Magnum 9 column (Whatman, Clifton, NJ) eluted isocratically with 15% MeOH in H20 for 5 min, and then with a gradient from 15 to 25% MeOH in H20 in 30 min at 4 mL/min. Detection was by UV absorbance at 254 or 285 nm. System 2 was two 4.6 mm X 25 cm Partisil 10 SCX strong cation exchange columns (Whatman) eluted isocratidy with 0.04 M ammonium phosphate buffer, pH 2.0, at 1 mL/min. Detection was by fluorescence with excitation at 290 nm and emission at 380 nm. System 3 was two 3.9 mm X 30 cm CISpBondapak reversephase columns (Waters) in series, eluted isocratically with 15% MeOH in H20 for 5 min, and then with a gradient from 15 to 25% MeOH in H20 in 30 min, at 1 mL/min using curve 6. Detection

0893-228x/89/2702-0423$01.50/00 1989 American Chemical Society

424

Chem. Res. Toxicol., Vol. 2, No. 6, 1989

9

NPYR,

Wang et al.

I

NrO

1

2) O.1N HCI, 75OC

k

li

I3 Figure 1. Formation of DNA adducts from a-acetoxyNPYR. Only one of two possible enantiomers of adducts 11-13 is shown. was by UV absorbance at 254 or 285 nm. System 4 was two new 4.6 mm X 25 cm Partisil 10 SCX columns in series eluted isocratically with 0.125 M ammonium phosphate, pH 2.0, at a flow rate of 1.5 mL/min. Detection was by fluorescence as in system 2. UV spectra were recorded on a Hewlett-Packard 8452A diode array spectrophotometer coupled to an IBM XT personal computer. MS were run on a Hewlett-Packard Model 5988A instrument, using electron impact or desorption chemical ionization, with NH&l added to the samples. NMR spectra were recorded on a Bruker Instruments superconducting Fourier NMR spectrometer, AM 360 WB. Chemicals. TAG was obtained from Aldrich Chemical Co., Milwaukee, WI. a-AcetoxyNPYR, 4-iodobutanol, and ethyl 4iodobutyrate were synthesized (13-15). Calf thymus DNA and hog liver esterase were obtained from Sigma Chemical Co., St. Louis, MO. Warning: a-AcetoxyNPYR is a strong mutagen and presumed carcinogen. It should be handled with care. Reaction of 4-Iodobutyraldehyde and TAG. 4-ChlOrObutyraldehyde (1.5 g, 0.014 mol) (16) and NaI (2.5 g, 0.017 mol) were dissolved in 11mL of acetone, and the mixture was heated under reflux for 9 h. The acetone was removed, and 150 mL of EhO was added. The precipitate was filtered and the EhO filtrate was washed three times with 75 mL of 1M NGZO3and then dried (MgSO,). The Et20 was removed to give crude 4-iodobutyraldehyde (1.5 g, 54%): NMR (CDC13) 6 9.8 (s, 1, CHO), 3.2 (t, 2, CHzI), 2.6 (m, 2, CH2CHO), 2.1 (m, 2, CH2CH2CH2).It was used without further purification due to its instability. 4-Iodobutyraldehyde (1g, 5.1 mmol) and TAG (1g, 2.4 mmol) were dissolved in 5 mL of anhydrous DMF. The mixture was stirred at room temperature for 21 h and then heated at 37 "C for 48 h. The DMF was removed by rotary evaporation, and the residue was dissolved in 15 mL of CHCIS. This solution was extracted with three 500-mL portions of H20. The combined aqueous extracts were extracted with three 250-mL portions of CHC13 and then concentrated to dryness. The residue was dissolved in 12 mL of 10 mM sodium cacodylate buffer, pH 7.0, and heated at 100 "C for 1h. The hydrolysate was analyzed by HPLC system 1. Material eluting at the approximate retention time of 5,15-18 min, was collected and analyzed by HPLC systems 2 and 3. Compounds 5 (2.6 mg, 0.5% yield) and 6 (2.5 mg, 0.5% yield) were purified by repetitive collections from systems 1 and 2. Preparation of 7-(3-Carboxypropyl)guanine(7) a n d 7(4-Hydroxybuty1)guanine (14). A mixture of TAG (409 mg, 1 mmol) and ethyl 4-iodobutyrate (409 mg, 1.7 mmol) in 15 mL of DMF was heated at 110 "C for 2.5 h. The DMF was removed in vacuo to give a brown residue. Thirty milliliters of CHC1, was added, and the mixture was allowed to stand at room temperature

for 3 h. A solid formed which was collected by filtration. It was recrystallized from ethanol to yield 40 mg of crude product. This was recrystaked from 20% aqueous EtOH to afford white shiny crystals of 7-(3-carboxypropyl)gineethyl ester, mp >280 "C dec, which was used without further purification. The ester (27 mg, 0.1 "01) was heated in 6 mL of refluxing 0.04 N NaOH for 10 min. An equal volume of H 2 0 was added, and the mixture was cooled to room temperature. The pH of the solution was adjusted to 4-5 by addition of 0.1 N HC1. 7-(3-Carboxypropy1)guanine (15 mg, 0.06 mmol, 60% yield) precipitated as a white solid, mp >280 "C [lit. (17)274 "C dec]. For MS, a sample was derivatized by heating at 90 "C for 1 h with excess bis(trimethylsily1)trifluoroacetamideand a catalytic amount of pyridine. The electron impact MS of the tris-TMS derivative had m/e (relative intensity) 453 (M+, 19) and 438 (100). UV and NMR spectral properties are presented in the text. The same procedure was used to prepare 14, mp >215 "C, in 10% yield, from TAG and 4-iodobutanol: MS, electron impact, m / e (relative intensity) 223 (M+, 10.3), 151 (100); UV (nm) pH 1, 286, 233 (sh), pH 13, 283, 245 (sh), MeOH, 280.5, 240.5. Other spectral properties are described in the text. Reaction of a-AcetoxyNPYR with DNA. Mixtures of calf thymus DNA (10 mg), a-acetoxyNPYR (30 mg, 0.19 mmol), and esterase (28 units) in 2 mL of 0.1 M phosphate buffer, pH 7.0, were heated at 37 OC for intervals as illustrated in Figure 6. The reaction mixture was added to 2 mL of HzO, extracted with 5 mL of CHC13/isoamylalcohol (201), and centrifuged at 14000g for 20 min. The supernatant was removed, and 10 mL of ethanol was added to precipitate the DNA. The DNA was dissolved in 1 mL of 10 mM sodium cacodylate buffer, pH 7.4, and heated at 100 "C for 1h, or at 37 "C to release adducts 5-7, 11, and 12, which were determined by HPLC system 4. After cooling, 40 pL of 1 N HC1 was added to precipitate the DNA. The DNA was dissolved in 0.5 mL of 0.1 N HC1, and the solution was heated at 75 "C for 45 min. It was analyzed by HPLC system 4 for adduct 13 and guanine. Retention times were as follows (min): 5 (17), 6 (12.1), 7 (8.8), 11 (12.7), 12 (13.4), 13 (19.6), and 14 (15.0). Stability of Adducts in DNA. A mixture of calf thymus DNA (41.4 mg), a-acetoxyNPYR (126 mg,0.8 mmol), and esterase (112 units) in 8 mL of 0.1 M phosphate buffer, pH 7.4, was incubated at 37 "C for 12 h. Ten milliliters of H 2 0 was added, and the mixture was extracted with 20 mL of CHC13/isoamylalcohol (201) and then centrifuged at 14000g for 20 min. The aqueous layer was removed, and to it was added 40 mL of cold ethanol to precipitate the DNA. The dried DNA was dissolved in 10 mL of 10 mM sodium cacodylate buffer, pH 7.0. Aliquots of 0.8 mL were incubated at 37 OC for 0,2,5,8, 12,16,20, 24,28, and 40 h. One 0.4mL aliquot

Chem. Res. Toxicol., Val. 2, No. 6, 1989 425

Reaction of DNA with a-Acetoxy-N-nitrosopyrrolidine

Table I. Proton NMR Chemical Shift Data for Adducts 6, 7, and 14 0

1-CHz 2-CHz 3-CHz 4-CHO 4-CH20H C8-H H2N HN a

6

7

4.17, t, J = 6.7 Hz 2.03, m 2.40, t, J = 7.2 Hz 9.60, s

4.18, t, J = 6.7 Hz 1.99, m 2.15, t, J = 7.0 Hz

4.17, t, J = 6.9 Hz 1.77, m 1.33, m

7.90, s 6.11, s 10.7, s

7.87, s 6.10, s 10.7, 8

CH2 3.4;* OH 4.42, t, J = 5.2 Hz 7.90, s 6.10, s 10.7, s

14

Spectra were run in DMSO. *Partiallyobscured by H20.

I

I4

!i!

=

3

d L

s 8

e

B

I

200

250

300 350 Wavelength (nm)

400

Figure 3. UV spectra of peak 3 of Figure 2. 0

4 8 12 16 20 24 Retention Time (min)

'2'

DMSO

Figure 2. Chromatogram obtained upon HPLC analysis in system 2 of products formed in the reaction of TAG with 4iodobutyraldehyde. A preliminary purification was carried out with HPLC system 1 as described under Experimental Section. Peaks 2,3, and 4 were identified as adducts 7,6, and 5 of Figure 1, respectively. of each resulting solution ("aliquot A") was mixed with 40 pL of cold 1N HCl to precipitate the DNA. The "aliquot A" supernatants were analyzed by HPLC system 4 to determine levels of spontaneously released adducts. The remaining 0.4-mL aliquot of each solution ("aliquot B") was heated at 100 O C for 1 h to release adducts 5-7,11, and 12. The DNA in aliquot B was then precipitated by addition of 40 pL of 1N HC1 and dissolved in 0.5 mL of 0.1 N HC1. The solution was heated at 75 O C for 45 min to release guanine and adduct 13. The neutral thermal and acid hydrolysates of aliquot B were analyzed by HPLC system 4. The levels of adducts in aliquot B represented the amounts remaining in DNA plus the amounts that had been lost by spontaneous depurination at each time point. The levels of adducts in aliquot A represented only that portion which was spontaneously released by depurination at 37 O C . The difference between the amounts in aliquots B and A represented the level of each adduct remaining in DNA at each time point.

Results Reaction of TAG with Ciodobutyraldehyde, followed by neutral thermal hydrolysis, gave a complex mixture of products, which was partially purified by reverse-phase HPLC. The region corresponding in approximate retention time to 5 was collected and analyzed on a strong cation exchange HPLC column with fluorescence detection. This gave the chromatogram illustrated in Figure 2. Peak 4 had the same retention time as 5, which had previously been characterized as a product of the reaction of a-acetoxyNPYR with DNA (10). It was collected and its UV and proton NMR spectra were determined. They were

4:2

4:O 3:8 3:6 3:4 3.'2 3:O 218 2:6 2:4

11.0 10.0

9.0

8.0

7.0

212 2:O

6.0 5.0 PPM

1:8

I I

4.0

3.0

2.0

Figure 4. Proton NMR spectrum of peak 3 of Figure 2. essentially identical with the published spectra of 5 (10) confirming the structure. The UV spectra of peak 3 of Figure 2 are illustrated in Figure 3. The pH dependency of the spectra indicated that it was a 7-substituted guanine derivative. The spectra were very similar to those of 5, 11, 12, and other 7-substituted guanines (10, 18). The proton NMR spectrum of peak 3 is shown in Figure 4, and chemical shift data are summarized in Table I. The singlets at 9.60 and 7.90 ppm were assigned as an aldehyde proton and the C-8proton of guanine, respectively. The singlets at 10.7 and 6.11 ppm disappeared upon addition of D20 and were assigned as N-1H and NH2 of guanine. The assignments of the three pairs of methylene protons were established by irradiating at 2.0 ppm; the triplets at 2.40 and 4.17 ppm collapsed to singlets. These proton NMR data were entirely consistent with the assigned structure, 6. The DCI-MS of peak 3

426 Chem. Res. Toxicol., Vol. 2, No. 6,1989

A

Wang et al.

1 G"a

20 i

0Retention Time (min)

Figure 5. HPLC separation using sytem 4 of (A) standards and (B) products released upon neutral thermal hydrolysis of DNA that had been allowed to react with a-acetoxyNPYR in the presence of esterase. Upon NaBH, treatment of the thermal hydrolysate, adduct 6 disappeared, adduct 14 formed, and the

other adducts remained unchanged.

showed the following peaks: m / e (relative intensity) 222 [(M + l)+, 21, 152 (62), 136 (52), 93 (94), 72 (100). Since the molecular ion was small, we carried out further experiments to confirm the structure of peak 3. Peak 3 wtts treated with Na13H4. The product was purified by HPLC, and its proton NMR spectrum and MS were obtained. The NMR spectral data are summarized in Table I. The assignments, which were confirmed by decoupling experiments, were consistent with its structure, 744-hydroxybuty1)guanine (14). The DCI-MS of 14 had the following peaks: m / e (relative intensity) 224 [(M + l)', 1001, 166 (92), 124 (22), 108 (30). The HPLC retention time and spectral properties of 14, obtained by Na13H4 reduction of peak 3, were identical with those of 14 prepared by reaction of TAG with 4-iodobutanol. Taken together, these data firmly establish the structure of peak 3 as 7-(4-oxobuty1)guanine (6). The UV spectra of peak 5 of Figure 2 indicated that it was also a 7-substituted guanine. Although it was apparently pure by HPLC analysis, the proton NMEt was not that of a 1:l adduct. Rather, there appeared to be more than one alkyl moiety per guanine. Reproducible MS data could not be obtained. It was not reduced with Na13H4 and did not have an aldehyde proton. Its structure was not further pursued. Upon standing in H20 at room temperature, 6 was gradually converted to a new product. Its W spectra were similar to that of 6 with maxima at 250,279, and 283 nm at pH 1, 12, and 7, respectively. We suspected that the new product, which was shown to be identical with peak 2 of Figure 2, was 7-(3-~arboxypropyl)guanine (7). This was synthesized by reaction of TAG with ethyl 4-iodobutyrate, followed by hydrolysis. The proton NMR spectrum, UV spectra, and HPLC retention time of the synthetic standard were identical with those of the new product, confirming its structure as 7. The proton NMR spectral data of 7 are summarized in Table I. In order to investigate the formation of adducts 5-7,11, and 12 in the reaction of a-acetoxyNPYR with DNA, we developed conditions (HPLC system 4) to separate them. This system also separated 14 from the other compounds, as illustrated in Figure 5A. HPLC analysis of the products formed by neutral thermal hydrolysis of DNA that had been incubated with a-acetoxyNPYR and esterase for 12

0

A

4

8

12 16 20 24 28 32 36 40

Time (h)

Figure 6. Formation of adducts 5 (m), 6 ( O ) , 7 (A),11 (X), 12 (A),and 13 ( 0 )in the reaction of a-acetoxyNPYR ( 0 )with DNA

for various lengths of time.

h gave the chromatogram illustrated in Figure 5B. The presence of adducts 5,11, and 12 in this mixture had been previously confirmed. The presence of 6 was confirmed by treating the mixture with NaBH4, which converted 6 to 14. The presence of 7 was established by collecting it and analyzing with HPLC system 3. It coeluted with standard 7 and had an identical UV spectrum. Figure 6 illustrates the formation of adducts 5-7 and 11-13 with time in incubations of a-acetoxyNPYR and DNA in the presence of esterase. Adducts 5-7,11, and 12 were quantified by HPLC analysis of neutral thermal hydrolysates, while adduct 13, which is stable to neutral thermal hydrolysis, was quantified by acid hydrolysis and HPLC analysis. Most of the a-acetoxyNPYR had disappeared by 3 h, as determined by HPLC in system 3 (see Figure 6). Adducts 5 and 6 were formed in the greatest amounts, and their concentrations were maximal after 6 h. At later time points, their concentrations decreased due to spontaneous depurination (see below). The concentrations of adducts 7 and 11-13 gradually increased with time. The stability of the adducts in DNA was investigated in order to determine their rates of spontaneous depurination as well as the extent to which adduct 6 might cyclize to 5 or undergo other reactions such as cross-linking. These reactions seemed possible because adduct 6 had a free aldehyde group. DNA that had been allowed to react with a-acetoxyNPYR for 12 h in the presence of esterase was isolated, purified, and then incubated at 37 "C and pH 7.0. Aliquots were analyzed at various time points from 0 to 40 h. In this analysis, we measured the adducts released from DNA by spontaneous depurination as well as the adducts remaining in the DNA at each time point (see Experimental Section). This allowed us to determine not only the half-life of each adduct in DNA at 37 "C but also the stabilities and possible interconversion in DNA of adducts 5 and 6. The half-lives (h) of the adducts in DNA at 37 "C were as follows: 5 (54), 6 (33), 7 (45), 11 (32), and 12 (31). For adducts 5 and 6, the sums of the amounts remaining in DNA and the amounts lost by spontaneous depurination at the beginning and end of the 40-h time period (e.g., amounts in "aliquot B") were as follows (mmol/mol of guanine): 5 (25.3,O h; 26.1,40 h) and 6 (20.5, 0 h; 18.5, 40 h). Thus, the levels of adducts 5 and 6 did not significantly change over the 40-h time period studied. Other than depurination, adducts 5 and 6 in DNA ap-

Reaction of DNA with a-Acetoxy-N-nitrosopyrrolidine peared to be relatively unreactive.

Dlscusslon Hunt and Shank observed a highly fluorescent putative DNA adduct in the livers of rats treated with NPYR but were unable to obtain sufficient material for structural characterization (11). We detected the same adduct in hepatic DNA of NPYR-treated rats and established that it was identical with one of the products of reaction of a-acetoxyNPYR with DNA, adduct 5 of Figure 1 (10). The structure of 5 was determined by its NMR, UV, and mass spectral properties (10). Although this structural assignment seemed unlikely to be incorrect, it was based in part on the expected chemistry of a hypothetical intermediate, the aldehydic diazohydroxide 4. One goal of this study was to develop an independent synthesis of 5. 4-Iodobutyraldehyde can be viewed as a stable model for 4. It would be expected to react initially with N-7 of guanosine, with nucleophilic displacement of iodide, analogous to displacement of nitrogen from 4. Thus, the isolation of 5 as a product of the reaction of 4-iodobutyraldehyde with TAG confirms our structural assignment. Adduct 5 was apparently formed by a concerted process in which nucleophilic displacement of iodide by N-7 of TAG and ring closure at C-8 occurred simultaneously. Little evidence was obtained for ring closure of the guanine adduct 6 or its precursor in DNA, as discussed further below. A second goal of this study was to obtain larger quantities of 5 to use as a standard for its quantitation in vivo. Although the yield of 5 from the reaction of TAG with 4-iodobutyraldehyde was low, approximately 0.5%,it was about 10 times greater than the yield of 5 from the reaction of a-acetoxyNPYR with DNA. The reactions of simple alkanediazohydroxides with DNA give a spectrum of products among which 7-alkylguanines and phosphotriesters generally predominate (19). Little is known about the reactions with DNA of aldehydic diazohydroxides such as 4 which are formed in the metabolic activation of cyclic nitrosamines. Although adducts like 6 would be expected as products of this reaction, investigations of their formation were precluded by the lack of available standards. The isolation of 6 from the reaction of TAG with 44odobutyraldehyde provides a solution to this problem. Using 6 as a standard, we were able to demonstrate its formation in the reaction of a-acetoxyNPYR with DNA. These results, together with the data on production of 5 in this reaction, strongly support the intermediacy of the aldehydic diazohydroxide 4, or a related carbonium ion, in the hydrolysis of a-acetoxyNPYR and in metabolism of NPYR by a-hydroxylation. The chemistry of compounds such as 6, which are 7substituted guanines containing a free aldehyde group, has not been previously studied. 7-(2-Oxoethyl)guanine, formed by the reaction with deoxyguanosine of chloroethylene oxide, is the only other known 7-substituted guanine having an aldehyde group; it appears to exist mainly as a cyclic hemiacetal in which the aldehyde has reacted with the oxygen at position 6 (20). Intramolecular or intermolecular reactions of the aldehyde group of 6 might be expected, leading perhaps to intrastrand or interstrand cross-links. Although this requires further investigation, our initial studies suggest that these reactions do not readily occur. Adduct 6 was stable to the neutral thermal hydrolysis conditions used for its release from DNA, and upon storage at room temperature, the only further reaction observed was gradual oxidation to the corresponding acid 7. Our studies of the stability in DNA

Chem. Res. Toxicol., Vol. 2, No. 6, 1989 427 of adduct 6 suggest that depurination, rather than further reactions such as cross-linking or cyclization, is the major route for its disappearance. The loss of 6 from DNA, as indicated in Figure 6, can be accounted for mainly by depurination, which occurs with a half-life of 33 h, and to a lesser extent by oxidation to 7. The studies on the stability of 6 in DNA also indicated that intramolecular cyclization to adduct 5 was minimal. Therefore, the mechanism of formation of 5 in DNA probably involves concerted attack of N-7 and C-8 of deoxyguanosine on diazohydroxide 4 or a related electrophile, rather than initial displacement of nitrogen by N-7 followed by ring closure, as previously suggested (10). Initial reaction of 4 at N-7 of deoxyguanosine gives predominately 6. Adducts 5 and 6 were the major products of the esterase-catalyzed reaction of a-acetoxyNPYR with DNA. Their concentrations were maximal after 6 h whereas the concentrations of the minor adducts 7 and 11-13 increased gradually with time. Adduct 7 was most likely formed by oxidation of 6. Adducts 11-13 are produced by reaction of DNA with crotonaldehyde (10) (10). Since 13 is stable in DNA, its gradually increasing concentrations can be accounted for by reaction of DNA with crotonaldehyde released upon hydrolysis of a-acetoxyNPYR. The levels of 11 and 12 detected between 3 and 36 h are a composite of their formation by reaction of DNA with crotonaldehyde and their loss due to spontaneous depurination. The half-life of adduct 5 in DNA, as determined in the present study, was 54 h. In our previous study, we observed a half-life of 13.6 h for adduct 5 (IO). The only apparent difference between the two studies was the initial concentration of adduct 5 in DNA 25.3 mmol/mol of guanine in the present study compared to approximately 4 mmol/mol of guanine in the previous one. It is unclear why initial adduct concentrations would affect half-lives. This requires further investigation. In the study summarized in Figure 6, most of the aacetoxyNPYR had disappeared by 3 h. However, the levels of adducts 5 and 6 increased between 3 and 6 h. This suggests that a relatively stable intermediate was formed which reacted with DNA to give 5 and 6. Nothiig is known about the stability of the aldehydic diazohydroxide 4, but its half-life would be expected to be short. It is possible that 4 may have reacted with DNA to give an intermediate such as a phosphotriester, which then reacted with N-7 of deoxyguanosine, leading to 5 and 6. Thus, 5 and 6 may have been formed both by direct reaction with diazohydroxide 4 and by a second pathway involving a more stable intermediate. Our investigations of adduct levels in hepatic DNA of NPYR-treated rats have shown that the concentration of adduct 5, 24 h after treatment with NPYR, is approximately 10000 times as great as that of adduct 13. The relatively low concentration of the latter is believed to be due to competitive pathways, such as reaction with thiols which scavenge crotonaldehyde before it can react with DNA. Such pathways should not interfere with the in vivo formation of adducts 6 and 7. The standards and HPLC systems developed in this study will allow us to quantify levels of these adducts in rats treated with [3H]NPYR.

Acknowledgment. This study was supported by Grant CA-44377 from the National Cancer Institute. We thank Susan LaGreca for her outstanding technical assistance.

References (1) Assembly of Life Sciences. Committee on Nitrite and Alterna-

tive Curing Agenta in Food. (1981) N-Nitroso compounds: environmental distribution and exposure of humans. In The Health

428 Chem. Res. Toxicol., Vol. 2, No. 6, 1989 Effects of Nitrate, Nitrite, and N-Nitroso Compounds, Chapter 7, National Academy Press, Washington, DC. (2) Preussmann, R. (1984) Occurrence and exposure to N-nitroso compounds and precursors. In N-Nitroso Compounds: Occurrence, Biological Effects and Releuance to Human Cancer (0'Neill, I. K., Von Borstel, R. C., Miller, C. T., Long, J., and Bartsch, H., Eds.) pp 3-15, International Agency for Research on Cancer, Lyon, France. (3) Hoffmann, D., Brunnemann, K. D., Adams, J. D., and Hecht, S. S. (1984) Formation and analysis of N-nitrosamines in tobacco products and their endogenous formation in consumers. In NNitroso Compounds: Occurrence, Biological Effects and Releuance to Human Cancer (ONeill, I. K., Von Borstel, R. C., Miller, C. T., Long, J., and Bartsch, H. Eds.) pp 743-762, International Agency for Research on Cancer, Lyon, France. (4) Preussmann, R., and Stewart, B. W. (1984) N-Nitroso carcinogens. In Chemical Carcinogens (Searle, C. E., Ed.) 2nd ed., pp 643-828, American Chemical Society, Washington, DC. (5) Hecht, S. S., McCoy, G. D., Chen, C. B., and Hoffmann, D. (1981) The metabolism of cyclic nitrosamines. In N-Nitroso Compounds (Scanlan, R. A., and Tannenbaum, S. R., Eds.) ACS Symposium Series 174, pp 247-273, American Chemical Society, Washington, DC. (6) Hecht, S. S., Chen, C. B., and Hoffmann, D. (1978)Evidence for metabolic a-hydroxylation of N-nitrosopyrrolidine. Cancer Res. 38,215-218. (7) Hecker, L. I., Farrelly, J. G., Smith, J. H., Saavedra, J. E. and Lyon, P. A. (1979) Metabolism of the liver carcinogen N-nitrosopyrrolidine by rat liver microsomes. Cancer Res. 39, 2679-2686. (8) Wang, M. Y., Chung, F.-L., and Hecht, S. S. (1988) Identification of crotonaldehyde as a hepatic microsomal metabolite formed by a-hydroxylation of the carcinogen N-nitrosopyrrolidine. Chem. Res. Toxicol. 1, 28-31. (9) Chung, F.-L., and Hecht, S. S. (1983) Formation of cyclic 1,N2-adductsby reaction of deoxyguanosine with a-acetoxy-Nnitrosopyrrolidine, 4-(carbethoxynitrosamino)butanal,and crotonaldehyde. Cancer Res. 43, 1230-1235.

Wang et al. (10) Chung, F.-L., Wang, M., and Hecht, S. S. (1989) Detection of exocyclic N7,C8 guanine adducts in hydrolysates of hepatic DNA of rats treated with N-nitrosopyrolidine and in calf thymus DNA reacted with a-acetoxy-N-nitrosopyrrolidine.Cancer Res. 49, 2034-2041. (11) Hunt, E. J., and Shank, R. C. (1982) Evidence for DNA adducts in rat liver after administration of N-nitrosopyrrolidine. Biochem. Biophys. Res. Commun. 104,1343-1348, (12) Chung, F.-L., Young, R., and Hecht, S. S. (1989) Detection of cyclic 1,N2-propanodeoxyguanosineadducts in DNA of rats treated with N-nitrosopyrrolidine and mice treated with crotonaldehyde. Carcinogenesis 10, 1291-1297. (13) Saavedra, J. E. (1979) Oxidation of nitrosamines. 1. Formation of N-nitrosoimminiumions through the oxidative decarboxylation of N-nitrosoproline, N-nitrosopipecolic acid, and N-nitrososarcosine. J. Org. Chem. 44, 4511-4516. (14) Blomquist, A. T., and Buck, C. J. (1959) Many-membered carbon rings. XVII. Synthesis and acyloin cyclization of b,b-diphenylazelaic ester. J. Am. Chem. SOC.81, 672-676. (15) Bohlmann, F., Sucrow, W., Jastrow, H., and Koch, H. J. (1961) Polyacetylene compounds. XXXII. Further polyynes from Centaurea ruthenica. Chem. Ber. 94, 3179-3188. (16) Loftfield, R. B. (1951) y-Chlorobutyraldehyde and its diethyl acetal. J. Am. Chem. SOC.73, 1365-1366. (17) Kruger, F. W. (1972) New aspects in metabolism of carcinogenic nitrosamines. In Topics in Chemical Carcinogenesis (Nakahara, W., Takayama, S., Sugimura, T., and Odashima, S., Eds.) pp 213-232, University of Tokyo Press, Tokyo. (18) Singer, B., and Grunberger, D. (1983) Molecular Biology of Mutagens and Carcinogens, p 304, Plenum Press, New York. (19) Margison, G. P., and OConnor, P. J. (1979) Nucleic Acid Modification by N-Nitroso Compounds. In Chemical Carcinogens and DNA (Grover, P. L., Ed.) Vol. 1, pp 111-159, Boca Raton, FL. (20) Scherer, E., Van Der Laken, C. J., Gwinner, L. M., Laib, R. J., and Emmelot, P. (1981) Modification of deoxyguanosine by chloroethylene oxide. Carcinogenesis 2, 671-677.