Nitrosation of Glycine Ethyl Ester and Ethyl ... - ACS Publications

Feb 26, 2004 - Terence A. Lawson,† and Sidney S. Mirvish*,†,‡. Eppley Institute for Research in Cancer, 6805 University of Nebraska Medical Cent...
1 downloads 0 Views 92KB Size
Download PDF
416

Chem. Res. Toxicol. 2004, 17, 416-423

Nitrosation of Glycine Ethyl Ester and Ethyl Diazoacetate To Give the Alkylating Agent and Mutagen Ethyl Chloro(hydroximino)acetate Lin Zhou,† James Haorah,† Sheng C. Chen,† Xiaojie Wang,† Carol Kolar,† Terence A. Lawson,† and Sidney S. Mirvish*,†,‡ Eppley Institute for Research in Cancer, 6805 University of Nebraska Medical Center, Omaha, Nebraska 68198, and Departments of Pharmaceutical Sciences and of Biochemistry and Molecular Biology, University of Nebraska Medical Center, Omaha, Nebraska 68198 Received October 10, 2003

Whereas nitrosation of secondary amines produces nitrosamines, amino acids with primary amino groups and glycine ethyl ester were reported to react with nitrite to give unidentified agents that alkylated 4-(p-nitrobenzyl)pyridine to produce purple dyes and be direct mutagens in the Ames test. We report here that treatment of glycine ethyl ester at 37 °C with excess nitrite acidified with HCl, followed by ether extraction, gave 30-40% yields of a product identified as ethyl chloro(hydroximino)acetate [ClC(dNOH)COOEt, ECHA] and a 9% yield of ethyl chloroacetate. The ECHA was identical to that synthesized by a known method from ethyl acetoacetate, strongly alkylated nitrobenzylpyridine, and may have arisen by Nnitrosation of glycine ethyl ester to give ethyl diazoacetate, which was C-nitrosated and reacted with chloride to give ECHA. Nitrosation of ethyl diazoacetate also yielded ECHA. Ethyl nitroacetate was not an intermediate as its nitrosation did not produce ECHA. ECHA reacted with aniline to give ethyl (hydroxamino)(phenylimino)acetate [PhNdC(NHOH)CO2Et]. This product was different from ethyl [(phenylamino)carbonyl]carbamate [PhNHC(dO)NHCO2Et], which was synthesized by reacting ethyl isocyanatoformate (OCN‚CO2Et) with aniline. ECHA reacted with guanosine to give a derivative, which may have been a guanine-C(dNOH)CO2Et derivative. ECHA showed moderate toxicity and weak but significant mutagenicity without activation in Salmonella typhimurium TA-100 (mean, 1.31 × control value for 12-18 µg/plats) and for V79 mammalian cells (1.5-1.7 × control value for 60-100 µM). In conclusion, gastric nitrosation of glycine derivatives such as peptides with a N-terminal glycine might produce ECHA analogues that alkylate bases of gastric mucosal DNA and thereby initiate gastric cancer.

Introduction Gastric nitrite derived from dietary nitrate and nitrite may be involved in the etiology of gastric, esophageal, nasopharyngeal, oral, colonic, and bladder cancer (1, 2). Salted, dried fish and meat products containing nitrate and nitrite may be a cause of gastric cancer (1, 3). Prospective studies indicated that nitrite-preserved meat is a cause of colon cancer in adults (4) and of brain cancer and leukemia in children (5, 6). Such associations may be due to the acid-catalyzed gastric formation of nitrosamines and nitrosamides (collectively called N-nitroso compounds) formed by the nitrosation of secondary amines and N-alkylamides, because N-nitroso compounds can alkylate DNA bases and thereby cause somatic mutations leading to cancer (1, 7). Nitrosation of primary amines to yield agents that could alkylate DNA has not been intensively investigated (but see ref 8), although nitrosation of amino acids and peptides with primary amino groups has been studied since 1886 (9-11). Some of the products were substituted diazoacetic acids (NdNdCR‚COOH) (12). * To whom correspondence should be addressed. E-mail: smirvish@ unmc.edu. † Eppley Institute for Research in Cancer. ‡ Departments of Pharmaceutical Sciences and of Biochemistry and Molecular Biology.

Nitrosation of glycocholic acid and reaction of the product with DNA produced O6-carboxymethylguanine and O6-methylguanine, attributed to the formation of diazoacetate (NdNdCHCOO-) and its reaction with DNA guanine (13, 14). This may be a source of the O6-methylguanine that occurs in human gastric and colonic DNA (13, 15). In 1991, Meier et al. (16) reported that glycine, GEE1 (1; Scheme 1), and aspartame react with acidified nitrite to produce agents that alkylate NBP (2) to give purple dyes (16, 17). These dyes arise from products alkylated at the pyridine nitrogen, and their production indicates the presence of alkylating agents (Scheme 2) (18, 19). The products of these nitrosations were direct mutagens in the Ames test (20-22). Such reactions could be significant because the stomach contains free amino acids (23) and peptides, which could be nitrosated under the acidic conditions in this organ. We describe here a study of the agents that are formed by the nitrosation of GEE and alkylate NBP. We identified ECHA (3; Scheme 1) as the chief such agent and studied some of its chemical and toxicological properties. Preliminary reports of some of these studies have been presented (24, 25). 1 Abbreviations: DCTP, 3,4-dichlorothiophenol; DMEM, Dulbecco’s modified Eagle’s medium; ECHA, ethyl chloro(hydroxyimino)acetate; GEE, glycine ethyl ester; NBP, 4-(p-nitrobenzyl)-pyridine.

10.1021/tx0300481 CCC: $27.50 © 2004 American Chemical Society Published on Web 02/26/2004

Nitrosation of Glycine Ethyl Ester

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 417

Scheme 1. Suggested Mechanism for Nitrosation of GEE (1) To Give ECHA (3) and Ethyl Chloroacetate (6)

Scheme 2. NBP Test: Production of Purple Dyes on Alkylation of NBP (2)

Scheme 3. Synthesis of ECHA (3) from Ethyl Acetoacetate (4)

Experimental Procedures General Procedures. 1H and 13C NMR spectra were obtained using solutions in CDC13 with tetramethylsilane as internal standard and were mostly measured with a Varian XL-500 instrument (Palo Alto, CA). Mass spectra were measured using high-resolution electron ionization and low-resolution fast atom bombardment techniques at the facilities listed in the Acknowledgments. All high-resolution MS peaks agreed with the calculated masses within 3 millimass units. Infrared spectra were measured with a Perkin-Elmer model 1420 instrument (Norwalk, CT). Unless mentioned otherwise, chemicals were obtained from Aldrich (Milwaukee, WI) and procedures were carried out at room temperature. Solutions were adjusted to the desired pH with HCl or NaOH. Solutions in organic solvents were dried over Na2SO4 before further workup and were evaporated in a rotary evaporator at 20% cell survival. The cells were rinsed with serum-free DMEM and cultured as log-phase growth in DMEM containing 7.5% serum. On day 10, 200 cells were plated in each of three 100 mm dishes. After another 7 days, the colonies were stained and plating efficiencies were counted. Also on day 10, 3 × 105 cells were plated in each of eight dishes and incubated with 45 µM 6-thioguanine (Sigma Chemicals, St. Louis, MO). After another 10 days, 6-thioguanine resistant colonies were stained and counted. Results were expressed as number of mutants/106 cells. Statistics. The significance of differences between groups in the mutagenesis tests was analyzed by the Wilcoxon rank order test.

Results Synthetic Studies. GEE was nitrosated with NaNO2/ HCl at 37 °C. When the reaction mixture was treated with sulfamic acid and then extracted with ether, the extracts contained a relatively stable product that crystallized readily when the extract was concentrated. Recrystallization produced needles of mp 76-78 °C. Elemental analysis indicated the formula C4H6ClNO3. The 1H NMR spectrum indicated that the ethoxy group

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 419

of GEE was still present and that the hydrogen atom that was not part of the ethyl group was attached to oxygen or nitrogen. The 13C NMR spectrum confirmed the presence of a carbethoxy (COOEt) group and showed a peak at 158.6 ppm suggesting a carbonyl type group. The IR spectrum included two carbonyl type groups (at 1745 and 1725 cm-1). The mass spectrum confirmed the molecular formula and showed pairs of fragments due to 35Cl and 37Cl, as well as fragments with loss of OH, ethyl, ethoxy, and COOEt groups. This evidence indicated that the product was ECHA. To confirm the structure of the GEE product, we repeated a 1902 synthesis of ECHA by Jovitchitch (35, 36). Ethyl acetoacetate (4; Scheme 3) was nitrosated at the central carbon to give a 92% yield of ethyl 2,3diketobutyrate-2-oxime 5. Treatment with nitric acid led to replacement of the acetyl group with a nitro group to give ethyl nitrohydrximinoacetate (a nitrolic acid, 15), which reacted with HCl to give ECHA in a 22% yield from oxime 5. This sample of ECHA showed the same mp of 76-78 °C as GEE-derived ECHA. The mixed mp of the two products was also 76-78 °C. Identity with the GEEderived product was confirmed by the similar 1H and 13C NMR spectra of the two products. After GEE was treated with NaNO2/HCl and sulfamic acid as in the Experimental Procedures, Nitrosation of GEE To Give ECHA, a 1/500 dilution of the resulting solution in 0.01 N HCl showed A580 ) 0.66 in the NBP test, similar to the previous report (16). An NBP test on 20 µg of ECHA in pH 2 buffer showed A580 ) 0.88. On this basis, the NBP test after nitrosation of GEE suggested a possible yield of 75% for ECHA (but see later). The positive NBP test for the crude product was not due to ethyl diazoacetate (7), a known product of GEE nitrosation (37), because even 100 µL of 7 did not react positively in the NBP test. Nitrosation of 5 mmol of ethyl diazoacetate (7) followed by ether extraction produced 260 mg of solid. Recrystallization yielded 35 mg of pure ECHA. The mp, 13C NMR spectrum, and high-resolution fast atom bombardment MS were identical to those of GEE-derived ECHA. The yield of crude product corresponded to an ECHA yield of 34%. Ethyl chloroacetate (6) was identified as a minor product of GEE nitrosation by demonstrating that a GFEE nitrosation product that was extracted by ether showed a GC peak with the retention time of authentic 6, which showed a mass spectrum identical to that of authentic 6, with peaks at m/z 122 (M+), 94, and 77 (100%). From the area of the GC peak, its yield from GEE was estimated to be 9%. Compound 6 reacted weakly in the NBP test, with 93 µmol giving A580 ) 0.09 and hence was not the chief product of GEE nitrosation responsible for its positive NBP test. Under the conditions used for GEE nitrosation (Experimental Procedures, Nitrosation of GEE To Give ECHA), nitrosation of ethyl chloroacetate failed to yield ECHA, as shown by the absence of the 1H NMR peaks of ECHA in an ether extract of the product. When GEE was treated with NaNO2/HCl and the reaciton mixture was reacted with DCTP at pH 8, a product was identified as S-carbethoxymethyl-DCTP. Its yield from GEE was 2.8%. Ethyl chloroacetate reacted with DCTP to give the same S-carboxymethyl adduct as that obtained from the nitrosation of GEE, as indicated by the similar 1H NMR spectra.

420

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

To estimate the yield of ECHA from GEE, GEE was nitrosated as in the Experimental Procedures, Nitrosation of GEE To Give ECHA, but with D2O in place of H2O as the solvent. After sulfamic acid was added, we measured the 1H NMR spectra of the crude mixture and of the mixture after addition of GEE (1), ethyl chloroacetate (6), or ethyl glycolate (16). The added compounds helped to provisionally identify the products. Integration of the NMR peaks (measured without the additions) indicated that 30-40% of the GEE had been converted to ECHA (similar to the yield based on the weight of recrystallized product) and that 30-40% of the GEE had not reacted. This estimate of the GEE yield is probably more accurate than that obtained from the NBP test. This insensitive method did not detect ethyl chloroacetate (6), presumably because of its low yield, or ethyl glycolate (16), a likely product of the reaction (Scheme 1). We also studied the nitrosation of ethyl nitroacetate to indicate if it could be an intermediate in ECHA formation from GEE. The product of this reaction gave a strong positive response in the NBP test, with 1 µL of the aqueous reaction mixture giving an absorbance of 1.11. The 1H and 13C NMR spectra of the ether extract of this solution indicated the presence of two unidentified compounds containing ethoxy groups, neither of which was ECHA. Chemical Properties of ECHA. Some chemical properties of ECHA were studied because the NBP test indicated that ECHA is an alkylating agent and little had been reported previously about its properties. An NBP test on 13 µg of ECHA gave an absorbance of 0.63 when the standard 200 µL of triethylamine was added at the end of the NBP test. When an additional 200 or 400 µL of triethylamine was used, the absorbance rose to 0.70 and 0.71, respectively. The NBP test was used to determine the stability of ECHA solutions in aqueous buffer kept at 37 °C. We found complete decomposition of ECHA after incubation for 15 min at pH 4-9, a half-life of 30 min at pH 3, and no decomposition after 3 h at pH 2. ECHA (152 µmol) reacted with silver nitrate to give 149 µmol silver chloride, i.e., this treatment liberated 1 mol of chloride/mol of ECHA. Treatment of ECHA with alkali gave a product that was identified as oxalic acid (9) by GC-MS of its methyl ester. Thus, dimethyl oxalate and the methylated product from ECHA showed identical GC retention times of 8.5 min and similar mass spectra. The reaction of ECHA with aniline gave a product that was provisionally identified as ethyl (hydroximino)phenylaminoacetate (13) or its conjugated tautomer, ethyl (hydroxamino)phenyliminoacetate (mp, 106-108 °C) by its elemental analysis consistent with C10H12N2O3, by its 1H NMR, 13C NMR, and IR spectra, and by highresolution MS (Scheme 5). In a similar reaction, ethyl isocyanatoformate (12) reacted with aniline to give ethyl (phenylamino)carbonylcarbamate (ethyl N-phenylallophanate, 14), which showed mp 106-107 °C, identical to the reported mp (Scheme 5) (38). The 13C NMR spectrum of 14 differed markedly from that of 13, even though the mixed mp of 13 and 14 was undepressed at 106 °C. The 1H and 13C NMR spectra of 13 and 14 did not interconvert when these compounds were heated to 106 °C in melting point tubes, showing that the nondepression of the mixed mp was not due to conversion of 13 to 14 or vice versa during the heating. As another model for reactions with DNA, we studied the reaction of ECHA with guanosine at pH 6. This pH

Zhou et al. Table 1. Mutagenicity of ECHA in V79 Cellsa pH of medium

ECHA concn (µM)

7.4 7.4 5.2

60 100 100

mutants/106 cells (mean ( SD) control treated Pb 105 ( 14 105 ( 14 131 ( 22

173 ( 22 162 ( 18 195 ( 31

0.0009 0.0009 0.0019

a For eight dishes/group in all groups. b Significance of difference between treated and control groups.

was used rather than pH 7.4 to increase stability of the ECHA. Samples of the reaction product were hydrolyzed by heating at pH 1 or pH 6 under conditions known to liberate O6- and 7-alkylguanines, respectively, from DNA (39, 40) and were then subjected to HPLC with UV detection at 254 nm. HPLC of the pH 6 hydrolysate showed principal peaks at 16 and 37-39 min. The latter peak contained 12% of the total UV-absorbing material. Analysis by HPLC-electrospray ionization-MS showed that the 16 min fraction was unhydrolyzed guanosine, with prominent MS peaks at m/z 284 (guanosine + H) and 152 (guanine). Electrospray ionization-MS of the 36 min fraction showed the largest MS peak at m/z 267 consistent with C9H11N6O4, corresponding to an adduct of guanine with -C(dNOH)COOEt. The yield of this product was estimated to be ∼12% from guanosine if we assume that all products showed molar extinction coefficients at 260 nm similar to that of guanosine and that the HPLC peak contained only a single product. The second largest MS peak derived from the 36 min fraction occurred at m/z 152 and was attributed to guanine. Analysis of the pH 1 hydrolysate gave results almost identical to those for the pH 6 hydrolysate. Mutagenicity of ECHA. When ECHA was examined in the Ames test using S. typhimurium TA-100 without activation, it showed toxicity at 20 µg ECHA/plate (60 µM) in the top agar and slight but consistent direct mutagenicity. A combination of two experiments yielded six results for untreated controls with 154 ( 4 mutants/ plate and 10 results with 203 ( 7 mutants/plate for plates treated with 8-18 µg (53-120 nmol) ECHA/plate (A ( B ) mean ( SE). Even though the mean value for the treated plates was only 31% higher than the mean background level, these results were significantly different with P < 0.01. Higher concentrations of ECHA were toxic, with faint background lawns at 20 µg/plate and all bacteria killed by 64 µg/plate. In the same two experiments, 4 µg of methylnitronitrosoguanidine produced mean values of 1090 and 1670 mutants/plate. Addition of hamster liver S-9 fraction did not increase the activity. Because ECHA is more stable at acidic than at neutral pH values, we also tested the mutagenicity after ECHA solutions in citrate buffer at pH 5 were incubated with the bacteria for 30 min, brought to pH 7, and mixed with the top agar. Under these conditions, 5 µg of ECHA produced a mean mutation rate that was 22% above the background level and the bacteria were all killed by 20 µg ECHA/plate. We also tested ECHA by the V79 mammalian cell mutagenicity assay for mutations leading to resistance to 6-thioguanine. The test was carried out at physiological pH 7.4 and at pH 5.2, the lowest pH at which the cells remained viable. In a preliminary toxicity test, survival was >50% for ECHA concentrations up to 100 µM at both pH 5.2 and pH 7.4. In the mutagenicity assay (Table 1), 60 and 100 µM ECHA showed significantly increased numbers of mutants at pH 7.4 (1.65 and 1.54 × back-

Nitrosation of Glycine Ethyl Ester

ground, respectively), and 100 µM ECHA also showed significant mutagenicity at pH 5.2 (1.4 × background). Table 1 gives the P values.

Discussion The principal product of the nitrosation of GEE (1) under our conditions was ECHA (3) and not ethyl diazoacetate (7), a known product of this nitrosation (41). Ethyl chloroacetate (6) was a minor product formed in 9% yield. ECHA reacted strongly, 6 reacted weakly, and 7 did not react at all in the NBP test. It is surprising that ECHA had not been reported before as a product of GEE nitrosation. Nitrosation of GEE to give ethyl diazoacetate (7) was first reported by Curtius in 1888 (9). A standard synthesis of 7 involves nitrosation of GEE‚HCl with NaNO2/H2SO4 at -5 °C in the presence of dichloromethane, which extracts 7 as it is produced (41, 42). As far as we are aware, no one had reported the use in this synthesis of HCl instead of H2SO4 as the acid that generated nitrous acid from sodium nitrite. In the standard synthesis of 7, solvent extraction as it is formed, the low reaction temperature, and the limited concentration of chloride would minimize further aqueous nitrosation of 7 to give (after reaciton with chloride) ECHA. In contrast, we performed the synthesis of 3 using HCl as the counteracid and conducted the reaction at 37 °C in the absence of an organic solvent because these conditions were modeled on those in the human stomach. ECHA can also be prepared by (i) a synthesis involving nitrosation of ethyl acetoacetate (4) (27), the method used here to confirm the structure of our product, and (ii) by reacting ketene with NOCl and adding ethanol to give ethyl hydroximinoacetate (HONdCHCOOEt), followed by treatment with chlorine (43). ECHA has been used as a synthetic intermediate, e.g., in the preparation of isooxazolines (44). We recommend our synthesis from GEE as a facile one step method for preparing ECHA. Further studies would be needed to establish the mechanism of ECHA formation from GEE and ethyl diazoacetate (7). Meanwhile, we speculate that ECHA is produced as shown in Scheme 1. We suggest that nitrosation of GEE gives ethyl diazoacetate (7) (41, 42), which reacts further with nitrite to yield C-nitroso compound 18. The ready protonation of diazoacetate 7 at the R-carbon to give diazonium ion 17 suggests that nitrosation at this position can also occur. Compound 18 could then be attacked by chloride ion with elimination of nitrogen to give 19, which would tautomerize to give ECHA (3). This mechanism is supported by the finding that 7 also reacted with NaNO2/HCl to form ECHA. One would probably not have predicted that 7 would be sufficiently stable under acidic conditions to be nitrosated, but diazoacetates are relatively stable, at least in alkali (12). In a side reaction, 7, after protonation to give diazonium ion 17, could react with chloride ion to give ethyl chloroacetate (6). A similar reaction with water would yield ethyl glycolate (16), although this was not detected by an insensitive NMR method. ECHA could not arise from ethyl chloroacetate (6) because 6 did not react with nitrite to give ECHA. Because ECHA formation from GEE requires a central methylene group, similar products are not expected from esters of other amino acids, all of which have a methine group in the R-position. The production of ECHA from GEE might have occurred via the formation of ethyl nitroacetate (8) by a

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 421

reaction of GEE with nitric acid, produced by the decomposition of nitrous acid (45). Ethyl nitroacetate might then be nitrosated to yield nitrolic acid 15 (Scheme 3) (nitrolic acids are defined as R-nitronitroso compounds). This reaction seemed possible because aliphatic nitro compounds can be C-nitrosated to form nitrolic acids (46, 47). Nitrolic acid 15 could then react with chloride to give ECHA in a reaction involved in the synthesis of ECHA from ethyl acetoacetate (Scheme 3). However, ECHA was not detected as a product of the nitrosation of ethyl nitroacetate (8). Our results indicate that, to obtain a maximum response, the NBP test should be performed with 400 instead of the recommended (16) 200 µL of triethylamine. The greater stability of ECHA at pH e3 than at higher pH values agrees with the report (16) that the reaction mixture after nitrosation of GEE was more stable at pH 2.5 than at pH 7 [in both our study and that by Meier et al. (16), stability was measured by the NBP test]. We initially considered it likely that the GEE-derived agent that reacted with NBP would also react with the thiophenol DCTP (10) and that the structure of the S-alkyl DCTP adduct would be easy to determine (48) and would help us to identify the alkylating agent formed from nitrosated GEE. In fact, nitrosation of GEE and reaction of the product mixture with DCTP gave a 2.8% yield of Scarbethoxymethyl-DCTP (11) (Scheme 4). As ethyl chloroacetate (6) was produced in a 9% yield from GEE and also reacted with DCTP to give adduct 11, this adduct may have arisen from 6. To help understand how ECHA might alkylate DNA bases, we studied the reaction of ECHA with aniline. We obtained a product provisionally identified as 13 (Scheme 5). During its synthesis, this product might have undergone a Beckman type rearrangement to give 14 (Scheme 5). However, 14 synthesized from 12 was not identical to 13. These experiments suggested that the adduct between ECHA and a DNA base R1NHR2 would have the structure R1R2NC(dNOH)COOEt. In fact, ECHA reacted with guanosine to give an adduct of guanine with a mass spectrum consistent with addition of the -C(NdOH)COOEt group. However, because a large excess of ECHA was used in this reaction, we cannot exclude the possibility that the adduct was produced from a minor impurity in the ECHA. ECHA showed weak but significant direct mutagenicity for S. typhimurium TA-100 and for mammalian V79 cells (Table 1). The only previous publication on the toxicology of ECHA reports that ECHA caused allergic contact dermatitis (49). The weak mutagenic activity of ECHA may have been due to its toxicity (seen at 60 µM in the Ames test and 100 µM in the V79 test), which prevented higher concentrations from being tested. On the basis of the results in the Ames test of 5 µg of ECHA, which produced a mean of 50.5 mutants/plate more than the results in untreated controls, the activity was 1800 mutants/µmol ECHA. When Shephard et al. (20) measured the mutagenicity of nitrosated GEE in strain TA100, they reported results of 2000 mutants/mol GEE, similar to those found here for ECHA, but they did not list the concentrations tested or the observed number of mutants in this or related reports (20-22). Although GEE is not known to occur naturally, it may be a model for N-terminal glycine in peptides. This might be nitrosated in the acidic stomach to give the peptide

422

Chem. Res. Toxicol., Vol. 17, No. 3, 2004

derivatives ClC(dNOH)‚CONH‚CHR‚CO‚‚‚‚, which could react with DNA bases. When the GEE nitrosation mixture was analyzed by the NBP test after varied reaction times, absorbance increased at a constant rate for up to 80 s (16). This rate probably corresponded to that for ECHA formation from GEE, was proportional to nitrite concentration squared, and showed a third-order stoichiometric rate constant (k2 in Table 1 of ref 50) of 0.2 M-2 s-1 (16). This value was ∼5 times the value of k2 for the nitrosation of proline, which is converted to N-nitrosoproline in the stomach of humans who ingest nitrate (51). This kinetic finding supports the view that nitrosation of peptides with N-terminal glycine to give ECHA analogues could occur in the human stomach. The proportionality to [nitrite]2 indicates that the active nitrosating agent was N2O3 (50). Although two successive nitrosations are involved (Scheme 1), presumably only one of these reactions is rate limiting. Our study on ECHA stability suggests that ECHApeptide derivatives would be stable under the acidic conditions of the stomach and may persist long enough inside gastric mucosal cells (which are probably at pH ∼7.4) to react with DNA bases. Therefore, it may be useful to prepare peptide analogues of ECHA, to test the carcinogenicity of ECHA in rodents, and to search for adducts of human DNA analogous to aniline adduct 13. At least five types of agents that could form adducts with DNA bases are now known to be formed by the nitrosation of amino acids and their esters: (i) A β-lactone that alkylates NBP is produced from aspartic acid (52). (ii) Nitrosation in the presence of chloride may produce R-chloroacids that can (in the case of ethyl chloroacetate) alkylate the thiol group of DCTP but not the pyridine N of NBP (see Results). The R-chloroacid derived from methionine is mutagenic and is probably an alkylating agent due to formation of a thiouronium intermediate (53). (iii) Nitrosation of tryptophan yields mutagens that may be N-nitrosoindoles (54). (iv) Nitrosation of phenylalanine gave a 4% yield of a nitrolic acid, 1-nitro-1oximino-2-phenylethane [PhCH2C(dNOH)NO2] (55). (v) Nitrosation of GEE in the presence of chloride produces ECHA (Scheme 1). The mutagenicity of ECHA and its ability to alkylate aniline and, apparently, guanosine suggest that it might be carcinogenic. Overall, these findings indicate that nitrosation of amino acids might contribute to the etiology of cancers linked with the consumption of nitrate, nitrite, and nitrite-preserved meat and fish products (1, 4).

Acknowledgment. We thank B. Gold of this Institute for his advice, S. S. Hecht and S. G. Carmella (University of Minnesota Cancer Center, Minneapolis, MN) for the MS of the guanine adducts, J. Vennerstrom (College of Pharmacy, University of Nebraska Medical Center) for the GC-MS of ethyl chloroacetate, and R. Cerny and R. Shoemaker (Center for Mass Spectrometry and Chemistry Department, University of NebraskaLincoln) and I. Vidavsky (Washington University, St. Louis) for the remaining MS. This study was supported by Grants RO1-CA-RO1-71483 and P30-CA-367273 from the National Cancer Institute, Grant 94B28 from the American Institute for Cancer Research, Grant 2000-23 from the Nebraska Department of Health and Human Services, and (for MS) National Cancer Institute Grant CA-77598 (University of Minnesota Cancer Center), National Science Foundation Grant DIR-9017262 (Uni-

Zhou et al.

versity of Nebraska-Lincoln), and NIH Grant P41RR0954 (Washington University, St. Louis).

References (1) Mirvish, S. S. (1995) Role of N-nitroso compounds (NOC) and N-nitrosation in etiology of gastric, esophageal, nasopharyngeal and bladder cancer and contribution to cancer of known exposures to NOC. [Erratum: (1995) 97, 271]. Cancer Lett. 93, 17-48. (2) Badawi, A. F., Hosny, G., el-Hadary, M., and Mostafa, M. H. (1998) Salivary nitrate, nitrite and nitrate reductase activity in relation to risk of oral cancer in Egypt. Dis. Markers 14, 91-97. (3) Mirvish, S. S. (1983) The etiology of gastric cancer: Intragastric nitrosamide formation and other theories. J. Natl. Cancer Inst. 71, 629-647. (4) Sandhu, M. S., White, I. R., and McPherson, K. (2001) Systematic review of the prospective cohort studies on meat consumption and colorectal cancer risk: A meta-analytical approach. Cancer Epidemiol., Biomarkers Prev. 10, 439-446. (5) Bunin, G. R. (1998) Maternal diet during pregnancy and risk of brain tumors in children. Int. J. Cancer Suppl. 11, 23-25. (6) Peters, J. M., Preston-Martin, S., London, S. J., Bowman, J. D., Buckley, J. D., and Thomas, D. C. (1994) Processed meats and risk of childhood leukemia (California, U.S.A.). Cancer Causes Control 5, 195-202. (7) Lijinsky, W. (1992) Chemistry and Biology of N-Nitroso Compounds, Cambridge University Press, Cambridge. (8) Obiedzinsdi, M. W., Wishnok, J. S., and Tannenbaum, S. R. (1980) N-Nitroso compounds from reactions of nitrite with methylamine. Food Cosmet. Toxicol. 18, 585-589. (9) Curtius, T. (1888) Uber Diazoessigsaure und ihre Derivate. J. Prakt. Chem. 38, 396-440. (10) Curtius, T., and Darapsky, A. (1906) U ¨ ber Diazoacetylaminoessiga¨urea¨thylester. Ber. Dtsch. Chem. Ges. 39, 1373. (11) Curtius, T., and Thompson, J. (1906) Einwirkung von Ammoniak auf Diazoacetyl-glycinester und Diazoacetyl-glycylglycinester. Ber. Dtsch. Chem. Ges. 39, 1383. (12) Challis, B. C. (1989) Chemistry and biology of nitrosated peptides. Cancer Surv. 8, 363-384. (13) Harrison, K. L., Fairhurst, N., Challis, B. C., and Shuker, D. E. G. (1997) Synthesis, characterization, and immunochemical detection of O6-(carboxymethyl)-2′-deoxyguanosine: A DNA adduct formed by nitrosated glycine derivatives. Chem. Res. Toxicol. 10, 652-659. (14) Shuker, D. E. G., and Margison, G. P. (1997) Nitrosated glycine derivatives as a potential source of O6-methylguanine in DNA. Cancer Res. 57, 366-369. (15) EUROGAST Study Group (1994) O6-Methylguanine in blood leucocyte DNA: an association with the geographic prevalence of gastric cancer and with low levels of serum pepsinogen A, a marker of severe chronic atrophic gastritis. Carcinogenesis 15, 1815-1820. (16) Meier, I., Shephard, S. E., and Lutz, W. K. (1990) Nitrosation of aspartic acid, aspartame, and glycine ethyl ester. Alkylation of 4-(p-nitrobenzyl)pyridine (NBP) in vitro and binding to DNA in the rat. Mutat. Res. 238, 193-201. (17) Lutz, W. K., Meier, I., and Shephard, S. E. (1990) Alkylation of 4-(p-nitrobenzyl) pyridine in vitro by nitrosated primary amines and amino acids, and investigation of DNA-binding in rat stomach after endogenous nitrosation of methylamine, tyramine, glycine ethyl ester, and aniline. Drug Dev. Eval. 16, 163-178. (18) Epstein, J., Rosenthal, R. W., and Ess, R. J. (1955) Use of gamma(4-nitrobenzyl) pyridine as analytical reagent for ethyleneimines and alkylating agents. Anal. Chem. 27, 1435-1439. (19) Archer, M. C., and Eng, V. W. S. (1981) Quantitative determination of carcinogens and mutagens as alkylating agents following chemical activation. Chem.-Biol. Interact. 33, 207-214. (20) Shephard, S. E., Hegi, M. E., and Lutz, W. K. (1987) In Vitro Assays to Detect Alkylating and Mutagenic Activities of Dietary Components Nitrosated In Situ. IARC Sci. Publ. 84, 232-236. (21) Shephard, S. E., Meier, I., and Lutz, W. K. (1991) Alkylating Potency of Nitrosated Amino Acids and Peptides. IARC Sci. Publ. 105, 383-387. (22) Shephard, S. E., Wakabayashi, K., and Nagao, M. (1993) Mutagenic activity of peptides and the artificial sweetener aspartame after nitrosation. Food Chem. Toxicol. 31, 323-329. (23) Komorowska, M., Szafran, H., Popiela, T., and Szafran, Z. (1981) Free amino acids of human gastric juice. Acta Physiol. Pol. 32, 559-567. (24) Zhou, L., and Mirvish, S. S. (2000) Identity of alkylating agent formed by nitrosation of glycine ethyl ester. Proc. Am. Assoc. Cancer Res. 41, 839.

Nitrosation of Glycine Ethyl Ester (25) Mirvish, S. S., and Zhou, L. (2000) Identity of alkylating agent formed by nitrosation of glycine ethyl ester. Proceedings of the Division of Chemical Toxicology, American Chemical Society 219th National Meeting, TOXI-112, American Chemical Society, Washington, DC. (26) Jovitschitsch, M. (1895) Darstellung des isonitroso-acetessigesters und eines isomeren aus acetessigester. Chem. Ber. 28, 2683-2687. (27) Jovitschitsch, M. (1902) Zu den synthesen der acetessigesterderivate. Chem. Ber. 35, 151-157. (28) Gomori, G. (1955) Preparation of buffers for use in enzyme studies. Methods. Enzymol. 1, 138-146. (29) Vogel, A. I. (1961) A Textbook of Quantitative Inorganic Analysis, Longmans, Green and Co., London. (30) Maron, D. M., and Ames, B. N. (1983) Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173-215. (31) Van Zeeland, A. A., and Simons, J. W. I. M. (1976) Linear doseresponse relationship after prolonged expression times in V79 Chinese hamster cells. Mutat. Res. 35, 129-138. (32) Hansen, M. B., Nielsen, S. E., and Berg, K. (198) Reexamination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J. Immunol. Methods 119, 203210. (33) Kolar, C., and Lawson, T. (1997) Mutagenicity of carcinogenic nitrosamines when activated by hamster and human pancreatic duct epithelial cells. Cancer Lett. 117, 149-154. (34) Hanssen, M. B., Nielsen, S. E., and Berg, K. (1989) Reexamination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J. Immunol. Methods 119, 203210. (35) Michael, F., and Klemer, H. (1952) Synthesis of N-glycosyl amino acids. Chem. Ber. 85, 1083-1086. (36) Heyns, K., and Noack, H. (1963) Die Umsetzung von L-tryptophan und L-histidin mit hexosen. Chem. Ber. 97, 415-418. (37) Fieser, L. F., and Fieser, M. (1967) Organic Syntheses, Wiley & Sons, New York. (38) Al Sabbagh, M. M., Calmon, M., and Calmon, J. P. (1984) Nouvelles syntheses de phenyl-4 allophanaates. Bull. Soc. Chim. Fr. Part 2 3-4, 73-77. (39) Lawson, T. A. (1990) Activation of N-nitrosobis(2-oxopropyl)amine by liver and nasal mucosa tissue from intact and castrated male rats. Cancer Lett. 53, 39-43. (40) Augusto, O., Cavalieri, E. L., Rogan, E. G., RamaKrishna, N. V. S., and Kolar, C. (1990) Formation of 8-methylguanine as a result of DNA alkylation by methyl radicals generated during horseradish peroxidase-catalyzed oxidation of methylhydrazine. J. Biol. Chem. 265, 22093-22096. (41) Fieser, L., and Fieser, M. (1967) Reagents for Organic Synthesis, pp 367-370, J. Wiley, New York.

Chem. Res. Toxicol., Vol. 17, No. 3, 2004 423 (42) Searle, N. E. (1963) Ethyl diazoacetate. Organic Syntheses, Collect. Vol. 4, pp 424-426, Wiley & Sons, New York. (43) Rave, T. W. (1972) Process of preparing aliphatic monofunctional oximes. U.S. Patent No. 3,686,280. (44) Pitts, W. J., Wityak, J., Smallheer, J. M., Tobin, A. E., Jetter, J. W., Buynitsky, J. S., Harlow, P. P., Solomon, K. A., Corjay, M. H., Mousa, S. A., Wexler, R. R., and Jadhav, P. K. (2000) Isoxazolines as potent antagoinsts of the integrin Rvβ3. J. Med. Chem. 43, 27-40. (45) Turney, T. A. and Wright, G. A. (1959) Nitrous acid and nitrosation Chem. Rev. 59, 497-513. (46) Matt, C., Wagner, A., and Mioskowski, C. (1997) Novel transformation of primary nitroalkanes and primary alkyl bromides to the corresponding carboxylic acids. J. Org. Chem. 62, 234235. (47) Matt, C., Gissot, A., Wagner, A., and Mioskowski, C. (2000) Nitrolic acids: efficient precursors of nitrile oxides under neutral conditions. Tetrahedron Lett. 41, 1191-1194. (48) Preussmann, R., Arjungi, K. N., and Ebers, G. (1976) In vitro detection of nitrosamines and other indirect alkylating agents by reaction with 3,4-dichlorothiophenol in the presence of rat liver microsomes. Cancer Res. 36, 2459-2462. (49) Hausen, B. M. (1992) Occupational allergic contact dermatitis from ethyl chloro oximido acetate. Contact Dermatitis 27, 277278. (50) Mirvish, S. S. (1975) Formation of N-nitroso compounds: Chemistry, kinetics, and in vivo occurrence. Toxicol. Appl. Pharmacol. 31, 325-351. (51) Ohshima, H., and Bartsch, H. (1981) Quantitative estimation of endogenous nitrosation in humans by monitoring N-nitrosoproline excreted in the urine. Cancer Res. 41, 3658-3662. (52) Garcia-Santos, M. D. P., Calle, E., and Casado, J. (2001) Amino acid nitrosation products as alkylating agents. J. Am Chem. Soc. 123, 7506-7510. (53) Chen, W., Weisburger, J. H., Fiala, E. S., Spratt, T. E., Carmella, S. G., Chen, D., and Hecht, S. S. (1996) Gastric carcinogenesis: 2-chloro-4-methylthiobutanoic acid, a novel mutagen in salted, pickled Sanma hiraki fish, or similarly treated methionine. Chem. Res. Toxicol. 9, 58-66. (54) Brown, N. K., Nguyen, T. T., Taghizadeh, K., Wishnok, J. S., and Tannenbaum, S. R. (1992) Mutagenic decomposition products of nitrosated 4-chloroindoles. Chem. Res. Toxicol. 5, 797-801. (55) Loeppky, R. N., Teuten, E. L., Tang, E., and Power, N. (2003) Nitrolic Acids: Possible Endogenous Toxicants from the Nitrosation of Amino Acids, American Chemical Society Division of Chemical Toxicology Abstracts of the 226th National Meeting, American Chemical Society, Washington, DC.

TX0300481