Identification of the Major Protein Adduct Formed in Rat Liver After Thioacetamide Administration1

[CANCER RESEARCH 41. 3430-3435, September 1981] 0008-5472/81 70041-0000$02.00 Identification of the Major Protein Adduct Formed in Rat Liver after Thioacetamide Administration1

Martin C. Dyroff2 and Robert A. Neal3

Center in Environmental Toxicology, Department of Biochemistry. Vanderbilt University School of Medicine. Nashville, Tennessee 37232

ABSTRACT that metabolism of thioacetamide-S-oxide by rat liver microsomes results in the formation of a reactive metabolite(s) which The in vivo covalent binding of the hepatocarcinogen thioa- binds to calf thymus DNA, calf liver RNA, polyguanylate, and cetamide to rat liver protein has been examined. Following administration of 3H- or '"C-labeled thioacetamide, the modified polyadenylate (23). In the present investigation, we have de termined the structure of the major amino acid adduct of liver amino acids present in the hepatic cytosolic proteins were proteins formed on administration of [3H]- or [1-'"C]thioaceta- isolated by enzymatic digestion and ion-exchange chromatog- mide to rats. raphy. Approximately 70% of the radioactivity covalently bound to cytosolic protein was recovered in a compound which upon MATERIALS AND METHODS acid hydrolysis yielded lysine and radiolabeled acetate. Addi tional studies indicated the structure of this adduct was N-e- Chemicals. [3H]Thioacetamide (specific activity, 25 mCi/mmol) and acetyllysine. [1-'4C]thioacetamide (specific activity, 1 mCi/mmol) were synthesized from [3H]acetonitrile (3.75 Ci/mmol) and [1-'4C]acetonitrile (10 mCi/ mmol) as described previously (17, 25). The [3H]acetonitrile was the INTRODUCTION product of New England Nuclear, Boston, Mass., and the [1-'"C]ace- Thioacetamide produces centrilobular hepatic necrosis (1, tonitrile was obtained from ICN Chemical Radioisotopes Div., Irvine, 8), bile duct proliferation (4, 6), and liver cirrhosis (4, 7) in rats. Calif. The purity of the radiolabeled thioacetamide was determined by thin-layer chromatography (17) to be greater than 98%. [D3]Thioacet- Thioacetamide is rapidly metabolized in vivo in the rat to sulfate (15), acetamide (20), and thioacetamide-S-oxide (2,16). Acet- amide was synthesized using [D3]acetonitrile (Pfaltz and Bauer, Inc., Stamford, Conn.) as described previously (25). PhÃ©nobarbital, 3- amide does not produce hepatotoxic effects such as those methylcholanthrene, pronase, leucine amino peptidase, and amino acid observed with thioacetamide even at very high doses (3). standards were obtained from Sigma Chemical Co., St. Louis, Mo.; W- However, an equimolar dose of thioacetamide-S-oxide pro e-acetyllysine was obtained from Vega-Fox Biochemicals, Tucson, duces a more severe necrosis and at an earlier time after Ariz.; dimethylformamide dimethylacetal and 3% OV-17 on Chromo- administration than does thioacetamide (8, 16). Following ad sorb W(HP) were from Pierce Chemical Co., Rockford, III.; Porapak Q ministration of [3H]- (16, 20) or [1-14C]thioacetamide (16), co was from Alltech Associates, Arlington Heights, III.; Sephacryl S-200 valent binding of radioactivity to cellular macromolecules can was obtained from Pharmacia Fine Chemicals, Inc., Piscataway, N. J.; be detected. After administration of [35S]thioacetamide or Dowex 50W-X8 was from Bio-Rad Laboratories, Richmond, Calif.; and [35S]thioacetamide-S-oxide to rats, only trace amounts of co the 4.6-mm x 25-cm Zorbax octadecyl silane column was the product of DuPont, Wilmington, Del. Omnifluor was obtained from New England valently bound radioactivity (16) can be detected. However, Nuclear, and Nuclear-Chicago solubilizer and ACS were from Amer- Raw and Rockwell (19) have reported a significantly higher sham/Searle Corp., Arlington Heights, III. labeling of the protein in the nucleolus as compared to other Animals. Male Sprague-Dawley rats (HarÃanIndustries) (150 to 160 subcellular fractions in rabbits 2 hr after administration of g) were fed a commercial diet of Purina rat chow ad libitum. The [35S]thioacetamide. cytochrome P-450 monooxygenase activity of rat liver was induced by Rees ef al. (20) proposed that the incorporation of radioac i.p. injection of phÃ©nobarbital (80 mg/kg) in 1.15% KCI at 72, 48, and tivity from [3H]thioacetamide into macromolecules was the re 24 hr or 3-methylcholanthrene (40 mg/kg) in corn oil 48 hr prior to administration of [3H]thioacetamide (2.67 mmol/kg; 2 Â¿iCi/g body sult of conversion of the metabolite acetamide to acetate which weight), [1-14C]thioacetamide (2.67 mmol/kg; 0.33 Â¿iCi/g body entered the metabolic pool and became incorporated into weight), or [D3]thioacetamide (2.67 mmol/kg). Following thioacetamide endogenous molecules. However, Porter ef al. (16) have dem administration, the rats were fasted but allowed unlimited access to onstrated that administration of a dose of radiolabeled aceta water until sacrifice 12 hr later. mide, estimated to approximate the levels formed in vivo after Subcellular Fractionation. Rats were stunned by a blow to the head thioacetamide administration, does not result in significant and decapitated. The livers were perfused in situ with ice-cold 1.15% radiolabeling of hepatic macromolecules. KCI prior to excision and immersion in ice-cold 1.15% KCI. A portion Recent results have suggested that the reactive metabolite of each liver was removed for determination of its content of radioac of thioacetamide which becomes covalently bound is a further tivity by combustion in a Packard Tri-Carb tissue oxidizer. Subcellular oxidation product of thioacetamide-S-oxide, perhaps thioacet- fractionation of the remainder of each liver was performed as described amide-S-dioxide (8, 17), formed in a reaction catalyzed by previously (5). Purified nuclei were prepared by the hypertonic sucrose amine oxidase (18, 26), the cytochrome P-450-containing method of Spelsberg ef al. (21). All subcellular fractions obtained by centrifugation were resuspended in the homogenization buffer to a final monooxygenases (8, 16), or both. It has also been reported concentration of 2 g wet weight liver per ml. Aliquots of each subcellular 1This work was supported by Grants ES 00075 and ES 00267. fraction were analyzed for protein by the method of Lowry ef al. (11 ), 2Recipient of Grant ES 07028. using bovine serum albumin as the standard, and for radioactivity, by 3To whom requests for reprints should be addressed. liquid scintillation counting after solubilization with Nuclear-Chicago Received March 20, 1981; accepted June 11,1981. tissue solubilizer and neutralization with glacial acetic acid.

3430 CANCER RESEARCH VOL. 41

Determination of Covalent Binding. Protein and other macromole- Amino Acid. The radioactive compound recovered from the amino acid cules present in aliquots of the subcellular fractions corresponding to analyzer and purified from buffer components was derivatized using 1 g, wet weight, of liver were precipitated by adding 20 volumes of dimethylformamide dimethylacetal as described by Thenot and Horning cold 95% ethanol. The precipitated material was homogenized in the (22). The derivatized compound was chromatographed on a 2-mm x 6- 20 volumes of ethanol using a Son/all Omnimixer at top speed for 30 ft column of 3% OV-1 7 on Chromosorb W(HP). The Chromatographie sec and then centrifuged for 10 min at 20,000 x g. This cycle was conditions were: flame ionization detector, 275Â°; injector, 250Â°; and repeated until the radioactivity in the supernatant was less than 2 times column, 200Â° for the first 10 min and then temperature programmed background (5). The radioactivity associated with the remaining pre 15Â°/min to a final temperature of 250Â°. The carrier gas was helium cipitated material was considered to be covalently bound. This material maintained at a flow rate of 20 ml/min. Alternatively, the 3H-labeled was solubilized by heating for 1 hr at 60Â°in N NaOH and was assayed adduct was initially esterified with methanol:N HCI at 100Â° for 15 min. for protein and radioactivity. After removal of the solvent, the resulting mixture was reacted with Aliquots of the cytosolic fraction were chromatographed on a 1.5- x trifluoroacetic anhydride in mÃ©thylÃ¨nechloride at room temperature for 50-cm column of Sephacryl S-200. The column was eluted with 0.05 15 min. The solvent was removed under a stream of nitrogen, and the M sodium phosphate buffer, pH 7.4, and the effluent was monitored for sample was dissolved in methanol and analyzed on a 2-mm x 6-ft UV absorbance at 280 nm and for radioactivity. Aliquots of the hepatic column of 3% OV-17 in the gas Chromatograph. The Chromatographie cytosolic fraction were also analyzed by sodium dodecyl sulfate:poly- conditions were: flame ionization or electron capture detector, 260Â°; acrylamide gel electrophoresis on 7.5% gels according to the method injector, 210Â°; and column, 140Â° initially and then programmed 10Â°/ of Laemmli ef al. (9). The gels were stained with Coomassie Brilliant min to a final temperature of 250Â°. The carrier gas was nitrogen and Blue and scanned for absorbance at 595 nm. Subsequently, the gels maintained at a flow rate of 20 ml/min. Radioactivity was monitored by were sliced in 2-mm pieces, the slices were solubilized using 30% bubbling the column effluent through 18 ml of ACS scintillation cocktail H2C>2and 70% HCICX,according to the method of Mahin and Lofberg for timed intervals and counting the timed fractions. (12), and analyzed for radioactivity. A Finnigan Model 3200 GC-MS, operated in the electron impact Enzymatic Hydrolysis of Hepatic Cytosolic Protein. Cytosolic frac mode, was used to obtain the mass spectrum of the thioacetamide- tions from the livers of rats administered [3H> or [1-14C]thioacetamide modified amino acid which was derivatized using the procedure de were dialyzed for 24 hr at 4Â°against 0.05 M phosphate buffer, pH 7.4, scribed by Thenot and Horning (22). The operating conditions of the containing 1.15% KCI until the radioactivity in the dialysate was at gas Chromatograph were as described above. The analysis was per background levels. The cytosolic protein was subsequently digested formed at an electron energy of 70 eV, an ionizing current of 85 jtia, with pronase (10:1) (w/w) for 24 hr at 37Â°and then with microsomal and an ion source temperature of 80Â°. leucine amino peptidase (100:1 ) (w/w) for an additional 12 hr. Prior to The purified thioacetamide-modified amino acid was also subject to ion exchange chromatography, undigested protein was removed by acid hydrolysis (6 N HCI) in a vacuum for 24 hr at 110Â°, and the addition of 10% sulfosalicylic acid followed by centrifugaron. hydrolysate was chromatographed on a 2-mm x 6-ft column of Porapak Isolation of the Major Thioacetamide-modified Amino Acids. The Q. The Chromatographie conditions were: flame ionization detector, enzymatic digest of cytosolic protein was chromatographed on a 0.9- 180Â°; injector, 160Â°; column, 150Â° initially and then temperature x 133-cm water-jacketed column packed with Dowex 50W-X8 in the programmed at 5Â°/min to a final temperature of 170Â°. The carrier gas pyridinium salt form. The buffers used were 0.2 N pyridine formate, pH was helium maintained at a flow rate of 20 ml/min. Elution of radioac 3.25, for the first 110 min and 0.2 N pyridine acetate, pH 4.25, for the tivity was monitored as described above. next 55 min, followed by 1.2 N pyridine acetate, pH 6.40, for 200 min. Statistical tests used were multivariant analysis of variance and The buffer flow rate was 60 ml/hr, and the column temperature was Duncan's new multiple-range test (10). 50Â°. Effluent fractions were collected and analyzed for radioactivity and for ninhydrin-positive material (14). The fractions representing the RESULTS major peak of radioactivity were pooled and lyophilized. The residue was resuspended in distilled water and lyophilized to remove the last Covalent Binding of [3HJThioacetamide. Twelve hr after traces of pyridinium salts. The residue was dissolved in distilled water, administration, approximately 2.9 Â±0.3%, 3.7 Â±0.3%, and and the pH was adjusted to 2, analysis being on a Beckman Model 5.6 Â± 0.6% of the radioactivity from i.p. administered [3H]- 121 amino acid analyzer. Further purification of the modified amino thioacetamide was found in the livers of control, phÃ©nobarbi acids was accomplished by collecting effluent fractions from the amino tal-, and 3-methylcholanthrene-pretreated animals, respec acid analyzer column, determining radioactivity, and pooling the radio active fractions. The pooled sample containing radioactivity was puri tively. On a g wet weight basis, the values were 0.45 Â±0.04%, fied of buffer components by a combination of reverse-phase high- 0.50 Â±0.04%, and 0.69 Â±0.01%, respectively. Expressed in performance liquid chromatography and ion-exchange chromatogra this way, there is no significant difference between the values phy. Thus, the sample containing the radioactivity was chromato obtained in the control and the phenobarbital-pretreated ani graphed on a 4.6- x 250-mm column of Zorbax octadecyl silane eluted mals. However, a significant difference exists between the with distilled water at a flow rate of 60 ml/hr. The column fractions Table 1 containing radioactivity were pooled, lyophilized to dryness, and re- In vivo covalent binding of thioacetamide to macromolecules of rat liver suspended in a minimum volume of distilled water. The pH of this subcellular fractions solution was adjusted to 2, and the sample was applied to a 0.5- x 2- cm column of Dowex 50W-X8. The column was washed with 10-column Radioactivity bound to macromolecules (nmol/mg pro tein) following animal pretreatment volumes of N acetic acid, and the radioactivity was eluted with 0.2 N pyridine acetate, pH 5.5. Fractions were collected, and those contain 3-Methylcholan- ing radioactivity were pooled and lyophilized. All procedures used in Subcellular fraction Untreated Phenobarbital threne Â±0.5s the isolation of the major amino acid adduct from the liver cytosolic NuclearMitochoMdriul Â±0.4 Â±0.3 7.7 Â±1.2 6.0 Â±1.1 6.4 Â±0.6 protein of rats treated with radiolabeled thioacetamide were followed 10.4 Â±0.66 8.7 Â±1.7* Microsomal 8.1 Â±0.6 exactly for the isolation of the adduct formed after administration of Cytosolic6.0 8.8 Â±0.56.7 6.8 Â±0.57.0 6.8 Â±0.4 [D3]thioacetamide. Gas Chromatography and GC-MS4 of the Thioacetamide-modified Mean Â±S.D. of individual determinations for 3 animals in each treatment group. Significantly different at p < 0.05 when compared to the other values within 4 The abbreviation used is: GC-MS, gas chromatography-mass spectrometry. an individual treatment.

SEPTEMBER 1981 3431

Downloaded from cancerres.aacrjournals.org on October 3, 2021. © 1981 American Association for Cancer Research. M. C. Dyroff and R. A. Neal value obtained in the 3-methylcholanthrene-pretreated animals mide-modified amino acids from cytosolic protein by acid hy and the other 2 groups ( p < 0.01 ). The subcellular distribution drolysis yielded as the sole product a volatile, radioactive of covalently bound radioactivity is presented in Table 1. Within compound which eluted in the void volume of the Dowex 50W- the untreated and 3-methylcholanthrene pretreatment groups, X8 column. This volatile compound was identified by gas there is a significantly higher binding of radioactivity to the chromatography as acetate. microsomal macromolecules as compared to the nuclear, mi- Enzymatic digestion of the dialyzed cytosol followed by chro tochondrial, or cytosolic macromolecules (p < 0.05). matography on a column of Dowex 50W-X8 revealed the The covalent binding of radioactivity from [3H]thioacetamide presence of one major and a number of minor radioactive to cytosolic protein as a function of molecular weight was also peaks (Chart 1). Seventy % of the radioactivity eluted from the examined. Both gel exclusion chromatography and sodium column as a single peak (76 to 94 ml) in the region of the acidic dodecyl sulfate:polyacrylamide gel electrophoresis indicate the amino acids. An aliquot of this radioactive peak was analyzed radioactivity from [3H]thioacetamide appears to be more or less on a Beckman Model 121 amino acid analyzer (Chart 2). All of uniformly bound to proteins of different molecular weights (data the radioactivity applied to the column was recovered in a not shown). symmetrical peak which eluted between proline and glycine. Isolation of the Major Thioacetamide-modified Amino Next, 20 nmol of the thioacetamide-modified amino acid Acid. Because of the relative abundance and greater ease of were subject to acid hydrolysis (24 hr in 6 N HCI at 110Â°). An manipulation relative to membranous subcellular fractions, cy aliquot was analyzed by gas chromatography and the remain tosolic protein was examined for the identity of the amino der, on the amino acid analyzer. The amino acid analysis acid(s) to which the radioactivity from [3H]- and [1-l4C]thioa- indicated complete loss of the peak which corresponded to the cetamide was bound. Initial attempts to isolate the thioaceta- radioactive adduct and the appearance of a peak correspond-

so IOO 200 250 300 VOLUME (ml) Chart 1. Ion-exchange chromatography, on a 0.9- x 133-cm column of Dowex 50W-X8. of an enzymatic digestion of dialyzed cytosol isolated from the liver of a rat (untreated) administered [1-'4C]thioacetamide. See "Materials and Methods" for experimental details. Two-mi fractions were collected and analyzed for radioactivity (â€¢)and for ninhydrin-positive material { ).

400-

200-

60 TO TIME (minutes) Chart 2. Ion-exchange chromatography, on a Beckman 121 amino acid analyzer, of the components of the major radioactive peak (76 to 94 ml) isolated from the Dowex 50W-X8 ion exchange column (Chart 1). One aliquot of the pooled fractions representing this peak was applied to the amino acid analyzer, and the effluent was monitored for ninhydrin-positive material. In a second experiment, an identical aliquot was applied to the analyzer, and fractions representing 1 min of elution were collected and monitored for radioactivity. The standardization of the timed fraction collection to the automated ninhydrin analysis was established by injection of a reference amino acid of known retention time (aspartic acid, 28 min), collection of 1-min timed fractions, and examination of these fractions for ninhydrin-positive material. The radioactivity profile was then adjusted for the difference in retention times observed for automated analysis and manual fraction collection.

3432 CANCER RESEARCH VOL. 41

of standard /V-e-acetyllysine were mixed and applied to the Â® amino acid analyzer. Effluent fractions were monitored for radioactivity and for ninhydrin-positive material (Chart 5). All of the radioactivity was found to coelute with the ninhydrin-posi tive fractions representing N-e-acetyllysine. Analysis of Adduct by GC-MS. The results to this point suggested the structure of the major thioacetamide:amino acid adduct was A/-â‚¬-acetyllysine.Verification of this tentative struc ture was sought by gas chromatography and GC-MS. The amino acid derivatization technique described by Thenot and Horning (22) was used to produce a product amenable to gas chromatography. Chart 6 shows the results of gas chromatog raphy of the 14C-labeled adduct isolated as described in Chart 2 and standard A/-â‚¬-acetyllysine which had been derivatized 70 80 90 130 TIME (minutes) Chart 3. Ion-exchange chromatography on a Beckman 121 amino acid ana lyzer of 10 nmol of the isolated and purified 3H-labeled adduct (Chart 2) (A) and .700 7 nmol of the 3H-labeled adduct after acid hydrolysis (24 hr, 6 N MCI) (B). Prior to application to the amino acid analyzer, the acid hydrolysate of the adduct was evaporated to dryness on a rotary evaporator and resuspended in the amino acid .600 analyzer sample application buffer (pH 2.2). No radioactivity was detectable in this sample. 500 250

400 200

.300 I50

.200 IOO i

IOO 50

IO 20 30 40 TIME (minutes)

Charts. Cochromatography on a Beckman 121 amino acid analyzer of a mixture of 100 nmol of W-f-acetyllysine:5 nmol of the 14C-labeled adduct formed in vivo and isolated as described in Charts 1 and 2. One-min fractions were collected and monitored for radioactivity (â€¢)and for ninhydrin-positive material ( ). (No adjustment has been made in this chart for the difference in retention times observed for automated ninhydrin analysis and manual fraction collection with subsequent ninhydrin analysis as noted in Chart 2.)

TIME (minutes) Chart 4. Gas Chromatograph of the acid hydrolysate of the purified 3H-labeled adduct and of standard acetic acid on a column of Porapak Q. The acid hydrolysate was directly injected on the column, and radioactivity was monitored by bubbling the column effluent through 18 ml ACS scintillation cocktail for timed intervals. The standard of acetic acid was injected separately and monitored by flame ionization detection (FID).

ing to lysine (Chart 3). Another unsymmetrical peak appeared Chart 6. Gas Chromatograph of the dimethylformamlde dimethylacetal-deri- at Fractions 75 to 79. This is thought to be a spurious peak vatized N-t-acetyllysine and "C-labeled adduct. The derivatized ""C-labeled unrelated to the adduct. Gas chromatography (Chart 4) indi adduct was injected onto the OV-17 column, and timed fractions were collected cated that the radioactive product released on acid hydrolysis for scintillation counting by bubbling the column effluent as it came from the flame ionization detector through a mixture of ethanolamine:methoxyethanol: of the adduct was acetate. Omnifluor (3:6:10). A sample of the derivatized W-e-acetyllysine was injected Five nmol of the radioactive adduct (estimated by the specific separately, and the column effluent was monitored by flame ionization detection activity of the administered [1 -14C]thioacetamide) and 100 nmol (FID).

SEPTEMBER 1981 3433

Downloaded from cancerres.aacrjournals.org on October 3, 2021. © 1981 American Association for Cancer Research. M. C. Dyroll and R. A. Neal posed largely of peptides. Acid hydrolysis (6 N HCI, 110Â°, 24

NCH3 - CH = hr) of an identical aliquot also indicated this peak of radioactiv CH2CH-0- CH - -NH-C-CHj196,K>*157-CH2 -CH2 ity is composed primarily of peptide material. Attempts at C847341iÃ,-CH2 digesting the peptide material to its component amino acids 1100;11 using a variety of proteases were unsuccessful.

50:i !r DISCUSSION !XNi40 r TiÃ¯' 183\ M* kL200 \257JJ Previous results in vivo have indicated the liver necrosis from 50 100 150: 250 3 m/e thioacetamide is the result of microsomal metabolism of thioa Chart 7. The mass spectrum of the dimethylformamide dimethylacetal-deri- cetamide and thioacetamide-S-oxide to a toxic metabolite(s) vatized "C-labeled adduct. The mass spectrum is consistent with N-a-(N',N'- (8). It is reasonable to expect that binding of a toxic metabolite dimethylaminomethylene)-N-i-acetyllysine methyl ester (structure shown). The of thioacetamide and thioacetamide-S-oxide to various mole relative intensity scale has been expanded 10 times above m/e 170. cules in the liver may be responsible, directly or indirectly, for the hepatic necrosis seen on exposure to these compounds. and analyzed separately. As can be seen, the radioactivity has The hepatic microsomal fraction generally exhibited slightly the same retention time as the product formed upon derivati- higher levels of covalent binding of radioactivity following ad zation of authentic W-e-acetyllysine. The derivatized 14C-labeled ministration of [1-14C]- or [3H]thioacetamide than did other adduct was also analyzed by GC-MS. The mass spectrum of subcellular fractions. However, the fact that the other subcel the isolated and derivatized radioadduct (Chart 7) was consist lular fractions were labeled to nearly the same extent supports ent with that of A/-a-{A/',A/'-dimethylaminomethylene)-A/-â‚¬-ace- the hypothesis that the reactive form(s) of thioacetamide is a tyllysine methyl ester. Both the 3H- and D3-labeled adducts compound with a sufficiently long half-life to allow diffusion formed in the liver cytosolic protein of rats administered 3H- throughout all parts of the cell. and D3-labeled thioacetamide were also derivatized and sub Of the radioactivity which is covalently bound to cytosolic jected to spectral analysis by gas chromatography and GC- protein, approximately 70% has been identified as an adduct MS. The gas Chromatographie retention times, as determined formed with lysine residues, specifically /V-e-acetyllysine. Ad by flame ionization detection, were the same as for the 14C- ditionally, although the recovery of radioactivity from the ion labeled adduct and standard A/-â‚¬-acetyllysine. However, the exchange column of the enzymatic digestion of the microsomal retention time of the 3H on gas chromatography was different fraction isolated from rat livers was low (67%), 66% of the than for /V-t-acetyllysine and the deuterium of A/-â‚¬-fD3-ace- recovered radioactivity was identified as W-e-acetyllysine, con fy/]acetyllysine was not identifiable as the M3+ ion upon GC- sistent with the value obtained for the cytosolic fraction. MS. When authentic A/-e-[D3-acefy/]acetyllysine was derivatized The adduct is apparently formed in a reaction between a with dimethylformamide dimethylacetal and analyzed by GC- reactive electrophilic metabolite of thioacetamide and the e- MS, the M3+ peak was not identifiable in the mass spectrum. amino group of lysine residues. Another candidate amino acid A possible reason for the results seen with the 3H- and D3- for reaction with the reactive electrophilic metabolite would be labeled adducts was exchange of the methyl hydrogens of the cysteine. However, the structure of the proposed adduct be thioacetamide portion of the adduct under the strong acid tween a cysteine residue and the reactive metabolite would conditions and high temperature of the derivatization proce probably be a thioimidate or thioester. Both of these com dures. Subsequently, a new procedure was developed which pounds would be much less stable than an amide and would achieved derivatization of the adduct under milder conditions. probably hydrolyze under physiological conditions. The e- Reaction of the isolated amino acid adduct with methanolic- amino groups of lysine residues are predominantly located on HCI and trifluoroacetic anhydride was found to produce a the surface of the proteins (13). Thus, their availability plus the derivative amenable to gas chromatography. Using this pro stability of the product may explain the predominance of A/-e- cedure, it was found that the radioactivity of the derivatized 3H- acetyllysine as the major adduct of the reactive metabolite(s) labeled adduct has the same retention time on gas chromatog of thioacetamide with cytosolic protein. raphy as the derivatized standard A/-â‚¬-acetyllysine. A reaction between a primary amino group, such as the e- The structure of the major thioacetamide-derivatized amino amino group of lysine, and the previously hypothesized reactive acid of the proteins of the microsomal fraction of rat liver was intermediate of thioacetamide, namely, thioacetamide-S-diox- also examined using the techniques utilized for the analysis ide (8, 17), might also be anticipated to result in the formation of the major amino acid adduct of the cytosolic proteins. of an acetamidine derivative. A/-e-Acetamidinolysine can be Although the recovery of radioactivity from the enzymatic formed by reacting proteins with ethyl acetimidate (24). When digestion of the microsomal fraction was low (67%) compared bovine serum albumin which had been reacted with ethyl [3H]- to the results obtained with the cytosolic fraction (>90%), 66% acetimidate was subject to acid hydrolysis (6 N HCI for 24 hr) of the recovered radioactivity was identified as A/-e-acetyllysine or to enzymatic digestion under identical conditions as those using gas chromatography and GC-MS. The poor recovery of used in the isolation of the major thioacetamide adduct formed radioactivity from this particular fraction was probably the result in vivo, the only labeled product isolated was A/-â‚¬-acetamidi- of incomplete digestion of the protein. nolysine. Neither labeled acetate nor A/-e-acetyllysine was de An aliquot of the second major peak (196 to 202 ml) re tected. Thus, it appears that the reaction of the reactive metab covered from the Dowex 50W-X8 ion exchange column (Chart olite of thioacetamide with the e-amino group of lysine does not 1) was also analyzed for its amino acid composition on the proceed via an amidine-type intermediate and that the mech amino acid analyzer. The results indicated this peak was com anism of adduct formation is not consistent with nucleophilic

3434 CANCER RESEARCH VOL. 41

Downloaded from cancerres.aacrjournals.org on October 3, 2021. © 1981 American Association for Cancer Research. Hepatic Protein Adduci of Thioacetamide attack of the e-amino group on the carbonyl carbon of thioa- 9. Laemmli, U. K. Cleavage of structural proteins during the assembly of the cetamide-S-dioxide with the loss of SO22~ (sulfoxylate aniÃ³n). head of bacteriophage T4 Nature (Lond.), 227: 680-685, 1970 10. Li, J. C. R. Statistical Inference I. Ann Arbor. Mich.: Edwards Brothers, Inc., There is a suggestive temporal relationship between the 1964. observation of hepatic necrosis and the detection of covalent 11. Lowry. O. H., Rosebrough. N. J., Farr, A. L.. and Randall, R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193: 265-275, binding of radioactivity to hepatic protein following administra 1951. tion of radiolabeled thioacetamide (16). However, whether the 12. Mahin, D. T., and Lofberg, R. T. A simplified method of sample preparation covalent binding of a thioacetamide metabolite(s) to hepatocyte for determination of tritium, carbon-14. or sulfur-35 in blood or tissue by liquid scintillation counting. Anal. Biochem., 16: 500-509. 1966. macromolecules is responsible for the observed hepatic necro 13. Matthews, B. W. X-ray structure of Proteins. In: H. Neurath, R. L. Hill, and sis is unknown at this time. Furthermore, what role, if any, C. Boeder (eds.), The Proteins; Ed. 3, pp. 403-590. New York: Academic covalent binding to protein and other macromolecules [includ Press, Inc., 1977. 14. Moore, S., Spackman, D. H., and Stein. W. H. Chromatography of amino ing DMA (23)] has in the development of the liver tumors acids on sulfonated polystyrene resins. Anal. Chem., 30:1185-1190, 1958. observed after chronic administration of thioacetamide is also 15. Nygaard, O., Eldjarn, L., and Nakken, K. F. Studies on the metabolism of thioacetamide-S35 in the intact rat. Cancer Res., 14: 625-628, 1954. unknown. 16. Porter. W. R., Gudzinowicz, M. J., and Neal, R. A. Thioacetamide-induced hepatic necrosis. II. Pharmacokinetics of thioacetamide and thioacetamide- ACKNOWLEDGMENTS S-oxide in the rat. J. Pharmacol. Exp. Ther.. 208: 386-391, 1979. 17. Porter, W. R.. and Neal, R. A. Metabolism of thioacetamide and thioacetamide-S-oxide by rat liver microsomes. Drug Metab. Dispos.. 6: 379-388, We thank Donna Smith for performing the amino acid analyses and Tracy Wright for her assistance with the GC-MS analyses. 1978. 18. Poulsen, L. L., and Ziegler, D. M. Microsomal mixed-function oxidase- dependent renaturation of reduced ribonuclease. Arch. Biochem. Biophys., REFERENCES 183: 563-570, 1977. 19. Raw, O., and Rockwell, P. A protein-bound form of thioacetamide in liver 1. Ambrose, A. M., DeEds. F.. and Rather, L. F. Toxicity of thioacetamide in nucleoli. Biochem. Biophys. Res. Commun., 90: 721-725, 1979. rats. J. Ind. Hyg. Toxicol.. 31: 158-161, 1949. 20. Rees, K. R., Rowland, G. F., and Varcoe, J. S. The metabolism of tritiated 2. Ammon. V. R., Berninger, H.. Haas, H. J., and Landsberg, L. Thioacetamid- thioacetamide in the rat. Int. J. Cancer. 1: 197-206, 1966. sulfoxid, ein Stoffwechselprodukt des Thioacetamids. Arzneim Forsch., i 7: 21. Spelsberg, T. C., Knowles, J. T., and Moses, H. L. Specific methods for the 521-523, 1967. isolation of nuclei from chick oviduct. Methods Enzymol., 31: 263-279, 3. Dessau, F. I., and Jackson, B. Acetamide induced liver cell alterations in 1974. rats. Lab. Invest., 4: 387-397, 1955. 22. Thenot, J. P., and Horning, E. C. Amino acid N-dimethylaminomethylene 4. Fitzhugh, O. G., and Nelson, A. A. Liver tumors in rats fed thiourea or alkyl esters. New derivatives for GC and GC-MS studies. Anal. Lett.. 5: 519- thioacetamide. Science (Wash. D. C.), 708. 626-628, 1948. 529, 1972. 5. Guengerich. F. P. Studies on the activation of a model furan compoundâ€” 23. Vadi, H. V., and Neal, R. A. Microsomal activation of thioacetamide-S-oxide toxicity and covalent binding of 2-(N-ethyl-carbamoylhydroxymethyl)furan. to a metabolite(s) that covalently binds to calf thymus DNA and other Biochem. Pharmacol., 26: 1909-1915, 1977. polynucleotides. Chem.-Biol. Interact., 35: 25-38. 1981. 6. Gupta, D. N. Production of cancer of the bile ducts with thioacetamide. 24. Wofsy, L.. and Singer, S. J. Effects of the amidination reaction on antibody Nature (Lond.), ) 75: 257, 1955. activity and on the physical properties of some proteins. Biochemistry, 2: 7. Gupta, D. N. Nodular cirrhosis and metastasing tumours produced in the 104-116, 1963. liver of rats by prolonged feeding with thioacetamide. J. Pathol. Bacteriol., 25. Walter, W., and Kubersky, H. P. Oxidation products of thiocarboxamides. 72:415-426. 1956. XI. Infrared spectroscopic studies of thioacetamide-S-oxide. Ann. Chem. 8. Hunter, A. L., Holscher, M. A., and Neal, R. A. Thioacetamide-induced (Justus Liebigs), 694: 70-77, 1966. hepatic necrosis. I. Involvement of the mixed-function oxidase enzyme 26. Ziegler, D. M.. and Poulsen, L. L. Hepatic mixed-function amine oxidase. system. J. Pharmacol. Exp. Ther.. 200: 439-448, 1977. Methods Enzymol., 52: 142-151, 1978.

SEPTEMBER 1981 3435

Downloaded from cancerres.aacrjournals.org on October 3, 2021. © 1981 American Association for Cancer Research. Identification of the Major Protein Adduct Formed in Rat Liver after Thioacetamide Administration

Martin C. Dyroff and Robert A. Neal

Cancer Res 1981;41:3430-3435.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/41/9_Part_1/3430

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

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

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/41/9_Part_1/3430. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.