Protein Modifications by Activated Carcinogens

The postulated fragmentation of RNase by N-acetoxy-2fluorenylacetamide (N-acetoxy-2-FAA), an activated metabolite of the carcinogen, N-hydroxy-24iuorenylacetamide, has been investigated. The interaction of N-acetoxy-Z-FAA with RNase resulted in the formation of two new proteins separable from RNase by electrophoresis and ion exchange chromatography. These proteins contained no new NH2terminal amino acids, and their amino acid composition was identical with that of native RNase, indicating that the protein had not been modified by cleavage of the peptide chain. The alternative possibility that the formation of the additional proteins was due to a decrease in the positive charge of RNase was examined by the determination of the incorporation of 3H and W from N-[3H]acetoxy-[9-14C]2-FAA into RNase. The ratio of bound 3H to bound 14C, which measures the relative extent of acetylation and arylamidation in the modified proteins, suggested that acetylation rather than arylamidation was the predominant reaction modifying RNase. The conclusion that the acetyl group of N-[3H]acetoxy-[Q-14C]2-FAA had been transferred to RNase was confirmed (a) by the recovery of [3H]acetic acid from the hydrolysates of the modified protein and (b) by the isolation of e-N-acetyl-L-lysine from enzymatic hydrolysates of the modified protein. The isolation of the N-acetylated amino acid accounted in part for the decrease in the positive charge of RNase after reaction with N-acetoxy-2-FAA. The results indicate that N-acetoxy3-FAA is an acetyl donor and modifies proteins primarily by acetylation rather than arylamidation.

The interaction of N-acetoxy-Z-FAA with RNase resulted in the formation of two new proteins separable from RNase by electrophoresis and ion exchange chromatography.
These proteins contained no new NH2terminal amino acids, and their amino acid composition was identical with that of native RNase, indicating that the protein had not been modified by cleavage of the peptide chain. The alternative possibility that the formation of the additional proteins was due to a decrease in the positive charge of RNase was examined by the determination of the incorporation of 3H and W from N-[3H]acetoxy-[9-14C]2-FAA into RNase.
The ratio of bound 3H to bound 14C, which measures the relative extent of acetylation and arylamidation in the modified proteins, suggested that acetylation rather than arylamidation was the predominant reaction modifying RNase.
The conclusion that the acetyl group of N-[3H]acetoxy-[Q-14C]2-FAA had been transferred to RNase was confirmed (a) by the recovery of [3H]acetic acid from the hydrolysates of the modified protein and (b) by the isolation of e-N-acetyl-L-lysine from enzymatic hydrolysates of the modified protein.
The isolation of the N-acetylated amino acid accounted in part for the decrease in the positive charge of RNase after reaction with N-acetoxy-2-FAA.
The results indicate that N-acetoxy3-FAA is an acetyl donor and modifies proteins primarily by acetylation rather than arylamidation.
The biological activity of certain hepatoca.rcinogenic arylamides appears to depend on a two-st'ep mechanism of metabolic activation.
The arylamide is first N-hydroxylated (2) and then esterified to an N-sulfate (3), and possibly also to an N-phosphate (4) or N-acetate (5). These esters are highly unstable and decompose spontaneously to an electrophilic arylamidonium * This work was supported by United States Public Health Service Grant CA02571 from the National Cancer Institute.
A preliminary account has been published (I).
ion which arylamidates DNA, RNA, and proteins (6,7). Although these macromolecular interactions have been documented in detail, the critical reaction that initiates the malignant transformation of the cell is not known.
In a previous study, we have described the arylamidation of nuclear proteins, including histones, of rat liver by N-2-fluorenylacetamide in vivo (8). These data prompted us to suggest that arylamidation might alter the histones in such a way that they were no longer able to inhibit the transcription of nuclear DNA (9). However, the arylamidation of the nuclear proteins under these conditions was of such a low order that the elucidation of the structural modifications of the nuclear proteins by presently available analytical techniques was not feasible. We have reinvestigated the problem of the structural modification of proteins by arylamidation in a model system in which we reacted N-acetoxy-2fluorenylacetamide (N-acetoxy-2-FAA) with RNase. This model system was chosen because it had been reported that N-acetoxy-2-FBA, an ester that is well characterized and available in pure form, reacts through arylamidation with methionine residues of RNase (10). Our objective was to determine whether this would cleave the peptide chain of RNase as had been suggested (7, ll), and, if so, whether histories that had been modified by arylamidation were fragmented in a similar manner. Although we were able to demonstrate arylamidation of RNase by N-acetoxy-2-FAA, to a minor extent, no cleavage of the protein was observed.
Further examination of the modified protein showed that acetylation, rather than arylamidation, appeared to be the predominant reaction between N-acetoxy-2-FAA and RNase and that it accounted, at least in part, for the altered properties of RNase that had been exposed to the ester. The experiments leading to these conclusions form the basis of this report. with N-acetoxy-2-FAA (10). e-N-Acetyl-L-lysine (m.p. 247-249") was prepared by the acetylation of the copper complex of L-lysine (13). ar-N-Acetyln-lysine (m.p. 245-249") was obtained from e-N-benzyloxycarbonyl-cY-N-acetyl-L-lysine (13,14). The infrared spectrum of the product was identical with that of an authentic sample. 0-Acetyl-L-serine was prepared by the acetylation of n-serine (2.0 g, 19 mmoles,Calbiochem) in glacial acetic acid saturated with HCl (15). The acetylation of the amino acid was carried out in suspension, instead of in solution as reported (15)  The characteristic infrared absorption band at 3450 cm-r (-OH-) present in n-serine was absent in O-acetyl-n-serine.  The compound rearranged in 0.5 N NHIOH to N-acetyl-nnthreonine, m.p. 130" (16). The identity of the compound as the 0-acetylated derivative of t,hreonine was confirmed by the synthesis of N-acetyl-L-threonine We have consistently obtained prod-Jr1.5". c = 2Y0 in HzO. which thev obtained bv the rearrangement ucts with infrared spectra superimposable on that of our ana-of O-acetyl-n-threonine in 0.5 N NH,OH and regarded as N&etyllytical sample that gave the specific rotation listed in the text.
Moreover, in our experience, the been reported (17).
0 --f N acetyl shift in 0.5 N NH,OH results in racemization. Both compounds were prepared by the acetylation, with acetic anhydride, of L-or n-threonine in alkali in a manner similar to that described for the N-acetylation of DL-or L-serine (18,19). N-Acetyl-n-and N-acetyl-n-threonine were hydrolyzed with 4 N HCl to n-threonine, 10~12~ = -27.8", and n-threonine, [a]: = +27.8", respectively.
The conditions for the hydrolysis, isolation, and purification of the amino acids were those described for the preparation of O-methyl-n-or L-threonine from O-methyl-N-chloroacetyl-n-or L-threonine (20). Conditions for iUodi$ication of RNase by N-Acetoxy-SFAA-Ribonuclease A (XII-A, low phosphate, Sigma) was purified either by gel filtration on Sephades G-25 or by ion exchange chromatography on Bio-Gel CM30 (Bio-Rad) ( Table I). The eluted protein was reacted with N-acetoxy2FAA in the media listed in Table I. The ester dissolved in acetone (20 ml) was added to RNase (80 mg) in Tris-HCl buffer (pH 7.4) (20 ml). In one experiment, the ester was dissolved in 0.5 ml of ethanol and added at room temperature to 20 ml of the magnetically stirred protein solution (Experiment 8, Table I). The reactions were carried out at 22' for 24 hr. The modified proteins were precipitated by cooling the solution in ice and adding 2 to 3 volumes of cold acetone.
The precipitate was collected by centrifugation and redissolved in a minimum amount of water.
After a second acetone precipitation, the modified proteins were collected and examined by disc electrophoresis or ion eschange chromatography.
Disc Electrophoresis-The methods for the electrophoretic resolution of the modified proteins were modifications of the techniques of Reisfeld et al. (21) and of Shepherd and Gurley (22). Solutions A to F were used to prepare the stacking and the resolving gels: Solution A-40Y0 Prep/Cry1 (Canalco) and 0.4Yo ethylene diacrglamate; Solution B4.75 ml of glacial acetic acid and 1.15 ml of TEMED in 400 ml of water titrated to pH 4.0 with KOH and diluted to 500 ml with water; Solution C-0.56% ammonium persulfate; Solution D-14.0Yo Prep/Cry1 and 0.8% ethylene diacrylamate; Solution E-4.75 ml of glacial acetic acid and 2.3 ml of TEMED in 400 ml of water titrated to pH 6.0 with KOH and diluted to 500 ml with water; Solution F-saturated aqueous riboflavin.
The resolving gel (16.6 ml) consisted of 7.0 g of urea dissolved in Solutions A, B, and C (6.0, 2.0, and 3.4 ml, respectively).
The solution was degassed and cast in silanized Pyrex tubes (6.0 mm, inner diameter) to a depth of 6.0 to 8.0 cm. After polymerization (2 hours), excess persulfate was removed by electrophoresis of the gels in 0.02 M potassium acetate buffer (pH 4.0) at 1 ma per tube for 2 hours. The stacking gel (16.6 ml) consisted of 7.0 g of urea dissolved in Solutions D, E, and F (6.0, 2.0, and 3.4 ml, respectively).
Polymerization was carried out for 0. carried out at 1.5 ma per tube for 1.5 hours. The proteins were stained with Amido Black (0.25cc) in 7.57! acetic acid and destained electrophoretically in the same acid. The absorbance profile of the prot.eins was measured with the use of a microdensitometer (Canalco, model F).
Identification and Measurement of oJfethylthio-Z-FAA-The amounts of the o-methylmercaptoamide released spontaneously from the protein during the react'ion of RKase with N-acetoxy-[9-%]2-FAA were measured by inverse isotope dilution (23). Carrier 3-CH3-S-2-FAA (0.4 g) was added t.o the reaction mixture, and the proteins were precipitated with acetone and removed by centrifugation.
The supernatant liquid was concentrated at reduced pressure, and the residue was extracted with a mixture of benzene-n-hexane (15:85). The solvent was evaporated at reduced pressure, and the o-CH3-S-[9-14C]2-FAA was purified by thin layer chromat,ographp.
Enzymatic ilctivity and dmino Acid Composition of Modified RNase-The enzymatic activity of the modified protein was assayed by the method of Roth with ribonucleic acid (type XI, Sigma) as the substrate (24). The hydrolysis of RNA by ribonuclease A (type XII-A, Sigma) served as a control.
The amino acid analyses of t,he proteins were carried out on duplicate samples with a Beckman/Spine0 amino acid analyzer model 120B (25, 26). The proteins were hydrolyzed in sealed tubes with 6 N HCl (3 mg of protein per ml) at 110" for 22 hours.
The NHz-terminal amino acids of the modified proteins were identified by reacting the proteins with FDNB (27). The proteins were hydrolyzed in 6 K HCI (2 ml) at 110" for 18 hours. The DNP amino acids were extracted with ether and chromatographed on Silica Gel GF264 with four different solvents (28).
Estimation of [3H]Acetyl Groups Released from Modi$ed Proteins by Alkaline or Acid Hydrolysis-The proteins that had been modified by the reaction of RNase with N-[3H]acetoxy-2-FAA were hydrolyzed in alkali in order to release the labeled acetyl group that was bound to the proteins in ester linkage.
The protein solution (0.5 ml) was added to 2 N KaOH (5 ml), and the mixture stood at 22" for 2 to 18 hours (Table III).
Sodium acetate (4 mmoles) was added, and the pH was adjusted to 3.0 with H3P04. The solution was steam-distilled, the distillate (50 ml) was titrated with 0.1 N NaOH, and its radioactivity was measured. The amounts of [3H]acetyl groups released from the protein were calculated from the specific radioactivity of the distilled [3H]acetic acid (29).
The amounts of [3H]acetyl groups bound to the RNase in amide linkage were determined by acid hydrolysis of the modified proteins at 100" for 6 hours. The protein solution (0.5 ml) was added to 8.5 ml of 6 s HzS04, and the solution was heated in stoppered tubes in a boiling water bath. After cooling, 4.0 mmoles of sodium acetate were added, the mixture was subjected to steam distillation, and the specific radioactivity of the [3H]acetic acid in the distillate was measured.
Isolation and Characterization of Acetylated Amino Acids-The modified proteins from the reaction of RNase with N-[3H]acetoxy-2-FAA were hydrolyzed with Pronase B (Calbiochem) and trypsin (30,31).
The hydrolysate was chromatographed on a column (1 x 7 cm) of AG 5OW-X4 (Bio-Rad) with 0.2 M ammonium acetate (pH 5.5) as the eluenO (32). The compounds eluted with the buffer were subjected to thin layer chromatography on Silica Gel GF254 with ethanol-water (63 : 37) as solvent (28). The acetylated amino acids were located by a radioscan of the chromatogram and eluted with water.
The eluate was lyophilized, the residue was dissolved in citrate buffer, and [14C]aspartic acid, [14C]glycine, and [%]leucine were added as markers.
The acet-ylated amino acids were then resolved by chromatography on a Beckman-Spinco 120B amino acid analyzer. The effluent was split into two equal portions.
One of these was used for amino acid analysis.
The other was analyzed for 3H and 'AC' by liquid scintillation spectrometry. The [3H]amino acid in the elut.ion profile was tentatively identified by its position relative to the positions of the markers.
Estimation of e-N-Acetyl-L-lysine-The amount of E-N-[~H]acetyl-L-lysine in the enzymatic hydrolysates of the modified proteins was measured by inverse isotope dilution (39). e-A-Acetyl-L-lysine (30 mg) was added to a portion of the radioactive eluate that had been obtained by chromatography of the hydrolysates on AG 5OW-X4.
The solution was brought to :I boil, and basic cupric carbonate was added.
The copper comples was collected and decomposed with H2S. The e-N-[3H]acetyl-L-lysine was isolated by thin layer chromatography on silica gel and purified to constant specific radioactivity by successive thin layer chromatographies with three different solvents (Table IV).
Protein and Radioactivity Measurements-Protein contents were estimated by the modified Folin method with RNase A (type XII, Sigma) as the standard (33).
All samples were counted in duplicate and corrected for quenching by means of an external standard.

RESULTS
Modi$cation of RNase by N-Acetoxy-d-FL-L--The init'ial reactions of RNase with N-acetoxy-2-FAA were carried out in media consisting of 0.1 M Tris-HCl (pH 7.4)) 7 JI urea, and 50yc acetone. These conditions were selected so as to unfold the RNase and to expose the amino acid residues of the protein to t'he ester (34). Furthermore, this concentration of acetone insured the solubility of the ester as well as that of the protein.
S-Acetosy-2-FAA and RNase are soluble in media with an acet,one content between lS and 65y0. The molar ratio of ester to protein in these experiments was 8.0. At the completion of the incubation the RNase was precipitated by excess acetone, and the precipitate was found to contain 86y0 of the initial protein.
The precipitated protein was resolved by disc electrophoresis into three components, designated A, B, and C (Fig. I), whereas native RKase moved largely as a single component.
The acetone-precipitable, heterogeneous RNase is subsequently referred to as modified RNase. Protein A had the same mobility as unreacted RKase and accounted for 45% of the absorbance of the electrophoretic profile. Prot.eins B and C were less basic than protein A and accounted for 42 and 13%, respectively, of the total stained protein.
The elect,rophoretic profiles of equal amounts of untreated RNase and of modified RNase (Fig. 1) indicated that proteins B and C were formed at the expense of native RNase.
The possibility that proteins B and C were products of the cleavage of RNase was investigated in four different esperiments. First, we compared the hydrolysis of RNA by nat'ire and by modified RNase. These measurements showed that, modified RNase had retained 96% of the enzymatic act'ivity of native RNase.
Second, we analyzed the supernatant' liquid of the deproteinized reaction mixture for the presence of o-CH&[9J4C]2-FAA.
If the mercaptoamide were found, it would indicate that N-acetoxy-2-FAA had reacted with methionyl residues of RSase and that the resulting sulfonium ion had decomposed spontane- ously, thereby releasing the mercaptoamide and cleaving the peptide bond at, the carboxyl end of the methionine residue (7, 11). However, little or no o-CH3-S-[9-14C]2-FAA was detectable after the first chromatography (Fig. 2) and no radioactivity was associated with the mercaptoamide upon rechromatography. Third, modified RSase was analyzed for new NHt-terminal amino acids by the DKP-method.
Since only di-DNP-lysine was extracted with ether from the acid hydrolysate of the proteins, proteins A, B, and C had the same NHz-terminal amino acid, L-lysine, as native RNase and no peptide bonds involving other amino acids had been cleaved. Fourth, electrophoresis of modified RNase in gels of different pore size (Fig. 3) showed no appreciable differences in the patterns or in the relative amounts of the three proteins.
Since the pore size of the 9.6% gel is greater than that of the 14.6% gel (35), the similarity of the patterns eliminated the possibility that proteins A, B, and C were separated on the basis of differences in molecular size. All of these data led us to conclude that the modification of the RNase that we had observed was not attributable to cleavage of peptide bonds and fragmentation of the native RNase. As an alternative, we considered the possibility that the interaction of N-acetoxy-2-FAA with RNase yielded three proteins which differed only in their net positive charge. This would account for the electrophoretic separation of modified RNase into three components.
If the above explanation were correct, it would also be expected that modified RNase would be resolvable by ion exchange chromatography.
Accordingly, we chromatographed modified RNase on Bio-Gel CM-30 with a linear gradient of increasing NaCl concent.rat.ion (36). As shown in Fig. 4, the modified RNase was fractionated into three components which appeared to be comparable in their relative charge and amounts by guest on March 24, 2020 http://www.jbc.org/ Downloaded from to proteins A, B, and C obtained by electrophoresis. Therefore, ester. However, these experiments were complicated by the fact we retained the same designation (A, B, and C) for the three that native RNase, even after purification by gel chromatogproteins resolved by ion exchange chromatography.
Protein A raphy, still contained a second protein (< 5%) that was eluted was eluted in the same position as native RNase (Fig. 4). Pro-in the same position as protein B and that appeared to be inteins B and C were eluted with lower concentrations of NaCl than creased in amount by exposure of the RNase to 7 M urea ( 3, Table I). The conditions for t,he formation of proteins A, B, and C from N-acetoxy-2-FAB and RNase are summarized in Table I (Table II), acetylation of proteins B and C was approximately 5-to IO-fold greater than arylamidation in media containing acetone + 7 M urea or acetone alone. Since the dissociat.ion of N ,O-diacetylarylhydroxyla-  mines to amidonium ions is suppressed in organic solvents (37), high concentrations of acetone in the reaction medium would be expected to favor acetylation of RNase over arylamidation. However, acetylation was also predominant when the reaction was carried out in an essentially aqueous medium (Experiment 3, Table II) where the rate of formation of amidonium ions, and therefore arylamidation, is maximal (37). These results indicated that N-acetoxy-a-FAA is an acetyl donor and may modify proteins by acetylation.
The acetylation of RNase by N-[3H]acetoxy-2-FAA was confirmed by the recovery of [3H]acetic acid from acid or alkaline hydrolysates of modified RNase. The identity of the 3H in the hydrolysate with [$H]acetic acid was proved by redistillation of the acid without change of the specific radioactivity. Minor amounts of [3H]acetic acid, equivalent to 8.3% of bound 3H, were detected without subjecting modtied RNase to acid or alkaline hydrolysis (Table III).
This [3H]acetic acid came either from acetate that was adsorbed to the modified protein or from acetyl groups that were cleaved off acetylated amino acids under the conditions of the analysis.
Since this quantity of [3H]acetic acid could not be unequivocally assigned to acetyl groups released from an ester or amide linkage, it was subtracted from the [3H]acetic acid obtained after acid or alkaline hydrolysis of modified RNase. Alkaline hydrolysis for 2 and 18 hours yielded 3 and 12% of the bound 3H, respectively, as [3H]acetic acid (Table III). These alkali-labile acetyl groups were presumably bound in ester linkage to tyrosine, serine and/or threonine (34). Acid hydrolysis liberated 78% of the bound [3H]acetyl goups in modified RNase. Since the ester linkage of certain N , O-diacylhydroxyamino acids appears to be resistant to mild acid hydrolysis (38), the contribution to the above value by acetyl groups bound to RNase in ester linkage is uncertain.
A minimum value for acetyl groups in amide linkage (67%) may be calculated by subtracting the acetyl groups released from modified RNase by alkali (11%) from those liberated by acid (78%).
In any event, it seems clear that the major share of the hydrolyzable acetyl groups in modified RNase was derived from amide linkages.
A functional group of the amino acids of proteins that is available for acetylation to an amide under physiological conditions (39) is the c-amino group of lysine. Therefore, we attempted the isolation and identification of e-N-acety-L-lysine in modified RNase by two different methods.
In the first experiment, protein B was hydrolyzed with Pronase and trypsin and the hy- drolysate was chromatographed on AG 5OW-X4. Nearly 80% of the radioactivity was eluted with 0.20 bf ammonium acetate (Fig. 5). This fraction which contained neutral and acidic amino acids was chromatographed as a band on silica gel, and the radioactive acetylated amino acid was tentatively ident,ified by reference to authentic acetylated amino acids that were run concurrently (Fig. 6) on the amino acid analyzer. As indicated by the elution profile (Fig. 7), the radioactivity was associated almost exclusively with the fraction in which e-l\;-acetyl-L-lysine was eluted.
No radioactivity was detected in the fractions in which ac-hi-acetyl-L-lysine or 0-acetyl-L-serine were located. In a second experiment, the amounts of +N-[3H]acetyl-L-lysine in a Pronase plus trypsin hydrolysate of modified RNase m-ere determined by inverse isotope dilution.
The compound was chromatographed to constant specific radioactivity (Table IV). It was calculated that 60% of the 3H in the hydrolysate was associated with e-N-[3H]acetyl-L-lysine.
These data confirmed our conclu   sion that N-acetosy-2-FAA is an acetyl donor and, in the case of RNase, acetylates primarily the e-amino group of lysine. DISCUSSION Our experiments show that acetylation is clearly the major reaction in the modification of RNase by N-acetoxy-a-FAA. However, examination of the data indicates that acetylation explains only in part the decrease in the positive charge of RNase and thus the formation of proteins B and C. This conclusion is based on the following considerations.
As shown in Experiments 2 and 3 of Table II, 100 pmoles of protein B contained 67 and 60 pmoles of t3H]acetyl groups, respectively.
Since a decrease in 1 unit of charge per 100 pmoles RNase would require the incorporation of at least 100 pmoles of acetyl groups, acetylation by N-acetosy-2-FAA is evidently insufficient to account quantitatively for the observed decrease in the positive charge of RNase, and thus for the formation of protein B. Application of these considerations to t,he formation of protein C leads to the same conclusion.
We are forced to assume that arylamidation and other unknown interactions of N-acetoxy-2-FAA with RNase contribute to the alteration of the charge properties of the protein.
The release of [3H]acetyl groups from modified RNase by alkaline hydrolysis indicates that N-acetoxy-a-FAA esterified hydroxyamino acids in RNase, although the particular amino acids have not been identified. 0-Acetyl-L-tyrosine can be eliminated from consideration since it is deacetylated in 1 hour by 2 N NaOH at 25" (40). In our experiments, only trace amounts of [3H]acetyl groups were obtained from modified RNase under these conditions (Table III).
L-Serine and L-threonine would also be available for esterification to 0-acetyl-L-serine and O-acetyl-Lthreonine (34). These acetylated amino acids were not found in enzymatic hydrolysates of modified RNase despite the fact that 11 y. of bound [3H]acetyl groups were liberated from the modified protein by alkaline hydrolysis (Table III).
The absence of the 0-acetylated amino acids might have been due to their rearrangement to the N-acetylated derivatives at the pH maintained during enzymatic hydrolysis (16). If rearrangement had occurred, the N-acetylated compounds would have been washed through AG SOW-X4 and would have appeared in the minor radioactive fraction eluted from the resin with water (Fig. 5). Because 0-acetylation of serine and threonine by N-acetoxy-2-FAA represents a relatively minor reaction in comparison to N-acetylation of lysine and because the charge properties of RNase would not be affected by 0-acetylation, the identification of the products of the rearrangement, i.e. N-acetyl-L-serine and N-acetyl-L-threonine, was not further pursued.
It has been postulated that N-acetoxy-2-FAA decomposes to an arylamidonium ion through an activated ion pair in which the acetoxy group is spatially separated from the remainder of the molecule (37). It has also been shown that the rate of formation of the amidonium ion is a function of the water content of the medium (37). Our data are in agreement with these findings since arylamidation of proteins B and C was increased by 50% when the reaction between N-acetoxy-a-FAA and RNase was carried out in 2.4% ethanol instead of in 50 y. acetone (Table II).
It has been reported that certain N-acetoxy-N-arylamides, such as N-acetoxy-4-acetylaminobiphenyl and N-acetoxy-2-acetylaminophenanthrene, acetylate, rather than arylamidate, guanosine (6). It, also, has been inferred that the acetylation proceeds through an ion pair similar to that proposed for arylamidation (37). However, our data indicate that acetylation of RNase by N-acetoxy-2-FAA was decreased by 11% under conditions which increased the extent of arylamidation by 50%. Because of the lack of correspondence between arylamidation and acetylation, it appears unlikely that acetylation proceeds through an activated ion pair resembling that proposed for arylamidation. The possibility remains to be explored whether the mechanism of the acetylation of proteins by N-acetoxy-2-FAA consists instead of a nucleophilic attack of the nitrogen of the e-amino group of lysine on the partially positively charged carbonyl carbon atom of N-acetoxy-2-FAA, as shown in Fig. 8. Heretofore, efforts to explain the mode of action of the carcinogenic N ,O-diacetylarylhydroxylamines and their phosphate and sulfate analogues have centered on the arylamidation of proteins and of nucleic acids. In this report, we present evidence that N-acetoxy-2-FAA is an acyl donor for proteins and that its capacity for acetylation exceeds that for arylamidation by several orders of magnitude.
It seems possible, also, that the phosphate and sulfate analogues of N-acetoxy-2-FAA which are considered to be biologically active forms of N-hydroxy-2-FAA may phosphorylate or sulfonate, or both, proteins.
If the chemical reactions underlying the action of carcinogenic esters of hydroxamic acids were the transfer of the acetyl, phosphoryl, or sulfonyl group from the ester to cellular receptors, the hydroxamic acid would be regenerated and be available for re-esterification as indicated in Fig. 8. By a repetition of this cyclic process, a few donor molecules could modify a large number of receptors.
In contrast, in modification by arylamidation the hydroxamic acid is not regenerated and a cyclic process is, therefore, not operative.
The limiting factor in modification by arylamidation would be the amount of activated carcinogen present in the cell.
We have previously presented evidence that N-Sfluorenylacetamide or its activated metabolite(s) arylamidate rat liver histones at a trace level (8). On the basis of our present data, it seems possible that the histones and other nuclear proteins were extensively modified in these earlier experiments by the mechanism suggested above. There is evidence that reversible acetylation or phosphorylation, or both, of histones and acidic nuclear proteins removes the inhibition that these proteins exert on the transcriptional process (9). Further experiments are needed to determine whether N-acetoxy-2-FAA and related carcinogens donate acetyl, phosphoryl, or sulfonyl groups to chromosomal proteins and whether the transfer of these groups affects the control of transcription by these proteins.