Differences in Lysine Adduction by Acrolein and Methyl Vinyl Ketone

Acrolein is a highly toxic environmental pollutant that readily alkylates the ... this issue, we compared the chemistry of protein adduction by acrole...
0 downloads 0 Views 192KB Size
Download PDF
NOVEMBER 2005 VOLUME 18, NUMBER 11 © Copyright 2005 by the American Chemical Society

Communications Differences in Lysine Adduction by Acrolein and Methyl Vinyl Ketone: Implications for Cytotoxicity in Cultured Hepatocytes Lisa M. Kaminskas,†,‡,§ Simon M. Pyke,‡ and Philip C. Burcham*,†,§ Department of Clinical and Experimental Pharmacology, and the School of Chemistry and Physics, The University of Adelaide, Adelaide, South Australia 5005 Received September 2, 2005

Acrolein is a highly toxic environmental pollutant that readily alkylates the -amino group of lysine residues in proteins. In model systems, such chemistry involves sequential addition of two acrolein molecules to a given nitrogen, forming bis-Michael-adducted species that undergo aldol condensation and dehydration to form N-(3-formyl-3,4-dehydropiperidino)lysine. Whether this ability to form cyclic adducts participates in the toxicity of acrolein is unknown. To address this issue, we compared the chemistry of protein adduction by acrolein to that of its close structural analogue methyl vinyl ketone, expecting that the R-methyl group would hinder the intramolecular cyclization of any bis-adducted species formed by methyl vinyl ketone. Both acrolein and methyl vinyl ketone displayed comparable protein carbonylating activity during in vitro studies with the model protein bovine serum albumin, confirming the R,β,-unsaturated bond of both compounds is an efficient Michael acceptor for protein nucleophiles. However, differences in adduction chemistry became apparent during the use of electrospray ionization-MS to monitor reaction products in a lysine-containing peptide after modification by each compound. For example, although a Schiff base adduct was detected following reaction of the peptide with acrolein, an analogous species was not formed by methyl vinyl ketone. Furthermore, while ions corresponding to mono- and bis-Michael adducts were detected at the N-terminus and lysine residues following peptide modification by both carbonyls, only acrolein modification generated ions attributable to cyclic adducts. Despite these differences in adduction chemistry, in mouse hepatocytes, the two compounds exhibited very comparable abilities to induce rapid, concentration-dependent cell death as well as protein carbonylation. These findings suggest that the acute toxicity of short-chain R,β-unsaturated carbonyl compounds involves their ability to form acyclic Michael addition adducts rather than Schiff conjugates or heterocyclic adducts.

Introduction Acrolein, a highly toxic pollutant formed during the combustion of organic matter, is strongly implicated in * To whom correspondence should be addressed. Phone, 61-8-9346 2986; fax, 61-8-9346 3469; e-mail, [email protected]. † Department of Clinical and Experimental Pharmacology. ‡ School of Chemistry and Physics. § Present addresses: P.C.B., Pharmacology Unit, School of Medicine and Pharmacology, The University of Western Australia, Nedlands, WA 6009, Australia; phone, 61-8-9346 2986; fax, 61-8-9346 3469; e-mail, [email protected]. L.M.K., Department of Pharmaceutics, Victorian College of Pharmacy, Monash University, 381 Royal Pde, Parkville, VIC 3052, Australia; phone, 61-3-9903 9610; e-mail, [email protected].

acute lung injury seen in victims of smoke inhalation (1, 2). It is also produced endogenously during membrane peroxidation and thus contributes to a range of chronic diseases that involve oxidative stress (3-5). In particular, recent research in animals has uncovered a role for acrolein in the pathological events triggered by physical trauma to the spinal cord (6, 7). The pronounced toxicity of acrolein reflects its ability to alkylate nucleophilic centers in cell macromolecules, including lysine, histidine, cysteine, and guanine residues (8). Since adduction typically proceeds via Michael addition chemistry, the resulting carbonyl-retaining adducts are able to participate in secondary cross-linking reactions (9, 10). Schiff

10.1021/tx0502387 CCC: $30.25 © 2005 American Chemical Society Published on Web 10/14/2005

1628

Chem. Res. Toxicol., Vol. 18, No. 11, 2005

Communications Scheme 1

base adducts also form during reactions of acrolein with primary amines in proteins and DNA, although they are usually less significant in quantitative terms (8). In vitro studies of the reactivity of acrolein with either NR-protected lysine or lysine-containing model peptides have revealed novel adduction chemistry whereby Michael adduction by two molecules of aldehyde occurs on the same -amino group of a given lysine side chain (Scheme 1) (11). The resulting bis-adducts then undergo sequential aldol condensation and dehydration reactions to generate the heterocycle N-(3-formyl-3,4-dehydropiperidino)lysine (FDP-lysine) (5, 11, 12). Furuhata et al. also identified an analogous reaction in which Michael addition of a second acrolein molecule occurs on Schiff-adducted lysine, although the resulting methylpyridinium adduct does not appear to be major in quantitative terms (13). FDP-lysine appears to be particularly immunogenic, and antibodies with activity toward this lesion have detected acroleinmodified proteins in a range of diseased tissues known to sustain chronic oxidative stress, including diabetic kidneys and atherosclerotic arteries (4, 5). Immunochemical evidence for lysine adduction by acrolein has also been obtained in aortic tissue of cyclophosphamidetreated rats (14) as well as in spinal cord proteins from guinea pigs subjected to spinal trauma (7). Despite such immunochemical evidence for lysine adduction by acrolein, definitive chemical characterization of acroleinderived protein adducts in biological systems is lacking. Consequently, the role played by cyclic adducts such as FDP-lysine in the toxicity of acrolein, in contrast to the contribution by acyclic mono- and bis-Michael-adducted species, is ill-defined. Since R,β-unsaturated aldehydes can sometimes react with macromolecules in vitro to form oligomeric adducts of unknown physiological relevance (15, 16), the question of the biological and toxicological significance of doubly adducted species such as FDPlysine requires careful attention. The R,β-unsaturated ketone methylvinyl ketone (MVK)1 is a close structural analogue of acrolein that may be useful in clarifying the chemical events underlying the toxicity of short-chain unsaturated carbonyls. MVK’s toxicological properties have received less attention than those of acrolein, although Lash and colleagues have established that it is highly toxic to rat kidney tubule cells, inducing protein alkylation, glutathione loss, ATP depletion, and impairment of mitochondrial respiration 1Abbreviations: BSA, bovine serum albumin; LDH, lactate dehydrogenase; MVK, methyl vinyl ketone; PPE, preproenkephalin fragment 128-140.

(17, 18). These effects closely resemble those elicited by acrolein in kidney cells (18, 19). The chemistry of macromolecular modification by MVK is poorly characterized, although it has been shown to generate a range of DNA adducts involving deoxyguanosine, including a cyclic 1,N2 propano-adduct that structurally resembles DNA adducts formed by acrolein (20). MVK’s reactivity with proteins is completely unexplored, although on account of its electrophilic unsaturated bond it can be expected to modify protein nucleophiles via strong Michael adduction (17, 18). Consequently, we hypothesized that during reactions with the -amine of lysine, MVK can be expected to form mono- and bis-Michael adducts analogous to those generated by acrolein (Scheme 1). However, since the methyl group of MVK seemed likely to hinder aldol condensation of bis-adducted species, the yield of cyclic adducts analogous to FDP-lysine was expected to be lower from MVK. To test these predictions, we compared the chemistry of adduction by these compounds in a model lysine-containing peptide. Furthermore, the inherent toxic potential of acrolein and MVK was evaluated in cultured cells, with a view to drawing preliminary conclusions concerning the toxicological significance of any differences in adduction chemistry between these closely related compounds.

Experimental Procedures Reagents. Preproenkephalin fragment 128-140 (PPE) was purchased from ICN Biomedicals (Costa Mesa, CA). Acrolein and methyl vinyl ketone were supplied by the Aldrich Chemical Co. (Milwaukee, WI). Primary and secondary antibodies and all other reagents were purchased from the Sigma Chemical Co. (St. Louis, MO). Animals. Male Swiss mice were purchased at 4-6 weeks of age from Laboratory Animal Services at the Waite Institute of the University of Adelaide. The mice were maintained at a constant ambient temperature of 21 °C on a 12 h light/dark cycle and were allowed free access to food and water. All animal use in these experiments was approved by the Animal Ethics Committee of the University of Adelaide. Peptide Modification and Mass Spectrometry. As in our recent studies with acrolein, the preproenkephalin fragment 128-140 (PPE, GGEVLGKRYGGFM, mass ) 1370 g/mol) served as a model lysine-containing peptide in these experiments (21). PPE was dissolved in 300 µL of Nanopure water to a concentration of 100 µM before an equivalent volume of water containing 10 mM MVK was added. The tubes were then sealed and placed in a 37 °C incubator. At various time intervals following the commencement of peptide modification (0.5, 1.0, 1.5, 3, 6, and 22 h), 25 µL aliquots of reaction mixture were removed and diluted with 25 µL of water followed by 50 µL

Communications volumes of 50% acetonitrile/2% glacial acetic acid. Parallel incubations were carried out using acrolein at peptide/carbonyl ratios equivalent to those specified for MVK. The samples were then analyzed via electrospray ionization MS using a Finnigan LCQ mass spectrometer in positive ion monitoring mode. The spray voltage of the instrument was set at 4.8 kV with a capillary temperature of 200 °C. The cylinder gas (N2) was maintained at 100 psi. Samples were introduced into the mass spectrometer at a constant flow rate of 8 µL/min. Mass spectra were collected over a m/z range of 400-2000. Cell Culture Experiments. Mouse hepatocytes were isolated via collagenase digestion by a method described previously (22). Following isolation and three rounds of washing in KrebsHenseleit buffer, cells were resuspended at 1 × 106 cells/mL in RPMI-1640 medium supplemented with L-glutamine (0.03%), bovine serum albumin (BSA, 0.2%), and penicillin/streptomycin (50 U/L and 50 µg/mL, respectively). After allowing 2 h for hepatocyte attachment to collagen-coated 60 mm polystyrene dishes (3 mL suspension per plate), the monolayers were washed twice with 2 mL volumes of PBS (50 mM, pH 7.4) to remove inadherent cells. Experiments were then commenced by exposing cells to BSA-free RPMI-1640 that was supplemented with 0, 20, 50, 100, 150, or 300 µM concentrations of either acrolein or MVK. The unsaturated carbonyl compounds were added from stock solutions prepared daily in PBS. The dishes were then returned to the incubator with media samples taken for determination of lactate dehydrogenase (LDH) leakage at 1, 2, and 4 h. LDH activities in 10 µL aliquots of media were assessed using a microplate fluorescence assay (23). Protein Carbonylation by MVK and Acrolein. The reactivity of acrolein and MVK toward the model protein bovine serum albumin (BSA) was compared using an immunochemical assay based on adduct derivatization with 2,4-dinitrophenylhydrazine (24). Solutions of acrolein or MVK were added to Eppendorf tubes that each contained 5 mg of BSA dissolved in phosphate buffer (50 mM, pH 7.0) to yield final carbonyl concentrations of 0, 30, 100, 300, or 1000 µM in a total volume of 1 mL. The tubes were then incubated for 2 h at 37 °C prior to chilling on ice. Protein was then precipitated by adding 100 µL trichloroacetic acid solution (20%) to each tube after which the samples were centrifuged for 5 min at 7000g. The resulting pellets were resuspended in 1 mL volumes of 8 M urea, and then 20 µL aliquots of each sample were diluted with an equal volume of derivatization solution (0.5% w/v 2, 4-dinitrophenylhydrazine in 10% trifluoroacetic acid). After allowing derivatization to proceed for 30 min at room temperature, the samples were neutralized by adding 36 µL of Tris base solution (2.0 M) containing 40% glycerol. A 7.6 µL volume of the resulting solution was then diluted with 2.4 µL of water before 10 µL of SDS-PAGE sample buffer was added to each tube. Following heat denaturation and cooling to room temperature, 10 µL aliquots (i.e., 5 µg of protein) were resolved on 12% acrylamide minigels (4% stacking gel) at 200 V for 45 min. After transfer to nitrocellulose (100 V, 30 min), membranes were blocked with 10% nonfat milk in PBS for 60 min and then treated overnight at 4 °C with 1/1000 dilutions of rabbit anti-dinitrophenyl serum (24). Following washing and exposure to alkaline phosphatasecoupled goat anti-rabbit IgG serum (Sigma, 1/4000 dilution, 60 min), membranes were developed using SigmaFast BCIP/NBT substrate tablets. Densitometry analysis of the resulting immunoblots was performed using Kodak Digital Science image analysis software (Vs. 1.3, Eastman Kodak Co., Rochester, NY). In a related experiment, protein carbonylation was assessed in mouse hepatocyte proteins following 30 min exposure of cells to BSA-free RPMI medium containing 0, 20, 50, 100, 150, or 300 µM MVK or acrolein. At the end of the incubations, cells were washed in cold PBS, and then 1 mL of cold 3% perchloric acid was added to each dish. The cells were then scraped from the dishes and centrifuged for 4 min at 4000g. The resulting protein pellets were dissolved in 8 M urea prior to estimating protein concentrations via absorbance measurements at 260 and 280 nm using a quartz cuvette (25). Sample aliquots containing

Chem. Res. Toxicol., Vol. 18, No. 11, 2005 1629

Figure 1. Both acrolein and MVK efficiently carbonylate BSA. BSA (5 mg/mL) was incubated with 0, 30, 100, 300, or 1000 µM concentrations of acrolein or MVK for 2 h at 37 °C. Following derivatization with 2,4-dinitrophenylhydrazine, aliquots comprising 5 µg of protein were resolved via SDS-PAGE prior to Western blotting using anti-dinitrophenyl serum. The various lane contents are (1) control, (2) 30 µM acrolein, (3) 100 µM acrolein, (4) 300 µM acrolein, (5) 1000 µM acrolein, (6) 30 µM MVK, (7) 100 µM MVK, (8) 300 µM MVK, and (9) 1000 µM MVK. 40 µg of protein were then derivatized and assessed for carbonyl groups using the same procedures as described above.

Results Protein-Carbonylating Reactivity of Acrolein and MVK. To facilitate comparisons between the adduction chemistry and toxicities of MVK and acrolein, it was necessary to ensure the two compounds exhibited comparable reactivity as Michael acceptors toward a model protein. Figure 1 shows a Western blot obtained during analysis of the carbonyl content of BSA that had been subjected to a 2 h incubation at 37 °C in the presence of 0, 30, 100, 300, or 1000 µM concentrations of acrolein (Lanes 2-5) or MVK (Lanes 6-9). While a low carbonyl content was detected in native BSA (Lane 1), both acrolein and MVK produced strong, concentration-dependent increases in protein carbonyls. Densitometry revealed that 30, 100, 300, and 1000 µM acrolein increased band intensities by 1.4-, 3.0-, 5.7-, and 9.1-fold relative to native BSA (Lanes 2-5, respectively, cf. Lane 1). In Lanes 6-9, modification with 30, 100, 300, and 1000 µM MVK increased band intensities respectively by 2.2-, 3.1-, 5.0-, and 5.6-fold relative to controls. Since these findings indicate that acrolein and MVK exhibit comparable intrinsic reactivity toward the model protein, further experiments were performed to see if the latter generated a less diverse range of adducts at lysine compared to acrolein. Chemistry of Peptide Modification by Acrolein and MVK. Consistent with insights gained by Uchida et al. during characterization of lysine adduction by acrolein (11, 12), reaction of the lysine-containing peptide preproenkephalin fragment 128-140 (PPE) with acrolein for 3 h prior to analysis by electrospray ionization-MS resulted in the detection of several reaction products (Figure 2A). The major ions could be attributed to the MH+ of the unmodified peptide (m/z 1370), the expected mono-Michael adduct (m/z 1426), a bis-Michael-adducted species (m/z 1482), as well as a strong ion at m/z 1464 consistent with the cyclic adduct FDP-lysine (Scheme 1 and Figure 2A). A minor Schiff adduct was also detected at m/z 1408. Reaction with MVK also led to new ions during MS analysis of PPE (Figure 2). A mass spectrum collected from PPE following an extended 22 h reaction with MVK is shown in Figure 2B. Ions corresponding to the MH+ of singly-Michael-adducted PPE and also doubly-Michaeladducted PPE are readily apparent (m/z 1440 and 1510, respectively) (Figure 2B). These two ions also dominated mass spectra collected at earlier time points during the modification of PPE by MVK (i.e., spectra collected at 0.5, 1.0, 1.5, 3, and 6 h; data not shown). However, in contrast to the spectrum obtained from acrolein-modified PPE

1630

Chem. Res. Toxicol., Vol. 18, No. 11, 2005

Communications

Figure 2. Mass spectra obtained during ESIMS analysis of (A) acrolein- and (B) MVK-modified PPE. The spectra shown were obtained following 3 h (Panel A) and 22 h (Panel B) reactions of PPE (100 µM) with 10 mM concentrations of the respective unsaturated carbonyls. The structures of adducts formed on lysine are shown for the dominant ions.

(Figure 2A), an ion corresponding to a Schiff conjugate was not detected during analysis of the MVK-modified peptide (Figure 2B). Moreover, the expected MH+ of a heterocyclic species formed upon ring closure and dehydration of the bis-Michael adduct was not present in spectra collected from MVK-modified PPE at any time point (e.g., in Figure 2B, such an ion would be detected at m/z 1492). Thus, the collective data in Figure 2 confirm that although both acrolein and MVK readily generated singly- and doubly-Michael-adducted species in a model peptide, the methyl group possessed by MVK hindered aldol condensations to form cyclic species analogous to the FDP-lysine species generated by acrolein. PPE contains an N-terminal primary amine in addition to the -NH2 group of the central lysine; hence, it was possible that adduction occurred on both nucleophilic centers. To assess this likelihood, MS/MS fragmentation analysis was performed on the two major ions detected during MS analysis of the MVK-modified peptide (Figure 3). In Figure 3, the mass of fragments generated during the loss of each residue is shown either above the peptide sequence (Y + 2 cleavages) or below the sequence for B-type cleavages. Adduction sites were revealed by the loss of an additional 70 m/z units during cleavage of the respective residue. MS/MS analysis of the ion corre-

sponding to singly adducted peptide (m/z 1440) revealed that two species contributed to this ion, namely, peptides containing Michael adducts on the central lysine as well as those modified on the N-terminal glycine (shown, respectively, in panels A and B in Figure 3). The m/z 1510 ion corresponding to doubly adducted peptide also comprised three distinct species as per panels C, D, and E of Figure 3. These were modified PPE molecules containing bis-Michael-adducted lysine (Figure 3C), species comprising a bis-Michael-adducted N-terminus (Figure 3D), and peptide derivatives with mono-Michael adducts on both the lysinyl and N-terminal residues (Figure 3E). Comparative Toxicity of MVK and Acrolein in Mouse Hepatocytes. To see whether the differences in adduction chemistry between MVK and acrolein are relevant to the cellular effects of these compounds, we compared their toxicity and protein carbonylating activity in cultured mouse hepatocytes. The assumption underlying these experiments was that if an ability to form Schiff conjugates or heterocyclic adducts was crucial to the acute toxicity of short-chain unsaturated carbonyls, then acrolein would exhibit greater cytotoxicity than MVK. Despite this expectation, acrolein (Figure 4A) and MVK (Figure 4B) elicited very similar toxicity in terms of the extent and time-course of lactate dehydrogenase leakage

Communications

Figure 3. Assignment of ions detected during MS/MS analysis of the two dominant species detected during ESIMS analysis of MVK-modified PPE. B cleavages are indicated by the upper brackets (ion masses indicated above the peptide sequence), while Y + 2 cleavages are shown by the lower brackets (ion masses indicated below the sequence).

Figure 4. Induction of time- and concentration-dependent cell death by acrolein (Panel A) and MVK (Panel B). Mouse hepatocytes were incubated for up to 4 h with culture medium alone (b) or various concentrations of each carbonyl compound: 20 µM (O), 50 µM (1), 100 µM (3), 150 µM (9), or 300 µM (0). Media aliquots were taken at 0, 1, 2, and 4 h and assessed for lactate dehydrogenase (LDH) leakage as an indicator of cell death. Each data point represents the mean ( SEM of three independent observations. The immunoblots (40 µg of protein/ lane) depict carbonylation of hepatocellular proteins by acrolein (Panel C) and MVK (Panel D) following a 30 min incubation. The arrow depicts the 55 kDa band mentioned in the text. The lane contents in Panels C and D are Lane 1, control; Lane 2, 20 µM carbonyl compound; Lane 3, 50 µM; Lane 4, 100 µM; Lane 5, 150 µM; and Lane 6, 300 µM.

from cells exposed to equivalent concentrations of the two compounds (Figure 4). The 4 h concentration-response curves for acrolein and MVK were normalized to controls,

Chem. Res. Toxicol., Vol. 18, No. 11, 2005 1631

and then carbonyl concentrations producing half-maximal lethality (LC50) were estimated using the statistical analysis software (Prism v.4). No statistically significant difference was observed between the 4 h LC50 values obtained for acrolein and MVK (143 ( 28 µM versus 136 ( 15 µM, respectively, p > 0.05, unpaired t-test), indicating that the inherent cytotoxicity of the two compounds toward hepatocytes was very similar. Panels C and D of Figure 4 depict Western blots obtained during the immunochemical analysis of protein carbonyls following a 30 min exposure of hepatocytes to various concentrations of acrolein and MVK. In keeping with endogenous exposure to oxidants and carbonylating agents, high levels of protein carbonyls were detected in untreated hepatocytes (Lane 1 of Figure 4C,D). Exposure of cells to 20-300 µM concentrations of acrolein (Figure 4C, Lanes 2-6) and MVK (Figure 4D, Lanes 2-6) produced strong concentration-dependent increases in protein carbonyls. Although carbonylation involved a wide range of proteins, an approximately 55 kDa species was a common target for both carbonyls and was sufficiently well-defined to permit analysis via densitometry. In acrolein-treated cells, the intensity of the 55 kDa band increased by 1.5-, 1.7-, 8.1-, 9.6-, and 22-fold relative to controls upon exposure to respective aldehyde concentrations of 20, 50, 100, 150, and 300 mM (Figure 4C, Lanes 2-6). MVK elicited comparable increases in the intensity of this band, with 2.6-, 3.3-, 4.7-, 8.0-, and 6.8-fold increases over controls seen at MVK concentrations of 20, 50, 100, 150, and 300 µM, respectively (Figure 4D, Lanes 2-6). Experiments are currently underway to establish the identity of this 55 kDa protein.

Discussion The present work provides new insights into the toxicological properties of methyl vinyl ketone (MVK), a close structural analogue of the ubiquitous environmental pollutant acrolein. Although acrolein’s toxicology has been more widely studied than that of MVK, the latter compound deserves attention given its wide use in the chemical industry during the synthesis of steroids, styrene, biodegradable polymers, and vitamin A. In addition to the potential for exposure via industrial usage, MVK also forms within the atmosphere via photochemical oxidation of biogenic and anthropogenic hydrocarbons (26, 27). Consistent with its volatility and chemical reactivity, inhalational exposure to airborne MVK can damage the upper respiratory tract (28). Although previous studies have characterized key adducts formed during the reaction of MVK with DNA (20), the present work presents new insights into the chemistry of protein adduction by MVK. As R,β-unsaturated carbonyl compounds, MVK and acrolein are bifunctional electrophiles capable of reacting with protein amines via either their carbonyl group (to form Schiff base adducts) or their unsaturated bond to form Michael addition adducts. In the case of acrolein, the Schiff adduct (additional mass of 38 units) was detected as a minor reaction product at m/z 1408 during ESIMS analysis of modified peptide (Figure 2A), yet in the case of MVK modifications, no ion indicative of Schiff base formation was detected (Figure 2B). In contrast, both carbonyl compounds reacted via Michael-addition chemistry to form mono- and bis-adducts (Figure 2). In the case of MVK-modified PPE, the ion attributed to singly adducted

1632

Chem. Res. Toxicol., Vol. 18, No. 11, 2005

peptide was of stronger intensity than its doubly adducted counterpart (m/z 1440 vs 1510). This may suggest that bis adduction events are less favored for MVK than is the case for acrolein. However, this conclusion is tentative, since differences in the ionization properties of singly and doubly adducted species might influence ion abundance during ESIMS analysis. The most significant finding from the present work was that, compared to their acrolein-derived counterparts, bisMichael adducts generated by MVK were much less prone to spontaneous ring closure and dehydration to form heterocyclic adducts. Thus, even after extended modification using a large excess of MVK, counterparts to the acrolein-derived heterocycle FDP-lysine were not detected in the MVK-modified peptide. This finding is significant given that the inherent cytotoxicities of acrolein and MVK were similar in hepatocytes (Figure 4). Assuming its inability to form heterocyclic adducts extends to the cellular setting, the finding that MVK was as cytotoxic to cells as acrolein strongly implies that formation of cyclic adducts is not vital to the acute toxicity of these compounds. This was not a foregone conclusion, since Furuhata et al. have previously demonstrated that FDP-lysine (i.e., derived from acrolein) is a reactive species that is readily attacked by thiol nucleophiles including the cysteine-containing tripeptide glutathione (29). Since this raised the possibility that FDP-lysine might contribute to thiol depletion and also form protein cross-links (i.e., via reactions with cysteine residues in adjacent proteins), it seemed feasible that FDP-lysine might amplify cell damage in acroleinexposed cells. On account of our finding that MVK triggered acute cell death despite an inability to form heterocyclic adducts, it seems reasonable to conclude that FDP-lysine does not amplify cell injury during acrolein toxicity. However, since this conclusion rests on extrapolation of observations in cell-free systems to much more complicated cellular models, future work should seek to develop robust analytical assays for definitive detection of individual adducts formed by MVK and acrolein to allow this inference to be tested directly. Since both MVK and acrolein readily generated acyclic Michael adducts at N-containing nucleophiles, future work should address the role of these adducts in the acute toxicity of unsaturated carbonyl compounds. Such a role is consistent with our finding that hydrazino drugs block acrolein-mediated cell killing at least in part via “adducttrapping” reactions between drugs and acrolein-adducted proteins (21, 30). During in vitro experiments in hepatocytes, cytoprotection was achieved when the drug was added to cells that had been preexposed to a toxic concentration of acrolein precursor (allyl alcohol) for just 30 min, but were washed free of the toxicant prior to drug addition. Cytoprotection under these conditions was accompanied by intense “trapping” of acrolein-adducted proteins, indicating that Michael addition adducts were abundant on cell proteins after a brief exposure to acrolein (21). Whether levels of FDP-lysine would have accumulated by this time is unknown, but our study of the kinetics of peptide modification by acrolein revealed that acyclic mono- and bis-acrolein adducts predominated within the first 30 min of commencing modification by acrolein.2 If such kinetics of protein modification are found to extend to the cellular environment, future work 2L.

M. Kaminskas, unpublished observation.

Communications

should focus on clarifying the chemical fate of acyclic adducts in the early stages of acute acrolein toxicity. In conclusion, this work provides new insights into the similarities between the chemistry of protein adduction by two related R,β-unsaturated carbonyl compounds, acrolein and MVK. Both compounds generated acyclic mono and bis Michael adducts at the lysinyl and Nterminal amines of a model peptide, although only acrolein-derived bis adducts were susceptible to intramolecular aldol condensation to form heterocyclic adducts. Since MVK and acrolein displayed comparable carbonylating efficiency in both a model protein and isolated hepatocytes and exhibited similar cytotoxic potency in the cells, it can be inferred that the ability to form cyclic adducts from bis-adducted species is not fundamental to their acute toxicity.

Acknowledgment. Funding assistance received from the Research Committee of the Faculty of Health Sciences is gratefully acknowledged.

References (1) Hales, C. A., Barkin, P. W., Jung, W., Trautman, E., Lamborghini, D., Herrig, N., and Burke, J. (1988) Synthetic smoke with acrolein but not HCl produces pulmonary edema. J. Appl. Physiol. 64, 1121-1133. (2) Hales, C. A., Musto, S. W., Janssens, S., Jung, W., Quinn, D. A., and Witten, M. (1992) Smoke aldehyde component influences pulmonary edema. J. Appl. Physiol. 72, 555-561. (3) Lovell, M., Xie, C., and Markesbery, W. (2001) Acrolein is increased in Alzheimer’s disease and is toxic to primary hippocampal cultures. Neurobiol. Aging 22, 187-194. (4) Suzuki, D. and Miyata, T. (1999) Carbonyl stress in the pathogenesis of diabetic nephropathy. Intern. Med. 38, 309-314. (5) Uchida, K., Kanematsu, M., Sakai, K., Matsuda, T., Hattori, N., Mizuno, Y., Suzuki, D., Miyata, T., Noguchi, N., Niki, E., and Osawa, T. (1998) Protein-bound acrolein: Potential markers for oxidative stress. Proc. Natl. Acad. Sci. U.S.A. 95, 4882-4887. (6) Luo, J., and Shi, R. (2004) Acrolein induces axolemmal disruption, oxidative stress, and mitochondrial impairment in spinal cord tissue. Neurochem. Int. 44, 475-486. (7) Luo, J., Uchida, K., and Shi, R. R. (2005) Accumulation of acroleinprotein adducts after traumatic spinal cord injury. Neurochem. Res. 30, 291-295. (8) Esterbauer, H., Schaur, R., and Zollner, H. (1991) Chemistry and biochemistry of 4-hydroxynonenal, malondialdehyde and related aldehydes. Free Radical Biol. Med. 11, 81-128. (9) Kuykendall, J. and Bogdanffy, M. (1992) Efficience of DNAhistone crosslinking induced by saturated and unsaturated aldehydes in vitro. Mutat. Res. 283, 131-136. (10) Kozekov, I. D., Nechev, L. V., Sanchez, A., Harris, C. M., Lloyd, R. S., and Harris, T. M. (2001) Interchain cross-linking of DNA mediated by the principal adduct of acrolein. Chem. Res. Toxicol. 14, 1482-1485. (11) Uchida, K., Kanematsu, M., Morimitsu, Y., Osawa, T., Noguchi, N., and Niki, E. (1998) Acrolein is a product of lipid peroxidation reaction. Formation of free acrolein and its conjugate with lysine residues in oxidized low-density lipoproteins. J. Biol. Chem. 273, 16058-16066. (12) Uchida, K. (1999) Current status of acrolein as a lipid peroxidation product. Trends Cardiovasc. Med. 9, 109-113. (13) Furuhata, A., Ishii, T., Kumazawa, S., Yamada, T., Nakayama, T., and Uchida, K. (2003) N-(3-methylpyridinium)lysine: Identification of a major antigenic adduct generated in acroleinmodified protein. J. Biol. Chem. 278, 48658-48665. (14) Arikketh, D., Niranjali, S., and Devaraj, H. (2004) Detection of acrolein-lysine adducts in plasma low-density lipoprotein and in aorta of cyclophosphamide-administered rats. Arch. Toxicol. 78, 397-401. (15) Stone, K., Uzieblo, A., and Marnett, L. J. (1990) Studies of the reaction of malondialdehyde with cytosine nucleosides. Chem. Res. Toxicol. 3, 467-472. (16) Stone, K., Ksebati, M. B., and Marnett, L. J. (1990) Investigation of the adducts formed by reaction of malondialdehyde with adenosine. Chem. Res. Toxicol. 3, 33-38.

Communications (17) Lash, L., Tokarz, J., and Pegouske, D. (1995) Susceptibility of primary cultures of proximal tubular and distal tubular cells from rat kidney to chemically induced toxicity. Toxicology 103, 85103. (18) Lash, L. and Woods, E. (1991) Cytotoxicity of allkylating agents in isolated rat kidney proximal tubular and distal tubular cells. Arch. Biochem. Biophys. 286, 46-56. (19) Zaki, E. L., Springate, J. E., and Taub, M. (2003) Comparative toxicity of ifosfamide metabolites and protective effect of mesna and amifostine in cultured renal tubule cells. Toxicol. in Vitro 17, 397-402. (20) Eder, E., Hoffman, C., and Deininger, C. (1991) Identification and characterization of deoxyguanosine adducts of methyl vinyl ketone and ethyl vinyl ketone. Genotoxicity of the ketones in the SOS chromotest. Chem. Res. Toxicol. 4, 50-57. (21) Burcham, P., Fontaine, F. R., Kaminskas, L. M., Petersen, D. R., and Pyke, S. M. (2004) Protein adduct-trapping by hydrazinophthalazine drugs: Mechanisms of cytoprotection against acroleinmediated toxicity. Mol. Pharmacol. 65, 655-664. (22) Harman, A. W., McCamish, L. E., and Henry, C. A. (1987) Isolation of hepatocytes from postnatal mice. J. Pharmacol. Methods 17, 157-163. (23) Kaminskas, L. M., Pyke, S. M., and Burcham, P. C. (2004) Reactivity of hydrazinophthalazine drugs with the lipid peroxidation products acrolein and crotonaldehyde. Org. Biomol. Chem. 2, 2578-2584.

Chem. Res. Toxicol., Vol. 18, No. 11, 2005 1633 (24) Keller, R. J., Halmes, N. C., Hinson, J. A., and Pumford, N. R. (1993) Immunochemical detection of oxidized proteins. Chem. Res. Toxicol. 6, 430-433. (25) Bollag, D. M., Rozycki, M. D., and Edelstein, S. J. (1996) Protein Methods, 2nd ed., Wiley-Liss, New York. (26) Cerqueira, M. A., Pio, C. A., Gomes, P. A., Matos, J. S., and Nunes, T. V. (2003) Volatile organic compounds in rural atmospheres of central Portugal. Sci. Total Environ. 313, 49-60. (27) Doyle, M., Sexton, K. G., Jeffries, H., Bridge, K., and Jaspers, I. (2004) Effects of 1,3-butadiene, isoprene, and their photochemical degradation products on human lung cells. Environ. Health Perspect. 112, 1488-1495. (28) Morgan, D., Price, H. C., O’Connor, R. W., Seely, J. C., Ward, S. M., Wilson, R. E., and Cunningham, M. C. (2000) Upper respiratory tract toxicity of inhaled methylvinyl ketone in F344 rats and B6C3F1 mice. Toxicol. Sci. 58, 182-194. (29) Furuhata, A., Nakamura, M., Osawa, T., and Uchida, K. (2002) Thiolation of protein-bound carcinogenic aldehyde. An electrophilic acrolein-lysine adduct that covalently binds to thiols. J. Biol. Chem. 277, 27919-27926. (30) Kaminskas, L. M., Pyke, S. M., and Burcham, P. C. (2004) Strong protein adduct trapping accompanies abolition of acrolein-mediated hepatotoxicity by hydralazine in mice. J. Pharmacol. Exp. Ther. 310, 1003-1010.

TX0502387