Formation of Acrolein-derived 2′-Deoxyadenosine Adduct in an Iron-induced Carcinogenesis Model*

Acrolein is a representative carcinogenic aldehyde found ubiquitously in the environment and formed endogenously through oxidation reactions, such as lipid peroxidation and myeloperoxidase-catalyzed amino acid oxidation. It shows facile reactivity toward DNA to form an exocyclic DNA adduct. To verify the formation of acrolein-derived DNA adduct under oxidative stress in vivo, we raised a novel monoclonal antibody (mAb21) against the acrolein-modified DNA and found that the antibody most significantly recognized an acrolein-modified 2′ -deoxyadenosine. On the basis of chemical and spectroscopic evidence, the major antigenic product of mAb21 was the 1,N6-propano-2′ -deoxyadenosine adduct. The exposure of rat liver epithelial RL34 cells to acrolein resulted in a significant accumulation of the acrolein-2′ -deoxyadenosine adduct in the nuclei. Formation of this adduct under oxidative stress in vivo was immunohistochemically examined in rats exposed to ferric nitrilotriacetate, a carcinogenic iron chelate that specifically induces oxidative stress in the kidneys of rodents. It was observed that the acrolein-2′ -deoxyadenosine adduct was formed in the nuclei of the proximal tubular cells, the target cells of this carcinogenesis model. The same cells were stained with a monoclonal antibody 5F6 that recognizes an acrolein-lysine adduct, by which cytosolic accumulation of acrolein-modified proteins appeared. Similar results were also obtained from myeloperoxidase knockout mice exposed to the iron complex, suggesting that the myeloperoxidase-catalyzed oxidation system might not be essential for the generation of acrolein in this experimental animal carcinogenesis model. The data obtained in this study suggest that the formation of a carcinogenic aldehyde through lipid peroxidation may be causally involved in the pathophysiological effects associated with oxidative stress.

Lipid peroxidation leads to the formation of a broad array of different products with diverse and powerful biological activities. Among them are a variety of different aldehydes (1). The primary products of lipid peroxidation, lipid hydroperoxides (2), can undergo carbon-carbon bond cleavage via alkoxyl radicals in the presence of transition metals, giving rise to the formation of short chain, unesterified aldehydes (2,3) or a second class of aldehydes still esterified to the parent lipid. These reactive aldehydic intermediates readily form covalent adducts with cellular macromolecules, including DNA, leading to disruption of important cellular functions and mutations. The important agents that give rise to the modification of DNA may be represented by ␣,␤-unsaturated aldehydic intermediates, such as 2-alkenals and 4-hydroxy-2-alkenals (4, 5). 2-Alkenals represent a group of highly reactive aldehydes containing two electrophilic reaction centers. A partially positive carbon 1 or 3 in such molecules can attack nucleophiles, such as protein and DNA, which leads to the formation of cyclic adducts or cross-links. Among all of the ␣,␤-unsaturated aldehydes, acrolein is by far the strongest electrophile (4). Acrolein is widely found in the environment and is formed in cells via lipid peroxidation (6). Its high reactivity indeed makes this aldehyde a dangerous substance for the living cell. A number of reports have appeared describing the damaging effects of acrolein on the tracheal ciliatory movement (7) and the pulmonary wall (8). It has also been shown that acrolein reduces the colony-forming efficiency of mammalian cells, forms cyclic adducts with nucleosides in vitro, and is a potent mutagen (4). Moreover, acrolein was shown to initiate urinary bladder carcinogenesis in rats (9).
We have studied the role of reactive aldehydes in the pathophysiological effects associated with oxidative stress. During the course of this study, we found that, upon reaction with protein, acrolein primarily reacted with lysine residues to form a N ⑀ -(3-formyl-3,4-dehydropiperidino)lysine (FDP-lysine) 1 adduct as the major product (6). In addition, using a monoclonal antibody (mAb5F6) against FDP-lysine, we demonstrated that the FDP-lysine adduct recognized by the antibody indeed constituted the atherosclerotic lesions, in which intense positivity was primarily associated with macrophage-derived foam cells (10). Based on these and the in vitro observations (6) that FDP-lysine was detected in the oxidatively modified low density lipoprotein with Cu 2ϩ and that a metal-catalyzed oxidation of arachidonate was associated with the formation of acrolein, we proposed that polyunsaturated fatty acids might represent the potential sources of acrolein under oxidative stress. On the other hand, Anderson et al. (11) have also shown that acrolein * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. could be formed from a myeloperoxidase (MPO)-catalyzed oxidation of threonine in vitro.
An iron chelate, ferric nitrilotriacetate (Fe 3ϩ -NTA), is known to be a strong inducer of renal proximal tubular necrosis (12). It also induces a high incidence of renal adenocarcinoma in rodents after repeated intraperitoneal administration of Fe 3ϩ -NTA (12)(13)(14). Oxidative stress has been suggested to play a critical role in the Fe 3ϩ -NTA-induced acute nephrotoxicity (15), which finally leads to a high incidence of renal adenocarcinoma in rodents (16). In the present study, to verify the formation of acrolein-derived DNA adduct under oxidative stress in vivo, we raised a new type of monoclonal antibody that specifically recognizes an acrolein-derived DNA adduct and examined the formation of this adduct in the Fe 3ϩ -NTA-induced carcinogenesis model. Furthermore, utilizing MPO knockout mice, we examined whether the MPO-catalyzed oxidation system was involved in the production of acrolein in this experimental animal carcinogenesis model.

Cell Culture
RL34 cells were obtained from the Japanese Cancer Research Resources Bank. The cell line, which is a nonmalignant epithelial cell, has been histologically and biochemically shown to possess characteristics of well differentiated liver parenchymal cells. The cells were grown as monolayer cultures in Dulbecco's modified Eagle's medium supplemented with 5% heat-inactivated fetal bovine serum, penicillin (100 units/ml), streptomycin (100 g/ml), L-glutamine (588 g/ml), and 0.16% NaHCO 3 at 37°C in an atmosphere of 95% air and 5% CO 2 . Cells postconfluency on glass coverslips was exposed to acrolein in Dulbecco's modified Eagle's medium containing 5% fetal bovine serum.

General Procedures
NMR spectra were recorded using a Bruker AMX400 (400 MHz) instrument. Fluorescence spectra were recorded with a Hitachi F-2000 spectrometer. Fast atom bombardment-mass spectrometry was measured with a JEOL JMS-700 (MStation) instrument. Liquid chromatography-mass spectrometry (LC-MS) was measured with a Jasco Platform II-LC instrument.

Identification of an Antigenic Acrolein-2Ј-Deoxyadenosine Adduct
The reaction mixture (100 l) of 2Ј-deoxyadenosine (2 mM) and acrolein (10 mM) was injected and fractionated (every 1 min) with a reversephase HPLC using a Develosil ODS-HG-5 column (4.6 ϫ 250 mm). The gradient (flow rate, 0.8 ml/min; solvent A ϭ 5% aqueous acetonitrile containing 0.05% acetic acid; solvent B ϭ acetonitrile containing 0.05% acetic acid) was as follows: 0 -5 min, 100% A; 5-32 min, linear gradient to 60% B; 32-40 min, linear gradient to 100% B; 40 -45 min, linear gradient to 100% A. The elution profiles were monitored by absorbance at 260 nm. The collected fractions were evaporated and dissolved in PBS, and the immunoreactivity with mAb21 was then examined by competitive enzyme-linked immunosorbent assay (ELISA). The major product in the immunoreactive fraction was isolated with a reversephase HPLC using a Develosil ODS-HG-5 column (20.0 ϫ 250 mm) equilibrated in 1% acetonitrile containing 0.05% acetic acid at a flow rate of 6.0 ml/min. The chemical structure of the product was characterized by mass spectrometry, 1 H and 13

Reaction of Calf Thymus DNA with Aldehydes
Calf thymus DNA (1 mg/ml) was mixed with aldehydes (10 mM) and incubated in phosphate buffer (pH 7.4) at 37°C for 24 h. The reaction was stopped by ethanol precipitation, and the precipitated DNA was washed with cold 70% ethanol. The amount of the DNA was determined by measuring the fluorescence intensity (excitation, 415 nm; emission, 505 nm) formed by the reaction of the 2Ј-deoxyribose moiety with 3,5-diaminobenzoic acid.

Preparation of a Monoclonal Antibody against
Acrolein-modified DNA The acrolein-treated calf thymus DNA (0.5 mg/ml) was electrostatically complexed with an equal volume of methylated BSA (0.5 mg/ml) (17). The complex was emulsified with an equal volume of complete Freund's adjuvant. Six-week-old female Balb/c mice were intraperitoneally immunized with this emulsion (100 l). After 2 weeks, the mice were boosted with the acrolein-treated DNA/methylated BSA emulsified with an equal volume of incomplete Freund's adjuvant. In the final boost, 100 l of acrolein-treated DNA/methylated BSA in PBS was intravenously injected. Three days after the final boost, the mouse was sacrificed, and the spleen was removed for fusion with P3/U1 myeloma cells. The fusion was carried out by polyethylene glycol, and the cells were cultured in hypoxanthine/aminopterin/thymidine selection medium. After a week, culture supernatants of the hybridomas were screened by ELISA using acrolein-modified DNA and untreated DNA as the coating agents. The positive hybridomas were cloned by the limiting dilution method. After repeated screening and cloning, four specific clones were obtained. Among them, a clone (named mAb21) was used in the following experiment because of its specificity and great ability for cell growth.

ELISA
The indirect noncompetitive ELISA procedure has been previously described (18). Briefly, 100 l of antigens in PBS were coated in wells and kept at 4°C overnight. After washing and blocking with 4% Block Ace (Dainihon Seiyaku, Osaka, Japan), 100 l of the mAb21 (1:100 dilution) in PBS containing 0.05% Tween 20 (TPBS) was then added, and the wells were incubated at 37°C for 2 h. After washing, 100 l of anti-mouse IgG goat antibody peroxidase labeled (1/5000 in TPBS) was added, and the wells were incubated at 37°C for 1 h. After washing, 100 l of o-phenylenediamine solution (0.5 mg/ml containing 0.03% H 2 O 2 in citrate-phosphate buffer) was added. The color developing reaction was stopped by the addition of 50 l of 2 N H 2 SO 4 . The binding of the antibody to the antigen was evaluated by measuring the absorbance at 492 nm.
Competitive indirect ELISA was performed for estimating the crossreactivity of the low molecular weight compounds with the antibody.
The acrolein-modified calf thymus DNA (0.05 g/well) was used as the coating antigen. For the competitive reaction, 50 l of competitors in PBS was mixed with an equal volume of the antibody (1:100 dilution) in PBS containing 4% BSA. The competitor solution was kept at 4°C overnight, and 90 l of the mixture was used as the primary antibody. The cross-reactivity of the antibody to the competitors was expressed as B/B 0 , where B is the amount of mAb21 bound to the coating antigen in the presence of the competitor, and B 0 is the amount in the absence of a competitor.

Immunocytochemical Detection of Acrolein Adducts
Acrolein-2Ј-Deoxyadenosine Adduct-The cells were fixed overnight in PBS containing 2% paraformaldehyde and 0.2% picric acid at 4°C. The membranes were permeabilized by exposing the fixed cells to PBS containing 0.3% Triton X-100. To eliminate the binding of the antibody to the RNA containing acrolein-adenosine adduct, coverslips were treated with 100 g/ml ribonuclease A (Qiagen) in TEN buffer (10 mM Tris-HCl, pH 7.4; 1 mM EDTA, pH 7.6; 400 mM NaCl) for 60 min at 37°C. The coverslips were then washed once in PBS and incubated with 10 g/ml proteinase K (Wako) in 100 mM Tris-HCl, pH 7.5, and 10 mM EDTA for 10 min at room temperature. To provide maximal exposure of mAb21 to regions of DNA where acrolein-2Ј-deoxyadenosine lesions may exist, the DNA was denatured at room temperature with 2 M HCl for 5 min and then neutralized with 2.5 volumes of 1 M Tris-base. To prevent nonspecific antibody binding, the coverslips were washed two times in PBS and blocked for 1 h at room temperature with 5% BSA in TPBS. The cells were then incubated in primary antibody (mAb21) at 4°C overnight. The cells were then incubated for 1 h in the presence of fluorescein isothiocyanate (FITC)-labeled anti-mouse IgG (Dako Japan Co., Ltd., Kyoto, Japan), rinsed with PBS containing 0.3% Triton X-100, and mounted on glass slides using 50% glycerol in PBS. Images of cellular immunofluorescence were acquired using a Zeiss LSM5 PASCAL confocal laser scanning microscope with a 40ϫ objective (excitation, 488 nm; emission, 518 nm). The DNA was stained with propidium iodide, while the damage was co-localized using a FITClabeled secondary antibody.
Acrolein-Lysine Adduct-The cells were fixed overnight in PBS containing 2% paraformaldehyde and 0.2% picric acid at 4°C. The membranes were permeabilized by exposing the fixed cells to PBS containing 0.3% Triton X-100. The cells were then sequentially incubated in PBS solutions containing blocking serum (5% normal goat serum) and immunostained with anti-protein-bound acrolein mAb5F6 (10). The cells were then incubated for 1 h in the presence of FITC-labeled anti-mouse IgG (Dako Japan Co.), rinsed with PBS containing 0.3% Triton X-100 and were mounted on glass slides using 50% glycerol in PBS. Images of cellular immunofluorescence were acquired using a Zeiss LSM5 PASCAL confocal laser scanning microscope with a 40ϫ objective (excitation, 488 nm; emission, 518 nm). DNA was stained with propidium iodide, while the damage was co-localized using a FITC-labeled secondary antibody.

Animal Experiments
Male Wistar rats (Shizuoka Laboratory Animal Center, Shizuoka) weighing 130 -150 g (6 weeks old) and mice lacking MPO (MPOϪ/Ϫ mice), the sixth-generation progeny of a backcross into C57BL/6J mice (B6 mice) originally generated by Aratani et al. (19), were used. The Fe 3ϩ -NTA solution was prepared immediately before use by the method described by Toyokuni et al. (15) with a slight modification. Briefly, ferric nitrate enneahydrate and the nitrilotriacetic acid disodium salt were each dissolved in deionized water to form 80 and 160 mM solutions, respectively. They were mixed at the volume ratio of 1:2 (molar ratio, 1:4), and the pH was adjusted with sodium hydrogen carbonate to 7.4. The animals received a single intraperitoneal injection of Fe 3ϩ -NTA (15 mg of iron/kg of body weight). They were sacrificed by decapitation at 0, 24, and 48 h after the administration. Both kidneys of each animal were immediately removed. One of them was fixed in Bouin's solution, embedded in paraffin, cut at 3-m thickness, and used for immunohistochemical analyses by an avidin-biotin complex method with alkaline phosphatase. Briefly, after deparaffinization with xylene and ethanol, normal rabbit serum (Dako; diluted to 1:75) for the inhibition of the nonspecific binding of the secondary antibody, a primary antibody (0.5 g/ml), biotin-labeled rabbit anti-mouse IgG serum (Vector Laboratories; diluted to 1:300), and avidin-biotin complex (Vector; diluted to 1:100) were sequentially used. The procedures using PBS or the IgG fraction (0.5 g/ml) of normal mouse serum instead of mAb21 showed no or negligible positive responses.

Formation of Acrolein during Peroxidation of PUFAs-Based
on the in vitro observations that the acrolein-lysine adduct (FDP-lysine) was detected in the oxidatively modified low density lipoprotein with Cu 2ϩ , PUFAs were suggested to represent the potential sources of acrolein under oxidative stress (6,10). To evaluate the effectiveness of lipid peroxidation as the origin of acrolein, we first examined the generation of acrolein during in vitro peroxidation of PUFAs, including linoleate, arachidonate, linolenate, cis -5,8,11,14,17-eicosapentaenoic acid, and cis-4,7,10,13,16,19-docosahexaenoic acid, by a reversed-phase HPLC following derivatization with DNPH. As shown in Fig.  1A, the HPLC analysis of the authentic acrolein showed that the DNPH derivative of acrolein was eluted at 32 min (chromatogram a). The same peak indicated by the arrow was detected in the UV chromatogram obtained from the HPLC analysis of the peroxidized arachidonate with Fe 2ϩ /ascorbate (chromatogram b). Formation of acrolein was also confirmed by LC-MS analysis following derivatization with biotin-LC-hydrazide (Fig. S1). A time course analysis of acrolein generation in the metal-catalyzed peroxidation of PUFAs revealed that significant amounts of acrolein were generated during peroxida-  tion (Fig. 1B), suggesting that lipid peroxidation could represent the major source of acrolein under oxidative stress.
Preparation of a Monoclonal Antibody against Acrolein-modified DNA-It has been suggested that acrolein modification of DNA may contribute to so-called spontaneous mutagenesis, thereby playing a role in aging and cancer. Hence, to examine whether acrolein endogenously generated under oxidative stress is involved in the formation of acrolein-DNA adduct, we attempted to detect the acrolein-modified DNA by an immunochemical procedure. To raise a monoclonal antibody specific to acrolein-modified DNA, mice were immunized with the acrolein-modified calf thymus DNA, and after the fusion, the hybridomas were selected by the reactivities of the culture supernatant to the immunogen. We finally obtained one clone (clone No. 21), which exhibited the most distinctive recognition of the acrolein-modified DNA over native DNA (Fig. 2A). As shown in Fig. 2B, incubation of calf thymus DNA (1 mg/ml) with 10 mM acrolein resulted in a time-dependent increase in immunoreactivity with the monoclonal antibody 21 (mAb21). Despite extensive screening of hybridomas, which produce monoclonal antibodies specific to the acrolein-modified DNA, it is still conceivable that the antibody recognizes epitopes originating from other reactive aldehydes. Hence, we examined the immunoreactivity of the antibody to aldehyde-treated DNA by a direct ELISA and found that among the aldehydes tested, acrolein was the only source of immunoreactive structures generated in the DNA (Fig. 2C). It is notable that the antibody scarcely reacted with the DNA treated with an acrolein analogue, crotonaldehyde. In addition, the binding of the acroleinmodified DNA to mAb21 was scarcely inhibited by the reaction mixtures of acrolein/2Ј-deoxyguanosine, acrolein/2Ј-deoxythymidine, and acrolein/2Ј-deoxycytidine but was most signifi-cantly inhibited by the reaction mixture of acrolein/2Ј-deoxyadenosine (Fig. 2D). In addition, to attest the selectivity observed for free acrolein-nucleoside adducts is identical with the situation in double-stranded DNA, we examined the immunoreactivity of the acrolein-modified oligonucleoside 50-mers (dA 50 , dG 50 , and dC 50 ) by a direct ELISA and found that the antibody reacted only with the acrolein-modified dA 50 (Fig. S2). These data strongly suggest that mAb21 selectively recognizes an acrolein-2Ј-deoxyadenosine adduct as the major epitope.
Identification of an Antigenic Acrolein-2Ј-Deoxyadenosine Adduct-To identify an acrolein-2Ј-deoxyadenosine adduct recognized by mAb21, the immunoreactivity with the reaction products of acrolein with 2Ј-deoxyadenosine was characterized. As shown in Fig. 3A, the ELISA analysis of the HPLC fractions for immunoreactivity with mAb21 showed that the antibody had immunoreactivity with only one fraction (product a) eluted at 4 min. The structure of the antigenic product a was characterized by LC-MS, UV, and 1 H and 13 C NMR. The LC-MS analysis of the product showed a pseudomolecular ion peak at m/z 308 (MϩH) ϩ , suggesting that it was composed of one molecule of 2Ј-deoxyadenosine and one molecule of acrolein. Finally, the 1 H and 13 C NMR data (see "Experimental Procedures") determined the antigenic adduct to be 1,N 6 -propano-2Јdeoxyadenosine (Fig. 3B), which was previously identified as the major acrolein-2Ј-deoxyadenosine adduct by Smith et al. (20). The isolated 1,N 6 -propano-2Ј-deoxyadenosine inhibited the antibody binding to the coated antigen in a dose-dependent manner (Fig. 3C). The reaction of acrolein with other 2Ј-deoxynucleosides such as 2Ј-deoxycytidine and 2Ј-deoxy- guanosine also gave similar propano-type adducts as the major products, but they scarcely inhibited the binding of the acrolein-modified DNA to mAb21 (Fig. S3). These data suggest that the acrolein-2Ј-deoxyadenosine adduct (1,N 6 -propano-2Ј-deoxyadenosine) is an intrinsic epitope of mAb21.
Formation of Acrolein-2Ј-Deoxyadenosine Adduct in Rat Liver Epithelial RL34 Cells Exposed to Acrolein-It is known that acrolein is extremely toxic to living cells (4). As shown in Fig. 4A, the treatment of rat liver epithelial RL34 cells with acrolein resulted in dose-and time-dependent decreases in MTT reduction levels. Essentially all of the cells were detached from culture dishes within 8 h at a dose of 50 M acrolein (data not shown). The acrolein-mediated cell death was accompanied by marked changes in cellular morphology, most notably the swelling of the cells, which changes were evident in phase contrast microscopy. It was anticipated that the cytotoxicity of acrolein was due to its facile reactivity toward cellular macromolecules, including protein and DNA. To examine the correlation between cytotoxicity and the generation of these acrolein-derived adducts, we tested whether the acrolein-2Јdeoxyadenosine adduct was formed in cultured cells in vitro. As shown in Fig. 4B, when the cells were exposed to acrolein (50 M) for 1 h, immunoreactivity with mAb21 was mainly observed in the nuclei. In addition, we also observed that acrolein generated the epitopes recognized by an antibody (mAb5F6) raised against acrolein-modified protein (Fig. 4C). Thus, the acrolein modification of the protein and DNA seems to be closely associated with the cytotoxicity of acrolein. Meanwhile, we have also observed that acrolein induces depletion of glutathione and activation of mitogen-activated protein kinases, such as c-Jun N-terminal kinases and p38, in RL34 cells. 2 These cellular events might be associated with apoptotic cell death. Indeed, Li et al. (21) have reported that acrolein, at lower doses, significantly induces apoptosis of human alveolar macrophages. These observations suggest that there may be an alternative mechanism of acrolein-induced cell death, independent of modification of protein and DNA.
Formation of Acrolein-2Ј-Deoxyadenosine Adduct in the Renal Proximal Tubules of Rats Exposed to Fe 3ϩ -NTA-The formation of acrolein-derived epitopes in vivo was immunohistochemically assessed in a rat renal carcinogenesis model with Fe 3ϩ -NTA. The monoclonal antibodies against the acroleinlysine adduct (mAb5F6) and 8-oxo-2Ј-deoxyguanosine (mAbN45.1) were also used for comparison. It has been shown that iron overload using Fe 3ϩ -NTA induces acute renal proximal tubular necrosis, a consequence of oxidative tissue damage, that eventually leads to a high incidence of renal adenocarcinoma in rodents (13,14). The kidneys were excised at the time of sacrifice and then fixed with Bouin's fixative. The hematoxylin and eosin-stained sections of the paraffinembedded tissues were analyzed for histological damage. The morphological changes in the kidneys of rats treated with Fe 3ϩ -NTA versus time are very similar to previous reports on ddY mice (15,22). As shown in Fig. 5A, untreated control animals showed weak immunoreactivity in all of the proximal tubules. One hour after the administration of Fe 3ϩ -NTA (15 mg of iron/kg of body weight), nuclear stainings were found in some of the renal proximal tubular cells in a patchy distribution. Intense immunoreactivities were also observed at 3 and 24 h. Notably, the proximal tubular cells at higher magnification revealed the localization of the acrolein-DNA adduct in the nuclei of degenerating cells (Fig. 5B), suggesting that the formation of acrolein and its DNA adduct may be responsible for the cellular damage. Preabsorption of the antibodies with the acrolein-modified DNA completely abolished the immunostainings (data not shown), indicating the specific reactivity of these antibodies with their epitopes. The data are expressed as the percentages of control culture conditions. B, immunocytochemical detection of acrolein-2Ј-deoxyadenosine adduct in RL34 cells exposed to acrolein. The cells were incubated with 50 M acrolein for 1 h at 37°C. Panels a-c, control; panels d-f, acrolein-treated cells. Propidium iodide (PI) (DNA, red) is shown in panels a and d, FITC (acrolein-2Ј-deoxyadenosine, green) is shown in panels b and e, and the corresponding combined (superimposed) images are shown in panels c and f (yellow represents co-localization). C, immunocytochemical detection of acrolein-lysine adduct in RL34 cells exposed to acrolein. The cells were incubated with 100 M acrolein for 1 h at 37°C. Panels a-c, control; panels d-f, acrolein-treated cells. Propidium iodide (DNA, red) is shown in panels a and d, FITC (acrolein-lysine, green) is shown in panels b and e, and the corresponding combined (superimposed) images are shown in panels c and f (yellow represents co-localization).
The pattern of distribution of the acrolein-2Ј-deoxyadenosine adduct was compared with those of the acrolein-lysine adduct and 8-oxo-2Ј-deoxyguanosine. As shown in Fig. 6, the acroleinlysine immunoreactivities also appeared in some of the renal proximal tubular cells 1 h after the administration of Fe 3ϩ -NTA, whereas their patterns of distribution appeared to be significantly different. In contrast to the nuclear staining with mAb21, the acrolein-lysine adduct immunoreactive with mAb5F6 was mainly detected in the cytoplasm. This pattern of distribution in the rat kidney was consistent with that of the distribution of other lipid peroxidation products and their conjugates with cytosolic proteins (15,23,24). In addition, these and the observation that the acrolein-derived adducts co-localized with 8-oxo-2Ј-deoxyguanosine, a marker of oxidative stress, strongly suggest that acrolein was generated under oxidative stress.
MPO-catalyzed Oxidation of Threonine as an Alternative Source of Acrolein-On the other hand, it has been demonstrated that acrolein is also generated from the MPO-catalyzed oxidation of an amino acid (threonine) in vitro (11). In addition, MPO appeared to catalyze lipid peroxidation (25). These and the observations that neutrophil and macrophage infiltration is present in the kidney of Fe 3ϩ -NTA-treated rats 3 suggest that MPO might play a potential role in the generation of acrolein in this iron-induced carcinogenesis model. Consistent with the previous findings, acrolein was detected in the MPO-catalyzed oxidation of threonine as the major product (Fig. 7A). A maximal generation of acrolein was observed at 6 h followed by a gradual decline to 24 h was observed in the MPO-H 2 O 2 -Cl Ϫ system. This transient accumulation of acrolein in the MPOcatalyzed threonine oxidation system was attested to be due to the secondary reaction of the primary product (acrolein) to the substrate (threonine) present in the reaction mixture (data not shown). These data supported the previous findings that MPOcatalyzed threonine oxidation could be a potential source of acrolein in vivo. To investigate whether MPO is involved in the generation of acrolein in the Fe 3ϩ -NTA-induced carcinogenesis model, the MPO-deficient mice were exposed to Fe 3ϩ -NTA, and the formation of the acrolein-derived DNA adduct was immunohistochemically examined. The Fe 3ϩ -NTA potently induced lipid peroxidation monitored by 2-thiobarbituric acid reactive substances (TBARS) formation in the kidneys of MPO-deficient mice (Fig. 8A). The amount of TBARS reached the peak at 1 h after administration of Fe 3ϩ -NTA (5 mg Fe/kg) and gradually decreased thereafter. As shown in Fig. 8B, the acrolein-2Ј-3 S. Toyokuni, unpublished observation.

FIG. 5.
Formation of acrolein-2-deoxyadenosine adduct in the renal proximal tubules of rats exposed to Fe 3؉ -NTA. A, time-dependent formation of acrolein-2Ј-deoxyadenosine adduct (ϫ200). B, higher magnification of the proximal tubules of rats exposed to Fe 3ϩ -NTA for 3 h (ϫ400). deoxyadenosine immunoreactivities appeared in the kidneys of MPO-deficient mice exposed to Fe 3ϩ -NTA. These data provide direct evidence that MPO is not essential for the generation of acrolein, at least, in this experimental model of carcinogenesis. DISCUSSION The previous observations that the metal-catalyzed oxidation of low density lipoprotein is associated with the formation of the apo B-bound acrolein suggested that substantial amounts of acrolein might be generated during the peroxidation of polyunsaturated fatty acids (6). In the present study, we also observed that the peroxidation of polyunsaturated fatty acids with an iron/ascorbate-mediated free radical-generating system produces multiple products that react with DNPH, as monitored by reverse-phase HPLC, and one of the products co-migrates with the DNPH derivative of authentic acrolein (Fig. 1). These data strongly suggest that the peroxidation of polyunsaturated fatty acids represents a potential endogenous pathway for the production of acrolein. Although the mechanism for the formation of acrolein during lipid peroxidation has not yet been experimentally resolved, it is not unlikely that acrolein represents a major causal factor that contributes to the development of tissue damage under oxidative stress.
Among all of the ␣,␤-unsaturated aldehydes, acrolein exhibits the greatest reactivity with nucleophiles, such as proteins (5). Upon reaction with protein, acrolein selectively reacts with the side chains of the cysteine, histidine, and lysine residues.
Of these, lysine generates the most stable product. The ␤-substituted propanals and Schiff's base cross-links had been suggested as the predominant adduct; however, the major adduct formed upon the reaction of acrolein with protein was identified as a novel lysine product, FDP-lysine, which requires the attachment of two acrolein molecules to one lysine side chain (6). In addition, we have recently shown that the FDP-lysine adduct generated in the acrolein-modified protein represents a thiol-reactive electrophile (26). The formation of a similar FDPtype adduct (dimethyl-FDP-lysine) has been reported in the lysine modification with the acrolein analogue, crotonaldehyde (23). In our previous study, we obtained a murine monoclonal antibody, mAb5F6, that clearly distinguished the acroleinmodified BSA and oxidized low density lipoprotein from native BSA and low density lipoprotein and showed that it recognized the acrolein-lysine adduct (FDP-lysine) as the major epitope (6). Using this antibody, it was clearly demonstrated that atherosclerotic lesions contained antigenic materials in the granular cytoplasmic elements of foam cells and the thickening neointima of arterial walls (6). In addition, the presence of antigenic materials has also been observed in the nigral neurons of patients with Parkinson's disease (27), in which oxidative stress and mitochondrial respiratory failure with a resultant energy crisis have been implicated as two major contributors to nigral neuronal death.
It has been shown that upon reaction with DNA, acrolein primarily reacts with 2Ј-deoxyguanosine and forms the 1,N 2 -  8. Formation of acrolein-derived adducts in the renal proximal tubules of MPO-deficient mice exposed to Fe 3؉ -NTA. A, time-dependent formation TBARS in the kidney of MPO-deficient mice exposed to Fe 3ϩ -NTA. B, formation of acrolein-2Ј-deoxyadenosine adduct in the kidney of MPO-deficient mice exposed to Fe 3ϩ -NTA for 1 h. Formation of acrolein-lysine and acrolein-2Ј-deoxyadenosine (acrolein-dA) immunoreactive with mAb5F6 and mAb21, respectively, was immunohistochemically examined (ϫ200).
propano-2Ј-deoxyguanosine adduct (28 -30). In later studies, the 1,N 2 -propano-2Ј-deoxyguanosine adduct was detected in the DNA extracted from rat and human liver (31)(32)(33) and from lymphocyte DNA in patients undergoing treatment with cyclophosphamide, a precursor of acrolein (34,35). It is striking to note that the acrolein-derived 2Ј-deoxyguanosine adducts are mutagenic in vitro and in vivo (36,37). This may be correlated to our previous findings that the acrolein adduct (FDP-lysine) was markedly generated in the livers of rats exposed to the carcinogenic chemicals, such as butylated hydroxyanisole, 4-dimethylaminoazobenzene, and 2-acetylaminofluorene (38). In addition, we confirmed the potency of acrolein to promote the transformation of benzo[a]pyrene-treated C3H/10T1/2 cells in vitro (38). These observations raised the possibility that acrolein generation under oxidative stress could be closely associated with carcinogenesis. Furthermore, the causality of acrolein in the Fe 3ϩ -NTA-induced oxidative tissue damage may also be suggested by our previous observation that the Fe 3ϩ -NTA exposure to rats dramatically induced gene expression of GST P1 subunit in the proximal tubular cells (39). Berhane et al. (40) have shown that GST P1 is highly active in catalyzing the reactions with ␣,␤-unsaturated aldehydes, particularly with acrolein. In addition, these authors have shown that the GST P1 isozyme introduced in the cells by electroporation increase their resistance to acrolein. Thus, it can be hypothesized that the induction of GST P1 in the proximal tubular cells, the target organs of Fe 3ϩ -NTA, may represent the cellular response to the formation of acrolein and its DNA adduct.
To verify the formation of acrolein and to investigate the potential role of acrolein modification of DNA in vivo, an experimental model of iron overload using Fe 3ϩ -NTA was utilized. This iron chelate was originally used for an experimental model of iron overload (41). Later, repeated injections of Fe 3ϩ -NTA were reported to induce acute and subacute renal proximal tubular necrosis and a subsequent high incidence (60 -92%) of renal adenocarcinoma in male rats and mice (13,14). A single injection of Fe 3ϩ -NTA causes a number of time-dependent morphological alterations in the structure and the function of the renal proximal tubular cells and their mitochondria. During the early stage of the injury, typical cellular changes are the loss of brush border, cytoplasmic vesicles, mitochondrial disorganization, and dense cytoplasmic deposits in the proximal tubular cells. Most of the damaged epithelia show the typical appearance of necrotic cells, and more than half of the proximal tubular cells are removed. It has been suggested that oxidative stress is one of the basic mechanisms of Fe 3ϩ -NTAinduced acute renal injury and is closely associated with renal carcinogenesis (16). To detect the acrolein-derived DNA adducts in this carcinogenesis model, an immunochemical approach has been taken in this study. Immunologic detection is a powerful tool that can be used to evaluate the presence of a desired target and its subcellular localization. A major advantage of this technique over biochemical approaches is the evaluation of small numbers of cells or archival tissues that may otherwise not be subject to analysis. We successfully raised a murine monoclonal antibody, mAb21, that clearly distinguished the acrolein-modified DNA from the native DNA. This antibody appeared to be highly specific for the acrolein-modified DNA and did not cross-react with other aldehyde-modified DNA (Fig. 2). In addition, we identified the 1,N 6 -propano-2Јdeoxyadenosine as the major epitope of mAb21 (Fig. 3). Using this antibody, we demonstrated the formation of the acrolein-2Ј-deoxyadenosine adduct in rat liver epithelial RL34 cells exposed to acrolein (Fig. 4). Furthermore, the in vivo study demonstrated that the Fe 3ϩ -NTA treatment resulted in a significant accumulation of the acrolein adduct in the nuclei of the proximal tubular cells, the target cells of this carcinogenesis model (Fig. 5). The same cells were also stained with an alternative mAb5F6 that recognizes an acrolein-lysine adduct (FDP-lysine), by which cytosolic accumulation of acrolein-modified proteins appeared (Fig. 6). As far as we know, this is the first report that immunohistochemically proved the in vivo formation of acrolein-derived DNA adducts in the target organ of the carcinogenic protocol.
On the other hand, it has been suggested that MPO in the presence of H 2 O 2 and Cl Ϫ is an alternative source of acrolein in vivo (11). MPO is an enzyme found mainly in neutrophils and to a lesser degree in monocytes (42). In chemoattractant-activated neutrophils, MPO transforms H 2 O 2 generated during the oxidative burst into highly cytotoxic hypochlorous acid (HOCl) in the presence of chloride ions (43). This MPO-H 2 O 2 -Cl Ϫ system appears to function in the killing of microbes by neutrophils (19,44). It may also be involved in their cytotoxicity against tumor cells (45,46) and in the tissue damage at sites of inflammation where neutrophils can release both MPO and H 2 O 2 (47)(48)(49)(50). It has been proposed that the pathway for acrolein synthesis involves the initial chlorination of the ␣-amino group of threonine by HOCl (11). In the absence of the enzymatic system, HOCl alone could convert threonine into acrolein, strongly implicating HOCl in the MPO-H 2 O 2 -Cl Ϫ system (11). We have also examined the generation of acrolein in the threonine oxidation and observed that, in contrast to the peroxidation of polyunsaturated fatty acids, the MPO-H 2 O 2 -Cl Ϫ system transiently generated acrolein, which was proven to be due to the secondary reaction of the product (acrolein) with the catalyst (MPO) and/or the substrate (threonine) present in the reaction mixture (Fig. 7). Thus, although nonenzymatic lipid peroxidation and MPO-catalyzed threonine oxidation showed considerably distinct profiles in kinetics, these data supported the previous findings that both systems could be potential origins of acrolein in vivo. To investigate whether MPO is involved in the generation of acrolein in the Fe 3ϩ -NTA-induced carcinogenesis model, the MPO-deficient mice were exposed to Fe 3ϩ -NTA, and the formation of acrolein-derived DNA adducts was immunohistochemically examined. Fe 3ϩ -NTA induced lipid peroxidation monitored by TBARS formation in the kidneys of MPO-deficient mice (Fig. 8A). The amount of TBARS reached the peak at 1 h after the administration of Fe 3ϩ -NTA and gradually decreased thereafter. In parallel with the TBARS formation, the acrolein-2Ј-deoxyadenosine adducts were significantly accumulated in the kidneys of MPO-deficient mice exposed to Fe 3ϩ -NTA for 1 h (Fig. 8B). These data provide direct evidence that MPO is not essential for the generation of acrolein, at least in this experimental model of renal carcinogenesis.
In summary, to verify the acrolein generation under oxidative stress in vivo, we raised a new type of monoclonal antibody (mAb21) against the acrolein-modified DNA and found that the antibody most significantly recognized the 1,N 6 -propano-2Јdeoxyadenosine adduct. The formation of this adduct was attested to in rat liver epithelial RL34 cells exposed to acrolein and in rats exposed to Fe 3ϩ -NTA, a carcinogenic iron chelate that specifically induces oxidative stress in the kidney of rodents. These data suggest that the formation of a carcinogenic aldehyde during lipid peroxidation may be causally involved in the pathophysiological effects associated with oxidative stress.