Mass Spectroscopic Characterization of Protein ... - ACS Publications

122

Chem. Res. Toxicol. 2006, 19, 122-129

Mass Spectroscopic Characterization of Protein Modification by Malondialdehyde Takeshi Ishii,‡,| Shigenori Kumazawa,*,‡,§ Toyo Sakurai,| Tsutomu Nakayama,§ and Koji Uchida| Graduate School of Bioagricultural Sciences, Nagoya UniVersity, Chikusaku, Nagoya 464-8601, Japan, and Department of Food and Nutritional Sciences, UniVersity of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan ReceiVed August 19, 2005

Malondialdehyde (MDA), a naturally occurring dialdehyde produced in the membrane by lipid peroxidation, is a strong alkylating agent of primary amino groups. We recently raised a monoclonal antibody (mAb1F83) directed to the lipofuscin-like MDA-lysine adduct and demonstrated the presence of immunoreactivity to the antibody in the atherosclerotic lesions, in which intense positivity was associated primarily with macrophage-derived foam cells (Yamada et al., (2001) J. Lipid Res. 42, 1187-1196). To identify the structure of the epitope in the protein recognized by mAb1F83, in the present study, we exposed chain B from bovine insulin (insulin B chain) to MDA and characterized the MDA adducts by mass spectrometry. The MDA-modified insulin B chain was digested with V8 protease, and the resulting peptides were subjected to liquid chromatography-electrospray ionization-mass spectrometry (LCESI-MS/MS). The MS/MS analyses confirmed the formation of N-propenal- (+54 Da) and dihydropyridine-type (DHP, +134 Da) adducts in both Lys29 and the N-terminus of insulin B chain. The ELISA analysis of HPLC fractions of peptides, including the DHP adducts using mAb1F83, showed that the immunoreactivity of the DHP-lysine adduct was more significant than the DHP-N-terminus adduct. The results of this study chemically characterized that the MDA adducts such as DHP-type adducts generated in the -amino group of lysine and N-terminal amino acid residues in the protein and the structure of the epitope recognized by mAb1F83 were DHP-lysine adducts in protein. Introduction The oxidative modification of protein and subsequent accumulation of the modified proteins have been found in cells during aging, oxidative stress, and in various pathological states including premature diseases, muscular dystrophy, rheumatoid arthritis, and atherosclerosis (1, 2). Important agents that give rise to the modification of a protein may be represented by reactive aldehydic intermediates, such as keto aldehydes, 2-alkenals, and 4-hydroxy-2-alkenals (3, 4). These reactive aldehydes are considered important mediators of cell damage due to their ability to covalently modify biomolecules, which can disrupt important cellular functions and can cause mutations (3). Furthermore, the modification of aldehydes to apolipoprotein B in low-density lipoproteins (LDL)1 has been strongly implicated in the mechanism by which LDL is converted to an atherogenic form that is taken up by macrophages, leading to the formation of foam cells (5, 6). Malondialdehyde (MDA) is often the most abundant individual aldehyde resulting from lipid peroxidation and occurs * To whom correspondence should be addressed: Shigenori Kumazawa, Department of Food and Nutritional Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan. Tel: 81-54-264-5523. Fax: 81-54-264-5523. E-mail: [email protected]. ‡ These authors contributed equally to this work. | Nagoya University. § University of Shizuoka. 1 Abbreviations: MDA, malondialdehyde; DHP, dihydropyridine; ELISA, enzyme-linked immunosorvent assay; HPLC, high-performance liquid chromatography; insulin B chain, chain B from bovine insulin; LC-ESIMS, liquid chromatography-electrospray ionization-mass spectrometry; PBS, phosphate-buffered saline; LDL, low-density lipoproteins; V8 protease, Staphylococcus aureus V8 protease.

Figure 1. Chemical structures of N-propanal-, and DHP-type MDA adducts. The numbers represent the increments of molecular weight as the adducts are formed.

in biological materials in various covalently bound forms (7-11). MDA primarily forms adducts with lysine residues of proteins or with amine headgroups of phospholipids, such as phosphatidylserine and phosphatidylethanolamine, and reacts with DNA bases to produce adducts of deoxyguanosine, deoxyadenosine, and deoxycytidine (12-14). The major reaction of MDA comprises addition to lysine residue, generating N(2-propenal)lysine (N-propenal-Lys) (Figure 1) (9, 15). This adduct has been detected as the major form in which endogenous MDA is excreted in urine in rats and humans (9, 10). MDA also forms some fluorescent products such as the dihydropyridine (DHP)-type adducts as shown in Figure 1, a model of fluorescent components in lipofuscin (16-18). Other fluorescent products including N,N′-disubstituted 1-amino-3-iminopropenetype and pyridyl DHP-type lysine-lysine cross-link have been also reported (7, 19). More recently, a new cross-link and unique histidin adduct has been identified in bovine serum albumin incubated with MDA (20). Mass spectrometry (MS) is a important analytical technique for the structural characterization of proteins. MS, in particular the application of electrospray ionization (ESI) coupled on-line

10.1021/tx050231p CCC: $33.50 © 2006 American Chemical Society Published on Web 12/06/2005

Characterization of Protein Modification by Malondialdehyde

with high-performance separation techniques such as highperformance liquid chromatography (HPLC), has had a dramatic effect on the sensitivity and the speed with which the primary structure of proteins can be determined. Many applications, such as verification of the expression of the protein sequence and identification of post-translational modifications, have been reported (21-23). Previously, we raised a monoclonal antibody (mAb1F83) directed to the MDA-modified protein and identified a lipofuscin-like fluorophore as the major epitope (24). This antibody was used to conclusively demonstrate that the fluorophore forms on oxidatively modified low-density lipoproteins (LDL). In addition, we demonstrated that the materials immunoreactive to mAb1F83 indeed constituted the atherosclerotic lesions, in which intense positivity was associated primarily with macrophage-derived foam cells. However, the MDA adducts such as DHP-lysine generated in the protein and the structure of the epitope in the protein recognized by mAb1F83 have not been chemically characterized. The purpose of this study was to identify and to further characterize the protein modifications formed upon MDA treatment by mass spectroscopic techniques. Thus, a model peptide, chain B from bovine insulin (insulin B chain), was exposed to MDA, and the modified protein was digested with Staphylococcus aureus V8 protease (V8 protease) and analyzed by electrospray ionization-liquid chromatography/mass spectrometry/mass spectrometry (ESI-LC/MS/MS) to characterize the modification behavior of MDA to protein.

Experimental Procedures Materials. The sodium salt of MDA (3-hydroxy-2-propenal, sodium salt) was prepared by hydrolysis of malondialdehyde bis (diethyl acetal) (25). Anti-MDA-lysine antibodies that recognize lipofuscin-like fluorophore epitopes were raised by immunizing a mouse with lipofuscin-like MDA-lysine adducts treated with keyhole limpet hemocyanin as previously reported (24). Horseradish peroxidase-linked anti-rabbit IgG immunoglobulin was purchased from Amersham Pharmacia Biotech U.K., Ltd. (Buckinghamshire, U.K.). Insulin B chain was obtained from Sigma Chemical Co. (St. Louis, MO). BlockAce was obtained from Snow Brand Milk Products Co., Ltd. (Hokkaido, Japan). V8 protease, Tween 20, and 1,2-phenylenediamine were purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Reaction of Insulin B Chain with MDA. Insulin B chain (1 mg/mL) was incubated in a 1:35 molar ratio with 1 mM MDA in 1 mL of 50 mM sodium phosphate buffer (pH 7.2) at 37 °C for 24 h. The reaction was terminated by centrifugal filtration (Microcon 3, molecular weight cutoff of 3000; Millipore) to remove the lowmolecular weight reactants. Control experiments were performed under the same conditions without reaction of MDA. Enzyme-Linked Immunosorbent Assay (ELISA) for V8 Protease Digests of MDA-Modified Insulin B Chain. The fractionated samples were dissolved in 100 µL of 50 mM sodium phosphate buffer (pH 7.2). A 50 µL aliquot of the antigen solution was added to each well of a 96-well microtiter plate and incubated for 1 h at 37 °C. The antigen solution was then removed, and the plate was washed with phosphate-buffered saline (PBS) containing 0.05% Tween 20 (PBS/Tween) and water. Each well was filled with 200 µL of blocking solution (40 mg/mL BlockAce) for 1 h at 37 °C. The primary antibody (mAb1F83) was then added to the wells at 100 µL/well of 1 µg/mL solution for 2 h at 37 °C. The plate was then washed once with PBS/Tween. After discarding the supernatants and washing three times with PBS/Tween and water, 100 µL of a 5 × 103 dilution of goat anti-mouse IgG conjugated to horseradish peroxidase in PBS/Tween was added. After incubation for 1 h at 37 °C, the supernatant was discarded, and the plates were washed three times with PBS/Tween and water. Enzyme-linked

Chem. Res. Toxicol., Vol. 19, No. 1, 2006 123

Figure 2. Sequences of chain B from bovine insulin (molecular mass 3495.9 Da) and its fragments formed by the enzymatic digestion of Staphylococcus aureus V8 protease. Molecular mass of Fragment 1 is 1529.7 Da, 914.4 Da for Fragment 2, and 1085.6 Da for Fragment 3, respectively. The amino acid residues modified by MDA are represented in bold characters. Label (*) denotes oxidation of cysteine residues (Cys-SO3H).

antibody bound to the well was revealed by adding 100 µL/well of 1,2-phenylenediamine (0.5 mg/mL) in 0.1 M citrate/phosphate buffer (pH 5.0) containing 0.003% H2O2. The reaction was terminated by the addition of 50 µL of 2 M sulfuric acid, and the absorbance at 492 nm was read on a micro-ELISA plate reader. HPLC Analysis of Native and MDA-Modified Insulin B Chains. Native and MDA-modified insulin B chains were analyzed by a reversed-phase HPLC. Separation of products was carried out on a nanospace SI-1 HPLC system (Shiseido, Tokyo, Japan) with a FP-1520 fluorescence detector (Jasco Co., Tokyo, Japan), using a Capcell Pak C18 UG300 column (Shiseido). These samples were eluted with a linear gradient of 0.1% acetic acid in water (solvent A)-acetonitrile (solvent B) (time ) 0, 20% B; 0-5 min, 20% B; 5-25 min, 35% B; 25-35 min, 45% B). The flow rate was 0.2 mL/min, and the column temperature was controlled at 40 °C. The elution profiles were monitored by absorbance at 215 nm and by fluorescence intensity (excitation, 387 nm; emission, 455 nm). Peptide Mapping. Native and MDA-modified insulin B chains (0.5 mg/mL) were digested with V8 protease in 0.1 mL of 50 mM sodium phosphate buffer (pH 7.2) at 37 °C for 24 h using an enzyme/substrate ratio of 1:50 (w/w). Peptide samples were analyzed by a reversed-phase HPLC, which was the same system described above, using a Capcell Pak C18 UG120 column (Shiseido). These samples were eluted with a linear gradient of 0.1% acetic acid in water (solvent A)-acetonitrile (solvent B) (time ) 0, 10% B; 0-5 min, 10% B; 5-45 min, 60% B). The flow rate was 0.2 mL/min, and the column temperature was controlled at 40 °C. The elution profiles were monitored by absorbance at 215 nm and by fluorescence intensity (excitation, 387 nm; emission, 455 nm). The fractions containing peptides (5-30 min) were collected at intervals of 30 s and dried in a centrifugal concentrator CC-105 (Tomy Seiko, Tokyo, Japan). Liquid Chromatography-Electrospray Ionization-Tandem Mass Spectrometry (LC-ESI-MS/MS) Analysis. The LCESI-MS (MS/MS) analyses were performed on an LCQ ion trap mass system (Thermo Electron, San Jose, CA) equipped with an electrospray ion source using a spray voltage of 5 kV and a capillary temperature of 260 °C. Spectra were acquired in the positive ion mode, with a scan range from m/z 200 to 2000. Collision-induced dissociation (CID) experiments in the positive ion mode were performed by setting the relative collision energy at 30% using helium as collision gas.

Results Modification Behavior of MDA to Insulin B Chain. To investigate the modification behavior of MDA to protein, we used the insulin B chain as a model peptide, which contains two main possible modification sites of MDA, the N-terminal amino acid residue (Phe1) and one lysine residue (Lys29) (Figure 2). As shown in Figure 3A, incubation of insulin B chain (1 mg/mL) with 1 mM MDA resulted in a time-dependent loss of primary amino groups of the insulin B chain. Approximately 55% of the amino groups was lost after 24 h of incubation,

124 Chem. Res. Toxicol., Vol. 19, No. 1, 2006

Ishii et al.

Figure 4. HPLC profiles of native and MDA-modified insulin B chains: (A) native insulin B chain; (B) chromatogram of MDAmodified insulin B chain detected by UV 215 nm; (C) chromatogram of MDA-modified insulin B chain detected by fluorescein (excitation, 387 nm; emission, 455 nm). Molecular mass and modified structure of each peak are shown in Table 1. Table 1. Summary of MDA-Modified Insulin B Chain

Figure 3. MDA modification of insulin B chain. Insulin B chain (1 mg/mL) was incubated with 1 mM MDA in 50 mM sodium phosphate buffer (pH 7.4) at 37 °C. (A) Loss of primary amino group; (B) increase in fluorescence intensity (excitation, 387 nm; emission, 455 nm); (C) Immunoreactivity with mAb1F83.

peak

MW

∆ma (Da)

fluorescence

proposed modification

1 2 3 4 5 6

3496 3630 3550 3550 3630 3684

134 54 54 134 188

o o o

native DHP-type N-propenal-type N-propenal-type DHP-type N-propenal- + DHP-type

a

suggesting that MDA reacted not only with the lysine residue but also with the N-terminal amino acid residues in insulin B chain. The loss of amino groups of insulin B chain was accompanied by an increase in protein fluorescence (Figure 3B) and immunoreactivity with mAb1F83 (Figure 3C), suggesting that the mAb1F83 may recognize DHP-type adducts generated in the MDA-modified insulin B chain. Subsequently, aliquots of native and MDA-modified insulin B chains were analyzed by HPLC. As shown in Figure 4, several peaks were detected on the chromatogram by UV (215 nm) and fluorescence (excitation, 387 nm; emission, 455 nm) detection. The individual peaks were then analyzed by LC-ESI-MS, and the results are summarized in Table 1. LC-ESI-MS analysis of the native insulin B chain gave a molecular mass of 3496 Da (peak 1), which was in agreement with the theoretical molecular mass derived from the sequence for insulin B chain. The peaks 3 and 4 gave the same molecular mass, [M + H]+ at m/z ) 3550 (+54 Da), which corresponded to the generation of N-propenal-type adducts, whereas the peaks 2 and 5 gave the same molecular mass, [M + H]+ at m/z ) 3630 (+134 Da), which corresponded to generation of DHP-type adducts, both of which showed a strong fluorescence. Similarly, peak 6, whose molecular mass was [M + H]+ at m/z ) 3684, which corresponded to a 188 Da increase in the mass value of native

∆m (Da) are the increased weights against native insulin B chain.

insulin B chain, suggested a 54 and 134 Da increase in the mass value at two sites in native insulin B chain. The corresponding mass shift and the different chromatographic behavior suggested that at least two sites react with MDA, and this seemed to be reasonably explained by the formation of N-propenal- and DHPtype adducts. Detection of MDA-Modified Fragments in ProteaseDigested Insulin B Chain by LC-ESI-MS Analysis. The native and MDA-treated insulin B chains were digested with V8 protease and then analyzed by LC-ESI-MS. V8 protease digestion of insulin B chain theoretically generates three fragments, namely, FVNQHLC*GSHLVE (C*, Cys-SO3H) (F1), ALYLVC*GE (C*, Cys-SO3H) (F2), and RGFFYTPKA (F3) (Figure 2). As shown in Figure 5A, limited proteolysis of native insulin B chain with V8 protease indeed gave three fragments, a, b, and g. Peaks a, b, and g were identified to be F3 (RGFFYTPKA), F1 (FNNQHLCGSHLVE), and F2 (ALYLVCGE), respectively. The MDA-modified insulin B chain digested with V8 protease was also analyzed by HPLC with detection by UV and fluorescence. By modification of MDA, the peaks a (F3) and b (F1) were significantly decreased ,and new peaks (peaks d and f) with florescence appeared (Figure 5B,C). In addition, both fluorescent peaks exhibited significant


Figure 5. HPLC profiles of digested native and MDA-modified insulin B chains: (A) digested native insulin B chain; chromatograms of digested MDA-modified insulin B chain (B) detected by UV 215 nm, (C) detected by fluorescein (excitation, 387 nm; emission, 455 nm); and (D) ELISA analysis of HPLC fractions for immunoreactivity with mAb1F83. Molecular mass and modified structure of each peak are shown in Table 2.

immunoreactivity with mAb1F83 (Figure 5D). Interestingly, fluorescence intensity of the peak d was weaker than the peak f, but immunoreactivity of the peak d was more significant than the peak f. On the basis of their mass value, we presumed that the fragments e and f originated from the modification of the sequence F1 and the fragments c and d originated from the modification of the sequence F3 (Figure 6). MDA was anticipated to form two different types of adducts, namely, the N-propenal-, and DHP-type adducts (Figure 1), resulting in a 54 and 134 Da increase in the peptide mass, respectively. Relative to the calculated masses of the unmodified fragments, two fragments (c and e) increased the mass 54 Da, which corresponded to a N-propenal-type adduct (Figure 6). We also detected two fragments d and f, which increased the mass 134 Da (Figure 6). Identification of the MDA Modification Site in Insulin B Chain by MS/MS Analysis. Since F1 and F3 contain only one possible target amino acid (N-terminus (Phe1) and Lys29, respectively), the mass increments of 54 Da (fragments c and e) and 134 Da (fragments d and f) in peptide fragments F1 and F3 probably originated from the modification of the N-terminus (Phe1) and Lys29 with MDA molecules. Hence, to further characterize the MDA modification of the insulin B chain, both


Figure 6. Mass spectra of peptides from native and MDA-modified insulin B chains digested with V8 protease. Spectra a-g correspond to the alphabetical symbols in the HPLC chromatograms in Figure 5.

unmodified fragments (a, b, and g) and MDA-modified fragments (c, d, e, and f) were analyzed by ESI-LC/MS/MS without additional chromatography. The MS/MS analysis of the [M + H]+ at m/z ) 1086.6 from fragment a, the [M + H]+ at m/z ) 1531.8 from fragment b, and the [M + H]+ at m/z ) 915.5 from fragment g confirmed the identity of the sequences and their lack of modifications (data not shown). Figure 7 shows the MS/MS spectra of peaks c-f. MS/MS analysis of peak c (F3 + 54 Da) revealed the singly charged N-terminal product ions (b3-b7) and H2O lost product ions (b6-18 and b7-18) (Figure 7A). Furthermore, compared to the native peptide, fragment ions (y3-y5, y7, and b82+) and H2O or NH3 lost fragment ions (y4-18, y5-17, b8-18, and [b8-18]2+) showed a 54 Da increase. These results confirmed that the N-propenaltype adduct is associated with Lys29. Figure 7B shows the MS/ MS spectrum of peak d (F3 + 134 Da). Several N-terminal product ions (b2-b6) and an H2O lost fragment ion (b7-18) were observed. In addition, the fragment ions (y2-y6 and b82+) and H2O or NH3 lost fragment ions (y3-17 and [y8-18]2+) showed a 134 Da increase compared to those in the native peptide. These data indicated that the DHP-type adduct is associated with Lys29. The MS/MS spectrum of peak e (F1 + 54 Da) revealed several singly charged C-terminal product ions (y6-y9, y11, and y12), and H2O or NH3 lost product ions (y1018 and y11-17) were observed (Figure 7C). Furthermore, the N-terminal fragment ions (b4∼12 and a-series ions) and NH3 lost fragment ions (b10-17, b11-17, and b12-17) showed a 54 Da increase, suggesting that the N-propenal-type adduct is


Ishii et al.

Figure 7. ESI-LC/MS/MS spectra of the peptides from the MDA-modified insulin B chain digested with V8 protease. (A) MS/MS spectrum of the [M + H]2+ ion at m/z 571.2 from peak c (F3 + 54 Da); (B) MS/MS spectrum of the [M + H]2+ ion at m/z 611.0 from peak d (F3 + 134 Da). Label (*) denotes modification by MDA. (C) MS/MS spectrum of the [M + H]+ ion at m/z 1585.6 from peak e (F1 + 54 Da); (D) MS/MS spectrum of the [M + H]+ ion at m/z 1665.8 from peak f (F1 + 134 Da). Label (*) denotes modification by MDA.

on N-terminal amino acid residue Phe1 in the sequence. Figure 7D shows the MS/MS spectrum of peak f (F1 + 134 Da). The MS/MS spectrum revealed singly charged C-terminal product ions (y6, y8, y10, and y12), and H2O lost product ions (y1018 and y11-18) were observed in the MS/MS spectrum. The N-terminal fragment ions (b4∼12 and a-series ions) and H2O or NH3 lost fragment ions (b4-17, b6-17, b7-18, b8-18, b918, b10-18, b11-18, and b12-17) showed a 134 Da increase compared to those in the native peptide. These results suggest that the DHP-type adduct is associated with the N-terminal amino acid residue (Phe1). Table 2 summarizes the results of MS/MS analysis of the V8 protease-digested MDA-modified insulin B chain. Thus, from MS/MS analysis, we identified the formation of N-propenal- and DHP-type adducts in both N-terminus (Phe1) and Lys29 on the insulin B chain. These findings confirmed the formation of MDA adducts such as DHPtype adduct in the amino groups in the protein and revealed

Table 2. Summary of the V8 Protease Digested Peptides from MDA-Modified Insulin B Chain peak

[M + H]+

∆ma (Da)

fluorescence

proposed modification

a b c d e f g

1086.6 1531.8 1140.7 1220.6 1585.6 1665.8 915.5

54 134 54 134 -

o o -

F3 F1 F3 + N-propenal-type F3 + DHP-type F1 + N-propenal-type F1 + DHP-type F2

a

∆m (Da) is the increase in weight against native insulin B chain.

that the epitope recognized by mAb1F83 was of a lipofuscinlike DHP-lysine fluorophore.

Discussion Using the mAb (MDA-Lys) against MDA-modified lysine residues, Haberland et al. (26) demonstrated immunocytochemi-


cally the presence of MDA-lysine residues in atherosclerotic lesions that colocalized with apoB. In addition, by immunocytochemical application of different antibodies, Palinski et al. (27, 28) and Rosenfeld et al. (29) confirmed the presence of MDA-lysine in atherosclerotic lesions and demonstrated several different oxidation-specific epitopes in the lesion but not in normal areas of rabbit aortas. Previously, the materials immunoreactive to the mAb1F83 raised against MDA-modified protein indeed constituted the atherosclerotic lesions of human aorta (24). Thus, the monoclonal antibodies developed in these studies commonly recognize the MDA linked to lysine residues in arterial lesions. However, because of the different experimental procedures for antigen generation (e.g., concentration of MDA, pH, time of incubation, and protein carrier), the antibodies may recognize different MDAlysine epitopes. In addition, the MDA adducts such as DHPlysine generated in the proteins have not been chemically characterized. A recent mass spectrometry approach, using ESI-LC/MS and MALDI-TOF MS, has permitted the direct characterization of the structures of the antibodies recognizing the epitope in the protein. By using a combination of immunochemical and mass spectrometric methods, Crabb et al. (30) demonstrated that 4-hydroxy-2-nonenal (HNE) treatment of the purified bovine cathepsin B results in selective modification of active site residues, Cys29 (A chain) and His150 (B chain), with the significantly reduced enzyme activity. Previously, we demonstrated that HNE treatment of the purified glyceraldehyde-3phosphate dehydrogenase (GAPDH) results in the increase of immunoreactivity with the anti-HNE-histidine antibody and modification of amino acid residues primarily located on the surface of the GAPDH molecule, with the generation of HNEMichael adducts (31). In other studies, based on the mass spectroscopic identification of a novel acrolein-lysine adduct, we examined the specificity of mAb5F6 and found that the antibody recognized N-(3-methylpyridinium)lysine (MPlysine) in the proteins (32). This study revealed that MDA adducts such as the DHPtype adduct were generated in the -amino group of lysine in the protein and that the structure of the epitope recognized by mAb1F83 was the DHP-lysine adducts, using mass spectroscopic techniques. To investigate the modification behavior of MDA to protein, we analyzed a model peptide (insulin B chain) treated with MDA by the LC-ESI-MS analysis. The insulin B chain has two main possible modification sites, the N-terminal amino acid residue (Phe1) and one lysine residue (Lys29) (Figure 2). Major products obtained by binding of MDA to lysine residues have been reported to be N-propenal- and DHP-lysine adducts, which corresponded, respectively, to the 54 and 134 Da increase in the mass value of unmodified lysine residues. Peaks 3 and 4 exhibited the same molecular mass of m/z ) 3550, which corresponded to a 54 Da increase in the mass value of the native insulin B chain, and agreed with the addition of N-propenal (Table 1), although they displayed a different chromatographic behavior as shown in Figure 4. Furthermore, peaks 2 and 5 exhibited the same molecular mass of m/z ) 3630, which corresponded to a 134 Da increase in the mass value of the native insulin B chain, and agreed with the addition of DHP-type adducts (Figure 3). On the other hand, peak 6, whose molecular mass was 3684, which corresponded to a 188 Da increase in the mass value of native insulin B chain, was expected to increase the mass value 54 and 134 Da at two sites in the native insulin B chain. The corresponding mass shift suggested reaction with MDA at least at two sites, which seemed


to be reasonably explained by the formation of N-propenal- and DHP-type adducts. To identify the modification sites of MDA in insulin B chain, MDA-treated insulin B chain (insulin B chain/MDA ) 1:1 ∼35, mol/mol) were digested with V8 protease, and the resulting peptides were subjected to LC-ESI-MS/MS. This technique identified four peptides, which contained the N-propenal- and DHP-type adducts at the N-terminus and Lys29 (Figures 5 and 7), and revealed that the N-terminus (Phe1) was detected as a single modification site of MDA in a 1:1 ratio sample (data not shown). On the basis of their reactivity and mass value, we presumed that the peaks 3 (N-propenal-type) and 5 (DHP-type) originated from the modification at an N-terminus and the peaks 2 (N-propenal-type) and 4 (DHP-type) originated from the modification at a Lys29 (Figure 4). These results suggest that formation of MDA adducts to protein is influenced by the amino acid sequence of proteins. The N-propenal-type adduct is a major product by incubation of insulin B chain with MDA (Figure 4B). This adduct was generated on the N-terminus and Lys29 (Figure 7A,C). Previously, we found that N-acetyl-N-propenal lysine was generated as the predominant product by incubation of N-acetyl lysine with MDA (15). In addition, polyclonal antiserum raised against the MDA-modified protein was found to contain antibody populations that could be purified by affinity gel prepared by covalent attachment of N-acetyl-N-2-propenal lysine. We concluded that the affinity-purified anti-N-propenal lysine antibody was highly specific to the N-propenal lysine. This antibody revealed that N-propenal-type adduct is formed in human LDL upon reaction with MDA or Cu2+. These results and observations suggest that the N-propenal-type adduct represents one of a major form of MDA covalently attached to proteins. Previously, Slatter et al. reported two concurrent mechanisms for the formation of DHP-type adducts from MDA and propylamine (18). Formation of an N-propenal-type adduct is faster than that of DHP-type adducts initially, because direct formation of DHPtype adducts via the proposed transient intermediates requires an initial chemical cleavage of MDA into acetaldehyde and formic acid. Formation of DHP-type adducts is assumed to be irreversible as an N-propenal-type adduct and MDA go down to trace levels after the reaction had proceeded, and there is no apparent degradation of DHP-type adducts. Therefore generation of N-propenal lysine may influence generation and accumulation of DHP-lysine in proteins. MDA-derived DHP contribute to the presence of elevated levels of fluorescence not only in the protein incubated with MDA but also in oxdatively modified LDL and human atherosclerotic aorta. Lipofuscin is a general term used to describe fluorescent material that accumulates in cells as function of age (29, 33). Lipid peroxidation has been generally regarded as being involved in the formation of lipofuscin with characteristic fluorescence (8). Previously, we demonstrated that 4-methyl-1,4-dihydropyridine-3,5-dicarbaldehyde (DHP)-derivative was generated as the predominant product by incubation of N-acetyl lysine with MDA, and the structure of the epitope recognized by mAb1F83 was DHP-derivative by competitive ELISA and NMR analysis (24). However, the MDA adducts such as the DHP-lysine generated in a protein and the structure of the epitope in the protein recognized by mAb1F83 have not been chemically characterized. As shown in Figure 5D, the ELISA analysis of HPLC fractions for immunoreactivity with mAb1F83 showed that the antibody had immunoreactivity not only with the DHP-lysine adduct but also with the peak corresponding to DHP-N-terminus adduct. More strikingly, the


ratio of immunoreactivities to the yields of these products (DHP-lysine and DHP-N-terminus adducts) suggested that the antibody might be rather specific to the DHP-lysine adducts. These results demonstrated that the reaction of insulin B chain with MDA resulted in DHP-type adducts and provide further direct evidence that the major epitope recognized by mAb1F83 was the DHP-lysine adduct. The possibility that MDA may play a role in the pathogenesis of atherosclerosis has also been suggested by the facts that (i) high concentrations of MDA can be generated during the oxidation of LDL phospholipids (3, 4, 35), (ii) the structural and functional changes associated with the oxidation of LDL can also be produced by direct interaction of LDL with MDA (33), (iii) the reaction of MDA with a critical number of lysine residues of LDL apoB produces internalization by the scavenger receptor of human monocyte-macrophages and subsequent intracellular accumulation of lipoprotein-derived cholesteryl ester (8, 36, 37), (iv) the level of MDA-modified LDL increases in the plasma of patients with atherosclerosis (38), and (v) antioxidant therapy slows the progress of early atherosclerotic lesions (39, 40). Previously, we characterized the formation of the fluorophore in oxidized LDL and in atherosclerotic lesions of human aorta using mAb1F83 against the MDA-derived lipofuscin-like fluorophore (24). This study provides further direct evidence that the major epitope recognized by mAb1F83 was DHP-lysine adduct. These finding and this study strongly suggest that DHP-lysine accumulates in atherosclerotic lesions of human aorta. In summary, we have chemically characterized the modification behavior of MDA to protein by mass spectroscopic techniques. In addition, we showed that mAb1F83 is specific to the DHP-lysine adduct generated in the MDA-treated protein. Previously, we demonstrated that the materials immunoreactive to mAb1F83 indeed constitute the atherosclerotic lesions, in which intense positivity was associated primarily with macrophage-derived foam cells (24). These findings provided further direct evidence that DHP-type adducts are the major fluorophore in oxidized LDL and in atherosclerotic lesions of human aorta, and the reaction between the MDA and lysine residues of proteins might represent a process common to the formation of lipofuscin-like fluorophores during aging and its related diseases. Acknowledgment. We thank Atsunori Furuhata and Takahiro Shibata (Nagoya University) for the helpful technical suggestions. This work was supported by a research grant from the Ministry of Education, Culture, Sports, Science and Technology, by the COE program in the 21st Century in Japan, and by central Shizuoka, Cooperation of Innovation Technology and Advanced Research in Evolutional Area.

References (1) Shacter, E. (2000) Quantification and significance of protein oxidation in biological samples. Drug. Metab. ReV. 32, 307-326. (2) Stadtman, E. R., and Levine, R. L. (2000) Protein oxidation. Ann. N. Y. Acad. Sci. 899, 191-208. (3) Esterbauer, H., Schaur, R. J., and Zollner, H. (1991) Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radical Biol. Med. 11, 81-128. (4) Uchida, K. (2000) Role of reactive aldehyde in cardiovascular diseases. Free Radical Biol. Med. 28, 1685-1696. (5) Steinberg, D., Parthasarathy, S., Carew, T. E., Khoo, J. C., and Witztum, J. L. (1989) Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N. Engl. J. Med. 320, 915924. (6) Steinberg, D. (1995) Role of oxidized LDL and antioxidants in atherosclerosis. AdV. Exp. Med. Biol. 369, 39-48.

Ishii et al. (7) Chio, K. S., and Tappel, A. L. (1969) Synthesis and characterization of the fluorescent products derived from malonaldehyde and amino acids. Biochemistry 8, 2821-2827. (8) Haberland, M. E., Fogelman, A. M., and Edwards, P. A. (1982) Specificity of receptor-mediated recognition of malondialdehydemodified low density lipoproteins. Proc. Natl. Acad. Sci. U.S.A. 79, 1712-1716. (9) McGirr, L. G., Hadley, M., and Draper, H. H. (1985) Identification of NR-acetyl--(2-propenal)lysine as a urinary metabolite of malondialdehyde. J. Biol. Chem. 260, 15427-15431. (10) Draper, H. H., Hadley, M., Lissemore, L., Laing, N. M., and Cole, P. D. (1988) Identification of N-(2-propenal)lysine as a major urinary metabolite of malondialdehyde. Lipids 23, 626-628. (11) Slatter, D. A., Paul, R. G., Murray, M., and Bailey, A. J. (1999) Reactions of lipid-derived malondialdehyde with collagen. J. Biol. Chem. 274, 19661-19669. (12) Marnett, L. J. (2002) Oxy radicals, lipid peroxidation and DNA damage. Toxicology 27, 181-182, 219-222. (13) Luczaj, W., and Skrzydlewska, E. (2003) DNA damage caused by lipid peroxidation products. Cell. Mol. Biol. Lett. 8, 391-413. (14) Stone, K., Uzieblo, A., and Marnett, L. J. (1990) Studies of the reaction of malondialdehyde with cytosine nucleosides. Chem. Res. Toxicol. 3, 467-472. (15) Uchida, K., Sakai, K., Itakura, K., Osawa, T., and Toyokuni, S. (1997) Protein modification by lipid peroxidation products: formation of malondialdehyde-derived N-(2-propenol)lysine in proteins. Arch. Biochem. Biophys. 346, 45-52. (16) Burcham, P. C., and Kuhan, Y. T. (1996) Introduction of carbonyl groups into proteins by the lipid peroxidation product, malondialdehyde. Biochem. Biophys. Res. Commun. 220, 996-1001. (17) Kikugawa, K., and Ido, Y. (1984) Studies on peroxidized lipids. V. Formation and characterization of 1,4-dihydropyridine-3,5-dicarbaldehydes as model of fluorescent components in lipofuscin. Lipids 19, 600-608. (18) Slatter, D. A., Murray, M., and Bailiey, A. J. (1998) Formation of a dihydropyridine derivative as a potential cross-link derived from malondialdehyde in physiological systems. FEBS Lett. 421, 180-184. (19) Itakura, K., Uchida, K., and Osawa, T. (1996) A novel fluorescent malondialdehyde-lysine adduct. Chem. Phys. Lipids 84, 75-79. (20) Slatter, D. A., Avery, N. C., and Bailey, A. J. (2004) Identification of a new cross-link and unique histidine adduct from bovine serum albumin incubated with malondialdehyde. J. Biol. Chem. 279, 6169. (21) Dell, A., and Morris, H. R. (2001) Glycoprotein structure determination by mass spectrometry. Science 291, 2351-2356. (22) Lim, H., Eng, J., Yates, J. R., Tollaksen, S. L., Giometti, C. S., Holden, J. F., Adams, M. W. W., Reich, C. I., Olsen, G. J., and Hays, L. G. (2003) Identification of 2D-gel proteins: a comparison of MALDI/ TOF peptide mass mapping to mu LC-ESI tandem mass spectrometry. J. Am. Soc. Mass Spectrom. 14, 957-970. (23) Wu, C. C., and Yates, J. R. (2003) The application of mass spectrometry to membrane proteomics. Nat. Biotechnol. 21, 262-267. (24) Yamada, S., Kumazawa, S., Ishii, T., Nakayama, T., Itakura, K., Shibata, N., Kobayashi, M., Sakai, K., Osawa, T., and Uchida, K. (2001) Immunochemical detection of a lipofuscin-like fluorophore derived from malondialdehyde and lysine. J. Lipid Res. 42, 11871196. (25) Marnett, L. J., and Tuttle, M. A. (1980) Comparison of the mutagenicities of malondialdehyde and the side products formed during its chemical synthesis. Cancer Res. 40, 276-282. (26) Haberland, M. E., Fong, D., and Cheng, L. (1988) Malondialdehydealtered protein occurs in atheroma of Watanabe heritable hyperlipidemic rabbits. Science 241, 215-217. (27) Palinski, W., Rosenfeld, M. E., Butler, S., Yla¨-Herttuala, S., Gurtner, G. C., Socher, S. S., Butler, S. W., Parthasarathy, S., Carew, T. E., and Steinberg, D. (1989) Low-density lipoprotein undergoes oxidative modification in vivo. Proc. Natl. Acad. Sci. U.S.A. 86, 1372-1376. (28) Palinski, W., Ord, V. A., Plump, A. S., Breslow, J. L., Steinberg, D., and Witztum, J. L. (1994) ApoE-deficient mice are a model of lipoprotein oxidation in atherogenesis. Demonstration of oxidationspecific epitopes in lesions and high titers of autoantibodies to malondialdehyde-lysine in serum. Arterioscler. Thromb. 14, 605-616. (29) Rosenfeld, M. E., Palinski, W., Yla-Herttuala, S., Butler, S., and Witztum, J. L. (1990) Distribution of oxidation specific lipid-protein adducts and apolipoprotein B in atherosclerotic lesions of varying severity from WHHL rabbits. Arteriosclerosis 10, 336-349. (30) Crabb, J. W., O’Neil, J., Miyagi, M., West, K., and Hoff, H. F. (2002) Hydroxynonenal inactivates cathepsin B by forming Michael adducts with active site residues. Protein Sci. 11, 831-840. (31) Ishii, T., Tatsuda, E., Kumazawa, S., Nakayama, T., and Uchida, K. (2003) Molecular basis of enzyme inactivation by an endogenous


(32)

(33) (34) (35) (36)

electrophile 4-hydroxy-2-nonenal: identification of modification sites in glyceraldehyde-3-phosphate dehydrogenase. Biochemistry 42, 34743480. Furuhata, A., Ishii, T., Kumazawa, S., Yamada, T., Nakayama, T., and Uchida, K. (2003) N-(3-Methylpyridinium)lysine, a major antigenic adduct generated in acrolein-modified protein. J. Biol. Chem. 278, 48658-48665. Hoff, H. F., and O’Neil, J. (1993) Structural and functional changes in LDL after modification with both 4-hydroxynonenal and malondialdehyde. J. Lipid Res. 34, 1209-1217. Esterbauer, H., Dieber-Rotheneder, M., Waeg, G., Striegel, G., and Ju¨rgens, G. (1990). Biochemical, structural, and functional properties of oxidized low-density lipoprotein. Chem. Res. Toxicol. 3, 77-92. Esterbauer, H., Gebicki, J., Puhl, H., and Ju¨rgens, G. (1992) The role of lipid peroxidation and antioxidants in oxidative modification of LDL. Free Radical Biol. Med. 13, 341-343. Haberland, M. E., Olch, C. L., and Fogelman, A. M. (1984) Role of lysines in mediating interaction of modified low-density lipoproteins with the scavenger receptor of human monocyte macrophages. J. Biol. Chem. 259, 11305-11311.

Chem. Res. Toxicol., Vol. 19, No. 1, 2006 129 (37) Fogelman, A. M., Schecter, I., Seager, J., Hokom, M., Child, J. S., and Edwards, P. A. (1980) Malondialdehyde alteration of low-density lipoproteins leads to cholesteryl ester accumulation in human monocyte-macrophages. Proc. Natl. Acad. Sci. U.S.A. 77, 2214-2218. (38) Holvoet, P., Perez, G., Zhao, Z. E., Brouwers, E., Bernar, H., and Collen, D. (1995). Malondialdehyde-modified low-density lipoproteins in patients with atherosclerotic disease. J. Clin. InVest. 95, 26112619. (39) Carew, T. E., Schwenke, D. C., and Steinberg, D. (1987) Antiatherogenic effect of probucol unrelated to its hypocholesterolemic effect: evidence that antioxidants in vivo can selectively inhibit low-density lipoprotein degradation in macrophage-rich fatty streaks and slow the progression of atherosclerosis in the Watanabe heritable hyperlipidemic rabbit. Proc. Natl. Acad. Sci. U.S.A. 84, 7725-7729. (40) Kang, M. H., Kawai, Y., Naito, M., and Osawa, T. (1999) Mode of action of sesame lignans in protecting low-density lipoprotein against oxidative damage in vitro. J. Nutr. 129, 1885-1890.

TX050231P

Mass Spectroscopic Characterization of Protein ... - ACS Publications

Recommend Documents