- Email: [email protected]
Generation of a Polyclonal Antibody Against Lipid Peroxide-Modified Proteins
Free Radical Biology & Medicine, Vol. 23, No. 2, pp. 251–259, 1997 Copyright q 1997 Elsevier Science Inc. Printed in the USA. All rights reserved 0891-5849/97 $17.00 / .00
PII S0891-5849(96)00615-6
Original Contribution GENERATION OF A POLYCLONAL ANTIBODY AGAINST LIPID PEROXIDE-MODIFIED PROTEINS
JONG G. KIM,* FADI SABBAGH,* NALINI SANTANAM,* JOSIAH N. WILCOX,† RUSSELL M. MEDFORD,† and SAMPATH PARTHASARATHY* *Department of Gynecology and Obstetrics, and the †Department of Medicine, Emory University, Atlanta, GA 30322, USA (Received 6 September 1996; Accepted 26 November 1996)
Abstract—A specific polyclonal antibody against the lipid peroxide (LOOH)-modified rabbit serum albumin (RSA) was generated in rabbits. The antibody selectively recognized the modified protein in a concentration-dependent manner and did not cross react with aldehyde-modified proteins or proteins directly oxidized with the free radical generator 2,2 *-azobis (2-amidinopropane) hydrochloride (AAPH). Oxidized low-density lipoprotein (Ox-LDL), but not native LDL, was also recognized by the antibody in a concentration-dependent manner. The antibody also cross reacted with several other proteins modified by LOOH suggesting that the antibody is directed towards a common epitope and not towards the protein sequence. Western blot analysis of normal human plasma showed that at least three different proteins are recognized by the antibody. RAW cells, preincubated with LOOH, were immunostained with the antibody and the antigenic epitopes were present intracellularly, while controls lacking in the primary antibodies failed to show immunoreactivity. Atherosclerotic arteries from cholesterol-fed monkeys and human atherosclerotic lesions were also immunostained by the antibody. The immunoreactivity was co-localized in areas rich in foam cell macrophages. These results suggest that LOOH-modified proteins present an unique antigenic epitope that may represent a primary product of interaction of LOOH with proteins. q 1997 Elsevier Science Inc. Keywords—Atherosclerosis, Lipid peroxidation, Aldehydes, Low-density lipoprotein, Oxidized low-density lipoprotein
of aldehydes, intact lipid peroxides (LOOH), and the hydroxy products derived from LOOH. Aldehydes generated from lipid peroxides are highly reactive. Aldehydes readily modify protein thiols, lysine and other residues.5 – 7 Such aldehyde modified proteins are highly antigenic and antibodies to such modified epitopes have provided a powerful tool for the detection of aldehydemodified proteins in the atherosclerotic artery.8 – 10 However, the generation of aldehydes from lipid peroxides may represent a terminal decomposition stage which may require metal ions such as iron or copper, or their chelated derivatives such as heme. In contrast, LOOH, primary products of lipid oxidation, are also highly reactive and readily react with amino groups of proteins, amino lipids such as phosphatidylethanolamine (PtdEtn), and other lipophilic amino compounds. In addition, they can react with a wide variety of amino acids.11 When lipid oxidation
INTRODUCTION
Oxidative stress has been implicated in a number of diseases. Activation of cellular oxidases has been suggested to result in the loss of antioxidant protection and eventually to the oxidation of lipids of cell membranes and circulating lipoproteins. The deleterious effects of peroxidized lipids and their degradation products have been well recognized in the pathogenesis of atherosclerosis.1 – 3 Oxidation has now been deemed a potential risk factor for cardiovascular diseases.4 A number of methods are available for the detection of lipid peroxides. These include the determination of malondialdehyde (MDA) levels that are determined as thiobarbituric acid products (TBARS), determination Address correspondence to: Sampath Parthasarathy, Ph.D., Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, GA 30322. 251
/ 2b2c 2469 Mp 251 Monday Jun 02 10:16 PM EL–FRB 2469
252
J. G. KIM et al.
occurs, the close proximity of LOOH to membrane proteins and apoproteins of lipoproteins may suggest that initial modifications other than aldehyde-mediated modifications may be more relevant in biological systems. In other words, the generation of LOOH on a cell membrane or lipoproteins is more likely to generate, at least in initial stages, proteins that are modified directly by LOOH rather than proteins that are modified by their extensive degradation products, such as aldehydes. Lipid peroxides react with proteins in a number of different ways. Lipid peroxides and oxygen radicals can directly damage specific amino acids.12 – 14 The direct oxidation of amino acids may generate carbonyl groups that can be detected by antibodies against dinitro phenylhydrazine derivatives of the carbonyl groups.15 Oxidation can also crosslink proteins, so that large molecular weight molecules can be formed.16 Products of lipid peroxidation such as aldehydes can react with specific amino acids resulting in the modification of these amino acids.5 – 7,17,18 We have described a modification that readily ensued during the interaction of linoleic acid hydroperoxide with proteins such as albumin or even polylysine.19 This interaction generated fluorescent products similar to those that are naturally formed when LDL is subjected to oxidation. Similar modification ensues when unsaturated PtdEtn is subjected to oxidation.19,20 In the current study, we describe the generation and characterization of a polyclonal antibody to proteins modified by LOOH. We have modified rabbit serum albumin with LOOH and generated a polyclonal antibody that recognizes proteins modified by LOOH, but fails to recognize unmodified proteins. By immunohistochemistry, this antibody recognizes epitopes present in the atherosclerotic arteries of human and cholesterol fed monkeys, and RAW macrophage cells preincubated with lipid peroxide. The antibody is effective in Western blot analysis and can be used to detect the presence of modified epitopes even in normal plasma. More importantly, this antibody did not cross react with proteins modified by aldehydes. MATERIAL AND METHODS
Materials Rabbit serum albumin (RSA), soybean lipoxygenase, linoleic acid, goat anti-rabbit IgG conjugated with alkaline phosphatase, octylglucoside, hexanal, tetramethoxypropane and p-nitrophenyl phosphate were obtained from Sigma chemical company (St Louis, MO, USA). Nonenal was purchased from Aldrich chemical company (Milwaukee, WI, USA), and 2,2 *-azobis (2amidinopropane) hydrochloride (AAPH) was purchased from Polysciences, Inc. (Warrington, PA,
USA). Nonfat milk powder was purchased from BioRad Chemical Company (Hercules, CA, USA). Tween-20 and 96 well microtiter plates were purchased from Fisher Chemical Company (Pittsburgh, PA, USA). A synthetic, lysine-rich peptide of the sequence [YVTKSYNETKIKFDKYKAEKSHDEL] was generously provided by Dr. Richard Smith. Methods Preparation of lipid peroxide-modified albumin Linoleic acid was converted into 13-hydroperoxy linoleate by treatment with soybean lipoxygenase (SLO) as described.19,20 Linoleic acid hydroperoxide was immediately reacted with immunoglobulin (IgG)-free RSA and incubated at 377C for 2 d. In a typical reaction, 100 nmols of linoleic acid was treated with 30 units of SLO in 1 ml of phosphate buffered saline (PBS) and the reaction was followed by measuring the increase in absorption at 234 nm. Usually, the reaction is complete within 30 min.19 The lipid peroxide (13-HPODE) was then treated with 100 mg of lipid-free and IgG-free albumin (or other proteins) in the presence of 50 mM EDTA for 2 d at 377C. The formation of fluorescent products was established by measuring the fluorescence at excitation wavelength of 330 nm and the emission at 430 nm. The product was extracted with chloroform and methanol by the method of Bligh and Dyer 21 to remove unreacted LOOH and then washed several times with ice cold acetone. The final product, LOOH modified RSA (LOOH/RSA) was soluble in aqueous buffers and has fluorescent characteristics similar to that of Ox-LDL. The generation of LOOH, as well as the modification of the protein, were performed in the absence of any added metals to limit the formation of aldehydes. Preparation of LDL and Ox-LDL LDL was isolated from heparinized plasma of normal human donors using a table-top Beckman TL-100 ultracentrifuge and a TLA-100.4 rotor.22 The isolated LDL was dialyzed against PBS at 47C (200 1 volumes) for 6 h. LDL (100 mg/ml) was oxidized by incubation in PBS with 5 mM copper at 377C. After 24 h incubation the solution was transferred to a glass tube and delipidated by the extraction of the lipids by the method of Bligh and Dyer.21 The delipidated LDL protein was dissolved in octylglucoside (100 ml of 10 mg/ml was added to the protein, along with 5 ml of 1 N NaOH) as described by Parthasarathy et al.23 Preparation of aldehyde-modified proteins IgG-free RSA containing 1 mg protein (or other proteins) in 100 ml of PBS was taken, and the total volume was in-
/ 2b2c 2469 Mp 252 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins
creased to 1 ml using PBS. Nonenal or hexanal, 100 ml of 25 mM in ethanol, was added to the protein and mixed well and incubated at 377C for 24 h. Then 4 ml of ice cold acetone was added to the solution and the tube was kept in the freezer for 1 h. After centrifugation at 1500 rpm for 10 min at 47C, the supernatant was removed and this step was repeated three more times. The precipitated protein was dried in vacuum, dissolved in PBS and used in ELISA assays. Preparation of MDA-modified proteins 100 ml of tetramethoxypropane was taken and 0.5 ml of 6 N HCl was added to the tube and the tube was heated at 607C for 30 min. The pH was adjusted to 6.4 using 4 N NaOH, and the total volume adjusted to 2.7 ml using PBS. Then 25 ml of the prepared solution was added to 2 mg of IgG-free rabbit serum albumin or other proteins, and incubated at 377C. After 3 h of incubation the solution was dialyzed against PBS and used in ELISA. Preparation of AAPH-modified proteins Fatty acidfree RSA containing 100 mg protein in 0.5 ml of PBS with 0, 1.0, 5.0, and 10.0 mM AAPH was incubated for 4 h at 377C. Then epitopes generated in AAPHmodified RSA and LOOH/RSA were compared by ELISA. Preparation of the antibody Three male rabbits with 3–4 kg body weight were purchased from Myrtle Rabbitry (Thompson Station, TN, USA). For primary immunization, 1.5 mg/ml of LOOH modified rabbit serum albumin was dissolved in PBS and mixed with Freunds complete adjuvant (Sigma), was then injected by subcutaneous injection. Booster immunization was continued with antigen in PBS and mixed with Freunds incomplete adjuvant at 4 week interval. Rabbits were bled 10–14 d following the immunizations and blood was allowed to stand for 4 h at room temperature and 47C overnight. After removal of clot and debris by centrifuging 20 min at 3000 rpm, the serum was assayed by ELISA and stored at 0207C. Monthly titers were followed and blood was drawn terminally at 6 months after initial immunization. ELISA assay for LOOH-modified proteins Wells of ELISA plate were coated with 50 ml per well of different dilutions of the lipid peroxide modified RSA and incubated in 377C for 3 h. Plates were washed three times with 0.05% Tween-20 in PBS and blocked for 3 h with 300 ml of 1% nonfat milk powder along with 0.05% Tween-20 in PBS. After blocking, the plates were washed three times with 0.05% Tween-20 in PBS, and anti-LOOH/RSA sera was diluted 1:250 and 50 ml
253
was added to each well and incubated in 377C for 3 h or overnight. After washing three times with 0.05% Tween-20 in PBS, anti-rabbit IgG conjugated with alkaline phosphatase was diluted 1:38,000 and 50 ml was added to each well and incubated at 377C for 2 h. After washing six times with 0.05% Tween-20 in PBS, 50 ml of p-nitrophenyl phosphate was added to each well and incubated at 377C and the plates were checked at 15min intervals for 2 h. The curve was constructed by plotting OD reading vs. concentration of LOOH/RSA. ELISA assay for aldehyde-modified peptides Modified peptides were plated in the 96 well plates overnight at room temperature. Wells were blocked by 3% nonfat dry milk in PBS for 2 h and washed three times with PBS. Anti-LOOH/RSA antibody was diluted 1:250 with PBS containing 3% BSA was added to each well. After 2 h incubation at 377C, wells were washed with PBS three times and goat anti-rabbit IgG conjugated with alkaline phosphatase was diluted to 1:38,000 and added. After 2 h incubation at 377C, wells were washed again and p-nitrophenyl phosphate was added. Color development was determined by a ELISA plate reader. Immunohistochemistry of monkey and human artery Frozen segments of the abdominal aorta from a group of male cynomolgus monkeys were generously provided as a gift from Drs. Thomas Clarkson and Koudy Williams of Bowman-Gray University, Department of Comparative Zoology. These animals had been fed a moderately high fat diet for over 5 years consisting of high protein monkey chow supplemented with 8.2% dried egg yolk and 10% lard. The final diet contained 37.5% saturated fat, 44.9% monounsaturated fat, 17.5% polyunsaturated fat with 0.25% cholesterol. The average serum cholesterol level of these animals was 306 mg/dL. These cholesteol levels are sufficient to produce a range of atherosclerotic lesions in the monkeys. Human aortas displaying different degrees of atherosclerotic development were obtained from organ donors at the time of tissue harvest and were collected with the approval of the Emory University Human Subjects Committee. Human and cynomologus monkey aortas were fixed in paraformaldehyde and frozen in O.C.T. prior sectioning to 5 mm on a cryostat. Tissue sections were then immunostained using the antibody at a dilution of 1:500 followed by biotinylated goat anti-rabbit IgG (Fisher Scientific) used at a dilution of 1:200 and visualized by the Vector ABC-AP system using Vector Red as a chromogen (Vector Laboratories). Western blot analysis Western blotting was performed after separation of proteins in a 7.5% cross-
/ 2b2c 2469 Mp 253 Monday Jun 02 10:16 PM EL–FRB 2469
254
J. G. KIM et al.
linked acrylamide gels using 2.5 mg protein. Transblotted samples were detected using 1:250 diluted anti-LOOH / RSA as the primary antibody and a peroxidase-conjugated goat anti-rabbit IgG as the secondary antibody. The cross reactive bands were visualized by chemiluminescence detection. Immunostaining of LOOH treated cells RAW macrophages were incubated with 100 mM 13-HPODE for 1, 2, or 3 d as follows. Confluent cells in 24 well plates were treated with 13-HPODE for 1, 2, or 3 d in serumfree DMEM. Fresh 13-HPODE was added on the second and third day, to respective cell dishes. At the end of the first, second, or third day, cells were fixed with Bouins solution for 10 min and immunohistochemistry was performed using antibodies against LOOH/RSA. After fixation, cells were washed three times with PBS. Anti-LOOH/RSA antibody was diluted 1:250 with PBS containing 3% BSA and was added to each well. For negative control, primary antibody was not added. After 2 h incubation at room temperature, wells were washed with PBS three times and goat anti-rabbit IgG conjugated with alkaline phosphatase was diluted to 1:100 and added. After 2 h incubation at room temperature, wells were washed again and incubated with fast red. After color development, the reaction was terminated and the cells were photographed using a Nikon microscope with a camera attachment. RESULTS
First, we determined whether the antibody recognized the LOOH-modified protein and not the unmodified protein using an ELISA assay. Unmodified IgGfree RSA was used as control. Preliminary results showed that the antibody titer in animals increased with time and plateaued at about 4 months. The results presented in Fig. 1 show that the antibody at 1:250 dilution selectively recognized the modified protein in a concentration-dependent manner. As little as 10 ng of RSA ( õ 2 nM modified RSA) was able to generate recognition significantly different from similar levels of the control protein. Unmodified control protein was barely recognized by the antibody even at high concentrations. Controls without the primary or the secondary antibodies did not show any recognition. Other modified proteins (LOOH-modified bovine and human serum albumin, catalase and cytochrome C) were also recognized by the antibody (results not given). Western blot analysis also showed that the antibody specifically recognized the LOOH-modified albumin (Fig. 2A). Fatty acid-free albumin, used as control albumin, was not recognized. A number of antibodies have been described for al-
Fig. 1. Recognition of LOOH/RSA, but not RSA by anti-LOOH/ RSA antibody. Microtiter wells were coated with increasing concentrations of LOOH modified RSA (solid bar) or RSA (open bar). After blocking with BSA, 50 ml of 1:250 dilution of anti-LOOH/ RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phosphatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent averages of a triplicate set of wells from one of at least 6 separate trials.
dehyde-modified proteins.6,8,9 Our previous studies have demonstrated that it is unlikely that aldehydes contributed to the modification during the incubation of LOOH with proteins.19 However, the incubation is of long duration and it is likely that aldehydes could have contributed to the generation of the antigenic epitopes and to the antibody. To establish whether the antibody to LOOH/RSA recognized proteins modified by aldehydes, we used a synthetic peptide rich in lysine residues. We used the peptide, as our results demonstrated that plasma proteins, such as albumin, from commercial sources may already possess similar epitopes, either as a result of their in vivo presence or as a result of their generation during in vitro purification procedures (results not given). ELISA assay of MDA, hexanal, nonenal, and LOOH-modified synthetic peptide is shown in Table 1. The antibody failed to recognize the native peptide or other aldehyde-modified peptides to any significant extent. In contrast, LOOHmodified peptide was avidly recognized by the antibody. Western blot analysis of aldehyde- and LOOH-modified synthetic peptide is shown in Fig. 2B. Again, none of the aldehyde-modified peptides were recognized to any detectable extent (lanes not shown due to lack of reaction), whereas LOOH-modified peptide was readily recognized by the antibody. There were multiple bands of immunoreactive peptides suggesting crosslinks. The apoprotein B100 (apo B) component undergoes extensive modification during the oxidation of LDL.2,24,25 A number of studies have demonstrated that the modification includes new antigenic epitopes gen-
/ 2b2c 2469 Mp 254 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins
255
Fig. 2. Western blot analysis of LOOH-modified RSA, aldehyde-modified peptide and human plasma using LOOH/RSA antibody. Proteins were separated by 7.5% polyacrylamide gel electrophoresis and transblotted. Membranes were incubated with anti-LOOH/ RSA antibody and the reaction was determined by chemiluminescence. (A) L: RSA, R: LOOH modified RSA; (B) LOOHmodified peptide; (C) human plasma.
erated by the covalent modification of lysine residues by aldehydes.8 – 10 To determine whether Ox-LDL also contains lysines modified by intact LOOH, we determined the recognition of LDL and Ox-LDL by the antibody to LOOH/RSA using an ELISA assay. Figure 3 shows that Ox-LDL was recognized by the antibody in a concentration-dependent manner. The antibody was able to recognize as little as 0.25 mg of Ox-LDL protein ( õ 0.5 nM apo B). Native LDL, prepared in the presence of butylated hydroxtoluene (BHT) was not recognized even at 2.5 mg concentrations. In sep-
arate experiments we determined that the lipid component of Ox-LDL was also recognized by the antibody suggesting that the amino phospholipids, such as phosphatidylethanolamine (PtdEtn) may be similarly modified. The Western blot analysis of LDL and Ox-LDL showed that, as expected, the antibody was able to recognize Ox-LDL and not native LDL (gel not shown).
Table 1. Recognition of LOOH-Modified Peptide and Not Aldehyde-Modified Peptides by Anti-LOOH/RSA Antibody mg Peptide
Native Peptide
LOOHPeptide
MDAPeptide
NonenalPeptide
HexanalPeptide
0 0.25 1.25 2.5
57 46 38 42
51 190 327 358
53 50 57 62
51 56 60 46
55 64 84 103
Microtiter wells were coated with increasing concentrations of unmodified, or modified peptide. After blocking wells with nonfat milk powder, 100 ml of 1:250 dilution of anti-LOOH/RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phosphatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent averages of a triplicate set of optical density readings of wells from one of at least two separate trials.
Fig. 3. Recognition of Ox-LDL, but not LDL by anti-LOOH/RSA antibody. Microtiter wells were coated with increasing concentrations of Ox-LDL (open bar) or LDL (solid bar). After blocking with BSA, 50 ml of 1:250 dilution of anti-LOOH/RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phosphatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent averages of a triplicate set of wells from one of at least three separate trials OD of each well.
/ 2b2c 2469 Mp 255 Monday Jun 02 10:16 PM EL–FRB 2469
256
J. G. KIM et al.
Western blot analysis of normal human plasma showed that at least three different proteins are recognized by the antibody (Fig. 2C). The identities of these proteins are unknown, although from the mobility on the gel it is suspected that these proteins may represent albumin and proteolytic products derived from apoprotein B. The ability of the antibody to recognize modified proteins in tissues was tested under two conditions. In the first study, RAW cells preincubated with LOOH were used. Immunohistochemistry ( Fig. 4 ) showed that cells incubated with LOOH were immunostained with the antibody and the antigenic epitopes were present intracellularly. Control cells and controls lacking in the primary antibodies failed to show immunoreactivity. Atherosclerotic arteries from cholesterol-fed monkeys and from atherosclerotic human subjects were immunostained with the antibody. Figures 5A and 5B show intense immunoreactivity, which was co-localized in areas rich in foam cell macrophages as determined by counter staining for macrophages. Deposits of immunoreactive material are found in similar sites in early atherosclerotic lesions in human aortas and in hypercholesterolemic monkeys. In both human and the experimental monkey lesions, the antigenic epitope was localized in the foamy intimal macrophages as well as the endothelium overlying early lesions (Fig. 5). DISCUSSION
The formation of lipid peroxides in biological systems is a complex process and can be brought about by a variety of means. These include enzymatic (lipoxygenases, cyclooxygenases, peroxidases, and other oxygenases) as well as by non enzymatic means by way of generation of reactive oxygen species. The generation of these products has been suggested to be involved etiologically as well as consequentially in a number of diseases. A plethora of studies have documented using a variety of experimental approaches to demonstrate and document the presence and involvement of lipid peroxidation products in normal and abnormal pathology. However, the transient nature of LOOH themselves suggested that the actual concentrations of these highly reactive products may vary depending on the presence of redox metals, thiols, nitric oxide, proteins, and a host of other biological materials. Lipids in cell membranes are surrounded by membrane proteins and lipids are also carried in the plasma in the form of plasma lipoproteins. This suggested to us that lipophilic proteins may be one of the first targets of interaction with LOOH. Accordingly, we described a specific modification that results during the
incubation of LOOH with proteins.19 This study also showed that the amino groups of lysines may be involved, as similar modification could be shown during the interaction of polylysine with LOOH. The results presented in this study show that the modification of proteins by LOOH generated an antigenic epitope. We used LOOH-modified IgG-free rabbit albumin itself to generate antibodies in rabbits thus eliminating the possibility that the antibody could be due to the protein backbone. The poor recognition of unmodified albumin in our studies clearly demonstrates that the antibody did not recognize an epitope in the native protein. In fact, several batches of commercial albumin preparations and albumin from human plasma appear to have detectable levels of the antigenic epitopes suggesting that such epitopes may even be present in vivo. Albumin has been suggested to carry products of lipid peroxidation in vivo.26 It is possible that albumin, an amphiphilic plasma protein that readily complexes with free fatty acids and other lipids is more vulnerable to oxidative damage. The epitope for this antibody was generated by the long incubation of protein with lipid peroxide. During this modification, a number of other reactive epitopes can be generated including changes in the conformation of the native protein. Thus, to eliminate the possibility that LOOH-induced conformational changes in RSA could have generated immunological response and generated the antibodies in rabbits, we used a synthetic peptide derived from the sequence of apo B. This peptide, described in this study contains six lysine residues. This peptide, upon modification with LOOH, was readily recognized by the antibody. It is unlikey that a short peptide derived from the sequence of a totally different protein could have the same ‘‘conformation’’ as LOOH/RSA. It is more likely that LOOH/RSA and LOOH-modified peptides shared similar antigenic epitopes. The interaction of proteins with LOOH has been suggested to result in proteolysis as well as aggregation.12 In Western blot analysis, LOOH/RSA and the apo B peptide showed several immunologically reactive bands. More of aggregation could be seen with the small peptide. All the fragments of albumin or apo B peptides reacted with the antibody. These fragments indicated that the antibody recognized a common epitope and not any specific fragments derived from the original polypeptide. Aldehyde-modified proteins have been demonstrated in the atherosclerotic artery by immunohistochemistry.8 – 10,27 The results presented show that the LOOH/RSA antibody is specific for the proposed epitope and does not recognize aldehyde-modified proteins. Free radicals can directly interact with proteins
/ 2b2c 2469 Mp 256 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins
Fig. 4. Immunohistochemical staining of RAW cells preincubated with LOOH. RAW cells were incubated in a 24 well tissue culture plates in the presence of lipid peroxide. After 3 d of culture, cells containing epitopes reacting with anti-LOOH/RSA was determined by immunohistochemistry. Left: in the absence of the primary antibody, Right: in the presence of the primary antibody.
Fig. 5. Immunohistochemical localization of LOOH-modified antigenic epitopes in early atherosclerotic lesions from human and hypercholesterolemic cynomolgus monkeys. The LOOH-modified RSA antibody was used for immunohistochemistry to localize LOOH-modified antigenic epitopes in human (A) and hypercholesterolemic monkey aortic lesions (B). LOOH-modified antigenic epitopes were found in endothelial cells and in macrophage foam cells of the atherosclerotic human aortas. In both human and hypercholesterolemic monkey lesions the antigenic epitope is localized in macrophages in a similar distribution.
/ 2b2c 2469 Mp 257 Monday Jun 02 10:16 PM EL–FRB 2469
257
258
J. G. KIM et al.
philic proteins interact with LOOH and undergo modification as described. Autoantibody to oxidized LDL has been reported to be present even in normal plasma. These antibodies recognize MDA-LDL, at least under in vitro conditions. It remains to be established whether these antibodies also recognize LOOH-modified proteins. Acknowledgements — The authors would like to thank Drs. Clarkson and Williams of the Bowman-Gray School of Medicine for their generous gift of samples of atherosclerotic aortas from hypercholesterolemic cynomologus monkeys. JNW is supported by NIH grants HL47838–04 and HL48667.
REFERENCES Fig. 6. Recognition of LOOH-modified, but not AAPH-modified RSA by anti-LOOH/RSA antibody. Microtitier wells were coated with unmodified, AAPH-modified or LOOH-modified RSA. After blocking with BSA, 50 ml of 1:250 dilution of anti-LOOH/RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phophatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent mean { standard deviation of a duplicate set of wells from three separate trials.
and generate other modifications that may be antigenic. To ensure that the antibody recognizes specifically those modifications that are induced by lipid peroxides, we tested the ability of the antibody to detect RSA modified by AAPH. The antibody failed to recognize AAPH-modified albumin, whereas it readily recognized LOOH-modified RSA (Fig. 6). To our knowledge the current antibody represents the recognition of an unique antigen that is not represented by aldehydemodified proteins. The ability of the antibody to immunostain the atherosclerotic artery would suggest that these antigenic epitopes are derived in the macrophagerich tissue, either by the uptake of oxidized LDL or by in situ formation. Thus, these studies would indicate that direct modification of proteins by LOOH may be an ongoing process in the atherosclerotic artery. Our studies also show that the neo-antigenic epitopes are present even in normal plasma. However, aldehyde-modified proteins have not been detected in plasma by antibodies to aldehyde-modified proteins. Recently, it was reported that immunologically detectable epitopes could be demonstrated in patients with unstable angina by antibodies directed towards MDALDL. The MDA was proposed to be generated by aggregated platelets and not during the oxidation of LDL. In plasma, it is unlikely that large concentrations of free aldehydes are generated from LOOH, as very little redox metals are available in the free form. On the other hand, LOOH can be derived from dietary sources, 28,29 and from plasma cells as well as from oxidative enzymes, such as myeloperoxidase. It is likely that lipo-
1. Steinberg, D.; Parthasarathy, S.; Carew, T. E.; Khoo, J. C.; Witztum, J. L. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N. Engl. J. Med. 320:915–924; 1989. 2. Parthasarathy, S., ed. Modified lipoproteins in the pathogenesis of atherosclerosis. Austin: R.G. Landes Company; 1994. 3. Penn, M. S.; Chisolm, G. M., III. Oxidized lipoproteins, altered cell function and atherosclerosis. Atherosclerosis 108 Suppl:S21–S29; 1994. 4. Forrester, J. S.; Merz, C. N.; Bush, T. L.; Cohn, J. N.; Hunninghake, D. B.; Parthasarathy, S.; Superko, H. R. 27th Bethesda Conference: matching the intensity of risk factor management with the hazard for coronary disease events. Task Force 4. Efficacy of risk factor management. J. Am. Coll. Cardiol. 27:991– 1006; 1996. 5. Esterbauer, H.; Jurgens, G.; Quehenberger, O.; Koller, E. Autoxidation of human low density lipoprotein: loss of polyunsaturated fatty acids and vitamin E and generation of aldehydes. J. Lipid Res. 28:495–509; 1987. 6. Steinbrecher, U. P. Oxidation of human low density lipoprotein results in derivatization of lysine residues of apolipoprotein B by lipid peroxide decomposition products. J. Biol. Chem. 262:3603–3608; 1987. 7. Shechter, I.; Fogelman, A. M.; Haberland, M. E.; Seager, J.; Hokom, M.; Edwards, P. A. The metabolism of native and malondialdehyde-altered low density lipoproteins by human monocytemacrophages. J. Lipid Res. 22:63–71; 1981. 8. Haberland, M. E.; Fong, D.; Cheng, L. Malondialdehyde-altered protein occurs in the atheroma of Watanabe heritable hyperlipidemic rabbits. Science 241:215–218; 1988. 9. Jurgens, G.; Ashy, A.; Esterbauer, H. Detection of new epitopes formed upon oxidation of low-density lipoprotein, lipoprotein (a) and very-low-density lipoprotein. Use of an antiserum against 4-hydroxynonenal-modified low-density lipoprotein. Biochemical J. 265:605–608; 1990. 10. Palinski, W.; Rosenfeld, M. E.; Yla¨ Herttuala, S.; Gurtner, G. C.; Socher, S. S.; Butler, S. W.; Parthasarathy, S.; Carew, T. E.; Steinberg, D.; Witztum, J. L. Low density lipoprotein undergoes oxidative modification in vivo. Proc. Natl. Acad. Sci. USA 86:1372–1376; 1989. 11. Fletcher, B. L.; Tappel, A. L. Fluorescent modification of serum albumin by lipid peroxidation. Lipids 6:172–175; 1971. 12. Fong, L. G.; Parthasarathy, S.; Witztum, J. L.; Steinberg, D. Nonenzymatic oxidative cleavage of peptide bonds in apoprotein B100. J. Lipid Res. 28:1466–1477; 1987. 13. Davies, K. J. Protein modification by oxidants and the role of proteolytic enzymes. Biochem. Soc. Trans. 21:346 – 353; 1993. 14. Stadtman, E. R. Oxidation of free amino acids and amino acid residues in proteins by radiolysis and by metal-catalyzed reactions. Annu. Rev. Biochem. 62:797–821; 1993.
/ 2b2c 2469 Mp 258 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins 15. Levine, R. L.; Williams, J. A.; Stadtman, E. R.; Shacter, E. Carbonyl assays for determination of oxidatively modified proteins. Methods Enzymol. 233:346–357; 1994. 16. Heinecke, J. W.; Li, W.; Francis, G. A.; Goldstein, J. A. Tyrosyl radical generated by myeloperoxidase catalyzes the oxidative cross-linking of proteins. J. Clin. Invest. 91:2866 – 2872; 1993. 17. Uchida, K.; Toyokuni, S.; Nishikawa, K.; Kawakishi, S.; Oda, H.; Hiai, H.; Stadtman, E. R. Michael addition-type 4-hydroxy2-nonenal adducts in modified low-density lipoproteins: markers for atherosclerosis. Biochem. 33:12487–12494; 1994. 18. Uchida, K.; Stadtman, E. R. Modification of histidine residues in proteins by reaction with 4-hydroxynonenal. Proc. Natl. Acad. Sci. USA 89:4544–4548; 1992. 19. Fruebis, J.; Parthasarathy, S.; Steinberg, D. Evidence for a concerted reaction between lipid hydroperoxides and polypeptides. Proc. Natl. Acad. Sci. USA 89:10588–10592; 1992. 20. Parthasarathy, S. Lipid peroxidation-induced protein modifications in the oxidation of low density lipoprotein. In: Packer, L.; Cadenas, E., eds. Biological oxidants and antioxidants. Stuttgart: Hippokrates Verlag; 1994:255–259. 21. Bligh, E. G.; Dyer, W. J. Rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911–917; 1959. 22. Santanam, N.; Parthasarathy, S. Paradoxical actions of antioxidants in the oxidation of low density lipoprotein by peroxidases. J. Clin. Invest. 95:2594–2600; 1995. 23. Parthasarathy, S.; Fong, L. G.; Otero, D.; Steinberg, D. Recognition of solubilized apoproteins from delipidated, oxidized low
24.
25.
26.
27.
28. 29.
259
density lipoprotein (LDL) by the acetyl-LDL receptor. Proc. Natl. Acad. Sci. USA 84:537–540; 1987. Steinbrecher, U. P.; Witztum, J. L.; Parthasarathy, S.; Steinberg, D. Decrease in reactive amino groups during oxidation or endothelial cell modification of LDL. Correlation with changes in receptor-mediated catabolism. Arteriosclerosis 7:135–143; 1987. Steinbrecher, U. P.; Parthasarathy, S.; Leake, D. S.; Witztum, J. L.; Steinberg, D. Modification of low density lipoprotein by endothelial cells involves lipid peroxidation and degradation of low density lipoprotein phospholipids. Proc. Natl. Acad. Sci. USA 81:3883–3887; 1984. Houglum, K.; Filip, M.; Witztum, J. L.; Chojkier, M. Malondialdehyde and 4-hydroxynonenal protein adducts in plasma and liver of rats with iron overload. J. Clin. Invest. 86:1991–1998; 1990. Rosenfeld, M. E.; Palinski, W.; Yla Herttuala, S.; Butler, S.; Witztum, J. L. Distribution of oxidation specific lipid-protein adducts and apolipoprotein B in atherosclerotic lesions of varying severity from WHHL rabbits. Arteriosclerosis 10:336–349; 1990. Staprans, I.; Pan, X. M.; Miller, M.; Rapp, J. H. Effect of dietary lipid peroxides on metabolism of serum chylomicrons in rats. Am. J. Physiol. 264:G561–8; 1993. Staprans, I.; Rapp, J. H.; Pan, X. M.; Feingold, K. R. The effect of oxidized lipids in the diet on serum lipoprotein peroxides in control and diabetic rats. J. Clin. Invest. 92:638 – 643; 1993.
/ 2b2c 2469 Mp 259 Monday Jun 02 10:16 PM EL–FRB 2469
PII S0891-5849(96)00615-6
Original Contribution GENERATION OF A POLYCLONAL ANTIBODY AGAINST LIPID PEROXIDE-MODIFIED PROTEINS
JONG G. KIM,* FADI SABBAGH,* NALINI SANTANAM,* JOSIAH N. WILCOX,† RUSSELL M. MEDFORD,† and SAMPATH PARTHASARATHY* *Department of Gynecology and Obstetrics, and the †Department of Medicine, Emory University, Atlanta, GA 30322, USA (Received 6 September 1996; Accepted 26 November 1996)
Abstract—A specific polyclonal antibody against the lipid peroxide (LOOH)-modified rabbit serum albumin (RSA) was generated in rabbits. The antibody selectively recognized the modified protein in a concentration-dependent manner and did not cross react with aldehyde-modified proteins or proteins directly oxidized with the free radical generator 2,2 *-azobis (2-amidinopropane) hydrochloride (AAPH). Oxidized low-density lipoprotein (Ox-LDL), but not native LDL, was also recognized by the antibody in a concentration-dependent manner. The antibody also cross reacted with several other proteins modified by LOOH suggesting that the antibody is directed towards a common epitope and not towards the protein sequence. Western blot analysis of normal human plasma showed that at least three different proteins are recognized by the antibody. RAW cells, preincubated with LOOH, were immunostained with the antibody and the antigenic epitopes were present intracellularly, while controls lacking in the primary antibodies failed to show immunoreactivity. Atherosclerotic arteries from cholesterol-fed monkeys and human atherosclerotic lesions were also immunostained by the antibody. The immunoreactivity was co-localized in areas rich in foam cell macrophages. These results suggest that LOOH-modified proteins present an unique antigenic epitope that may represent a primary product of interaction of LOOH with proteins. q 1997 Elsevier Science Inc. Keywords—Atherosclerosis, Lipid peroxidation, Aldehydes, Low-density lipoprotein, Oxidized low-density lipoprotein
of aldehydes, intact lipid peroxides (LOOH), and the hydroxy products derived from LOOH. Aldehydes generated from lipid peroxides are highly reactive. Aldehydes readily modify protein thiols, lysine and other residues.5 – 7 Such aldehyde modified proteins are highly antigenic and antibodies to such modified epitopes have provided a powerful tool for the detection of aldehydemodified proteins in the atherosclerotic artery.8 – 10 However, the generation of aldehydes from lipid peroxides may represent a terminal decomposition stage which may require metal ions such as iron or copper, or their chelated derivatives such as heme. In contrast, LOOH, primary products of lipid oxidation, are also highly reactive and readily react with amino groups of proteins, amino lipids such as phosphatidylethanolamine (PtdEtn), and other lipophilic amino compounds. In addition, they can react with a wide variety of amino acids.11 When lipid oxidation
INTRODUCTION
Oxidative stress has been implicated in a number of diseases. Activation of cellular oxidases has been suggested to result in the loss of antioxidant protection and eventually to the oxidation of lipids of cell membranes and circulating lipoproteins. The deleterious effects of peroxidized lipids and their degradation products have been well recognized in the pathogenesis of atherosclerosis.1 – 3 Oxidation has now been deemed a potential risk factor for cardiovascular diseases.4 A number of methods are available for the detection of lipid peroxides. These include the determination of malondialdehyde (MDA) levels that are determined as thiobarbituric acid products (TBARS), determination Address correspondence to: Sampath Parthasarathy, Ph.D., Department of Gynecology and Obstetrics, Emory University School of Medicine, Atlanta, GA 30322. 251
/ 2b2c 2469 Mp 251 Monday Jun 02 10:16 PM EL–FRB 2469
252
J. G. KIM et al.
occurs, the close proximity of LOOH to membrane proteins and apoproteins of lipoproteins may suggest that initial modifications other than aldehyde-mediated modifications may be more relevant in biological systems. In other words, the generation of LOOH on a cell membrane or lipoproteins is more likely to generate, at least in initial stages, proteins that are modified directly by LOOH rather than proteins that are modified by their extensive degradation products, such as aldehydes. Lipid peroxides react with proteins in a number of different ways. Lipid peroxides and oxygen radicals can directly damage specific amino acids.12 – 14 The direct oxidation of amino acids may generate carbonyl groups that can be detected by antibodies against dinitro phenylhydrazine derivatives of the carbonyl groups.15 Oxidation can also crosslink proteins, so that large molecular weight molecules can be formed.16 Products of lipid peroxidation such as aldehydes can react with specific amino acids resulting in the modification of these amino acids.5 – 7,17,18 We have described a modification that readily ensued during the interaction of linoleic acid hydroperoxide with proteins such as albumin or even polylysine.19 This interaction generated fluorescent products similar to those that are naturally formed when LDL is subjected to oxidation. Similar modification ensues when unsaturated PtdEtn is subjected to oxidation.19,20 In the current study, we describe the generation and characterization of a polyclonal antibody to proteins modified by LOOH. We have modified rabbit serum albumin with LOOH and generated a polyclonal antibody that recognizes proteins modified by LOOH, but fails to recognize unmodified proteins. By immunohistochemistry, this antibody recognizes epitopes present in the atherosclerotic arteries of human and cholesterol fed monkeys, and RAW macrophage cells preincubated with lipid peroxide. The antibody is effective in Western blot analysis and can be used to detect the presence of modified epitopes even in normal plasma. More importantly, this antibody did not cross react with proteins modified by aldehydes. MATERIAL AND METHODS
Materials Rabbit serum albumin (RSA), soybean lipoxygenase, linoleic acid, goat anti-rabbit IgG conjugated with alkaline phosphatase, octylglucoside, hexanal, tetramethoxypropane and p-nitrophenyl phosphate were obtained from Sigma chemical company (St Louis, MO, USA). Nonenal was purchased from Aldrich chemical company (Milwaukee, WI, USA), and 2,2 *-azobis (2amidinopropane) hydrochloride (AAPH) was purchased from Polysciences, Inc. (Warrington, PA,
USA). Nonfat milk powder was purchased from BioRad Chemical Company (Hercules, CA, USA). Tween-20 and 96 well microtiter plates were purchased from Fisher Chemical Company (Pittsburgh, PA, USA). A synthetic, lysine-rich peptide of the sequence [YVTKSYNETKIKFDKYKAEKSHDEL] was generously provided by Dr. Richard Smith. Methods Preparation of lipid peroxide-modified albumin Linoleic acid was converted into 13-hydroperoxy linoleate by treatment with soybean lipoxygenase (SLO) as described.19,20 Linoleic acid hydroperoxide was immediately reacted with immunoglobulin (IgG)-free RSA and incubated at 377C for 2 d. In a typical reaction, 100 nmols of linoleic acid was treated with 30 units of SLO in 1 ml of phosphate buffered saline (PBS) and the reaction was followed by measuring the increase in absorption at 234 nm. Usually, the reaction is complete within 30 min.19 The lipid peroxide (13-HPODE) was then treated with 100 mg of lipid-free and IgG-free albumin (or other proteins) in the presence of 50 mM EDTA for 2 d at 377C. The formation of fluorescent products was established by measuring the fluorescence at excitation wavelength of 330 nm and the emission at 430 nm. The product was extracted with chloroform and methanol by the method of Bligh and Dyer 21 to remove unreacted LOOH and then washed several times with ice cold acetone. The final product, LOOH modified RSA (LOOH/RSA) was soluble in aqueous buffers and has fluorescent characteristics similar to that of Ox-LDL. The generation of LOOH, as well as the modification of the protein, were performed in the absence of any added metals to limit the formation of aldehydes. Preparation of LDL and Ox-LDL LDL was isolated from heparinized plasma of normal human donors using a table-top Beckman TL-100 ultracentrifuge and a TLA-100.4 rotor.22 The isolated LDL was dialyzed against PBS at 47C (200 1 volumes) for 6 h. LDL (100 mg/ml) was oxidized by incubation in PBS with 5 mM copper at 377C. After 24 h incubation the solution was transferred to a glass tube and delipidated by the extraction of the lipids by the method of Bligh and Dyer.21 The delipidated LDL protein was dissolved in octylglucoside (100 ml of 10 mg/ml was added to the protein, along with 5 ml of 1 N NaOH) as described by Parthasarathy et al.23 Preparation of aldehyde-modified proteins IgG-free RSA containing 1 mg protein (or other proteins) in 100 ml of PBS was taken, and the total volume was in-
/ 2b2c 2469 Mp 252 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins
creased to 1 ml using PBS. Nonenal or hexanal, 100 ml of 25 mM in ethanol, was added to the protein and mixed well and incubated at 377C for 24 h. Then 4 ml of ice cold acetone was added to the solution and the tube was kept in the freezer for 1 h. After centrifugation at 1500 rpm for 10 min at 47C, the supernatant was removed and this step was repeated three more times. The precipitated protein was dried in vacuum, dissolved in PBS and used in ELISA assays. Preparation of MDA-modified proteins 100 ml of tetramethoxypropane was taken and 0.5 ml of 6 N HCl was added to the tube and the tube was heated at 607C for 30 min. The pH was adjusted to 6.4 using 4 N NaOH, and the total volume adjusted to 2.7 ml using PBS. Then 25 ml of the prepared solution was added to 2 mg of IgG-free rabbit serum albumin or other proteins, and incubated at 377C. After 3 h of incubation the solution was dialyzed against PBS and used in ELISA. Preparation of AAPH-modified proteins Fatty acidfree RSA containing 100 mg protein in 0.5 ml of PBS with 0, 1.0, 5.0, and 10.0 mM AAPH was incubated for 4 h at 377C. Then epitopes generated in AAPHmodified RSA and LOOH/RSA were compared by ELISA. Preparation of the antibody Three male rabbits with 3–4 kg body weight were purchased from Myrtle Rabbitry (Thompson Station, TN, USA). For primary immunization, 1.5 mg/ml of LOOH modified rabbit serum albumin was dissolved in PBS and mixed with Freunds complete adjuvant (Sigma), was then injected by subcutaneous injection. Booster immunization was continued with antigen in PBS and mixed with Freunds incomplete adjuvant at 4 week interval. Rabbits were bled 10–14 d following the immunizations and blood was allowed to stand for 4 h at room temperature and 47C overnight. After removal of clot and debris by centrifuging 20 min at 3000 rpm, the serum was assayed by ELISA and stored at 0207C. Monthly titers were followed and blood was drawn terminally at 6 months after initial immunization. ELISA assay for LOOH-modified proteins Wells of ELISA plate were coated with 50 ml per well of different dilutions of the lipid peroxide modified RSA and incubated in 377C for 3 h. Plates were washed three times with 0.05% Tween-20 in PBS and blocked for 3 h with 300 ml of 1% nonfat milk powder along with 0.05% Tween-20 in PBS. After blocking, the plates were washed three times with 0.05% Tween-20 in PBS, and anti-LOOH/RSA sera was diluted 1:250 and 50 ml
253
was added to each well and incubated in 377C for 3 h or overnight. After washing three times with 0.05% Tween-20 in PBS, anti-rabbit IgG conjugated with alkaline phosphatase was diluted 1:38,000 and 50 ml was added to each well and incubated at 377C for 2 h. After washing six times with 0.05% Tween-20 in PBS, 50 ml of p-nitrophenyl phosphate was added to each well and incubated at 377C and the plates were checked at 15min intervals for 2 h. The curve was constructed by plotting OD reading vs. concentration of LOOH/RSA. ELISA assay for aldehyde-modified peptides Modified peptides were plated in the 96 well plates overnight at room temperature. Wells were blocked by 3% nonfat dry milk in PBS for 2 h and washed three times with PBS. Anti-LOOH/RSA antibody was diluted 1:250 with PBS containing 3% BSA was added to each well. After 2 h incubation at 377C, wells were washed with PBS three times and goat anti-rabbit IgG conjugated with alkaline phosphatase was diluted to 1:38,000 and added. After 2 h incubation at 377C, wells were washed again and p-nitrophenyl phosphate was added. Color development was determined by a ELISA plate reader. Immunohistochemistry of monkey and human artery Frozen segments of the abdominal aorta from a group of male cynomolgus monkeys were generously provided as a gift from Drs. Thomas Clarkson and Koudy Williams of Bowman-Gray University, Department of Comparative Zoology. These animals had been fed a moderately high fat diet for over 5 years consisting of high protein monkey chow supplemented with 8.2% dried egg yolk and 10% lard. The final diet contained 37.5% saturated fat, 44.9% monounsaturated fat, 17.5% polyunsaturated fat with 0.25% cholesterol. The average serum cholesterol level of these animals was 306 mg/dL. These cholesteol levels are sufficient to produce a range of atherosclerotic lesions in the monkeys. Human aortas displaying different degrees of atherosclerotic development were obtained from organ donors at the time of tissue harvest and were collected with the approval of the Emory University Human Subjects Committee. Human and cynomologus monkey aortas were fixed in paraformaldehyde and frozen in O.C.T. prior sectioning to 5 mm on a cryostat. Tissue sections were then immunostained using the antibody at a dilution of 1:500 followed by biotinylated goat anti-rabbit IgG (Fisher Scientific) used at a dilution of 1:200 and visualized by the Vector ABC-AP system using Vector Red as a chromogen (Vector Laboratories). Western blot analysis Western blotting was performed after separation of proteins in a 7.5% cross-
/ 2b2c 2469 Mp 253 Monday Jun 02 10:16 PM EL–FRB 2469
254
J. G. KIM et al.
linked acrylamide gels using 2.5 mg protein. Transblotted samples were detected using 1:250 diluted anti-LOOH / RSA as the primary antibody and a peroxidase-conjugated goat anti-rabbit IgG as the secondary antibody. The cross reactive bands were visualized by chemiluminescence detection. Immunostaining of LOOH treated cells RAW macrophages were incubated with 100 mM 13-HPODE for 1, 2, or 3 d as follows. Confluent cells in 24 well plates were treated with 13-HPODE for 1, 2, or 3 d in serumfree DMEM. Fresh 13-HPODE was added on the second and third day, to respective cell dishes. At the end of the first, second, or third day, cells were fixed with Bouins solution for 10 min and immunohistochemistry was performed using antibodies against LOOH/RSA. After fixation, cells were washed three times with PBS. Anti-LOOH/RSA antibody was diluted 1:250 with PBS containing 3% BSA and was added to each well. For negative control, primary antibody was not added. After 2 h incubation at room temperature, wells were washed with PBS three times and goat anti-rabbit IgG conjugated with alkaline phosphatase was diluted to 1:100 and added. After 2 h incubation at room temperature, wells were washed again and incubated with fast red. After color development, the reaction was terminated and the cells were photographed using a Nikon microscope with a camera attachment. RESULTS
First, we determined whether the antibody recognized the LOOH-modified protein and not the unmodified protein using an ELISA assay. Unmodified IgGfree RSA was used as control. Preliminary results showed that the antibody titer in animals increased with time and plateaued at about 4 months. The results presented in Fig. 1 show that the antibody at 1:250 dilution selectively recognized the modified protein in a concentration-dependent manner. As little as 10 ng of RSA ( õ 2 nM modified RSA) was able to generate recognition significantly different from similar levels of the control protein. Unmodified control protein was barely recognized by the antibody even at high concentrations. Controls without the primary or the secondary antibodies did not show any recognition. Other modified proteins (LOOH-modified bovine and human serum albumin, catalase and cytochrome C) were also recognized by the antibody (results not given). Western blot analysis also showed that the antibody specifically recognized the LOOH-modified albumin (Fig. 2A). Fatty acid-free albumin, used as control albumin, was not recognized. A number of antibodies have been described for al-
Fig. 1. Recognition of LOOH/RSA, but not RSA by anti-LOOH/ RSA antibody. Microtiter wells were coated with increasing concentrations of LOOH modified RSA (solid bar) or RSA (open bar). After blocking with BSA, 50 ml of 1:250 dilution of anti-LOOH/ RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phosphatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent averages of a triplicate set of wells from one of at least 6 separate trials.
dehyde-modified proteins.6,8,9 Our previous studies have demonstrated that it is unlikely that aldehydes contributed to the modification during the incubation of LOOH with proteins.19 However, the incubation is of long duration and it is likely that aldehydes could have contributed to the generation of the antigenic epitopes and to the antibody. To establish whether the antibody to LOOH/RSA recognized proteins modified by aldehydes, we used a synthetic peptide rich in lysine residues. We used the peptide, as our results demonstrated that plasma proteins, such as albumin, from commercial sources may already possess similar epitopes, either as a result of their in vivo presence or as a result of their generation during in vitro purification procedures (results not given). ELISA assay of MDA, hexanal, nonenal, and LOOH-modified synthetic peptide is shown in Table 1. The antibody failed to recognize the native peptide or other aldehyde-modified peptides to any significant extent. In contrast, LOOHmodified peptide was avidly recognized by the antibody. Western blot analysis of aldehyde- and LOOH-modified synthetic peptide is shown in Fig. 2B. Again, none of the aldehyde-modified peptides were recognized to any detectable extent (lanes not shown due to lack of reaction), whereas LOOH-modified peptide was readily recognized by the antibody. There were multiple bands of immunoreactive peptides suggesting crosslinks. The apoprotein B100 (apo B) component undergoes extensive modification during the oxidation of LDL.2,24,25 A number of studies have demonstrated that the modification includes new antigenic epitopes gen-
/ 2b2c 2469 Mp 254 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins
255
Fig. 2. Western blot analysis of LOOH-modified RSA, aldehyde-modified peptide and human plasma using LOOH/RSA antibody. Proteins were separated by 7.5% polyacrylamide gel electrophoresis and transblotted. Membranes were incubated with anti-LOOH/ RSA antibody and the reaction was determined by chemiluminescence. (A) L: RSA, R: LOOH modified RSA; (B) LOOHmodified peptide; (C) human plasma.
erated by the covalent modification of lysine residues by aldehydes.8 – 10 To determine whether Ox-LDL also contains lysines modified by intact LOOH, we determined the recognition of LDL and Ox-LDL by the antibody to LOOH/RSA using an ELISA assay. Figure 3 shows that Ox-LDL was recognized by the antibody in a concentration-dependent manner. The antibody was able to recognize as little as 0.25 mg of Ox-LDL protein ( õ 0.5 nM apo B). Native LDL, prepared in the presence of butylated hydroxtoluene (BHT) was not recognized even at 2.5 mg concentrations. In sep-
arate experiments we determined that the lipid component of Ox-LDL was also recognized by the antibody suggesting that the amino phospholipids, such as phosphatidylethanolamine (PtdEtn) may be similarly modified. The Western blot analysis of LDL and Ox-LDL showed that, as expected, the antibody was able to recognize Ox-LDL and not native LDL (gel not shown).
Table 1. Recognition of LOOH-Modified Peptide and Not Aldehyde-Modified Peptides by Anti-LOOH/RSA Antibody mg Peptide
Native Peptide
LOOHPeptide
MDAPeptide
NonenalPeptide
HexanalPeptide
0 0.25 1.25 2.5
57 46 38 42
51 190 327 358
53 50 57 62
51 56 60 46
55 64 84 103
Microtiter wells were coated with increasing concentrations of unmodified, or modified peptide. After blocking wells with nonfat milk powder, 100 ml of 1:250 dilution of anti-LOOH/RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phosphatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent averages of a triplicate set of optical density readings of wells from one of at least two separate trials.
Fig. 3. Recognition of Ox-LDL, but not LDL by anti-LOOH/RSA antibody. Microtiter wells were coated with increasing concentrations of Ox-LDL (open bar) or LDL (solid bar). After blocking with BSA, 50 ml of 1:250 dilution of anti-LOOH/RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phosphatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent averages of a triplicate set of wells from one of at least three separate trials OD of each well.
/ 2b2c 2469 Mp 255 Monday Jun 02 10:16 PM EL–FRB 2469
256
J. G. KIM et al.
Western blot analysis of normal human plasma showed that at least three different proteins are recognized by the antibody (Fig. 2C). The identities of these proteins are unknown, although from the mobility on the gel it is suspected that these proteins may represent albumin and proteolytic products derived from apoprotein B. The ability of the antibody to recognize modified proteins in tissues was tested under two conditions. In the first study, RAW cells preincubated with LOOH were used. Immunohistochemistry ( Fig. 4 ) showed that cells incubated with LOOH were immunostained with the antibody and the antigenic epitopes were present intracellularly. Control cells and controls lacking in the primary antibodies failed to show immunoreactivity. Atherosclerotic arteries from cholesterol-fed monkeys and from atherosclerotic human subjects were immunostained with the antibody. Figures 5A and 5B show intense immunoreactivity, which was co-localized in areas rich in foam cell macrophages as determined by counter staining for macrophages. Deposits of immunoreactive material are found in similar sites in early atherosclerotic lesions in human aortas and in hypercholesterolemic monkeys. In both human and the experimental monkey lesions, the antigenic epitope was localized in the foamy intimal macrophages as well as the endothelium overlying early lesions (Fig. 5). DISCUSSION
The formation of lipid peroxides in biological systems is a complex process and can be brought about by a variety of means. These include enzymatic (lipoxygenases, cyclooxygenases, peroxidases, and other oxygenases) as well as by non enzymatic means by way of generation of reactive oxygen species. The generation of these products has been suggested to be involved etiologically as well as consequentially in a number of diseases. A plethora of studies have documented using a variety of experimental approaches to demonstrate and document the presence and involvement of lipid peroxidation products in normal and abnormal pathology. However, the transient nature of LOOH themselves suggested that the actual concentrations of these highly reactive products may vary depending on the presence of redox metals, thiols, nitric oxide, proteins, and a host of other biological materials. Lipids in cell membranes are surrounded by membrane proteins and lipids are also carried in the plasma in the form of plasma lipoproteins. This suggested to us that lipophilic proteins may be one of the first targets of interaction with LOOH. Accordingly, we described a specific modification that results during the
incubation of LOOH with proteins.19 This study also showed that the amino groups of lysines may be involved, as similar modification could be shown during the interaction of polylysine with LOOH. The results presented in this study show that the modification of proteins by LOOH generated an antigenic epitope. We used LOOH-modified IgG-free rabbit albumin itself to generate antibodies in rabbits thus eliminating the possibility that the antibody could be due to the protein backbone. The poor recognition of unmodified albumin in our studies clearly demonstrates that the antibody did not recognize an epitope in the native protein. In fact, several batches of commercial albumin preparations and albumin from human plasma appear to have detectable levels of the antigenic epitopes suggesting that such epitopes may even be present in vivo. Albumin has been suggested to carry products of lipid peroxidation in vivo.26 It is possible that albumin, an amphiphilic plasma protein that readily complexes with free fatty acids and other lipids is more vulnerable to oxidative damage. The epitope for this antibody was generated by the long incubation of protein with lipid peroxide. During this modification, a number of other reactive epitopes can be generated including changes in the conformation of the native protein. Thus, to eliminate the possibility that LOOH-induced conformational changes in RSA could have generated immunological response and generated the antibodies in rabbits, we used a synthetic peptide derived from the sequence of apo B. This peptide, described in this study contains six lysine residues. This peptide, upon modification with LOOH, was readily recognized by the antibody. It is unlikey that a short peptide derived from the sequence of a totally different protein could have the same ‘‘conformation’’ as LOOH/RSA. It is more likely that LOOH/RSA and LOOH-modified peptides shared similar antigenic epitopes. The interaction of proteins with LOOH has been suggested to result in proteolysis as well as aggregation.12 In Western blot analysis, LOOH/RSA and the apo B peptide showed several immunologically reactive bands. More of aggregation could be seen with the small peptide. All the fragments of albumin or apo B peptides reacted with the antibody. These fragments indicated that the antibody recognized a common epitope and not any specific fragments derived from the original polypeptide. Aldehyde-modified proteins have been demonstrated in the atherosclerotic artery by immunohistochemistry.8 – 10,27 The results presented show that the LOOH/RSA antibody is specific for the proposed epitope and does not recognize aldehyde-modified proteins. Free radicals can directly interact with proteins
/ 2b2c 2469 Mp 256 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins
Fig. 4. Immunohistochemical staining of RAW cells preincubated with LOOH. RAW cells were incubated in a 24 well tissue culture plates in the presence of lipid peroxide. After 3 d of culture, cells containing epitopes reacting with anti-LOOH/RSA was determined by immunohistochemistry. Left: in the absence of the primary antibody, Right: in the presence of the primary antibody.
Fig. 5. Immunohistochemical localization of LOOH-modified antigenic epitopes in early atherosclerotic lesions from human and hypercholesterolemic cynomolgus monkeys. The LOOH-modified RSA antibody was used for immunohistochemistry to localize LOOH-modified antigenic epitopes in human (A) and hypercholesterolemic monkey aortic lesions (B). LOOH-modified antigenic epitopes were found in endothelial cells and in macrophage foam cells of the atherosclerotic human aortas. In both human and hypercholesterolemic monkey lesions the antigenic epitope is localized in macrophages in a similar distribution.
/ 2b2c 2469 Mp 257 Monday Jun 02 10:16 PM EL–FRB 2469
257
258
J. G. KIM et al.
philic proteins interact with LOOH and undergo modification as described. Autoantibody to oxidized LDL has been reported to be present even in normal plasma. These antibodies recognize MDA-LDL, at least under in vitro conditions. It remains to be established whether these antibodies also recognize LOOH-modified proteins. Acknowledgements — The authors would like to thank Drs. Clarkson and Williams of the Bowman-Gray School of Medicine for their generous gift of samples of atherosclerotic aortas from hypercholesterolemic cynomologus monkeys. JNW is supported by NIH grants HL47838–04 and HL48667.
REFERENCES Fig. 6. Recognition of LOOH-modified, but not AAPH-modified RSA by anti-LOOH/RSA antibody. Microtitier wells were coated with unmodified, AAPH-modified or LOOH-modified RSA. After blocking with BSA, 50 ml of 1:250 dilution of anti-LOOH/RSA antibody was added to each well. After wash, anti-rabbit IgG conjugated with alkaline phophatase was added to each well. After adding substrate p-nitrophenyl phosphate, the OD was measured at 405 nm using a microplate reader. Values represent mean { standard deviation of a duplicate set of wells from three separate trials.
and generate other modifications that may be antigenic. To ensure that the antibody recognizes specifically those modifications that are induced by lipid peroxides, we tested the ability of the antibody to detect RSA modified by AAPH. The antibody failed to recognize AAPH-modified albumin, whereas it readily recognized LOOH-modified RSA (Fig. 6). To our knowledge the current antibody represents the recognition of an unique antigen that is not represented by aldehydemodified proteins. The ability of the antibody to immunostain the atherosclerotic artery would suggest that these antigenic epitopes are derived in the macrophagerich tissue, either by the uptake of oxidized LDL or by in situ formation. Thus, these studies would indicate that direct modification of proteins by LOOH may be an ongoing process in the atherosclerotic artery. Our studies also show that the neo-antigenic epitopes are present even in normal plasma. However, aldehyde-modified proteins have not been detected in plasma by antibodies to aldehyde-modified proteins. Recently, it was reported that immunologically detectable epitopes could be demonstrated in patients with unstable angina by antibodies directed towards MDALDL. The MDA was proposed to be generated by aggregated platelets and not during the oxidation of LDL. In plasma, it is unlikely that large concentrations of free aldehydes are generated from LOOH, as very little redox metals are available in the free form. On the other hand, LOOH can be derived from dietary sources, 28,29 and from plasma cells as well as from oxidative enzymes, such as myeloperoxidase. It is likely that lipo-
1. Steinberg, D.; Parthasarathy, S.; Carew, T. E.; Khoo, J. C.; Witztum, J. L. Beyond cholesterol. Modifications of low-density lipoprotein that increase its atherogenicity. N. Engl. J. Med. 320:915–924; 1989. 2. Parthasarathy, S., ed. Modified lipoproteins in the pathogenesis of atherosclerosis. Austin: R.G. Landes Company; 1994. 3. Penn, M. S.; Chisolm, G. M., III. Oxidized lipoproteins, altered cell function and atherosclerosis. Atherosclerosis 108 Suppl:S21–S29; 1994. 4. Forrester, J. S.; Merz, C. N.; Bush, T. L.; Cohn, J. N.; Hunninghake, D. B.; Parthasarathy, S.; Superko, H. R. 27th Bethesda Conference: matching the intensity of risk factor management with the hazard for coronary disease events. Task Force 4. Efficacy of risk factor management. J. Am. Coll. Cardiol. 27:991– 1006; 1996. 5. Esterbauer, H.; Jurgens, G.; Quehenberger, O.; Koller, E. Autoxidation of human low density lipoprotein: loss of polyunsaturated fatty acids and vitamin E and generation of aldehydes. J. Lipid Res. 28:495–509; 1987. 6. Steinbrecher, U. P. Oxidation of human low density lipoprotein results in derivatization of lysine residues of apolipoprotein B by lipid peroxide decomposition products. J. Biol. Chem. 262:3603–3608; 1987. 7. Shechter, I.; Fogelman, A. M.; Haberland, M. E.; Seager, J.; Hokom, M.; Edwards, P. A. The metabolism of native and malondialdehyde-altered low density lipoproteins by human monocytemacrophages. J. Lipid Res. 22:63–71; 1981. 8. Haberland, M. E.; Fong, D.; Cheng, L. Malondialdehyde-altered protein occurs in the atheroma of Watanabe heritable hyperlipidemic rabbits. Science 241:215–218; 1988. 9. Jurgens, G.; Ashy, A.; Esterbauer, H. Detection of new epitopes formed upon oxidation of low-density lipoprotein, lipoprotein (a) and very-low-density lipoprotein. Use of an antiserum against 4-hydroxynonenal-modified low-density lipoprotein. Biochemical J. 265:605–608; 1990. 10. Palinski, W.; Rosenfeld, M. E.; Yla¨ Herttuala, S.; Gurtner, G. C.; Socher, S. S.; Butler, S. W.; Parthasarathy, S.; Carew, T. E.; Steinberg, D.; Witztum, J. L. Low density lipoprotein undergoes oxidative modification in vivo. Proc. Natl. Acad. Sci. USA 86:1372–1376; 1989. 11. Fletcher, B. L.; Tappel, A. L. Fluorescent modification of serum albumin by lipid peroxidation. Lipids 6:172–175; 1971. 12. Fong, L. G.; Parthasarathy, S.; Witztum, J. L.; Steinberg, D. Nonenzymatic oxidative cleavage of peptide bonds in apoprotein B100. J. Lipid Res. 28:1466–1477; 1987. 13. Davies, K. J. Protein modification by oxidants and the role of proteolytic enzymes. Biochem. Soc. Trans. 21:346 – 353; 1993. 14. Stadtman, E. R. Oxidation of free amino acids and amino acid residues in proteins by radiolysis and by metal-catalyzed reactions. Annu. Rev. Biochem. 62:797–821; 1993.
/ 2b2c 2469 Mp 258 Monday Jun 02 10:16 PM EL–FRB 2469
Antibody to LOOH-modified proteins 15. Levine, R. L.; Williams, J. A.; Stadtman, E. R.; Shacter, E. Carbonyl assays for determination of oxidatively modified proteins. Methods Enzymol. 233:346–357; 1994. 16. Heinecke, J. W.; Li, W.; Francis, G. A.; Goldstein, J. A. Tyrosyl radical generated by myeloperoxidase catalyzes the oxidative cross-linking of proteins. J. Clin. Invest. 91:2866 – 2872; 1993. 17. Uchida, K.; Toyokuni, S.; Nishikawa, K.; Kawakishi, S.; Oda, H.; Hiai, H.; Stadtman, E. R. Michael addition-type 4-hydroxy2-nonenal adducts in modified low-density lipoproteins: markers for atherosclerosis. Biochem. 33:12487–12494; 1994. 18. Uchida, K.; Stadtman, E. R. Modification of histidine residues in proteins by reaction with 4-hydroxynonenal. Proc. Natl. Acad. Sci. USA 89:4544–4548; 1992. 19. Fruebis, J.; Parthasarathy, S.; Steinberg, D. Evidence for a concerted reaction between lipid hydroperoxides and polypeptides. Proc. Natl. Acad. Sci. USA 89:10588–10592; 1992. 20. Parthasarathy, S. Lipid peroxidation-induced protein modifications in the oxidation of low density lipoprotein. In: Packer, L.; Cadenas, E., eds. Biological oxidants and antioxidants. Stuttgart: Hippokrates Verlag; 1994:255–259. 21. Bligh, E. G.; Dyer, W. J. Rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37:911–917; 1959. 22. Santanam, N.; Parthasarathy, S. Paradoxical actions of antioxidants in the oxidation of low density lipoprotein by peroxidases. J. Clin. Invest. 95:2594–2600; 1995. 23. Parthasarathy, S.; Fong, L. G.; Otero, D.; Steinberg, D. Recognition of solubilized apoproteins from delipidated, oxidized low
24.
25.
26.
27.
28. 29.
259
density lipoprotein (LDL) by the acetyl-LDL receptor. Proc. Natl. Acad. Sci. USA 84:537–540; 1987. Steinbrecher, U. P.; Witztum, J. L.; Parthasarathy, S.; Steinberg, D. Decrease in reactive amino groups during oxidation or endothelial cell modification of LDL. Correlation with changes in receptor-mediated catabolism. Arteriosclerosis 7:135–143; 1987. Steinbrecher, U. P.; Parthasarathy, S.; Leake, D. S.; Witztum, J. L.; Steinberg, D. Modification of low density lipoprotein by endothelial cells involves lipid peroxidation and degradation of low density lipoprotein phospholipids. Proc. Natl. Acad. Sci. USA 81:3883–3887; 1984. Houglum, K.; Filip, M.; Witztum, J. L.; Chojkier, M. Malondialdehyde and 4-hydroxynonenal protein adducts in plasma and liver of rats with iron overload. J. Clin. Invest. 86:1991–1998; 1990. Rosenfeld, M. E.; Palinski, W.; Yla Herttuala, S.; Butler, S.; Witztum, J. L. Distribution of oxidation specific lipid-protein adducts and apolipoprotein B in atherosclerotic lesions of varying severity from WHHL rabbits. Arteriosclerosis 10:336–349; 1990. Staprans, I.; Pan, X. M.; Miller, M.; Rapp, J. H. Effect of dietary lipid peroxides on metabolism of serum chylomicrons in rats. Am. J. Physiol. 264:G561–8; 1993. Staprans, I.; Rapp, J. H.; Pan, X. M.; Feingold, K. R. The effect of oxidized lipids in the diet on serum lipoprotein peroxides in control and diabetic rats. J. Clin. Invest. 92:638 – 643; 1993.
/ 2b2c 2469 Mp 259 Monday Jun 02 10:16 PM EL–FRB 2469