| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Lipid Research, Vol. 45, 1859-1867, October 2004
Copyright © 2004 by American Society for Biochemistry and Molecular Biology
* Department of Microbiology and Immunology, Medical University of South Carolina, Charleston, SC
Department of Chemistry and Biochemistry, University of South Carolina, Columbia, SC
Department of Medicine, Division of Endocrinology-Metabolism-Nutrition, Medical University of South Carolina, and Ralph H. Johnson Department of Veteran Affairs Medical Center, Charleston, SC
Published, JLR Papers in Press, July 16, 2004. DOI 10.1194/jlr.M400095-JLR200
1 To whom correspondence should be addressed. e-mail: virellag{at}musc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
(carboxymethyl)lysine-modified LDL (CML-LDL) as well as antibodies reacting with MDA-LDL, copper-oxidized LDL (OxLDL), CML-LDL, and advanced glycosylation end product (AGE)-modified LDL. OxLDL and AGE-LDL antibodies isolated from human sera recognize the same LDL modifications and do not react with modified non-LDL proteins. Rabbit antibodies have different reactivity patterns: MDA-LDL antibodies react strongly with MDA-LDL and MDA-BSA but weakly with OxLDL; OxLDL antibodies react strongly with OxLDL and weakly with MDA-LDL; CML-LDL antibodies react with CML-LDL > CML-BSA > AGE-LDL > OxLDL; AGE-LDL antibodies react strongly with AGE-LDL, react weakly with OxLDL, and do not react with CML-LDL. Thus, human and rabbit antibodies seem to recognize different epitopes. Capture assays carried out with all rabbit antibodies showed binding of apolipoprotein B-rich lipoproteins isolated from ICs, suggesting that laboratory-generated epitopes are expressed by in vivo-modified LDL, although they are not necessarily recognized by the human immune system. Thus, the definition of immunogenic forms of modified LDL eliciting human autoimmune responses requires the isolation and characterization of autoantibodies and modified LDL from human samples, whereas rabbit antibodies can be used to detect in vivo-modified human LDL.
Supplementary key words low density lipoprotein modifications • low density lipoprotein autoantibodies • modified low density lipoprotein antibodies • immunology of atherosclerosis • autoimmunity • atherosclerosis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
LDL oxidation affects both the lipid and protein components of LDL. Reactive aldehyde products formed during the oxidation of polyunsaturated fatty acids, such as malondialdehyde (MDA) and 4-hydroxynonenal (HNE), are capable of attaching covalently to the -amino groups of lysine residues of apolipoprotein B (apoB) (7–9). These modifications are present in OxLDL as well as in LDL isolated from atherosclerotic plaques, which reacts with monoclonal antibodies produced in guinea pigs against MDA and HNE-lysine (9, 10).
Detailed investigations have also been carried out with advanced glycosylation end product (AGE)-modified LDL. Advanced glycosylation involves a chain of chemical reactions that starts with the nonenzymatic addition of reducing sugars to protein amino groups (Schiff base, Amadori adducts). If the half-life of a protein is sufficiently long, additional reactions take place leading to the formation of a heterogeneous family of sugar-amino acid adducts collectively known as "advanced glycosylation end products" (11). LDL, like most plasma proteins, is susceptible to AGE modification (12). AGE-modified proteins are immunogenic (13), and the corresponding antibodies raised in laboratory animals have been used for the detection of AGE-modified proteins in serum (14) and tissues (14, 15). In humans, AGE-LDL antibodies have been purified from sera with high antibody titers and characterized as far as isotype distribution, avidity, and specificity (16). Of the two best characterized modifications of AGE proteins, N (carboxymethyl)lysine (CML) and N (carboxyethyl)lysine (CEL), CML seems to be the most immunogenic.
The comparison of data reported in the literature, based on animal immunization studies, with data generated from the study of purified human and mouse OxLDL autoantibodies suggests that MDA-lysine is a most significant epitope (17, 18), but significant differences seem to exist between the specificity of human and hyperimmune animal antibodies to modified lipoproteins. For example, Steinbrecher et al. (3) showed that immune guinea pig antibodies to carbamylated, acetylated, or ethylated autologous LDL reacted with the corresponding modified lysine residues and that the reactivity extended to other modified non-LDL proteins. Similarly, it has been reported that antibodies raised in rabbits against one AGE-modified protein react with other AGE-modified proteins (19). In contrast, affinity chromatography-purified human AGE-LDL antibodies react primarily with AGE-LDL, cross-react strongly with CML-LDL and OxLDL, but show very little to minimal reactivity with CML-human serum albumin (HSA) (16). These discrepancies raise important questions about species differences in epitope recognition and about the nature of the epitopes recognized by spontaneously formed human autoantibodies and antibodies induced in animals by immunization with chemically modified LDL. To resolve some of these questions, we have characterized the nature of the predominant modifications in human apoB-rich lipoproteins obtained from antigen-antibody complexes isolated from human sera and undertaken a comparative study of the reactivity of purified human autoantibodies to OxLDL and AGE-LDL with antibodies obtained in rabbits immunized with different modifications of human LDL.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Preparation of modified proteins
OxLDL was prepared according to our modification (21) of the protocol described by Steinbrecher (8). Human LDL was diluted in PBS to a concentration of 1,500 mg/l apoB and incubated with 40 µmol/l CuCl2. The degree of oxidation was monitored continuously by fluorescence emission using a fluorescence spectrophotometer (SLM-AMINCO® Series 2; Spectronic Instruments, Rochester, NY). LDL oxidation was stopped 4–6 h after the fluorescence values reached the peak (
1.1 fluorescence units).
AGE-LDL was prepared by a modification of the method described by Schmidt et al. (22). Freshly isolated LDL (1.2 mg/ml) was sterilized by passage through a 0.2 µm filter, added to 150 mM glucose-6-phosphate in 200 mM phosphate buffer, pH 7.8, containing 40 µM butyl-hydroxytoluene (BHT) and 540 µM EDTA, filter-sterilized a second time, and incubated for 8 weeks at 37°C. At the end of the incubation, the LDL solution was dialyzed for 24 h against three changes of 4 liters of 0.15 M NaCl, 0.3 mM EDTA, pH 8.0. Our AGE-LDL modifications contained CML and CEL (Table 1). The MDA-lysine content of AGE-LDL was usually below the detection limit, but occasionally small amounts of MDA-lysine were detected.
|
CML-modified proteins were prepared by incubation of the protein with glyoxylic acid and NaBH3CN in phosphate buffer at 37°C, as described previously for the preparation of CML-BSA (24). In each case, control protein was incubated under the same conditions but without glyoxylic acid.
Analysis of LDL modifications
Analysis of modified lipoproteins for their content of CML, CEL, and the advanced lipoxidation end products (ALEs) MDA-lysine and HNE-lysine was carried out by selected ion monitoring (SIM)-GC-MS as described by Requena et al. (11). Briefly, these four AGEs/ALEs were measured simultaneously in modified LDLs (AGE-LDL, OxLDL) and in LDL isolated from human serum (0.5–1.0 mg of protein) after borohydride reduction of the protein. The reduced sample was dialyzed, dried, delipidated [chloroform-methanol (2:1) containing 0.02% BHT], and, after the addition of heavy labeled internal standards, hydrolyzed at 110°C in 6 N HCl for 18 h. The hydrolysate was dried, passed over a 1 ml solid phase extraction C18-column (Sep-Pak; Waters, Milford, MA), and dried again, and the amino acids were derivatized as their trifluoroacetyl methyl esters for analysis by SIM-GC-MS (11, 14, 25). Quantification was performed by isotope dilution, and all modified amino acids are expressed as a ratio to the parent amino acid, lysine.
The degree of modification of MDA-LDL was estimated by the thiobarbituric acid reactive substances assay (26) using MDA as a standard. The extent of modification of CML-LDL and CML-albumins was measured by amino acid analysis (24).
Rabbit antibodies
Antibodies to modified LDL were obtained by immunization of New Zealand White female rabbits with different modifications of human LDL (AGE-LDL, CML-LDL, OxLDL, and MDA-LDL). The degree and nature of modification of the lipoproteins used for immunization are summarized in Table 1. Two to four rabbits were immunized per each modified human LDL. The lipoproteins were diluted in sterile saline to 2.5 mg/ml. Each LDL preparation was emulsified with an equal volume of complete Freund's adjuvant immediately before intramuscular inoculation. Each rabbit was immunized with 1 ml of emulsion. Two weeks later, the rabbits were inoculated a second time with an identical amount of modified LDL emulsified with incomplete Freund's adjuvant by the same route. A third inoculation was administered 2 weeks later using antigen diluted in saline and injected subcutaneously in four divided doses placed in the dorsal area. Two weeks after the third injection, the rabbits were bled and euthanized. The immunization and bleeding protocols were approved by the university's Institutional Animal Care and Use Committee.
The sera obtained at exsanguination, containing a mixture of antibodies to apoB and to the modified LDL, were fractionated on a protein G-Sepharose column. The eluted IgG fraction was then absorbed on a column of native LDL-Sepharose. The washout from that column contained a mixture of IgG antibodies to modified LDL and irrelevant IgG. The pattern of reactivity with the different modified LDL preparations and controls was consistent for each set of rabbits immunized with a given modified LDL. However, there were variations in antibody titer from rabbit to rabbit, and the antiserum with a higher titer was selected for further characterization.
An alternative protocol was designed to determine whether immobilized OxLDL contained CML epitopes. In this case, the IgG fraction from a CML-LDL antiserum was absorbed first on an immobilized native LDL column and the washout was fractionated on an OxLDL column. After washing the OxLDL column extensively with 0.01 M sodium bicarbonate buffer, pH 8.3, bound antibodies were eluted with 0.1 M sodium bicarbonate buffer containing 0.5 M NaCl, pH 8.3, and tested for reactivity with CML-LDL.
Determination of antibody specificity
The specificity of purified human and rabbit antibodies was determined through a variety of approaches. The basic pattern of reactivity for both human and rabbit antibodies was established by competition studies, in which different antigens were preincubated with a given antibody to determine their ability to inhibit binding with immobilized antigens. The inhibition studies were designed as modifications of our previously reported competitive enzymoimmunoassay (EIA) for oxidized LDL antibodies (27). The EIA plates were coated with 0.75 µg/well of protein, including modified forms of LDL (OxLDL, AGE-LDL, MDA-LDL, or CML-LDL), other modified proteins (MDA-HSA or CML-BSA), and native LDL, according to the experiment being performed. In all competition studies, the proteins preincubated with the different antibodies were adjusted to a final concentration of 200 µg/ml.
In addition, capture assays were carried out in experiments designed to complement the competition studies. These assays were carried out with purified rabbit IgG antibodies of different specificities absorbed to the plates at a 1:10 dilution. After washing off the unabsorbed antibodies, the plates were blocked with 5% BSA. Serial dilutions of modified LDL preparations were added to the plates, and peroxidase-conjugated goat anti-rabbit apoB (1:1,000 dilution; Academy Bio-Medical, Houston, TX) was used to detect bound LDL. Capture assays were also used to determine whether the rabbit antibodies had the potential to capture in vivo-modified LDL, obtained as described below.
Comparison of the degree of modification of soluble apoB-rich lipoproteins versus apoB-rich lipoproteins precipitated with 4% PEG
Fractionation of human sera with 4% PEG was performed as previously described (16). ApoB-containing lipoproteins were isolated from the supernatant and from the resuspended PEG precipitates. The supernatants were fractionated directly on heparin-agarose columns (Sigma-Aldrich Corp., St. Louis, MO) that retain apoB-containing lipoproteins. The PEG precipitates were first submitted to affinity chromatography on protein G-Sepharose, and the washout, containing all precipitated proteins other than IgG, was then fractionated on heparin-agarose columns. The lipoprotein-containing samples were pooled and dialyzed against saline containing 0.3 mM EDTA, pH 8.0, and analyzed for modifications as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
3-fold enrichment in the precipitated compared with the soluble proteins. The difference was even more marked in the case of MDA-lysine, which was rarely detected in soluble lipoproteins (6 of 37 samples), whereas it was always detected in precipitated lipoproteins (P < 0.0001). HNE-lysine was analyzed but not detected.
|
|
|
|
|
|
|
2-fold. As previously reported, OxLDL contains MDA-lysine, CML-lysine, and HNE-lysine (5), raising the possibility that the CML-LDL antibodies could recognize CML-lysine epitopes in OxLDL. To clarify this issue, the IgG fraction from a rabbit CML-LDL antibody was first absorbed with immobilized native LDL to eliminate apoB antibodies and then fractionated on a column of immobilized OxLDL. The antibody peak eluted from the OxLDL column was tested in a direct binding assay against different LDL modifications as well as native LDL and was found to react strongly and exclusively with CML-LDL. Capture assays confirmed the exquisite specificity of rabbit anti-CML antibodies for the CML-LDL preparation used as an immunogen (Fig. 6 , inset). At relatively high concentrations, AGE-LDL and OxLDL were also captured (Fig. 6).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
To define the immunogenic modifications of LDL that elicit spontaneous autoimmune responses in humans, we characterized the modified forms of LDL and the antibodies purified from isolated ICs. Our data indicate that a minimum of two antigen-antibody systems are involved in the formation of modified LDL-ICs. The significant enrichment of PEG-precipitated ICs in MDA-LDL and the detection of antibodies reacting with both OxLDL, which contains significant concentrations of MDA (5), and MDA-LDL indicate that the MDA modification of LDL is recognized by the human immune system. The equally significant enrichment of PEG precipitates in CML-LDL and the detection of antibodies reacting with AGE-modified LDL, which contains significant amounts of CML-LDL (16), and with CML-LDL indicate that CML-LDL is immunogenic for humans.
The pattern of reactivity of OxLDL and AGE-LDL antibodies isolated from serum was similar to that of antibodies isolated from PEG-precipitated ICs. Of interest is the fact that human OxLDL antibodies are highly specific for OxLDL and MDA-LDL, showing minimal cross-reactivity with MDA-BSA and virtually no reactivity with native LDL. This pattern of reactivity is similar to that reported by other investigators with a human MDA-LDL antibody (18), which reacted both with MDA-LDL and OxLDL and showed a low degree of cross-reactivity with MDA-BSA. On the other hand, AGE-LDL antibodies showed a marked preference for AGE-LDL relative to CML-LDL and cross-reacted with OxLDL. The preferential recognition of AGE-LDL over CML-LDL and the total lack of reactivity with CML-BSA suggests that human AGE-LDL antibodies recognize LDL-specific epitope(s) and that the configuration of the CML epitope(s) may not be identical in AGE-LDL and CML-LDL, perhaps as a consequence of different degrees of modification. In fact, apoB-containing lipoproteins separated as ICs from patient sera contain significantly lower levels of CML per mole of lysine than laboratory-generated AGE-LDL; higher degrees of CML modification could interfere by charge effects or by steric hindrance with the reactivity of AGE-LDL autoantibodies. The reactivity of human AGE-LDL antibodies with OxLDL seems to be a result of the presence of low concentrations of CML in OxLDL, concentrations very similar to those measured in the apoB-containing lipoproteins coprecipitated with ICs.
The reactivity of rabbit antibodies obtained after immunization with human MDA-LDL and human OxLDL was quite different from that of human OxLDL antibodies. Although rabbit MDA-LDL antibodies reacted equally well with different MDA-modified proteins and cross-reacted with OxLDL, human OxLDL antibodies seem to recognize MDA-lysine epitopes shared by OxLDL and MDA-LDL but not by other MDA-modified proteins.
Rabbit OxLDL antibodies had different patterns of reactivity when tested in competition and capture assays, probably as a consequence of the existence of several populations of antibodies reacting with different epitopes with different affinities. The competition assays, in which the binding of a given antibody to the corresponding antigen is inhibited by the addition of different antigens in relatively high concentrations, favor the detection of low-affinity antibodies. In contrast, capture assays require a higher affinity interaction between antigen and antibody, sufficient to keep the antigen and antibody as a complex through two washing cycles. Thus, low-affinity antibodies to MDA and CML-LDL would explain the extensive cross-reactivity with different types of modified LDL and native LDL. MDA-lysine epitopes are present in MDA-LDL and OxLDL (5), whereas CML-LDL epitopes can be detected in OxLDL, AGE-LDL, and native LDL, albeit over a wide range of concentrations (5, 16).
On the other hand, antibodies of higher affinity to a unique epitope of OxLDL would explain the results observed in the capture assays that indicate that rabbits recognize immunogenic structures of OxLDL that are not shared by MDA-LDL. A possible explanation is that MDA-LDL has two sets of epitopes, one shared with OxLDL that is immunodominant in humans, and one that is shared by different MDA-modified proteins and is immunodominant in rabbits. The higher degree of modification of MDA-LDL (20–25% substitution of lysine residues) may create clusters of MDA lysine that rabbits recognize as an epitope present in other MDA-modified proteins but absent in less modified OxLDL (0.4–0.7% substitution of lysine residues). The epitope recognized by rabbit OxLDL antibodies should also be different from the MDA-related epitope recognized by human OxLDL antibodies, because human antibodies react with both OxLDL and MDA-LDL. A second possibility to be considered is that rabbit OxLDL antibodies recognize epitopes not related to MDA-lysine, such as the one described by Palinski and coworkers (28).
Human AGE-LDL autoantibodies and rabbit AGE-LDL antibodies are protein-specific, because they react primarily with AGE-LDL and cross-react with CML-LDL but do not react with CML-HSA. Immunization of rabbits with human CML-LDL seems to result in the production of more than one antibody population. When tested by competition against immobilized CML-LDL and by capture assay, we detected a high-affinity antibody reactive primarily with CML-LDL but also reacting strongly with CML-BSA, moderately with AGE-LDL, and weakly with OxLDL. The lack of protein specificity of CML antibodies has been reported by other groups (19). When tested against immobilized AGE-LDL, the preferential reactivity with CML-LDL is not obvious and the cross-reactivity with OxLDL is enhanced, suggesting that the assay detects lower affinity antibodies that cross-react more extensively. The cross-reactivity with OxLDL seems to be a consequence of the generation of CML during copper oxidation of LDL, as proven by the isolation of CML-LDL antibodies from an immobilized LDL column.
Rabbit AGE-LDL antibodies are considerably more specific for AGE-LDL than are human antibodies (Table 3), although the capture assay showed definite cross-reactivity with OxLDL. The cross-reactivity with OxLDL could also be explained by the presence of CML in OxLDL, although this is not supported by the lack of reactivity with CML-LDL. Differences in epitope configuration related to the extent of CML modification could be responsible for the different reactivity with CML epitopes in OxLDL and CML-LDL, but it is also possible that rabbit AGE-LDL antibodies may recognize a non-CML epitope also expressed by OxLDL.
The differences in reactivity between rabbit antibodies to different modifications of LDL and human autoantibodies may result from differences between the epitopes generated during spontaneous oxidation in vivo and the epitopes generated by chemical modification of lipoproteins in the laboratory, as discussed above. The possibility that immunogenic epitopes other than MDA may be present in OxLDL and epitopes other than CML may be present in AGE-LDL is strongly suggested by the data generated with rabbit antibodies and needs to be further investigated. But the data generated by capture assays showing that modified lipoproteins isolated from PEG-precipitated ICs were specifically captured with all of the different rabbit modified LDL antibodies tested strongly suggests that there are species-specific differences in the recognition of different epitopes of modified forms of LDL. Thus, the definition of the immunogenic modifications and epitopes recognized by the human immune system has to be based on the characterization of human autoantibodies rather than on the study of antibodies induced in laboratory animals with in vitro-modified LDL. On the other hand, when developing antibodies for the detection of modified lipoproteins in human samples, laboratory-modified lipoproteins seem to be perfectly adequate, because the epitopes of in vitro-modified LDL recognized by rabbits are also present in modified LDL isolated from human sera.
|
ACKNOWLEDGMENTS |
|---|
Manuscript received March 8, 2004 and in revised form May 20, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. F. Saad, G. Virella, C. Chassereau, R. J. Boackle, and M. F. Lopes-Virella OxLDL immune complexes activate complement and induce cytokine production by MonoMac 6 cells and human macrophages J. Lipid Res., September 1, 2006; 47(9): 1975 - 1983. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Virella, M. B. Derrick, V. Pate, C. Chassereau, S. R. Thorpe, and M. F. Lopes-Virella Development of Capture Assays for Different Modifications of Human Low-Density Lipoprotein Clin. Vaccine Immunol., January 1, 2005; 12(1): 68 - 75. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Journal of Biological Chemistry |
| Molecular and Cellular Proteomics | ASBMB Today |
