Atheroprotective immunization with malondialdehyde-modified LDL is hapten specific and dependent on advanced MDA adducts: implications for development of an atheroprotective vaccine

Abstract

Immunization with homologous malondialdehyde (MDA)-modified LDL (MDA-LDL) leads to atheroprotection in experimental models supporting the concept that a vaccine to oxidation-specific epitopes (OSEs) of oxidized LDL could limit atherogenesis. However, modification of human LDL with OSE to use as an immunogen would be impractical for generalized use. Furthermore, when MDA is used to modify LDL, a wide variety of related MDA adducts are formed, both simple and more complex. To define the relevant epitopes that would reproduce the atheroprotective effects of immunization with MDA-LDL, we sought to determine the responsible immunodominant and atheroprotective adducts. We now demonstrate that fluorescent adducts of MDA involving the condensation of two or more MDA molecules with lysine to form malondialdehyde-acetaldehyde (MAA)-type adducts generate immunodominant epitopes that lead to atheroprotective responses. We further demonstrate that a T helper (Th) 2-biased hapten-specific humoral and cellular response is sufficient, and thus, MAA-modified homologous albumin is an equally effective immunogen. We further show that such Th2-biased humoral responses per se are not atheroprotective if they do not target relevant antigens. These data demonstrate the feasibility of development of a small-molecule immunogen that could stimulate MAA-specific immune responses, which could be used to develop a vaccine approach to retard or prevent atherogenesis.

Keywords: adaptive immunity; humoral immunity; malondialdehyde-acetaldehyde adducts; oxidation-specific epitopes.

Fig. 1.

Fluorometric characterization of fluorescent MAA-MSA, nonfluorescent MDA-MSA, and other immunogens showing the enrichment of MAA-type adducts in MAA-MSA and to a lesser extent in MDA-mLDL. Emission spectrum (and insert for lower fluorescent unit (FLU) signal range) upon excitation at 394 nm at 10 μg/ml of protein shows the characteristic maxima at 462 nm for the 1,4-dihydropyridine-3,5-dicarbaldehyde-type adducts. Formation of such adducts can be enriched through an MAA-modification, and one of the major products is the MDHDC-lysine adduct (schematically illustrated on top right). The fluorescence intensity at 462 nm was ∼20-fold and 100-fold higher in MAA-MSA than in MDA-mLDL and MDA-MSA preparations, respectively. PA-MSA (theoretical major adduct is schematically illustrated on bottom right) and unmodified MSA and mLDL did not yield MAA-specific fluorescence. Note that although we aimed at generating totally nonfluorescent MDA-MSA, we were not able to modify >80% of the lysines in MSA without generating trace amounts of MAA-type adducts.

Fig. 2.

Only MDA-modified lysine-containing peptides compete for binding to MDA-LDL of antisera generated by immunization with MDA-mLDL. A 1:80,000 dilution of pooled plasma from MDA-mLDL-immunized C57Bl/6J mice was preincubated with the indicated competitors and controls prior to detection of IgG1 binding to plated MDA-LDL. Unmodified LDL, poly-lysine (pLys), and peptides did not compete. MDA-LDL and MDA-modified polylysine (MDA-pLys) competed >90% of binding. The three lysine-containing peptides competed for up to 80% of IgG1 binding in an MDA-dependent manner, whereas MDA-modified peptides devoid of lysine did not compete beyond the nonspecific effects shown with a peptide-blank negative control (MDA modification without a peptide acceptor), as indicated with broken horizontal line. To increase the extent of MDA modification of the peptides, they were modified at indicated peptide-to-MDA molar ratios (mol/mol) at pH 7.4 for 60 min. Data are shown as B/B₀ of triplicates for a representative experiment of two.

Fig. 3.

MAA-MSA immunization confers atheroprotection. MAA-MSA immunization significantly reduced lesion area by 37% to PBS-treated mice and by 26%–29% to PA-MSA-, MSA-, and FA-immunized mice. Plot shows surface lesion area of the entire aorta relative to total aortic area (mean ± SEM; n = 15–19, as indicated within each bar). The right y-axis shows the degree of reduced surface lesion formation compared with the PBS mice. Statistical differences were analyzed using ANOVA with Newman-Keuls multiple comparisons posttest. P values are given in the plot.

Fig. 4.

Antigen-specific IgG1-to-IgG2c ratios demonstrate Th2-biased responses in immunized Ldlr^−/− mice. The y-axis shows the IgG1/IgG2c ratio determined on terminal plasma from different immunized mice as shown in supplementary Fig. II. The x-axis indicates the immunization group that was the source of plasma. Each of the charts in A–E indicate the antigen to which titers were measured: MDA-LDL (A), MAA-BSA (B), MDA-BSA (C), PA-BSA (D), and CuOxLDL (E). Ratios were calculated using the same plasma dilution within an immunization group, although different dilutions may have been used as more appropriate for a given antigen/immunization group (most ratios were calculated at 1:10,000 plasma dilution). For the ratio to be meaningful, both IgG1 and IgG2c binding were selected to be in the linear range. Data are given as mean ± SEM of 15–20 mice per group as indicated in Table 1.

Fig. 5.

MAA-MSA immunization induces and sustains an IgG1-dominant hapten-specific humoral immune response against MDHDC-lysine, demonstrating that MDHDC-lysine is an immunodominant epitope. A–F: Terminal antisera were pooled according to immunization groups, formally diluted, and tested for IgG1 (A, B), IgG2c (C), IgG2b (D), IgG3 (E), and IgM (F) binding to a semisynthetic MAA-modified biotinylated GDGDGK peptide [Bt-GDGDG-K(MAA)] (B–F) or the control peptide (A) by ELISA. The Bt-GDGDG-K(MAA) peptide was captured with plate-immobilized avidin to ensure reproducible epitope presentation and density. Data are curves of averaged triplicate determination in RLUs. G, H: A fixed 1:15,000 dilution of pooled terminal MAA-MSA antisera was preincubated in the presence or absence of synthetic GDGDG-K(MDHDC) (green diamonds; chemical structure shown in I) or GDGDG-K control (circles) peptide prior to direct plating of mixture on immobilized Bt-GDGDG-K(MAA) or MAA-BSA and detection of IgG1 binding by competition ELISA. Synthetic GDGDG-K(MDHDC) completely competes all IgG1 binding to immobilized Bt-GDGDG-K(MAA) illustrating that binding is specific to MDHDC-lysine. IgG1 binding to the more complex MAA-BSA antigen is less efficiently competed by GDGDG-K(MDHDC), showing that the MAA-MSA antisera contains Abs specific to MAA epitopes other than the MDHDC-lysine. Data are B/B_0, where each data point is the average of triplicate determination.

Fig. 6.

Pooled MAA-MSA antisera compete for modified LDL binding to macrophages. A: In vitro binding experiment showing that biotinylated MDA-LDL (circles), CuOxLDL (squares), and MAA-LDL (black/filled/solid diamonds) bound in a dose-dependent and saturable manner to J774 macrophages, whereas native LDL (reversed triangles) and PA-LDL (triangles) did not. Data are RLUs per 100 ms. Plot is a representative experiment of more than four. B: MAA-LDL, but not native LDL, competes for Bt-MAA-LDL (1.5 μg/ml) binding to J774 macrophages. Data are B/B₀ for a representative experiment of more than four. C: Plasma dilution curves showing that pooled antisera from MAA-MSA-immunized mice, but not PA-MSA-immunized mice, compete for Bt-MAA-LDL (2.5 μg/ml) binding to J774 macrophages. Data are mean ± SD of B/B₀ from two independent experiments for a total of four replicates. $, PA-MSA versus no antisera, P < 0.05; §, MAA-MSA versus no antisera, P < 0.0001; #, MAA-MSA versus PA-MSA, P < 0.001. D: The pooled antisera from MAA-MSA-immunized mice compete significantly more Bt-MAA-LDL (2.5 μg/ml) binding to J774 macrophages than the other six pooled antisera at 1:100. Data are mean ± SD of B/B₀ from two independent experiments for a total of four replicates. Statistical differences were analyzed using ANOVA with Bonferroni’s multiple comparisons posttest. P values are given in plot.

Fig. 7.

Cellular and systemic Th2-biased immunity in MAA-MSA-immunized Ldlr^−/− mice after 28 weeks of cholesterol feeding. A-C: Impaired splenic capacity for IFN-γ release (A), but not IL-5 release (B), in Ldlr^−/− mice immunized with modified immunogens. Mice were euthanized (7 weeks after last boost), and single-cell suspensions of whole spleens were cultured on plate-bound anti-CD3, to induce nonspecific maximal release of cytokines by memory Th cells. C: Plotted IFN-γ/IL-5 ratios show the most pronounced impairment of IFN-γ release capacity in the MAA-MSA-immunized mice. Data are mean ± SEM of four to nine mice as indicated in bars. To increase the statistical power, the three control groups (MSA, FA, and PBS; no difference according to separate ANOVA) were combined. D–G: Pronounced and concentration-dependent MAA-MSA-specific IL-5 release by splenocytes from MAA-MSA-immunized Ldlr^−/− mice. Mice immunized with indicated antigen are shown above each panel, and respective splenocytes were stimulated with indicated antigens and released IL-5 (D, F) and IFN-γ (E, G) quantified by ELISA. Splenocytes from MDA-mLDL-immunized mice responded to MDA-mLDL recall stimulation with mixed IL-5 (D; five of five mice) and IFN-γ (E; three of five mice) release. Splenocytes from MAA-MSA-immunized mice responded to MAA-MSA with an IL-5 dominated (F; four of five mice) release over IFN-γ (G; two of five mice). Data are mean ± SEM. P values are given in plots as * P < 0.05 and ** P < 0.01 and ns (not significantly different).

Fig. 8.

Plasma cytokines indicate strong systemic Th2-biased immunity in MAA-MSA-immunized Ldlr^−/− mice after 28 weeks of cholesterol feeding. Plasma IFN-γ (A) is markedly reduced in MAA-MSA compared with MDA-mLDL-, MDA-MSA-, and PA-MSA-immunized Ldlr^−/− mice, whereas plasma IL-5 (B) is identical. C: MAA-MSA-immunized mice also have a markedly reduced IFN-γ/IL-5 ratio. Plasma cytokines were reported as box plots of 15–20 mice/group (as indicated in Table 1, except that only 11 of 19 PBS plasmas were analyzed). P values are given in plots as * P < 0.05; ** P < 0.01; and *** P < 0.001.

Fig. 9.

Th2-biased response in MAA-MSA-immunized C57BL/6J mice. Mice were immunized with homologous MAA-MSA and relevant controls in FA, and 1 week after the third injection, a single-cell suspension of whole spleen was cultured with various stimuli (n = 8 per group). A, B: Concentration-dependent antigen-specific release after 72 h of IL-5 (A) and IFN-γ (B) in eight of eight MAA-MSA-immunized mice, but not in PA-MSA-immunized (IL-5 and IFN-γ: zero and four of eight mice, respectively) and MSA-immunized (IL-5 and IFN-γ: zero and two of eight mice, respectively) mice, upon in vitro recall stimulation with indicated immunogens (mean ± SEM). C: To “normalize” antigen-specific cytokine release, we plotted the relative cytokine release in response to antigen stimulation with 50 μg/ml MAA-MSA + anti-CD28 divided by cytokine release in parallel cultures stimulated with anti-CD3 and anti-CD28. D: Increased frequency of MAA-MSA-specific IL-5 secreting cells in spleens of MAA-MSA-immunized mice as assessed by ELISpot assay. Splenocytes (2 × 10⁶) of immunized and control mice were incubated overnight with 50 μg/ml immunogen (FA splenocytes were stimulated with MSA) or with a limiting concentration of the nonspecific stimuli Concanavalin A (1 μg/ml), and the frequencies of IL-5 and IFN-γ SFCs were assessed. The plot shows individual mice as the number of antigen-specific SFC divided by number of Concanavalin A SFCs, after subtraction of background spots.

Fig. 10.

Human MAA-specific Fab mAb stains human carotid artery and human natural IgM Abs in umbilical cord plasma bind specifically to MAA epitopes. A, B: Immunohistochemical staining of human carotid artery with an MAA-specific human Fab NAb cloned from a phage display library derived from human umbilical cord blood (A) or without a primary Ab (B). Bound Fab was detected with a secondary anti-histidine antigen tag-Bt conjugate. MAA epitopes in the carotid section are indicated by red color, and nuclei are counterstained with hematoxylin. C: Plot shows a representative competition ELISA with pooled human umbilical cord plasma from seven newborns. Binding of the human natural IgM Abs to plated MAA-BSA was efficiently competed with MAA-BSA (>98%) and MAA-LDL (∼90%), less efficiently with MDA-LDL (∼55%), and not at all with nonfluorescent MDA-BSA, PA-BSA, or native BSA and LDL. We estimated the dissociation constants (K_d) using the Klotz method and found strong avidities of human natural IgM to MAA-BSA; competition with MAA-BSA yielded K_d of 1.8 × 10⁻⁸ M, with MAA-LDL a K_d of 2.6 × 10⁻⁹ M, and with MDA-LDL a K_d of 1.3 × 10⁻⁷ M. Data are shown as B/B₀ curves, where each data point is the average of triplicate determination.