Retinoic Acid-Loaded PLGA Nanoparticle Formulation of ApoB-100-Derived Peptide 210 Attenuates Atherosclerosis

Abstract

We developed a vaccine formulation containing ApoB derived P210 peptides as autoantigens, retinoic acid (RA) as an immune enhancer, both of which were delivered using PLGA nanoparticles. The formula was used to induce an immune response in 12-week-old male Apoe^−/− mice with pre-existing atherosclerotic lesions. The nanotechnology platform PRINT® was used to fabricate PLGA nanoparticles that encapsulated RA inside and adsorbed the P210 onto the particle surface. In this study, we demonstrated that immunization of Apoe^−/− mice with the formulation was able to considerably attenuate atherosclerotic lesions, accompanied by increased P210 specific IgM and another oxidized lipid derived autoantigen, M2AA, specific IgG autoantibodies, and decreased the inflammatory response, as compared to the P210 group with Freund’s adjuvant. Our formulation represents an exciting technology to enhance the efficacy of the P210 vaccine.

Graphical Abstract

PLGA nanoparticles encapsulated retinoic acid and adsorbed peptide 210 on their surface. Apoe^−/− mice were immunized subcutaneously with the formula. The vaccine effectively attenuated atherosclerosis, increased P210 and M2AA specific autoantibodies.

INTRODUCTION

Atherosclerosis is a leading cause of morbidity and death worldwide, and treatment is complicated by a complex arterial pathology with multiple genetic and environmental risk factors¹. It is now medically/scientifically accepted that atherosclerosis is an immune-mediated inflammatory disease in arterial walls and that the immune system is involved in the modulation of the atherogenic process^2,3. Among all immune responses involved in the atherogenic process, autoimmune reactions may play an important role. Evidence suggests that a better understanding of the role of autoimmunity may provide an opportunity to prevent and attenuate atherosclerosis^4,5.

Low-density lipoproteins (LDLs) play a critical role in the development of atherosclerosis⁶. Initial trapping and retention of LDL in the aortic intima and subsequent modification by either enzymatic or non-enzymatic (i.e. oxidative) fashion in the arterial intima elicit a chronic inflammatory process⁷. Like pathogen-associated molecular patterns or PAMPs, these modified LDLs act as damage-associated molecular patterns (DAMPs), resulting in self-recognizing molecules that expose and activate the innate immune system⁸. In addition, oxidized LDL (oxLDLs) is immunogenic and induces adaptive immune responses and generates oxidized LDL-specific autoantibodies⁹. OxLDL autoantibodies are commonly detected in the circulation of both healthy subjects and patients with cardiovascular disease¹⁰ and have been identified within human and animal atherosclerotic lesions¹¹. The function of these autoantibodies remains to be fully elucidated, but current data suggest they may be correlated to disease severity¹² or may block autoantigens to alleviate the disease¹³.

Apolipoprotein B (ApoB) is the main protein moiety of LDL. The function of ApoB is to transport lipids including cholesterol to all cells. Several ApoB-100-derived peptides have been identified as autoantigens, and have previously been shown to reduce atherosclerosis when injected together with an adjuvant¹⁴. For example, ApoB derived peptide 210 (P210) can be used to activate an antigen-specific atheroprotective immune response¹⁵. One P210 vaccine formulation resulted in a 40% decrease in atherosclerosis and reduction in plaque inflammation in young hypercholesterolemic mice¹⁶. Current P210-based therapeutic approaches are still largely ineffective, especially for mice older than 12 weeks, when atherosclerotic plaque is already mature and established. Thus, it is important to develop alternative delivery systems and formulations to enhance efficacy for therapeutic administration. Hence, we developed a novel vaccine formulation by encapsulating retinoic acid (RA) and adsorbing ApoB peptide, delivered by PLGA PRINT nanoparticles to investigate the efficiency of the formulation against atherosclerosis in Apoe^−/− mice.

RA is a vitamin A metabolite and is crucial for proper immune regulation in homeostasis at multiple levels during the inflammatory response in adult life. It is suggested that RA is instrumental in the initial phase of secondary lymphoid organ formation¹⁷. RA promotes the differentiation of naïve T cells to regulatory T cells (Tregs)¹⁸. In addition to proper Th1 and Th2 responses, a balance between Tregs and Th17 cells is thought to be crucial for healthy immune homeostasis. A lack of Tregs, in combination with the excessive presence of Th17 cells, is often associated with autoimmune pathologies such as inflammatory bowel disease (IBD), rheumatoid arthritis, systemic lupus erythematosus (SLE), multiple sclerosis (MS), and asthma¹⁹. We hypothesized that inclusion of RA in this vaccine formula will help stimulate the differentiation of Tregs and increase autoantibody production.

Nanoparticles generated from poly(lactic-co-glycolic acid) (PLGA) have attracted much attention due to their clinically proven biocompatibility, biodegradability, and potential adjuvant activity during immunization protocols²⁰. Many different approaches have been utilized to fabricate PLGA NPs. Among them, PRINT stands out due to its ability to generate monodisperse NPs with well-defined size, shape, and modulus²¹. PLGA NPs have been investigated for vaccine delivery and demonstrated significant variation in modulating immune responses^22–24. NPs are used as either a delivery system to enhance antigen processing and/or as an immune-stimulant adjuvant to activate or augment immunity in both prophylactic and therapeutic approaches²⁰. However, studies exploring the therapeutic benefits of P210 adsorbed onto PLGA NPs in atherosclerosis are currently lacking. In this study, we characterized the production of autoantibodies against the ApoB-100 peptides at position p210. We further constructed a P210-adsorbed PLGA NP formulation of RA and investigated the therapeutic efficacy of the aforementioned nano vaccine system against atherosclerosis in Apoe-deficient mice. Our results showed that this PLGA-based, retinoic acid NP formulation led to a significant enhancement of ApoB-100-derived peptide P210 and M2AA antibodies against atherosclerosis.

MATERIALS AND METHODS

Reagents

ApoB-100 derived peptide P210 (ApoB-100 3136–3155): KTTKQSFDLSVKAQYKKNKH was synthesized by Genscript. Poly (D-lactide-co-glycolide) (PLGA, lactide:glycolide 50:50, Mw 16–54 K) was purchased from Sigma- Aldrich. Poly (ethylene terephthalate) (PET) sheets were purchased from KRS plastics (Tabor City, NC). Perfluoro polyether PRINT® molds were provided by Liquidia Technologies (Research Triangle Park, NC). All-trans-retinoic acid (RA), chloroform, acetonitrile and water for high-performance liquid chromatography (HPLC) were purchased from Fisher Scientific (Waltham, MA). All other chemicals were purchased from Sigma-Aldrich. Sterile ultrapure water was used throughout the study.

Particle Fabrication

The 1 μm cylindrical PLGA PRINT nanoparticles were fabricated based on a previously published method²⁵ with minor modification (Scheme 1). Briefly, PLGA of lactide:glycolide 50:50 was dissolved in chloroform, and the sample was diluted to 2 wt% (mass/mass) solution with chloroform. The pre-particle solution was composed of 100 mg/mL PLGA, 30 mg/mL retinoic acid dissolved in DMSO. A thin film was made by spreading 130 μL of the pre-particle solution on a poly-(ethylene terephthalate) (PET) sheet using a #5 Mayer Rod (R.D. Specialties). The solvent was evaporated with heat. Fluorocur®, d = 1000 nm × h = 1000 nm prefabricated molds and 2000 g/mol polyvinyl alcohol (PVOH) coated PET sheets were provided by Liquidia Technologies. The PLGA-retinoic acid film was then placed in contact with the patterned side of a mold containing 1 μm diameter cavities and then passed through a heated laminator at 130° C and 80 pounds per square inch to fill the cavities with a mixture of the polymer and RA. This heating process enables the PLGA polymer solution to fill the molds, thereby forming nanoparticles of desired size and shape. The mold was peeled from the PET sheet as they passed through the heated laminator. The filled side of the mold was then placed in contact with a sheet of PET coated with poly (vinyl alcohol) (PVOH, 2000 g/mol) and passed through the heated laminator to transfer the particles from the mold to the PVOH-coated PET sheet. The particles were removed by passing the PVOH coated PET sheet through motorized rollers and applying water to dissolve the PVOH to release the particles. The particle suspension was centrifuged at 5,000 × for 1 min, the supernatant discarded, and the pellets resuspended with 0.1 wt% PVOH aqueous solution. This purification step was repeated 5 times to remove excess PVOH. An aliquot of the particles was suspended in 10 mM KCl solution for zeta potential measurements. The PLGA particles (5 μl) was pipetted on a glass slide and was dried for Scanning Electron Microscopy. The rest of particle suspension was flash frozen in 0.1 wt% PVOH aqueous solution.

Scheme 1. The PRINT process:

(A) PET sheet casting: retinoic acid was dissolved in DMSO and then mixed with PLGA chloroform solution (red). A mayer rod is then used to draw a film from this solution on a PET sheet. The solvent is removed under heat generating a film. (B) Particle fabrication: a perfluoropolyether mold (green) in brought into contact with a PLGA (red) film, passed through a heated nip (gray) and then separated. The mold cavities were filled. (C) Particle harvesting: a filled mold was brought into contact with an excipient layer (yellow) and passed through the heated nip without splitting. After cooling down, the nanoparticles in the molds were removed.

Characterization of nanoparticles

Zeta potential of the RA-loaded and blank PLGA nanoparticles was measured by a Zetasizer (Malvern Instruments Nano-ZS, Malvern, UK) at room temperature. The size and shape of particles were confirmed by Scanning Electron Microscopy (SEM, Hitachi S-4700) after the specimens were coated with 1.5nm of gold-palladium alloy using a Cressington 108 auto sputter coater (Ted Pella, Inc. Redding, CA).

Particles were quantified by thermo gravimetric analysis using a TA Instrument’s Discovery TGA.

Retinoic acid loading efficiency and release study in vitro

Trans-retinoic acid (RA) was dissolved in DMSO and stored at −20° C and protected from light until used. After dissolving particles in DMSO, quantification of RA particles was performed by reverse HPLC. RA was detected via UV-VIS detector at 360 nm, and concentration was determined by comparing the peak area with a standard curve prepared by using a known amount of free RA.

For RA release studies, 300 μl of PLGA-RA solution was placed in dialytic bags (MWCO=10 kDa) and dialyzed against 1,000 ml of PBS buffer (10 mM, pH 7.4) at 37° C stirring at 100 rpm. At the designed time points, the particle suspension in the dialysis bags was removed and analyzed using a UV spectrophotometer at 360 nm for the remaining amount of RA. To determine the percent of RA released over time, the amount of RA remaining was compared to the initial amount of RA. Experiments were repeated in triplicate.

P210 adsorption onto the surface of PLGA nanoparticles

To determine the amount of adsorption of P210 onto the surface of PLGA nanoparticles, various P210/PLGA (w/w%) ratios were mixed by keeping the amount of P210 constant and varying nanoparticle amounts to obtain higher P210/PLGA ratios. They were mixed for 15 minutes at RT. The adsorption efficiency was determined by the amount of non-adsorbed P210 after centrifugation of the particles.

Immunization protocols

Because females have larger and more advanced atherosclerotic lesions (²⁶, male129 P2-Apoe^tm1Unc/J on a C57BL/6 background (stock number 002052) were originally purchased from Jackson Laboratories (Bar Harbor, ME) and bred in the animal facility at the University of North Carolina and housed under standard conditions of temperature (20–22° C), relative humidity (30–70%), a 12-hour light: dark cycle and free access to food and water. Animal care and experimental protocols were approved by the Institutional Animal Care and Use Committee of the University of North Carolina, Chapel Hill.

To examine the therapeutic effect of the P210 NP vaccine formulations, 12 week-old male Apoe^−/−mice were randomly divided into four intervention groups and a control group of untreated mice (n = 9–12 mice per group): A) Blank particles loaded with P210 (NP blank-P210); B) particles loaded with RA and P210 (NP-RA-P210); C) P210 emulsified first in complete Freund’s adjuvant (CFA) and then in incomplete Freund adjuvant²⁷ (CFA-P210); D) Particles loaded with RA (NP-RA); and E) Untreated Apoe^−/− mice. Each mouse was immunized subcutaneously in the flank with 100 μl solution containing P210 at a concentration of 100 μg/mL adsorbed onto the surface of 1 μm PLGA-RA or PLGA blank nanoparticles by using 550 μg/ml of the particles. In addition, P210 was emulsified in CFA/IFA (BD Difco, Sparks, MD) to compare the effect of these common vaccine adjuvants with PLGA particle formulations. As shown in scheme 2, the animals were primed and boosted 4 times at 4-week intervals using the same formulations, except for the CFA/P210 group where the four P210 boost formulations were emulsified in IFA. All mice were sacrificed at 28 weeks of age. Blood, heart, spleen and lymph nodes were collected for further analyses.

Scheme 2. Immunization strategy.

Apoe^−/− mice were divided into five groups. Group 1 and 2 were immunized with P210 adsorbed onto the surface of 1 μm PLGA nanoparticle encapsulated either with or without retinoic acid (RA), respectively. Mice in group 3 were immunized with P210 in complete Freund’s adjuvant (CFA). Group 4 was only given P210 and RA. Group 5 was untreated mice that were only injected with sterile saline. All mice were primarily injected sc. at week 12 and then were boosted four more times at 4-week intervals. Adjuvant for mice in group 3 was changed into incomplete Freund’s adjuvant (IFA) when used for boosters. Blood was drawn one day before the primary immunization (pre-immunization) and collected again just before sacrifice at 28 wks of age (post-immunization).

Atherosclerosis quantification

At the end of the study, mice were given an anesthetic, followed by perfusion under physiological pressure with PBS/heparin, and then 4% buffered paraformaldehyde (pH 7.4) for routine histological analysis. Hearts were harvested, placed in Optimal Cutting Temperature medium (OCT, Electron Microscopy Sciences, Hatfield, PA), and frozen at −80° C. Beginning at the first appearance of the tri-leaflet aortic valve, successive 10 μm transverse sections were cut. Sections were then stained with Oil Red O and counter-stained with hematoxylin. The lesion area for one anatomically defined position is where 3 valves show and was averaged for two different sections per animal. Lesion quantification was performed in 8 consecutive sections, each 50 μm apart from each other. Quantitative morphometric analysis of lesion area using NIH 1.59 Imaging Software was carried out by a histologist blinded to the sample groups.

Serum cholesterol and triglyceride analysis

Plasma cholesterol and triglycerides were measured using commercial kits from FUJIFILM Wako Diagnostics (Richmond, VA) according to the manufacturer’s instructions.

Measurement of antibodies against native and malondialdehyde (MDA)-modified P210

Malondialdehyde (MDA) is generated as a result of oxidative degradation of lipids with the formation of lipid peroxides. MDA is a mediator or marker of inflammation that has been associated with atherosclerosis^{28, 29}. Assessment of native and MDA-modified P210-specific antibody in plasma was carried out using an ELISA assay as previously described³⁰. MDA modified P210 was prepared by treatment with 0.5 mol/L MDA. To remove the unbound MDA, the P210 was dialyzed against 0.15 mol/L phosphate-buffered saline (PBS) containing 1 mmol/L EDTA of pH 7.4, using dialysis tubing with a 1,000 MW cut-off. The thiobarbituric acid reactive substance (TBARs) assay was used to assess MDA modification of the peptide³¹. The respective native and MDA-modified P210 in this study were designated as P210_nat and P210_MDA.

In brief, P210_nat or P210_MDA, diluted in PBS pH 7.4 (20 μg/mL), were adsorbed into the wells of an N-ethylmaleimide coated microtiter plate (Thermo Scientific Pierce) by incubation overnight at 4°C. The plates can bind sulfhydryl-containing molecules, also the α-group of peptides and the imidazole group of histidine as well as P210 because it contains a terminal histidine. After a washing step with PBS containing 0.05% Tween-20 (PBS-T), the uncoated areas were blocked with SuperBlock solution (Pierce) in Tris-buffered saline for 15 minutes at room temperature, followed by incubation with the mouse plasma samples, diluted 1/100 in PBS-T, for 3 hours at room temperature. After washing the wells with PBS-T, the addition of 200μL freshly made cysteine solution was added at 10μg/mL and incubated for 1 hour at room temperature to inactivate excess maleimide groups. After washing the plate, IgG or IgM antibodies to P210 were added and incubated for 2 hours for detection of IgG and 3 hours for detection of IgM. Following the washes, secondary antibodies were added (1/1,000 rat anti-mouse IgM conjugated with HRP (Invitrogen) or 1/3,000 goat anti-mouse IgG, conjugated with HRP (Invitrogen), respectively), for 2 hours at room temperature. The color reaction was developed using tetramethylbenzidine (TMB)/HRP substrate kit (BD Bioscience) for 11 minutes at room temperature for all plates. The plates were measured for absorbance at 405 nm. For each sample, two blank wells (non-coated with the peptides) were analyzed, and the mean OD405_nm value of those was subtracted from the OD405_nm value obtained with the peptide-coated wells.

1,4-Dihydropyridine-type malondialdehyde-acetaldehyde (M2AA) adduct autoantibody determination

It has been demonstrated that MDA can break down to form acetaldehyde (AA), and to form a stable malondialdehyde–acetaldehyde (M2AA) adduct^32,33. The M2AA antigen-based ELISA has not only proven to be more sensitive and specific than the crude M2AA antigen-based ELISA that is currently in use but has also been able to detect markedly increased anti-M2AA antibody titers in the serum of Apoe^−/−mice at a very early stage of atherosclerosis³².

Malondialdehyde-acetaldehyde adducts were synthesized as described previously³². Briefly, ε-aminocaproic acid (6-ACA), acetaldehyde, and malondialdehyde were incubated at 37° C for 72 hours. M2AA-6ACA was purified by HPLC and then conjugated with bovine serum albumin (BSA) as reported previously⁴⁵. M2AA-6ACA-BSA was adsorbed on wells in 96-well plates at 4° C overnight. After blocking with 3% BSA, serum samples diluted at 1:320 ratio with 1% BSA in PBS were incubated at 4° C overnight. After washing, peroxidase-labeled secondary antibody (EnVision+Single Reagents anti-mouse (IgG)-HRP (Code K4001, Dako North America, Inc., Carpentaria, CA) was added and incubated for 1 hour. After washing the wells, the 3,3′,5,5′-TMB/H₂O₂ substrate was added to the wells and remained at room temperature for 30 minutes. The plates were then read with a plate reader (Vmax Kinetic Microplate Reader, Molecular Devices, Sunnyvale, CA) at 650 nm.

Flow cytometry

To analyze blood lymphocytes, mouse peripheral blood was collected by submandibular bleeding into microtubes with 500 μl PBS containing 7 mM EDTA, loaded onto Lympholyte®-Mammal Separation Media (Cedarlane Laboratories Ltd, Burlington, NC), followed by centrifugation at 800×g for 20 minutes at room temperature. Lymphocytes at the interface were collected and washed for antibody staining. Single cell spleen and lymph node suspensions were prepared by grinding the tissue between two sterile glass slides in PBS. Red blood cells were lysed and cells filtered through a 70-μm filter.

Tregs were detected by staining with CD3-APC-eFluor780, CD4-FITC, CD25-PE-Cyanine 7 (eBioscience) for 30 minutes in the dark and then fixed in Cytofix/Cytoperm³⁴ buffer for one hour, followed by staining with FoxP3-PE-eFluor 610 antibody (eBioscience) for one hour in the dark. The stained cells were washed and re-fixed with 1% paraformaldehyde and stored in the dark at 4° C. Cells were run on Cyan flow cytometer (Dako), and data were analyzed using Submit software.

Detection of serum cytokine secretion levels

Serum IL-10 and IFN-γ were measured by enzyme-linked immunosorbent assay (ELISA) using commercially available kits (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions. Each assay was carried out in duplicate for each sample. Absorbance was measured at 450 nm by means of a spectrophotometer (SpectraMax M5, Molecular Devices).

Statistical analysis

Data are expressed as means ± SE. To compare groups, we used the one-way analysis of variance³⁵. Post hoc pairwise comparisons were performed by Tukey–Kramer honestly significant difference (HSD) test (JMP; SAS Institute).

RESULTS

Particle fabrication and characterization

One μm PLGA nanoparticles with or without RA were fabricated using PRINT technology. RA can be loaded to 18 wt% as determined by HPLC. Both blank and RA loaded particles were highly uniform in size and shape as visualized by SEM (Figure 1A and 1B). The zeta potential of blank PLGA particle was −12 mV, and loading of RA made the particles more negative and reached −21.5±0.6 mV.

Figure 1. Scanning Electron Microscope (SEM) image of trans-retinoic acid loaded on A) PLGA nanoparticles and B) blank PLGA NPs.

The diameters of both NPs were 1 μm. C). In vitro drug release of RA from PLGA nanocarriers in phosphate buffered saline (pH 7.4) at 37°C. The release study was performed in a dialysis membrane and the RA remaining in the dialytic bag was analyzed using a microplate reader. D). Peptide 210 was adsorbed onto the surface of PLGA nanoparticles in variable P210/PLGA (w/w%) ratios. Increasing P210/PLGA ratios were obtained by fixing P210 and reducing the nanoparticle mass. Adsorption efficiency was determined by the amount of non-adsorbed P210 after centrifugation of the nanoparticles.

In vitro release of RA from particles was assessed under pH 7.4 aqueous conditions (Figure 1C). The release of RA from 1 μm size PLGA nanoparticles displayed two-phases: a faster release of 65% in the first 3 hours and a much slower and extended release. The slow and sustained release of RA from particles ascribed to multiple mechanisms including penetration and diffusion of RA through a matrix of PLGA particles and suggests the stability of PLGA particles in the natural physiological environment.

To determine optimal adsorption of P210 onto the surface of PLGA nanoparticles, variable P210/PLGA (w/w%) ratios were obtained by keeping the amount of P210 constant and decreasing the nanoparticle amount to obtain higher P210/PLGA ratios. Up to a 4% w/w ratio resulted in a nearly 100% adsorption efficiency for each particle size (Figure 1D). The decrease in adsorption efficiency with increasing %w/w ratios (8% and 12%) was most likely caused by an excess of P210 over the available nanoparticle surface.

Effect of immunization on atherosclerosis

To evaluate the therapeutic effect of P210 NP vaccine formulations, male Apoe^−/−mice at 12 weeks of age were subcutaneously immunized with different formulations following the procedure in scheme 1. The lesion size in the aortic root, a well-established marker for atherosclerosis, was assessed at the end of treatment. Compared to the untreated group, vaccination with NPRA-P210 particle group led to a 53% reduction in the lesion size in the aortic root (P<0.05) (Figure 1A, 2E and 2F). The exclusion of RA in the particles increased the average lesion size although the difference did not reach significance (Figure 2B and 2F). The results suggest that P210 is an essential component in the particulate vaccine for attenuation of atherosclerosis. CFA is a potent vaccine adjuvant, and CFA combined P210 has been reported for its therapeutic effect on atherosclerosis³⁶. However, in the study, we observed a minor difference between CFA-210 and the untreated group (P=0.04) but clearer, and higher therapeutic efficacy occurred in NP-RA-P210 group. Furthermore, although both NP-P210 (BLK and RA) groups outperformed the CFA-P210 group (conventional adjuvant group) in their anti-atherosclerosis efficacy (Figure 2C and 2F), this difference lacks statistical significance (P > 0.05).

Figure 2. P210 and RA carried by PLGA nanoparticles retarded atherosclerotic lesion formation in the aorta.

A-E). Oil red O staining of representative aortas from each group. F). Quantification of aortic root lesion area. Each shape represents a single mouse. n=9 to 12 mice/group. All the statistical analysis of experiments comparing the five groups was evaluated by one-way analysis of variance. P210: peptide 210; NP: PLGA nanoparticles; RA: trans-retinoic acid; CFA: complete Freund’s adjuvant and incomplete Freund adjuvant; Untreated: Apoe^−/− mice injected only with sterile saline subcutaneously.

Plasma cholesterol and triglyceride levels

Similarly, the plasma cholesterol level in the above-untreated animals was 500 ± 33 mg/dL at the time of sacrifice. Immunization with both NP-P210 (BLK and RA) formulations significantly decreased plasma cholesterol levels with the NP-RA-P210 particle group outperforming the NP BLK-P210 particle group (Figure 3A). However, immunization with peptide 210 did not affect plasma triglyceride level between treated and untreated mice among the five groups (Figure 3B). In addition, there was not a significant change in body weight in the immunized mice (data not shown).

Figure 3. Plasma Chol and TG levels.

A). Plasma total Chol levels. B). Plasma TG levels. n=9 to 12 mice/group. Data are presented as mean ± SEM. All the statistical analysis of experiments comparing the five groups was evaluated by one-way ANOVA. Chol: cholesterol; TG: triglyceride.

Production of specific antibodies against P210in Apoe^−/−mice

One potential mechanism of atheroprotection is the development of protective antibodies. Immunization with P210 based formulations may induce changes of ApoB-specific autoantibodies to P210_Nat and P210_MDA. P210_nat and P210_MDA specific IgM and IgG autoantibody titers were determined in this study before the initial injection (pre-immunization) and at sacrifice (post-immunization). The results showed that mice in all groups significantly increased titers of IgM autoantibodies against P210_nat after immunization (Figure 4A). Among them, the largest net increase of IgM titers occurred in NP-RA-P210 group, and both NP BLK-P210 and NP-RA groups displayed a significant increase compared to the untreated group (Figure 4A and 4C). For instance, immunization with P210 resulted in an increase in IgM antibody levels against native peptides (0.236±0.0586 in pre-immunization versus 0.071±0.020 absorbance units in post-immunization) in NP-RA-P210 group (P<0.001) (Figure 4A).

Figure 4. Titers of P210-specific total IgM or IgG from immunized Apoe^−/− mice.

Plasma from Apoe^−/− mice was obtained before immunization (pre-immunization) and at termination (post-immunization). A. Titers of P210-specific total IgM. B. Titers of IgM against MDA-modified P210. C. The values of P210-specific IgM titers differed between post-immunization and pre-immunization. D. The values of MDA-modified P210-specific IgM titers differed between post-immunization and pre-immunization. E. Titers of P210-specific IgG. F. Titers of IgG against MDA-modified P210. Data represent arithmetic means ± SEM of specific antibodies. All the statistical analysis of experiments comparing the five groups was evaluated by one-way ANOVA.

A similar significant increase was observed in IgM against the corresponding MDA-modified P210 sequences (0.162±0.068 versus 0.326±0.035, P<0.05) absorbance units in NP-RA-P210 group and other three experimental groups compared to untreated (Figure 4B and 4D). The data suggest that a potential protective role of these autoantibodies in atherosclerosis.

On the contrary, both P210_nat and P210_MDA specific IgM autoantibody titers in NP-RA group did not significantly different compared to the untreated group. The results further demonstrated the essential role of P210 in the vaccine formula.

The production of IgG requires assistance from antigen-specific T-cells and gives insight into antigen-specific T-cell activation. However, there were no significant differences in IgG against either native peptide (Figure 4E) or MDA-modified peptide (Figure 4F) between pre-immunization and post-immunization. The data suggest that IgG production does not significantly increase by immunization and thus it is very likely that it does not contribute to attenuation of atherosclerosis.

PLGA nanoparticles without loaded RA increased the number of circulating Tregs

To explore whether the atheroprotection conveyed by immunization with P210 may be related to increased numbers of FoxP3-expressing Treg cells, we compared NP BLK-P210, NP-RA-P210, and untreated groups in the frequency of Tregs (CD4+CD25+FoxP3) in blood, spleen and lymph nodes. No significant difference was observed between NP-RA-P210 and the untreated group, but NP BLK-P210 significantly up-regulated the number of Tregs (P<0.05) in the blood (Figure 5), suggesting that RA did not significantly shift naïve T cells to the regulatory lineage in blood.

Figure 5. Tregs in peripheral blood, lymphocytes and spleen of Apoe^−/− mice.

Apoe^−/− mice immunized with P210 adsorbed onto the surface of PLGA nanoparticles containing either RA or no RA. Untreated mice served as controls. After euthanizing the mice, blood and tissue were collected. Circulating lymphocytes separated from blood, and lymphocytes isolated from lymph nodes and spleen were stained with CD3-APC-eFluor780, CD4-FITC. Cells were then fixed and permeabilized followed by intracellular staining with FoxP3-PE-eFluor 610 and flow cytometric analysis. Cells were gated on CD4 and CD3 and FoxP3 by the three groups of mice. (B) Quantitative analysis of Tregs in the three groups of mice. Results are expressed as mean ± SE. Statistical analysis of experiments comparing the two groups was evaluated by Student’s test.

Blood IFN- γ and IL-10 cytokines

To directly examine whether T-cell lineage differentiation is influenced by Th1 or Tregs, we measured plasma IFN-γ cytokine response in the atherosclerotic Apoe^−/− mice before (wk 0) and after 28 weeks of intervention (wk 28). The untreated Apoe^−/− mice have increasing IFN-γ level in blood with the disease development, an indication of a more severe inflammatory condition, which is consistent with the literature³⁷. Remarkably, treatment with each vaccine formulation was able to suppress IFN-γ release (Figure 6A), suggesting that the vaccine reduced inflammation. However, we did not observe significant differences in IL-10 release, an anti-inflammatory cytokine, between vaccine groups and the untreated group (Figure 6B). The result may provide indirect evidence that the attenuation of atherosclerosis was not mainly through enhanced Treg numbers or their function.

Figure 6. Effect of vaccination using PLGA-adsorbed P210 on plasma levels of IFN-γ and IL-10.

Plasma from Apoe^−/− mice was obtained before immunization (pre-immunization) and just before euthanasia (post-immunization) and the levels of the cytokines were examined using commercial ELISA kits. Levels of cytokines (ng/mL), A) IFN-γ and B) IL-10, are graphically represented as mean ± SE, n=9–12 mice per group.

Markedly increased serum M2AA-specific antibody

Additionally, mice immunized with NP-RA-P210 showed that anti-M2AA IgG titer considerably increased after immunization, compared to all other groups (Figure 7). The data clearly shows that retinoic acid enhances M2AA-specific IgG antibody production that may have contributed to reduction of atherosclerosis.

Figure 7. Serum M2AA production in Apoe^−/− mice.

The mice were immunized with the P210 vaccine in all experimental groups and controls. Results are expressed as mean ± SE. All statistical analysis of experiments comparing the five groups was evaluated by one-way ANOVA.

DISCUSSION

In this study, we tested our particulate vaccine formulation for enhanced therapeutic efficacy in reducing established atherosclerosis in Apoe^−/− mouse model. It is well-known that aging affects the magnitude and the persistence of antibody responses to protein vaccines^38,. We chose 12-week-old Apoe^−/− mice since the age of 12 weeks is a time point when fatty streaks and other early atherosclerotic lesions start to exhibit in the proximal aorta based on our results from a large number of Apoe^−/− mice and other investigators³⁹. Most investigators use young Apoe^−/− mice at 6–10 weeks old due to their ability to mount a robust immune response since old mice with advanced atherosclerosis may be less plastic and resistant to immunomodulation than young mice. The formulations for older mice with preexisting aortic lesions must be able to generate a more vigorous immune response than normally required for young mice. However, our results clearly demonstrated that atherosclerotic lesion size was significantly reduced by 53% due to immunization with P210 in 12-week-old Apoe^−/−mice, accompanied with increased production of IgM_nat and IgM_MDA, and IgG_M2AA autoantibody, and a decreased blood pro-inflammatory cytokine, INF-γ, suggesting that our formulation can boost a stronger immune response.

Our data showed that immunization with P210 reduced atherosclerosis by around 42–53% (P<0.05, vs untreated group). Our data are corroborated by other investigators who demonstrated that immunization with oxidized LDL inhibits the development of atherosclerosis. Nilsson et al. reported the immunization of 6-week-old Apoe^−/− mice with P210 by the subcutaneous (s. c.) route resulted in a significant reduction (37%) of atherosclerosis⁴⁰. Herbin et al. reported that the immunization of 12-week-old Apoe^−/− mice with a mixture of the ApoB100-derived peptides (P210, P240, and MDA-P210) by a s.c. mini-osmotic pump diffusing for 2 weeks, resulted in a significant reduction of lesion development in young mice (~40%)⁴¹. Chyuet al.⁴² described a significant reduction in atherosclerotic lesion development (50%) in Apoe^−/−mice at 6–7 weeks of age subcutaneously immunized with P210 coupled to BSA with Alum as an adjuvant. The authors thought that the atheroprotective effect was mediated by CD8⁺CD25⁺ (also known as CD8⁺ T-suppressor cells). Pierides et al.¹⁴ used 6–8 weeks old Apoe^−/− mice subcutaneously immunized with three peptides (P2, P45, and P210) from ApoB100 conjugated to BSA. Immunization with P45 and P210 significantly increased the levels of peptide-specific immunoglobulins IgG1, Treg populations, and IL-10 levels but had no effect on IFN-γ levels. Remarkably, the immunization with P45 and P210 reduced atherosclerosis progression (50%)^43,44. These studies failed to find a unified mechanism to explain the vaccine’s effect on reduced lesion size. Our results revealed that IgM antibodies in circulation in almost all experiment groups markedly increased after immunization, compared to the untreated group. In particular, the highest production of specific P210 IgM appeared in NP-RA-P210 groups. The results suggested that our vaccine formulation was effective for curbing further development of atherosclerosis in established atherosclerosis and increased IgM contributed to the alleviation. It is reported that in humans, levels of plasma IgM reactive to oxLDL have an inverse relation to carotid atherosclerotic plaque size⁴⁵ and to coronary artery disease⁴⁶. Evidence suggests that such atheroprotective effects may result from autoantibody blocking uptake of oxLDL by macrophages and thereby preventing foam-cell formation⁴⁷. In addition, IgM may represent natural antibodies present at birth that are likely evolutionarily conserved to protect against danger-associated molecular patterns (DAMP) present in sites of inflammation⁸. Likewise, our data revealed that serum MDA-modified P210 antibodies also increased in all immunization groups. It is likely that immunization with P210 also leads to a localized inflammatory condition that resulted in the generation of some P210_MDA. However, unlike native P210 antibodies, P210_MDA antibody titer remained unchanged in the untreated group. The different profiles between native and MDA-modified IgM antibody titers suggests that specific IgM antibody against P210_MDA have more therapeutic effects compared to the antibody against P210_nat. In other words, P210_MDA appears to be a more important target than P210_nat in terms of protection and alleviation of atherosclerosis by vaccination. Moreover, the fact that lack of P210-specific IgG production after immunization does not support a strong role for the humoral response in mediating the protective effect of active immunization using the P210 vaccine. Klingenberg reported that the induced IgG antibody titers did not correlate with the lesion size⁴⁸. Contrary to our and Klingenberg’s observations, Fredrikson et al. claimed that IgG autoantibodies against the native peptide are associated with decreased cardiovascular risk while autoantibodies against MDA-modified peptides reflect disease severity⁴⁹. Binder et al. found that native LDL is heavily modified with MDA and led to atheroprotection in both rabbits and mice⁵⁰ and the titer of IgG antibody against MDA-modified ApoB-100 was elevated, but the IgM titer was not changed¹⁶.

Whether the immunization with these p210-loaded PLGA PRINT particles has some effect on the adaptive immune response and to T cell-dependent or independent antigens is not fully understood. Our finding that high titers of IgM, but not IgG, yield a reduction in lesion size due to the p210 epitope may suggest that a T-cell–dependent switch to synthesis of IgG antibodies against epitopes in oxidized LDL is not involved in atheroprotective immune responses¹⁶. But it remains to be clarified whether this reflects a higher target specificity of p210 IgM. On the other hand, the fact that M2AA specific antibody significantly increased suggests that a T-cell–dependent switch is required for the synthesis of IgG antibodies with an assist by follicular helper T cells against epitopes in M2AA. The mechanism in which IgG response to the different autoantigens warrants for further study.

Another plausible explanation for the alleviation of atherosclerosis by our vaccine formulation is that a large amount of M2AA autoantibody in circulation elicited by our vaccine plays a critical role against disease. It has recently been recognized that MDA and AA react synergistically to form stable M2AA adducts⁵¹. M2AA adducts generate immuno-dominant epitopes and appear to play key roles in atherogenesis⁵¹. Since M2AA-modified proteins may pose an atherogenic threat to blood vessels, autoantibody against the M2AA-modified proteins may be beneficial to atherosclerosis, which may reflect an attempt by the immune system to remove harmful M2AA-modified proteins (oxidized LDL) from circulation. We do not know whether there is a link that exists between P210 antibody and M2AA antibody. However, our data suggest that retinoic acid selectively enhances M2AA-specific IgG antibody production, which may contribute to attenuation of atherosclerosis.

As mentioned before, some investigators found that activation of Tregs is another possible mechanism for downregulating the inflammatory response and consequently reducing atherosclerosis^52,53. For example, Herbin et al.’s⁵⁷ data showed an infusion of ApoB100 peptides for two weeks and did not alter antibody production but promoted a specific Treg cell response, whereas our data showed that Treg number increased in the peripheral circulation. But neither study showed increases in the secondary lymphoid organs, nor elevation of serum IL-10. It is not fully understood why immunization with ApoB100-derived peptide P210 reduced lesion development. Atherosclerosis antigen-specific Treg epitopes have not been identified. There was no evidence that peptide-specific Treg cells were generated after P210 immunization, and there is still the possibility that the increase in Treg cell number was mediated by aluminum hydroxide alone, previously shown to promote Treg cells in the absence of any peptide adjunction⁵³. We postulate that PLGA does not entirely share a common function with aluminum to promote Treg cells. Instead, the addition of PLGA profoundly induces a specific humoral immune response. The observation is further strengthened by the effect of retinoic acid (RA) as a test on Treg response. RA is thought to play a vital role in autoimmune diseases by shifting naïve T cells to the Treg lineage¹⁹. However, in our experiment, including RA in the PLGA particles did not completely support our hypothesis and did not significantly increase Treg numbers in blood and thus the attenuation of atherosclerosis was not through enhanced Treg numbers or their function.

In the P210 particulate vaccine formulations, we compared the effect of PLGA nanoparticles as a potential adjuvant by employing two commonly used adjuvants, CFA and IFA, which are effective in atherosclerosis vaccines and served as adjuvant controls. Although our data showed that PLGA NPs without any adjuvant in the formulation did not significantly increase vaccine efficacy, compared to CFA-P210 group, PLGA NPs, as an effective adjuvant, may still have benefit considering the fact that CFA/IFA cannot be used for humans. The PLGA particles possess adjuvant qualities which may be based on the following assumptions; 1) formation of a depot in the injection site for protection of the antigen; 2) condensation of the density of the antigens by adsorbing them on the surface of the NPs to assist in crosslinking and thus becoming more immunogenic; 3) possible aggregation of antigens leads to a sustained delivery and controlled release.

Our results showed that M2AA-specific IgG production profoundly increased in NP-RA-P210 group. It is very likely that RA contributes to the alleviation of atherosclerosis. However, RA did not show a significant impact on either P210-specific IgM or IgG production after immunization, and it also did not increase Treg production in circulation. Since a broad spectrum of anti-inflammatory functions of RA depends on the local microenvironment, the roles of RA in vaccination against atherosclerosis is promising and worth additional investigation.

In this study, we only measured the aortic root lesion size to evaluate atherosclerosis quantification using the valid method⁵⁵ based on the following reasons. 1) aortic root has the greatest proclivity to developing lesions and it is an important parameter for assessing atherosclerosis severity in the mouse⁵⁶. 2) immunization with native apoB-100 peptides reduced en face fatty lesions on early lesions likely in the aortic origin but not more advanced plaques in apoE-null mice^{16, 57}. In addition, immunization with both NP-P210 and NP-RA-P210 formulations significantly decreased plasma cholesterol levels whereas the NP-RA-P210 showed a better effect. We do not know the underlying mechanism currently. Zarei et al. thought that role of RA in reducing atherosclerotic plaque size may be a mechanism other than lowering serum lipids levels⁵⁸.

CONCLUSIONS

The present study provides evidence that after immunization with our vaccine formulations, production of P210 specific IgM antibodies and oxLDL specific M2AA autoantibodies increase, which may associate with the alleviation of atherosclerosis in advanced stages of the disease in Apoe^−/− mouse model. Our results suggest that RA-PLGA nanoparticles, as a delivery system and adjuvant, represent an exciting new formulation for retardation of established atherosclerosis in old mice.