VB-201, an oxidized phospholipid small molecule, inhibits CD14- and Toll-like receptor-2-dependent innate cell activation and constrains atherosclerosis

Authors


Summary

Atherosclerosis is an inflammatory disease of the vascular wall. Activated monocytes and dendritic cells (DC) in the intima layer of the vasculature promote atherogenesis. Toll-like receptor (TLR)-2 and TLR-4, which are predominantly expressed on these cells and mediate their activation, are essential for atherosclerosis development. In this study we demonstrate that VB-201, an oxidized phospholipid (Ox-PL) small molecule, inhibits TLR signalling restricted to TLR-2 and TLR-4 in human and mouse monocytes and DC. Mechanistically, we show that VB-201 binds directly to TLR-2 and CD14, the TLR-4 co-receptor, to impair downstream cues and cytokine production. In a rabbit model, oral administration of VB-201 constrained atherosclerosis progression. This effect was not due to reduced cholesterol abundance, as hyperlipidaemia was sustained. We suggest that VB-201 may counter inflammation where TLR-2 and/or CD14 complicity is essential, and is therefore beneficial for the treatment of atherosclerosis.

Introduction

Atherosclerosis is a complex disorder involving lipid retention, inflammation and oxidative stress [1-4]. Cells of the innate and adaptive immune arms collaborate to affect the size, composition and vulnerability of the atherosclerotic plaque [5-8]. Monocytes are recognized as the prime immune cells involved in the development of the atherosclerotic plaque [9, 10]. They traverse into the intima layer of the vascular wall, concomitantly converting into activated macrophages, and bind to uptake oxidized low-density lipoprotein (Ox-LDL) through the pattern recognition receptor (PRR) family members of scavenger receptors (SRs) CD36 and SR-A [11, 12]. This process engenders the formation of lipid-laden foam cells, a principal stage in inflammation initiation and subsequent construction of the necrotic core in atherosclerosis [13, 14]. Studies have shown that binding is not restricted to Ox-LDL and that oxidized phospholipids (Ox-PLs), which represent endogenous moieties formed during oxidative stress, can also interact with SRs [15, 16] and promote monocytes adhesion [17] and secretion of chemoattractants [18]. None the less, a growing body of evidence suggests that Ox-LDL and Ox-PLs can convey their proinflammatory activity through a different group of PRRs, namely Toll-like receptors (TLRs). For example, binding of Ox-LDL to CD36 promoted the formation of a hetrotrimeric CD36–TLR-4–TLR-6 complex and the induction of proinflammatory mediators by macrophages [19], while chemokine production induced by Ox-PLs in endothelial cells was TLR-4-dependent [20].

TLRs, a family of receptors imperative for the innate immune response against microbial invasion, can be divided into two major subgroups based on their cellular localization. Plasma membrane-expressed TLRs include TLR-1, -2, -4, -5 and -6, whereas the intracellular TLRs are TLR-3, -7, -8 and -9 [21, 22]. The interaction between TLRs and their cognate agonists instigates a cascade of signals that include the recruitment of adaptor molecules myeloid differentiation primary response gene/TIR-domain-containing adapter-inducing interferon-β (MyD88/TRIF) and downstream phosphorylation of mitogen-activated protein kinases (MAPK) and nuclear factor kappa B (NF-κB) [22, 23]. These events culminate in the secretion of proinflammatory cytokines, including interleukin (IL)-12/23, IL-6 and tumour necrosis factor (TNF)-α [24]. TLR-2 forms a heterodimer with TLR-1 to recognize triacylated lipopeptides from Gram-negative bacteria and mycoplasma [25], and with TLR-6 to bind diacylated lipopeptides from Gram-negative bacteria and mycoplasma [26]. TLR-4 couples with MD2 and CD14 to form a complex that binds lipopolysaccharide (LPS) from Gram-negative bacteria [27].

While the majority of studies attributed a proinflammatory role to Ox-PLs, reports have shown that Ox-PLs can protect mice from endotoxin-induced septic shock [28], down-regulate T cell effector function [29] and inhibit TLR-2- and TLR-4-mediated signalling events in dendritic cells (DC) [30, 31], alluding to an anti-inflammatory role for Ox-PLs.

Our previous study demonstrated that a novel synthetic, fully oxidized phospholipid small molecule named VB-201 reduces interferon (IFN)-γ production by antigen-specific pathogenic T cells and restricts central nervous system (CNS) inflammation in a mouse model of multiple sclerosis [32]. In the current report we propose a mechanism by which VB-201 conveys its immunomodulatory effect. Specifically, VB-201 inhibited TLR signalling restricted to TLR-2 and TLR-4 in monocytes and DCs. Precipitation experiments revealed that the effect on signal transduction can be attributed to the binding of VB-201 to TLR-2 and CD14. Furthermore, oral administration of VB-201 profoundly constrained atherosclerosis in a rabbit model without affecting lipid profiles. The results suggest an anti-inflammatory role for modified Ox-PLs in atherosclerosis.

Materials and methods

Animals

Twelve-week-old male New Zealand white (NZW) rabbits were purchased from Harlan Laboratories (Rehovot, Israel). All experiments were approved by the Institutional Animal Care and Use Committee of the Sheba Medical Center, Ramat Gan, Israel.

Reagents

The bioactive compound VB-201 [(R)-1-hexadecyl-2-(4′-carboxy)butyl-sn-glycero-3-phosphocholine] and its inactive derivate VB-207 [(R)-1-octyl-2-(4′-carboxy)butyl-sn-glycero-3-phosphocholine] (Fig. 1) were synthesized in VBL's chemical laboratory (Or Yehuda, Israel). (R)-1,2-O-isopropylideneglycerol was reacted with 1-bromoalkyl (C = 16 for VB-201 and C = 8 for VB-207) and potassium hydroxide in toluene by azeotropic distillation followed by removal of the acetonide moiety under acidic conditions. After evaporating the solvent, 1-alkyl-sn-glycerol was purified by recrystallization from hexane to synthesize VB-201 or by column chromatography on silica gel to synthesize VB-207. The primary hydroxyl group of 1-alkyl-sn-glycerol was protected selectively using triphenylchloromethane under mild basic conditions in dry tetrahydrofuran (THF) and dry acetonitrile mixture to yield 1-alkyl-3-trityl-sn-glycerol. The product was then etherified with 6-bromo-1-hexene and potassium hydroxide in toluene by azeotropic distillation. 1-alkyl-2-(5′-hexenyl)-3-trityl-sn-glycerol was isolated as an oil. The double bond of 1-alkyl-2-(5′-hexenyl)-3-trityl-sn-glycerol to carboxylic acid was oxidized using sodium periodate in the presence of a catalytic amount of potassium permanganate and the trityl protecting group was removed under acidic conditions to yield the (S)-1-alkyl-2-(4′-carboxy)butyl-glycerol.

Figure 1.

Chemical structure of VB-201 and VB-207.

The free acid was protected by esterification and reacted with phosphorus oxychloride followed by an addition of ethanolamine in THF and triethylamine. Ring opening with acetic acid followed by a reaction with methyl tosylate and removal of the ester group by saponification with sodium hydroxide and purification by silica gel column yielded (R)-1-alkyl-2-(4′-carboxy)butyl-sn-glycero-3-phosphocholine.

To label with biotin, VB-201 and VB207 and OVA (Sigma, Rehovot, Israel) were dissolved in 0·1 M 2-(N-morpholino)ethanesulphonic acid (MES) buffer (Thermo Scientific, Rockford, IL, USA) and conjugated using EDC [1-ethyl-3-(dimethylaminopropyl) carbodiimide HCL] (Thermo Scientific) at a molar ratio of 100 (VB-201/VB-207):1 (OVA):240 (EDC) for 2–3 h at room temperature (RT). Samples were transferred to 10 kDa dialysis cassettes (Thermo Scientific) and dialyzed overnight against phosphate-buffered saline (PBS). Coupling was validated using ovalbumin (OVA)-VB201- and OVA-VB207-coated enzyme-linked immunosorbent assay (ELISA) plates and rabbit anti-VB-201 polyclonal antibodies (Supporting information, Fig. S1A). OVA-VB201 and OVA-VB207 were then conjugated with amine-polyethylene glycol (PEG2)-biotin (in 0·1 M MES buffer) using EDC at a molar ratio of 1 (OVA-VB201/OVA-VB207) : 100 (amine-PEG2-biotin) : 700 (EDC). The reaction was allowed to proceed for 2–3 h at RT after which samples were again transferred to a 10 kDa dialysis cassette and dialyzed overnight against PBS. Biotin conjugation and equal amounts used for precipitation of BO-VB201 and BO-VB207 were validated by using ELISA plates coated with biotinylated OVA VB201 or OVA VB-207 (OB-VB201; OB-VB207 respectively) and Western blot with anti-biotin antibody (Supporting information, Fig. S1B). Radio-labelled VB-201 ([3H]2-VB-201) was prepared from unsaturated precursor Δ9-VB-201 [1-(9′-cis-hexadecenyl)-2-(4′-carboxy)butyl-sn-glycero-3-phosphocholine] by hydrogenation with tritium.

[3H]2-VB-201-cell association

Cells (106) were cultured for 2 h with [3H]2-VB-201 mixed with cold (3·4 μM) VB-201, and then washed twice with PBS. Cells were resuspended in 100 μl of lysis buffer and 20 μl were taken to measure VB-201 association using a β-counter.

Isolation of B cells, T cells and monocytes and in-vitro generation of DC

Venous blood samples were obtained from healthy male donors in compliance with the Institutional Review Board at the Sheba Medical Center, Ramat Gan, Israel. PBMCs were isolated on Ficoll-Paque PLUS (GE Healthcare, Uppsala, Sweden) using 50 ml Leucosep tubes (Greiner Bio-One, Frickenhausen, Germany). Cells were washed in PBS (Kibbutz Beit Haemek, Israel) and incubated at 4°C for 15 min in a buffer containing PBS and 0·5% bovine serum albumin (BSA) with human CD14, CD19 and CD4 microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) to isolate monocytes, B cells and T cells, respectively. To generate monocyte-derived DC (Mo-DC), CD14+ monocytes were counted, washed and seeded (106/ml) in medium containing RPMI-1640, L-glutamine, β-mercaptoethanol, 10% fetal calf serum (FCS), sodium pyruvate, non-essential amino acids, 0·01 M HEPES, antibiotics (penicillin, streptomycin) and 50 ng/ml human granulocyte–macrophage colony-stimulating factor (GM-CSF) and 20 ng/ml human IL-4 (both from PeproTech Asia, Rehovot, Israel). Medium was replaced every 2–3 days.

To generate mouse bone marrow-derived dendritic cells (BMDC), bone marrow was flushed with cold RPMI-1640 from mice femur and tibia. A cell suspension was prepared and erythrocytes were removed using red blood cell (RBC) lysis buffer (Kibbutz Beit Haemek). Cells were washed in PBS (Kibbutz Beit Haemek) and incubated for 15 min at 4°C in buffer containing PBS and 0·5% BSA with mouse B220 and CD90 microbeads (Miltenyi Biotec). Cells were then washed, resuspended in the same buffer and depleted on a Midi-Macs separation unit through an LS column (Miltenyi Biotec). The depleted cells were then counted, washed and seeded (106/ml) in medium containing RPMI-1640, L-glutamine, β-mercaptoethanol, 10% FCS, antibiotics (penicillin, streptomycin) and 20 ng/ml of mouse GM-CSF (PeproTech Asia). Medium was replaced every other day and cells were used for subsequent experiments on days 5–6 post-culture. DCs were enriched using CD11c microbeads (Miltenyi Biotec). Proper culture of BMDC was validated using anti-mouse major histocompatibility complex (MHC) class II-PE and anti-mouse CD11c-allophycocyanin (APC) (cat. no. 12–5321 and 17–0114, respectively; eBioscience, San Diego, CA, USA) (Supporting information, Fig. S2). Mouse splenic B and T cells were isolated using B220 and CD90 microbeads, respectively.

ELISA of IL-12/23p40 and IL-6

To measure the effect of VB-201 on IL-12/23p40 and IL-6 production, Mo-DC were collected 5–6 days post-culture, counted and seeded (106/ml). Cells were pretreated for 1 h with VB-201 followed by 24 h activation with 100 ng/ml LPS from Escherichia coli strain 055:B5 (Sigma, Israel) or 10 μg/ml PGN-SA (InvivoGen, San Diego, CA, USA) to induce cytokine production. IL-12/23p40 and IL-6 concentration in supernatant was measured by ELISA. Cells activated with solvent (0·5% ethanol in PBS) were used as control.

Plasmid transfection

Human embryonic kidney (HEK) 293 cells were transfected for 24–48 h with plasmids encoding human CD14 or human TLR2 (Origene, Rockville, MD, USA) using lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Transfection efficiency was determined by Western blotting and/or flow cytometry using anti-human CD14-fluorescein isothiocyanate (FITC) (cat. no. 11–0149; eBioscience) and anti-human TLR2-PE (cat. no. FAB2616P; R&D Systems, Minneapolis, MN, USA).

Activation of cells and Western blotting

Cells (106/ml) were pretreated for 20 min with VB-201 or VB-207 followed by 15 min activation with 100 ng/ml LPS, 200 ng/ml recombinant human IL-1β (PeproTech Asia), 10 μg/ml PGN-SA, 300 ng/ml Pam3CSK4, 0·5 μg/ml R848, 10 μg/ml cytosine–phosphate–guanosine oligodeoxynucleotide (CpG ODN) 1826 and 1 μg/ml flagellin (all from InvivoGen). Cells were washed and resuspended in lysis buffer containing 1:100 dithiothreitol (DTT), phosphatase and protease inhibitors (Thermo Scientific). Samples were loaded onto a precast Criterion TGX gel (Bio-Rad, Hemel Hempstead, UK) and transferred onto nitrocellulose membrane. Blots were blocked with 5% milk or BSA in Tris-buffered saline and Tween 20 (TBST) for 1 h, followed by incubation with primary and secondary antibodies. Membranes were developed using an ECL kit (Thermo Scientific). The following antibodies were used for immunoblotting:

Primary antibodies: p-p38 (cat. no. 4511; 1:1000), IkBα (cat. no. 9242; 1:1000), p-IKKα/β (cat. no. 2697; 1:1000) and myeloid differentiation primary response gene 88 (MyD88) (cat. no. 4283; 1:1000) were all from Cell Signaling Technology (Danvers, MA, USA). Tubulin (cat. no. T9026; 1:5000), extracellular-regulated kinase (ERK)1/2 (cat. no. M5670; 1:10 000) and p-ERK1/2 (cat. no. M8159; 1:10 000) were purchased from Sigma (Israel). Heat shock protein (HSP) 90 (cat. no. 13119; 1:500) and CD14 (cat. no. sc-58951; 1:500) were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). TLR-2 (cat. no. BAF2616; 1:500) and TLR-4 (cat. no. BAF1478; 1:200) were from R&D Systems (Minneapolis, MN, USA).

Secondary antibodies: HRP donkey anti-rabbit (1:5000) and HRP goat anti-mouse (1:3000) were from Jackson ImmunoResearch (West Grove, PA, USA). HRP donkey anti-goat (1:5000) was from Santa Cruz Biotechnology.

Precipitation

Cells were lysed using a 1% NP-40 lysis buffer containing 1:100 protease and phosphatase inhibitors, followed by 20 min incubation on ice and 15 min centrifugation at maximum speed. Samples were incubated overnight at 4°C with solvent, OB-VB201 or OB-VB207 in a rotator. Streptavidin agarose beads (Sigma, Israel) were added for 2 h. Protein elution was performed with lysis buffer without DTT for 10 min at RT. Sample loading, transfer and immunoblotting were performed as described above.

VB-201 cell-surface binding specificity by flow cytometry

Unless noted otherwise, streptavidin-APC (eBioscience) was used to detect binding in all flow cytometry experiments. To assess VB-201 interference with LPS binding, VB-201 was incubated for 20 min with cells (106/ml) after which 100 ng/ml of biotin-LPS (InvivoGen) was added for an additional 15 min, all at 4°C. To prevent binding of OB-VB201 to CD14 and TLR-2, anti-human CD14-neutralizing antibody clone 18D11 (cat. no. HM2224; HyCult biotech, Uden, the Netherlands), anti-human TLR-2-neutralizing antibody clone T8050-15C (cat. no. C10673; Lifespan Biosciences, Seattle, WA, USA) or a control anti-6x histidine antibody (cat. no. MAB050; R&D Systems) were used for 15 min before adding OB-VB201 for an additional 15 min.

Determination of cholesterol and triglyceride levels

Total plasma cholesterol and triglyceride levels were determined using an automated enzymatic technique (Roche Cobas Mira Analyzer, Indianapolis, IN, USA). Cholesterol in the plasma was measured using a microplate assay (Roche/Hitachi enzymatic assay reagent kits: thermo-triglycerides reagent TR22421 and cholesterol reagent 12016630 122).

VB-201 efficacy studies in atherosclerosis

Atherosclerosis was induced in NZW rabbits by a 14-week high-cholesterol diet (0·5% cholesterol diet Teklad 06060). VB-201 (0, 0·4, 1 and 4 mg/kg) was administered by oral gavage once a day for 14 weeks. Atorvastatin (50 mg/kg) was added to the diet to achieve a target dose of 2·5 mg/kg body weight.

Atherosclerosis evaluation

Lesion area was calculated using a computerized analysis method (Image Pro Plus software, version 4·5.1·29; Medical Cybernetics Corporation, Albemarle, NC, USA). Lesion area was determined using the Image-Pro Plus software, which measures the staining area covered by Sudan IV stain adhering to the lesioned tissue. The percentage of lesioned area was calculated relative to the size of the entire aorta.

Statistical analysis

Mean ± standard deviation was calculated using Sigma-Stat software and statistical significance was calculated using the two-tailed Student's t-test. One-way analysis of variance (anova) or t-test were used to compare experimental groups. Significance level was set at P < 0·05.

Results

VB-201 associates with myeloid mononuclear cells

We initially set out to determine the association of subpopulations of immune cells with VB-201. Isolated subsets from mouse and human mononuclear cells were exposed to [3H]2-radio-labelled VB-201 and subsequently analysed for VB-201 incorporation. Figure 2a,b demonstrates a profound difference between myeloid and lymphoid cells in VB-201 association. VB-201 was found to associate primarily with DC and monocytes, while only relatively low VB-201 counts were measured in lymphocytes. Cell fractionation indicated that VB-201 binding is localized to the membrane compartment (Supporting information, Fig. S3). Moreover, whereas preincubation of cells with ‘cold’ VB-201 or with bioactive phospholipid oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine (Ox-PAPC) [30, 31] inhibited radioactive-VB-201 incorporation, phosphocholine or VB-207, the inactive derivate of VB-201, did not affect the extent of incorporation (Fig. 2c,d). These results ascribe a high level of specificity to VB-201's molecular targeting in freshly isolated immune cell populations.

Figure 2.

VB-201 associates preferentially with myeloid mononuclear cells (MNC). Isolation and generation of cell subsets from (a) human peripheral blood mononuclear cells (PBMC) and (b) mouse spleen and bone marrow (BM) are described in the Materials and methods section. Cells (106) were cultured for 2 h with [3H]2-VB-201. (c,d) Bone marrow-derived dendritic cells (BMDC) were preincubated with the compounds indicated [20 μg/ml VB-201, 20 μg/ml phosphocholine (Ph.Chol), 80 μg/ml VB-207, 100 μg/ml oxidized 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine (Ox-PAPC)] for 1 h before adding [3H]2-VB-201 for 2 h. Radioactivity was evaluated in cell lysates. VB-201 association was measured using a β-counter. Results are mean ± standard deviation of triplicates. One of three experiments is shown.

VB-201 inhibition of TLR-signalling events is circumscribed to TLR-2 and TLR-4

Treatment with Ox-PLs was shown to inhibit Pam3CSK4- and LPS-induced signalling in human Mo-DC and macrophages [30, 33]. Accordingly, we investigated whether VB-201 can influence signals downstream to TLR-2 and TLR-4. Pretreatment of LPS (TLR-4)-activated human monocytes, mouse BMDC (Fig. 3a,b) and mouse peritoneal macrophages (Supporting information, Fig. S4) with VB-201 profoundly inhibited phosphorylation of ERK1/2, p38 and IKKα/β and rescued IkBα degradation. VB-201's inhibitory effect was not limited to MyD88-mediated signalling, as TRIF-mediated phosphorylation of interferon regulatory transcription factor (IRF)-3 was also reduced (Fig. 3a). VB-201 similarly inhibited PGN (TLR2/6)- and Pam3CSK4 (TLR-1/2)-induced downstream phosphorylation in human monocytes and mouse BMDC (Fig. 3a,b) as well as in mouse peritoneal macrophages (Supporting information, Fig. S4). This effect was not restricted to microbial motifs because VB-201 similarly inhibited signalling induced by a non-microbial TLR-2 ligand serum amyloid A (SAA) (Fig. 3c). In line with these results, VB-201 considerably inhibited the expression of the T helper type 1 (Th1)- and Th17-promoting cytokines IL-12/23 and of IL-6 in activated Mo-DC (Fig. 4). None the less, signalling events subsequent to cell activation through plasma membrane TLR-5, endosomal TLR-7 and TLR-9, and IL-1 receptor, which shares the adaptor molecule MyD88 with TLRs, were unfettered by VB-201 (Supporting information, Fig. S5). Taken together, these results indicate that inhibition of TLR signalling and cytokine secretion by VB-201 is restricted to TLR-2 and TLR-4 and takes place in the plasma membrane.

Figure 3.

VB-201 inhibits Toll-like receptor (TLR) signalling restricted to TLR-2 and TLR-4. Human monocytes (a) and mouse bone marrow-derived dendritic cells (BMDC) (b) were pretreated at the indicated VB-201 concentrations and then activated with TLR-2 and TLR-4 agonists. (c) Human monocytes were pretreated with VB-201 as in (a) and then activated with serum amyloid A (SAA) for the indicated time-points. Samples were analysed by Western blotting for inhibition of phosphorylation. α-Tubulin was used for loading control. One of at least three experiments is shown.

Figure 4.

VB-201 inhibits Toll-like receptor (TLR)-2- and TLR-4-mediated cytokine production. Monocyte-derived dendritic cells (Mo-DC) were pretreated at the indicated VB-201 concentrations for 1 h and then activated with lipopolysaccharide (LPS) or peptidoglycan (PGN) for 24 h. Collected supernatants were analysed for secretion of (a) interleukin (IL)-12/23p40 and (b) IL-6. Two representative subjects of more than eight are shown.

VB-201 binds CD14 and TLR-2

The above results warranted experiments to determine whether or not VB-201 binds TLR-2 and TLR-4 to antagonize Pam3CSK4/PGN and LPS-induced signalling. To enable interaction studies, VB-201 and its inactive R1-truncated form, VB-207, were labelled with biotinylated-OVA and initially tested for TLR-induced signalling-inhibition activity. Similar to VB-201, OB-VB201 displayed an inhibitory effect on TLR-4-mediated phosphorylation, but signalling events were intact in the presence of both VB-207 and OB-VB207 (Fig. 5a,b). To test if OB-VB201 and OB-VB207 can directly bind TLR-4 and TLR-2, the two molecules were applied to lysates from human primary monocytes and the CD14-deficient monocyte line THP-1 and precipitated using streptavidin beads. In both human primary monocytes and THP-1 cell lines, TLR-2 precipitated only with OB-VB201 and not with the trimmed OB-VB207 molecule (Fig. 5c,d). Next, we analysed precipitated samples for other components of the LPS recognition complex. Whereas TLR-4 could not be detected in OB-VB201-treated samples, CD14, an essential co-receptor for TLR-4 activity, was found to specifically bind OB-VB201 (Fig. 5c,d). Neither the scavenger receptor CD36, widely described as interacting with Ox-PLs, nor the promiscuous TLR downstream-signalling adaptor molecule, MyD88, were found to interact with OB-VB201 (Fig. 5c,d). As CD14 was also described as a co-receptor for TLR2, we wished to ascertain that their detection following treatment with OB-VB201 was not due to co-precipitation. Accordingly, we transfected 293 cells with either CD14 or TLR-2 and used OB-VB201 for precipitation. The results depicted in Fig. 5e,f reveal that CD14 detection was not due to its co-precipitation with TLR-2, and vice versa.

Figure 5.

VB-201 specifically binds Toll-like receptor (TLR)-2 and CD14. Mouse RAW 264·7 monocytic cells were pretreated at the indicated concentrations of (a) VB-201 and VB207 or (b) BO-VB201 and BO-VB207 and then activated with lipopolysaccharide (LPS). Samples were analysed by Western blotting for inhibition of phosphorylation. Heat shock protein (HSP)90 was used for loading control. (c) Lysates from human monocytes and (d) human monocytic cell line THP-1 were left untreated or incubated with BO-VB201 or BO-VB207. Precipitation was performed using streptavidin beads. (e,f) Human embryonic kidney (HEK) 293 cells were left untreated or transfected with plasmids expressing human CD14 or TLR-2 for 48 h. Cell lysates were incubated with BO-VB201 or BO-VB207 as negative control followed by precipitation with streptavidin beads. One of at least three experiments is shown.

Next, we performed a set of experiments to establish VB-201 binding to cell-surface-expressed CD14 and TLR-2. VB-201 impaired binding of biotinylated LPS to human monocytes in a dose-dependent manner (Fig. 6a). OB-VB201 stained CD14-transfected cells, whereas no staining was observed in cells incubated with OB-VB207 (Fig. 6b), and transfected cells expressing high levels of CD14 were stained distinctly brighter with OB-VB201 than were low-CD14-expressing cells (Fig. 6c). Moreover, anti-CD14-neutralizing antibody abrogated OB-VB201 staining of human monocytes (Fig. 6d). Finally, in contrast to the control isotype, anti-TLR-2-neutralizing antibody impaired BO-VB201 staining of human monocytes (Fig. 6e).

Figure 6.

VB-201 binds cell surface-expressed CD14 and Toll-like receptor (TLR)-2. (a) Human monocytes were incubated with VB-201 at the indicated concentration for 20 min before biotin–lipopolysaccharide (LPS) (100 ng/ml) was added for an additional 15 min. (b) Human embryonic kidney (HEK) 293 cells were left untreated or transfected for 48 h with CD14-expressing plasmid. BO-VB201 and BO-VB207 were used for 15 min. (c) As in (b), but showing stain with BO-VB201 on high- compared to low-CD14-expressing cells. (d,e) Human monocytes were incubated for 15 min with anti-human CD14 blocking antibody (d) or anti-human TLR-2 blocking antibody (e). For control, matched isotype antibody was used at the indicated concentrations before BO-VB201 was added for an additional 15 min. (a–c) Streptavidin–allophycocyanin ((APC) (1:200) was used to detect binding. One of at least three experiments is shown.

Taken together, these results indicate that CD14 and TLR-2 are two distinct and specific targets of VB-201.

VB-201 treatment constrains atherosclerosis development in NZW rabbits

Previous studies demonstrated the roles of TLR-2 and TLR-4 in animal models of atherosclerosis. We therefore tested VB-201's effect on the development of atherosclerosis. Treatment with VB-201 in the high-cholesterol diet atherosclerosis model in rabbits constrained disease development in a dose-dependent manner: 1 mg/kg treatment resulted in a 25% reduction in lesion area in the aorta compared with the control group, whereas the most effective dose of 4 mg/kg yielded a significant 50% reduction (Fig. 7a). This effect was not accompanied by a decrease in total cholesterol concentration. Moreover, efficacy was demonstrated to be correlated with VB-201 steady-state blood levels. We then investigated the effect of VB-201 on atherosclerosis compared with and combined with the eminent lipid-lowering drug atorvastatin. Given individually, VB-201 and atorvastatin resulted in a similar, respective reduction of 33 and 39% in aorta lesion area compared with controls (Fig. 7b). Combined treatment of VB-201 with atorvastatin significantly decreased total aorta lesion area by 53%. The effect of VB-201 on atherosclerosis in rabbits was not accompanied by a decrease in cholesterol abundance as was observed for atorvastatin treatment (Fig. 7b).

Figure 7.

VB-201 inhibits atherosclerosis progression in New Zealand white rabbits. (a) VB-201 was orally administered daily to rabbits at escalating doses (0·1, 1, 4 mg/kg) for 14 weeks, along with a high-cholesterol diet supplementation (0·5% cholesterol) to induce atherogenesis. (b) Rabbits were treated with phosphate-buffered saline (PBS), 4 mg/kg VB-201, 2·5 mg/kg atorvastatin or combined treatment of VB-201 and atorvastatin at the specified doses. VB-201 was orally administered daily for 14 weeks along with a high-cholesterol diet (0·5%), with or without 2·5 mg/kg atorvastatin (supplemented to the diet). (a,b) Upon killing, aorta lesion area and cholesterol plasma level were evaluated. Aorta lesion area (%) is presented as mean ± standaed error. Mean plasma cholesterol concentration (mg/dl) is indicated in each column in yellow (n = 7–8 *P ≤ 0·05 between PBS control and treated groups; no statistically significant differences were found between treated groups).

Discussion

Acute coronary syndrome and myocardial infarction are arterial diseases that manifest as atherosclerosis. These detrimental outcomes are caused by rupture of atherosclerotic plaque and the occlusion of arterial blood vessels as a result. Although chronic inflammation is the underlying mechanism for atherosclerosis pathogenesis, current available therapies largely address the atherosclerosis-related high-risk factor hyperlipidaemia and aim to lower LDL cholesterol.

Our study presents an alternative concept for the treatment of atherosclerosis in which inflammation, and not hyperlipidaemia, is targeted successfully using modified naturally occurring Ox-PL small molecules.

Monocytes/macrophages and DCs are implicated in the entire process of atherogenesis [34, 35]. Activation of DCs, and of macrophages in particular, relies for the most part on interactions between PRRs and their corresponding ligands. In atherosclerosis, macrophages were shown to have increased expression of TLR-2 and TLR-4 [36, 37] and a broad range of ligands, including microbial motifs such as LPS and endogenous danger-associated molecular patterns (DAMP) released from injured tissue and necrotic cells, are postulated to use these receptors for disease development. Within DAMP, Ox-PLs that are generated abundantly at inflammation sites used CD14 to induce ERK1/2 activation in macrophages and triggered expression of inflammation mediators via TLR-4 [38]. Moreover, under endoplasmatic reticulum stress, Ox-PL induction of macrophage apoptosis was TLR-2-dependent [39]. These findings indicate that Ox-PLs can perpetuate chronic inflammation in atherosclerosis. Paradoxically, few studies have shown that Ox-PLs can inhibit TLR-2- and TLR-4-mediated activation of macrophages and DCs [30, 33] and constrain inflammation in vivo [32].

We generated a family of modified, synthetic small molecule analogues called lecinoxoids (lecin for lecithin – i.e. phospholipid, and oxoid for oxidized) and showed that the lead compound, VB-201, antagonizes ligation of microbial components to CD14 and TLR-2 on monocytes and DCs, thereby inhibiting all anticipated downstream signalling events. Moreover, our data suggest that signalling inhibition can be applied to TLR-2-associated endogenous ligands, as was demonstrated using the inflammation marker SAA. Additional studies to assess VB-201's inhibitory effect on other TLR-2- and TLR-4-related DAMPs are under way.

The imperative role played by TLR-2 and TLR-4 in atherogenesis was demonstrated in several gene-targeted mouse strains. Mice bearing a missense mutation in the TLR-4 gene were resistant to atherosclerosis [40], and TLR-2 deficiency in ApoE−/− and LDLR−/− mice resulted in reduced lesion areas [41, 42]. Therefore, given VB-201's antagonistic effect on TLR-2- and TLR-4-mediated signalling, we postulated that it could inhibit atherogenesis in vivo. Our results show that VB-201 indeed constrains atherosclerosis in the rabbit model.

We identified several pertinent features with respect to VB-201 application in vivo. (i) VB-201 constrains atherogenesis without affecting lipid profiles; although high cholesterol is associated strongly with cardiovascular diseases and is treated to prevent possible future events, cholesterol levels following VB-201 treatment remained unchanged. Therefore, and as supported by the in-vitro signalling and cytokine production data, it is plausible that VB-201 cannot affect atherosclerosis progression other than by restraining inflammation in vivo. (ii) VB-201 has an additive effect to atorvastatin; treatment with atorvastatin had a distinct effect on plaque formation in rabbits, and concurrent dosing with VB-201 resulted in increased and more significant efficacy. This observation suggests that administration of VB-201 in parallel to statins, targeting inflammation and cholesterol abundance, respectively, is conducive for the treatment of atherosclerosis. (iii) VB-201 is available orally.

Ever since their introduction, statins have been the first choice in the prevention of cardiovascular events. Despite the wealth of options available for treating high cholesterol, a remedy is needed that will directly address chronic inflammation in atherosclerosis. Given its aforementioned merits, VB-201 may pave the way to an alternative approach according to which synthetic, modified Ox-PL small molecules could be used to counter inflammation for the prevention and treatment of atherosclerosis and sequelae in humans.

Acknowledgement

This study was funded by VBL Therapeutics.

Author contributions

I. M., E. F., Y. S., N. Y. and D. H participated in the research design. N. Y., Y. S., I. L., O. P.-M. and A. S. conducted the experiments. I. M., E. F., N. Y and D. H. performed the data analysis. I. M., E. F., J. G. and E. B. wrote or contributed to the writing of the manuscript.

Disclosures

The authors are employees and stock options holders of VBL therapeutics.

Ancillary