Therapeutic potential of a synthetic FABP4 inhibitor 8g on atherosclerosis in ApoE-deficient mice: the inhibition of lipid accumulation and inflammation

Abstract

Fatty-acid-binding proteins are small (14–15 kDa) proteins that bind reversibly with high affinity to hydrophobic ligands. Fatty-acid-binding protein 4 (FABP4), highly expressed in adipocytes and macrophages, plays an essential regulatory role in energy metabolism and inflammation. In a previous study, we reported N-(2-(4-(1-allyl-2,4-dioxo-2,3,4,5-tetrahydro-1H-pyrrolo(3,2-d)pyrimidin-6-yl)phenoxy)ethyl)picolinamide (8g), as an effective agent to prevent non-alcoholic fatty liver disease (NAFLD). In the present study, we found 8g to be a novel FABP4 inhibitor that significantly inhibited triglyceride accumulation and the expression of Fabp4 in 3T3-L1 adipocytes. In macrophages, 8g inhibited both Fabp4 and pro-inflammatory cytokine production. Importantly, in a co-culture system of adipocytes and macrophages, which mimics the functional interaction between adipocytes and macrophages within adipose tissue, 8g inhibited the expression of Fabp4 and cytokine production, and downregulated FABP4 and stress kinases. In line with the in vitro results, 8g markedly and dose-dependently decreased the expression of serum FABP4 and atherosclerotic lesion area in apolipoprotein E-deficient (ApoE-deficient) mice, and significantly reduced epididymal fat mass and plasma levels of triglycerides in diet-induced obesity (DIO) mice. All together, 8g is a novel FABP4 inhibitor demonstrated to ameliorate atherosclerosis through the reduction of lipid accumulation and inflammatory response, which may offer a potent therapeutic strategy against atherosclerosis and obesity.

---

Introduction

Lipids and lipid signaling are predominately involved in metabolic and inflammatory responses, and exhibit key roles in the pathogenesis of metabolic diseases including atherosclerosis, obesity and type 2 diabetes.^1–3 Cytoplasmic fatty-acid-binding proteins (FABPs), intracellular lipid chaperones, are a family of 14–15 kDa proteins that reversibly bind to hydrophobic ligands, such as long-chain fatty acid eicosanoids and other lipids.^4,5 The adipocyte FABP, also known as fatty-acid-binding protein 4 (FABP4) or adipocyte protein 2 (ap2), is an important member of the FABP family and is mainly expressed in adipocytes and macrophages, and regulated by fatty acids (FAs), insulin, and peroxisome-proliferator-activated receptor-c (PPARc) agonists.^6,7 In adipocytes, free fatty acids (FFAs) bind to FABP4/aP2 and lead to the differentiation of small adipocytes into large adipocytes, which store more triglycerides and also release FFAs and adipokines, including tumor necrosis factor (TNF) and interleukins (IL).⁸ In macrophages, FABP4 expression is induced by oxidized low density lipoprotein (ox-LDL) and Toll-like receptor activators.^9,10 With FFA-FABP4/aP2 activation, macrophages differentiate into foam cells, which store cholesterol esters and release cytokines such as TNF, IL, and monocyte chemoattractant protein (MCP-1).⁸

Considerable studies in FABP4 deficient mice have demonstrated the regulated roles of FABP4 in glucose and lipid metabolism in metabolic diseases. FABP4 deficient mice, when fed with a high-fat and high-calorie diet, become obese but develop neither insulin resistance nor diabetes, suggesting that FABP4 is a link between obesity, insulin resistance and diabetes.^11–13 A macrophage-specific lack of FABP4 has been reported to protect mice deficient in apolipoprotein E (ApoE-/-) against early and advanced phase atherosclerosis.^7,14 These findings indicate that FABP4 holds a central role in the development of metabolic syndromes through its distinct effects on adipocytes and macrophages, and its potential to regulate metabolic and inflammatory responses. Thus, suppression of FABP4 may serve as a therapeutic strategy for various components of metabolic syndromes, such as insulin resistance, type 2 diabetes, and atherosclerosis.

In a previous study, we reported 8g as an effective agent to prevent non-alcoholic fatty liver disease (NAFLD).³⁴ In the present study, we discovered that a synthetic FABP4 inhibitor 8g (Fig. 1A) suppressed the lipid accumulation in DIO mice and further ameliorated the symptoms of atherosclerosis in ApoE-/- mice. Importantly, aiming to understand the FABP4 inhibitor 8g on the lipid accumulation and the pro-inflammatory cytokine expression in atherosclerosis, the mechanisms of the adipocyte-macrophage co-culture system and THP-1 derived foam cell formation during cholesterol loading were explored.

Fig. 1 In vitro FABP4-inhibitory activity of 8g. (A) Structure of 8g. (B) 8g (2.23 μM) showed different FABP4 agonist activities compared to arachidonic acid (2.75 μM) in FABP4 reporter assays. (C) 8g forms H-bonding interactions with residues of Tyr-19, Arg-78, and Ser-53. (D) 8g forms hydrophobic interactions with residues of Met-20, Ala-33, Ala-36 and Ala-75. Data are the mean ± SEM (n = 3).

Results

In vitro FABP4-inhibitory activity of 8g

The inhibitory activity of FABP4 was determined by incubating different concentrations of 8g with fluor-labeled FABP4 protein (Fig. 1A), and arachidonic acid, a known ligand of FABP4, as a positive control, was also determined. The IC₅₀ of 8g is 2.23 μM in comparison to the 2.75 μM of arachidonic acid (Fig. 1B), suggesting that 8g has a potent inhibitory effect on FABP4. The crystal structure (PDB code: 2QM9)²⁰ for troglitazone bonded to FABP4 was employed as docking template. 8g docked well into the ligand binding site by AutoDock4.2.²¹ From the docking results, 8g formed H-bonding interactions with residues of Tyr-19, Arg-78, and Ser-53. And the pyrrolo(3,2-d)pyrimidine-2,4-dione group was able to align well with the thiazolidinedione of troglitazone. In addition, 8g could form hydrophobic interactions with residues of Met-20, Ala-33, Ala-36 and Ala-75. Interestingly, the aromatic rings of 8g may form π–π stacking interactions with itself (Fig. 1C and D).

8g reduced lipid accumulation in 3T3-L1 adipocytes

To examine the effect of 8g on the adipocytes, we cultured 3T3-L1 pre-adipocytes with IBMX (3-isobutyl-1-methylxanthine), dexamethasone and insulin to obtain mature adipocytes. 8g was added to test the suppression of lipid accumulation. At the same time, metformin and rosiglitazone were the positive and negative controls, respectively. Oil Red O staining results showed that 8g more effectively suppressed the lipid accumulation in comparison to metformin in a dose-dependent manner (Fig. 2A and S1A†). In contrast, rosiglitazone, as a PPARγ agonist, increased the lipid accumulation. Furthermore, the differentiated adipocytes were lysed for the determination of intracellular triglycerides. Consistent with the results of the Oil Red O staining, 8g more effectively and dose-dependently inhibited triglyceride accumulation in 3T3-L1 adipocytes compared to metformin at the same concentration (Fig. 2B and C). Rosiglitazone promoted the lipid accumulation, which is in line with the side effect of body weight gain in the treatment of diabetes.

Fig. 2 Effect of 8g on the lipid accumulation in 3T3-L1 adipocytes. (A) Cells were stained with Oil Red O. Magnification, ×200. (B) Cells were lysed and the triglyceride content was detected in metformin-treated, rosiglitazone-treated, and 8g-treated differentiated adipocytes at a concentration of 10 μM. (C) Triglyceride content was detected in differentiated 3T3-L1 adipocytes treated with different concentrations of 8g. (D) Effect of 8g on the expression of certain key lipogenic genes in differentiated 3T3-L1 adipocytes at a concentration of 10 μM. (E) Effect of 8g on the expression of Fabp4 in differentiated 3T3-L1 adipocytes at different concentrations (2.5, 5, and 10 μM). (F) Cells infected with siRNA or treated with 8g (2.5, 5, 10 μM) and BMS309403 (10 μM), then differentiated to mature adipocytes for 8 days, stained with Oil Red O. Magnification, ×200, the scale bar indicates 100 μm. Data are shown as the mean ± S.E.M. *P < 0.05, **P < 0.01, and ***P < 0.001.

8g down-regulated the expression of Fabp4 and other lipid metabolism genes in 3T3-L1 adipocytes

It was reported that the expression of certain key lipogenic genes including Fabp4, Pparg, C/EBPα, Leptin, lpl, FAS, FABP5 and Ppara would be upregulated during the differentiation of adipocytes.^5,22 To further investigate the effect of 8g on the lipid accumulation, the expression of key metabolic genes were measured by a quantitative real-time polymerase chain reaction. As shown in Fig. 2D, after differentiated 3T3-L1 cells were treated with 8g (10 μM), the expression of certain key lipogenic genes was significantly increased in comparison with control cells, including Fabp4, Leptin and C/EBPα. Interestingly, 8g also down-regulated the expression of Pparg, while BMS309403, a well-characterized FABP4 inhibitor, up-regulated the levels of Pparg. The decreased Pparg level consequently down-regulated the expression of lipogenic genes, such as CCAAT-enhancer-binding protein α (C/EBPα). Fas and Lpl gene expressions were modestly down-regulated with treatment of 8g, while Fabp5 and Ppara gene expressions were not decreased after treatment with 8g. In addition, Fabp4 gene expression was suppressed in a dose-dependent manner (Fig. 2E).

In order to determine whether 8g is selective against Fabp5, we chose LL/2 cells which expressed the Fabp5 gene and had little expression of Fabp4 gene to investigate the inhibition of 8g. As exhibited in Fig. S1B,† the Fabp5 gene was not reduced in LL2 cells treated with 8g, which indicated that 8g may be a selective inhibitor of FABP4.

8g reduced lipid accumulation similar to 3T3-L1 cells which deficiency of Fabp4

To investigate the effects of FABP4 deficiency on adipocyte differentiation, we chose 3T3-L1 cells to obtain mature adipocytes. 3T3-L1 cells were transfected with siRNA against Fabp4 mRNA. Cells were transfected or treated with 8g and at the same time, BMS309403 was added as a positive control. Differentiation of 3T3-L1 preadipocytes to mature adipocytes was performed as previously described.¹⁵ On day 8 after differentiation, the accumulation of lipids in transfection cells was decreased (Fig. 2F), while there was no effect of lipid accumulation in control cells. Oil Red O staining further showed that 8g more effectively suppressed lipid accumulation than BMS309403 at the same concentration (10 μM) and its inhibition was dose-dependent. Compared with the transfected 3T3-L1 cells, 8g showed a similar effect on the inhibition of triglyceride accumulation (Fig. 2F). These data indicate that 8g may be a potent FABP4 inhibitor.

8g decreased the expression of the Fabp4 gene and the production of cytokines in RAW264.7 macrophages

In macrophages, the expression of Fabp4 was up-regulated after LPS or PMA stimulation.⁷ RAW264.7 macrophages possess little PPARγ, yet lipopolysaccharide (LPS)/interferon (IFN)-induced iNOS was inhibited by synthetic PPARγ. Toll-like Receptor 4 (TLR4) was reported to be involved in the interaction between adipocytes and macrophages.¹⁷ We stimulated TLR4 with LPS in RAW264.7 cells and determined the production of cytokines. ELISA results showed 8g inhibited the expression of IL-6 at concentrations of 5 μM (**p < 0.01) and 10 μM (***p < 0.001), respectively, and decreased the production of TNF-α and IL-1β at a concentration of 10 μM (Fig. 3A–C). Furthermore, qPCR studies revealed that 8g suppressed the mRNA expression of Fabp4, TNF-α, IL-6 and IL-1β in the LPS-induced RAW264.7 macrophages (Fig. 3D). These data indicate that 8g also suppresses the expression of Fabp4 in the inflammatory reaction.

Fig. 3 Effect of 8g on the expression of cytokines and relative genes in RAW264.7 cells and co-culture system. The levels of IL-6 (A), TNF-α (B), and IL-1β (C) were determined in RAW264.7 cells. (D) Expression of Fabp4, IL-6, TNF-α, and IL-1β in RAW264.7 cells. The levels of IL-6 (E), TNF-α (F), and IL-1β (G) were determined in RAW264.7 and 3T3-L1 co-culture systems. (H) Expression of fabp4, IL-6, TNF-α, and IL-1β in 3T3-L1 adipocytes and RAW264.7 cell co-culture systems. Data are shown as the mean ± S.E.M. *P < 0.05, **P < 0.01, and ***P < 0.001.

8g inhibited the expression of Fabp4 and inflammatory mediators in the co-culture system

To examine the effects of 8g on the interaction between adipocytes and macrophages, we established a direct co-culture system using 3T3-L1 adipocytes and RAW264.7 macrophages, which mimic the functional interaction between adipocytes and macrophages in adipose tissue. Secretion levels of inflammatory mediators such as IL-6, TNF-α, and IL-1β were determined using an ELISA kit (Fig. 3E–G). The expression of these inflammatory mediators was markedly increased in the conditioned medium from the co-culture compared with that from the control culture. However, treatment with 8g at 10 μM in the co-culture system significantly decreased the secreted levels of inflammatory mediators of TNF-α, IL-6 and IL-1β. In addition, 8g did not affect the viability of RAW264.7, 3T3-L1 and THP-1 cells, the IC₅₀ values of 8g in these cells were respectively 58.7, 60.1 and 85.5 μM (Table S2†), suggesting that 8g has anti-inflammatory effects in the contact co-culture system of adipocytes and macrophages without obvious toxicity. Fabp4 and inflammatory gene expression in the 3T3-L1 and RAW264.7 co-cultured system was also determined to evaluate the effect of 8g on the inflammatory response (Fig. 3H). Quantitative real-time PCR results indicated that 8g (2.5, 5, and 10 μM) suppressed gene expression of Fabp4, TNF-α, IL-6 and IL-1β in a dose-dependent manner. The results showed that 8g inhibited the expression of Fabp4 and cytokine production in a co-culture system.

8g suppressed FABP4 and the activity of lipid sensitive protein kinases in the co-culture system

FABP4 might coordinate the lipid-mediated activation of stress kinases such as JNK or IKK under immune or metabolic stimuli, thus directly linking lipid signals to the established role of these kinases in pro-inflammation and anti-insulin action.^2,23 To examine whether 8g modified the inflammatory profile and insulin action by this mechanism, the activity of JNK, p38MAPK and IKKβ was analyzed using a western blot (Fig. 4). While the expression of FABP4 in co-culture system was significantly inhibited by 8g at 10 μM, there was also a significant reduction of JNK, p38MAPK and IKKβ activity in co-culture treated with 8g compared with vehicle-treated controls, especially at high concentration (10 μM). These data indicate that 8g may suppress the functional interaction between adipocytes and macrophages in adipose tissue.

Fig. 4 Effect of 8g on the expression of lipid sensitive protein kinases and FABP4 in a co-culture system. Expression of p-JNK/JNK, p-p38MAPK/p38MAPK, p-IKKβ/IKKβ and FABP4 in a co-culture system was analyzed by western blot. Data are shown as the mean ± S.E.M. *P < 0.05, **P < 0.01.

8g reduced lipid content in foam cells

Macrophage foam cells play a critical role in the development and progression of atherosclerosis. Expression of Fabp4 mRNA and proteins was up-regulated during this process.^9,24 As 8g decreased the lipid accumulation in DIO mice, we further investigate the effect of FABP4 inhibitor 8g on the lipid accumulation in THP-1 derived foam cells by treating with 8g (10 and 2.5 μM) and a FABP4 inhibitor BMS309403 (10 μM as positive control). The results showed that ac-LDL significantly promoted the accumulation of neutral lipids in PMA-treated THP-1 cells after incubation for 48 h as measured by Oil Red O staining (Fig. 5). However, 8g significantly reduced the accumulation of lipids in foam cells at a concentration of 2.5 μM, but with only a minor reduction at a higher concentration (10 μM). In addition, the effect of 8g on the foam cells was similar to the positive control BMS309403.

Fig. 5 Effect of 8g on the lipid accumulation in foam cells. Foam cells derived from THP-1 cells were treated with 8g (2.5 and 10 μM) and BMS309403 (10 μM) for 24 hours. Cells were stained with Oil Red O and quantitated at 500 nm. Data are shown as the mean ± S.E.M. *P < 0.05, **P < 0.01.

8g decreased the serum FABP4 and ameliorated atherosclerosis in ApoE-/- mice

To determine whether 8g can alter the development of vascular lesions, we performed a late intervention study in the ApoE-/- mouse model of atherosclerosis on a western diet. 8g at 5 and 10 mg kg⁻¹ was administered after 12 weeks of a western diet, when significant atherosclerosis had developed. Analysis of the en face aorta demonstrated marked reductions in atherosclerotic lesion area in the 8g-treated group compared with the vehicle group (42% in 5 mg kg⁻¹ treated group, 83% in 10 mg kg⁻¹ treated group, Fig. 6A and B). Atorvastatin, a safe drug for atherosclerosis, as a positive control at 5 mg kg⁻¹, showed 45% reduction in the atherosclerotic lesion area.

Fig. 6 Effect of 8g on the ApoE deficient mouse model. ApoE-/- mice were treated with 8g (5 and 10 mg kg⁻¹) and atorvastatin (5 mg kg⁻¹) for 4 weeks. (A) The aorta of mice in each group was isolated and stained by Oil Red O. (B) The atherosclerotic lesion area of each group was quantitated. (C–E) The levels of serum IL-6, TNF-α and IL-1β in mice. (F) The level of FABP4 in mice. Data are shown as the mean ± S.E.M. *P < 0.05, **P < 0.01, ***P < 0.001.

The effect of 8g on the physical and biochemical characteristics of ApoE-/- mice in atherosclerosis was investigated. Interestingly, 8g decreased the serum levels of TG and TC in ApoE-/- mice, and it significantly improved the total cholesterol in serum at 10 mg kg⁻¹ (*p < 0.05, **p < 0.01) (Table 1). 8g also increased HOMA-IR at 10 mg kg⁻¹ (*p < 0.05), suggesting that 8g might improve the insulin resistance in ApoE-/- mice. According to the area under the curve (AUC) of GTT, the total glucose showed the difference between ApoE-/- mice and 10 mg kg⁻¹-treated groups (P < 0.05), indicating that 8g might improve glucose tolerances in ApoE-/- mice (Table 1). Macrophages participated in the pathogenesis of atherosclerosis partly through the production of inflammatory mediators. Hence, the impact of 8g on several critical inflammatory cytokines, including TNF-α, IL-6, and IL-1β was determined; these inflammatory cytokines were significantly reduced in a dose-dependent manner. Particularly, IL-6 was significantly decreased by 85.43% (p < 0.01) at 5 mg kg⁻¹ and 99.57% at 10 mg kg⁻¹ (p < 0.001), indicating that 8g may inhibite the mitogen activated protein (MAP) kinase family, which is consistent with the previous study in a co-culture system (Fig. 6C–E). More importantly, the level of serum FABP4 in mice treated with 8g was significantly decreased both at 5 mg kg⁻¹ and 10 mg kg⁻¹ (Fig. 6F). 8g did not influence bodyweight, food intake rate, HbA1c, and FFA level in ApoE-/- mice (Table 1), which is consistent with previous observations made in mice with genetic deficiency of FABP4 in the ApoE-/- background.¹⁴

Table 1 Effects of 8g on the biochemical parameters of plasma in ApoE-/- mice for 8 weeks

	WT	ApoE-/-	AS (5 mg kg^−1)	8g (5 mg kg^−1)	8g (10 mg kg^−1)
n	10	10	10	10	10
Bodyweight (g)	36.3 ± 0.3	37.5 ± 1.2	36.6 ± 0.8	36.9 ± 1.0	36.5 ± 0.7
Food intake rate (g per mouse per day)	5.05 ± 0.12	5.45 ± 0.25	5.69 ± 0.31	5.40 ± 0.63	5.84 ± 0.27
Epididymal fat mass (g)	0.33 ± 0.03***	1.72 ± 0.17	1.43 ± 0.21	1.32 ± 0.16	1.20 ± 0.11*
TG (mM)	1.03 ± 0.04***	4.20 ± 0.43	2.37 ± 0.42	2.13 ± 0.17	1.63 ± 0.09*
TC (mM)	2.13 ± 0.19***	34.30 ± 4.79	27.87 ± 2.13	23.87 ± 4.06*	17.27 ± 3.38**
HOMA-IR	2.62 ± 0.17**	5.24 ± 0.18	5.39 ± 0.22	5.11 ± 0.10	4.57 ± 0.05*
GTT AUC (mg dl^−1 min^−1)	23 302 ± 533***	31 692 ± 1070	28 435 ± 1371	27 659 ± 724*	26 399 ± 873*
ITT AUC (mg dl^−1 min^−1)	6269 ± 194**	10 985 ± 967	9211 ± 678	8638 ± 537	8389 ± 325
HbA1c (%)	5.06 ± 0.06	5.15 ± 0.13	5.05 ± 0.21	5.06 ± 0.21	5.00 ± 0.14
FFA (ng L^−1)	19.77 ± 0.27**	27.88 ± 0.45	23.52 ± 1.52	24.29 ± 1.23	20.94 ± 2.58

---

8g ameliorated lipid accumulation in DIO mice

PPARγ is mainly expressed in adipose tissue where it stimulates adipogenesis and lipogenesis, adipocyte differentiation is probably inhibited by PPARγ activity. BMS309403 was reported to ameliorate dyslipidemia but not insulin resistance in DIO mice. In 3T3-L1 cells, we found that doses of 2.5, 5 and 10 μM 8g all reduce gene expression of Fabp4, yet only concentrations higher than 10 μM decrease TG accumulation in 3T3-L1 cells (Fig. 2), moreover, 8g decreased the expression of Pparg. These data suggest that an affect by 8g on PPARγ may be the mechanism for the decrease in TG. Then we carried out a study to investigate the effect of 8g on insulin sensitivity and metabolic parameters in DIO mice. We dosed DIO mice at 12 weeks of age with 8g at 5, 10, and 20 mg kg⁻¹ for 6 weeks. Among the results, the average body weight was reduced respectively by 7.4%, 8.5% and 9.7% without causing dramatic changes in accumulative food intake; the mean mass of epididymal fat mass decreased respectively by 9.28%, 24.23% and 36.08% (Table S3,† Fig. 7A and C). As shown in Fig. 7D and E, 8g administered at 10 mg kg⁻¹ and 20 mg kg⁻¹ significantly reduced plasma levels of TG in a dose-dependent manner. Particularly, levels of TG in 20 mg kg⁻¹-treated group almost reduced to the normal group; the reduced percentages of TG by the administration of 8g (20 mg per kg per day) were 23.60%. 8g also decreased the serum levels of TC in DIO mice, and it significantly improved the total serum cholesterol in a dose-dependent manner (*p < 0.05, **p < 0.01) (Table S3†). According to area under the curve (AUC) of GTT, total glucose showed a significant difference between HFD and 20 mg kg⁻¹-treated groups (P < 0.05), indicating that 8g might improve glucose tolerance in DIO mice (Fig. 7F); but the AUC of ITT for HFD and 20 mg kg⁻¹ groups was not significantly different (p > 0.05) (Table S3†), which is consistent with previous observations made in DIO mice with the treatment of BMS309403.³³ 8g did not influence HbA1c and FFA levels in DIO mice (Table S3†), which is consistent with previous observations made in ApoE-/- mice. These results provided the in vivo evidence to indicate that 8g inhibits adipogenesis/lipogenesis with adipose tissue as predicted from our findings in 3T3-L1 cells, and indicates that 8g ameliorated dyslipidemia in DIO mice.

Fig. 7 Effect of 8g administration on in vivo lipid metabolism in DIO mice. (A) Body weight, (B) food intake, (C) epididymal fat mass, (D) plasma TG, (E) plasma FFA, (F) glucose tolerance, (G) insulin tolerance. Data are shown as the mean ± S.E.M. *P < 0.05, **P < 0.01, ***P < 0.001.

Discussion

Atherosclerosis is a progressive disease that is associated with multiple cardiovascular risk factors. It was formerly considered as a bland lipid storage disease, but recently many studies demonstrated that it involved an ongoing inflammatory response. It was reported that inflammation played a fundamental role in mediating all stages of atherosclerosis from initiation through to progression and, ultimately, thrombotic complications.^25,26 The adipocyte/macrophage FABP4 holds a critical role in the regulation and dysregulation of metabolic and inflammatory responses since FABP4 coordinates the functional interactions between macrophages and adipocytes, two critical cell types in adipose tissue.^23,27 BMS309403, a selective inhibitor of FABP4, was reported to improve glucose tolerance in ob/ob mice, DIO mice and atherosclerosis in ApoE-/- mice.²⁸ Here, we have provided a potent therapeutic strategy for atherosclerosis through suppression of lipid accumulation and inflammation both in vitro and in vivo.

Expression of FABP4 is highly regulated during differentiation of adipocytes, and its mRNA is transcriptionally controlled by fatty acids, PPAR-γ agonists and insulin.²⁹ The epidermal FABP, FABP5 (also named as mal1), which is the second isoform expressed in adipocytes, was also present in macrophages. Furthermore, in FABP4-/- mice, compensatory regulation causes a dramatically increased expression of mal1 in adipocytes. However, the expression of fabp5 did not appear to be upregulated in FABP-/- primary macrophages.^7,12 Interestingly, our research showed that 8g dramatically decreased the mRNA level of Fabp4 in adipocytes, but the level of mRNA encoding Fabp5 did not increase at the same time. This result suggested that 8g also affected the expression of Fabp5 in adipocytes, and more work is needed to be done to explore the underlying mechanism.

FABP4 plays a critical role in both inflammation and metabolic response during nutrient overload.²³ In 2003, Guillaume Charrière et al. first found that preadipocytes could partially undergo phenotype conversion to macrophages in a cell-to-cell contact system.³⁰ In 2005, Takayoshi Suganami et al. developed an in vitro co-culture system composed of adipocytes and macrophages to examine the molecular mechanism whereby the cells communicate.³¹ Because of the integration ability of FABP4 in metabolic and inflammatory responses, we introduced the contact co-culture system composed of adipocyte and macrophages into the investigation of FABP4 inhibitors in our research. Through this system, we evaluated the effects of the FABP4 inhibitors on inflammatory and metabolic responses at the same time. More importantly, we further studied the influence of FABP4 inhibitors on cell communication. In our results, we found that 8g decreased the levels of mRNA as well as proteins of pro-inflammatory cytokines such as TNF-α, IL-1β, and IL-6 in both macrophages and contact co-culture system. Moreover, 8g was more potent in decreasing the production of cytokines in a co-culture system than in macrophages; these results suggested that 8g would be more effective for improving the inflammatory environment under the nutrient overload conditions.

Although our understanding of the precise mechanisms that underlie FABP4 action is incomplete, it is certain that inflammatory pathways, including the activation of c-Jun amino-terminal kinase (JNK) and IKKβ in adipocytes and macrophages are strongly influenced by FABP4. Mounting evidence indicates that activation of JNK, IKKβ and conventional protein kinase C (PKC) is central to mediate insulin resistance in response to various stresses that occur in obesity and other conditions of insulin resistance.² The mitogen activated protein (MAP) kinase family, including p38 MAPK, ERK1/2, and JNK, is necessary for macrophage foam cell formation.³² In our investigation, we found that 8g inhibited the expression of FABP4 and the phosphorylation of MAPKs and IKKβ in the co-culture system composed of adipocytes and macrophages, and further decreased the level of HOMA-IR in ApoE-/- mice. These results indicated that 8g might improve insulin resistance and foam cell formation in the process of atherosclerosis. We also found, when chronically administered in vivo, that 8g seemed to be effective in ameliorating dyslipidemia. However, no significant effects on insulin resistance were observed in DIO mice.

In the present study, FABP4 inhibitor 8g was demonstrated to ameliorate atherosclerosis through the reduction of lipid accumulation and inflammatory response, and therefore the results offer a novel therapeutic strategy for the treatment of cardiovascular diseases and maybe also for obesity and type 2 diabetes.

Experimental

Reagents

8g and BMS309403 were synthesised by the State Key Laboratory of Biotherapy, Sichuan University, China. Primary and secondary antibodies were purchased from Cell Signaling Technology. Ac-LDL was obtained from Biomedical Technologies. Oil Red O, IBMX, insulin, LPS were purchased from Sigma. Metformin, rosiglitazone and atorvastatin were purchased from Selleck.

FABP4-inhibitory assay

The inhibitory activity of 8g was assayed with a FABP4 Inhibitor/Ligand Screening Assay Kit (Cayman Chemical, Item Number 10010231). A Detection Reagent in this kit exhibits increased fluorescence at 475 nm when bound to FABP4, and a ligand and/or inhibitor of FABP4 can displace the Detection Reagent, thereby reducing the fluorescence. The assay was performed according to the supplier’s protocol.

Cell culture

3T3-L1 preadipocytes, RAW264.7 macrophage cell line (American Type Culture Collection, Manassas, Va) were maintained in Dulbecco’s modified Eagle’s medium (Hyclone, Logan, Utah, USA) containing 10% fetal bovine serum (FBS, Hyclone, US). Human monocytic cell line THP-1 cells were cultured in RPMI-1640 medium (Hyclone, Logan, Utah, USA) supplemented with 10% fetal bovine serum, penicillin (100 U ml⁻¹) and streptomycin (100 μg ml⁻¹) in a CO₂/O₂ incubator at 37 °C. Differentiation of 3T3-L1 (1.0 × 10⁵ cells) preadipocytes to mature adipocytes was performed as previously described,¹⁵ and used as differentiated 3T3-L1 adipocytes at day 8 to 10 after the induction of differentiation.

Oil Red O staining

At day 8 after the induction of 3T3-L1 differentiation, different concentrations of 8g (2.5, 5, 10 μM), metformin (10 μM), and rosiglitazone (10 μM) were added. After 24 h incubation, cells were washed in PBS and then fixed for 20 min at room temperature (25 °C) with 4% paraformaldehyde. Cells were then washed with deionized water and stained with Oil-Red-O solution. For quantification, Oil-Red-O staining was dissolved in isopropanol and absorbance was measured at 500 nm.¹⁶

Triglyceride assay

At day 8 after the induction of 3T3-L1 differentiation, different concentrations of 8g (1.25, 2.5, 5, 10, 20, 40 μM), metformin (10 μM), and rosiglitazone (10 μM) were added. After 24 h incubation, cells were washed and lysed for the triglyceride detection according to the manufacturer’s protocol using the Triglyceride Quantification Kit (BioVision, Mountain View, CA).

Detection of cytokines in RAW264.7 macrophages

RAW264.7 macrophages were treated with LPS (1 μg ml⁻¹) and different concentrations of 8g (2.5, 5 and 10 μM) for 24 hours, and the culture supernatant was collected. TNF-α, IL-6, and IL-1β concentration in the culture supernatants were measured by using a commercial immunoassay (mouse TNF-α, IL-6, IL-1β enzyme linked immunosorbent assay kits, R&D Systems, Minneapolis, MN).

Co-culture of adipocytes and macrophages

Co-culture of adipocytes and macrophages was performed as described with some modification.¹⁷ In the contact system, differentiated 3T3-L1 adipocytes (≈0.5 × 10⁶ cells) were cultured in a 35 mm dish and macrophages (1.0 × 10⁵ cells of RAW264.7 macrophages) were plated onto 3T3-L1 adipocytes. The cells were treated with LPS (1 μg ml⁻¹, Escherichia coli O111:B4, Sigma) and cultured for 24 h while 3T3-L1 and RAW264.7 cells were contacted to each other, then we harvested cells. Supernatant was collected and the concentrations of TNF-α, IL-6, and IL-1β were detected by using a commercial immunoassay. As a control, adipocytes and macrophages, the numbers of which were equal to those in the co-culture, were cultured separately and mixed after harvest.

Quantitative real-time polymerase chain reaction

Total RNA was extracted from cultured cells from 100 mm dishes (3T3-L1 cells, RAW264.7 cells or co-culture cells) using TRIzol reagent (Invitrogen) and quantitative real-time PCR was performed with an CFX96 Touch™ Real-Time PCR Detection System using PCR Master MixReagent (Bio-Rad Laboratories, Hercules, CA) and 4% of the starting 1 μg RNA with the primers at 450 nmol l⁻¹ per 20 μL reaction. Primers used in this study are described in Table S1.† Quantification was performed by the comparative CT method.

Western blot analysis

Cell lysates were prepared by washing co-cultured cells twice in ice-cold PBS followed by the addition of lysis buffer. Cells were lysed at 4 °C for 30 min and clarified by centrifugation. Lysates were separated by SDS-PAGE, transferred to a PVDF membrane, blocked in 5% non-fat milk in PBS containing 0.5% Tween-20, probed with primary antibodies and detected with horseradish peroxidase-conjugated anti-rabbit or anti-mouse antibodies (Cell Signaling Technology, Beverly, MA).

Ac-LDL-induced foam cell formation of THP-1 cells

THP-1 cells were transformed into foam cells by ac-LDL.¹⁸ THP-1 monocytic cells were seeded in 6-well plates at 106 cells/mL density and treated with 50 ng ml⁻¹ phorbol myristate acetate (PMA) to differentiate for 48 hours into basal macrophages. Basal macrophages were washed once with warm PBS and serum-starved for 24 hours, followed by incubation with acetylated (ac-) LDL (50 μg ml⁻¹) for 48 hours. Then, foam cells were washed once with warm PBS and incubated with 8g (10 μM, 2.5 μM) or BMS309403 (10 μM) for 24 hours. Cells were finally washed and stained with Oil Red-O and quantitated at 500 nm.

Transfection of the differentiated 3T3-L1 cells with siRNA

The siRNA transfection of 3T3-L1 adipocytes was performed with Lipofectamine RNAiMAX (Invitrogen, San Diego, CA) according to the manufacturer’s protocol. Transfection of siRNA was performed using Lipofectamine RNAiMAX (both from Life Technologies).

Animal protocol

The animal protocol was approved by the Animal Care and Use Committee of Sichuan University in China (IACUC number: 20100318). Male C57BL/6J mice at 3 weeks of age were purchased from the Jackson Laboratory (Bar Harbor, ME) and were maintained on a 60% HFD (D12492, Research Diets) until they reached 12 weeks of age. 8g was formulated in 60% HFD at three dose levels (5, 10, and 20 mg kg⁻¹), and was administered to mice for 6 weeks (n = 6 each group). Bodyweight and food intake was monitored every 3 days. Oral glucose tolerance tests (OGTT) and intraperitoneal insulin tolerance tests (ipITT) were performed after the sixth week of drug administration. For GTT and ITT, blood was collected after 16 h (for GTT) or 3 h (for ITT) of fasting, glucose (2 g per kg body weight) or insulin (0.5 U per kg body weight) was administered by oral gavage or by intraperitoneal injection, respectively. Blood was collected from the tail vein at 0, 15, 30, 60, 90, and 120 min post doses for glucose determination. At the end of the study, blood samples in the postprandial state were collected for analyses of metabolic parameters such as triglyceride and free fatty acid. Tissues of interest were freeze-clamped immediately. Epididymis fat mass was weighed using an analytical balance.

ApoE-/- mice (male, 4 weeks old) of the C57BL/6 genetic background were obtained from the Jackson Laboratory (Bar Harbor, ME). Animals were maintained in a pathogen-free barrier facility with a 12 hour light/dark cycle and had free access to food and water. To induce early atherosclerotic lesion development, ApoE-/- mice were fed with Western-type diet (D12079B, Research Diets Inc., New Brunswick, New Jersey, USA) for 12 weeks. Mice were randomly divided into four groups with each containing 10 animals and then treated with 8g (5 mg kg⁻¹ and 10 mg kg⁻¹), Atorvastatin (positive control, 5 mg kg⁻¹) or saline for 4 weeks. Animals were fasted overnight and euthanized using CO₂ and serum, plasma and tissue samples (aorta and fat) were collected and stored at −80 °C until analysis. OGTT and ipITT were performed after the forth week of drug administration.

Biomarkers and lesion analysis

The fasting serum glucose, total cholesterol (TC), and triglycerides (TG) were measured using commercially available reagents (Hitachi 7020 Automatic Analyzer, Japan). The fasting serum insulin, free fatty acid, FABP4 and pro-inflammatory cytokines were detected according to the manufacturer’s protocol (mouse insulin ELISA kit, Millipore Corporation; Free Fatty Acid Assay Kit, Cayman Chemical; mouse FABP4 ELISA Kit, Biovendor Laboratory Medicine Inc.; mouse TNF-α, IL-6, IL-1β ELISA kits, R&D Systems). The homeostasis model assessment (HOMA) of insulin resistance, a simple assessment of insulin sensitivity, was calculated with the following formula: fasting plasma glucose (mmol l⁻¹) × fasting insulin (mIU l⁻¹)/22.5. Hemoglobin A1c (HbA1c) was determined by an A1cNow+ device (Bayer HealthCare) according to the manufacturer’s protocol.

For analyses of atherosclerotic lesions, the heart of ApoE-deficient mice was perfused for 10 min with PBS and for 15 min with 4% formaldehyde. Then the aorta was dissected, opened longitudinally from the heart to the iliac arteries, and stained with Oil Red O. Images were analyzed as described by Kratzer et al.¹⁹

Statistical analysis

Data were reported as means ± standard errors of the mean (SEM). Depending on the data structure, analyses were performed by using the two-tailed Student’s t-test or analysis of variance (ANOVA). P values < 0.05 were considered statistically significant.

Conclusions

In this work, we discovered that 8g was a FABP4 inhibitor and exhibited a high affinity towards FABP4 and ameliorated the symptoms of atherosclerosis in ApoE-/- mice by oral administration. We evaluated this FABP4 inhibitor 8g in an adipocyte and macrophage co-culture system and studied the effect of 8g on the foam cell formation during cholesterol loading, aiming to investigate the effect of the FABP4 inhibitor on the lipid accumulation and the pro-inflammatory cytokine expression in atherosclerosis. Our results may represent a potent therapeutic strategy against severe atherosclerosis.

Acknowledgements

This work was supported by the National Key Programs of China during the 12th Five-Year Plan period (2012ZX09101103-033) and NSFC (81402782).

References

1. L. A. Cowart, Trends Endocrinol. Metab., 2009, 20, 34–42.
2. G. S. Hotamisligil, Nature, 2006, 444, 860–867.
3. D. Pratico and J. M. Dogne, Expert Rev. Cardiovasc. Ther., 2009, 7, 1079–1089.
4. A. V. Hertzel and D. A. Bernlohr, Trends Endocrinol. Metab., 2000, 11, 175–180.
5. A. W. Zimmerman and J. H. Veerkamp, Cell. Mol. Life Sci., 2002, 59, 1096–1116.
6. C. R. Hunt, J. H. Ro, D. E. Dobson, H. Y. Min and B. M. Spiegelman, Proc. Natl. Acad. Sci. U. S. A., 1986, 83, 3786–3790.
7. L. Makowski, J. B. Boord, K. Maeda, V. R. Babaev, K. T. Uysal, M. A. Morgan, R. A. Parker, J. Suttles, S. Fazio, G. S. Hotamisligil and M. F. Linton, Nat. Med., 2001, 7, 699–705.
8. M. Roden, Cell Metab., 2007, 6, 89–91.
9. Y. Fu, N. Luo and M. F. Lopes-Virella, J. Lipid Res., 2000, 41, 2017–2023.
10. M. R. Kazemi, C. M. McDonald, J. K. Shigenaga, C. Grunfeld and K. R. Feingold, Arterioscler., Thromb., Vasc. Biol., 2005, 25, 1220–1224.
11. M. Doi, T. Miyoshi, S. Hirohata, K. Nakamura, S. Usui, K. Takeda, M. Iwamoto, S. Kusachi, K. Kusano and H. Ito, Cardiovasc. Diabetol., 2011, 10, 44.
12. G. S. Hotamisligil, R. S. Johnson, R. J. Distel, R. Ellis, V. E. Papaioannou and B. M. Spiegelman, Science, 1996, 274, 1377–1379.
13. K. T. Uysal, L. Scheja, S. M. Wiesbrock, S. Bonner-Weir and G. S. Hotamisligil, Endocrinology, 2000, 141, 3388–3396.
14. J. B. Boord, Arterioscler., Thromb., Vasc. Biol., 2002, 22, 1686–1691.
15. M. J. Brady, P. M. Kartha, A. A. Aysola and A. R. Saltiel, J. Biol. Chem., 1999, 274, 27497–27504.
16. C. Lu, P. S. Ward, G. S. Kapoor, D. Rohle, S. Turcan, O. Abdel-Wahab, C. R. Edwards, R. Khanin, M. E. Figueroa, A. Melnick, K. E. Wellen, D. M. O’Rourke, S. L. Berger, T. A. Chan, R. L. Levine, I. K. Mellinghoff and C. B. Thompson, Nature, 2012, 483, 474–478.
17. T. Suganami, K. Tanimoto-Koyama, J. Nishida, M. Itoh, X. Yuan, S. Mizuarai, H. Kotani, S. Yamaoka, K. Miyake, S. Aoe, Y. Kamei and Y. Ogawa, Arterioscler., Thromb., Vasc. Biol., 2007, 27, 84–91.
18. J. T. Chai, J. E. Digby, N. Ruparelia, A. Jefferson, A. Handa and R. P. Choudhury, PLoS One, 2013, 8, e62934.
19. A. Kratzer, M. Buchebner, T. Pfeifer, T. M. Becker, G. Uray, M. Miyazaki, S. Miyazaki-Anzai, B. Ebner, P. G. Chandak, R. S. Kadam, E. Calayir, N. Rathke, H. Ahammer, B. Radovic, M. Trauner, G. Hoefler, U. B. Kompella, G. Fauler, M. Levi, S. Levak-Frank, G. M. Kostner and D. Kratky, J. Lipid Res., 2009, 50, 312–326.
20. R. E. Gillilan, S. D. Ayers and N. Noy, J. Mol. Biol., 2007, 372, 1246–1260.
21. D. S. Goodsell, G. M. Morris and A. J. Olson, J. Mol. Recognit., 1996, 9, 1–5.
22. J. M. Ntambi and K. Young-Cheul, J. Nutr., 2000, 130, 3122S–3126S.
23. L. Makowski and G. S. Hotamisligil, J. Nutr., 2004, 134, 2464s–2468s.
24. X. H. Yu, Y. C. Fu, D. W. Zhang, K. Yin and C. K. Tang, Clin. Chim. Acta, 2013, 424, 245–252.
25. P. Libby, Circulation, 2002, 105, 1135–1143.
26. Y. Mizuno, R. F. Jacob and R. P. Mason, J. Atheroscler. Thromb., 2011, 18, 351–358.
27. M. Furuhashi and G. S. Hotamisligil, Nat. Rev. Drug Discovery, 2008, 7, 489–503.
28. M. Furuhashi, G. Tuncman, C. Z. Gorgun, L. Makowski, G. Atsumi, E. Vaillancourt, K. Kono, V. R. Babaev, S. Fazio, M. F. Linton, R. Sulsky, J. A. Robl, R. A. Parker and G. S. Hotamisligil, Nature, 2007, 447, 959–965.
29. H. Tilg and A. R. Moschen, Mol. Med., 2008, 14, 222–231.
30. G. Charriere, B. Cousin, E. Arnaud, M. Andre, F. Bacou, L. Penicaud and L. Casteilla, J. Biol. Chem., 2003, 278, 9850–9855.
31. T. Suganami, J. Nishida and Y. Ogawa, Arterioscler., Thromb., Vasc. Biol., 2005, 25, 2062–2068.
32. S. O. Rahaman, D. J. Lennon, M. Febbraio, E. A. Podrez, S. L. Hazen and R. L. Silverstein, Cell Metab., 2006, 4, 211–221.
33. H. Lan, C. C. Cheng, T. J. Kowalski, L. Pang, L. Shan, C. C. Chuang, J. Jackson, A. Rojas-Triana, L. Bober, L. Liu, J. Voigt, P. Orth, X. Yang, G. W. Shipps Jr and J. A. Hedrick, J. Lipid Res., 2011, 52, 646–656.
34. J. Chen, L. Ma, L. Huang, C. Xie, H. Pei, L. He, S. Fu and L. Chen, Chem. Biol. Drug Des., 2015, 86, 66–79.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra05637g

‡ These authors contributed equally to this work.

---