Advertisement

Curcumin modulation of high fat diet-induced atherosclerosis and steatohepatosis in LDL receptor deficient mice

      Highlights

      • Curcumin reduces atherosclerosis and fatty liver in high fat-fed Ldlr−/− mice.
      • Curcumin reduces atherogenesis by suppressing aP2 and CD36 expression in macrophages.
      • Curcumin also alleviates dyslipidemia and expression of inflammatory cytokines.
      • Low doses of curcumin reduce atherogenesis; high doses increase atherosclerosis.

      Abstract

      Objective

      Consuming curcumin may benefit health by modulating lipid metabolism and suppressing atherogenesis. Fatty acid binding proteins (FABP-4/aP2) and CD36 expression are key factors in lipid accumulation in macrophages and foam cell formation in atherogenesis. Our earlier observations suggest that curcumin's suppression of atherogenesis might be mediated through changes in aP2 and CD36 expression in macrophages. Thus, this study aimed to further elucidate the impact of increasing doses of curcumin on modulation of these molecular mediators on high fat diet-induced atherogenesis, inflammation, and steatohepatosis in Ldlr−/− mice.

      Methods

      Ldlr−/− mice were fed low fat (LF) or high fat (HF) diet supplemented with curcumin (500 HF + LC; 1000 HF + MC; 1500 HF + HC mg/kg diet) for 16 wks. Fecal samples were analyzed for total lipid content. Lipids accumulation in THP-1 cells and expression of aP2, CD36 and lipid accumulation in peritoneal macrophages were measured. Fatty streak lesions and expression of IL-6 and MCP-1 in descending aortas were quantified. Aortic root was stained for fatty and fibrotic deposits and for the expression of aP2 and VCAM-1. Total free fatty acids, insulin, glucose, triglycerides, and cholesterol as well as several inflammatory cytokines were measured in plasma. The liver's total lipids, cholesterol, triglycerides, and HDL content were measured, and the presence of fat droplets, peri-portal fibrosis and glycogen was examined histologically.

      Results

      Curcumin dose-dependently reduced uptake of oxLDL in THP-1 cells. Curcumin also reduced body weight gain and body fat without affecting fat distribution. During early intervention, curcumin decreased fecal fat, but at later stages, it increased fat excretion. Curcumin at medium doses of 500–1000 mg/kg diet was effective at reducing fatty streak formation and suppressing aortic expression of IL-6 in the descending aorta and blood levels of several inflammatory cytokines, but at a higher dose (HF + HC, 1500 mg/kg diet), it had adverse effects on some of these parameters. This U-shape like trend was also present when aortic root sections were examined histologically. However, at a high dose, curcumin suppressed development of steatohepatosis, reduced fibrotic tissue, and preserved glycogen levels in liver.

      Conclusion

      Curcumin through a series of complex mechanisms, alleviated the adverse effects of high fat diet on weight gain, fatty liver development, dyslipidemia, expression of inflammatory cytokines and atherosclerosis in Ldlr−/− mouse model of human atherosclerosis. One of the mechanisms by which low dose curcumin modulates atherogenesis is through suppression of aP2 and CD36 expression in macrophages, which are the key players in atherogenesis. Overall, these effects of curcumin are dose-dependent; specifically, a medium dose of curcumin in HF diet appears to be more effective than a higher dose of curcumin.

      Keywords

      1. Introduction

      Atherosclerosis is the major cause of coronary heart disease (CHD) and the leading cause of morbidity and mortality in Western society. Atherosclerosis is a multi-factorial disease in which dysregulation of lipid metabolism and aberrant inflammatory responses in the arterial walls at predisposed sites plays a central role from initiation to progression and eventually rupture of the atherosclerotic plaque [
      • Ross R.
      Atherosclerosis: an inflammatory disease.
      ]. Curcumin, the major bioactive component of turmeric spice, through its antioxidant and anti-inflammatory properties, has been claimed to retard tumorigenesis and diabetes and to exhibit beneficial effects on the modulation of several factors such as lipids and cholesterol, which are involved in the development of cardiovascular disease. Limited animal studies have been conducted to test curcumin's effect on atherogenesis. Curcumin at a low dose (0.3 mg/mouse/d) for 4 mo attenuated the progression of atherosclerosis by a small but significant amount in apoE−/−/Ldlr−/− double knockout mice fed a high fat diet [
      • Olszanecki R.
      • Jawien J.
      • Gajda M.
      • Mateuszuk L.
      • Gebska A.
      • Korabiowska M.
      • et al.
      Effect of curcumin on atherosclerosis in apoE/LDLR-double knockout mice.
      ] without having a significant effect on cholesterol and triglyceride levels. Supplementing a high fat atherogenic diet of old Ldlr−/− mice with 200 mg curcumin/kg diet for 18 wks showed no visible intimal atherosclerotic lesions [
      • Poolsup N.
      • Suksomboon N.
      • Setwiwattanakul W.
      Efficacy of various antidiabetic agents as add-on treatments to metformin in type 2 diabetes mellitus: systematic review and meta-analysis.
      ]. In one rabbit study, supplementing an atherogenic diet with curcumin had no effect on fatty streak formation [
      • Ramirez-Tortosa M.C.
      • Mesa M.D.
      • Aguilera M.C.
      • Quiles J.L.
      • Baro L.
      • Ramirez-Tortosa C.L.
      • et al.
      Oral administration of a turmeric extract inhibits LDL oxidation and has hypocholesterolemic effects in rabbits with experimental atherosclerosis.
      ], whereas supplementing a high fat diet of rabbits with curcumin in another study for several weeks altered lipid composition of aortas and reduced fatty streak formation significantly in thoracic and abdominal aorta, but not in the aortic arc [
      • Quiles J.L.
      • Mesa M.D.
      • Ramirez-Tortosa C.L.
      • Aguilera C.M.
      • Battino M.
      • Gil A.
      • et al.
      Curcuma longa extract supplementation reduces oxidative stress and attenuates aortic fatty streak development in rabbits.
      ]. These effects of curcumin have been mainly attributed to its antioxidant activity. However, numerous other mechanisms may mediate curcumin's anti-atherogenic effect including the alteration of lipids and cholesterol in liver and circulation as well as the expression of inflammatory cytokines.
      Monocytes/macrophages and other immune cells as well as inflammatory cytokines and growth factors actively participate in the pathogenesis of atherosclerosis. Lipids accumulation in macrophages is a hallmark of foam cell formation and fatty streak development in atherogenesis. The influx of modified LDL in macrophages leading to foam cell formation is mediated by several transport proteins including SR-A and CD36. Recently, an adipocyte Protein 2 (aP2), also named Fatty Acid Binding Protein (FABP)-4, was identified in macrophages [
      • Pelton P.D.
      • Zhou L.
      • Demarest K.T.
      • Burris T.P.
      PPARgamma activation induces the expression of the adipocyte fatty acid binding protein gene in human monocytes.
      ]. aP2 regulates two central molecular pathways in macrophages: coordination of lipids/cholesterol trafficking processes and activation of inflammatory processes. Deficiency of aP2 alters the lipid composition in macrophages, increases CD36 expression, and enhances uptake of modified low density lipoprotein. Contrary to these findings, aP2-deficient macrophages display reduced IκB kinase and NF-κB activity, resulting in suppression of inflammatory function and production of inflammatory cytokines [
      • Makowski L.
      • Brittingham K.C.
      • Reynolds J.M.
      • Suttles J.
      • Hotamisligil G.S.
      The fatty acid-binding protein, aP2, coordinates macrophage cholesterol trafficking and inflammatory activity. Macrophage expression of aP2 impacts peroxisome proliferator-activated receptor gamma and IkappaB kinase activities.
      ]. Knockout of aP2 gene (aP2−/−) in apoE-deficient mice significantly reduced development of atherosclerotic lesions independent of any effects on lipid or glucose metabolism [
      • Makowski L.
      • Boord J.B.
      • Maeda K.
      • Babaev V.R.
      • Uysal K.T.
      • Morgan M.A.
      • et al.
      Lack of macrophage fatty-acid-binding protein aP2 protects mice deficient in apolipoprotein E against atherosclerosis.
      ,
      • Boord J.B.
      • Maeda K.
      • Makowski L.
      • Babaev V.R.
      • Fazio S.
      • Linton M.F.
      • et al.
      Combined adipocyte-macrophage fatty acid-binding protein deficiency improves metabolism, atherosclerosis, and survival in apolipoprotein E-deficient mice.
      ,
      • Nicholson A.C.
      • Hajjar D.P.
      • Zhou X.
      • He W.
      • Gotto Jr., A.M.
      • Han J.
      Anti-adipogenic action of pitavastatin occurs through the coordinate regulation of PPARgamma and Pref-1 expression.
      ]. Therefore, reducing expression or total inhibition of aP2 functions in macrophages through specific drugs, nutrients, and bioactive food components may provide preventive and therapeutic possibilities for atherosclerosis. We have reported that supplementing 3T3-L1 adipocytes with curcumin, the major polyphenol in turmeric spice, dose-dependently suppressed the accumulation of lipids in these cells [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ]. A few natural bioactive components of foods and nutrients have been tested for aP2 expression in mostly 3T3-L1 adipocytes. Since adipocytes and macrophages share several biological and molecular characteristics, information on the effects of bioactive compounds on those lipid-transport genes in adipocytes are also likely to be applicable to macrophages. It has been reported that 3T3-L1 adipocytes treated with an active form of vitamin D, (1,25(OH)2D3) or in combination with genistein or xanthohumol, the principal flavonoid found in the hops plant, reduced the expression of aP2 and lipid accumulation [
      • Nagy L.
      • Tontonoz P.
      • Alvarez J.G.
      • Chen H.
      • Evans R.M.
      Oxidized LDL regulates macrophage gene expression through ligand activation of PPARgamma.
      ]. Therefore, using a high fat diet in mice, we examined increasing doses of dietary curcumin to attenuate atherosclerosis by down-regulation of aP2, CD36, proinflammatory cytokines and vascular cell adhesion molecule and to prevent liver lipotoxicity.

      2. Methods

      2.1 Cell culture study

      Human THP-1 cell line (ATCC, Manassas, VA) was used to determine the curcumin's effect on uptake of oxidized low density lipoprotein (oxLDL) by macrophages. THP-1 cells were seeded on 24-well plates in RPMI-1640 medium and were differentiated by adding 100 nmol phorbol 12-myristate 13-acetate (PMA) over 24 h. Cells were then washed and incubated for another 24 h. The medium was replaced with new medium containing 0, 5, 10, and 20 μmol of curcumin for 24 h. The cells were treated with 5 μg/mL freshly prepared human DiI-oxLDL and incubated for 5 h. Cells were fixed with 4% formaldehyde and stained with Oil red O stain, after which photomicrographs were captured. For oxLDL uptake, after treatments macrophages were scraped with rubber policemen from the plastic dish. After washing with PBS, cells were fixed with 4% formaldehyde and analyzed by FACS using DiI-labeled oxLDL. This experiment was repeated in triplicates. Western blots for aP2 and CD36 were performed after incubation of THP-1 cells with increasing doses of curcumin for 24 h and exposing to 5 μg/mL freshly prepared human DiI-oxLDL for 5 h.

      2.2 Animals and diet

      A total of 120 male 8-wk old Ldlr−/− mice with a C57/BL6 genetic background were used in this study. Mice were bred and produced in-house by mating 6-wk old male and female mice (strain B6.129S7-LDLRtm1Her, Jackson Laboratory, Bar Harbor, ME) at the Comparative Biology Unit at the JM-USDA Human Nutrition Research Center on Aging at Tufts University. The mice were randomly divided into five groups (n = 24/group). One group of mice was fed AIN-93M (Table 1 Suppl.) diet containing 5.2% fat by weight and designated as the low fat negative control (LF-Con) group. The other four groups were fed high fat (HF) Western style diet containing 21% by weight milk fat, 0.2% cholesterol (Table 1 Suppl). One group of these HF-fed mice was designated as positive control (HF). The diets of the three other HF-fed groups were supplemented with curcumin (Sigma, 98% purity) at concentrations of 500 mg (LC), 1000 mg (MC), and 1500 mg (HC)/kg diet. The animals were fed these diets ad-libitum for 16 wks. Food intake was monitored three times during the course of the study, each time for three consecutive days. Feces of each mouse were collected at 30th, 70th and 100th days of the study period for total fat excretion analysis. All animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee at Tufts University (See detail in Suppl.).

      2.3 Tissue collection

      Prior to the sacrificing mice, their total body weight, body fat, and lean mass were measured by magnetic resonance imaging (Echo MRI, Houston, TX). Blood samples were collected in EDTA. Peritoneal macrophages were collected by peritoneal lavage. Peritoneal macrophages were incubated in RPMI medium for 2 h, fixed with formaldehyde, stained for total lipids with Nile red [
      • Greenspan P.
      • Mayer E.P.
      • Fowler S.D.
      Nile red: a selective fluorescent stain for intracellular lipid droplets.
      ], and analyzed by FACS. The total RNA and proteins were extracted from peritoneal macrophages. Expression of mRNA for aP2 and CD36 was measured by qRT-PCR, and expression of aP2 protein was determined by Western blotting.
      The thoracic cavity was opened and aortic root was dissected and separated from aortic arc at the right subclavical branching point. It was then embedded in optimal cutting temperature (OCT) compound and frozen in liquid nitrogen. For quantitative analysis, aortic trees from 6 to 13 mice were opened longitudinally, pinned down on black wax platform, fixed with 10% formaldehyde, stained with Oil red O stain. En face images of aortic tree were captured, and lesions were measured as a percent of the total luminal area using NIH Image-J Analyzer. Five micron serial frozen sections were prepared for histological examination of atheroma lesions at the aortic tricuspid valve level. The 20th frozen section and consecutive 3 sections, distally from aortic arc, were selected for different staining including Oil-Red O for lipid accumulation in foam cells, Gomori trichrome stain for collagen and smooth muscle cells, and immunologically stained with rabbit anti-mouse aP2 primary antibody for aP2 expression and with primary anti-VCAM-1 antibody followed by secondary Cy3-labeled goat-anti-rat IgG antibody for VCAM-1 and with Sytox Green for nucleic acid.
      Whole aorta samples from the subclavical branch point to the iliac artery bifurcation of 6 mice per group were collected, washed, and stored in RNAlater solution (Qiagen) for RNA extraction and RT-PCR analysis of cytokines. Livers were dissected out and weighed after which samples were fixed with formaldehyde for histological examination and frozen samples in liquid nitrogen for analysis of total lipids, cholesterol, triglycerides, high density lipoprotein (HDL) and glycogen. All the visceral, inguinal, and subcutaneous adipose tissues were dissected out and individually weighed to determine the impact of dietary intervention on distribution of fat in different depots.

      2.4 Chemical analysis

      Plasma samples were subjected to β-glucuronidase and sulfatase to deconjugate curcumin [
      • Vareed S.K.
      • Kakarala M.
      • Ruffin M.T.
      • Crowell J.A.
      • Normolle D.P.
      • Djuric Z.
      • et al.
      Pharmacokinetics of curcumin conjugate metabolites in healthy human subjects.
      ]. The liberated curcumin was then extracted and analyzed by high performance liquid chromatography [
      • Ma Z.
      • Haddadi A.
      • Molavi O.
      • Lavasanifar A.
      • Lai R.
      • Samuel J.
      Micelles of poly(ethylene oxide)-b-poly(epsilon-caprolactone) as vehicles for the solubilization, stabilization, and controlled delivery of curcumin.
      ]. The total amount of cholesterol [
      • MacLachlan J.
      • Wotherspoon A.T.
      • Ansell R.O.
      • Brooks C.J.
      Cholesterol oxidase: sources, physical properties and analytical applications.
      ], triglycerides [
      • Lehnus G.
      • Smith L.
      Automated procedure for kinetic measurement of total triglycerides (as glycerol) in serum with the Gilford System 3500.
      ], free fatty acids (FFA) [
      • Duncombe W.G.
      The colorimetric micro-determination of non-esterified fatty acids in plasma.
      ], glucose [
      • Costello J.
      • Scott J.M.
      • Bourke E.
      Enzymic method for determining the specific activity of glucose.
      ] and insulin in plasma were measured [
      • Morgan C.
      • Lazarow A.
      Immunoassay of insulin: two antibody system. Plasma insulin levels in normal, subdiabetic and diabetic rats.
      ]. Plasma levels of inflammatory cytokines were measured by ultra-sensitive multi-spot kit using Meso Scale Discovery's Sector Imager 2400 system (Gaithersburg, MD).
      The frozen samples of liver tissues were homogenized with potassium phosphate buffer. Total lipids in liver were extracted, weighed and the concentrations of total triglycerides, cholesterol [
      • MacLachlan J.
      • Wotherspoon A.T.
      • Ansell R.O.
      • Brooks C.J.
      Cholesterol oxidase: sources, physical properties and analytical applications.
      ] and glycogen were measured. After drying fecal samples at 60 °C, the total fat was extracted by Folch's extraction, dried with nitrogen, and weighed.
      Aorta samples were stored in RNAlater solution (Qiagen) after harvesting. RNA was isolated using RNeasy Mini kit and cDNA was synthesized by TaqMan Reverse Transcription Reagents. Gene expression was measured by quantitative RT-PCR. The reactions were run in triplicate, and the data were analyzed using the ΔΔCt method [
      • Schefe J.H.
      • Lehmann K.E.
      • Buschmann I.R.
      • Unger T.
      • Funke-Kaiser H.
      Quantitative real-time RT-PCR data analysis: current concepts and the novel “gene expression's CT difference” formula.
      ].

      2.5 Statistical analysis

      We used SAS for Windows, version 9.13 (SAS Institute). Statistical analysis was performed using general linear model (GLM) procedure with the adjustment of Tukey–Kramer for multiple comparisons test among the LF-Con, HF and curcumin treated groups. Significance was set at p < 0.05.

      3. Results

      3.1 Cell culture study

      Incubation of differentiated THP-1 cells with oxLDL increased the uptake of lipids as shown by lipid droplets with red color Oil red O stain (Fig. 1a). Incubation of THP-1 cells with increasing concentrations of curcumin for 24 h reduced the lipid droplets' accumulation (Fig. 1a). Similarly, the uptake of DiI-oxLDL was reduced by increasing doses of curcumin as fluorescence of DiI-labeled oxLDL quantitatively measured by FACS (Fig. 1b). Western blot also showed dose-dependent changes in aP2 expression but not CD36 by THP-1 macrophages exposed to oxLDL (Fig. 1c).
      Figure thumbnail gr1
      Fig. 1Curcumin dose-dependently suppressed lipids uptake in oxLDL-treated and PMA-activated THP-1 macrophages. THP-1 macrophages were incubated with increasing doses of curcumin for 24 h. a) Representative photomicrographs of THP-1 macrophages with lipid droplets stained red with Oil-red O stain. b) Quantitative FACS analysis of triplicates uptake of freshly isolated 5 μg/mL DiI-labeled oxLDL by THP-1 cells treated with increasing doses of curcumin (μmol). *p < 0.05, **p < 0.01 compared to control. c) Protein expression of aP2 by THP-1 cells (n = 4, duplicate, #p = 0.03 compared to THP-1 cells treated with 5 μg/mL DiI-oxLDL for 5 h, and *p = 0.09 compared to THP-1 cells treated with 2.5 μmol curcumin and 5 μg/mL DiI-oxLDL for 5 h) and expression of CD36 (n = 3, duplicate) by THP-1 macrophages treated with increasing doses of curcumin (μmol) for 24 h and exposed to 5 μg/mL DiI-oxLDL for 5 h. #p = 0.03 compared to cells exposed to oxLDL. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

      3.2 Animal study

      3.2.1 Body weight and food intake

      As we have reported in our earlier study [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ], supplementing the high fat diet of mice with curcumin at different concentrations (groups HF + LC, HF + MC, and HF + HC) slightly but significantly increased food intake relative to HF only (Fig. 1a, Suppl.), yet it attenuated the body weight gain significantly in HF + MC group (1000 mg/kg diet) (p = 0.04) as early as the 8th wk of supplementation and suppressed weight gain significantly throughout the study (Fig. 1b, Suppl.). In spite of a slightly higher intake of diets by HF + LC and HF + HC, the body weight gains at the end of the 16 wk in these groups were significantly lower than mice fed HF only (Fig. 1a, Suppl.).

      3.2.2 Body fat

      Total body fat measured by MRI and by weighting dissected fats from dead mouse showed a significant increase in body fat mass with HF diet. MRI analysis also showed that only curcumin supplementation at a concentration of 1000 mg/kg of diet (HF + MC) was effective in reducing total body fat (Fig. 1c, Suppl.). All the mice groups fed HF diets with or without curcumin supplementation had significantly reduced lean body mass compared to LF-Con (Fig. 1d Suppl.). The post-mortem dissected adipose tissues revealed that HF Group gained about 6 g more body fat, while the curcumin supplemented groups gained less fat; this differential gain was statistically significant in HF + LC group (Fig. 1e, Suppl.). The total amount of fat collected post-mortem from subcutaneous, inguinal, and visceral areas also showed a significant increase of each depot by HF diet and a significant reduction with curcumin supplementation (Fig. 1f, Suppl.). Specifically, curcumin supplementation in HF + LC group significantly reduced both inguinal and visceral fat and in HF + HC group, it significantly reduced visceral fat. Overall, subcutaneous and inguinal fat comprised the majority of fat in each dietary group, whereas visceral fat constituted a small proportion of adiposity. The differential distribution of fat in different depots was not affected by curcumin treatments.

      3.2.3 Fecal lipid excretion

      We made interesting observations on the fecal excretion of lipids regarding the concentration and duration of curcumin supplementation. Fecal samples collected at day 30 of dietary intervention showed that the HF-fed group excreted two times more lipids (p = 0.0004) than LF-Con group (Fig. 2a). In contrast, curcumin supplemented HF diets dose-dependently suppressed fat excretion in the feces even though the mice consumed a slightly higher amount of food per day (Fig. 1a, Suppl.). However, analysis of fecal samples collected on day 70 showed a weaker effect of curcumin and less difference among the groups on the total lipids excretion (Fig. 2b). Interestingly, analysis of fecal samples collected at day 100 showed an increase of total lipids in feces with curcumin supplemented dose-dependently (Fig. 2c).
      Figure thumbnail gr2
      Fig. 2Concentration of total extracted lipids in the feces. Fecal samples collected at three consecutive days from Ldlr−/− mice at: a) day 30 (n = 10/group); *p < 0.05 HF group compared to HF + curcumin supplemented groups and #p < 0.05 LF-Con compared to HF; b) day 70 (n = 19/group); §p < 0.07 HF tended to be higher than HF + LC and HF + MC, and c) day 100 (n = 18–21/group) **p < 0.0001 HF is lower compared to HF + MC and HF + HC, and #p < 0.05 LF-Con compared to HF. Data are Mean ± SEM.

      3.2.4 Plasma curcumin

      HPLC analysis of plasma showed no free curcumin in circulation (Data not shown). After treating plasma samples with both glucuronidase and sulfatase the HPLC analysis revealed a dose-dependent increase of conjugated forms of curcumin in plasma (Fig. 2, Suppl.).

      3.2.5 Aortic lesion

      The en-face analysis of images from the luminal side of aortic tree after staining with Oil Red O showed that LF-Con-fed mice had developed fatty streak lesions, which covered only 4.8% of luminal surfaces (Fig. 3a and b), but the Ldlr−/− mice fed HF diet developed more fatty streaks lesions, which covered more than 13% of luminal surface. Supplementing the high fat diet with 500 and 1000 mg/kg diet (HF + LC and HF + MC) reduced lesion development compared to HF-fed group, reaching statistical significance (p = 0.024) in HF + MC group. Surprisingly, HF + HC group, which received the highest curcumin dose (1500 mg/kg diet) with a high fat diet, had significantly more lesions in the aortic tree compared to HF + MC group (p = 0.016). The quantitative analysis of frozen sections of aortic roots stained with Oil Red O also showed a similar trend of HF and curcumin effects on atheroma lesions (Fig. 3c, upper panel). The subsequent adjacent sections stained with Gomori trichrome (2nd panel) showed the presence of sparse nuclei due to the presence of foam cells. Collagen and fibrotic tissues are stained blue/green. Staining of the adjacent section with aP2 antibody also revealed an increased aP2 expression, presumably by macrophages and adipocytes, with HF diet. Curcumin supplementation in a dose-dependent manner suppressed aP2 expression (3rd panel). The differences in the expression of VCAM-1 with dietary treatments are depicted with the intensity of red color (4th panel); and the nuclei are stained green. VCAM-1 expression also showed a similar trend as observed with the other stainings for foam cells and aP2. Quantitative analysis of lesions using aortic root sections stained with Oil Red O (Fig. 3d) showed that the HF diet increased lesion development significantly and that supplementation of with 1000 mg/kg (HF + MC group) tended to suppress lesion development in spite of high fat diet intake. Again, mice fed the highest curcumin concentration (HF + HC group) showed no suppressive effect.
      Figure thumbnail gr3
      Fig. 3Curcumin suppresses atherosclerosis in Ldlr−/− mice. a) Representative en face images of luminal side of descending aorta from different dietary groups of Ldlr−/− mice. Fatty lesions are stained red with Red oil O; b) The size of lesion area expressed as a percent of total lumen area. All the HF-fed mice had significantly higher percent of lesions compared to LF-Con (#p = 0.02). HF + MC-fed mice had significantly fewer lesions than HF, and HF + HC groups (*p = 0.02) (The number of animals/group for this analysis are denoted at the base of each bar-graph); c) Representative photomicrographs of aortic root sections of Ldlr−/− mice from different dietary groups are presented (n = 5). Five micron OCT frozen sections of aortic root were prepared and 20th section and consecutive 3 sections, distally from aortic arc, were differentially stained. Foam cells containing lipids in lesions are stained red with Red oil O (upper panels). The immediate adjacent section was stained with Gomori trichrome stain (middle panels), which stains cytoplasm and cardiac muscle in red, fibrotic tissue and collagens in blue. Accumulation of lipids in foam cells reduced cytoplasmic space thus showing less intense red color. Subsequent adjacent section was stained with aP2 antibody as shown in 3rd row panels. The intensity of aP2 expression as detected by rabbit anti-mouse aP2 antibody was substantially diminished in HF + MC groups. Immediate sections also immuno-stained with anti-VCAM-1 antibody (lower panel), shown in red with nuclei in green color. VCAM-1 expression was also up-regulated with HF diet and attenuated with medium level curcumin (HF + MC), but its expression increased with high levels of curcumin (HF + HC). d) The size of the aortic root's lesion area (stained with Red oil O) was measured in pixel units and expressed per pixel size of whole root perimeter. A trend toward reduction in lesions is present with HF + MC treatment (n = 5). #p < 0.05 compared to HF-fed groups. Data are mean ± SEM.

      3.2.6 Steatohepatosis

      Feeding mice a HF diet induced steatohepatosis or fatty liver. The HF diet increased liver weights, but not significantly. However, liver weights were significantly (p < 0.001) increased with HF diet supplemented with all three doses of curcumin (Fig. 4a). A larger liver after curcumin treatment may also reflect a more efficient liver in removal and metabolism of lipids, and thus contribute to the hypolipidemic effect of curcumin. Total lipids extracted from liver samples and expressed as a percentage of whole liver also showed increased lipid content with HF diets, which was significantly reduced with a high dose of curcumin in HF + HC group (Fig. 4b, and Fig. 5). HF diet also reduced the liver's glycogen reserve, and curcumin reduced it slightly further (Fig. 3b, Suppl. and Fig. 5). Since mice were Ldlr−/−, no LDL cholesterol was detected in liver. Whereas, HF diet increased total and HDL cholesterol and triglyceride levels in liver (Table 1) curcumin supplementation at high dose (HF + HC) significantly reduced these lipids in liver.
      Figure thumbnail gr4
      Fig. 4Curcumin effects on liver weight and lipid content. a) Liver weight as affected by dietary treatments (n = 18–21); #p < 0.001compared to curcumin supplemented groups; b) Percent of lipids present in whole liver (n = 5). *p < 0.05 compared to HF and HF + LC. Data are mean ± SEM.
      Figure thumbnail gr5
      Fig. 5Liver histopathology as affected by high fat diet and curcumin. Representative 20X images of liver sections from different dietary treatments (n = 5/group) stained with hematoxylin/eosin (H&E) to display the hepatic tissue architecture and presence of macro- and micro-vesicular steatosis (left panels, black arrows). Liver sections also stained with Gomori's trichrome to show presence of fibrotic tissue (middle panels, stained blue, yellow arrows), and stained with Periodic acid-Schiff (PAS), to display glycogen (right panels, cytoplasm in purple color, white arrows). LF-Con liver with normal hepatocytes (a–c), HF diet alone resulted in accumulation of large numbers of micro- and macro-vesicles representing accumulation of lipid droplets in hepatocytes and the distribution of steatosis, which is centri-lobular (d). The extent of macro-vesicular steatosis was lower with increasing dose of dietary curcumin (g, j, m). High fat diet induced fibrotic tissue with abnormal central veins (e). The extent of fibrotic tissue is declined with increased dose of dietary curcumin (h, k, n). PAS stains glycogen in cytoplasm purple with nucleus in dark blue. As shown in (f), glycogen reserve in high fat fed mice was depleted while curcumin supplementation prevented glycogen loss (i, l, o).
      Table 1Concentration of total cholesterol, HDL cholesterol and triglyceride in liver homogenate.
      mg/dLLF-ConHFHF + LCHF + MCHF + HC
      Total Chol.10.20 ± 1.2857.20 ± 8.36#47.00 ± 11.1936.80 ± 10.3025.60 ± 3.17*
      HDL Chol.5.40 ± 0.5129.20 ± 3.61#18.40 ± 6.6414.60 ± 4.2718.00 ± 1.30*
      Triglyceride93.00 ± 30.75489.60 ± 124.68#571.60 ± 215.63#403.80 ± 174.35253.80 ± 49.83*
      Data are mean ± SEM. (n = 5). Chol.: cholesterol, LF: low fat, HF: high fat, LC: 500 mg curcumin, MC: 1000 mg, HC: 1500 mg curcumin/kg diet.
      #Significantly different from LF-Con (p < 0.05).
      *Significantly different from HF (p < 0.05).
      Examination of photomicrographs from liver sections stained with H&E showed presence of micro- and macro-vesicular steatosis in hepatocytes of HF-fed mice compared to LF-Con group (Fig. 5d). Trichrome staining also showed fibrotic tissues (stained blue) in liver (Fig. 5e), which represents a loss of hepatocytes. The pathologic effect of HF diet was further depicted by glycogen loss (Fig. 5f). Supplementing the HF diet with increasing doses of curcumin reduced the size and prevalence of lipid droplets in hepatocytes and preserved cellular integrity and their glycogen reserves (Fig. 5g–o). Further, curcumin supplementation may have stimulated liver protein synthesis and contributed to liver enlargement (Fig. 4a).

      3.2.7 Aortic cytokines

      qRT-PCR analysis of mRNA extracted from the descending aorta of HF-fed mice relative to LF-Con group showed a significantly higher expression of IL-6 and MCP-1 (Fig. 4a&b, Suppl.), the two prominent inflammatory cytokine markers of the vascular system. Importantly, curcumin supplementation of HF diet attenuated the expression of these inflammatory markers in aortic tissue.

      3.2.8 aP2 and CD36 expression by peritoneal macrophages

      The FACS analysis of peritoneal macrophages stained with Nile red showed a significant increase of lipids in macrophages from mice fed HF diet, which was reduced by curcumin supplementation (Fig. 6a). qRT-PCR analysis of mRNA extracted from peritoneal macrophages for aP2 and CD36 genes are expressed as a percent relative to LF-Con. The HF diet significantly increased mRNA expression of the aP2 gene (Fig. 6b) which was significantly reduced with highest level of curcumin supplementation (p < 0.05). However, Western blot analysis of aP2 protein from peritoneal macrophages showed no clear trend from dietary fat or curcumin supplementation (Fig. 6b). The mRNA expression of CD36 by peritoneal macrophages was also reduced significantly with high levels of curcumin supplementation (HF + HC) (Fig. 6c; p < 0.05).
      Figure thumbnail gr6
      Fig. 6Accumulation of lipids in peritoneal macrophages as affected by the expression of aP2 and CD36: a) Peritoneal macrophages (n = 10) were fixed with formaldehyde, stained with Nile red, and analyzed by FACS for lipid accumulation (#p < 0.05 compared to HF, and *p < 0.05 compared to HF and HF + LC); b) RNA was extracted from freshly isolated peritoneal macrophages (n = 4 in duplicates), and mRNA expression of aP2 was measured by qRT-PCR and expressed as a percent relative to LF-Con (*p < 0. compared to LF-Con and HF + HC groups). Protein was extracted from peritoneal macrophages (n = 4, duplicates) and analyzed for expression of aP2 by Western blotting and expressed as a percent relative to LF-Con. c) mRNA expression of CD36 by macrophages (*p < 0.05 expression of CD36 from HF + HC compared to LF-Con and HF). Data are Mean ± SEM.

      3.2.9 Blood chemistry and inflammatory markers

      Feeding Ldlr−/− mice with HF diet significantly increased plasma levels of total cholesterol (p < 0.0001) (Table 2). This increase was also present and did not decrease significantly in groups of Ldlr−/− mice fed high fat diet supplemented with the three different doses of curcumin. The concentrations of plasma triglycerides from HF-fed groups appeared to be also higher than those of LF-Con. However, only plasma from HF + LC statistically tended (p < 0.07) to be higher than LF-Con. As shown in Table 2, HF diet significantly increased the plasma levels of FFA by more than 2-fold (p < 0.05). Curcumin supplementation of HF diet significantly reduced FFA in plasma in HF + LC and HF + HC groups. Blood insulin levels appeared to be increased with HF diet and curcumin supplementation at the highest dose (1500 mg/kg diet) significantly reduced the plasma insulin levels in HF + HC groups. High fat feeding also showed a trend toward increasing blood glucose levels in Ldlr−/− mice, and supplementing diet with a high concentration of curcumin (HC) tended to reduce the levels back to those observed in LF-Con mice (p = 0.07).
      Table 2Concentration of free fatty acids, insulin, glucose, triglycerides, and cholesterol in plasma of LDLR-/-mice fed LF and HF diet containing different doses of curcumin.
      LF-ConHFHF + LCHF + MCHF + HC
      Chol (mg/dL)474 ± 21816 ± 23#821 ± 7#837 ± 9#797 ± 26#
      TG (mg/dL)202 ± 33312 ± 46397 ± 84§371 ± 109204 ± 32
      Free FA (μM)1028 ± 3933285 ± 750†2021 ± 346*2136 ± 5361656 ± 125*
      Insulin (ng/mL)0.69 ± 0.201.22 ± 0.281.02 ± 0.191.14 ± 0.190.61 ± 0.11*
      Glucose (mg/dL)277 ± 37359 ± 44332 ± 15325 ± 22257 ± 53§
      Data are Mean ± SEM. (n = 5–8) TG: triglyceride, Chol: cholesterol, LF: low fat, HF: high fat, LC: 500 mg curcumin, MC: 1000 mg curcumin, HC: 1500 mg curcumin/kg diet. #Significantly different from LF-Con (p < 0.0001). †Significantly different from LF-Con (p < 0.05). *Significantly different from HF (p < 0.05). §A trend of differences from LF-Con and HF + HC (p = 0.07).
      In HF + LC group the plasma levels of IL-1β tended to be, and TNF-α became significantly lower than HF group (Table 2, Suppl.). KC/GRO (keratinocyte chemoattractant/growth related oncogene), a cytokine from C-X-C family of chemokines [
      • Boisvert W.A.
      • Rose D.M.
      • Johnson K.A.
      • Fuentes M.E.
      • Lira S.A.
      • Curtiss L.K.
      • et al.
      Up-regulated expression of the CXCR2 ligand KC/GRO-alpha in atherosclerotic lesions plays a central role in macrophage accumulation and lesion progression.
      ], showed a trend of increase with HF diet, and a trend of decrease with curcumin supplementation. The HF diet tended to increase IL-10 levels and low dose curcumin supplementation of HF diet (HF + LC) significantly (p < 0.05) increased the production of this cytokine in plasma. Trends toward high levels of IL-6, IL-12p70, and IFN-γ were present with high curcumin diet (HF + HC).

      4. Discussion

      In our earlier study we noted that curcumin suppressed lipid uptake by 3T3-L1 adipocyte [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ], which could have been partly attributed to curcumin's suppression of fatty acid binding protein-4 (FABP-4) or aP2, which has been discovered to be a key player in lipid accumulation and foam cell formation, both of which lead to atherogenesis [
      • Makowski L.
      • Brittingham K.C.
      • Reynolds J.M.
      • Suttles J.
      • Hotamisligil G.S.
      The fatty acid-binding protein, aP2, coordinates macrophage cholesterol trafficking and inflammatory activity. Macrophage expression of aP2 impacts peroxisome proliferator-activated receptor gamma and IkappaB kinase activities.
      ,
      • Chen H.
      • Lin A.S.
      • Li Y.
      • Reiter C.E.
      • Ver M.R.
      • Quon M.J.
      Dehydroepiandrosterone stimulates phosphorylation of FoxO1 in vascular endothelial cells via phosphatidylinositol 3-kinase- and protein kinase A-dependent signaling pathways to regulate ET-1 synthesis and secretion.
      ,
      • Chung Y.W.
      • Kim H.K.
      • Kim I.Y.
      • Yim M.B.
      • Chock P.B.
      Dual function of protein kinase C (PKC) in 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced manganese superoxide dismutase (MnSOD) expression: activation of CREB and FOXO3a by PKC-alpha phosphorylation and by PKC-mediated inactivation of Akt, respectively.
      ]. To test this concept, we conducted a preliminary cell culture study and we found that curcumin does-dependently suppresses oxLDL uptake by human macrophage cell line differentiated by PMA. In agreement with our observation, recently Min et al. reported that curcumin through inhibition of CD36 expression dose-dependently inhibited oxLDL uptake by RAW-264.7 murine macrophages [
      • Min K.J.
      • Um H.J.
      • Cho K.H.
      • Kwon T.K.
      Curcumin inhibits oxLDL-induced CD36 expression and foam cell formation through the inhibition of p38 MAPK phosphorylation.
      ]. In our earlier study with C57/BL6 mice, we also found that 500 mg curcumin/kg diet was effective at reducing body weight gain, adiposity, fatty liver, blood lipids and glucose all of which can contribute to the prevention of atherosclerosis [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ]. To further elucidate these in vitro and in vivo observations, we conducted the current study with Ldlr−/− mice, an animal model of human atherosclerosis, to examine whether in addition to suppressing blood lipids, increasing doses of curcumin through suppressing aP2 prevents the accumulation of lipids in macrophages and suppresses foam cell formation, inflammation and atherogenesis.
      We found that increasing the curcumin dose in the HF diet of these mice dose-dependently increased curcumin levels in plasma, but all were in conjugated forms as glucuronide and sulfate, indicating that supplemental curcumin is absorbed from the gut in a dose-dependent manner; however, no free form of curcumin was detectable in the plasma. While several in vivo biological effects of dietary curcumin have been reported, it is plausible that the transient presence of the free form in tissues or curcumin metabolites contributes to curcumin's biological effects, at least in animal models [
      • de Ruyter J.C.
      • Olthof M.R.
      • Seidell J.C.
      • Katan M.B.
      A trial of sugar-free or sugar-sweetened beverages and body weight in children.
      ].
      While plasma levels of conjugated curcumin increased dose-dependently, increasing curcumin doses from 500 mg to 1500 mg/kg in the diet did not further suppress the high fat diet-induced body weight gain. Interestingly, probably as a consequence of a higher caloric density, food intake of mice in HF and HF + curcumin groups was less than LF-Con. Nevertheless, HF + curcumin groups consumed significantly more food than HF-Con, yet they gained less body weight and body fat. These observations about body weight loss without food restriction are interesting and are in agreement with earlier reports on curcumin's effect on body weight gain [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ,
      • Asai A.
      • Miyazawa T.
      Dietary curcuminoids prevent high-fat diet-induced lipid accumulation in rat liver and epididymal adipose tissue.
      ,
      • Weisberg S.P.
      • Leibel R.
      • Tortoriello D.V.
      Dietary curcumin significantly improves obesity-associated inflammation and diabetes in mouse models of diabesity.
      ,
      • Dam H.
      • Glavind J.
      Factors influencing capillary permeability in the vitamin e deficient chick.
      ]. Supplementing the diet of ob/ob mice with curcumin resulted in weight loss whereas in C57/BL mice, curcumin reduced the weight gain in high fat-fed mice, probably due to curcumin's inhibition of lipid metabolic pathways regulated by AMPK [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ,
      • Zang M.
      • Xu S.
      • Maitland-Toolan K.A.
      • Zuccollo A.
      • Hou X.
      • Jiang B.
      • et al.
      Polyphenols stimulate AMP-activated protein kinase, lower lipids, and inhibit accelerated atherosclerosis in diabetic LDL receptor-deficient mice.
      ] and key transcription proteins involved in adipogenesis such as PPAR-γ or C/EBPα? [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ,
      • Weisberg S.P.
      • Leibel R.
      • Tortoriello D.V.
      Dietary curcumin significantly improves obesity-associated inflammation and diabetes in mouse models of diabesity.
      ,
      • Shao W.
      • Yu Z.
      • Chiang Y.
      • Yang Y.
      • Cha iT.
      • Holtz w
      • et al.
      Curcumin prevents high fat diet induced insulin resistance and obesity via attenuating Lipogenesis in liver and inflammatory pathway in adipocytes.
      ] which cause an increase in basal metabolism, energy expenditure, and weight loss. As expected, the LF diet, due to its low caloric density, caused less weight gain in LF-Con group even though they consumed significantly more food than HF-fed mice.
      Our data on total body fat measurements using MRI and total body fat tissues after dissection from different depots indicate that decreased body weight gain in low and medium curcumin-supplemented groups may be related in part to curcumin's suppression of adipogenesis without having a significant effect on lean body mass. Subcutaneous and inguinal fat comprised the largest percentage of total body fat compared to a minor amount of visceral fat. The HF diet proportionally increased adiposity in all three depots and as seen in our earlier study [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ] we did not see any preferential shift in fat distribution by curcumin supplementation among the different fat depots.
      To determine if curcumin's effect on lower body weight gain is not due to its effect on the absorption of lipids from the gut, we collected mice droppings and measured the amount of lipids per g of dry weight from each of the three consecutive-day-collections. Relative to LF-Con, the lipids excretion in the feces of mice fed HF diet was higher by more than threefold. While food intake in HF-fed mice supplemented with different doses of curcumin was slightly higher, the lipid excretion in feces was lower in a dose-dependent manner. Thus, high food intake, lower excretion of lipids in the feces, yet a lower body weight gain further suggest that curcumin may up-regulate energy expenditure in a short period of dietary intervention. However, this trend did not remain constant over the course of the study. Lipid content of fecal samples from HF-fed mice decreased from 270 mg/g on day 30 to 90 mg/g on day 70 and 50 mg/g feces on day 100 indicating up-regulation of fat absorption from the gut, possibly resulting from increased bile production [
      • Prakash U.N.
      • Srinivasan K.
      Fat digestion and absorption in spice-pretreated rats.
      ]. In contrast, excretion of lipids in the fecal samples from all three HF + curcumin groups increased in a dose-dependent manner. This observation suggests that at the initial stages of curcumin supplementation of HF diets, curcumin might have stimulated and increased liver output of bile acids, which in turn facilitated more lipid absorption when increasing the dose of curcumin. However, long-term exposure to increasing doses of curcumin along with high levels of fat in the diet resulted in the formation of fatty liver and fibrosis, which in turn might have down-regulated bile acid output from the liver resulting in less fat absorption from the gut and more excretion of lipids in the feces. In support of the latter notion, we found that total cholesterol content of the liver at the end of the study was dose-dependently lowered by increasing curcumin in the diet (Table 1). Alternatively, long-term exposure to curcumin might have altered the interaction of gut microbiome with curcumin and/or digestive enzymes together with bile production, all of which are involved in fat absorption. This interesting phenomenon needs to be further investigated.
      We found that several factors including FFA, insulin, glucose, triglycerides and cholesterol, all known to be involved in the development of metabolic syndrome, were elevated by a high fat diet in HF group (Table 2) and that curcumin supplementation dose-dependently counteracted these high fat effects. For example, a high level of curcumin in HF + HC group resulted in low levels of FFA and insulin in plasma by more than 50% compared to HF group, which is similar to the previously reported studies on curcumin supplementation in rats and other mouse strains [
      • Srinivasan K.
      • Sambaiah K.
      • Chandrasekhara N.
      Spices as beneficial hypolipidemic food adjuncts: a review.
      ,
      • Dou X.
      • Fan C.
      • Wo L.
      • Yan J.
      • Qian Y.
      • Wo X.
      Curcumin up-regulates LDL receptor expression via the sterol regulatory element pathway in HepG2 cells.
      ]. Elevated levels of FFA in plasma, known to be risk factors for metabolic syndrome, are associated with insulin resistance, dyslipidemia, atherogenesis, and type II diabetes [
      • Roden M.
      • Price T.B.
      • Perseghin G.
      • Petersen K.F.
      • Rothman D.L.
      • Cline G.W.
      • et al.
      Mechanism of free fatty acid-induced insulin resistance in humans.
      ,
      • Boden G.
      Free fatty acids, insulin resistance, and type 2 diabetes mellitus.
      ]. The liver is the main organ responsible for removing FFA from circulation for catabolic β-oxidation to generate energy or for anabolic synthesis of triglycerides by esterification and storage [
      • Soler-Argilaga C.
      • Infante R.
      • Renaud G.
      • Polonovski J.
      Factors influencing free fatty acid uptake by the isolated perfused rat liver.
      ,
      • Jang E.M.
      • Choi M.S.
      • Jung U.J.
      • Kim M.J.
      • Kim H.J.
      • Jeon S.M.
      • et al.
      Beneficial effects of curcumin on hyperlipidemia and insulin resistance in high-fat-fed hamsters.
      ]. In this study, curcumin supplementation in HF-fed mice significantly attenuated the deleterious effects of HF diet on liver histology (Fig. 5) and lipid profile (Table 1), thus maintaining the liver's function to remove FFA from circulation (Table 2).
      Insulin resistance is a pathological condition that is associated with obesity and high levels of glucose in circulation. In this study, the HF diet increased body weight and adiposity, which is known to induce a chronic proinflammatory condition that can lead to the development of insulin resistance [
      • Dali-Youcef N.
      • Mecili M.
      • Ricci R.
      • Andres E.
      Metabolic inflammation: connecting obesity and insulin resistance.
      ]. As has been reported in C57/BL mice [
      • Shao W.
      • Yu Z.
      • Chiang Y.
      • Yang Y.
      • Cha iT.
      • Holtz w
      • et al.
      Curcumin prevents high fat diet induced insulin resistance and obesity via attenuating Lipogenesis in liver and inflammatory pathway in adipocytes.
      ] and hamsters [
      • Jang E.M.
      • Choi M.S.
      • Jung U.J.
      • Kim M.J.
      • Kim H.J.
      • Jeon S.M.
      • et al.
      Beneficial effects of curcumin on hyperlipidemia and insulin resistance in high-fat-fed hamsters.
      ], we found in our study that a HF diet almost doubled the blood insulin levels (Table 2). Of interest, curcumin supplementation of HF diets reduced insulin levels, which was parallel to the trend toward a reduction of blood glucose, triglyceride, and cholesterol levels. This reflects curcumin's attenuation of liver pathology due to HF diet induction of non-alcoholic steatohepatosis (NASH). These effects of curcumin are evident from histological examination of liver tissues, which clearly show the presence of micro-vesicular fat throughout the liver in HF group which are attenuated with increasing doses of curcumin. It is important to note that hepatic steatosis might be responsible for insulin resistance as shown in human and in rodent models; conversely, the treatment of hepatic steatosis alleviated insulin resistance [
      • Morino K.
      • Petersen K.F.
      • Shulman G.I.
      Molecular mechanisms of insulin resistance in humans and their potential links with mitochondrial dysfunction.
      ,
      • Samuel V.T.
      • Liu Z.X.
      • Qu X.
      • Elder B.D.
      • Bilz S.
      • Befroy D.
      • et al.
      Mechanism of hepatic insulin resistance in non-alcoholic fatty liver disease.
      ].
      Increase in the liver's lipid content due to HF diet consumption increased liver weights in the HF group; however, supplementing HF diet with curcumin not only increased liver weight but also increased its weight relative to body weight, which indicates a probable induction of metabolic enzymes/proteins to conjugate and eliminate curcumin metabolites. In support of this notion, we found that the accumulation of lipids in livers of HF + HC group was significantly lower even though liver weight was higher than LF-Con group. A larger liver may remove and metabolize lipids from plasma more efficiently and thus may be in part responsible for the hypolipidemic effect of curcumin. The mechanisms by which curcumin modulate liver lipid metabolism and reduces hyperlipidemia is largely not known. It has been suggested that curcumin directly or indirectly inhibits HMG-CoA reductase, thus reduces cholesterol synthesis, or upregulates bile acid formation by inducing 7-α-hydroxylase, or up-regulate liver specific fatty acid binding protein [
      • Zingg J.M.
      • Hasan S.T.
      • Meydani M.
      Molecular mechanisms of hypolipidemic effects of curcumin.
      ]. Since curcumin also reduced lipid levels in plasma (Table 2) and in liver (Fig. 4b) of Ldlr−/− mice in this study, combination of additional mechanisms, yet to be determined, may involve in this process.
      Accumulation of glycogen as stored energy could be another factor contributing to liver weight changes. We measured liver glycogen and stained liver sections with periodic acid Schiff for visualization of glycogen under the microscope. The concentration of this energy source in the liver was reduced with HF diet, and with the development of fatty liver, the glycogen reserves of the liver were reduced in all HF-fed mice, with and without curcumin supplementation. Examination of liver sections stained with trichrome revealed the presence of fibrotic tissue predominantly around the central veins. Even though we did not quantify the extent of fibrosis, the development of fibrotic tissue may have contributed to the increase of liver weight and liver dysfunction.
      Several in vitro and in vivo studies have suggested that curcumin has a variety of physiological and pharmacological activities such as antioxidant, anti-inflammatory, and anticancer properties. It has also been suggested that curcumin, through several complex mechanisms, may have hypocholesterolemic properties [
      • Ejaz A.
      • Wu D.
      • Kwan P.
      • Meydani M.
      Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
      ,
      • Asai A.
      • Miyazawa T.
      Dietary curcuminoids prevent high-fat diet-induced lipid accumulation in rat liver and epididymal adipose tissue.
      ,
      • Rao D.S.
      • Sekhara N.C.
      • Satyanarayana M.N.
      • Srinivasan M.
      Effect of curcumin on serum and liver cholesterol levels in the rat.
      ,
      • Kim M.
      • Kim Y.
      Hypocholesterolemic effects of curcumin via up-regulation of cholesterol 7a-hydroxylase in rats fed a high fat diet.
      ,
      • Manjunatha H.
      • Srinivasan K.
      Hypolipidemic and antioxidant effects of dietary curcumin and capsaicin in induced hypercholesterolemic rats.
      ] and may provide protection against the development of atherosclerosis in animal models. Since the liver is directly responsible for lipid and lipoprotein metabolism, overloading this metabolic organ with high levels of fat and cholesterol results in disturbance of lipid homeostasis and lipid stress, which can lead to the pathology of fatty liver disease and other associated maladies such as diabetes and cardiovascular disease [
      • Bechmann L.P.
      • Hannivoort R.A.
      • Gerken G.
      • Hotamisligil G.S.
      • Trauner M.
      • Canbay A.
      The interaction of hepatic lipid and glucose metabolism in liver diseases.
      ,
      • Shapiro H.
      • Bruck R.
      Therapeutic potential of curcumin in non-alcoholic steatohepatitis.
      ]. In this study, we found that supplementing HF diet of LDLR−/Ldlr−/− mice with up to 1000 mg curcumin/kg diet (HF + MC), not only reduced steatohepatosis but significantly attenuated aortic fatty streak development. Paradoxically, the highest curcumin dose used (1500 mg/kg diet or 6 mg curcumin/mouse/d) increased rather than further decreased the development of aortic fatty streaks. Although it was not statistically significant, this negative effect of a high dose of curcumin was also observable when the aortic roots were examined for fatty streak formation by staining with Oil Red O, trichrome, aP2 and VCAM-1 antibodies. Previously, in a study, rabbits were given relatively low doses of 1.66 and 3.2 mg curcumin per/kg bw; while these doses reduced cholesterol levels, they had no effect on the degree of atherosclerosis induced by a high fat diet [
      • Ramirez-Tortosa M.C.
      • Mesa M.D.
      • Aguilera M.C.
      • Quiles J.L.
      • Baro L.
      • Ramirez-Tortosa C.L.
      • et al.
      Oral administration of a turmeric extract inhibits LDL oxidation and has hypocholesterolemic effects in rabbits with experimental atherosclerosis.
      ,
      • Quiles J.L.
      • Mesa M.D.
      • Ramirez-Tortosa C.L.
      • Aguilera C.M.
      • Battino M.
      • Gil A.
      • et al.
      Curcuma longa extract supplementation reduces oxidative stress and attenuates aortic fatty streak development in rabbits.
      ]. Interestingly, supplementation of apoE/LDLR double knockout mice with a very low dose of curcumin or 0.3 mg/d/mouse for 16 wks significantly reduced atherogenesis without reducing total cholesterol or triglyceride levels [
      • Olszanecki R.
      • Jawien J.
      • Gajda M.
      • Mateuszuk L.
      • Gebska A.
      • Korabiowska M.
      • et al.
      Effect of curcumin on atherosclerosis in apoE/LDLR-double knockout mice.
      ]. Supplementing a high fat diet of 34-wk old Ldlr−/− mice with 200 mg curcumin/kg diet for 18 wks reduced fatty streaks in the aortic arc [
      • Poolsup N.
      • Suksomboon N.
      • Setwiwattanakul W.
      Efficacy of various antidiabetic agents as add-on treatments to metformin in type 2 diabetes mellitus: systematic review and meta-analysis.
      ]. Most recently, a study [
      • Zhao J.F.
      • Ching L.C.
      • Huang Y.C.
      • Chen C.Y.
      • Chiang A.N.
      • Kou Y.R.
      • et al.
      Molecular mechanism of curcumin on the suppression of cholesterol accumulation in macrophage foam cells and atherosclerosis.
      ] reported that when the diet of apoE−/− mice contained normal levels of fat (4.5%), a daily oral dose of 20 mg curcumin/kg bw for 4 wks attenuated atherosclerosis progression, reduced systemic inflammation, and decreased the serum levels of cholesterol and triglycerides. Our medium dose of 1000 mg curcumin/kg diet of mouse is approximately equivalent to consumption of 500 mg curcumin/d by a 60 kg human. To calculate this, we assumed a 35 g mouse from HF + MC group consumed on average of 3.6 g food or 3.6 mg curcumin/d, which is about 102.85 mg per kg of mouse bw. By multiplying this value by 3/37 km factors [
      • Diepvens K.
      • Westerterp K.R.
      • Westerterp-Plantenga M.S.
      Obesity and thermogenesis related to the consumption of caffeine, ephedrine, capsaicin, and green tea.
      ] it would be 8.34 mg/kg bw in human or 500 mg curcumin intake by a 60 kg human. It is important to note that oral intake of curcumin by human subjects had mixed results in blood lipid profiles. The intake of 1 g or 4 g of curcumin/d for 6 mo by middle-aged subjects with progressive memory decline did not alter lipid profiles [
      • Baum L.
      • Cheung S.K.
      • Mok V.C.
      • Lam L.C.
      • Leung V.P.
      • Hui E.
      • et al.
      Curcumin effects on blood lipid profile in a 6-month human study.
      ]. However, a low dose (15 mg/d) in patients with acute coronary syndrome showed a trend toward a reduction in total and LDL cholesterol levels [
      • Alwi I.
      • Santoso T.
      • Suyono S.
      • Sutrisna B.
      • Suyatna F.D.
      • Kresno S.B.
      • et al.
      The effect of curcumin on lipid level in patients with acute coronary syndrome.
      ]. These results suggest that curcumin's effectiveness at attenuating atherosclerosis and altering lipid profiles may depend on species, strain of animal model, presence of a high fat diet and the dose of curcumin administered.
      Another interesting finding was the curcumin modulation of a fatty acid binding protein, aP2 and CD36 in macrophages. aP2 is a cytosolic fatty acid chaperone, initially discovered to play important roles in lipid storage and metabolism in adipocytes, and later found to be critical in lipid accumulation in macrophages where it coordinates cholesterol trafficking and foam cell formation [
      • Makowski L.
      • Brittingham K.C.
      • Reynolds J.M.
      • Suttles J.
      • Hotamisligil G.S.
      The fatty acid-binding protein, aP2, coordinates macrophage cholesterol trafficking and inflammatory activity. Macrophage expression of aP2 impacts peroxisome proliferator-activated receptor gamma and IkappaB kinase activities.
      ]. The expression of aP2 mRNA and aP2 protein was increased along with lipid accumulation in both THP-1 cells exposed to oxLDL and peritoneal macrophages of mice fed a HF diet. In contrast, curcumin down-regulated the mRNA expression of aP2 and tended to decrease aP2 protein with low dose of curcumin in the diet and tended to increase with high dose of curcumin in the diet. This observation is also consistent with the concept of U-shape effect of increasing dose of curcumin in pathological outcomes. The curcumin's suppressive effect on aP2 expression in macrophages in this study is one potential mechanism by which curcumin attenuated the development of atherosclerosis. It has been demonstrated that aP2 deficiency protected apoE−/− mice from atherosclerosis [
      • Makowski L.
      • Boord J.B.
      • Maeda K.
      • Babaev V.R.
      • Uysal K.T.
      • Morgan M.A.
      • et al.
      Lack of macrophage fatty-acid-binding protein aP2 protects mice deficient in apolipoprotein E against atherosclerosis.
      ,
      • Boord J.B.
      • Fazio S.
      • Linton M.F.
      Cytoplasmic fatty acid-binding proteins: emerging roles in metabolism and atherosclerosis.
      ] and that a functionally genetic variation at aP2 locus resulted in decreased aP2 expression in humans and reduced risk for coronary heart disease [
      • Tuncman G.
      • Erbay E.
      • Hom X.
      • De Vivo I.
      • Campos H.
      • Rimm E.B.
      • et al.
      A genetic variant at the fatty acid-binding protein aP2 locus reduces the risk for hypertriglyceridemia, type 2 diabetes, and cardiovascular disease.
      ,
      • Ordovas J.M.
      Identification of a functional polymorphism at the adipose fatty acid binding protein gene (FABP4) and demonstration of its association with cardiovascular disease: a path to follow.
      ]. In agreement with these discoveries, we found that HF diet up-regulated but curcumin supplementation down-regulated the expression of this fatty acid binding protein in the aortic root. The expression of aP2 is clearly displayed with immunohistochemical staining of fatty streak lesions in aortic valves where macrophages and foam cells are abundant (Fig. 3). In addition, curcumin supplementation of THP-1 cells exposed to oxLDL or peritoneal macrophages harvested from HF-fed mice supplemented with curcumin had a lower expression of CD36, a receptor involved in lipid and cholesterol transport. Curcumin suppression of CD36 expression has also been reported in HepG2 cells [
      • Peschel D.
      • Koerting R.
      • Nass N.
      Curcumin induces changes in expression of genes involved in cholesterol homeostasis.
      ], but it was induced in other cells such as hepatic stellate cells [
      • Na L.X.
      • Zhang Y.L.
      • Li Y.
      • Liu L.Y.
      • Li R.
      • Kong T.
      • et al.
      Curcumin improves insulin resistance in skeletal muscle of rats.
      ,
      • Zingg J.M.
      • Hasan S.T.
      • Cowan D.
      • Ricciarelli R.
      • Azzi A.
      • Meydani M.
      Regulatory effects of curcumin on lipid accumulation in monocytes/macrophages.
      ,
      • Mimche P.N.
      • Thompson E.
      • Taramelli D.
      • Vivas L.
      Curcumin enhances non-opsonic phagocytosis of Plasmodium falciparum through up-regulation of CD36 surface expression on monocytes/macrophages.
      ]. Although we did not measure CD36 expression in liver, it is plausible that curcumin supplementation might have suppressed lipid accumulation in the liver via suppressing CD36 expression.
      Curcumin, which is claimed to have anti-inflammatory properties, may influence vascular pathology by reducing proinflammatory cytokine expression [
      • Sharma R.A.
      • Gescher A.J.
      • Steward W.P.
      Curcumin: the story so far.
      ,
      • Kim K.-H.
      • Lee E.N.
      • Park J.K.
      • Lee J.-R.
      • Kim J.-H.
      • Choi H.-J.
      • et al.
      Curcumin attenuates TNF-alpha-induced expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1 and proinflammatory cytokines in human endometriotic stromal cells.
      ]. IL-6 and MCP-1 are two inflammatory cytokines involved in the pathogenesis of the vascular system. HF diets have been shown to up-regulate expression of these and other inflammatory cytokines. We found that mRNA expression of these inflammatory cytokines in aorta from HF group was higher than LF-Con. However, due to a limited number of samples, curcumin supplementation of HF diets only showed a trend toward reduction of these inflammatory cytokines in aorta.
      Nevertheless, using Meso Scale techniques, we could measure several cytokines in plasma to evaluate the impact of HF diet and curcumin on inflammation status. Overall, our results indicated that the HF diet up-regulated several inflammatory cytokines, and curcumin at medium dose (HF + MC group) reduced their production (IL-1β, IL-10, IL-6, IL-12p70, TNF-α, IFN-γ) back to those levels of LF-Con. However, the expression of several cytokines (IL-10, IL-6, IL-12p70, TNF-α, IFN-γ) tended to increase in plasma with high dose of curcumin, i.e. HF + HC. It should be noted that IL-10 is an anti-inflammatory cytokine that inhibits proinflammatory cytokines produced by T cells and a potent suppressor of monocyte and macrophage function [
      • Jukema J.W.
      • Cannon C.P.
      • de Craen A.J.
      • Westendorp R.G.
      • Trompet S.
      The controversies of statin therapy: weighing the evidence.
      ]. Therefore up-regulation of this cytokine in our study probably contributed to suppression of proinflammatory cytokines and atherogenesis.
      IFN-γ is known to activate macrophages and production of TNF-α with a diverse cell signaling pathway. Production of these cytokines is influenced by the secretion of IL-12. It is known that increased expression of IL-12 stimulates the secretion of TNF-α and IFN-γ [
      • Rossato M.
      • Curtale G.
      • Tamassia N.
      • Castellucci M.
      • Mori L.
      • Gasperini S.
      • et al.
      IL-10-induced microRNA-187 negatively regulates TNF-alpha, IL-6, and IL-12p40 production in TLR4-stimulated monocytes.
      ]. Indeed, in our study we noticed that HF diet increased the production of IL-12 which was in parallel with the increase of TNF-α and IFN-γ. Conversely, a decrease of IL-12 in mice fed medium levels of curcumin in the diet (1000 mg, HF + MC) reduced the production of these cytokines, which might have alleviated progression of HF-induced atherogenesis in this study. KC/GRO, a chemokine involved in development of inflammation, was up-regulated by HF and supplementing the HF diet with medium and high levels of curcumin suppressed its levels.
      Recently, metabolomic studies have demonstrated a biphasic response of breast cancer cells to curcumin doses [
      • Bayet-Robert M.
      • Morvan D.
      Metabolomics reveals metabolic targets and biphasic responses in breast cancer cells treated by curcumin alone and in association with docetaxel.
      ]. This biphasic characteristic of curcumin doses was also present in several measured parameters in our study. For example, the plasma levels of cytokines such as IL-6, TNFα, INF-γ and IL-12 (Table 2 suppl.) showed a reduced levels of these cytokines with medium dose of curcumin (1000 mg curcumin/kg diet), whereas the level of these cytokines increased rather than decreasing with higher dose of curcumin (1500 mg/kg, HF + HC) in the diet. Likewise, the extent of atheroma lesions in aorta and aortic roots (Fig. 3) was less with 1000 mg curcumin/kg diet (HF + MC) compared with when diet contained more curcumin (1500 mg/kg, HF + HC).

      5. Conclusion

      In conclusion, we have shown that higher doses of curcumin in the diet increase the conjugated forms of curcumin in plasma of Ldlr−/− mice, yet through a series of complex mechanisms, “curcumin or metabolites” exerted many biological and pharmacological effects on several organ systems including the liver, adipose tissue, and the cardiovascular system. Curcumin supplementation abrogated the high fat diet's adverse effect on fatty liver development, dyslipidemia, and atherosclerosis in this mouse model of human atherosclerosis. Our in vivo and in vitro data show that one of the mechanisms by which curcumin modulates atherogenesis is through suppression of aP2 and CD36 expression in macrophages, the main culprits in developing foam cells and fatty streak formation. In addition, curcumin's inhibitory effect on the development of high fat diet-induced fatty liver may play a central role in the attenuation of dyslipidemia, hyperglycemia and hyperinsulinemia induced by high fat diet in this mouse model. Further, as demonstrated in our study, curcumin's anti-inflammatory effects most likely play a significant role in the attenuation of atherosclerosis and steatohepatosis. The lower weight gain we observed during the study's early stages of curcumin supplementation can be attributed to curcumin's increase of basal metabolism as we and others have reported; however, at the later stages, curcumin supplementation may have hindered fat absorption from the gut resulting in lower weight gain and lower liver pathology. We also found that curcumin causes a biphasic dose effects, i.e. at the low and medium doses used in this study were more effective than a higher dose at attenuating atherosclerosis and inflammatory cytokines. This presents new challenges to determine optimal doses of curcumin with the highest beneficial effects on liver function, lipid metabolism, and cardiovascular health, particularly when these findings need to be translated for human use.

      Acknowledgments

      This material is based upon work supported by grant #10248826/2009-02916 National Institute of Food and Agriculture (NIFA)/U.S. Department of Agriculture and a grant from the USDA under agreement #58-1950-0-014. We acknowledge that a portion of data presented in Fig. 6 is adopted from our own publication as listed reference #57 in this manuscript. The authors would like to thank Professor Angelo Azzi for his valuable advice and discussions and Stephanie Marco for her assistance in the preparation of the manuscript. Any opinions, findings, conclusion, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture.

      References

        • Ross R.
        Atherosclerosis: an inflammatory disease.
        N Engl J Med. 1999; 340: 115-126
        • Olszanecki R.
        • Jawien J.
        • Gajda M.
        • Mateuszuk L.
        • Gebska A.
        • Korabiowska M.
        • et al.
        Effect of curcumin on atherosclerosis in apoE/LDLR-double knockout mice.
        J Physiol Pharmacol. 2005; 56: 627-635
        • Poolsup N.
        • Suksomboon N.
        • Setwiwattanakul W.
        Efficacy of various antidiabetic agents as add-on treatments to metformin in type 2 diabetes mellitus: systematic review and meta-analysis.
        ISRN Endocrinol. 2012; 2012: 798146
        • Ramirez-Tortosa M.C.
        • Mesa M.D.
        • Aguilera M.C.
        • Quiles J.L.
        • Baro L.
        • Ramirez-Tortosa C.L.
        • et al.
        Oral administration of a turmeric extract inhibits LDL oxidation and has hypocholesterolemic effects in rabbits with experimental atherosclerosis.
        Atherosclerosis. 1999; 147: 371-378
        • Quiles J.L.
        • Mesa M.D.
        • Ramirez-Tortosa C.L.
        • Aguilera C.M.
        • Battino M.
        • Gil A.
        • et al.
        Curcuma longa extract supplementation reduces oxidative stress and attenuates aortic fatty streak development in rabbits.
        Arterioscler Thromb Vasc Biol. 2002; 22: 1225-1231
        • Pelton P.D.
        • Zhou L.
        • Demarest K.T.
        • Burris T.P.
        PPARgamma activation induces the expression of the adipocyte fatty acid binding protein gene in human monocytes.
        Biochem Biophys Res Commun. 1999; 261: 456-458
        • Makowski L.
        • Brittingham K.C.
        • Reynolds J.M.
        • Suttles J.
        • Hotamisligil G.S.
        The fatty acid-binding protein, aP2, coordinates macrophage cholesterol trafficking and inflammatory activity. Macrophage expression of aP2 impacts peroxisome proliferator-activated receptor gamma and IkappaB kinase activities.
        J Biol Chem. 2005; 280: 12888-12895
        • Makowski L.
        • Boord J.B.
        • Maeda K.
        • Babaev V.R.
        • Uysal K.T.
        • Morgan M.A.
        • et al.
        Lack of macrophage fatty-acid-binding protein aP2 protects mice deficient in apolipoprotein E against atherosclerosis.
        Nat Med. 2001; 7: 699-705
        • Boord J.B.
        • Maeda K.
        • Makowski L.
        • Babaev V.R.
        • Fazio S.
        • Linton M.F.
        • et al.
        Combined adipocyte-macrophage fatty acid-binding protein deficiency improves metabolism, atherosclerosis, and survival in apolipoprotein E-deficient mice.
        Circulation. 2004; 110: 1492-1498
        • Nicholson A.C.
        • Hajjar D.P.
        • Zhou X.
        • He W.
        • Gotto Jr., A.M.
        • Han J.
        Anti-adipogenic action of pitavastatin occurs through the coordinate regulation of PPARgamma and Pref-1 expression.
        Br J Pharmacol. 2007; 151: 807-815
        • Ejaz A.
        • Wu D.
        • Kwan P.
        • Meydani M.
        Curcumin inhibits adipogenesis in 3T3-L1 adipocytes and angiogenesis and obesity in C57/BL mice.
        J Nutr. 2009; 139: 919-925
        • Nagy L.
        • Tontonoz P.
        • Alvarez J.G.
        • Chen H.
        • Evans R.M.
        Oxidized LDL regulates macrophage gene expression through ligand activation of PPARgamma.
        Cell. 1998; 93: 229-240
        • Greenspan P.
        • Mayer E.P.
        • Fowler S.D.
        Nile red: a selective fluorescent stain for intracellular lipid droplets.
        J Cell Biol. 1985; 100: 965-973
        • Vareed S.K.
        • Kakarala M.
        • Ruffin M.T.
        • Crowell J.A.
        • Normolle D.P.
        • Djuric Z.
        • et al.
        Pharmacokinetics of curcumin conjugate metabolites in healthy human subjects.
        Cancer Epidemiol Biomarkers Prev. 2008; 17: 1411-1417
        • Ma Z.
        • Haddadi A.
        • Molavi O.
        • Lavasanifar A.
        • Lai R.
        • Samuel J.
        Micelles of poly(ethylene oxide)-b-poly(epsilon-caprolactone) as vehicles for the solubilization, stabilization, and controlled delivery of curcumin.
        J Biomed Mater Res A. 2008; 86: 300-310
        • MacLachlan J.
        • Wotherspoon A.T.
        • Ansell R.O.
        • Brooks C.J.
        Cholesterol oxidase: sources, physical properties and analytical applications.
        J Steroid Biochem Mol Biol. 2000; 72: 169-195
        • Lehnus G.
        • Smith L.
        Automated procedure for kinetic measurement of total triglycerides (as glycerol) in serum with the Gilford System 3500.
        Clin Chem. 1978; 24: 27-31
        • Duncombe W.G.
        The colorimetric micro-determination of non-esterified fatty acids in plasma.
        Clin Chim Acta. 1964; 9: 122-125
        • Costello J.
        • Scott J.M.
        • Bourke E.
        Enzymic method for determining the specific activity of glucose.
        Anal Biochem. 1972; 46: 654-659
        • Morgan C.
        • Lazarow A.
        Immunoassay of insulin: two antibody system. Plasma insulin levels in normal, subdiabetic and diabetic rats.
        Diabetes. 1973; 31: 187
        • Schefe J.H.
        • Lehmann K.E.
        • Buschmann I.R.
        • Unger T.
        • Funke-Kaiser H.
        Quantitative real-time RT-PCR data analysis: current concepts and the novel “gene expression's CT difference” formula.
        J Mol Med (Berlin, Germany). 2006; 84: 901-910
        • Boisvert W.A.
        • Rose D.M.
        • Johnson K.A.
        • Fuentes M.E.
        • Lira S.A.
        • Curtiss L.K.
        • et al.
        Up-regulated expression of the CXCR2 ligand KC/GRO-alpha in atherosclerotic lesions plays a central role in macrophage accumulation and lesion progression.
        Am J Pathol. 2006; 168: 1385-1395
        • Chen H.
        • Lin A.S.
        • Li Y.
        • Reiter C.E.
        • Ver M.R.
        • Quon M.J.
        Dehydroepiandrosterone stimulates phosphorylation of FoxO1 in vascular endothelial cells via phosphatidylinositol 3-kinase- and protein kinase A-dependent signaling pathways to regulate ET-1 synthesis and secretion.
        J Biol Chem. 2008; 283: 29228-29238
        • Chung Y.W.
        • Kim H.K.
        • Kim I.Y.
        • Yim M.B.
        • Chock P.B.
        Dual function of protein kinase C (PKC) in 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced manganese superoxide dismutase (MnSOD) expression: activation of CREB and FOXO3a by PKC-alpha phosphorylation and by PKC-mediated inactivation of Akt, respectively.
        J Biol Chem. 2011; 286: 29681-29690
        • Min K.J.
        • Um H.J.
        • Cho K.H.
        • Kwon T.K.
        Curcumin inhibits oxLDL-induced CD36 expression and foam cell formation through the inhibition of p38 MAPK phosphorylation.
        Food Chem Toxicol Int J Publish Brit Indust Biol Res Assoc. 2013; 58C: 77-85
        • de Ruyter J.C.
        • Olthof M.R.
        • Seidell J.C.
        • Katan M.B.
        A trial of sugar-free or sugar-sweetened beverages and body weight in children.
        N Engl J Med. 2012; 367: 1397-1406
        • Asai A.
        • Miyazawa T.
        Dietary curcuminoids prevent high-fat diet-induced lipid accumulation in rat liver and epididymal adipose tissue.
        J Nutr. 2001; 131: 2932-2935
        • Weisberg S.P.
        • Leibel R.
        • Tortoriello D.V.
        Dietary curcumin significantly improves obesity-associated inflammation and diabetes in mouse models of diabesity.
        Endocrinology. 2008; 149: 3549-3558
        • Dam H.
        • Glavind J.
        Factors influencing capillary permeability in the vitamin e deficient chick.
        Science. 1942; 96: 235-236
        • Zang M.
        • Xu S.
        • Maitland-Toolan K.A.
        • Zuccollo A.
        • Hou X.
        • Jiang B.
        • et al.
        Polyphenols stimulate AMP-activated protein kinase, lower lipids, and inhibit accelerated atherosclerosis in diabetic LDL receptor-deficient mice.
        Diabetes. 2006; 55: 2180-2191
        • Shao W.
        • Yu Z.
        • Chiang Y.
        • Yang Y.
        • Cha iT.
        • Holtz w
        • et al.
        Curcumin prevents high fat diet induced insulin resistance and obesity via attenuating Lipogenesis in liver and inflammatory pathway in adipocytes.
        PLoS One. 2012; 7
        • Prakash U.N.
        • Srinivasan K.
        Fat digestion and absorption in spice-pretreated rats.
        J Sci Food Agric. 2012; 92: 503-510
        • Srinivasan K.
        • Sambaiah K.
        • Chandrasekhara N.
        Spices as beneficial hypolipidemic food adjuncts: a review.
        Food Rev Int. 2004; 20: 187-220
        • Dou X.
        • Fan C.
        • Wo L.
        • Yan J.
        • Qian Y.
        • Wo X.
        Curcumin up-regulates LDL receptor expression via the sterol regulatory element pathway in HepG2 cells.
        Planta Medica. 2008; 74: 1374-1379
        • Roden M.
        • Price T.B.
        • Perseghin G.
        • Petersen K.F.
        • Rothman D.L.
        • Cline G.W.
        • et al.
        Mechanism of free fatty acid-induced insulin resistance in humans.
        J Clin Invest. 1996; 97: 2859-2865
        • Boden G.
        Free fatty acids, insulin resistance, and type 2 diabetes mellitus.
        Proc Assoc Am Physicians. 1999; 111: 241-248
        • Soler-Argilaga C.
        • Infante R.
        • Renaud G.
        • Polonovski J.
        Factors influencing free fatty acid uptake by the isolated perfused rat liver.
        Biochimie. 1974; 56: 757-761
        • Jang E.M.
        • Choi M.S.
        • Jung U.J.
        • Kim M.J.
        • Kim H.J.
        • Jeon S.M.
        • et al.
        Beneficial effects of curcumin on hyperlipidemia and insulin resistance in high-fat-fed hamsters.
        Metabolism. 2008; 57: 1576-1583
        • Dali-Youcef N.
        • Mecili M.
        • Ricci R.
        • Andres E.
        Metabolic inflammation: connecting obesity and insulin resistance.
        Ann Med. 2013; 45: 242-253
        • Morino K.
        • Petersen K.F.
        • Shulman G.I.
        Molecular mechanisms of insulin resistance in humans and their potential links with mitochondrial dysfunction.
        Diabetes. 2006; 55: S9-S15
        • Samuel V.T.
        • Liu Z.X.
        • Qu X.
        • Elder B.D.
        • Bilz S.
        • Befroy D.
        • et al.
        Mechanism of hepatic insulin resistance in non-alcoholic fatty liver disease.
        J Biol Chem. 2004; 279: 32345-32353
        • Zingg J.M.
        • Hasan S.T.
        • Meydani M.
        Molecular mechanisms of hypolipidemic effects of curcumin.
        Biofactors. 2013; 39: 101-121
        • Rao D.S.
        • Sekhara N.C.
        • Satyanarayana M.N.
        • Srinivasan M.
        Effect of curcumin on serum and liver cholesterol levels in the rat.
        J Nutr. 1970; 100: 1307-1315
        • Kim M.
        • Kim Y.
        Hypocholesterolemic effects of curcumin via up-regulation of cholesterol 7a-hydroxylase in rats fed a high fat diet.
        Nutr Res Pract. 2010; 4: 191-195
        • Manjunatha H.
        • Srinivasan K.
        Hypolipidemic and antioxidant effects of dietary curcumin and capsaicin in induced hypercholesterolemic rats.
        Lipids. 2007; 42: 1133-1142
        • Bechmann L.P.
        • Hannivoort R.A.
        • Gerken G.
        • Hotamisligil G.S.
        • Trauner M.
        • Canbay A.
        The interaction of hepatic lipid and glucose metabolism in liver diseases.
        J Hepatol. 2012; 56: 952-964
        • Shapiro H.
        • Bruck R.
        Therapeutic potential of curcumin in non-alcoholic steatohepatitis.
        Nutr Res Rev. 2005; 18: 212-221
        • Zhao J.F.
        • Ching L.C.
        • Huang Y.C.
        • Chen C.Y.
        • Chiang A.N.
        • Kou Y.R.
        • et al.
        Molecular mechanism of curcumin on the suppression of cholesterol accumulation in macrophage foam cells and atherosclerosis.
        Mol Nutr Food Res. 2012; 56: 691-701
        • Diepvens K.
        • Westerterp K.R.
        • Westerterp-Plantenga M.S.
        Obesity and thermogenesis related to the consumption of caffeine, ephedrine, capsaicin, and green tea.
        Am J Physiol Regul Integr Comp Physiol. 2007; 292: R77-R85
        • Baum L.
        • Cheung S.K.
        • Mok V.C.
        • Lam L.C.
        • Leung V.P.
        • Hui E.
        • et al.
        Curcumin effects on blood lipid profile in a 6-month human study.
        Pharmacol Res. 2007; 56: 509-514
        • Alwi I.
        • Santoso T.
        • Suyono S.
        • Sutrisna B.
        • Suyatna F.D.
        • Kresno S.B.
        • et al.
        The effect of curcumin on lipid level in patients with acute coronary syndrome.
        Acta Medica Indones. 2008; 40: 201-210
        • Boord J.B.
        • Fazio S.
        • Linton M.F.
        Cytoplasmic fatty acid-binding proteins: emerging roles in metabolism and atherosclerosis.
        Curr Opin Lipidol. 2002; 13: 141-147
        • Tuncman G.
        • Erbay E.
        • Hom X.
        • De Vivo I.
        • Campos H.
        • Rimm E.B.
        • et al.
        A genetic variant at the fatty acid-binding protein aP2 locus reduces the risk for hypertriglyceridemia, type 2 diabetes, and cardiovascular disease.
        Proc Natl Acad Sci U S A. 2006; 103: 6970-6975
        • Ordovas J.M.
        Identification of a functional polymorphism at the adipose fatty acid binding protein gene (FABP4) and demonstration of its association with cardiovascular disease: a path to follow.
        Nutr Rev. 2007; 65: 130-134
        • Peschel D.
        • Koerting R.
        • Nass N.
        Curcumin induces changes in expression of genes involved in cholesterol homeostasis.
        J Nutr Biochem. 2007; 18: 113-119
        • Na L.X.
        • Zhang Y.L.
        • Li Y.
        • Liu L.Y.
        • Li R.
        • Kong T.
        • et al.
        Curcumin improves insulin resistance in skeletal muscle of rats.
        Nutr, Metab, Cardiovasc Dis NMCD. 2011; 21: 526-533
        • Zingg J.M.
        • Hasan S.T.
        • Cowan D.
        • Ricciarelli R.
        • Azzi A.
        • Meydani M.
        Regulatory effects of curcumin on lipid accumulation in monocytes/macrophages.
        J Cell Biochem. 2012; 113: 833-840
        • Mimche P.N.
        • Thompson E.
        • Taramelli D.
        • Vivas L.
        Curcumin enhances non-opsonic phagocytosis of Plasmodium falciparum through up-regulation of CD36 surface expression on monocytes/macrophages.
        J Antimicrob Chemother. 2012; 67: 1895-1904
        • Sharma R.A.
        • Gescher A.J.
        • Steward W.P.
        Curcumin: the story so far.
        Eur J Cancer. 2005; 41: 1955-1968
        • Kim K.-H.
        • Lee E.N.
        • Park J.K.
        • Lee J.-R.
        • Kim J.-H.
        • Choi H.-J.
        • et al.
        Curcumin attenuates TNF-alpha-induced expression of intercellular adhesion molecule-1, vascular cell adhesion molecule-1 and proinflammatory cytokines in human endometriotic stromal cells.
        Phytother Res PTR. 2012; 26: 1037-1047
        • Jukema J.W.
        • Cannon C.P.
        • de Craen A.J.
        • Westendorp R.G.
        • Trompet S.
        The controversies of statin therapy: weighing the evidence.
        J Am Coll Cardiol. 2012; 60: 875-881
        • Rossato M.
        • Curtale G.
        • Tamassia N.
        • Castellucci M.
        • Mori L.
        • Gasperini S.
        • et al.
        IL-10-induced microRNA-187 negatively regulates TNF-alpha, IL-6, and IL-12p40 production in TLR4-stimulated monocytes.
        Proc Natl Acad Sci U S A. 2012; 109: E3101-E3110
        • Bayet-Robert M.
        • Morvan D.
        Metabolomics reveals metabolic targets and biphasic responses in breast cancer cells treated by curcumin alone and in association with docetaxel.
        PLoS One. 2013; 8: e57971