NF-κB-SREBPs Pathway Mediates Cholesterol Efflux And Inflammatory Cytokines Production In Macrophages

Atherosclerosis (As) is a chronic inflammatory pathological processes.Inflammation plays an important role in the initiation, progress, and complications ofatherosclerosis. Nuclear factor-kappa B (NF-κB) is one of the key transcriptionfactors in inflammatory cytokine production. Activated NF-κB has demonstrated inhuman atherosclerotic lesions. Specificity silence or inhibition of NF-κB pathwaycould reduce atherosclerotic plaque area and its complications. The mechanisms thatNF-κB promotes atherosclerosis were mainly mediated by inflammatory factors, butrecent studies found that NF-κB could promote macrophage lipid accumulation,which provide us a new research direction to recognize the proatherogenicmechanisms of NF-κB.Sterol regulatory element binding proteins (SREBPs) is a key regulator of the genesthat control cholesterol biosynthesis and uptake. There are three forms of SREBP inmammals: SREBP-1a,-1c and-2. Although they undergo similar proteolyticactivation and share some target genes, SREBP-1a and-1c mainly stimulate fatty acidsynthesis, whereas SREBP-2acts primarily on the cholesterol biosynthetic genes andLDL receptor (LDLR) gene. The intron regions of SREBPs contain a newlydiscovered class of microRNAs (miRNAs), miR-33s. MiRNAs is a kind of smallsingle-stranded RNA molecules that have post-transcriptional regulatory activity.Studies have shown that miRNAs involved in the regulation of lipid metabolism,expression of inflammatory cytokines and formation of As. The main target genes ofmiR-33s is ATP-binding cassette transporter A1(ABCA1). As ABCA1is the mainmembrane transporter that mediates the outflow of cellular lipids, inhibiting itsexpression will results in intracellular accumulation of a large number of lipids and ultimately the formation of foam cells. In vivo studies demonstrated that antagonismof miR-33in mice promotes ABCA1expression and reverse cholesterol transport，butpromotes the regression of atherosclerosis. Thus, miR-33s has become an importanttarget for prevention and treatment of As diseases. However, the in vivo regulation ofmiRNAs is still in the initial stage, the mechanisms that control the expression ofmiR-33s in the body is far from clear.There is a growing body of evidence pointing to the closely relationship betweenthe innate immune response and a variety of diseases including As. NOD-likereceptors (NLRs) are pattern recognition receptors that feeling microbial andnon-microbial signal in cells. NLRs can form large protein complexes within the cells,referred to as the " inflammasome", including the NLRs, caspase1and the ASC. Themain role of inflammasome is to form an bracket to activate caspase1, which couldpromote pro-IL-1β and pro-IL-18transform into mature IL-1β and IL-18, respectively.IL-1β and IL-18correlate closely to the development of As, the absence of IL-1β orIL-18significantly restrict the development of the As in mice. However, themechanisms that regulate the inflammasome in vivo are still unclear.Through bioinformatics analysis, we found the promoter region of SREBPs, thehost gene of miR-33s, have the NF-κB response element (κBREs), suggesting thatNF-κB may be direct binding to SREBPs promoter and regulating the expression ofSREBPs/miR-33s. Therefore, in this study, we firstly observed the effects of NF-κBactivation on macrophages cholesterol efflux and inflammatory factors production inthe first part. Then, in the second part we explored the effect and regulatorymechanisms of NF-κB on SREBPs. We also investigated the the roles of miR-33s inNF-κB repressing cellular cholesterol efflux and the roles of SREBPs in NF-κBpromoting inflammatory cytokine release and inflammasome expression. Next, in thethird part we observed the effects of NF-κB-SREBPs pathway in As and inflammationreaction in vivo. Finally, we observed the effects of betulinic acid (BA), which hadproved to have significant anti-inflammatory effects, on NF-κB-SREBPs pathways invitro and in vivo, and explored its possible mechanisms in the fourth part. Our studiesprovide new insights for revealing the potential roles and mechanisms of NF-κB-SREBPs pathway in the development of As.Part I: Effect of NF-κB Activation on Cholesterol Efflux andInflammatory Cytokines Production in MacrophagesAims: To investigate the effect of NF-κB activation on macrophages cholesterolefflux and the expression of related membrane transporter ABCA1, and also observethe effect of NF-κB on inflammatory factors production.Methods: Human THP-1monocytes were treated with phorbol12-myristate13-acetate (PMA)(160nmol/liter) for24h to differentiate into macrophages. THP-1and RAW264.7macrophages were then replaced to the serum-free medium containingoxLDL (100μg/ml) to become fully differentiated macrophage foam cells beforetheir use in experiments. Both cells were divided into3groups: control group, LPSgroup (10ng/ml,24h) and LPS+PDTC group (LPS10ng/ml, PDTC50μM,24h), respectively. The cholesterol efflux was assessed by liquid scintillation counting.The protein and mRNA expression of ABCA1were examined by westernimmunoblotting assays and real-time quantitative PCR, respectively. Theconcentrations of tumor necrosis factor-α (TNF-α), IL-6and interleukin-1β (IL-1β) incell culture supernatants were measured by enzyme-linked immunosorbent assay(ELISA).Results: Apolipoprotein A1(apoA-I)-mediated cholesterol efflux wassignificantly reduced after treatment with LPS in THP-1and RAW264.7macrophage-derived foam cells. Accordingly, NF-κB activation clearly decreased thelevels of ABCA1mRNA and protein. Application of the NF-κB specific inhibitorPDTC significantly suppressed the LPS-induced down-regulation of cholesterol effluxand the expression of ABCA1. In addition, PDTC obviously inhibited the secretion ofinflammatory cytokines TNF-α, IL-1β and IL-6induced by LPS.Conclusion:①The activation of NF-κB inhibits cholesterol efflux and theexpression of ABCA1in macrophages.②NF-κB activation promotes the secretion of inflammatory cytokines, such as TNFα, IL-6and IL-1β etc.Part II: Molecular Mechanism of SREBPs-Mediated NF-κBRegulating Cholesterol Efflux and Secretion of InflammatoryCytokinesAims: To investigate the possible molecular mechanisms of NF-κB activation onmacrophages cholesterol efflux and secretion of inflammatory cytokinesMethods: Human THP-1monocytes were preincubated with PMA and ox-LDLto form foam cells. Cells were treated with LPS (10ng/ml) for3h and followed bytreatment of act D (5μg/ml) to stop the transcription. ABCA1mRNA expression wasexamined by real-time quantitative PCR to observe the changes of ABCA1mRNAstability. Cells was treated by LPS and observed the concentration-dependent ortime-dependent effect of NF-κB activation on miR-33s and their host genes SREBPs.Observed the expression changes of miR-33s/SREBPs after inhibition of NF-κBactivation by PDTC treatment. Co-transfection of SREBPs promoter reporterplasmid with expression vector p50(pRSV-NF-κB-1) and p65(pRSV-RelA) intoHEK-293T cells, cells were lysed and luciferase activities were measured by using thedual-luciferase reporter assay. Chromatin immunoprecipitation (ChIP) assays wasperformed to detect if NF-κB have a direct binding to the promoter regions of theSREBPs genes. Human THP-1cells were transfected with40nM miRIDIAN miRNAmimics (miR-33a/b) or with60nM miRIDIAN miRNA inhibitors (anti-miR-33a/b) toassess the effects of gain and loss of function of miR-33on ABCA1expression andcholesterol efflux. Treatment of macrophages with SREBPs siRNA to study the roleof SREBPs in the expression of inflammatory cytokines.Results: NF-κB activation significantly promotes ABCA1mRNA degradation,indicating that the reduction in mRNA is caused by destabilization. NF-κB activationincreased the SREBP-2/miR-33a and SREBP-1a/miR-33b mRNA expression in aconcentration-dependent manner when compared with a negative control. Accordingly, NF-κB also increase the SREBPs protein levels. The Luciferase-SREBPs reporterswere expressed at markedly higher levels as compared with the luciferase vectorwithout insert. Both deletions and mutations of κBRE abolished the enhancing effectsof NF-κB compared to the wild-type control construct. Further, ChIP assays with anantibody to the p65subunit of NF-κB demonstrated a direct binding of thistranscription factor to the promoter regions of the SREBPs genes. ABCA1mRNA andprotein levels were both reduced in THP-1cells transfected with excess wild-typehuman miR-33a/b mimic oligonucleotides compared to the control group. By contrast,the effects of NF-κB on ABCA1were abolished by anti-miR-33a/b. Accordingly,introduction of miR-33mimic into macrophages resulted in further repression of the[3H]-cholesterol efflux compared to the control oligonucleotides, whereas miR-33antisense oligonucleotides treatment led to a marked increase in [3H]-cholesterolefflux. The increase in IL-1β release observed in WT macrophages was significantlyreduced with SREBPs knockdown, but the repression of either SREBP-1or SREBP-2had no obvious effects on TNF-α, IL-6and IL-18expression. Treatment with siRNAfor SREBPs significantly down-regulated NF-κB-induced NLRP1protein expressionin THP-1cells without detectable effect on NLRP3. In addition, the effect of NF-κBon IL-1β expression in NLRP1siRNA cells was obviously suppressed.Conclusion:①SREBPs are direct NF-κB target genes.②NF-κB down-regulates ABCA1level and cholesterol efflux via promoting miR-33expression.③SREBPs increase NF-κB-induced IL-1β expression in macrophages throughactivation of NLRP1.Part III: Effects of NF-κB on Atherosclerotic Lesions andinflammatory cytokines expression in apoE-/-MiceAims: To observe the effect of NF-κB activation on atherosclerotic lesion, RCT,inflammatory response and the changes of molecules expression related to NF-κB-SREBPs pathway in apoE-KO mice. Methods: Six-week old male apoE-/-mice were obtained from LaboratoryAnimal Center of Peking University, China. All mice were fed a chow diet. At8weeks of age, apoE-/-mice were randomly divided into several groups (n=15pergroup). The LPS group was challenged intraperitoneally (i.p.) with LPS (2.5mg/kgbody wt) in200μL of PBS once every week. Mice in the PDTC group were injectedi.p. with PDTC (50mg/kg body wt)1h before LPS challenge. The control group wasreceived abdominal injections of PBS of the same volume at the same time. At week16, the mice were sacrificed, blood was obtained, and tissues were collected forfurther analysis. Triglyceride (TG), total cholesterol (TC), and HDL-C weredetermined by commercially enzymatic methods. Lipid accumulation in aorta andaortic sinus were evaluated by Oil Red O stain. The expression of CD68, a marker formacrophage, and NF-κB p65in lesions were tested by immuno-histochemistry.Masson’s staining for collagen. Plasma inflammatory cytokines (TNF-α、IL-1β andMCP-1etc.) expression was detected by ELISA. RCT efficiency of murinemacrophage cell was detected by liquid scintillation counting. The protein and mRNAexpression of SREBPs、miR-33、ABCA1、ABCG1、NLRPs and NF-κB in mouseperitoneal macrophages (MPMs) were examined by western immunoblotting assaysand real-time quantitative PCR, respectively.Results: Plasma TG, TC and LDL-C levels were increased in LPS groupcompared with the control group, and the HDL-C levels were decreased. ApoE-/-miceinjected with LPS and PDTC significantly reduced lesion size in comparison to theLPS group. Treatment of LPS-challenged apoE-/-mice with PDTC significantlyreduced cholesterol transport from macrophages to feces in the apoE-/-mice. Incontrast, both LPS and PDTC had no significant effect on cholesterol tracer recoveryin the liver and plasma. Treatment with PDTC significantly promoted the expressionof both ABCA1and ABCG1and reduced the expression of miR-33a in mouseperitoneal macrophages. Moreover, there was a reduction in CD68+macrophagescontent and increase in total lesional collagen content in PDTC group compared withLPS group. The NF-κB expression was decreased when treated with PDTC, andinhibition of NF-κB activity could reduce the inflammatory cytokines release. Analysis of the expression of SREBPs and NLRPs revealed a significant increased inLPS-challenged apoE-/-mice peritoneal macrophages, and this effect could be partlyreversed by application of the NF-κB specific inhibitor PDTC.Conclusion:①NF-κB activation in vivo can increase the xpression of miR-33and SREBPs, reduce the expression of ABCA1/ABCG1, inhibit the RCT in vivo andpromote the development of mice atherosclerotic lesions.②NF-κB activation canincrease the level of inflammatory cytokines in vivo and promote NLRPs expression.Part Ⅳ: Effect and Mechanism of Betulinic Acid on ABCA1Expression and Inflammatory Cytokines ProductionAims: To investigate the effect and molecular mechanism of BA on ABCA1expression and inflammatory cytokines production in vitro and in vivo.Methods: Human THP-1monocytes were preincubated with PMA and ox-LDLto form foam cells. Cells was pre-treated with BA (0,0.5,1,2μg/ml) for24hr or withBA (1μg/ml) for0,12,24and48hr, respectively, and then exposed them to LPS. Toobserved the effects of BA on cholesterol efflux and the expression of ABCA1inLPS-treated macrophages. HPLC was performed to determine cellular totalcholesterol, free cholesterol and cholesterol ester. Real-time quantitative PCR wasperformed the effect of BA on the expression of miR-33s and their host genesSREBPs. Cells were transfected with miR-33a/b mimics or with anti-miR-33a/b toassess the effects of miR-33in BA promoting ABCA1expression and cholesterolefflux. The protein expression of nuclear NF-κB was examined by westernimmunoblotting assays. THP-1macrophage-derived foam cells were pretreated withPDTC (50μM) or Bay11-7082(5μM) for24h, and cells were then incubated withLPS for another24h with or without pretreatment of BA. Expression of miR-33smRNA was confirmed by RT-PCR. Cellular cholesterol efflux was analyzed by liquidscintillation counting assays to access the effect of NF-κB in BA promotingcholesterol efflux. Protein levels of IκBα, p-IκBα and phosphorylation of p65were measured and analyzed. Furthermore, we examined the effect of BA onLPS-stimulated inflammatory cytokine production in macrophages to clarify theanti-inflammatory effect and mechanism of BA. In addition, we observed the effect ofBA on atherosclerotic lesion and inflammatory response in apoE-KO mice. Plasma TG,TC and LDL-C levels were increased in LPS group compared with the control group.Both TG and TC levels in animals treated with BA were decreased. Analysis of theplasma lipoproteins showed an increase in HDL cholesterol and a decrease in LDLcholesterol levels in group of BA treatment. LPS-injected apoE-/-mice treatedadditionally with BA have a significantly reduced lesion size. BA significantlyreduced the expression of miR-33and promoted the expression of ABCA1ascompared with those treated by LPS alone. BA also suppressed the activation ofNF-κB and decreased the plasma pro-inflammatory cytokines levelsResults: Pre-treated with BA clearly decreased the inhibition of cholesterolefflux and ABCA1mRNA and protein expression induced by LPS. Cellularcholesterol content decreased when cells were treated with BA. LPS significantlyincreased the miR-33s and SREBPs expression in THP-1macrophage-derived foamcells, whereas treatment with BA down-regulated the expression of miR-33s andSREBPs compared with those treated by LPS only. The effects of LPS on ABCA1were reversed by anti-miR-33a/b. Whereas ABCA1levels was not changedsignificantly in THP-1cells transfected with excess wild-type human miR-33a/bmimic oligonucleotides compared to the LPS group. In addition, the effect of BA onABCA1expression was reversed by transfected with miR-33a/b mimic. With the BAtreatment, the promoting effects of LPS at nuclear NF-κB p65protein levels wereeffectively abrogated. miR-33s expression in cells treated by the combination ofNF-κB specific inhibitor (PDTC or Bay11-7082) was significantly decreased, at thesame time, PDTC or Bay11-7082efficiently promote ABCA1expression and cellularcholesterol efflux induced by BA. BA treated cells showed significant decrease in thephospho-protein expression of IκBα and NF-κB p65. BA significantly inhibitedLPS-stimulated secretion of TNFα, IL-6and IL-1β. Conclusion:①Anti-inflammation and inhibition of plasma lipids are involvedin the mechanisms of BA treatment induces atherosclerosis regression in vivo.②NF-κB-SREBPs patherway involved in the reduction of atherosclerosis mediated byBA.