空气污染加重食物诱导性肥胖小鼠模型脂肪炎症和胰岛素抵抗_图文

1、Ambient Air Pollution Exaggerates Adipose Inflammationand Insulin Resistance in a Mouse Model ofDiet-Induced ObesityQinghua Sun,MD,PhD;Peibin Yue,MD,PhD;Jeffrey A.Deiuliis,PhD;Carey N.Lumeng,MD;Thomas Kampfrath,MS;Michael B.Mikolaj,MD;Ying Cai,MD;Michael C.Ostrowski,PhD;Bo Lu,PhD;Sampath Parthasarat

2、hy,MBA,PhD;Robert D.Brook,MD;Susan D.Moffatt-Bruce,MD,PhD;Lung Chi Chen,PhD;Sanjay Rajagopalan,MDBackground There is a strong link between urbanization and type 2diabetes mellitus.Although a multitude of mechanisms have been proposed,there are no studies evaluating the impact of ambient air pollutan

3、ts and the propensity to develop type 2diabetes mellitus.We hypothesized that exposure to ambient fine particulate matter (2.5m;PM 2.5exaggerates diet-induced insulin resistance,adipose inflammation,and visceral adiposity.Methods and Results Male C57BL/6mice were fed high-fat chow for 10weeks and ra

4、ndomly assigned to concentrated ambient PM 2.5or filtered air (n 14per groupfor 24weeks.PM 2.5-exposed C57BL/6mice exhibited marked whole-body insulin resistance,systemic inflammation,and an increase in visceral adiposity.PM 2.5exposure induced signaling abnormalities characteristic of insulin resis

5、tance,including decreased Akt and endothelial nitric oxide synthase phosphorylation in the endothelium and increased protein kinase C expression.These abnormalilties were associated with abnormalities in vascular relaxation to insulin and acetylcholine.PM 2.5increased adipose tissue macrophages (F4/

6、80cellsin visceral fat expressing higher levels of tumor necrosis factor-/interleukin-6and lower interleukin-10/N -acetyl-galactosamine specific lectin 1.To test the impact of exposed these mice to PM 2.5or saline intratracheally.PM 2.5induced YFP cell accumulation in visceral fat and potentiated YF

7、P cell adhesion in the microcirculation.Conclusion PM 2.5exposure exaggerates insulin resistance and visceral inflammation/vide a new link between air pollution and type 2diabetes mellitus.(Circulation .Key Words:air pollution diabetes mellitus macrophage Substantial epidemiological evi

8、dence implicates air pollu-tion as a major adverse risk factor with serious conse-quences on human health in both industrialized and develop-ing countries.1,2Recent data from large population cohorts have provided compelling associations between ambientPM 2.5pollution and increased cardiovascular mo

9、rbidity and mortality.3,4Substantial data exist implicating fine particulate matter (diameter 2.5m;PM 2.5as a major mediator of generated from power plants.associative studies have suggested that exposure to low concentrations of inhaled environmental pollutants might be associated with a propensity

10、 to chronic diseases,including type 2diabetes mellitus.7,8Although a number of factors such as increased caloric intake and reduced physical activity undoubtedly play an important part in the genesis of theseadipose tissue.911Accordingly,sure to PM 2.5exaggerates insulin resistance (IRand adipose in

11、flammation and tested this hypothesis in a model of diet-induced obesity.Received June 13,2008;accepted October 24,2008.From the Davis Heart and Lung Research Institute (Q.S.,P.Y.,J.A.D.,T.K.,M.B.M.,Y.C.,S.P.,S.R.,Division of Environmental Health Sciences (Q.S.,Division of Biostatistics (B.L.,Depart

12、ment of Molecular and Cellular Biochemistry (M.C.O.,and Division of Cardiothoracic Surgery (S.D.M.-B.,Colleges of Medicine and Public Health,Ohio State University,Columbus;Life Sciences Institute (C.N.L.and Department of Internal Medicine (R.D.B.,University of Michigan,Ann Arbor;and Department of En

13、vironmental Medicine (L.C.C.,New York University School of Medicine,New York.The online-only Data Supplement is available with this article at /cgi/content/full/CIRCULATIONAHA.108.799015/DC1.Correspondence to Sanjay Rajagopalan,MD,Wolfe Professor of Medicine and Radiology,D

14、avis Heart and Lung Research Institute,Room 110,473W 12th Ave,Columbus,OH 432101252.E-mail Sanjay.R ©2009American Heart Association,Inc.Circulation is available at DOI:10.1161/CIRCULATIONAHA.108.799015538Molecular Cardiology脂肪组织巨噬细胞肿瘤坏死因子环境空气污染在食物诱导性

15、肥胖的小鼠模型中加重脂肪炎症和胰岛素抵抗的研究心血管疾病 发病率关联性研究累计证据Editorial p492Clinical Perspective p546MethodsA detailed description of all materials and methods is provided in the online-only Data Supplement.Animals and Animal CareSix-week-old male C57BL/6mice(The Jackson Laboratories,Bar Harbor,Mewere equilibrated for2w

16、eeks before being fed high-fat chow(HFC;42%from fat-adjusted calorie diet,TD88137,Harlan, Indianapolis,Ind;n28for10weeks and then randomized to exposure(see below.Ten-week-old transgenic mice expressing yellow fluorescent protein(YFPunder a monocyte-specific pro-moter (c-fms CD115;c-fms YFP12were fe

17、d normal chow or wererendered diabetic by the use of a protocol similar to that stated above over10weeks(n8before being subjected to intratracheal PM2.5exposure.The Committees on Use and Care of Animals of OhioState University and New York University approved all experimentalprocedures.to PM2.5Whole

18、-Body Inhalational Protocolmice were exposed in vivo from October9,2006,to March26,2007,for a total duration of128days,to concentrated PM2.5composed of the northeastern regional background at the AJ LanzaLaboratory in the Department of Environmental Medicine at NewYork University,which was character

19、ized previously.13The concen-trated PM2.5was generated by a versatile aerosol concentrationenrichment system14that was modified for longer-term exposures.15The mice were exposed to PM2.5(n14for6h/d for5d/wk.Thefiltered air(FA;n14control mice were exposed to an identicalprotocol except a high-efficie

20、ncy particulate air filter(Pall LifeSciences,East Hills,NYwas positioned in the inlet valve to theexposure system to remove all of the particles from that airstream,asdetailed previously.16Intratracheal Installation Protocolc-fms YFP mice were randomized to intratracheally delivered PM2.5(1.6mg/kg,c

21、ollected from the filters after the exposure abovesuspended in25L sterile PBS or25L sterile PBS(controlovera5-minute period twice per week for10weeks,as detailedpreviously.17PM2.5in PBS was vortexed before instillation.Priorstudies have demonstrated that delivery with this route results insimilar le

22、vels of pulmonary distribution and bioavailability com-pared with ambient exposure.18Measures of Glucose HomeostasisMice were fasted overnight before the body weight was measuredand lipid profile and intraperitoneal glucose tolerance tests wereperformed before and after PM2.5whole-body exposure in C

23、57BL/6mice.The intraperitoneal glucose tolerance test also was performedin c-fms YFP mice after10weeks of normal chow/HFC feeding.Lipidprofile was tested by Cholestech LDX(Hayward,Calif.Fastingblood glucose concentration was determined with an Elite Glucom-eter(Bayer,Mishawaka,Ind,and a20-L sample o

24、f blood wastaken for basal blood insulin concentration measurement.Then,themice were administered an intraperitoneal injection of glucose(2mg/g body weight;blood glucose concentrations were measured;and20-L blood samples were taken at30,60,90,and120minutesafter injection.Serum insulin levels were me

25、asured with an insulinELISA kit(Alpco Diagnostics,Windham,NH.The homeostasismodel assessment of the IR index was then calculated as describedpreviously.19Magnetic Resonance ImagingIn vivo magnetic resonance imaging to assess visceral fat accumu-lation was performed with a T1-weighted gradient-echo s

26、equence onall C57BL/6mice when exposure ended.Magnetic resonance imag-ing was performed with a11.7-T Bruker magnet.Myograph ExperimentsAortic ring segments were subjected to graded doses of phenyleph-rine,acetylcholine,or insulin as described previously.20Adipose Stromal Vascular Fractionation andMa

27、crophage IsolationStromal vascular fraction pellet was harvested from epididymalfat pads,and the lymphocyte layers were incubated with F4/80antibody(Biolegend,San Diego,Califfollowed by anti-biotinsuperparamagnetic colloidal particles.Adipose tissue macro-phages(ATMswere separated with an MACS MS co

28、lumn/magnet according to the manufacturers instruction(MiltenyiBiotec,Germany.Confocal Microscopy Studies in Fixed and LiveAdipose TissueTo image ATMs in the C57BL/6group by immunofluorescence,epididymal fat pads were dissected and incubated in1%paraformal-dehyde overnight and incubated with anti-F4

29、/80and anti-caveolinantibodies.Tissue was imaged with an inverted confocal scanningmicroscope.To image live adipose tissue from c-fms YFP mice,epididymal fat tissue was minced into small pieces,washed thor-oughly,and incubated with antibodies.21Intravital Microscopy in c-fms YFP Micec-fms YFP mice w

30、ere anesthetized intraperitoneally by a mixture ofketamine(100mg/kgand xylazine(20mg/kg.The mesenterictissue and cremaster muscle were exteriorized.On an optically coherent mount.The tissues were superfusedwith prewarmed Ringers lactate(37°C,and the number ofadherent YFP cells in5to10movies was

31、 acquired with a40/0.80-W water-immersed objective using a Nikon EclipseFN1microscope(Nikon,Tokyo,Japanand Metamorph software(version,Metamorph,Downingtown,Pa.The number ofadherent cells in cremasteric muscle was determined by countingstationary leukocytes in a100-m vessel length per30seconds

32、;adherent cells in the mesenteric tissue were counted per imagefield(1.57105m2.Data AnalysesData are expressed as meanSEM unless otherwise indicated.Forresponses measured repeatedly at different time points or dose levels,a series of2-sample independent Students t tests were used to detectthe signif

33、icant differences between2treatment groups at every timepoint and dose level with the Bonferroni correction for multiplecomparison adjustment.Comparisons of other continuous variableswere conducted with an independent2-sample Student t test,withvalues of P0.05considered significant.The authors had f

34、ull access to and take responsibility for theintegrity of the data.All authors have read and agree to themanuscript as written.ResultsWhole-Body Exposure DataAmbient mean daily PM2.5concentration at the study site was7.7g/m3(SD,5.0g/m3.Mean concentration of PM2.5inthe exposure chamber was72.7g/m3(9-

35、fold concentra-tion from ambient levels.Because the mice were exposed for6hours a day,5days a week,the equivalent PM2.5concen-tration to which the mice were exposed in the chambernormalized over the exposure period(128dayswas13.0g/m3,which is well within the annual average PM2.5NationalAmbient Air Q

36、uality Standard(NAAQSof15g/m3.22TheSun et al Air Pollution Exaggerates Insulin Resistance539气管内的单细胞特异性预实验预实验证明丆此等剂量与环境暴露的结果近似血脂血糖耐受试验磁共振成像内脏脂肪堆积肌动扫描试验去氧肾上腺素用于评定去氧肾上腺素、醋丹素等的用量脂肪间质血管分馏和巨噬细胞分离附睾脂肪组织巨噬细胞共聚焦显微镜观察固定标本和活脂肪组织抗体荧光染色反向共聚焦扫描活体组织显微镜下观察麻醉国家环境空气质量标准general exposure characteristics have previously

37、 been described.13 Metabolic Impairment by PM 2.5Before exposure began,there was no statistically significant difference between the 2groups in body weight,chow consumption,lipid profile,and glucose tolerance response.Table I of the online Data Supplement summarizes the metabolic data in the FA and

38、PM 2.5groups at the end of 24weeks of exposure.C57BL/6mice exposed to PM 2.5demon-strated identical chow consumption (2.20.2g/d in FA versus 2.20.3g/d in PM 2.5and weight gain over the duration of the entire experiment (Table I in the Data Supplement.PM 2.5-exposed mice displayed abnormal in-dexes o

39、f glucose/insulin homeostasis as evidenced by ele-vated fasting glucose,insulin,homeostasis model assessment indexes,and abnormalities in lipid profile consistent with IR phenotype.Figure 1A displays responses to the intraperito-neal glucose tolerance test.PM 2.5mice demonstrated eleva-tions in gluc

40、ose levels at all time points after 20minutes.PM 2.5Impairs VascularEndothelium-Dependent ResponsesBecause the vascular endothelium is a sensor for multiple stimuli,including inhaled particulates,and plays an importantrole in IR development,we investigated endothelial responses to acetylcholine and

41、insulin (Figure 1B and 1C.PM 2.5-exposed C57BL/6mice demonstrated a decrease in peak relaxation and ED 50to acetylcholine and decreased peak relaxation to insulin compared with FA mice.Because these changes are generally consistent with a decrease in nitric oxide release in response to these agonist

42、s,we confirmed the effects of PM 2.5in reducing vascular nitric oxide bioavail-ability by the increment in tension in preconstricted aortic rings to N G -monomethyl-L -arginine (L-NMMA;104mol/L.PM 2.5-exposed mice demonstrated reduced constriction to this agent (Figure 1D,consistent with lower level

43、s of nitric oxide release.PM 2.5Exposure Impairs Insulin SignalingWe examined the effect of PM 2.5exposure on insulin signal-ing in the aorta and liver.The phosphorylation of Akt in intact aorta but not denuded aortic segments was reduced in the PM 2.5group compared with the FA group (Figure 2A and

44、2B.These results suggest that the differential effect of PM 2.5on aorta was secondary to differences in endothelial Akt phos-phorylation.We also investigated the time course of Akt phosphorylation in response to stimulation with insulin and found that Akt phosphorylation levels were decreased at 30m

45、inutes in PM 2.5compared with FA (Figure 2C.To inves-Ach, mol/L% R e l a x a t i o n50100150% I n c r e a s e i n T e n s i o n2.5-120-90-60-300Insulin, mol/L% R e l a x a t i o n2004006000306090120Time, minG l u c o s e , m g /m lABCDFigure 1.Glucose tolerance test and vaso-motor response in FA-or

46、PM 2.5-exposed C57BL/6mice fed HFC.A,Blood glucose levels by glucose tolerance test.B,Mean vasorelaxation of aortic rings in response to acetylcholine (Ach.C,Insulin-mediated vasorelaxation in aortic rings preconstricted with phenylephrine (107mol/L.D,Maxi-mum phenylephrine-induced vasoconstric-tion

47、 in aortic rings preincubated with the nitric oxide synthase inhibitor L-NMMA (104mol/L.n 7.*P 0.05vs FA.Ap-Akt BAktCp-Akt Akt4080120p -A k t /A k t 2.54080120p -A k t /A k t 2.5Insulin, minp -A k t /A k tFigure 2.PM 2.5exposure impairs insulin signaling via the PI3-kinase/Akt pathway in FA-or PM 2.

48、5-exposed C57BL/6mice fed HFC.A,Immunoblotting and analysis of Akt (Ser 473phosphorylation in intact aortic tissues.B,Immunoblotting and analysis of Akt (Ser 473phosphorylation in denuded aortic tissues.C,Time course of Akt phosphorylation in response to insulin (500mU/mLin intact aortic rings.n 5.*

49、P 0.05vs FA.540Circulation February 3,2009醋丹素苯肾上腺素新陈代谢损害tigate whether protein kinase C (PKCis involved in PM 2.5-mediated IR and vascular dysfunction,we investigated the expression of multiple PKC isoforms by immunohistochem-istry.PKC-II but not PKC-or PKC-was increased in PM 2.5compared with the F

50、A groups (data not shown.The increased PKC-II expression colocalized in aortic tissues,with anti-insulin receptor substrate-1suggesting a role for this pathway in PM 2.5-mediated IR (see the Data Supplement.Taken together,these data suggest that insulin signaling is abnormal in response to PM 2.5exp

51、osure.PM 2.5Exposure Induces Adipose Inflammation and Visceral AdiposityOn the basis of the strong evidence that adipose tissue inflammation contributes to IR,we examined the effects of PM 2.5exposure on systemic and local adipose tissue inflam-mation.Figure 3shows the plasma concentration of adipok

52、ines and inflammatory biomarkers in C57BL/6mice exposed to either PM 2.5or FA.Tumor necrosis factor-(TNF-,interleukin-6(IL-6,E-selectin,intracellular adhe-sion molecule-1(ICAM-1,plasminogen activator inhibi-tor-1,and resistin were significantly increased in the mice exposed to PM 2.5compared with FA

53、.To examine whether PM 2.5exposure altered visceral adiposity,magnetic resonance imaging measurements using T1-weighted techniques were performed before and after PM 2.5exposure.Before exposure,there was no statistical significance in body fat mass between the 2groups (data not shown.Although PM 2.5

54、exposure had no effect on subcutaneous and retroperitoneal fat mass,increased visceral and mesenteric fat mass was observed compared with FA group (Figure 4A through 4C.ATMs,known to be mediators of adipose tissue inflamma-tion and IR,were examined in exposed mice.PM 2.5exposure led to a marked incr

55、ease in F4/80macrophages in adiposetissue (Figures 5A,5B,and 6A.In addition,adipocyte size as measured by cross-sectional area was increased in PM 2.5-exposed mice (FA,883m 221;PM 2.5,973m 226;P 0.01(Figure 5C and 5D.In view of the increased inflammation seen in adipose tissue,we investigated the ex

56、pression of ATM-specific genes in the F4/80cells in the stromal vascular fraction.PM 2.5treatment resulted in signif-icant increases in the expression of proinflammatory genes TNF-and IL-6(M1with no change in nitric oxide synthase-2expression.In contrast,IL-10and the alternative (M2macrophage activa

57、tion marker galactose-N -acetylgalac-tosaminespecific lectin (Mgl1were markedly downregu-lated (Figure 6B.These results indicated that PM 2.5down- CC o n c e n t r a t i o n FA PM 2.5FA PM 2.5FA PM 2.5E-selectin ICAM-1 PAI-1 pg/ml ng/ml ng/ml51015C o n c e n t r a t i o n2.5 2.5TNF-IL-6pg/ml ng/ml 1

58、000200030004000C o n c e n t r a t i o nFA PM 2.5FA PM 2.5Resistin Leptin pg/ml pg/mlABFigure 3.Exposure to ambient PM 2.5increases circulatory adipokines and inam-matory biomarkers in C57BL/6mice fed HFC.A,TNF-and IL-6.B,E-selectin,ICAM-1,and plasminogen activator inhibi-tor-1(PAI-1.C,Resistin and leptin.n 7.*P 0.05vs FA.Figure 4.PM 2.5exposure increases visceral fat mass measured by magnetic reso