reduced vascular nitric oxide-cgmp signaling contributes to adipose tissue inflammation during...

Reduced Vascular Nitric Oxide–cGMP SignalingContributes to Adipose Tissue Inflammation During

High-Fat FeedingPriya Handa, Sanshiro Tateya, Norma O. Rizzo, Andrew M. Cheng, Vicki Morgan-Stevenson,

Chang-Yeop Han, Alexander W. Clowes, Guenter Daum, Kevin D. O’Brien,Michael W. Schwartz, Alan Chait, Francis Kim

Objective—Obesity is characterized by chronic inflammation of adipose tissue, which contributes to insulin resistance anddiabetes. Although nitric oxide (NO) signaling has antiinflammatory effects in the vasculature, whether reduced NOcontributes to adipose tissue inflammation is unknown. We sought to determine whether (1) obesity induced by high-fat(HF) diet reduces endothelial nitric oxide signaling in adipose tissue, (2) reduced endothelial nitric oxide synthase(eNOS) signaling is sufficient to induce adipose tissue inflammation independent of diet, and (3) increased cGMPsignaling can block adipose tissue inflammation induced by HF feeding.

Methods and Results—Relative to mice fed a low-fat diet, an HF diet markedly reduced phospho-eNOS andphospho-vasodilator-stimulated phosphoprotein (phospho-VASP), markers of vascular NO signaling. Expression ofproinflammatory cytokines was increased in adipose tissue of eNOS�/� mice. Conversely, enhancement of signalingdownstream of NO by phosphodiesterase-5 inhibition using sildenafil attenuated HF-induced proinflammatory cytokineexpression and the recruitment of macrophages into adipose tissue. Finally, we implicate a role for VASP, a downstreammediator of NO-cGMP signaling in mediating eNOS-induced antiinflammatory effects because VASP�/� micerecapitulated the proinflammatory phenotype displayed by eNOS�/� mice.

Conclusion—These results imply a physiological role for endothelial NO to limit obesity-associated inflammation inadipose tissue and hence identify the NO-cGMP-VASP pathway as a potential therapeutic target in the treatment ofdiabetes. (Arterioscler Thromb Vasc Biol. 2011;31:2827-2835.)

Key Words: endothelium � nitric oxide � adipose � inflammation

Obesity is linked to a state of chronic low-grade inflam-mation in adipose and other tissues that in turn contrib-

utes to insulin resistance and thereby predisposes to type-2diabetes. At the cellular level, nutrient excess activates theinhibitor of nuclear factor kappa B kinase subunit �/nuclearfactor-�B (NF-�B) pathway, a key mediator of inflammationin many cell types.1 In visceral adipose tissue, this phenom-enon is associated with infiltration of macrophages2–4 thatcan impair insulin signaling through release of proinflamma-tory cytokines such as interleukin-6 (IL-6) and tumor necrosisfactor-� (TNF-�).5 Monocyte chemoattractant protein-1 (alsoknown as CC chemokine ligand 2), produced by the adiposetissue in response to dietary overload, is believed to be crucialin recruiting macrophage precursors into the adipose tissuevia interaction with its receptor, CC chemokine receptor 2(CCR-2), on the surface of the monocytes.5 Adipose tissueconsists of adipocytes along with a stromal vascular fraction,

the latter containing resident and infiltrating macrophagesand an extensive network of vascular endothelial cells. Inaddition to transporting nutrients and adipokines, growingevidence suggests that the vasculature itself can activelyinfluence tissue inflammation triggered by obesity.6,7

Among early responses observed in association withobesity-induced insulin resistance is a reduction of vascularnitric oxide (NO) content, an effect that precedes the onset ofvascular and then peripheral tissue (such as liver, muscle, andadipose) inflammation and insulin resistance during dietaryexcess.8 This observation raises the possibility that a dimin-ished level of vascular NO may itself increase tissue suscep-tibility to the inflammatory effects of nutrient excess duringhigh-fat (HF) feeding. Consistent with this hypothesis is alarge body of literature documenting antiinflammatory effectsof vascular NO.7,9

The enzyme endothelial nitric oxide synthase (eNOS)synthesizes NO from L-arginine and molecular oxygen in

Received on: December 2, 2010; final version accepted on: August 29, 2011.From the Department of Medicine (P.H., S.T., N.O.R., A.M.C., V.M.-S., C.-Y.H., K.D.O., M.W.S., A.C., F.K.), Department of Surgery (A.W.C.,

G.D.), and Diabetes and Obesity Center of Excellence (P.H., S.T., N.O.R., A.M.C., V.M.-S., C.-Y.H., M.W.S., A.C., F.K.), University ofWashington, Seattle, WA.

Correspondence to Francis Kim, MD, Department of Medicine, Box 358055, 815 Mercer St, University of Washington, Seattle, WA 98109. [email protected]

© 2011 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org DOI: 10.1161/ATVBAHA.111.236554

2827 by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from

http://atvb.ahajournals.org/













vascular tissues.9 eNOS-derived NO induces vasodilatory,antiinflammatory, antithrombotic, and antiproliferative ef-fects7,9 that may extend to surrounding nonvascular tissues.Although the role of eNOS-derived NO in adipose tissue hasbeen studied with respect to its importance in mitochondrialbiogenesis10 and adipocyte differentiation,11 it is not knownwhether reduced vascular NO signaling plays a role inobesity-associated adipose tissue inflammation.

Signal transduction induced by endothelial-derived NOinvolves activation of soluble guanylyl cyclase in vascularsmooth muscle and many other cell types, thereby increasingcGMP production, which in turn activates the serine-threonine kinase cGMP-dependent protein kinase (PKG).Among several substrates for PKG action is vasodilator-stimulated phosphoprotein (VASP), which is phosphorylatedby PKG on a specific serine residue (Ser-239). Althoughactivation of VASP by PKG is implicated in several effects ofNO-cGMP-PKG signaling,12–14 whether VASP contributes tothe antiinflammatory effects of this pathway has yet to beinvestigated.

Intracellular levels of cGMP are governed not only byguanylyl cyclase activity but also by phosphodiesterase con-version of cGMP back to GMP. Sildenafil is an inhibitor ofphosphodiesterase type 5, which attenuates intracellular ca-tabolism of cGMP and was shown previously to restrictvascular inflammation and insulin resistance in mice fed anHF diet.15

In the current studies, we sought to investigate whetherreduced vascular NO signaling contributes to development ofobesity-associated adipose tissue inflammation. We foundthat HF feeding reduces biomarkers of vascular NO signalingin adipose tissue and that disruption of either eNOS or VASPsignaling induces adipose tissue inflammation independentof diet. Combined with our finding that increased NO-cGMP signaling induced by sildenafil attenuates adiposeinflammation induced by HF feeding in normal mice, ourfindings implicate a physiological role for vascular NOsignaling to restrain adipose tissue inflammation inducedby nutrient excess.

Materials and MethodsPlease see supplemental material, available online athttp://atvb.ahajournals.org, for expanded Materials and Methodsand Results sections.

Study ProtocolsAdult male C57BL6 wild-type (WT) and eNOS–/– mice on aC57BL6 background were purchased from the Jackson Labora-tory (Bar Harbor, ME).15 VASP-deficient mice and their WTlittermates were also bred onto the C57BL6 background.12 Age-matched groups (6 –12 weeks old) were maintained in atemperature-controlled facility with a 12-hour light-dark cycleand were fed ad libitum an equicaloric diet that was matched formicronutrient content that was either low (10% saturated fat) orhigh (60% saturated fat) in fat (Research Diets, New Brunswick,NJ, nos. D12492 and D12450B). In a separate study involvingWT mice, sildenafil tablets (100 mg) were ground into a powderand mixed into a highly palatable “treat” containing peanut butterand an HF food pellet to deliver sildenafil orally at a dose of 30mg/kg per day, as described previously.15

At the conclusion of the protocol, animals were euthanized afteran overnight fast, with an overdose of CO2 followed by cervical

dislocation. Epididymal adipose tissue was quickly removed andsnap-frozen on dry ice and stored at �80°C until analysis. A portionof the epididymal fat tissue was collected in RNAlater (Ambion,Austin, TX) and stored at �20°C until RNA preparation. Protein wasextracted from epididymal fat samples and quantified using a kit(Micro BCA Protein Assay Kit, Pierce, Rockford, IL). All proce-dures were approved by the University of Washington InstitutionalAnimal Care and Use Committee.

MaterialsAnti–phosphorylated eNOS (serine 1177), anti–phosphorylated p65-subunit of NF-�B, anti–phosphorylated I�B-�, antiphosphorlylated-VASP Ser239, PKG, and �-actin antibodies were from Cell Signal-ing Technology (Beverly, MA); anti-eNOS antibody was obtainedfrom Transduction Laboratories, BD Biosciences (Lexington, KY);Mac-2 antibody was from Cedarlane; and anti-GAPDH was fromSanta Cruz Biotechnology.

Statistical AnalysisStatistical analysis was performed using the GraphPad Prism Version5.0a (GraphPad Software Inc, La Jolla, CA). Data are expressed asmean�SEM, and values of P�0.05 were considered statisticallysignificant. A 2-tailed t test was used to compare mean values for2-group comparisons. To compare responses between sildenafil- andvehicle-treated mice fed either of the 2 diets, data were analyzed by2-way analysis of variance, and the Bonferroni post hoc comparisontest was used to compare mean values between groups.

ResultsHigh-Fat Feeding Reduces eNOS and VASPPhosphorylation and Increases Inflammation inAdipose TissuePrevious evidence indicates that in normal mice, HF feedingimpedes the phosphorylation of eNOS (a marker of eNOSactivation) in the thoracic aorta and that this effect isaccompanied by reduced phosphorylation of Akt,8 (theserine-threonine kinase that phosphorylates and thereby acti-vates eNOS). Similarly, Akt phosphorylation was reducedand NF-�B activation was increased in adipose tissue after 14weeks of HF feeding.8 Together, these observations suggestthat reduced vascular NO signaling may contribute to adiposetissue inflammation during HF feeding. To test this hypoth-esis, we first determined the effect of HF feeding on phospho-eNOS content in epididymal white adipose tissue (EWAT).We found that in normal mice, HF feeding for 14 weeks (atime period known to increase NF-�B signaling in EWAT8)significantly decreased eNOS phosphorylation relative tototal eNOS (Figure 1A). Total eNOS levels tended to de-crease with HF feeding but did not reach statistical signifi-cance (Figure 1A and Supplemental Figure IA).

In vascular tissues, NO-cGMP activates PKG signaling,which leads to VASP phosphorylation at Ser239, an effectimplicated in the vasodilatory function of NO. To examinethe effect of HF feeding on the activation status of adownstream mediator of NO-cGMP signaling in EWAT, wemeasured levels of PKG and phospho-VASP. We found thatcompared with low-fat (LF)–fed controls, mice fed an HF dietfor 14 weeks displayed reduced levels of both PKG andphospho-VASP in EWAT, suggesting a decrease in signalingthrough the vascular NO-PKG-VASP pathway (Figure 1Band Supplemental Figure IB). To further assess NO signalingin adipose tissue, basal NO levels were measured by theelectron spin resonance spectroscopy technique, using the

2828 Arterioscler Thromb Vasc Biol December 2011

by guest on July 28, 2016http://atvb.ahajournals.org/Downloaded from

http://atvb.ahajournals.org


spin trap Fe(DETC)2 as previously described.8 An HF dietwas associated with reduced adipose NO content (Figure 1C),consistent with decreased eNOS phosphorylation.

Combined with our finding of increased TNF-� mRNAcontent in the same EWAT samples (Figure 1D), these findingsextend previous evidence that HF feeding induces NF-�Bactivation in EWAT8 and suggest that this effect is associatedwith reduced vascular NO content and signaling. We alsoobserved an increase in levels of mac-2, a macrophage marker,in EWAT of mice fed an HF diet relative to LF-fed controls(Figure 1E), suggesting that HF feeding increases adipose tissuemacrophage content, consistent with previous work.16

eNOS Deficiency Increases NF-�B-DependentSignaling in EWAT, Independently of DietTo test whether the absence of eNOS mimics the effect of HFfeeding in adipose tissue, we compared adipose tissue fromeNOS�/� and WT mice after an LF and HF diet for 4 weeks.Importantly, the HF diet for 4 weeks does not induce NF-�Bactivation in adipose tissue of WT mice.8 Relative to WTcontrols, EWAT from eNOS�/� mice exhibited increasedTNF-�, IL-6, and monocyte chemoattractant protein-1(MCP-1) mRNA expression (Figure 2A) and increased phos-phorylation of p65 subunit of NF-�B (Figure 2B), despiteshowing no differences in body weight between the 2

groups.15 Similarly, expression of mRNA encoding mono-cyte/macrophage/dendritic cell markers, such as CD68, F4/80, and CD11c, was increased in EWAT from eNOS�/�mice compared with WT mice (Figure 2A). By comparison,M2 markers (arginase-1 and Ym-1, indicative of antiinflam-matory macrophages) were not significantly different be-tween WT and eNOS�/� mice irrespective of diet (Figure2A), with the exception of IL-10, which showed a decrease inthe HF-fed WT mice relative to LF. We also assessed mac-2protein levels in the WT and eNOS�/� mice and found that,consistent with the increased mRNA expression of macro-phage markers and inflammatory cytokines, mac-2 proteinwas increased in eNOS�/� mice relative to littermatecontrols (Figure 2C). These data collectively indicate thatgenetic eNOS deficiency mimics the effect of HF feeding toincrease expression of proinflammatory cytokines and re-cruits macrophages to adipose tissue.

Effect of Sildenafil on HF-Induced SystemicInsulin ResistanceBecause genetic absence of eNOS predisposes adipose tissueto inflammation, we sought to determine whether increasedNO-cGMP signaling exerts a protective effect during HFfeeding. To test this hypothesis in vivo, age-matched WTmale mice were fed either an LF or an HF diet for 14 weeks,

Figure 1. High-fat (HF) feeding diminishes phosphorylation of endothelial nitric oxide synthase (eNOS) and vasodilator-stimulated phos-phoprotein (VASP) in adipose tissue. A, Protein lysates from epididymal fat from low-fat (LF)– and HF-fed mice assessing phosphoryla-tion of Ser1177 in eNOS. Quantitation shows phosphorylation of eNOS (p-eNOS) at Ser1177 relative to eNOS. *P�0.05 LF vs HF. B,Protein lysates made from epididymal white adipose tissue (EWAT) from LF- and HF-fed mice were analyzed by Western blots todetect phosphorylation of VASP at Ser-239 (p-VASP239) and protein kinase G (PKG). Representative blots are shown. Relative phos-phorylation of VASP at Ser-239 normalized to VASP and PKG normalized to GAPDH (GAP) are shown. *P�0.05 LF vs HF, n�3 to 4. C,Relative levels of NO in adipose tissue. *P�0.05 LF vs HF, n�4 to 6. D, Relative mRNA expression of tumor necrosis factor-� (TNF-�)normalized to GAPDH. *P�0.05 LF vs HF (n�4). E, Representative blots and quantitation showing Mac-2 and �-actin protein levels inLF and HF EWAT. *P�0.05 LF vs HF.

Handa et al NO-cGMP Signaling Restricts Adipose Inflammation 2829



a period previously shown to cause both obesity and adiposetissue inflammation in mice.8 During the last 4 weeks of thefeeding protocol, mice in each diet group received a dailydose of either vehicle or sildenafil (30 mg/kg/d orally).Although the effect of HF feeding to raise fasting plasma

glucose and insulin levels was attenuated by daily sildenafiltreatment (Figure 3B and 3C), this intervention did notsignificantly alter body weight (Figure 3A). Thus, the bene-ficial effects of sildenafil in this model are not due to reducedweight gain.

Figure 2. Deficiency of endothelial nitric oxide synthase (eNOS) promotes inflammation in adipose tissue. A, Quantitative mRNA analy-sis of tumor necrosis factor-� (TNF-�), interleukin-6 (IL-6), monocyte chemoattractant protein-1 (MCP-1), CC chemokine receptor 2(CCR-2), CD68, F4/80, CD11c, arginase-1, and IL-10 in adipose tissue from wild-type and eNOS�/� mice on low-fat (LF) and high-fat(HF) diets for 4 weeks. The expression levels were normalized to GAPDH. *P�0.05 wild-type (WT) vs eNOS on LF, †P�0.05 WT vseNOS on HF, ‡WT LF vs HF (n�5–6 per group). B, Protein levels of phosphorylated (p-) p65 relative to �-actin in epididymal white adi-pose tissue (EWAT) of WT and eNOS�/� animals. *P�0.05 WT vs eNOS (n�3–4). C, Protein levels of Mac-2 relative to �-actin inEWAT of WT and eNOS�/� animals. *P�0.05 WT vs eNOS (n�3–4).




To better characterize glucose homeostasis in these mice, weperformed insulin tolerance tests on a separate cohort of vehicleor sildenafil-treated mice fed the HF diet for 14 weeks. Com-pared with the vehicle-treated group, sildenafil treatment wasassociated with a greater insulin-mediated decrease of bloodglucose, implying increased insulin sensitivity (SupplementalFigure IIA). This pattern was evident regardless of whetherinsulin-induced decrease of plasma glucose was expressed inabsolute terms or as the percentage change during the insulintolerance test (Supplemental Figure IIB). Thus, sildenafil ap-pears to ameliorate systemic insulin resistance in mice fed an HFdiet, as observed in an earlier study.17

Effect of Sildenafil on Adipose TissueInflammation in a Diet-Induced Model of ObesityTo confirm that sildenafil increased cGMP levels in adiposetissue, we performed Western blot analysis using an antibodythat recognizes phospho-VASP at Ser239, because VASPphosphorylation correlates with cGMP levels.18 As expected,sildenafil treatment restored phospho-VASP levels in HFEWAT, comparable to LF, demonstrating that sildenafiltreatment improved NO-cGMP signaling in adipose tissue(Supplemental Figure III).

As expected, 14 weeks of HF feeding was associated withincreased NF-�B signaling in adipose tissue, as assessed bothby measuring levels of mRNA encoding MCP-1, TNF-�, andIL-6 and by measures of inflammatory immune cells, such asCCR-2, CD68, F4/80, and CD11c (Figure 4A), and immuno-blotting for phospho-I�B-�, a marker for NF-�B activation(Figure 4B). Importantly each of these effects was attenuatedby sildenafil treatment (Figure 4A and B). By comparison,HF feeding was associated with a decrease of M2 macro-phage marker expression (arginase-1, Ym-1) in EWAT (Fig-ure 4A), and sildenafil treatment did not significantly alterthis effect, suggesting that the protective action of sildenafilinvolves abolishing the recruitment of M1-activated macro-phages rather than retention or recruitment of M2 macro-phages. Consistent with these findings, immunoblot of pro-tein lysates from high fat fed vehicle and sildenafil-treatedmice using an antibody specific for mac-2 (Figure 4C)showed that levels of mac-2 in sildenafil-treated mice werereduced relative to controls. Thus, sildenafil treatment re-duces adipose tissue macrophage infiltration and inflamma-tion induced by HF feeding.

Effect of VASP on Adipose Tissue InflammationTo test the hypothesis that VASP signaling is necessary forthe antiinflammatory effects of NO-cGMP signaling onadipose inflammation, we used VASP-deficient and WTlittermate mice fed an LF and HF diet for 4 weeks. Asexpected, in WT mice, 4 weeks of HF feeding was notassociated with significant elevation in adipose inflammationcompared with LF controls. On the other hand, we found thatin EWAT from VASP-null mice, levels of mRNA encodingproinflammatory cytokines MCP-1, IL-6, and TNF-� weresignificantly elevated compared with WT controls, as was thephosphorylation of the p65 subunit of NF-�B, indicative ofincreased NF-�B activation (Figure 5A and 5B). Further-more, inflammatory myeloid markers such as CCR2,CD68, F4/80 and CD11c were also upregulated in EWATfrom VASP-null mice compared with littermate controls.The expression of M2 macrophage markers (such asarginase-1 and IL-10 mRNA) showed no significant dif-ferences in the EWAT of VASP�/� mice relative tocontrols (Figure 5A). In addition, mac-2 protein levelswere higher in the VASP null mice relative to controls(Figure 5C). Like eNOS deficiency, therefore, the absenceof VASP mimics the effect of HF feeding to induceproinflammatory activation in adipose tissue.

DiscussionOvernutrition induces proinflammatory activation in adiposeand many other tissues, and a growing body of evidencesuggests that this response links obesity to insulin resistanceand diabetes.1 On the basis of evidence that endothelial nitricoxide signaling has antiinflammatory properties in the vas-culature and that vascular nitric oxide levels decline early inthe course of diet-induced obesity, we hypothesized anetiologic role for reduced vascular NO-cGMP signaling in thepathogenesis of adipose tissue inflammation and insulinresistance in diet-induced obesity. In this study, we found thatin normal mice, HF feeding reduced vascular NO signaling inwhite adipose tissue in a manner that is temporally associatedwith adipose inflammation and insulin resistance.8 Further-more, genetic deficiency of eNOS induced adipose tissueinflammation even on an LF diet, suggesting that intactendothelial NO signaling limits inflammatory signaling inthis tissue. Conversely, the deleterious inflammatory activa-tion of adipose tissue induced by HF diet—measured in terms

Figure 3. Metabolic parameters of mice on 14 weeks of diet-induced obesity treated with vehicle or sildenafil. A, Weights of the low-fat (LF)and high-fat (HF) mice treated with either vehicle (open bars) or sildenafil (black bars). B, Fasting glucose levels. C, Fasting insulin levels.*P�0.05 vehicle LF vs HF, †P�0.05 vehicle- vs sildenafil-treated HF-fed mice, ‡P�0.05 LF vs HF-fed sildenafil-treated mice (n�6).




of both proinflammatory gene expression and adipose tissuemacrophage accumulation—was prevented in WT mice bytreatment with sildenafil, a drug that increases signalingdownstream of vascular NO. This effect was accompanied byimproved insulin sensitivity. Furthermore, we showed thatdeficiency of VASP, a downstream target of the NO-cGMPpathway, recapitulated the effect of eNOS deficiency to

increase adipose tissue inflammation. Together, these resultsprovide direct evidence in support of the hypothesis thatvascular NO-cGMP signaling protects against adipose in-flammation and that reduced eNOS/VASP function contrib-utes to adipose tissue inflammation during HF feeding.

Adipose tissue is a complex endocrine organ composed ofmultiple cell types, including adipocytes, macrophages, and

Figure 4. Adipose tissue inflammation in vehicle- and sildenafil-treated low and high-fat (HF)–fed animals. A, mRNA levels of a panel ofmacrophage, inflammatory, and antiinflammatory markers in the adipose tissue. Vehicle-treated (open bars) and sildenafil-treated (blackbars) low-fat (LF)– and HF-fed mice were assessed for monocyte chemoattractant protein-1 (MCP-1), tumor necrosis factor-� (TNF-�),interleukin-6 (IL-6), CC chemokine receptor 2 (CCR-2), CD68, F4/80, CD11c, arginase-1 (ARG-1), and Ym-1 relative to GAPDH.*P�0.05 vehicle LF vs HF, †P�0.05 vehicle- vs sildenafil-treated HF-fed mice (n�6 each). B, Cell lysates from vehicle- or sildenafil-treated LF- and HF-fed mice were analyzed for phosphorylation of I�B-� and GAPDH by immunoblotting. Densitometric quantitation show-ing fold difference in phosphorylated (p-) I�B-� relative to GAPDH in epididymal white adipose tissue (EWAT) of vehicle- and sildenafil-treated LF and HF mice. *P�0.05 vehicle LF vs HF, †P�0.05 vehicle- vs sildenafil-treated HF-fed mice (n�3). C, Western blotsshowing Mac-2 and GAPDH in EWAT from vehicle- and sildenafil-treated HF-fed mice. Quantitation of Mac-2 was normalized toGAPDH. *P�0.05 vehicle- vs sildenafil-treated HF fed mice.




endothelial cells. In adipose tissue of lean animals, a healthyhomeostasis exists among these cell types. HF feeding (oreNOS/VASP deletion) can disrupt this homeostasis by exert-ing proinflammatory effects that, in turn, promote macro-phage recruitment into adipose tissue. Our finding thatmacrophage accumulation in adipose tissue is increased in

mice lacking eNOS/VASP, even when fed an LF diet,whereas it is reduced by sildenafil treatment in HF-fed WTmice, suggests that obesity-induced macrophage influx re-quires a reduced NO signal.

Furthermore, in a severe model of obesity such as db/dbmouse model, adipose tissue inflammation and macrophage

Figure 5. Vasodilator-stimulated phosphoprotein (VASP) deficiency enhances adipose inflammation. A, Relative mRNA levels of macro-phage, inflammatory, and antiinflammatory markers from epididymal fat tissue from wild-type (WT) and VASP-deficient mice on low-fat(LF) and high-fat (HF) diets for 4 weeks. *P�0.05 WT vs VASP�/� mice (n�4–6). B, Protein levels of phosphorylated (p-) p65 subunitof nuclear factor-�B relative to �-actin. *P�0.05 WT vs VASP�/� on LF, †P�0.05 WT vs VASP�/� on HF (n�3–4). C, Protein levelsof Mac-2 protein relative to �-actin. *P�0.05 WT vs VASP�/� mice (n�3–4). TNF-� indicates tumor necrosis factor-�; IL, interleukin;CCR2, CC chemokine receptor 2; ARG-1, arginase-1.




infiltration are reduced by sildenafil (Supplemental FigureVB and VC). Sildenafil, however, did not reduce fastinginsulin levels (data not shown) or reduce glucose levels(Supplemental Figure VA). We recognize that the db/dbmouse represents a severe form of diabetes and that restora-tion of leptin signaling is actually required for “normaliza-tion” of global metabolic parameters.19 We speculate thatsildenafil’s effects are mainly limited to the NF- �B pathwayand sildenafil does not appear to “alter” the leptin signalingpathway.

In the vasculature, nitric oxide plays a crucial antiinflam-matory role by regulating the expression of the chemokineMCP-120 and adhesion molecules, such as intercellular adhe-sion molecule-1 and vascular cell adhesion molecule-1. Thesemolecules are important for adhesion of monocytes to thevessel wall, initiating the first step of atherosclerosis.7,9

Consequently, mice lacking eNOS show increased expressionof adhesion molecules and thereby exhibit increasedleukocyte-endothelial interactions and diet–induced athero-sclerosis.9 Recent reviews have drawn parallels between themechanisms regulating recruitment of monocytes to theartery wall and trafficking of macrophages to the adiposetissue suggesting that similar chemokine (MCP-1)/chemokinereceptor (CCR-2) systems may be involved in both pro-cesses.5,21 This hypothesis is in agreement with our findingsthat decreased vascular NO/cGMP signaling leads to in-creased expression of MCP-1 and CCR-2 in the EWAT andconsequent establishment of an inflammatory milieu in theadipose tissue.

The NO signaling cascade involves activation of solubleguanylyl cyclase, leading to the synthesis of cGMP, which, inturn activates the cGMP-dependent protein kinase (PKG).One of the targets of PKG is VASP, which plays an importantrole in cell migration and cytoskeletal dynamics.14 PKGpreferentially phosphorylates VASP at Ser239, and this effectis involved in the growth-inhibitory effects of NO on smoothmuscle cells.12 Downstream effects of VASP include regula-tion of actin polymerization and cell-cell contact in endothe-lial cells.22 In a rodent model of diabetes, reduced phosphor-ylation of VASP at Ser239 correlates with insulinresistance,13 but whether VASP signaling mediates the inhib-itory effect of NO/cGMP on NF-�B signaling had not beenpreviously explored. Our finding that deficiency in VASPpromotes proinflammatory M1 activation in the EWATsuggests that it plays an important role in antiinflammatoryeffects of vascular NO-cGMP signaling by attenuatingNF-�B signaling. However, based on our finding that VASPis expressed in several of the cell types present in adiposetissue, such as endothelial cells, macrophages, and adipocytes(data not shown), it is not possible to determine whether it isthe vascular, myeloid, or adipocyte-specific VASP that isimportant for the antiinflammatory effects. Only a cell typespecific ablation of VASP can definitively allow us todetermine that. Activation of M2 phenotype has been asso-ciated with IL-4/STAT6 pathway23; however, the associationbetween VASP and M2 pathway has yet to be fully explored.

The interaction between NO and inflammation is compli-cated because at high levels, as generated by inducible NOS,NO is cytotoxic, relative to the much lower (nmol/L) levels

produced by eNOS that activate the soluble guanylyl cyclase/cGMP/PKG pathway, which are cytoprotective.7 cGMP poolsare regulated not only by guanylyl cyclase activity but also bythe phosphodiesterase conversion of cGMP back to GMP.Sildenafil is a phosphodiesterase-5 inhibitor that inhibits thisconversion. Sildenafil’s effect in promoting signaling down-stream of NO has been useful not only in treating erectiledysfunction but also in treating pulmonary hypertension andcongestive heart failure. It may also augment ischemia-induced angiogenesis24 and immune regulation.25 Recentstudies have established a connection between cGMP signal-ing and insulin action, wherein chronic sildenafil treatmentleads to improved insulin sensitivity in mice fed a high-fatdiet.17 Furthermore, it attenuates endothelial dysfunction-induced diabetes in rat26 and humans.27 Targetingphosphodiesterase-5 by using sildenafil circumvents the pos-sible ill effects associated with excessive NO signaling, asproduced by inducible NOS or as happens even when eNOSis overexpressed in apolipoprotein E�/� mice.28 The dele-terious effects associated with eNOS overexpression areexplained by increased oxidative stress.29 This potentialpitfall can be avoided by targeting the downstream moleculesin the NO-cGMP signaling pathway, such as activation ofsoluble guanylyl cyclase or PKG. Consistent with this, it wasrecently shown that overexpression of constitutively activePKG protects female mice from the ill effects of diet-inducedobesity.30

eNOS-derived NO appears to be crucial for the productionof adequate endogenous levels of cGMP. In the absence ofenhanced pools of cGMP, sildenafil was unable to exertanti-inflammatory effects. This was reflected in the lack ofimproved phosphorylation of VASP in eNOS�/� micetreated with sildenafil and no reduction in markers of adiposeinflammation relative to controls (Supplemental Figure IV).These observations allowed us to draw 2 important conclu-sions: (1) eNOS-derived NO-cGMP is critical for sildenafil’santi-inflammatory effects, and (2) phosphorylation of VASPin response to enhanced cGMP stores is required to exertprotective effects of NO-cGMP. This suggests that VASPdoes not act independent of NO-cGMP, but needs enhancedcGMP signaling to alleviate adipose inflammation. Theseobservations are consistent with the reports that sildenafil hadno effect on erectile function in eNOS�/� mice.31

Taken together, our findings support a model in whichthe effect of diet-induced obesity to reduce vascularNO-cGMP signaling contributes to the mechanism under-lying adipose tissue inflammation. Therapeutic strategiesdesigned to increase vascular NO/cGMP signaling maytherefore be of value in prevention and treatment ofobesity-associated diseases. Finally, in these studies, weshow that deficiency of vascular nitric oxide primes theadipose tissue toward a proinflammatory state; that pharma-cological improvement of NO-cGMP signaling reverses adi-pose inflammation during HF feeding; and that eNOS-derived NO (and cGMP) is essential for VASPphosphorylation, which in turn is vital for sildenafil action,underlining the role of VASP as the effector of NO-cGMPsignaling that limits adipose tissue inflammation.




Sources of FundingThis study was supported by National Institutes of Health GrantsDK073878 (to F.K.), DK083042 and DK052989 (to M.W.S.),HL52459 (to A.W.C.), and HL94352 and HL92969 (to A.C.) and bygrants from the John L. Locke, Jr, Charitable Trust and from theKenneth H. Cooper Endowed Professorship in Preventive Cardiol-ogy (to F.K.).

DisclosuresNone.

References1. Olefsky JM, Glass CK. Macrophages, inflammation, and insulin resistance.

Annu Rev Physiol. 2010;72:219–246.2. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante

AW. Obesity is associated with macrophage accumulation in adiposetissue. J Clin Invest. 2003;112:1796–1808.

3. Lumeng CN, Bodzin JL, Saltiel AR. Obesity induces a phenotypic switchin adipose tissue macrophage polarization. J Clin Invest. 2007;117:175–184.

4. Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, Sole J, Nichols A,Ross JS, Tartaglia LA, Chen H. Chronic inflammation in fat plays acrucial role in the development of obesity-related insulin resistance.J Clin Invest. 2003;112:1821–1830.

5. Rocha VZ, Libby P. Obesity, inflammation, and atherosclerosis. Nat RevCardiol. 2009;6:399–409.

6. Rutkowski JM, Davis KE, Scherer PE. Mechanisms of obesity and relatedpathologies: the macro- and microcirculation of adipose tissue. FEBS J.2009;276:5738–5746.

7. Liu VW, Huang PL. Cardiovascular roles of nitric oxide: a review ofinsights from nitric oxide synthase gene disrupted mice. Cardiovasc Res.2008;77:19–29.

8. Kim F, Pham M, Maloney E, Rizzo NO, Morton GJ, Wisse BE, Kirk EA,Chait A, Schwartz MW. Vascular inflammation, insulin resistance, andreduced nitric oxide production precede the onset of peripheral insulinresistance. Arterioscler Thromb Vasc Biol. 2008;28:1982–1988.

9. Huang PL. eNOS, metabolic syndrome and cardiovascular disease.Trends Endocrinol Metab. 2009;20:295–302.

10. Nisoli E, Clementi E, Paolucci C, Cozzi V, Tonello C, Sciorati C, BracaleR, Valerio A, Francolini M, Moncada S, Carruba MO. Mitochondrialbiogenesis in mammals: the role of endogenous nitric oxide. Science.2003;299:896–899.

11. Yan H, Aziz E, Shillabeer G, Wong A, Shanghavi D, Kermouni A,Abdel-Hafez M, Lau DC. Nitric oxide promotes differentiation of ratwhite preadipocytes in culture. J Lipid Res. 2002;43:2123–2129.

12. Chen L, Daum G, Chitaley K, Coats SA, Bowen-Pope DF, Eigenthaler M,Thumati NR, Walter U, Clowes AW. Vasodilator-stimulated phospho-protein regulates proliferation and growth inhibition by nitric oxide invascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 2004;24:1403–1408.

13. Russo I, Del Mese P, Doronzo G, Mattiello L, Viretto M, Bosia A,Anfossi G, Trovati M. Resistance to the nitric oxide/cyclic guanosine5�-monophosphate/protein kinase G pathway in vascular smooth musclecells from the obese Zucker rat, a classical animal model of insulinresistance: role of oxidative stress. Endocrinology. 2008;149:1480–1489.

14. Trichet L, Sykes C, Plastino J. Relaxing the actin cytoskeleton foradhesion and movement with ENA/VASP. J Cell Biol. 2008;181:19–25.

15. Rizzo NO, Maloney E, Pham M, Luttrell I, Wessells H, Tateya S, DaumG, Handa P, Schwartz MW, Kim F. Reduced NO-cGMP signaling con-

tributes to vascular inflammation and insulin resistance induced byhigh-fat feeding. Arterioscler Thromb Vasc Biol. 2010;30:758–765.

16. Kirk EA, Sagawa ZK, McDonald TO, O’Brien KD, Heinecke JW.Monocyte chemoattractant protein deficiency fails to restrain macrophageinfiltration into adipose tissue [corrected]. Diabetes. 2008;57:1254–1261.

17. Ayala JE, Bracy DP, Julien BM, Rottman JN, Fueger PT, WassermanDH. Chronic treatment with sildenafil improves energy balance andinsulin action in high fat-fed conscious mice. Diabetes. 2007;56:1025–1033.

18. Oelze M, Mollnau H, Hoffmann N, Warnholtz A, Bodenschatz M, Smo-lenski A, Walter U, Skatchkov M, Meinertz T, Munzel T. Vasodilator-stimulated phosphoprotein serine 239 phosphorylation as a sensitivemonitor of defective nitric oxide/cGMP signaling and endothelial dys-function. Circ Res. 2000;87:999–1005.

19. Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, LakeyND, Culpepper J, Moore KJ, Breitbart RE, Duyk GM, Tepper RI, Mor-genstern JP. Evidence that the diabetes gene encodes the leptin receptor:identification of a mutation in the leptin receptor gene in db/db mice. Cell.1996;84:491–495.

20. Tsao PS, Wang B, Buitrago R, Shyy JY, Cooke JP. Nitric oxide regulatesmonocyte chemotactic protein-1. Circulation. 1997;96:934–940.

21. Surmi BK, Hasty AH. The role of chemokines in recruitment of immunecells to the artery wall and adipose tissue. Vascul Pharmacol. 2010;52:27–36.

22. Benz PM, Blume C, Moebius J, Oschatz C, Schuh K, Sickmann A, WalterU, Feller SM, Renne T. Cytoskeleton assembly at endothelial cell-cellcontacts is regulated by �II-spectrin-VASP complexes. J Cell Biol. 2008;180:205–219.

23. Odegaard JI, Chawla A. Alternative macrophage activation and metabo-lism. Annu Rev Pathol. 2011;6:275–297.

24. Senthilkumar A, Smith RD, Khitha J, Arora N, Veerareddy S, LangstonW, Chidlow JH, Barlow SC, Teng X, Patel RP, Lefer DJ, Kevil CG.Sildenafil promotes ischemia-induced angiogenesis through a PKG-dependent pathway. Arterioscler Thromb Vasc Biol. 2007;27:1947–1954.

25. Serafini P, Meckel K, Kelso M, Noonan K, Califano J, Koch W, DolcettiL, Bronte V, Borrello I. Phosphodiesterase-5 inhibition augments endog-enous antitumor immunity by reducing myeloid-derived suppressor cellfunction. J Exp Med. 2006;203:2691–2702.

26. Schafer A, Fraccarollo D, Pfortsch S, Flierl U, Vogt C, Pfrang J, KobsarA, Renne T, Eigenthaler M, Ertl G, Bauersachs J. Improvement ofvascular function by acute and chronic treatment with the PDE-5 inhibitorsildenafil in experimental diabetes mellitus. Br J Pharmacol. 2008;153:886–893.

27. Aversa A, Vitale C, Volterrani M, Fabbri A, Spera G, Fini M, RosanoGM. Chronic administration of sildenafil improves markers of endothelialfunction in men with type 2 diabetes. Diabet Med. 2008;25:37–44.

28. Ozaki M, Kawashima S, Yamashita T, Hirase T, Namiki M, Inoue N,Hirata K, Yasui H, Sakurai H, Yoshida Y, Masada M, Yokoyama M.Overexpression of endothelial nitric oxide synthase accelerates athero-sclerotic lesion formation in ApoE-deficient mice. J Clin Invest. 2002;110:331–340.

29. Takaya T, Hirata K, Yamashita T, Shinohara M, Sasaki N, Inoue N, YadaT, Goto M, Fukatsu A, Hayashi T, Alp NJ, Channon KM, Yokoyama M,Kawashima S. A specific role for eNOS-derived reactive oxygen speciesin atherosclerosis progression. Arterioscler Thromb Vasc Biol. 2007;27:1632–1637.

30. Nikolic DM, Li Y, Liu S, Wang S. Overexpression of constitutivelyactive PKG-I protects female, but not male mice from diet-inducedobesity. Obesity (Silver Spring). 2011;19:784–791.

31. Champion HC, Bivalacqua TJ, Takimoto E, Kass DA, Burnett AL.Phosphodiesterase-5a dysregulation in penile erectile tissue is amechanism of priapism. Proc Natl Acad Sci U S A. 2005;102:1661–1666.




Schwartz, Alan Chait and Francis KimChang-Yeop Han, Alexander W. Clowes, Guenter Daum, Kevin D. O'Brien, Michael W.

Priya Handa, Sanshiro Tateya, Norma O. Rizzo, Andrew M. Cheng, Vicki Morgan-Stevenson,Inflammation During High-Fat Feeding

cGMP Signaling Contributes to Adipose Tissue−Reduced Vascular Nitric Oxide

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2011 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Arteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/ATVBAHA.111.2365542011;

2011;31:2827-2835; originally published online September 8,Arterioscler Thromb Vasc Biol.

http://atvb.ahajournals.org/content/31/12/2827World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://atvb.ahajournals.org/content/suppl/2011/09/08/ATVBAHA.111.236554.DC1.htmlData Supplement (unedited) at:

http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

http://www.lww.com/reprints

Information about reprints can be found online at: Reprints:

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theArteriosclerosis, Thrombosis, and Vascular Biologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


http://atvb.ahajournals.org/content/31/12/2827

http://atvb.ahajournals.org/content/suppl/2011/09/08/ATVBAHA.111.236554.DC1.html

http://www.ahajournals.org/site/rights/

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://atvb.ahajournals.org//subscriptions/


1

Supplemental Material:

Western blot:

SDS gel electrophoresis was performed using a 4% by 20% gradient gel8.

Quantification of Western blots was performed using Image J Processing and

Analysis (NIH).

Quantitative RT-PCR Analyses

RNA was extracted using RNAeasy kit (Qiagen). For gene expression analysis,

real-time RT-PCR reactions were conducted using TaqMan Gene Expression

Analysis as described previously8 (Applied Biosystems) and normalized to

GAPDH levels.

Nitric oxide measurement

NO content was measured in epididymal adipose tissue using Electron Spin

Resonance Spectroscopy technique as described previously8.

Mice: Male db/db and age-matched lean controls, db/+m (2–3 mo of age; Harlan)

were housed in a temperature-controlled room with a 12:12-h light-dark cycle and

maintained with access to food (rodent chow) and water ad libitum.

Insulin sensitivity testing:

Insulin sensitivity test was performed after an overnight fast. Mice were injected

intraperitoneally with human insulin (0.75 unit/kg body weight), after which blood

was collected from a tail vein and blood glucose levels were monitored 0, 15, 30,

60, 90 and 120 minutes after injection using glucometer (FreeStyle,

2

TheraSense).

Plasma Insulin and Blood glucose measurement:

Blood samples were placed on ice until separation of plasma by centrifugation.

Insulin levels were determined using an ELISA kit (Crystal Chem Inc), and

glucose levels were determined by glucometer .

Supplementary results:

Effect of Sildenafil on eNOS null mice:

To test the hypothesis that eNOS-derived NO (and cGMP) is required for the

protective effect of sildenafil in the adipose tissue, 8 week old eNOS-/- mice on a

low fat diet were administered either vehicle or sildenafil (30 mg/Kg/day) for 4

weeks. At the end of 4 weeks, we assessed the metabolic parameters

(supplementary figure IVA), and found that there was no effect on body weight

and fasting plasma glucose. Further, in the adipose tissue in eNOS-/- mice that

had received sildenafil or vehicle, we found that there was no difference in the

sildenafil treated animals relative to controls at the level of NF-!B activation as

assessed by phosphorylation of p65 subunit or levels of mac-2 protein. Further,

there was no increase in phospho-VASP protein levels. The

proinflammatory/macrophage mRNA expression (TNF-", MCP-1, IL-6, CCR-2,

CD11c and F480) was found to be unchanged between the sildenafil and vehicle

treated animals (supplementary figure IVB &C). Taken together, these

observations suggest that eNOS-derived NO-cGMP signaling is crucial for the

3

anti-inflammatory effects of sildenafil exerted by phosphorylation of VASP.

Effect of sildenafil on db/db mice:

We asked if sildenafil is able to protect a model of genetic obesity and

insulin resistance such as db/db mice. We treated 12 week old db/db obese mice

and db/+m, their lean controls, on chow, with 30 mg/kg/day sildenafil for 4 weeks.

At the end of the study period, we assessed the metabolic parameters such as

weight and fasting glucose, and found no differences between the vehicle and

sildenafil treated mice (supplementary figure VA). We also performed insulin

tolerance test and found that there was no effect of sildenafil (data not shown).

We next examined the effect on adipose tissue inflammation. Relative to db/+m

lean mice, the db/db mice showed extensive inflammation as assessed by mRNA

expression of inflammatory genes: TNF-", IL-6, MCP-1 and inflammatory

myeloid markers such as CD11b, CD11c, CCR2, CD68 and F480, and protein

levels of p-p65 subunit of NF-!B and Mac-2. While there was no appreciable

effect of the sildenafil on the db/+m mice, there was a significant decrease in

mRNA expression levels of inflammatory and immune cell markers

(Supplementary figure VB), and also phosphorylation of p65 subunit of NF-!B

and mac-2 protein levels in db/db mice with sildenafil treatment relative to vehicle

treated db/db mice (Supplementary figure VC). Further, we find that the levels of

phospho-VASP at ser-239 were significantly elevated in the sildenafil treated

db/db mice compared to vehicle treated db/db mice (Supplementary figure VC),

suggesting that the anti-inflammatory effects of sildenafil were mediated by

4

activation of the NO-cGMP-VASP pathway.

Supplementary Figure legends:

Supplementary figure I: Densitometric quantitation of eNOS or VASP from LF

and HF diet fed mice: A. eNOS levels were compared to #-actin , B. VASP levels

were compared to #-actin (n=3-4).

Supplementary Figure II: Glucose homeostasis in high-fat fed vehicle

versus sildenafil treated mice. A: Insulin sensitivity test: B. % drop of the initial

glucose. Vehicle treated (black circles) and sildenafil treated (black squares)

animals on 14 weeks of high fat diet are shown. IP denotes intraperitoneal

injection. *P<0.05 vehicle vs. sildenafil treated high-fat fed mice. (n=4-5 per

group).

Supplementary figure III. Effect of sildenafil on phosphorylation of VASP at

Ser-239 on mice fed LF and HF for 14 weeks: Cell lysates from vehicle or

sildenafil treated LF and HF-fed mice were analysed for phosphorylation of VASP

by immunoblotting. Densitometric quantitation showing fold difference in

phosphorylated VASP relative to GAPDH in EWAT of vehicle and sildenafil

treated LF and HF mice. *p<0.05 vehicle LF vs. HF. † P<0.05 vehicle vs.

sildenafil treated HF-fed mice.

Supplementary figure IV: Effect of sildenafil on eNOS null mice: A. Metabolic

5

parameters were assessed. Weight of eNOS null mice treated with vehicle or

sildenafil. (n=3-6 each). Fasting blood glucose levels of eNOS null mice treated

with vehicle and sildenafil. (n=5-6 each). B. mRNA expression analysis of the

eNOS null mice administered either vehicle or sildenafil. Genes such as TNF-",

MCP-1, IL-6, CCR-2, CD11c and F480 were assessed. (n=5-6 each). C. Protein

analysis of the eNOS deficient mice treated with vehicle or sildenafil. i.

phosphorylated-p65 subunit of NF-!B. ii. Mac-2, iii. phosphorylated-VASP

(Ser239). Densitometric quantitation of proteins relative to GAPDH are shown

(n=5-6 each).

Supplementary figure V: Effect of administration of sildenafil or vehicle on

db/+m and db/db mice: A. Metabolic parameters were assessed: Weight (n=6,

db/db mice, n=3, db/+m) fasting blood glucose (n=6, db/db mice, n=3, db/+m). *

p<0.05, significant difference between db/+m and db/db vehicle treated animals.

B. mRNA expression analysis of the db/+m and db/db animals, such as TNF-",

MCP-1, IL-6, CCR-2, CD68, F480, CD11c and CD11b. p<0.05 *, significant

difference between db/+m and db/db vehicle treated animals, p<0.05 †,

significant difference between db/db vehicle treated animals and db/db sildenafil

treated mice. (n=6 each db/db mice, n=3 each db/+m). C. Protein analysis of the

db/db mice treated with vehicle or sildenafil. i. phosphorylated-p65 subunit of NF-

!B, ii. Mac-2, iii. phosphorylated-VASP (Ser239). Densitometric quantitation of

proteins relative to GAPDH are shown (n=4 each). p<0.05 *, significant difference

between db/db mice treated with sildenafil and db/db animals treated with

6

vehicle.

Supplementary figure I: Densitometric quantitation of eNOS or VASP from LF

and HF diet fed mice: A. eNOS levels were compared to !-actin , B. VASP

levels were compared to !-actin. p values are shown.

A. B.

0.0

0.5

1.0

1.5

Rela

tive V

AS

P levelsp=0.0692 p=0.75

LF HF

A. B.

!"##$%&%'()*+,-."*%,//,

Supplementary Figure II: Glucose homeostasis in high-fat fed vehicle versus sildenafil treated mice. A: Insulin sensitivity test: B. % drop of the initial glucose.

Vehicle treated (black circles) and sildenafil treated (black squares) animals on 14 wk of high fat diet are shown. IP denotes intraperitoneal injection. *P<0.05 vehicle vs. sildenafil

treated high-fat fed mice. (n=4-5 per group).

!" !"

!"

!" "!"

!"##$%&%'()*+,-."*%,///,

!" 0,

#123!4567,#*8(%9',

Supplementary figure III. Effect of sildenafil on phosphorylation of VASP at Ser-239 on mice fed LF and HF for 14 wk: Cell lysates from vehicle or sildenafil treated LF and

HF-fed mice were analysed for phosphorylation of VASP by immunoblotting. Densitometric quantitation showing fold difference in phosphorylated VASP relative to

GAPDH in EWAT of vehicle and sildenafil treated LF and HF mice. *p<0.05 vehicle LF vs. HF. † P<0.05 vehicle vs. sildenafil treated HF-fed mice(n=3).

Supplementary IV

A.

B.

C.

Supplementary figure IV: Effect of sildenafil on eNOS null mice: A. Metabolic parameters were assessed. Weight of eNOS null mice treated with vehicle or

sildenafil. (n=3-6 each). Fasting blood glucose levels of eNOS null mice treated with vehicle and sildenafil. (n=5-6 each). B. mRNA expression analysis of the eNOS null mice

administered either vehicle or sildenafil. Genes such as TNF- !, MCP-1, IL-6, CCR-2, CD11c and F480 were assessed. (n=5-6 each). C. Protein analysis of the eNOS deficient mice treated with vehicle or sildenafil. i. phosphorylated-p65 subunit of NF- "B. ii. Mac-2, iii.

phosphorylated-VASP (Ser239). Densitometric quantitation of proteins relative to GAPDH are shown (n=5-6 each).

!"##$%&%'()*+,-,

.,.,

.,

!"#

$"#

%"#

&#############&#############'# &#############'# &#############'#

&#############'#&############'#

Supplementary figure V: Effect of administration of sildenafil or vehicle for 4 weeks on db/+m and db/db mice: A. Metabolic parameters were assessed: Weight (n=6, db/db

mice, n=3, db/+m) fasting blood glucose (n=6, db/db mice, n=3, db/+m). * p<0.05, significant difference between db/+m and db/db vehicle treated animals. B. mRNA expression analysis

of the db/+m and db/db animals, such as TNF- !, MCP-1, IL-6, CCR-2, CD68, F480, CD11c and CD11b. p<0.05 *, significant difference between db/+m and db/db vehicle treated animals, p<0.05 †, significant difference between db/db vehicle treated animals and db/db

sildenafil treated mice. (n=6 each db/db mice, n=3 each db/+m). C. Protein analysis of the db/db mice treated with vehicle or sildenafil. i. phosphorylated-p65 subunit of NF- "B, ii.

Mac-2, iii. phosphorylated-VASP (Ser239). Densitometric quantitation of proteins relative to GAPDH are shown (n=4 each). p<0.05 *, significant difference between db/db mice treated with sildenafil and db/db animals treated with vehicle.

&#############'#&#############'#

'#

., .,

reduced vascular nitric oxide-cgmp signaling contributes to adipose tissue inflammation during...

Documents