Basic Science

Toll-Like Receptor 4 Mutation Protects Obese Mice Against Endothelial Dysfunction by Decreasing NADPH Oxidase Isoforms 1 and 4Significance

Chao-Fan Liang, Jacky TC Liu, Yu Wang, Aimin Xu, Paul M. Vanhoutte
https://doi.org/10.1161/ATVBAHA.112.301087
Arteriosclerosis, Thrombosis, and Vascular Biology. 2013;33:777-784
Originally published March 13, 2013

Jump to

Abstract

Objective—To analyze the role of toll-like receptor 4 in modulating metabolism and endothelial function.

Approach and Results—Type 2 diabetic mice with mutated toll-like receptor 4 (DWM) were protected from hyperglycemia and hypertension, despite an increased body weight. Isometric tension was measured in arterial rings with endothelium. Relaxations to acetylcholine were blunted in aortae and mesenteric arteries of Leprdb/db mice, but not in DWM mice; the endothelial NO synthase dimer/monomer ratio and endothelial NO synthase phosphorylation levels were higher in DWM preparations. These differences were abolished by apocynin. Contractions to acetylcholine (in the presence of L-NAME) were larger in carotid arteries from Leprdb/db mice than from DWM mice and were inhibited by indomethacin and SC560, demonstrating involvement of cyclooxygenase-1. The release of 6-ketoprostaglandin F1α was lower in DWM mice arteries, implying lower cyclooxygenase-1 activity. Apocynin, manganese(III) tetrakis(1-methyl-4-pyridyl) porphyrin, catalase, and diethyldithiocarbamate inhibited endothelium-dependent contractions. The mRNA and protein levels of NADPH oxidase isoforms NOX1 and NOX4 were downregulated in DWM mice arteries. The in vivo and in vitro administration of lipopolysaccharide caused endothelial dysfunction in the arteries of wild-type, but not toll-like receptor 4–mutated mice.

Conclusions—Toll-like receptor 4 plays a key role in obesity and diabetes-associated endothelial dysfunction by increasing oxidative stress.

Introduction

Obesity and diabetes mellitus are associated with augmented circulating levels of free fatty acids, low-grade chronic inflammation,1 and endothelial dysfunction attributable to impaired release of endothelium-derived relaxing and augmented production of endothelium-derived contracting (EDCF) factors.2 Thus, the bioavailability of NO is reduced by obesity and diabetes mellitus,3 and relaxations attributable to endothelium-derived hyperpolarization (EDH) are impaired at the early stage of diabetes mellitus.4 Finally, increased release of endothelium-derived vasoconstrictor prostanoids and reactive oxygen species (ROS) also contributes to endothelial dysfunction in obesity and diabetes mellitus.5

Proinflammatory cytokines (tumor necrosis factor-α [TNF-α], interleukin 6 [IL-6], resistin, and lipocalin-2) link obesity to metabolic and vascular dysfunction.1 Bacterial endotoxin lipopolysaccharide (LPS) and saturated fatty acids share as pattern recognition target toll-like receptors 4 (TLR4), which activate inflammatory pathways, induce cytokine expression in the endothelium,6,7 and augment the expression of NADPH oxidase in macrophages.8 The absence of TLR4 protects mice against obesity-induced insulin resistance.7

NADPH oxidase is a major source of endothelial ROS.9,10 ROS decrease the bioavailability of NO11 and facilitate the production of EDCF.12

It is unknown whether or not TLR4 signaling contributes to the impaired NO- and EDHF-mediated relaxations and the amplified EDCF-mediated contractions seen in obesity and diabetes mellitus. Therefore, the present experiments tested the hypothesis that TLR4 signaling mediates, at least in part, the deleterious effects of diet-induced and genetic obesity leading to endothelial dysfunction.

Materials and Methods

Materials and Methods are available in the online-only Supplement.

Results

General Characteristics

There were no significant differences in food intake between wild-type (WT) and TLR4−/− mice (Figure IA in the online-only Data Supplement). The high-fat diet evoked comparable increases in body weight in 12-week-old WT and TLR4−/− mice (Figure IB in the online-only Data Supplement). In DWM mice, the food intake and the body weight were higher starting at the age of 8 weeks (Figure IC in the online-only Data Supplement) compared with Leprdb/db mice. Both feeding (Figure IC in the online-only Data Supplement) and fasting (Figure ID in the online-only Data Supplement) glucose levels of DWM mice were significantly lower than those of Leprdb/db mice. After injection of glucose, the glycemia rose more slowly and returned back to normal more quickly in DWM mice than in Leprdb/db mice (Figure IE and IF in the online-only Data Supplement). After injection of insulin, blood glucose dropped more quickly in DWM mice than in Leprdb/db mice (Figure IG and IH in the online-only Data Supplement).

The mRNA expressions of IL1-β, IL-6, and TNF-α were significantly lower in the epididymal fat of DWM mice compared with Leprdb/db mice (Figure II in the online-only Data Supplement).

Systolic arterial blood pressures were comparable between 16-week-old lean WT and TLR4−/− mice. WT mice were fed high-fat diet, and Leprdb/db mice fed standard chow exhibited an elevated systolic arterial blood pressure (Figure IIIA in the online-only Data Supplement). DWM mice showed significantly lower systolic arterial blood pressures than Leprdb/db mice (Figure IIIA in the online-only Data Supplement).

Vascular Responsiveness

Mutation of TLR4

Relaxations to increasing concentrations of acetylcholine (10−10 to 10−4 mol/L) were reduced significantly in aortae of 16-week-old WT mice compared with TLR4−/− preparations (Figure 1).

Figure 1.
Figure 1.

Endothelium-dependent responses in isolated aortae of toll-like receptor (TLR)4+/+ and TLR4−/− mice. Concentration-response curves to acetylcholine in arteries with endothelium of 16-week-old mice fed normal (top) or high-fat (bottom) diet. Left: Relaxations obtained in control solution during contractions to U46619 (1–3×10−8 mol/L) in aortic rings. Center: Relaxations obtained in the presence of L-NAME (10−4 mol/L) and indomethacin (10−5 mol/L) during contractions to U46619 in mesenteric arterial rings. Right: Contractions obtained in the presence of L-NAME (10−4 mol/L) in quiescent carotid arterial rings. Relaxations are expressed as percentage of contraction to U46619. Contractions are expressed as the percentage of a reference contraction to 60 mmol/L KCl obtained at the beginning of the experiment; *P<0.05 vs respective control; n=4 to 6.

Increasing concentrations of acetylcholine (in the presence of L-NAME [10−4 mol/L; nonselective NOS inhibitor] and indomethacin [10−5 mol/L; nonselective cyclooxygenase inhibitor] to assess responses attributable to EDH)26 caused relaxations in mesenteric arteries, which were significantly larger in preparations of TLR4−/− than in those of control mice (Figure 1).

In the presence of L-NAME (10−4 mol/L, to optimize endothelium-dependent contractions),18 acetylcholine induced increases in tension in the quiescent carotid arteries of WT and TLR4−/− mice (Figure 1). These contractions were significantly attenuated in TLR4−/− mice arteries.

High-Fat Diet

After 12 weeks of high-fat diet, relaxations to acetylcholine (10−10 to 10−4 mol/L) were significantly impaired in aortae or mesenteric arteries (in the presence of L-NAME and indomethacin) of WT mice (Figure 1). The high-fat diet did not significantly affect relaxations to acetylcholine in either aortae or mesenteric arteries of TLR4−/− mice (Figure 1).

Contractions induced by acetylcholine in the presence of L-NAME were enhanced significantly in WT carotid arteries after high-fat feeding (Figure 1).

Double Mutation of Leptin Receptor and TLR4

Vascular Smooth Muscle

Contractions to 60 mmol/L KCl were not significantly different in the aortae of 12-week-old Leprdb/db and DWM mice (1.83±0.49 and 1.79±0.36 g, respectively). Likewise, contractions to increasing concentration of U46619 were not significantly different in Leprdb/db and DWM mice aortae without endothelium (Figure IIIC in the online-only Data Supplement).

Relaxations to increasing concentrations of sodium nitroprusside were not significantly different between Leprdb/db and DWM mice aortae (Figure IIIB in the online-only Data Supplement).

NO-Dependent Relaxations and eNOS Coupling

The relaxations to acetylcholine (10−10 to 10−4 mol/L) were significantly larger in DWM compared with Leprdb/db mice aortae (Figure 2A). After incubation with apocynin (10−4 mol/L; antioxidant), the relaxations to acetylcholine were potentiated only in Leprdb/db mice aortae, and the difference in relaxations between the 2 strains was no longer significant (Figure 2A). The relaxations to acetylcholine were abolished by L-NAME (10−4 mol/L) in the aortae of both strains (Figure IIIE in the online-only Data Supplement).

Figure 2.
Figure 2.

NO bioavailability and endothelial NO synthase (eNOS) in Leprdb/db and Type 2 diabetic mice with mutated toll-like receptor 4 (DWM) mice aortae. A, Responses to acetylcholine in the absence and presence of apocynin (10−4 mol/L) in aortae with endothelium of 12-week-old Leprdb/db and DWM mice. B, Left: Representative Western blots of basal and acetylcholine- (10−6 mol/L) evoked eNOS phosphorylation at Ser 1177, and total eNOS levels; Right: p-eNOS expression levels normalized to the abundance of eNOS expression. C, Left: Representative Western blots of the eNOS dimer and monomer protein expression levels; Right: Mean data of the eNOS dimer/monomer ratios. D, Left: Representative Western blots showing the effect of apocynin on eNOS dimer and monomer expression levels; Right: Mean data of the eNOS dimer/monomer ratios. *P<0.05 vs basal conditions with same treatment in Leprdb/db mice preparations; #P<0.05 vs respective controls under control conditions in Leprdb/db mice arteries; %P<0.05 Leprdb/db mice vs DWM mice. n=4 to 6.

The basal and the acetylcholine-evoked phosphorylation of endothelial NO synthase (eNOS) at Serine 1177 was significantly higher in DWM than in Leprdb/db mice aortae (Figure 2B). Apocynin enhanced the acetylcholine-evoked phosphorylation of eNOS only in Leprdb/db preparations. The expression of total eNOS was comparable in Leprdb/db and DWM preparations and not affected by apocynin (Figure 2B).

The eNOS dimer/monomer ratio was significantly higher in DWM than in Leprdb/db mice aortae (Figure 2C). Apocynin increased the eNOS dimer levels and decreased those of eNOS monomer in Leprdb/db aortae to comparable levels as those observed in DWM preparations (Figure 2D). The eNOS dimer and monomer levels were not altered significantly by apocynin in DWM preparations (Figure 2D).

EDH-Mediated Relaxations

After incubation with L-NAME (10−4 mol/L) and indomethacin (5×10−6 mol/L), acetylcholine caused concentration-dependent relaxations of mesenteric arteries, which were larger in DWM than in Leprdb/db preparations (Figure IV in the online-only Data Supplement). In both cases, the relaxations were attenuated significantly by either 1-[(2-Chlorophenyl)diphenylmethyl]-1H-pyrazole (10−7 mol/L; intermediate-conductance calcium-activated potassium channels channel blocker) or 6,12,19,20,25,26-hexahydro- 5,27:13,18:21,24-Trietheno-11,7-metheno-7H-dibenzo[b,m][1,5,12,16]tetraazacyclotricosine-5,13-diium ditrifluoroacetate hydrate (10−7 mol/L; small-conductance calcium-activated potassium channels channel blocker) and nearly abolished by the combination of these 2 inhibitors (Figure IV in the online-only Data Supplement). Incubation with apocynin potentiated the relaxations in Leprdb/db, but not in DWM mice arteries, and in the presence of the antioxidant, the relaxations were no longer significantly different between the 2 types of preparations (Figure IV in the online-only Data Supplement).

Endothelium-Dependent Contractions

In the presence of L-NAME, acetylcholine evoked concentration-dependent contractions in quiescent carotid arteries of 12-week-old Leprdb/db mice (Figure 3A). The increases in tension caused by acetylcholine were significantly smaller in DWM preparations (Figure 3A). The contractions were abolished by apocynin (Figure 3A and 3C). Removal of the endothelium, as well as incubation with indomethacin, SC560 (3×10−7 mol/L; selective cyclooxygenase [COX]-1 inhibitor) and S18886 (10−7 mol/L; thromboxane-prostanoids receptor antagonist), but not NS398 (10−6 mol/L; preferential COX-2 inhibitor), significantly inhibited the acetylcholine-induced contractions (Figure V in the online-only Data Supplement). Diphenyliodium (10−5 mol/L; flavoenzyme oxidoreductases inhibitor), diethyldithiocarbamate (10−4 mol/L; superoxide dismutase inhibitor), catalase-polyethylene glycol (1200 U mL−1), and manganese(III) tetrakis(1-methyl-4-pyridyl) porphyrin (10−4 mol/L; cell permeable superoxide dismutase mimetic) significantly decreased the endothelium-dependent contractions to acetylcholine in both type of preparations, whereas superoxide dismutase (120UmL−1) and deferoxamine (10−4 mol/L; inhibitor of formation of hydroxyl radicals) had no significant effect (Figure 3C).

The mRNA expression and protein levels of COX-1 were comparable in Leprdb/db and DWM mice carotid arteries (Figure 3B).

The production of 6-ketoPGF1α under basal conditions and during stimulation with acetylcholine (10−6 mol/L) was significantly lower in DWM than in Leprdb/db mice carotid arteries (Figure 4A). Incubation with apocynin abolished the difference in the production of 6-ketoPGF1α (Figure 4A).

Figure 3.
Figure 3.

Endothelium-dependent contractions and cyclooxygenase-1 (COX-1) expression in carotid arteries of Leprdb/db and DWM mice. A, Cumulative concentration-responses curve to acetylcholine in quiescent carotid arteries obtained after incubation with L-NAME (10−4 mol/L), in the presence or absence of apocynin (10−4 mol/L). Data are expressed as the percentage of the reference contraction to KCl obtained at the beginning of the experiment. B, mRNA (bottom) and protein (top) expressions of COX-1 in carotid arteries (with endothelium) of Leprdb/db and DWM mice assayed by real time–polymerase chain reaction and Western blotting, respectively. C, Effect of antioxidants including apocynin (10−4 mol/L), diphenyliodium (DPI, 10−6 mol/L), diethyldithiocarbamate (DETCA, 10−4 mol/L), superoxide dismutase (SOD, 120 U mL−1), catalase-monomethoxypolyethylene glycol (catalase-PEG, 1200 U mL−1), manganese(III) tetrakis(1-methyl-4-pyridyl) porphyrin (MnTMPyP, 10−4 mol/L), and deferoxamine (10−4 mol/L) on contractions to acetylcholine (shown as areas under the curve); *P<0.05 vs respective control as indicated; #P<0.05 vs controls under basal conditions, n=6.

Figure 4.
Figure 4.

Production of prostanoids and reactive oxygen species and expression of NADPH oxidase in the carotid arteries of Leprdb/db and DWM mice. A, Basal and acetylcholine- (10−6 mol/L, 3 minutes) induced production of 6-ketoprostaglandin F1α (PGF1α), in the absence or presence of apocynin (10−4 mol/L). B, Effect of H2O2 (10−5 mol/L) on the production of 6-ketoPGF1α. C, mRNA levels of NADPH oxidase isoform 1 (NOX1) and 4 (NOX4) in the carotid arteries of Leprdb/db and DWM mice at 12 weeks. D, Representative Western Blots showing NADPH oxidase isoform 1 (NOX1) and 4 (NOX4) in the carotid arteries of Leprdb/db and DWM mice at 12 weeks. E, Membrane fractions of 12-week-old Leprdb/db and DWM mice carotid arteries were suspended in PBS, and NADPH oxidase activity was measured by lucigenin-chemiluminescence assay. Basal and acetylcholine (10−6 mol/L) -stimulated production of superoxide anions measured in the absence and presence of indomethacin (10−5 mol/L). Data are expressed as areas under the curve. *P<0.05 vs respective controls; #P<0.05 vs controls under basal conditions; n=4 to 6.

NADPH Oxidase and Superoxide Anion Production

The mRNA (Figure 4C) and protein (Figure 4D) expressions of NADPH oxidase isoforms 1 and 4 were significantly lower in DWM than in Leprdb/db mice carotid arteries.

The extracellular free radical presence, estimated by lucigenin-enhanced chemiluminescence, was lower under basal conditions in membrane fractions of DWM than in Leprdb/db preparations (Figure 4E). Acetylcholine increased the ROS production in preparations of Leprdb/db, but not of DWM mice, but this response was not affected significantly by indomethacin (Figure 4E).

LPS Administration

The glucose and insulin tolerances were comparable in WT mice 4 hours after the in vivo intraperitoneal injection of either vehicle or LPS (25 mg/kg; Figure VI in the online-only Data Supplement).

NO-Dependent Relaxations

The acetylcholine-induced concentration-dependent relaxations were not significantly different between the aortae of 8-week-old WT and TLR4−/− mice (Figure 5). The in vivo injection of LPS significantly attenuated the relaxations to acetylcholine in WT, but not in TLR4−/− preparations (Figure 5). The acetylcholine-induced relaxations in WT aortae were normalized by apocynin (10−4 mol/L), which did not significantly affect the response in TLR4−/− preparations (Figure 5). The injection of LPS did not significantly affect the relaxations to sodium nitroprusside (Figure VII in the online-only Data Supplement).

Figure 5.
Figure 5.

Effect of in vivo lipopolysaccharide (LPS) administration on ex vivo endothelium-dependent responses to acetylcholine in 8-week-old toll-like receptor (TLR)4+/+ and TLR4−/− mice. Wild-type and TLR4-mutated mice were injected with either vehicle or lipopolysaccharide (LPS, 25 mg/kg) 4 hours before harvesting the preparations. Responses of arterial rings with endothelium to increasing concentrations of acetylcholine were obtained ex vivo in the absence or presence of apocynin (10−4 mol/L, 30 minutes, in vitro). Top, relaxations of aortae during contractions to U46619 (1–3×10−8 mol/L); center, relaxations of mesenteric arteries during contractions to U46619 after incubation with L-NAME (10−4 mol/L) and indomethacin (5×10−6 mol/L); bottom, contractions in quiescent carotid arteries incubated with L-NAME. Relaxations are expressed as percentage of the contraction to U46619, and increases in tension are expressed as the percentage of a reference contraction to 60 mmol/L KCl obtained at the beginning of the experiment; *P<0.05 vs respective controls; n=4 to 6.

In vitro incubation with LPS (10−6 mol/L for 4 hours) significantly attenuated the relaxations to acetylcholine in WT, but not in TLR4−/− aortae (Figure 6).

Figure 6.
Figure 6.

Effects of in vitro incubation with lipopolysaccharide (LPS). Endothelium-dependent responses in arteries of 8-week-old wild-type and toll-like receptor (TLR)4−/− mice, before or after incubation with LPS (10−6 mol/L for 4 hours). Left, relaxations to acetylcholine during contractions to U46619 (1–3×10−8 mol/L) in aortae; center, relaxations to acetylcholine during contractions to U46619 in mesenteric arteries after incubation with L-NAME (10−4 mol/L) and indomethacin (5×10−6 mol/L); right, contractions to acetylcholine in the presence of L-NAME (10−4 mol/L) in quiescent carotid arteries. Data expressed as areas under the curve; *P<0.05 vs vehicle or control in wild-type mice preparations; n=6 to 8.

EDH-Mediated Relaxations

In mesenteric arteries (in the presence of L-NAME and indomethacin) of 8-week-old WT and TLR4−/− mice, acetylcholine induced comparable concentration-dependent and endothelium-dependent relaxations (Figure 5). The in vivo injection of LPS reduced the relaxations to acetylcholine in WT, but not in TLR4−/− arteries (Figure 5). The in vitro administration of apocynin prevented this inhibitory effect of LPS (Figure 5).

In vitro incubation with LPS significantly reduced the relaxations to acetylcholine in WT, but not in TLR4−/− arteries (Figure 6).

Endothelium-Dependent Contractions

In the presence of L-NAME (10−4 mol/L), acetylcholine evoked comparable and minimal contractions in rings with endothelium of carotid arteries of lean WT mice and TLR4−/− mice (Figure 5). After the in vivo injection of LPS, it induced significantly larger increases in tension in WT, but not in TLR4−/− preparations, which were inhibited by apocynin (Figure 5). However, in vivo injection of LPS did not significantly affect the contractions to the thromboxane-prostanoids receptor agonist U46619 (Figure VIII in the Supplement).

After in vitro incubation with LPS, acetylcholine induced significantly larger increases in tension in WT preparations only (Figure 6).

Discussion

The current experiments were undertaken to determine whether or not TLR4, a major target for the initiation of inflammatory responses,13 contributes to the endothelial dysfunction resulting from diet-induced obesity and diabetes mellitus. Responses of arteries from WT mice were compared with those obtained in preparations from mice with mutated, nonfunctional TLR4, subjected to a high-fat diet. Furthermore, a unique animal model was created, with combined nonfunctional mutation of the leptin and TLR4 receptors. Although these DWM mice displayed a significantly heavier body weight, they were protected from hypertension and hyperglycemia demonstrating that overweight, per se, does not cause these pathological changes. They can be attributed to the decreased production of cytokines illustrated by the lower expression of TNF-α, IL-1β, and IL-6 in the DWM mice adipose tissue. The lower plasma level of free fatty acids and cholesterol, also attributable to the TLR4 mutation,14 explained why these animals did not develop the equivalent of a metabolic syndrome. The enhanced insulin sensitivity in DWM mice directly illustrates the link between TLR4 and insulin resistance.

The present study confirms that high-fat diet14,15 and diabetes mellitus16 increase arterial blood pressure in mice and demonstrates that loss of function of TLR4 prevents obesity- and diabetes-induced hypertension but does not affect this parameter in animals receiving standard diet.

Three types of endothelium-dependent responses were investigated in the present study.

First, NO-mediated acetylcholine-induced relaxations were determined in aortae in which, endothelium-dependent relaxations are attributable exclusively to the production of NO by eNOS.17 The involvement of NO was confirmed by the abolition of the responses by L-NAME, a nonselective inhibitor of NO synthases. In the aortae of 16-week-old WT mice given high-fat diet, or those of Leprdb/db mice, relaxations to acetylcholine, but not those to sodium nitroprusside, were blunted, demonstrating endothelial dysfunction. However, such relaxations were not altered in preparations of TLR4−/− or DWM mice, suggesting that TLR4 loss-of-function protects against the reduction in NO bioavailability attributable to diet-induced obesity and diabetes mellitus. The latter reduction is attributable to TLR4-induced impairment of eNOS activation. This conclusion is supported by the Western blotting data in Leprdb/db mice demonstrating a reduced phosphorylation both under basal conditions and on stimulation with acetylcholine, and an increased eNOS monomer level, 2 measures of the activation of the enzyme.17 In mice, uncoupling of eNOS can be caused by hypercholesterolemia.18 The present study demonstrates that loss of function of TLR4 rescues diabetic obese mice from the impaired relaxations. The likely mechanism underlying this effect is a reduced superoxide anion production leading to prevention of eNOS uncoupling.19–21 This interpretation is supported by the observation that in Leprdb/db aortae the antioxidant apocynin, which scavenges superoxide anions,10 recoupled eNOS allowing phosphorylation of the enzyme on exposure to acetylcholine, thus potentiating relaxations to the muscarinic agonist.

Second, the involvement of endothelium-dependent hyperpolarization was assessed by determining the inhibitory effect of incubation with 1-[(2-Chlorophenyl)diphenylmethyl]-1H-pyrazole and 6,12,19,20,25,26-hexahydro- 5,27:13,18:21,24-Trietheno-11,7-metheno-7H-dibenzo[b,m][1,5,12,16]tetraazacyclotricosine-5,13-diium ditrifluoroacetate hydrate on relaxations to acetylcholine, obtained in the presence of L-NAME and indomethacin.22 The present findings confirm that EDH-mediated relaxations are impaired in arteries of obese and diabetic mice.2 Activation of TLR4 triggers the production of inflammatory cytokines.7 In arteries of diabetic mice, TNF-α deficiency reverses the blunting of EDH-mediated relaxations.4 Taken in conjunction with the reduced level of TNF-α, IL-1β, and IL-6 observed in the adipose tissue of DWM mice, the prevention of the impairment of EDH-mediated relaxations observed in their arteries illustrates that TLR4 signaling contributes to this aspect of endothelial dysfunction resulting from obesity and diabetes mellitus.

An elevated glucose level inhibits acetylcholine-induced, EDH-mediated responses in the rat mesenteric artery because of the increased production of ROS.23 The administration of the antioxidant apocynin prevented the reduction in EDH-mediated relaxation, suggesting that the impaired response in arteries with functional TLR4 signaling also results from an increased ROS production.

Third, the aberrant production of vasoactive factors causing endothelial dysfunction includes the increased release of EDCFs from endothelial cells.24 In the mouse, EDCF-mediated responses are most pronounced in the carotid artery,25 and therefore this preparation was selected to examine endothelium-dependent contractions. Various arachidonic acid metabolites, including thromboxane A2, prostaglandin F2α, and prostacyclin, contribute to endothelium-dependent contractions that are ultimately attributable to activation of thromboxane-prostanoids receptors of the vascular smooth muscle.26 In the present study, the levels of 6-ketoPGF1α, the stable metabolite of prostacyclin, were measured to estimate the production of EDCF.24,26,27

The findings that the endothelium-dependent contractions to acetylcholine were abolished by indomethacin and a thromboxane-prostanoids receptor antagonist (S18886) demonstrate an EDCF-mediated response.24,26,27 Augmented endothelium-dependent contractions were observed in carotid arteries of older or high-fat fed WT and Leprdb/db mice, but not in TLR4−/− and DWM preparations, implying an amplification of the phenomenon by functional TLR4. The acetylcholine-induced endothelium-dependent contractions were blocked by a selective COX-1, but not by a COX-2 inhibitor, confirming the key role of COX-1 in EDCF-mediated responses in mice.28 TLR4 mutation did not significantly affect the COX-1 expression at either the mRNA or protein levels, implying that it does not modulate the presence of the enzyme. However, because the production of 6-ketoPGF1α was increased in Leprdb/db arteries, TLR4 must upregulate COX-1 activity. The antioxidant apocynin reduced the production of 6-ketoPGF1α, and hence the endothelium-dependent contractions,24,26,27 suggesting that ROS play a key role in activating cyclooxygenase to generate prostanoids when TLR4 is functional. Thus, the reduced endothelium-dependent contractions in the carotid arteries of the DWM mice are consistent with the lower production of ROS suggested by the lucigenin-chemiluminescence measurements in isolated membrane preparations of these blood vessels.28,29 In small mesenteric arteries of diabetic mice, the enhanced contractile activity is also associated with increased oxidative stress and cyclooxygenase production.16 The lower ROS production in DWM mice arteries is accompanied by a lesser mRNA expression and protein presence of NOX1 and NOX4, the major isoforms of NADPH oxidase30 that contribute to oxidative stress in the vascular wall. LPS, the prototypical agonist of TLR4, stimulates the production of superoxide anions in cultured endothelial cells, and this response is reduced by silencing RNAs for either NOX1 or NOX4 (Figure IX in the online-only Data Supplement). These observations indirectly support the interpretation that the decreased ROS production in DWM mice arteries may well result from a lesser activity of NADPH oxidase.

The finding that the endothelium-dependent contractions were prevented by the antioxidant apocynin5 and the inhibitor of flavoenzyme oxidoreductases (including NADPH oxidase)30 diphenyliodium31 is consistent with the key initial role of ROS in the EDCF-mediated contraction. The inhibitory effects of the cell permeable agents, diethyldithiocarbamate, catalase-polyethylene glycol, and manganese(III) tetrakis(1-methyl-4-pyridyl) porphyrin, observed in the present experiments imply that the dismutation of superoxide anions to H2O2 is also required. The observation that the addition of exogenous superoxide dismutase, which does not permeate cell membranes, did not prevent the endothelium-dependent contractions to acetylcholine, in confirmation of earlier studies in the rat aorta,12 indicates that the dismutation occurs intracellularly. The comparable contractions to exogenous H2O2 observed in the carotid arteries of Leprdb/db and DWM mice suggest that augmented levels of free radicals in the endothelial cells rather than their direct effect on vascular smooth muscle play a crucial role in the dysfunction observed in Leprdb/db preparations.

The present study thus demonstrates that loss-of-function in TLR4 protects mice from endothelial dysfunction resulting from obesity, whether genetic or diet induced, by both potentiating endothelium-dependent relaxations and decreasing endothelium-dependent contractions. These protective effects of the TLR4 mutation can be attributed to a reduced production of ROS because the different aspects of endothelial dysfunction observed in obese/diabetic mice with functional TLR4 are prevented by apocynin. Elevated ROS formation in the vascular wall is a key feature of cardiovascular disease and a likely contributor to endothelial dysfunction and vascular inflammation.32 ROS generated by NOX are the initial source of endothelial dysfunction in diabetes mellitus.5,9,19 NOX1 contributes significantly to gastrointestinal inflammation, hypertension, and restenosis after angioplasty.33

LPS is the prototypical ligand for TLR414 and triggers the signaling cascade leading to the activation of nuclear factor-κB and the production of proinflammatory cytokines, including TNF-α and IL-1β 8. Indeed, mutation of the TLR4 gene prevents LPS signal transduction in vitro and in vivo.34 Chronic elevation of the circulating levels of LPS exists in obese subjects and contributes to insulin resistance and its related cardiometabolic complications.35,36 The present experiments demonstrate that LPS administration in vivo to lean WT mice does not change glucose tolerance and insulin sensitivity but impairs endothelium-dependent relaxations and enhances endothelium-dependent contractions, mimicking the vascular phenotype observed in obese/diabetic mice. However, LPS fails to elicit these vascular responses in mice with mutation of TLR4, implying that the derogative effects of LPS on vascular tone are TLR4 dependent. These findings suggest that TLR4 activation-induced endothelial dysfunction is a direct effect and not secondary to metabolic changes. This conclusion is strengthened by the demonstration that in vitro administration of LPS had identical effects on endothelial function as its in vivo injection, and this again only in WT arteries. The ex vivo reversal by apocynin of these responses induced by LPS indicates again that ROS are the key link between TLR4 and endothelial dysfunction.

In conclusion, the present findings suggest that activation of TLR4 promotes the transcription of NADPH oxidase 1 and 4, resulting in elevated reactive oxidative stress. In endothelial cells, the increased level of ROS reduces eNOS coupling leading to a reduced NO-production and bioavailability, impairs EDH-mediated responses, and increases the activity of COX-1 with augmented EDCF-mediated contractions. These effects combine to initiate endothelial dysfunction in arteries of obese and diabetic mice. The present study identifies TLR4 as being responsible for the endothelial dysfunction resulting from obesity and diabetes mellitus.

Acknowledgments

We thank the late Dr Francosis Li for his contribution to the experiments.

Sources of Funding

This project is supported by the University of Hong Kong, the Research Grant Council of the Hong Kong Special Administrative Region (HKU777208M, NHKU 735/08, and HKU4/CRF/10), and the World Class University program (R31-20029) funded by the Ministry of Education, Science and Technology, South Korea.

Disclosures

None.

Footnotes

  • Received February 27, 2012.
  • Accepted January 22, 2013.

References

  1. 1.↵
    1. Antuna-Puente B,
    2. Feve B,
    3. Fellahi S,
    4. Bastard JP
    . Adipokines: the missing link between insulin resistance and obesity. Diabetes Metab. 2008;34:2–11.
    OpenUrlCrossRefPubMed
  2. 2.↵
    1. Vanhoutte PM,
    2. Shimokawa H,
    3. Tang EH,
    4. Feletou M
    . Endothelial dysfunction and vascular disease. Acta Physiol (Oxf). 2009;196:193–222.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Gruber HJ,
    2. Mayer C,
    3. Mangge H,
    4. Fauler G,
    5. Grandits N,
    6. Wilders-Truschnig M
    . Obesity reduces the bioavailability of nitric oxide in juveniles. Int J Obes (Lond). 2008;32:826–831.
    OpenUrlCrossRefPubMed
  4. 4.↵
    1. Park Y,
    2. Capobianco S,
    3. Gao X,
    4. Falck JR,
    5. Dellsperger KC,
    6. Zhang C
    . Role of EDHF in type 2 diabetes-induced endothelial dysfunction. Am J Physiol Heart Circ Physiol. 2008;295:H1982–H1988.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    1. Shi Y,
    2. Vanhoutte PM
    . Oxidative stress and COX cause hyper-responsiveness in vascular smooth muscle of the femoral artery from diabetic rats. Br J Pharmacol. 2008;154:639–651.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Kim F,
    2. Tysseling KA,
    3. Rice J,
    4. Pham M,
    5. Haji L,
    6. Gallis BM,
    7. Baas AS,
    8. Paramsothy P,
    9. Giachelli CM,
    10. Corson MA,
    11. Raines EW
    . Free fatty acid impairment of nitric oxide production in endothelial cells is mediated by IKKbeta. Arterioscler Thromb Vasc Biol. 2005;25:989–994.
    OpenUrlAbstract/FREE Full Text
  7. 7.↵
    1. Shi H,
    2. Kokoeva MV,
    3. Inouye K,
    4. Tzameli I,
    5. Yin H,
    6. Flier JS
    . TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest. 2006;116:3015–3025.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Kim JS,
    2. Yeo S,
    3. Shin DG,
    4. Bae YS,
    5. Lee JJ,
    6. Chin BR,
    7. Lee CH,
    8. Baek SH
    . Glycogen synthase kinase 3beta and beta-catenin pathway is involved in toll-like receptor 4-mediated NADPH oxidase 1 expression in macrophages. FEBS J. 2010;277:2830–2837.
    OpenUrlCrossRefPubMed
  9. 9.↵
    1. Somers MJ,
    2. Burchfield JS,
    3. Harrison DG
    . Evidence for a NADH/NADPH oxidase in human umbilical vein endothelial cells using electron spin resonance. Antioxid Redox Signal. 2000;2:779–787.
    OpenUrlPubMed
  10. 10.↵
    1. Brandes RP,
    2. Kreuzer J
    . Vascular NADPH oxidases: molecular mechanisms of activation. Cardiovasc Res. 2005;65:16–27.
    OpenUrlAbstract/FREE Full Text
  11. 11.↵
    1. Rubanyi GM,
    2. Vanhoutte PM
    . Oxygen-derived free radicals, endothelium, and responsiveness of vascular smooth muscle. Am J Physiol. 1986;250(5 pt 2):H815–H821.
    OpenUrl
  12. 12.↵
    1. Tang EH,
    2. Vanhoutte PM
    . Prostanoids and reactive oxygen species: team players in endothelium-dependent contractions. Pharmacol Ther. 2009;122:140–149.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Raetz CR,
    2. Garrett TA,
    3. Reynolds CM,
    4. et al
    . Kdo2-Lipid A of Escherichia coli, a defined endotoxin that activates macrophages via TLR-4. J Lipid Res. 2006;47:1097–1111.
    OpenUrlAbstract/FREE Full Text
  14. 14.↵
    1. Kim F,
    2. Pham M,
    3. Luttrell I,
    4. Bannerman DD,
    5. Tupper J,
    6. Thaler J,
    7. Hawn TR,
    8. Raines EW,
    9. Schwartz MW
    . Toll-like receptor-4 mediates vascular inflammation and insulin resistance in diet-induced obesity. Circ Res. 2007;100:1589–1596.
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    1. Reckelhoff JF,
    2. Fortepiani LA
    . Novel mechanisms responsible for postmenopausal hypertension. Hypertension. 2004;43:918–923.
    OpenUrlCrossRef
  16. 16.↵
    1. Pannirselvam M,
    2. Wiehler WB,
    3. Anderson T,
    4. Triggle CR
    . Enhanced vascular reactivity of small mesenteric arteries from diabetic mice is associated with enhanced oxidative stress and cyclooxygenase products. Br J Pharmacol. 2005;144:953–960.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Michel T,
    2. Vanhoutte PM
    . Cellular signaling and NO production. Pflugers Arch. 2010;459:807–816.
    OpenUrlCrossRefPubMed
  18. 18.↵
    1. Musicki B,
    2. Liu T,
    3. Lagoda GA,
    4. Strong TD,
    5. Sezen SF,
    6. Johnson JM,
    7. Burnett AL
    . Hypercholesterolemia-induced erectile dysfunction: endothelial nitric oxide synthase (eNOS) uncoupling in the mouse penis by NAD(P)H oxidase. J Sex Med. 2010;7:3023–3032.
    OpenUrlCrossRefPubMed
  19. 19.↵
    1. Rajagopalan S,
    2. Kurz S,
    3. Münzel T,
    4. Tarpey M,
    5. Freeman BA,
    6. Griendling KK,
    7. Harrison DG
    . Angiotensin II-mediated hypertension in the rat increases vascular superoxide production via membrane NADH/NADPH oxidase activation. Contribution to alterations of vasomotor tone. J Clin Invest. 1996;97:1916–1923.
    OpenUrlCrossRefPubMed
  20. 20.↵
    1. Förstermann U,
    2. Münzel T
    . Endothelial nitric oxide synthase in vascular disease. Circulation. 2006;113:1708–1714.
    OpenUrlAbstract/FREE Full Text
  21. 21.↵
    1. Landmesser U,
    2. Dikalov S,
    3. Price SR,
    4. McCann L,
    5. Fukai T,
    6. Holland SM,
    7. Mitch WE,
    8. Harrison DG
    . Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension. J Clin Invest. 2003;111:1201–1209.
    OpenUrlCrossRefPubMed
  22. 22.↵
    1. Félétou M,
    2. Vanhoutte PM
    . EDHF: an update. Clin Sci. 2009;117:139–155.
    OpenUrlCrossRefPubMed
  23. 23.↵
    1. Ozkan MH,
    2. Uma S
    . Inhibition of acetylcholine-induced EDHF response by elevated glucose in rat mesenteric artery. Life Sci. 2005;78:14–21.
    OpenUrlCrossRefPubMed
  24. 24.↵
    1. Félétou M,
    2. Vanhoutte PM,
    3. Verbeuren TJ
    . The thromboxane/endoperoxide receptor (TP): the common villain. J Cardiovasc Pharmacol. 2010;55:317–332.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Zhou Y,
    2. Varadharaj S,
    3. Zhao X,
    4. Parinandi N,
    5. Flavahan NA,
    6. Zweier JL
    . Acetylcholine causes endothelium-dependent contraction of mouse arteries. Am J Physiol Heart Circ Physiol. 2005;289:H1027–H1032.
    OpenUrlAbstract/FREE Full Text
  26. 26.↵
    1. Vanhoutte PM
    . Endothelium-dependent contractions in hypertension: when prostacyclin becomes ugly. Hypertension. 2011;57:526–531.
    OpenUrlCrossRef
  27. 27.↵
    1. Félétou M,
    2. Huang Y,
    3. Vanhoutte PM
    . Endothelium-mediated control of vascular tone: COX-1 and COX-2 products. Br J Pharmacol. 2011;164:894–912.
    OpenUrlCrossRefPubMed
  28. 28.↵
    1. Münzel T,
    2. Afanas’ev IB,
    3. Kleschyov AL,
    4. Harrison DG
    . Detection of superoxide in vascular tissue. Arterioscler Thromb Vasc Biol. 2002;22:1761–1768.
    OpenUrlAbstract/FREE Full Text
  29. 29.↵
    1. Li FY,
    2. Lam KS,
    3. Tse HF,
    4. Chen C,
    5. Wang Y,
    6. Vanhoutte PM,
    7. Xu A
    . Endothelium-selective activation of AMP-activated protein kinase prevents diabetes-induced impairment in vascular function and re-endothelialization via induction of heme oxygenase-1 in mice. Circulation. 2012; 126:1267–1277.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Bedard K,
    2. Krause KH
    . The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiology Rev. 2005;146:1061–8.
    OpenUrl
  31. 31.↵
    1. Qian L,
    2. Gao X,
    3. Pei Z,
    4. Wu X,
    5. Block M,
    6. Wilson B,
    7. Hong JS,
    8. Flood PM
    . NADPH oxidase inhibitor DPI is neuroprotective at femtomolar concentrations through inhibition of microglia over-activation. Parkinsonism Relat Disord. 2007;13 suppl 3:S316–S320.
    OpenUrl
  32. 32.↵
    1. Csányi G,
    2. Taylor WR,
    3. Pagano PJ
    . NOX and inflammation in the vascular adventitia. Free Radic Biol Med. 2009;47:1254–1266.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Lassègue B,
    2. Griendling KK
    . NADPH oxidases: functions and pathologies in the vasculature. Arterioscler Thromb Vasc Biol. 2010;30:653–661.
    OpenUrlAbstract/FREE Full Text
  34. 34.↵
    1. Du X,
    2. Poltorak A,
    3. Silva M,
    4. Beutler B
    . Analysis of Tlr4-mediated LPS signal transduction in macrophages by mutational modification of the receptor. Blood Cells Mol Dis. 1999;25:328–338.
    OpenUrlCrossRefPubMed
  35. 35.↵
    1. Cani PD,
    2. Amar J,
    3. Iglesias MA,
    4. et al
    . Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes. 2007;56:1761–1772.
    OpenUrlAbstract/FREE Full Text
  36. 36.↵
    1. Manco M,
    2. Putignani L,
    3. Bottazzo GF
    . Gut microbiota, lipopolysaccharides, and innate immunity in the pathogenesis of obesity and cardiovascular risk. Endocr Rev. 2010;31:817–844.
    OpenUrlCrossRefPubMed

Significance

This study reveals that loss-of-function of toll-like receptor 4 in mice can alleviate dietary obesity- and diabetes-associated endothelial dysfunction by decreasing the production of reactive oxygen species (generated by NADPH oxidase isoforms 1 and 4) and vasoconstrictor prostaglandins (generated by cyclooxygenase-1) in endothelial cells. Because endothelial dysfunction precedes several cardiovascular complications, toll-like receptor 4 inactivation may be a therapeutic target for the prevention and treatment of vascular disease associated with obesity and diabetes mellitus.

View Abstract

This Issue

Jump to

Article Tools

Share this Article

  • Email
  • Toll-Like Receptor 4 Mutation Protects Obese Mice Against Endothelial Dysfunction by Decreasing NADPH Oxidase Isoforms 1 and 4Significance
    Chao-Fan Liang, Jacky TC Liu, Yu Wang, Aimin Xu and Paul M. Vanhoutte
    Arteriosclerosis, Thrombosis, and Vascular Biology. 2013;33:777-784, originally published March 13, 2013
    https://doi.org/10.1161/ATVBAHA.112.301087
    del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Related Articles

Cited By...

Subjects