Local Inflammation and Hypoxia Abolish the Protective Anticontractile Properties of Perivascular Fat in Obese Patients
Background— Inflammation in adipose tissue has been implicated in vascular dysfunction, but the local mechanisms by which this occurs are unknown.
Methods and Results— Small arteries with and without perivascular adipose tissue were taken from subcutaneous gluteal fat biopsy samples and studied with wire myography and immunohistochemistry. We established that healthy adipose tissue around human small arteries secretes factors that influence vasodilation by increasing nitric oxide bioavailability. However, in perivascular fat from obese subjects with metabolic syndrome (waist circumference 111±2.8 versus 91.1±3.5 cm in control subjects, P<0.001; insulin sensitivity 41±5.9% versus 121±18.6% in control subjects, P<0.001), the loss of this dilator effect was accompanied by an increase in adipocyte area (1786±346 versus 673±60 μm2, P<0.01) and immunohistochemical evidence of inflammation (tumor necrosis factor receptor 1 12.4±1.1% versus 6.7±1%, P<0.001). Application of the cytokines tumor necrosis factor receptor-α and interleukin-6 to perivascular fat around healthy blood vessels reduced dilator activity, resulting in the obese phenotype. These effects could be reversed with free radical scavengers or cytokine antagonists. Similarly, induction of hypoxia stimulated inflammation and resulted in loss of anticontractile capacity, which could be rescued by catalase and superoxide dismutase or cytokine antagonists. Incubation with a soluble fragment of adiponectin type 1 receptor or inhibition of nitric oxide synthase blocked the vasodilator effect of healthy perivascular adipose tissue.
Conclusions— We conclude that adipocytes secrete adiponectin and provide the first functional evidence that it is a physiological modulator of local vascular tone by increasing nitric oxide bioavailability. This capacity is lost in obesity by the development of adipocyte hypertrophy, leading to hypoxia, inflammation, and oxidative stress.
Received September 15, 2008; accepted January 26, 2009.
Metabolic syndrome is a precursor to type 2 diabetes mellitus and cardiovascular disease, with a prevalence of almost 40% in the adult population.1 Central obesity is believed to be the main cause of metabolic syndrome, and this is reflected in newer definitions of the condition with large waist circumference as a prerequisite.2 Although associations of obesity with hypertension,3 insulin resistance,4 and cardiovascular disease5 are well described, the underlying mechanisms are poorly understood. Two areas of research that may provide insight into these are the vasoactive properties of perivascular adipose tissue (PVAT) and the inflammatory changes that occur in fat as obesity develops. Demonstrated in 1991,6 it is now accepted that healthy PVAT has an anticontractile effect.7–10 The mechanism appears to be both endothelium dependent via release of nitric oxide8,10 and endothelium-independent via generation of hydrogen peroxide.8 In obesity and metabolic syndrome, there is also a conformational change in adipose tissue: Infiltration and activation11 of circulating macrophages leads to inflammation characterized by upregulation of tumor necrosis factor alpha (TNF-α).12 Until recently, the primary consequences of this inflammation were thought to be a reduction in insulin sensitivity and an increase in nonesterified free fatty acids. However, given the profound effects of cytokines on arterial function,13 it has been suggested that the changes seen in adipose tissue in obesity are likely to have detrimental effects on the ability of PVAT to regulate vascular tone.14,15 The consequences of this are profound: An increase in tone could enhance peripheral resistance and blood pressure.15 Furthermore, a reduction in downstream flow would limit nutritive delivery contributing to insulin resistance,16 whereas reduced exposure of triglycerides to endothelial lipoprotein lipase would contribute to dyslipidemia.17 Therefore, the present series of studies was designed to test the hypothesis that in obesity and the metabolic syndrome, local adipocyte inflammation affects the anticontractile properties of PVAT.
Clinical Perspective p 1670
Ten patients with obesity and at least 3 additional characteristics of the metabolic syndrome and 10 healthy lean subjects gave full written informed consent and participated in the study, which was approved by the local research ethics committee. Metabolic syndrome was defined according to the 2001 National Cholesterol Education Program Adult Treatment Panel guidelines.18 Fasting venous blood samples were taken to assess glycemic and lipid profiles and inflammatory markers. Blood pressure was measured with subjects seated, after 15 minutes of rest, by a semiautomatic machine (OMRON 705CP, White Medical, Clifton-Upon-Dunsmore, United Kingdom) with a mean of 3 readings recorded. Anthropometric measurements and bioimpedance were also recorded.
High-sensitivity C-reactive protein, leptin, interleukin (IL)-6, TNF, and adiponectin were assayed by an in-house ELISA technique (hs-CRP, Abcam, Cambridge, Mass; remainder, R&D Systems, Minneapolis, Minn). Insulin levels were measured by a radiolabeling technique described previously.19 The homeostasis model assessment (HOMA)20 was used to estimate β-cell function (HOMA-B) and peripheral insulin sensitivity (HOMA-S).
A subcutaneous gluteal fat biopsy sample was obtained from each subject under local anesthesia, which allowed tissue (2×1.5×1.5 cm) to be harvested and placed immediately in physiological saline solution (PSS). Small arteries, 100 to 150 μm in diameter, were dissected from the fat under a dissecting microscope, transferred to an arteriographic bath chamber, and cannulated.21
The chamber was placed on the stage of an inverted microscope and superfused with PSS, gassed with 5% CO2/95% air (pH 7.4) at 37°C, at a superfusion rate of 20 mL/min. PSS composition was (in mmol/L) 139 NaCl, 4.7 KCl, 25 NaHCO3, 1.17 KH2PO4, 1.17 MgSO4, 0.026 EDTA, 1.6 CaCl2, and 5.5 glucose. Lumen diameter was recorded with the use of a Video Dimension Analyzer (Living Systems Instrumentations, Burlington, Vt) connected to a chart recorder. Vessels were pressurized to 60 mm Hg with a pressure servo system and equilibrated at 37°C for 1 hour before challenge with 60 mmol/L potassium-enriched PSS (KPSS).
Pressure Myography: Pharmacological Assessment
Each vessel was stimulated as follows: (1) Cumulative addition of norepinephrine (Sigma-Aldrich, Dorset, UK): 10−9, 3×10−9, 10−8, 3×10−8, 10−7, 3×10−7, 10−6, 3×10−6, and 10−5 mol/L, with 5 minutes’ incubation per concentration. (2) Cumulative response to acetylcholine (Sigma; in mol/L): 10−9, 3×10−9, 10−8, 3×10−8, 10−7, 3×10−7, 10−6, 3×10−6, and 10−5, to a vessel preconstricted with 10−5 norepinephrine. (3) After 1 hour of incubation with 5×10−5 mol/L NG-monomethyl-l-arginine (L-NMMA; Sigma), a nitric oxide synthase inhibitor, the responses to acetylcholine were repeated as in step 2.
From the same biopsy, a larger arterial segment (diameter 250 to 350 μm) was dissected. One segment of this artery was cleaned of PVAT, whereas on the adjacent segment, PVAT was left intact. Both arteries were mounted on 40-μm wires in a wire myograph (Danish MyoTech, Aarhus, Denmark). After an initial 30-minute incubation, vessel wall tension and diameter were normalized in a standardized procedure22 and stabilized for 1 hour.
Arteries were first challenged with KPSS 60 mmol/L to establish a baseline contractile response. A cumulative dose response to norepinephrine was then constructed (as with the pressure myography protocol). Contractile responses to norepinephrine are expressed as a percentage of KPSS contraction, consistent with other studies.7,8,10,23
Bioassay experiments studied the release of relaxing factors from PVAT.8 We used arteries with PVAT intact (PVAT+) as donors and arteries stripped of PVAT as recipients (PVAT−; Figure 1A). Arterial segments came from the same branch of the artery from the biopsy. Both arteries were preconstricted with norepinephrine 1×10−6 mol/L. We then removed solutions from both baths and transferred solution from the bath that contained the donor artery (PVAT+) to the bath containing the recipient artery (PVAT−). The subsequent change in response to tension in the recipient artery (PVAT−) was then recorded. This was performed in both healthy control subjects (n=3) and obese patients (n=3).
We used pharmacological protocols to investigate whether reproduced aspects of the inflammatory obese environment were able to attenuate the anticontractile effects of PVAT. Thus, from healthy participants, arterial segments both with and without PVAT were incubated with an adiponectin type 1 receptor–blocking fragment, the cytokine TNF, and the inhibitor of nitric oxide synthase, L-NMMA. We also investigated whether PVAT function, which is damaged in patients with obesity and metabolic syndrome, could be restored by incubation with the anti-TNF antibody infliximab. The incubation time was 45 minutes for TNF (Sigma; 0.001 μg/mL, 5.9×10−8 mol/L). For adiponectin blockade, an incubation time of 45 minutes was used (5 μg/mL, 1.6×10−4 mol/L Acrp-30 [N-20]-P adiponectin receptor blocking peptide; Santa Cruz Biotechnology, Santa Cruz, Calif). One hour of incubation was used with L-NMMA (Sigma; 5×10−5 mol/L). For anti-TNF, an incubation time of 10 minutes was used (50 μg/mL infliximab, 3.4×10−4 mol/L; Schering-Plough Ltd, Hertfordshire, UK).
Male Wistar rats aged 12 to 15 weeks were killed by stunning, followed by cervical dislocation according to Home Office regulations. Mesenteric arteries (diameter 300 μm) were exposed and dissected. Part of the artery was cleaned of fat, whereas part of the artery was left with fat intact. Arteries were mounted and normalized on the wire myograph, and a cumulative concentration-response curve to norepinephrine was constructed (described above).
Protocols with healthy animal tissue were designed to investigate mechanisms of both inflammatory damage to PVAT and subsequent restoration of function. Incubations performed for arteries with and without PVAT included the following: (1) Adiponectin blocking peptide; (2) single cytokines (TNF, IL-6, or IL-1); (3) TNF and anti-TNF blocking antibody; (4) IL-6 and IL-6 blocking antibody; (5) IL-6 and TNF, followed by incubation with superoxide dismutase and catalase, followed by incubation with adiponectin blocking peptide; (6) hypoxia; and (7) hypoxia followed by incubation with anti-TNF blocking antibody (infliximab) or anti-IL-6 blocking peptide.
Protocols were as follows: TNF incubation, 0.001 μg/mL (5.9×10−8 mol/L; Sigma) for 45 minutes. TNF antagonism, incubation for 30 minutes with TNF-α followed by addition of 50 μg/mL infliximab (3.4×10−4 mol/L; Schering-Plough Ltd) for 10 minutes. IL-6, 0.01 μg/mL (Sigma) for 45 minutes. IL-6 was blocked by an IL-6 antagonist (Sigma) at a concentration of 0.2 μg/mL for 20 minutes. IL-1, 0.02 μg/mL for 45 minutes. Adiponectin blockade, incubation with 5 μg/mL (1.6×10−4 mol/L) Acrp-30 (N-20)-P adiponectin blocking peptide (Santa Cruz Biotechnology) for 10 minutes. Superoxide dismutase and catalase (Sigma), 80 and 120 U/mL, respectively, with 45 minutes’ incubation. The effect of hypoxia on PVAT was investigated by gassing mounted arteries with 95% nitrogen and 5% CO2 for 2.5 hours. Norepinephrine dose-response curves were constructed both before and after hypoxic incubation.
Vasodilating Effects of Exogenous Adiponectin
Pressure myography was used to investigate the ability of adiponectin, secreted outside of the artery by PVAT, to exert a vasodilatory effect by moving transluminally to stimulate endothelial nitric oxide production. A dose-response curve to norepinephrine was constructed with rat mesenteric arteries. The artery was then preconstricted at an ED50 concentration for norepinephrine before the addition of adiponectin (BioVendor, Heidelberg, Germany; 10 μg/mL) to the bath solution. Because the vessels are tied off firmly at both ends, adiponectin-mediated effects must be via adventitia or the smooth muscle of the vessel wall.
Sagittal sections from the biopsy that incorporated skin and subcutaneous fat were placed in 4% paraformaldehyde in phosphate-buffered saline for 24 hours and then processed on an 18-hour cycle in a Shandon automatic tissue processor. The sample was embedded in paraffin wax and serially sectioned at 5 μm. Dewaxed and rehydrated sections were immunostained for type 1 and type 2 TNF receptors (TNFRc1 and TNFRc2; rabbit polyclonal antibodies, Abcam). Tissue sections were incubated for 18 hours with primary antibodies, followed by biotinylated secondary antibodies and streptavidin-horseradish peroxidase–conjugated complexes (DAKO, Glostrup, Denmark) and developed with SG chromogen (Vector Laboratories, Peterborough, UK). Immunostaining for TNFRc1 required heat-induced antigen retrieval in an autoclave (at 125 psi) in 10 mmol/L citrate buffer (pH 6.0). Color images were captured with a Go-3 QImaging camera (QImaging Corp, Vancouver, Canada) mounted on a Leitz Diaplan microscope and converted to grayscale images. Quantitative analysis of immunostaining was obtained with a macro subroutine in an Image-Pro version 6.2 image-analysis program. The extent of staining was expressed as a percentage of the entire area photographed. Adipocyte size was quantified by manually tracing the margins of 100 consecutive adipocytes on immunostained sections to avoid selection bias (total: 2000 cells).
Data are presented as mean±SEM. For the cumulative dose-response experiments on human tissue (Figures 2A, 2C, 3D, and 3⇓E), differences between dose-diameter relations (pressure myography) and dose-tension relations (wire myography) were analyzed as repeated-measures ANOVA. For the pharmacological assays (Figures 3A through 3C and Figures 4 through 6⇓⇓), differences in tension at specific points on the dose-response curve were analyzed with Student’s t test. A probability value <0.05 was considered statistically significant. Analyses were performed with SPSS software (SPSS, Inc, Chicago, Ill).
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Study Design and Participants
Compared with healthy subjects, obese patients with metabolic syndrome had significantly greater waist circumference, body mass index, total body fat content, fasting glucose, pancreatic β-cell workload (HOMA-B), and leptin levels and lower bioimpedance, peripheral insulin sensitivity (HOMA-S), and high-density lipoprotein cholesterol (P<0.05). Blood pressure, markers of systemic inflammation, serum adiponectin levels, age (control group 49.6±3.65 years versus metabolic syndrome group 54.5±2.2 years), and sex (8 males and 2 females in the control group versus 7 males and 3 females in the metabolic syndrome group) were not significantly different (Table).
Anticontractile Capacity and Release of Relaxing Factor From PVAT Is Reduced in Metabolic Syndrome
In healthy volunteers, the presence of intact PVAT exerted a significant anticontractile effect on small arteries (P<0.05; Figure 2A). The effect was independent of endothelium-derived hyperpolarizing factor, because there was no difference in contraction to KPSS 60 mmol/L when vessels were studied in the presence or absence of PVAT (Figure 2B). In obese patients with metabolic syndrome, the anticontractile capacity of PVAT was abolished completely, with no difference in contractility to norepinephrine, whether PVAT was intact or removed (Figure 2C). The presence of intact PVAT in this group had no influence on contractility to KPSS 60 mmol/L (Figure 2D).
The bioassay protocol (Figure 1A) that we performed on control participant tissue elicited a profound vasodilation (50.4±8.8% increase in diameter; Figure 1B and 1C) of recipient arteries after transfer of solution taken from baths that contained arteries with intact PVAT. When solution transfer was performed on arteries from obese patients with metabolic syndrome, little change in arterial diameter occurred (12.4±6% increase in diameter, P<0.05 versus change elicited by control patients; Figure 1B and 1D).
Adipose Inflammation and Adipocyte Hypertrophy in Metabolic Syndrome
Quantitative image analysis demonstrated a significant increase in expression of TNF-α receptor 1 in adipose tissue from obese patients with metabolic syndrome (12.4±1.1%) compared with control participants (6.7±1%; P<0.001). There were no significant differences in TNF-α receptor 2 expression (control participants 4.6±0.9% versus obese patients 8.7±3.5%). Adipocyte cross-sectional area measured in 2000 consecutive cells was increased significantly in obese patients with metabolic syndrome compared with healthy participants (control 673±60 versus obese 1786±346 μm2, P=0.006).
Factors Damaging the Anticontractile Pathway in Healthy Control Subjects
Application of a fragment of the human type 1 adiponectin receptor to arteries with intact PVAT from healthy subjects completely abolished the anticontractile capacity of PVAT, restoring the contractile profile to that of a healthy artery devoid of PVAT (Figure 3A). Also, in healthy participants, incubation of arteries with intact PVAT with L-NMMA (Figure 3B) or the cytokine TNF (Figure 3C) caused an attenuation of the anticontractile response, mimicking the obese phenotype. Using pressure myography, we confirmed a defect in small-artery endothelial function in patients with obesity and metabolic syndrome (Figure 3D), assessed as a surrogate measure by the ability of a preconstricted artery to dilate in response to increasing doses of acetylcholine. Subsequent incubation with L-NMMA showed the defect in endothelial function in obesity to be caused by downregulation of nitric oxide synthase (Figure 3E).
Restoration of PVAT Anticontractile Function in Unhealthy Obese Patients
We attempted to restore the anticontractile effect of PVAT in patients with obesity and metabolic syndrome by incubation with the anti-TNF antibody infliximab, but this had no effect on the capacity of PVAT to reduce contractility of human arteries (supplemental Figure I).
Mechanisms of Damage to and Restoration of PVAT Function in an Animal Model
Intact PVAT exerted an anticontractile effect on small mesenteric arteries from healthy Wistar rats (Figure 4A). In agreement with our studies in small human arteries, a soluble fragment of the rodent adiponectin type 1 receptor almost completely abolished the anticontractile capacity of healthy rat PVAT (Figure 4B). We found a selective effect of inflammatory cytokines on the anticontractile effect of PVAT: Incubation with TNF-α (Figure 5A) and IL-6 (supplemental Figure II) but not IL-1 (Figure 5B) attenuated the anticontractile effect of healthy PVAT. The effects of TNF and IL-6 were fully reversible by subsequent incubation with their antagonists, the anti-TNF antibody infliximab and a bovine IL-6 antibody, respectively (supplemental Figure III). After incubation with TNF and IL-6 together, we were able to restore the anticontractile effect of PVAT using the free radical scavengers superoxide dismutase and catalase (Figure 5C). Further application of the soluble fragment of the adiponectin receptor 1 reduced the anticontractile effect (Figure 5D). Cytokine application had no effect on the contractility of arteries without PVAT (data not shown).
Effects of Hypoxia on PVAT Function in an Animal Model
The physiological effects of hypoxia on vascular contractility and PVAT function were investigated by incubation of arterial segments for 2.5 hours with 95% N2/5% CO2. Hypoxia significantly attenuated the anticontractile effect of PVAT (P<0.05; Figure 6A). We assessed whether the effect of hypoxia on PVAT was mediated by inflammation. After a period of hypoxia, arteries both with and without PVAT were incubated with an anti-TNF-α antibody (Figure 6B) or anti-IL-6 antibody (supplemental Figure II). These incubations restored the anticontractile capacity of the PVAT in both cases. In arteries without fat, neither hypoxic incubation nor subsequent incubation with anti-TNF or anti-IL-6 after hypoxia had significant effects on contractility (supplemental Figure III).
Bioactivity of Exogenous Adiponectin
The ability of adiponectin to act as a paracrine factor by diffusion through the vascular smooth muscle of the media was examined with pressure myography. Rat mesenteric arteries, tied firmly at both ends and pressurized to 60 mm Hg, were preconstricted with norepinephrine to an ED50 concentration. Adiponectin applied directly to the bath solution, but not the intraluminal compartment, elicited a vasodilation (71.5±7.8%, n=3) within 2 minutes of exposure to the artery (supplemental Figure IV, representative tracing).
Our data provide the first evidence in humans to support the hypothesis that obesity-related changes in adipose tissue have direct effects on the vasoactive properties of PVAT.14,15 We report abnormalities in adiponectin-mediated anticontractile properties of PVAT, endothelium-mediated dilation of the artery, and changes to the inflammatory and hypoxic profile of subcutaneous adipose tissue in obesity and metabolic syndrome. Additionally, using in vitro animal and human small-artery studies, hypoxia and inflammation were shown to attenuate the local vasoactive properties of PVAT by oxidative stress.
In healthy individuals, PVAT mediated an anticontractile effect. Although observed previously in medium-sized arteries,24 this is the first demonstration of this effect in human small arteries, and we are also the first to demonstrate a complete loss of the anticontractile effect in patients with obesity and metabolic syndrome. Subsequently, in vitro, we were able to replicate the biological changes that occur in adipose tissue in obesity and have delineated the pathway by which obesity-related damage to PVAT function occurs.
Although it is accepted that adipose inflammation has a central role in insulin resistance,25,26 there are no data investigating the mechanisms or effects that this inflammation will have on the vasoactive properties of PVAT. When human arteries with intact PVAT were exposed to TNF, there was a reduction in the anticontractile function of PVAT, which produced the unhealthy obese phenotype. Exogenous TNF damages small-artery endothelial function by downregulation of nitric oxide synthase,13 which is responsible for the endothelium-dependent mechanism by which PVAT mediates an anticontractile effect in healthy rat adipose tissue.8 Inhibition of nitric oxide synthase with L-NMMA attenuated the anticontractile capacity of healthy human PVAT, as with TNF, thus negating the vasoactive function of the fat. Our study of arteries with pressure myography confirmed that in obese patients with metabolic syndrome, the defect in endothelial function was caused by downregulation of nitric oxide synthase.
In an animal model, exposure to cytokines had a selective effect on PVAT, with reduction in function after incubation with TNF and IL-6 but not IL-1. The damage to the anticontractile function caused by these cytokines was completely reversible by their blocking antibodies but also by superoxide dismutase and catalase, which suggests the inflammatory damage is mediated by oxidative stress. The animal model also permitted investigation into the effects of hypoxia on PVAT function. It is thought that adipocyte hypertrophy, quantified in patients in the present study and other studies,27 causes cellular hypoxia, which stimulates adipocyte and macrophage production of inflammatory cytokines, including TNF and IL-6.28,29 We observed complete loss of the anticontractile effect after 2.5 hours under hypoxic conditions, but we were able to rescue this function of the fat physiologically by subsequent incubation with antibodies to either TNF or IL-6. Although the data appear robust, we acknowledge that the sequential nature of these pharmacological protocols increases the risk of a type 1 statistical error. In arteries without PVAT, neither hypoxia nor incubation with anti-TNF or anti-IL-6 after hypoxia had any effect on contractility, which suggests that the ability of anti-TNF or anti-IL-6 to restore the anticontractile capacity of PVAT was mediated through actions on adipose tissue. In obese patients, however, incubation with anti-TNF was unable to restore anticontractile function in PVAT. This may be due to the prolonged period of inflammation associated with disease, but this negative finding mirrors observations seen when anti-TNF is used to treat other cardiovascular disorders.30,31
Adiponectin displays many characteristics observed of the relaxing factor released from human PVAT. It is released exclusively by adipocytes and increases nitric oxide synthase activity through endothelial receptors.32 Furthermore, both hypoxia33 and inflammation34 reduce its production. Adiponectin was recently discounted as a potential candidate for this role on the grounds that PVAT surrounding mesenteric arteries from the adiponectin receptor 1 knockout mouse remains able to elicit an anticontractile effect.35 However, when we incubated arteries with intact PVAT from healthy individuals with an adiponectin type 1 receptor–blocking fragment, the anticontractile properties of the fat were abolished entirely, which suggests that in human subcutaneous adipose tissue, adiponectin is the predominant mediator of the relaxation. This was reinforced by a similar finding in our animal model. We have also shown that adiponectin, if released adjacent to a small artery by perivascular adipocytes, is able to effect a vasodilation.
The concept that adiponectin is central to the function of healthy PVAT is not incompatible with the findings from the study with the adiponectin type 1 receptor knockout mouse. Two pathways have been delineated by which PVAT mediates an effect on rat aorta: An endothelium-dependent pathway that involves nitric oxide and an endothelium-independent pathway that involves hydrogen peroxide.8 Given its association with nitric oxide synthase, adiponectin clearly would occupy a position in the former of these categories. In the adiponectin type 1 receptor knockout mouse, it is likely that the endothelium-independent pathway is upregulated in response to the genetic defect and thus is responsible for the anticontractile properties of the PVAT observed. Alternatively, in this model, adiponectin may act through either the adiponectin type 2 receptor or T-cadherin, a glycosylphosphatidylinol-linked cell surface molecule that has been suggested as a potential receptor for adiponectin and is highly expressed on vascular endothelium.36
The predominant physiological mechanism by which perivascular fat modulates contraction in human subcutaneous small arteries has been identified. Adipocytes release adiponectin, which maintains endothelial nitric oxide production32 and thereby reduces vascular tone. That blocking the effect of adiponectin completely abolishes this response implies that in humans, this is the major component of the anticontractile pathway. Further investigation is clearly indicated to validate this and to examine whether this is the case in other circulatory beds. In obesity and metabolic syndrome, however, the anticontractile capacity of PVAT is lost, because individual adipocyte size is greater, and consequently, there is local tissue hypoxia and inflammation. Such an environment provides the stimulus for adipocyte and macrophage release of inflammatory cytokines. Both hypoxia33 and inflammation34 are known to reduce adiponectin production from adipocytes, and in human and animal preparations, we have demonstrated that these stimuli abrogate the vasorelaxant properties of PVAT through increased oxidative stress. The reduction in adiponectin generation or function is likely to be responsible for the observed downregulation of nitric oxide synthase that contributes to endothelial dysfunction. However, given that cytokines are known to have a direct effect on endothelium, there is likely to be an additive effect also.13 The relative in vivo contributions of reduced adiponectin bioaction, increased local production of reactive oxygen species, and upregulation of tissue cytokines that impair local PVAT function is not entirely clear and will also require further investigation.
The present findings are also consistent with in vivo clinical studies in obesity that have shown reductions in flow-mediated dilation37 and both visceral and cutaneous capillary recruitment38 consistent with an enhanced vasoconstriction and impaired endothelial function. However, although the data provide evidence to support mechanisms by which obesity damages the influence of local adipose tissue on small arteries, the physiological consequences of this functional damage remain unknown. It has already been suggested that loss of PVAT function would result in increased peripheral vascular resistance and blood pressure.15 Furthermore, as proposed by Yudkin et al,16 the loss of a vasodilatory paracrine effect from adipose tissue may limit downstream microcirculatory nutritive flow, contributing to insulin resistance that is already compromised by the direct effects of TNF-α on adipocytes. A similar hypothesis has been applied to triglyceride metabolism.17 Whether this is the case remains to be established, but experiments using in vivo correlates are now clearly indicated to further our understanding of adipose tissue pathophysiology in obesity.
We thank all the volunteers who participated in this study and the staff of the Manchester Wellcome Trust Clinical Research Facility.
Sources of Funding
This study was funded by the European Society of Hypertension and the Wellcome Trust. Dr Heagerty has received speaking honoraria from Pfizer, Bayer, and Daiichi Sankyo. Dr Malik has received speaking honoraria from Pfizer, Astra-Zeneca, and Lilly. Dr Greenstein is a European Society of Hypertension Research Fellow.
Alexander CM, Landsman PB, Teutsch SM, Haffner SM. NCEP-defined metabolic syndrome, diabetes, and prevalence of coronary heart disease among NHANES III participants age 50 years and older. Diabetes. 2003; 52: 1210–1214.
International Diabetes Federation. The IDF consensus worldwide definition of the metabolic syndrome (article online). Available at: http://www.idf.org/home/index.cfm?node=1429. Accessed July 9, 2008.
Hall JE. The kidney, hypertension, and obesity. Hypertension. 2003; 41: 625–633.
Hunt KJ, Resendez RG, Williams K, Haffner SM, Stern MP. National Cholesterol Education Program versus World Health Organization metabolic syndrome in relation to all-cause and cardiovascular mortality in the San Antonio Heart Study. Circulation. 2004; 110: 1251–1257.
Dubrovska G, Verlohren S, Luft FC, Gollasch M. Mechanisms of ADRF release from rat aortic adventitial adipose tissue. Am J Physiol Heart Circ Physiol. 2004; 286: H1107–H1113.
Lohn M, Dubrovska G, Lauterbach B, Luft FC, Gollasch M, Sharma AM. Periadventitial fat releases a vascular relaxing factor. FASEB J. 2002; 16: 1057–1063.
Malinowski M, Deja MA, Golba KS, Roleder T, Biernat J, Wos S. Perivascular tissue of internal thoracic artery releases potent nitric oxide and prostacyclin-independent anticontractile factor. Eur J Cardiothorac Surg. 2008; 33: 225–231.
Kern PA, Ranganathan S, Li C, Wood L, Ranganathan G. Adipose tissue tumor necrosis factor and interleukin-6 expression in human obesity and insulin resistance. Am J Physiol Endocrinol Metab. 2001; 280: E745–E751.
Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care. 1998; 21: 2191–2192.
Mulvany MJ, Halpern W. Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ Res. 1977; 41: 19–26.
Verlohren S, Dubrovska G, Tsang SY, Essin K, Luft FC, Huang Y, Gollasch M. Visceral periadventitial adipose tissue regulates arterial tone of mesenteric arteries. Hypertension. 2004; 44: 271–276.
Kern PA, Di Gregorio GB, Lu T, Rassouli N, Ranganathan G. Adiponectin expression from human adipose tissue: relation to obesity, insulin resistance, and tumor necrosis factor-alpha expression. Diabetes. 2003; 52: 1779–1785.
Winkler G, Kiss S, Keszthelyi L, Sapi Z, Ory I, Salamon F, Kovacs M, Vargha P, Szekeres O, Speer G, Karadi I, Sikter M, Kaszas E, Dworak O, Gero G, Cseh K. Expression of tumor necrosis factor (TNF)-alpha protein in the subcutaneous and visceral adipose tissue in correlation with adipocyte cell volume, serum TNF-alpha, soluble serum TNF-receptor-2 concentrations and C-peptide level. Eur J Endocrinol. 2003; 149: 129–135.
Chung ES, Packer M, Lo KH, Fasanmade AA, Willerson JT. Randomized, double-blind, placebo-controlled, pilot trial of infliximab, a chimeric monoclonal antibody to tumor necrosis factor-α, in patients with moderate-to-severe heart failure: results of the anti-TNF Therapy Against Congestive Heart Failure (ATTACH) trial. Circulation. 2003; 107: 3133–3140.
Dominguez H, Storgaard H, Rask-Madsen C, Steffen Hermann T, Ihlemann N, Baunbjerg Nielsen D, Spohr C, Kober L, Vaag A, Torp-Pedersen C. Metabolic and vascular effects of tumor necrosis factor-alpha blockade with etanercept in obese patients with type 2 diabetes. J Vasc Res. 2005; 42: 517–525.
Wang ZV, Scherer PE. Adiponectin, cardiovascular function, and hypertension. Hypertension. 2008; 51: 8–14.
Wang B, Jenkins JR, Trayhurn P. Expression and secretion of inflammation-related adipokines by human adipocytes differentiated in culture: integrated response to TNF-alpha. Am J Physiol Endocrinol Metab. 2005; 288: E731–E740.
Fesus G, Dubrovska G, Gorzelniak K, Kluge R, Huang Y, Luft FC, Gollasch M. Adiponectin is a novel humoral vasodilator. Cardiovasc Res. 2007; 75: 719–727.
Hug C, Wang J, Ahmad NS, Bogan JS, Tsao TS, Lodish HF. T-cadherin is a receptor for hexameric and high-molecular-weight forms of Acrp30/adiponectin. Proc Natl Acad Sci U S A. 2004; 101: 10308–10313.
Hamburg NM, Larson MG, Vita JA, Vasan RS, Keyes MJ, Widlansky ME, Fox CS, Mitchell GF, Levy D, Meigs JB, Benjamin EJ. Metabolic syndrome, insulin resistance, and brachial artery vasodilator function in Framingham Offspring participants without clinical evidence of cardiovascular disease. Am J Cardiol. 2008; 101: 82–88.
Metabolic syndrome is a clinical phenotype of obesity with dyslipidemia, elevated glucose levels, and hypertension. Clinically, associated vascular abnormalities occur at almost every level, ranging from impaired flow-mediated dilation to reductions in capillary recruitment. Using material taken from gluteal fat biopsy samples, data are presented here to explain the mechanisms underlying this increase in vasoconstriction. In health, adipose tissue surrounding small arteries (perivascular adipose tissue) releases adiponectin, which exerts a local anticontractile effect by maintenance of endothelial nitric oxide synthase. In patients with obesity and metabolic syndrome, this anticontractile capacity of the perivascular adipose tissue is entirely lost, and as a result, there is an increase in basal tone of the small artery. Using a combination of immunohistochemistry and pharmacological assays, we show that the loss of anticontractile capacity of the fat is due to adipocyte hypertrophy, which initiates a cascade of hypoxia, inflammation, and oxidative stress.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.108.821181/DC1.