Skip Navigation

Adiponectin is a novel humoral vasodilator

  1. Gábor Fésüs,
  2. Galyna Dubrovska,
  3. Kerstin Gorzelniak,
  4. Reinhart Kluge,
  5. Yu Huang,
  6. Friedrich C. Luft and
  7. Maik Gollasch*
  1. Charité University Medicine Berlin, Germany
  2. Medical Clinic for Nephrology and Internal Intensive Care, Campus Virchow Klinikum, 13353 Berlin, Germany
  3. Franz Volhard Clinic at the Max Delbrück Center for Molecular Medicine, HELIOS Klinikum Berlin, 13125 Berlin, Germany
  4. German Institute of Human Nutrition (DIFE) Potsdam-Rehbrücke, Max Rubner Laboratory, 14558 Nuthetal, Germany
  5. Department of Physiology, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, Hong Kong, China
  1. * Corresponding author. Medical Clinic for Nephrology and Internal Intensive Care, Campus Virchow Klinikum, Augustenburger Platz 1, 13353 Berlin, Germany. Tel.: +49 30 450 653 263; fax: +49 30 450 553 909. maik.gollasch{at}charite.de
  • Received January 2, 2007.
  • Revision requested May 10, 2007.
  • Accepted May 29, 2007.
 
Next Section

Abstract

Objectives Perivascular adipose tissue secretes an adipocyte-derived relaxing factor(s) (ADRF) that opens Kv channels in rat arteries. Visceral fat accumulation causes adipocyte dysfunction, including hyposecretion of adiponectin. We tested the hypothesis that ADRF might be adiponectin and that adiponectin plays a role in the paracrine control of vascular tone by perivascular adipose tissue.

Methods and results We studied Sprague–Dawley rats, wild-type and adiponectin gene-deficient (Apn 1−/−) mice, and New Zealand obese (NZO) mice. In rat aortas, recombinant adiponectin at serum levels (2–5 μg/ml) inhibited serotonin-induced contractions. The effects were abolished by Kv channel inhibition with 4-aminopyridine (4-AP, 2 mM). Similar effects were observed in NZO mouse mesenteric arteries. To study vascular function in Apn 1−/− mice, the mesenteric vascular bed was isolated, cannulated, and perfused at a constant 4–5-ml/min flow in the absence and presence of serotonin. 4-AP (2 mM) induced a similar increase in perfusion pressure in the Apn 1−/− perfused isolated mesenteric vascular bed, compared to wild-type mice. Removal of perivascular fat increased the vasoconstrictor responses, but abolished the 4-AP effects. The anti-contractile effects of perivascular fat were similar in mesenteric artery and aortic rings from Apn 1−/− and wild-type mice. Despite high adiponectin levels, the anti-contractile effects of perivascular fat were diminished in mesenteric arteries of NZO mice with age.

Conclusion Adiponectin is a novel humoral vasodilator that relaxes aortic and mesenteric rings by opening Kv channels. Similar to the rat, perivascular adipose tissue of the mouse harbors an ADRF, which is malfunctional in NZO mice and is not adiponectin.

Keywords

Key words

Previous SectionNext Section

Time for primary review 30 days

Previous SectionNext Section

1 Introduction

Perivascular fat is routinely removed for isometric vessel contraction studies. Soltis and Cassis showed that perivascular fat significantly attenuates vascular responsiveness of rat isolated aortic ring preparations to norepinephrine [1]. We confirmed the inhibitory action of perivascular fat on vascular contraction in response to several hormonal agonists [2–5]. We also found that this effect is mediated by a transferable “adipocyte-derived relaxing factor” (ADRF) [2] that acts through delayed rectifier K+ (Kv) channel activation in vascular smooth muscle cells [4,5]. The ADRF action is not dependent on the endothelium, the cyclooxygenase or P450 pathway, activation of adenosine receptors, or presence of functional leptin receptors [2]. ADRF release is a Ca2+-dependent process and does not involve neuronal N-type Ca2+ and Na+ channels or vanilloid/cannabinoid and calcitonin gene related peptide receptors, suggesting that ADRF is released from perivascular adipocytes without involvement of perivascular nerves in rats [3]. The molecular identity of ADRF is unknown; however, protein bands with relative masses of 74.0, 59.8, 54.4, 28.7, and 13.8 kDa have been identified that represent candidates [6]. Whether or not perivascular adipose tissue of the mouse also harbors an ADRF is unknown. Adipose tissue produces various biologically active substances including adiponectin. Adiponectin is a 30-kDa protein, which inhibits angiogenesis [7], reduces atherosclerosis in apolipoprotein E-deficient mice [8], and prevents vascular restenosis after angioplasty [9]. Adiponectin serum levels are decreased under conditions of obesity, insulin resistance, type 2 diabetes, and essential hypertension [10–13]. Hypoadiponectinemia is associated with impaired endothelium-dependent vasodilation [14,15]. Hypoadiponectinemia is also found in spontaneously hypertensive rats (SHR), which show reduced anti-contractile effects of perivascular fat in mesenteric artery rings compared to Wistar Kyoto rats (WKY) [16,17]. However, whether or not adiponectin is produced by perivascular adipose tissue and modulates vascular function directly is unknown.

Previous SectionNext Section

2 Methods

2.1 Adiponectin determinations and detection of adiponectin receptors by RT-PCR

Adiponectin was measured in plasma and homogenates of subcutaneous fat, visceral abdominal epidydimal fat, perivascular fat of the superior branch of the mesenteric artery, and perivascular fat of the aorta, by using a specific ELISA kit for rats (B-Bridge International, Inc, Sunnyvale, CA, USA). Adipose tissue was homogenized in RIPA buffer containing sodium fluoride (100 mM), pepstatin (5 μg/ml), leupeptin (5 μg/ml), aprotinin (5 μg/ml), sodium orthovanadate (5 mM), phenylmethylsulphonyl fluoride (2 mM). Homogenates were centrifuged (10 min, 10,000 g, 4 °C) and aqueous phases were used for measuring adiponectin (1 μg total protein per well). Adiponectin concentrations were expressed as ng per microgram of protein.

Gene expression of adiponectin receptors was determined by Real Time RT-PCR in fat-removed mouse mesenteric arteries. Snap frozen tissue was homogenized by ultra sonic disruption (Sonopulse HD 70, Bandelin Electronic, Germany). Total RNA was isolated using mirVana miRNA Isolation Kit (Ambion, Austin, TX, USA) followed by determination of quality and quantity with the Agilent 2100 bioanalyzer (Agilent Technologies, Walbronn, Germany) and reverse transcription (1 μg) into cDNA (Superscript Reverse Transcriptase, Invitrogen, Karlsruhe, Germany). Relative quantitation of gene expression was performed with the 7500 Fast Real Time PCR System (Applied Biosystems, Darmstadt, Germany) using the standard curve method. Rodent glyceraldehyde-3-phosphate dehydrogenase (GAPDH, # 4308313) was chosen as the endogenous control. Ready-to-use TaqMan Gene Expression Assays for mouse adiponectin receptor 1 (AdipoR1, # Mm01291331_m1) and mouse adiponectin receptor 2 (AdipoR2, # Mm01184028_m1) were purchased from Applera Deutschland GmbH (Darmstadt, Germany).

2.2 Isometric contractions

Local authorities approved the studies according to established guidelines. Male Sprague–Dawley rats (200–300 g, 6–8 weeks) were killed under ether. The thoracic aortas were removed, quickly transferred to cold (4 °C) oxygenated (95% O2/5% CO2) physiological salt solution (PSS), and dissected into 5-mm rings. Perivascular fat and connective tissue were either removed ((−) fat) or left intact ((+) fat). After 1-h equilibration, aortic ring contractile force was measured isometrically using standard bath procedures as described. The basal tone was continuously monitored and adjusted to 1 g [2]. The composition of PSS (in mM) was 119 NaCl, 4.7 KCl, 1.2 KH2PO4, 25 NaHCO3, 1.2 Mg2SO4, 11.1 glucose, 1.6 CaCl2. The bath solution was continuously oxygenated with a gas mixture of 95% O2 plus 5% CO2, and kept at 37 °C (pH, 7.4). The bath solution volume was 20 ml.

Male and female wild-type or Apn 1−/− mice (20–25 g, 8–12 weeks) were killed under ether [15]. Male NZO mice were age-matched with male C57BL/6 mice and killed under ether [18,19]. The thoracic aorta and superior branch of the mesenteric artery were excised. The arteries were removed, quickly transferred to cold (4 °C) oxygenated (95% O2/5% CO2) PSS, and dissected into 2-mm rings whereby perivascular fat and connective tissue were either removed or left intact. The perivascular fat was removed with scissors being careful not to damage the adventitia [2,5]. Each ring was positioned between two stainless steel wires (diameter 0.0394 mm) in a 5-ml organ bath of a Small Vessel Myograph (DMT 610M, Danish Myo Technology, Aarhus, Denmark) [5]. The organ bath was filled with PSS. The rings were placed under a tension equivalent to that generated at 0.9 times the diameter of the vessel at 100 mm Hg. This normalization procedure was performed to obtain the passive diameter of the vessel at 100 mm Hg. The software Myodaq 2.01 M610+ was used for data acquisition and display. The rings were precontracted and equilibrated for 60 min until a stable resting tension was acquired. Tension is expressed as a percentage of the steady-state tension (100%) obtained with isotonic external 60 mM KCl. In some arterial rings the endothelial layer was mechanically disrupted by gently rubbing the luminal surface of a ring back and forth several times with scissors. Endothelium integrity or functional removal was confirmed by the presence or absence, respectively, of the relaxant response to 1 μM acetylcholine. Each experiment was carried out on rings prepared from different rats.

2.3 Vascular reactivity in mesenteric vascular beds

The mesenteric vascular bed was perfused [20]. Briefly, the superior mesenteric artery was cannulated at its junction with the abdominal aorta, put on a plastic grid in an organ chamber, and perfused using a peristaltic pump (WPI, Germany) at constant flow (4–5 ml/min) with oxygenated (95% O2–5% CO2) PSS. Perfusion pressure was continuously determined by a pressure transducer (Living Systems Instrumentation, Burlington, VT, USA) and recorded on a polygraph. Since flow was maintained at a constant rate, changes in perfusion pressure were used as an index of changes in the resistance of the arteries, and a rise or a decrease in the perfusion pressure indicated either vasoconstriction or vasodilatation. After an equilibration period of 60 min and before starting the experiment, the vascular mesenteric bed was contracted with KCl of 60 mM to assess its contractility. Subsequently, the mesentery was contracted with serotonin (100 to 300 nM) in the presence or absence of 4-AP (2 mM). In some experiments, the perivascular adipose tissue was carefully removed with scissors without damaging the adventitia using a Nikon stereomicroscope. The mesenteric bed was kept on ice during this procedure.

2.4 Materials and statistics

Human and mouse adiponectin were purchased from BioVendor GmbH (Heidelberg, Germany). Glibenclamide, apamin, serotonin, and iberiotoxin were obtained from Sigma (Taufkirchen, Germany). 4-Aminopyridine and tetraethylammonium chloride (TEA+) were from Merck Biosciences (former Calbiochem) (Schwalbach, Germany). All drugs were applied by adding to the bath solution (PSS). The adiponectin ELISA kit was from B-Bridge International, Inc. (Sunnyvale, CA, USA). Drs. S. Kihara and K. Walsh (Boston University School of Medicine, Boston, MA, USA) supplied the Apn 1−/− mice. All values are given as mean±SEM. Paired and unpaired Student's t tests or ANOVA were used as appropriate. A value of p<0.05 was considered statistically significant; n represents the number of arteries or samples tested. The investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health and the ethics policies of our university and the Land Berlin.

Previous SectionNext Section

3 Results

3.1 Adiponectin levels in serum and in perivascular adipose tissue

Serum adiponectin levels were 2.5±0.74 μg/ml in rats. Adiponectin levels were also measured in various rat adipose tissues. In subcutaneous fat of rats, the concentration of adiponectin was 0.223±0.109 ng/μg total protein. Similar adiponectin levels were detected in abdominal visceral and perivascular fat. The adiponectin concentrations were 0.255±0.069 ng/μg total protein, 0.220±0.053 ng/μg total protein, and 0.249±0.026 ng/μg total protein in abdominal visceral fat, in perivascular fat of the superior branch of the mesenteric artery, and in periaortic adipose tissue, respectively (p>0.05 for each comparison, n=4–6 each). These data suggest that subcutaneous adipose tissue is probably not the only source for adiponectin in the body. Instead, the substance may also be produced by abdominal visceral and perivascular adipose tissue.

3.2 Adiponectin is a novel humoral vasodilator

Serotonin produced stable contractions of aortic rings in the absence of perivascular fat with half-maximal effects at 2 μM [2]. The contractions were stable within 45 min (Fig. 1A) and reversible after serotonin washout [2]. Recombinant human adiponectin at serum levels (3 μg/ml) inhibited serotonin-dependent contractions of rat aortas (Fig. 1B, n=10). Half-maximal effects were observed at 7.9±2.2 μg/ml (IC50, n=6, Fig. 1E). Removal of endothelium did not affect the effects of adiponectin (IC50, 6.9±0.5 μg/ml, n=6, Fig. 1E). Adiponectin effects (3 μg/ml) were abolished by Kv channel inhibition with 4-AP. Fig. 1B shows that 4-AP (2 mM) reversed the vasorelaxant effects of adiponectin with an overshoot. 4-AP increased force to ∼250% of precontracted values by serotonin (100%) (n=6). However, similar increases in force by 4-AP were observed in the absence of adiponectin (n=6, p>0.05). Pretreatment of aortic rings with 4-AP (2 mM) prevented the vasorelaxant effects of adiponectin (Fig. 1C); adiponectin (3 μg/ml) produced vasorelaxation by 67.7±5.0% (n=10) and 4.4±1.3% (n=5) in the absence and presence of 4-AP, respectively (p<0.05). Inhibition of ATP-dependent (KATP) K+ channels with glibenclamide (3 μM) did not affect the vasorelaxant effects of adiponectin (Fig. 1D, n=4). Similar effects were observed in mesenteric arteries and aortas of mice. Fig. 2A and B shows that recombinant mouse adiponectin at serum levels (5 μg/ml) inhibited serotonin-dependent contractions of mouse mesenteric arteries (n≥9). The effects were inhibited by 4-AP (2 mM). These data show that adiponectin represents a novel potent arterial vasodilator in both rats and mice. Furthermore, the data suggest that vasorelaxation by adiponectin is produced by opening KV channels. Since Kv channels in arterial smooth muscle cells are targets for ADRF [4,5] and adiponectin levels are high in perivascular adipose tissue, we next investigated whether or not adiponectin is ADRF. We used Apn 1−/− mice to study the role of adiponectin in the paracrine control of vascular tone by perivascular adipose tissue.

Fig. 2
View larger version:
Fig. 2

Serotonin-dependent vasocontractions are reduced in the presence of adiponectin in mouse mesenteric arteries. Perivascular fat was removed. Panel A. Original recordings of wild-type arteries. Horizontal bars show the presence of the drugs. Panel B. Summary of the results for wild-type arteries. *, p<0.05. Panel C: Summary of the results for arteries of 25-week-old NZO mice. *, p<0.05.

Fig. 1
View larger version:
Fig. 1

Relaxation of rat aortic rings by adiponectin is shown. Panel A. Serotonin produced stable vasocontraction in an aortic ring without perivascular fat ((−) fat). Panel B. Reversal of adiponectin-dependent vasorelaxation by 4-aminopyridine (4-AP) in a (−) fat aortic ring. Panel C. Pretreatment of the vessel with 4-AP prevented adiponectin-dependent vasorelaxation. Panel D. Glibenclamide did not inhibit adiponectin-dependent vasorelaxation. Horizontal bars show the presence of the drugs. Panel E. Dose–response curves for the effects of adiponectin on rat aortic rings with (closed circles) and without endothelium (open squares). Tension was expressed as a percentage of the steady-state tension (100%) obtained with serotonin (2 μM).

3.3 Adiponectin is not ADRF

The mesenteric vascular bed of wild-type mice was first cannulated and perfused at a constant flow of 4–5 ml/min in the absence and presence of perivascular adipose tissue (Fig. 3). KCl (60 mM) induced similar increases in perfusion pressure in the presence ((+) fat) (n=10) and absence ((−) fat) (n=7) of perivascular fat (Fig. 3A and B). However, serotonin (100 and 300 nM) induced stronger increases in perfusion pressure in isolated (−) fat mesenteric vascular beds, compared to (+) fat mesenteric vascular beds of wild-type mice (Fig. 3A and C).

Fig. 3
View larger version:
Fig. 3

Perivascular fat limits vascular reactivity to serotonin in mouse mesenteric vascular beds. Panel A. Shown are original recordings of perfusion pressure for perfused isolated mesenteric vascular beds in the absence (fat −) and presence (fat +) of perivascular fat. Dashed lines represent 30 mm Hg. Horizontal bars show the presence of the drugs. The perivascular fat was carefully removed in fat − beds. The left panels show fat − and fat + mesenteric vascular beds. Panel B. Increase in perfusion pressure by application of KCl (60 mM). Panel C. Increase in perfusion pressure by external application of serotonin. All values are normalized to 60 mM KCl responses (100%). WT, wild-type mice; Apn 1−/−, adiponectin deficient mice. *, p<0.05; n.s., p≥0.05.

We next cannulated and perfused (+) fat mesenteric vascular beds of wild-type and Apn 1−/− mice at a constant flow of 4–5 ml/min (Fig. 4). KCl (60 mM) induced similar increases in perfusion pressure in (+) fat isolated mesenteric vascular beds of wild-type (n=10) and Apn 1−/− mice (n=5) (Figs. 4A and 3B). In addition, serotonin (100 and 300 nM) induced similar increases in perfusion pressure in (+) fat isolated mesenteric vascular beds of wild-type and Apn 1−/− mice (Figs. 4A and 3C). Kv channel inhibition with 4-AP (2 mM) increased perfusion pressure in (+) fat mesenteric vascular beds of both Apn 1−/− mice and wild-type mice (Fig. 4B). In contrast, 4-AP (2 mM) did not increase perfusion pressure in (−) fat mesenteric beds of wild-type mice (Fig. 4B).

Fig. 4
View larger version:
Fig. 4

No change in vascular reactivity to serotonin in mesenteric vascular beds of Apn 1−/− mice. Panel A. Shown are original recordings of perfusion pressure for perfused isolated mesenteric vascular beds. Dashed lines represent 30 mm Hg. Horizontal bars show the presence of the drugs. Panel B. Effects of 4-aminopyridine (4-AP) on perfusion pressure in the presence of serotonin (5-HT). WT, wild-type; Apn 1−/−, adiponectin deficient mice. *, p<0.05; n.s., p≥0.05.

We then performed isometric contraction measurements of isolated aortic and mesenteric arterial rings and generated dose–response curves to serotonin for both rings with (+) fat and without (−) fat (Fig. 5). KCl produced similar contractions of (−) fat and (+) fat mesenteric rings of wild-type mice (Fig. 5A). However, the contractile response to serotonin was markedly reduced in intact (+) fat vessels compared to (−) fat vessels (Fig. 5B). Similar effects were observed in mesenteric rings from Apn 1−/− mice (Fig. 5C and D). Kv channel inhibition with 4-AP (2 mM) abrogated the anti-contractile effects of perivascular fat and induced similar increases in the response of (+) fat mesenteric rings of Apn 1−/− mice to serotonin, compared to wild-type mice. Length–tension curves of differently sized mesenteric rings were not influenced by the presence of perivascular fat (Fig. 1 online). The small-conductance Ca2+-activated K+ channel blocker apamin [2,3,5] (1 μM, n=6), the KATP channel blocker glibenclamide [5] (3 μM, n=8), the inward rectifier K+ channel blocker Ba2+ (3 μM, n=3) [5,21], and blockers of large-conductance Ca2+-activated potassium (BK) channels, i.e. iberiotoxin [22,23] (100 nM, n=6) and TEA+ [22,24], (1 mM, n=6) did not influence the anti-contractile effect of fat in mesenteric rings of wild-type mice. In addition, the anti-contractile effects of perivascular fat were similar between mesenteric rings from BKβ1-deficient mice (n=4), lacking the β1 subunit of the BK channel [25].

Fig. 5
View larger version:
Fig. 5

The anti-contractile effects of perivascular fat in mesenteric arteries occur in the absence of adiponectin and are inhibited by 4-aminopyridine (4-AP). Panels A, C. Reducing the difference between the K+ equilibrium potential and the membrane potential by application of isotonic bath solution containing 60 mM KCl abolished the difference in contractile response between (+) fat (open bar) and (−) fat (filled bar) rings. Panels B, D. Dose–response curves to serotonin in intact (+) fat (open circles), (+) fat in the presence of 2 mM 4-AP (filled triangles), and (−) fat preparations without periadventitial fat (closed squares). Serotonin induced stronger contractions in (–) fat rings, compared to (+) fat rings of both wild-type and Apn 1−/− mice (*). 4-AP elevated the contractions in (+) fat isolated mesenteric rings of both wild-type (WT) and Apn 1−/− mice (**). *, p<0.05; **, p<0.05; n.s., p≥0.05.

Similar anti-contractile effects of perivascular fat were observed in wild-type and Apn 1−/− mouse aortas. KCl produced similar contractions of (−) fat and (+) fat rings of both wild-type and Apn 1−/− mice. The contractile response to serotonin was markedly reduced in intact (+) fat vessels compared to (−) fat vessels (Fig. 2 online).

3.4 Anti-contractile effects of perivascular fat decrease with age in NZO mice, but are unrelated to adiponectin

Dose–response curves to serotonin were determined in (−) fat and (+) fat mesenteric rings isolated from 20-week-, 25-week-, and 34-week-old NZO mice. As previously reported [18,19,26], body weight and visceral obesity of NZO mice increased with age (Fig. 3 online). The total body weight of the animals was 30.0±1.0 g, 39.0±1.3 g, and 59.3±1.9 g, and the mesentery wet weight was 0.40±0.10 g, 0.63±0.04 g, and 1.90±0.08 g, respectively.

Serotonin induced similar contractions in (–) fat rings from 20-, 25- and 34-week-old mice (Fig. 6A). The contractile response to serotonin was markedly reduced in intact (+) fat vessels compared to (−) fat vessels at all ages. However, these anti-contractile effects decreased with age (Fig. 6B). Notably, the reduction in the anti-contractile effects was not associated with decreased adiponectin levels in NZO mice. Instead, serum adiponectin levels showed a tendency to increase with age. Serum adiponectin levels were 20.3±3.6 μg/ml in 48-week-old NZO mice (n=6), which is higher than in 12-week-old NZO mice. Serum adiponectin levels were 15.7±0.87 μg/ml in 12-week-old NZO mice, which is similar to 12-week-old and 48-week-old wild-type mice (16.0±2.2 μg/ml, and 14.3±3.6 μg/ml, respectively, n=6 each).

Fig. 6
View larger version:
Fig. 6

The anti-contractile effects of perivascular fat in mesenteric arteries of NZO mice decreased with age. Panel A. Dose–response curves to serotonin in (−) fat preparations without periadventitial fat isolated from 20-, 25-, and 34-week-old NZO mice. Serotonin induced similar contractions in (–) fat rings from 20-, 25-, and 34-week-old NZO mice. The perivascular fat was carefully removed; the left panel shows a (−) fat mesenteric ring in the bath chamber. Panel B. Dose–response curves to serotonin in intact (+) fat preparations. Serotonin induced stronger contractions in (+) fat rings of 34-week-old mice, compared to (+) fat rings of 25-week-old mice (*). Serotonin induced stronger contractions in (+) fat rings of 25-week-old mice, compared to 20-week-old mice (*). The perivascular fat was left intact. The left panel shows a (+) fat mesenteric ring. *, p<0.05.

In addition, adiponectin levels were similar in 20-, 25-, and 34-week-old NZO mice (18.2±8.0, 20.0±4.8, 15.1±8.2 μg/ml, n=5 each). RT-PCR revealed the expression of the adiponectin receptors AdipoR1 and AdipoR2 mRNA in fat-removed NZO mouse mesenteric arteries; however, with no difference in the degree of expression level between 20- and 34-week-old NZO mice (not shown). Despite diminished anti-contractile effects to perivascular fat (Fig. 6), (−) fat mesenteric rings of NZO mice showed normal vasodilatory responses to recombinant mouse adiponectin (5 μg/ml) (Fig. 2C; n≥5). The vasodilatory effects were similar compared to control mice (p<0.05).

Previous SectionNext Section

4 Discussion

Our study produced four novel findings. First, we found that serum concentrations of adiponectin (IC50, ∼8 μg/ml) produced vasorelaxation of rat aorta and mouse mesenteric arteries. The effects were not dependent on the endothelium and may be mediated AdipoR1 and/or AdipoR2. 4-AP inhibited the vasorelaxation, suggesting that opening of Kv channels was involved. Our report is the first to show a direct vasodilating action of adiponectin and to demonstrate a mechanism for this effect. Second, we demonstrate that like the rat, the mouse also exhibits ADRF. Third, despite high-normal adiponectin levels and more perivascular fat, the anti-contractile effects of perivascular fat were diminished in mesenteric arteries of NZO mice, suggesting malfunction of ADRF in this model. Fourth, our investigations of Apn 1−/− mice suggest that adiponectin is not ADRF.

We demonstrated earlier that periadventitial adipose tissue markedly attenuates the contractile response to serotonin, phenylephrine, and angiotensin II in aortic and mesenteric ring preparations of rats [2,5]. Inhibition of the contractile response by fat depended on the amount of fat on each ring. The effects were not dependent on the endothelium [2,5] We and others showed that periadventitial fat appears to release ADRF into the organ bath that induces vasorelaxation by opening smooth muscle K+ channels [3,6,27], with a major role of Kv channels in mesenteric arteries [5]. Malfunction of this pathway may result to increased vascular reactivity in obesity and hypertension [4,28]. In accord, the anti-contractile effects of perivascular fat are reduced in SHR mesenteric artery rings compared with WKY rings [17]. Noteworthy, serum adiponectin levels are low in SHR compared to WKY [16]. Furthermore, Apn 1−/− mice develop hypertension on an atherogenic diet [14], which is aggravated by salt intake [29]. These findings prompted us to suggest that adiponectin might be involved in the reduced paracrine effects of perivascular fat.

We found that adiponectin levels in rat serum were ∼2.5 μg/ml, as previously reported for rats [30] and humans [31,32]. We studied the effects of adiponectin at these concentrations on vascular ring preparations and observed that serum concentrations of adiponectin relaxed rat aortas (IC50, ∼8 μg/l). The effects were reversible upon adiponectin washout, independent of the endothelium, and inhibited by 4-AP. Similar effects were observed in mouse mesenteric arteries. These results suggest that adiponectin represents a novel vasodilator in rats and mice at physiological concentrations that produces vasorelaxation by opening Kv channels.

We next studied adiponectin levels in different adipose tissues of rats. In subcutaneous fat of rats, the concentration of adiponectin was ∼0.2 ng/μg total protein, which is similar to adiponectin levels in human subcutaneous adipose tissue [33,34]. We found that adiponectin concentrations were similar in perivascular fat of the superior branch of the mesenteric artery and perivascular fat of the aorta, compared to visceral abdominal fat and subcutaneous fat. These results suggest that perivascular adipose tissue might serve as a paracrine adiponectin source to control vascular tone. Adiponectin represents a possible ADRF candidate because of its presence in perivascular adipose tissue, its ability to induce vasorelaxation of aortic and mesenteric rings, and its effects on Kv channels that are blocked by 4-AP. This notion is also supported by detection of a 28.7-kDa protein band in ADRF-containing bath solutions of rat aorta, which is close to the molecular mass of adiponectin (30 kDa) [6].

To test adiponectin as possible ADRF candidate, we studied Apn 1−/− mice. We perfused the isolated mesenteric vascular bed and at a constant flow in the absence and presence of serotonin. Inhibition of Kv channels with 4-AP induced a similar increase in perfusion pressure by serotonin in mesenteric vascular bed of Apn 1−/− mice, compared to wild-type mice. Our experiments with fat-removed mesenteric vascular beds showed that these effects were solely dependent on the presence of perivascular fat. Our report is the first to show a direct vasodilating action of perivascular fat on resistance vessels exhibiting myogenic tone [35] in a perfused organ and under flow. We next studied isometric contractions of (+) fat and (−) fat mesenteric arterial rings from Apn 1−/− mice and wild-type mice. We found that periadventitial adipose tissue markedly attenuated the contractile response to serotonin. The attenuation was not different between mesenteric rings from Apn 1−/− mice and from wild-type mice. Similar effects were observed in aortas from Apn 1−/− and wild-type mice. Furthermore, the anti-contractile effect of perivascular fat was antagonized by 4-AP. Blockers of other K+ channels (BK, Kir, KATP) in smooth muscle cells were not effective. The effects of 4-AP were not different between rings from Apn 1−/− mice and from wild-type mice. These results indicate that perivascular adipose tissue of the mouse harbors an ADRF, which is similar to ADRF in rats [5,17]. Despite high-normal adiponectin levels and more perivascular fat (Fig. 3 online), the anti-contractile effects of perivascular fat were diminished in mesenteric arteries of NZO mice, which have a severe metabolic syndrome associated with hypertension [18,19]. At first glance, the high-normal levels of adiponectin in NZO mice were surprising. However, hypoadiponectinemia is not observed in all human populations and rodent models with obesity, type 2 diabetes, and/or hypertension [36–39]. Nevertheless, the discordance between the adiponectin levels and anti-contractile effects of perivascular fat in NZO mice suggest that ADRF is not adiponectin. Moreover, the response of NZO mesenteric rings to adiponectin was normal. However, our data clearly demonstrate that ADRF is malfunctional in NZO mice. Therefore, we suggest that differences in visceral perivascular adipose tissue function may contribute to the increased vascular resistance observed in NZO mice.

Release of adiponectin is of interest in various pathophysiological conditions, including hypoadiponectinemia, obesity, and obesity-associated hypertension [11,40,41]. Our results demonstrate that NZO mice feature a metabolic syndrome without hypoadiponectinemia. Our results also demonstrate that the paracrine control of vascular tone by perivascular adipose tissue is disturbed in these mice and is not mediated by adiponectin but possibly by other substances released from adipocytes or other adventitial cells. The study of Apn 1−/− mice has advanced our search since we found a hitherto fore not appreciated vasodilatory mechanism for adiponectin. Nevertheless, the role and nature of ADRF in health and cardiovascular disease remains to be identified.

Previous SectionNext Section

Supplementary data

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.cardiores.2007.05.025.

Previous SectionNext Section

Acknowledgements

We thank Diana Herold for technical assistance. The Deutsche Forschungsgemeinschaft, Deutsche Akademische Austauschdienst and Research Grant Council of Hong Kong supported this work. Dr. Shinji Kihara and Dr. Kenneth Walsh kindly provided Apn 1−/− and wild-type mice.

Previous Section
 

References

  1. [1]↵
    1. Soltis E.E.,
    2. Cassis L.A.
    (1991) Influence of perivascular adipose tissue on rat aortic smooth muscle responsiveness. Clin Exp Hypertens A 13:277–296.
    MedlineWeb of Science
  2. [2]↵
    1. Lohn M.,
    2. Dubrovska G.,
    3. Lauterbach B.,
    4. Luft F.C.,
    5. Gollasch M.,
    6. Sharma A.M.
    (2002) Periadventitial fat releases a vascular relaxing factor. FASEB J 16:1057–1063.
    Abstract/FREE Full Text
  3. [3]↵
    1. Dubrovska G.,
    2. Verlohren S.,
    3. Luft F.C.,
    4. Gollasch M.
    (2004) Mechanisms of ADRF release from rat aortic adventitial adipose tissue. Am J Physiol Heart Circ Physiol 286:H1107–H1113.
    Abstract/FREE Full Text
  4. [4]↵
    1. Gollasch M.,
    2. Dubrovska G.
    (2004) Paracrine role for periadventitial adipose tissue in the regulation of arterial tone. Trends Pharmacol Sci 25:647–653.
    CrossRefMedline
  5. [5]↵
    1. Verlohren S.,
    2. Dubrovska G.,
    3. Tsang S.Y.,
    4. Essin K.,
    5. Luft F.C.,
    6. Huang Y.,
    7. et al.
    (2004) Visceral periadventitial adipose tissue regulates arterial tone of mesenteric arteries. Hypertension 44:271–276.
    Abstract/FREE Full Text
  6. [6]↵
    1. Yang L.,
    2. Hu B.R.,
    3. Xiang J.Z.,
    4. Wang J.L.
    (2005) Adventitium-derived relaxing factor may be a protein factor secreted by adipocytes with non-species-specificity and not limited to periadventitial fat. Chin J Pharmacol Toxicol 19:401–406.
  7. [7]↵
    1. Brakenhielm E.,
    2. Veitonmaki N.,
    3. Cao R.,
    4. Kihara S.,
    5. Matsuzawa Y.,
    6. Zhivotovsky B.,
    7. et al.
    (2004) Adiponectin-induced antiangiogenesis and antitumor activity involve caspase-mediated endothelial cell apoptosis. Proc Natl Acad Sci U S A 101:2476–2481.
    Abstract/FREE Full Text
  8. [8]↵
    1. Okamoto Y.,
    2. Kihara S.,
    3. Ouchi N.,
    4. Nishida M.,
    5. Arita Y.,
    6. Kumada M.,
    7. et al.
    (2002) Adiponectin reduces atherosclerosis in apolipoprotein E-deficient mice. Circulation 106:2767–2770.
    Abstract/FREE Full Text
  9. [9]↵
    1. Matsuda M.,
    2. Shimomura I.,
    3. Sata M.,
    4. Arita Y.,
    5. Nishida M.,
    6. Maeda N.,
    7. et al.
    (2002) Role of adiponectin in preventing vascular stenosis. The missing link of adipo-vascular axis. J Biol Chem 277:37487–37491.
    Abstract/FREE Full Text
  10. [10]↵
    1. Iwashima Y.,
    2. Katsuya T.,
    3. Ishikawa K.,
    4. Ouchi N.,
    5. Ohishi M.,
    6. Sugimoto K.,
    7. et al.
    (2004) Hypoadiponectinemia is an independent risk factor for hypertension. Hypertension 43:1318–1323.
    Abstract/FREE Full Text
  11. [11]↵
    1. Adamczak M.,
    2. Wiecek A.,
    3. Funahashi T.,
    4. Chudek J.,
    5. Kokot F.,
    6. Matsuzawa Y.
    (2003) Decreased plasma adiponectin concentration in patients with essential hypertension. Am J Hypertens 16:72–75.
    CrossRefMedlineWeb of Science
  12. [12]↵
    1. Okamoto Y.,
    2. Kihara S.,
    3. Funahashi T.,
    4. Matsuzawa Y.,
    5. Libby P.
    (2006) Adiponectin: a key adipocytokine in metabolic syndrome. Clin Sci (Lond) 110:267–278.
    Medline
  13. [13]↵
    1. Matsuzawa Y.
    (2006) Therapy insight: adipocytokines in metabolic syndrome and related cardiovascular disease. Nat Clin Pract Cardiovasc Med 3:35–42.
    CrossRefMedlineWeb of Science
  14. [14]↵
    1. Ouchi N.,
    2. Ohishi M.,
    3. Kihara S.,
    4. Funahashi T.,
    5. Nakamura T.,
    6. Nagaretani H.,
    7. et al.
    (2003) Association of hypoadiponectinemia with impaired vasoreactivity. Hypertension 42:231–234.
    Abstract/FREE Full Text
  15. [15]↵
    1. Maeda N.,
    2. Shimomura I.,
    3. Kishida K.,
    4. Nishizawa H.,
    5. Matsuda M.,
    6. Nagaretani H.,
    7. et al.
    (2002) Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8:731–737.
    CrossRefMedlineWeb of Science
  16. [16]↵
    1. Potenza M.A.,
    2. Marasciulo F.L.,
    3. Tarquinio M.,
    4. Quon M.J.,
    5. Montagnani M.
    (2006) Treatment of spontaneously hypertensive rats with rosiglitazone and/or enalapril restores balance between vasodilator and vasoconstrictor actions of insulin with simultaneous improvement in hypertension and insulin resistance. Diabetes 55:3594–3603.
    Abstract/FREE Full Text
  17. [17]↵
    1. Galvez B.,
    2. de Castro J.,
    3. Herold D.,
    4. Dubrovska G.,
    5. Arribas S.,
    6. Gonzalez M.C.,
    7. et al.
    (2006) Perivascular adipose tissue and mesenteric vascular function in spontaneously hypertensive rats. Arterioscler Thromb Vasc Biol 26:1297–1302.
    Abstract/FREE Full Text
  18. [18]↵
    1. Ortlepp J.R.,
    2. Kluge R.,
    3. Giesen K.,
    4. Plum L.,
    5. Radke P.,
    6. Hanrath P.,
    7. et al.
    (2000) A metabolic syndrome of hypertension, hyperinsulinaemia and hypercholesterolaemia in the New Zealand obese mouse. Eur J Clin Invest 30:195–202.
    CrossRefMedlineWeb of Science
  19. [19]↵
    1. Jurgens H.S.,
    2. Schurmann A.,
    3. Kluge R.,
    4. Ortmann S.,
    5. Klaus S.,
    6. Joost H.G.,
    7. et al.
    (2006) Hyperphagia, lower body temperature, and reduced running wheel activity precede development of morbid obesity in New Zealand obese mice. Physiol Genomics 25:234–241.
    Abstract/FREE Full Text
  20. [20]↵
    1. McGregor D.D.
    (1965) The effect of sympathetic nerve stimulation of vasoconstrictor responses in perfused mesenteric blood vessels of the rat. J Physiol 177:21–30.
    FREE Full Text
  21. [21]↵
    1. Robertson B.E.,
    2. Bonev A.D.,
    3. Nelson M.T.
    (1996) Inward rectifier K+ currents in smooth muscle cells from rat coronary arteries: block by Mg2+, Ca2+, and Ba2+. Am J Physiol 271:H696–H705.
    Medline
  22. [22]↵
    1. Gollasch M.,
    2. Ried C.,
    3. Bychkov R.,
    4. Luft F.C.,
    5. Haller H.
    (1996) K+ currents in human coronary artery vascular smooth muscle cells. Circ Res 78:676–688.
    Abstract/FREE Full Text
  23. [23]↵
    1. Rusch N.J.,
    2. Liu Y.,
    3. Pleyte K.A.
    (1996) Mechanisms for regulation of arterial tone by Ca2+-dependent K+ channels in hypertension. Clin Exp Pharmacol Physiol 23:1077–1081.
    MedlineWeb of Science
  24. [24]↵
    1. Nelson M.T.,
    2. Patlak J.B.,
    3. Worley J.F.,
    4. Standen N.B.
    (1990) Calcium channels, potassium channels, and voltage dependence of arterial smooth muscle tone. Am J Physiol 259:C3–C18.
    MedlineWeb of Science
  25. [25]↵
    1. Pluger S.,
    2. Faulhaber J.,
    3. Furstenau M.,
    4. Lohn M.,
    5. Waldschutz R.,
    6. Gollasch M.,
    7. et al.
    (2000) Mice with disrupted BK channel beta1 subunit gene feature abnormal Ca(2+) spark/STOC coupling and elevated blood pressure. Circ Res 87:E53–E60.
    MedlineWeb of Science
  26. [26]↵
    1. Kluge R.,
    2. Giesen K.,
    3. Bahrenberg G.,
    4. Plum L.,
    5. Ortlepp J.R.,
    6. Joost H.G.
    (2000) Quantitative trait loci for obesity and insulin resistance (Nob1, Nob2) and their interaction with the leptin receptor allele (LeprA720T/T1044I) in New Zealand obese mice. Diabetologia 43:1565–1572.
    CrossRefMedlineWeb of Science
  27. [27]↵
    1. Gao Y.J.,
    2. Zeng Z.H.,
    3. Teoh K.,
    4. Sharma A.M.,
    5. Abouzahr L.,
    6. Cybulsky I.,
    7. et al.
    (2005) Perivascular adipose tissue modulates vascular function in the human internal thoracic artery. J Thorac Cardiovasc Surg 130:1130–1136.
    Abstract/FREE Full Text
  28. [28]↵
    1. Gao Y.J.,
    2. Holloway A.C.,
    3. Zeng Z.H.,
    4. Lim G.E.,
    5. Petrik J.J.,
    6. Foster W.G.,
    7. et al.
    (2005) Prenatal exposure to nicotine causes postnatal obesity and altered perivascular adipose tissue function. Obes Res 13:687–692.
    MedlineWeb of Science
  29. [29]↵
    1. Ohashi K.,
    2. Kihara S.,
    3. Ouchi N.,
    4. Kumada M.,
    5. Fujita K.,
    6. Hiuge A.,
    7. et al.
    (2006) Adiponectin replenishment ameliorates obesity-related hypertension. Hypertension 47:1108–1116.
    Abstract/FREE Full Text
  30. [30]↵
    1. Hattori Y.,
    2. Akimoto K.,
    3. Gross S.S.,
    4. Hattori S.,
    5. Kasai K.
    (2005) Angiotensin-II-induced oxidative stress elicits hypoadiponectinaemia in rats. Diabetologia 48:1066–1074.
    CrossRefMedlineWeb of Science
  31. [31]↵
    1. Ogawa Y.,
    2. Kikuchi T.,
    3. Nagasaki K.,
    4. Hiura M.,
    5. Tanaka Y.,
    6. Uchiyama M.
    (2005) Usefulness of serum adiponectin level as a diagnostic marker of metabolic syndrome in obese Japanese children. Hypertens Res 28:51–57.
    CrossRefMedlineWeb of Science
  32. [32]↵
    1. Park K.G.,
    2. Park K.S.,
    3. Kim M.J.,
    4. Kim H.S.,
    5. Suh Y.S.,
    6. Ahn J.D.,
    7. et al.
    (2004) Relationship between serum adiponectin and leptin concentrations and body fat distribution. Diabetes Res Clin Pract 63:135–142.
    CrossRefMedlineWeb of Science
  33. [33]↵
    1. Bullo M.,
    2. Salas-Salvado J.,
    3. Garcia-Lorda P.
    (2005) Adiponectin expression and adipose tissue lipolytic activity in lean and obese women. Obes Surg 15:382–386.
    CrossRefMedlineWeb of Science
  34. [34]↵
    1. Sabin M.A.,
    2. Holly J.M.,
    3. Shield J.P.,
    4. Turner S.J.,
    5. Grohmann M.J.,
    6. Stewart C.E.,
    7. et al.
    (2006) Mature subcutaneous and visceral adipocyte concentrations of adiponectin are highly correlated in prepubertal children and inversely related to body mass index standard deviation score. J Clin Endocrinol Metab 91:332–335.
    Abstract/FREE Full Text
  35. [35]↵
    1. Koop E.A.,
    2. Gebbink M.F.,
    3. Sweeney T.E.,
    4. Mathy M.J.,
    5. Heijnen H.F.,
    6. Spaan J.A.,
    7. et al.
    (2005) Impaired flow-induced dilation in mesenteric resistance arteries from receptor protein tyrosine phosphatase-mu-deficient mice. Am J Physiol Heart Circ Physiol 288:H1218–H1223.
    Abstract/FREE Full Text
  36. [36]↵
    1. de Assis S.,
    2. Wang M.,
    3. Goel S.,
    4. Foxworth A.,
    5. Helferich W.,
    6. Hilakivi-Clarke L.
    (2006) Excessive weight gain during pregnancy increases carcinogen-induced mammary tumorigenesis in Sprague–Dawley and lean and obese Zucker rats. J Nutr 136:998–1004.
    Abstract/FREE Full Text
  37. [37]↵
    1. Hussein G.,
    2. Nakagawa T.,
    3. Goto H.,
    4. Shimada Y.,
    5. Matsumoto K.,
    6. Sankawa U.,
    7. et al.
    (2007) Astaxanthin ameliorates features of metabolic syndrome in SHR/NDmcr-cp. Life Sci 80:522–529.
    CrossRefMedlineWeb of Science
  38. [38]↵
    1. Kennedy L.,
    2. Bittel D.C.,
    3. Kibiryeva N.,
    4. Kalra S.P.,
    5. Torto R.,
    6. Butler M.G.
    (2006) Circulating adiponectin levels, body composition and obesity-related variables in Prader–Willi syndrome: comparison with obese subjects. Int J Obes (Lond) 30:382–387.
    CrossRefMedline
  39. [39]↵
    1. Huypens P.,
    2. Quartier E.
    (2006) Adiponectin expression is paradoxically increased in gold-thioglucose-induced obesity. Horm Metab Res 38:486–490.
    CrossRefMedlineWeb of Science
  40. [40]↵
    1. Zoccali C.,
    2. Mallamaci F.,
    3. Panuccio V.,
    4. Tripepi G.,
    5. Cutrupi S.,
    6. Parlongo S.,
    7. et al.
    (2003) Adiponectin is markedly increased in patients with nephrotic syndrome and is related to metabolic risk factors. Kidney Int Suppl, S98–S102.
  41. [41]↵
    1. Huang K.C.,
    2. Chen C.L.,
    3. Chuang L.M.,
    4. Ho S.R.,
    5. Tai T.Y.,
    6. Yang W.S.
    (2003) Plasma adiponectin levels and blood pressures in nondiabetic adolescent females. J Clin Endocrinol Metab 88:4130–4134.
    Abstract/FREE Full Text
« Previous | Next Article »Table of Contents

This Article

  1. Cardiovasc Res (2007) 75 (4): 719-727. doi: 10.1016/j.cardiores.2007.05.025
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free
  4. Supplementary Data

Responses

  1. submit an e-letter
  2. No responses published

Navigate This Article

Current Content

  1. August 1, 2010, 87 (3)
  1. Current Issue
  1. Alert me to new issues