Dahl SS rats demonstrate enhanced aortic perivascular adipose tissue-mediated buffering of vasoconstriction through activation of NOS in the endothelium

Abstract

Perivascular adipose tissue (PVAT) mediates buffering of vasoconstriction through activation of endothelium-derived factors. We hypothesized that the PVAT of Dahl salt-sensitive (Dahl SS) rats has reduced ability to buffer vasoconstriction. Vascular reactivity experiments were performed on aortic rings with PVAT intact (+PVAT) or removed (−PVAT), and endothelium intact (+ENDO) or removed (−ENDO) from Dahl SS rats and control SS.13^BN rats (Dahl SS rats that have had chromosome 13 completely replaced with that of the Brown Norway rat, rendering this strain insensitive to high-salt or high-fat diet-induced hypertension). Endothelial dysfunction, assessed by ACh-mediated vasorelaxation, was confirmed in aortic rings of Dahl SS rats. The +PVAT+ENDO aortic rings had indistinguishable phenylephrine-induced vasoconstriction between genotypes. In both strains, removal of PVAT significantly enhanced vasoconstriction. Dahl SS rat −PVAT+ENDO aortic rings displayed exaggerated vasoconstriction to phenylephrine vs. SS.13^BN rats, indicating that PVAT-mediated buffering of vasoconstriction was greater in Dahl SS rats. Removal of both the ENDO and PVAT restored vasoconstriction in both strains. The nitric oxide synthase (NOS) inhibitor, N^ω-nitro-l-arginine methyl ester (l-NAME), produced a similar effect as that seen with −ENDO. These data indicate that the function of the PVAT to activate endothelium-derived NOS is enhanced in Dahl SS compared with SS.13^BN rats and, most likely, occurs through a pathway that is distinct from ACh-mediated activation of NOS. PVAT weight and total PVAT leptin levels were greater in Dahl SS rats. Leptin induced a significantly decreased vasoconstriction in −PVAT+ENDO aortic rings from Dahl SS rats, but not SS.13^BN rats. In contrast to our initial hypothesis, PVAT in Dahl SS rats buffers vasoconstriction by activating endothelial NOS via mechanisms that may include the involvement of leptin. Thus, the PVAT serves a vasoprotective role in Dahl SS rats on normal-salt diet.

the blood vessel wall has the capability to regulate vascular smooth muscle tone. Notably, the endothelium, which is the single-cell lining that covers the lumen of all blood vessels, is known for its role in maintaining vascular tone homeostasis (17). This is evidenced by findings that endothelial dysfunction or removal of the endothelium exaggerates responsiveness of the vessel to numerous vasoconstrictor agents (18, 36). Furthermore, particularly in large arteries like the aorta, the endothelium buffers vascular smooth muscle reactivity mainly via the action of nitric oxide (NO) produced by endothelial NO synthase (NOS3, eNOS) (2, 38).

More recently, it has become apparent that the endothelium is not the only component in the vessel wall that is capable of regulating smooth muscle tone. The perivascular adipose tissue (PVAT), once thought to be functionally benign, has been shown to modulate vascular tone in an endothelium-derived NOS-dependent manner (6, 22). Studies from humans and animal models have demonstrated that isolated vascular rings with both the PVAT and the endothelium intact (+PVAT+ENDO) display blunted vasoconstriction in response to adrenergic agonists compared with vascular rings without the PVAT but with the endothelium still intact (−PVAT+ENDO) (15, 29). Additionally, studies using aortic rings from adult male Wistar rats demonstrated that the PVAT promotes vasorelaxation in an endothelium-dependent manner (22). Furthermore, treatment with the nonselective NOS inhibitor, N^G-monomethyl-l-arginine (l-NMMA), completely abrogated the PVAT-mediated buffering of norepinephrine-induced vasoconstriction in aortas from Wistar rats (39). These findings indicate that the PVAT assists the endothelium in regulating aortic vascular tone by enhancing NOS function.

The knowledge that the PVAT functionally tempers vascular reactivity raises the question of whether it becomes dysfunctional alongside the endothelium in cardiovascular disease. One of the most well-established rodent models that has a genetic predisposition to vascular disease is the Dahl salt-sensitive (SS) rat strain (3, 44). Indeed, on a normal-salt diet regimen, these animals develop endothelial dysfunction as adults. We have shown that at ∼16 wk of age, Dahl SS rats maintained on a standard, normal-salt chow presented with significantly reduced endothelium-dependent vasorelaxation in isolated aortic rings compared with the genetic control strain, the SS.13^BN rat (38). The SS.13^BN is in essence the Dahl SS rat with the exception that chromosome 13 has been completely replaced with that of the Brown Norway rat, rendering this strain insensitive to high-salt or high-fat diet-induced hypertension (5, 8, 37). Therefore, we hypothesized that under normal-salt conditions that the PVAT from Dahl SS rats is dysfunctional in its ability to buffer vasoconstriction compared with SS.13^BN rats. Experiments were designed to examine the endothelial and NOS dependence of PVAT-mediated buffering of vasoconstriction in aortic tissue of Dahl SS and SS.13^BN rats. Furthermore, it is known that leptin derived from the PVAT activates NOS in the endothelium and aids in mediating PVAT-buffering capacity in aortic tissue (12, 17); therefore, we examined the PVAT weight and adipocyte area, levels of leptin in the PVAT, leptin receptor expression, and leptin sensitivity of vasoconstriction in aortic tissue from both rat strains.

MATERIALS AND METHODS

Animals.

All animal experiments were conducted in accordance with the National Institutes of Health's “Guide for the Care and Use of Laboratory Animals” with all animal use protocols approved by the Georgia Regents University and the University of Alabama at Birmingham Institutional Animal Care and Use Committee. SS.13^BN rats were purchased from Charles River Laboratories (Kingston, NY). Dahl SS rats were obtained from Harlan Laboratories (Indianapolis, IN) or bred in-house from purchased Dahl SS breeders (Charles River Laboratories, Wilmington, MA). The genetic background of the Dahl SS breeders was confirmed by a genome-wide scan described by Moreno et al. (31). For this purpose, rat tail DNA was extracted, and the presence of microsatellite markers specific against the Dahl SS genome was assessed with PCR using simple sequence-length polymorphism primers. Rats were maintained on Harlan Teklad 8,604 or 8,640 grain-based diet having 0.3–0.4% Na (Teklad, Madison, WI), and water was provided ad libitum. Male rats were used for all experiments.

Animals were age-matched, and at 16–21 wk old, body weight and epididymal fat mass was measured, and rats were euthanized under pentobarbital sodium (Nembutal; 0.5 mg/kg; Abbott Laboratories, Abbott Park, IL) anesthesia. Adiposity index was calculated by dividing epididymal fat mass by body weight, then multiplying by 100, and presented as a percentage. The thoracic aorta and accompanying PVAT were isolated and weighed separately. PVAT weight was normalized to aortic weight and expressed as PVAT:aorta weight. Tissues were snap-frozen in liquid N₂.

Mean arterial blood pressure.

The day before vascular reactivity experiments, rats were placed under isoflurane anesthesia (Butler Schein Animal Health, Dublin, OH) for an average of 20 min to implant indwelling catheters in the left carotid artery, which were then exposed at the nape of the neck. Catheters consisted of V/1 tubing attached to V/3 tubing (Scientific Commodities, Lake Havasu City, AZ). Approximately 2.5 cm of the V/3 end of the catheter was inserted into the carotid. Catheters were filled with sterile heparin-saline solution (300 mg/ml; Pfizer, New York City, NY) and stoppered with a metal plug. Rats were placed in restraint cages, and catheters were connected to pressure transducers (MLT0699; ADInstruments, Colorado Springs, CO) coupled to a computerized data acquisition system (PowerLab, ADInstruments). Animals were allowed to acclimate to restraint for at least 1 h. Once hemodynamic readings stabilized, mean arterial blood pressure data were collected; n = 3 rats/group.

Vascular reactivity studies.

Depending on the experimental question, PVAT was dissected away from the vessel and/or vascular endothelium was removed by gently rubbing the vessel lumen with curved forceps. Aortic rings (∼3 mm) were mounted on pins for isometric wire myography (Danish Myo Technology A/S, Aarhus, Denmark) in physiological saline solution (PSS; in mM): 130 NaCl, 4.7 KCl, 1.2 MgSO₄, 1.2 KH₂PO₄, 14.9 NaHCO₃, 5.6 dextrose, 0.024 EDTA tetrasodium salt dehydrate, and 1.6 CaCl₂ (Sigma, St. Louis, MO), as previously described (20). The baseline force was set to 28 mN, and all aortic rings were within ∼5% of each other prior to confirming viability of vascular segments by preconstricting with 10⁻⁶ M phenylephrine (PE) followed by relaxation using 10⁻⁴ M of ACh. Only those vessels that relaxed >80% to ACh were considered to have a sufficiently functional endothelium to proceed with generating concentration-response curves. A lack of ACh-dependent vasorelaxation was confirmed in all of the endothelium-denuded vessels. Aortic rings were incubated for 15 min in the presence or absence of the nonselective nitric oxide synthase (NOS) inhibitor N^ω-nitro-l-arginine methyl ester (l-NAME; 10⁻⁴ M; Sigma), or in the presence or absence of rat leptin (20 ng/ml; Sigma) or indomethacin (10⁻⁵ M; Sigma) for 20 min. Following the incubation period, a cumulative-concentration response curve was generated to PE (1 × 10⁻⁹ to 3 × 10^−4.5 M; Sigma). Vessel baths were then washed with PSS followed by assessment of constriction curves to KCl-induced depolarization. In a separate group of constriction studies, +PVAT+ENDO or −PVAT+ENDO aortic rings were submaximally preconstricted with 10⁻⁷ M PE followed by the addition of cumulative concentrations of l-NAME (1 × 10⁻¹² to 1 × 10⁻³ M) to assess the effect of PVAT on endogenous NO bioavailability. Constriction data are presented as % increase in force as calculated using the equation [(response to vasoconstrictor − baseline prior to constriction)/baseline prior to constriction] × 100. N values are given in the results section and figure legends.

Additional experiments were performed to assess endothelial-dependent vasorelaxation to ACh (ACh; 1 × 10⁻⁹ to 3 × 10^−4.5 M; Sigma) in aortic rings constricted with 10⁻⁶ M PE. These same aortic rings were washed with PSS, incubated with 10⁻⁴ M l-NAME for 15 min, and then constricted with 10⁻⁶ M PE for the purpose of conducting endothelium-independent vasorelaxation curves generated with the NO-donor sodium nitroprusside (SNP; 1 × 10⁻¹⁰ to 3 × 10^−4.5 M; Sigma). These data are presented as maximum relaxation (E_max) as a percentage of PE and sensitivity as logEC₅₀. N values are given in the results section and figure legends.

Aortic histology.

Paraffin-embedded aortas with adherent PVAT were cross-sectioned into 4-μm-thick sections and mounted on Superfrost slides. Adipocytes were stained with Gomori's blue trichrome and visualized using brightfield microscopy (Olympus BX40; Olympus America, Melville, NY). Photographs were obtained with a digital camera (Olympus DP12; Olympus America). In an experimenter-blinded fashion, the area (μm²) of individual adipocytes (≥36 adipocytes per animal) was determined for each animal. The average adipocyte area was determined for each rat by calculating the mean of the areas of all individual adipocytes counted. Adipocyte area was determined using Metamorph software (Molecular Devices, Sunnyvale, CA).

Tissue homogenization.

Thoracic aortas were isolated from anesthetized animals, and PVAT was dissected away from the vessels. Tissues were snap-frozen in liquid nitrogen and stored at −80°C until assays were performed. Using a hand-held motorized pestle, tissues were homogenized in lysis buffer [50 mM Tris (pH 7.4), 250 mM sucrose, 0.1 mM EDTA, 0.1 mM EGTA, 10% glycerol, 0.1% SDS, 0.5% Triton X-100, 0.5% sodium deoxycholate, 0.1% BME, 0.001 mg/ml phenylmethanesulfonyl fluoride (PMSF), and 0.01 mg/ml each of leupeptin, pepstatin, and aprotinin] at a ratio of 100 mg tissue/ml buffer. The samples were then snap-frozen, thawed, and sonicated for 10 × 1 s bursts on ice. Additional PMSF was added to the homogenate prior to incubation on a rocker at 4°C for 30 min. After centrifugation at 17,000 g at 4°C for 20 min, supernatant was collected and stored at −80°C for enzyme immunoassay (EIA) or Western blot analysis.

Leptin peptide and receptor level determinations.

Quantification of leptin peptide levels in PVAT was determined by EIA (kit no. 1007609; Cayman Chemicals, Ann Arbor, MI). PVAT samples (n = 6 per group) were prepared as described above and diluted 1:10 prior to performing the assay. Absorbance was measured using an Epoch colorimetric plate reader (Bio-Tek Instruments, Winooski, VT), and protein concentrations were calculated using Gen 5 Data Analysis Software (version 2.04, Bio-Tek Instruments,). Total leptin levels in the PVAT were calculated by normalizing leptin levels to milligrams of total PVAT protein and then multiplied by relative PVAT:aorta weight.

Leptin receptor density of thoracic aortas was assayed by Western blot analysis. Samples were run on 8% SDS-polyacrylamide gels, transferred to polyvinylidene fluoride membranes (Immobilon-FL) and blotted using antibodies against the long and short leptin receptor (Ob-R) (no. 3583 rabbit polyclonal; Abcam, Cambridge, MA), and GAPDH (no. 47724 mouse monoclonal; Santa Cruz Biotechnology, Dallas, TX). The following secondary antibodies were used: goat anti-mouse (1 μg/ml) for monoclonal antibodies, and goat anti-rabbit (1 μg/ml) for polyclonal antibodies (Rockland Immunochemicals, Gilbertsville, PA). Membranes were imaged using the Odyssey CLx Infared Imaging System (Li-Cor Imaging Systems, Lincoln, NE). Band intensities were quantified using Image Studio Software (Li-Cor Imaging Systems); n = 6 rats/group.

Statistical analysis.

All data are expressed as means ± SE. Data were graphed and statistically analyzed using a t-test or two-way ANOVA with Bonferroni post hoc tests using GraphPad Prism software (La Jolla, CA), where appropriate. To examine total response to vasoconstrictive agonists in aortic ring preparations, area under the curve (AUC) was calculated using GraphPad Prism. Statistical significance was defined as P < 0.05. The critical value of P was two-tailed.

RESULTS

Body weight, adiposity index, and blood pressure.

Body weight was significantly less in the Dahl SS (n = 8) vs. SS.13^BN rats (n = 8) (380 ± 8 g vs. 465 ± 5 g, P = 0.04), whereas epididymal fat mass was similar between rat strains (4.1 ± 0.3 g vs. 4.0 ± 0.2 g, respectively). There was no significant difference in the adiposity index in Dahl SS vs. SS.13^BN rats (1.05 ± 0.06% vs. 0.88 ± 0.07%, P = 0.09). Mean arterial blood pressure was similar between SS.13^BN and Dahl SS rats (142 ± 5 mmHg vs. 140 ± 8 mmHg, respectively).

PVAT-mediated buffering of vasoconstriction in SS.13^BN and Dahl SS rat aortas.

Aortic rings with an intact PVAT and endothelium (+PVAT+ENDO) had significantly reduced maximal (E_max) and sensitivity (logEC₅₀) to PE-induced vasoconstriction in both SS.13^BN (Fig. 1A, Tables 1 and 3) and Dahl SS rats (Fig. 1B, Tables 1 and 3) compared with aortic rings with the PVAT removed (−PVAT+ENDO). Interestingly, this PVAT-mediated buffering was more pronounced in the Dahl SS rats when assessed using area under the curve (AUC) analysis (Fig. 1C, Table 2). As a result, although aorta rings of −PVAT+ENDO constricted more in the Dahl SS rat group (Fig. 1C, Tables 1–3), the presence of the PVAT resulted in there being similar vasoconstriction between SS.13^BN and Dahl SS rat strains (Fig. 1C, Tables 1–3).

Fig. 1.

Phenylephrine (PE)-induced vasoconstriction curves in aortic rings with the perivascular adipose tissue (PVAT) removed (−PVAT) or intact (+PVAT) and the endothelium intact (+ENDO) from SS.13^BN rats (n = 8–10) (A) and Dahl SS rats (n = 5–9) (B). C: area under the curve (AUC) analysis to examine total vascular response to vasoconstriction in aortic ring preparations. Number of multiple comparisons = 2. *P < 0.05 vs. SS.13^BN aortic rings–PVAT+ENDO; †P < 0.05 vs. Dahl SS aortic rings–PVAT+ENDO.

Table 1.

Maximum vasoconstriction response (E_max) to phenylephrine in aortic preparations from SS.13^BN and Dahl SS rats with or without the perivascular adipose tissue or the endothelium in the presence or absence of NOS inhibition with l-NAME

Aortic Preparation	SS.13BN	Dahl SS
Emax; % Increase in Force Response to Phenylephrine
−PVAT+ENDO	109.2 ± 6.6	154.2 ± 9.0*
+PVAT+ENDO	70.6 ± 8.5†	69.0 ± 5.0†
+PVAT-ENDO	103.3 ± 10.0§	151.8 ± 10.1*§
−PVAT-ENDO	116.1 ± 7.4	148.9 ± 17.3
+PVAT+ENDO+l-NAME	97.9 ± 10.8§	174.8 ± 14.9*§
−PVAT+ENDO+l-NAME	121.2 ± 12.3	139.6 ± 10.4

Values are expressed as means ± SE. PVAT, perivascular adipose tissue; ENDO, endothelium; NOS, nitric oxide synthase; l-NAME, N^ω-Nitro-[SCAP]L[R]-arginine methyl ester.

^*P < 0.05 vs. respective SS.13^BN aortic preparation;

^†P < 0.05 vs. −PVAT+ENDO within same rat strain;

^§P < 0.05 vs. +PVAT+ENDO within same rat strain.

For SS.13BN rats, n = 8–10. For Dahl SS rats, n = 5–9.

Table 3.

Vascular sensitivity reported as logEC₅₀ to phenylephrine in aortic preparations from SS.13^BN rats and Dahl SS rats with or without PVAT or ENDO in the presence or absence of NOS inhibition with l-NAME

Aortic Preparation	SS-13BN	Dahl SS
LogEC50 (M) Determined From the % Increase in Force to Phenylephrine
−PVAT+ENDO	−7.06 ± 0.08	−7.55 ± 0.19*
+PVAT+ENDO	−5.97 ± 0.30†	−6.09 ± 0.14†
+PVAT-ENDO	−7.62 ± 0.21§	−8.21 ± 0.17*§
−PVAT-ENDO	−8.01 ± 0.16†	−8.31 ± 0.14*†
+PVAT+ENDO+l-NAME	−7.13 ± 0.19§	−7.79±.025§
−PVAT+ENDO+l-NAME	−7.83 ± 0.13†	−7.79 ± 0.09

Values are expressed as means ± SE.

^*P < 0.05 vs. respective SS.13^BN aortic preparation;

^†P < 0.05 vs. −PVAT+ENDO within same rat strain;

^§P < 0.05 vs. +PVAT+ENDO within same rat strain.

For SS.13BN rats, n = 8–10. For Dahl SS rats, n = 5–9.

Table 2.

Area under the curve analysis in response to phenylephrine-induced vasoconstriction in aortic preparations from SS.13^BN rats and Dahl SS rats with or without PVAT or the ENDO and in the presence or absence of NOS inhibition with l-NAME

Aortic Preparation	SS.13BN	Dahl SS
AUC; % Increase in Force, Response to Phenylephrine
−PVAT+ENDO	269.9 ± 14.4	396.3 ± 41.2*
+PVAT+ENDO	156.6 ± 25.1†	149.2 ± 11.2†
+PVAT-ENDO	324.3 ± 33.3§	479.5 ± 28.2*§
−PVAT-ENDO	383.7 ± 28.7†	490.1 ± 52.3*
+PVAT+ENDO+l-NAME	273.9 ± 30.3§	481.4 ± 69.3*§
−PVAT+ENDO+l-NAME	376.1 ± 34.8†	469.1 ± 49.1

Values are expressed as means ± SE.

AUC, area under the curve.

^*P < 0.05 vs. SS.13^BN;

^†P < 0.05 vs. −PVAT+ENDO within same rat strain;

^§P < 0.05 vs. +PVAT+ENDO within same rat strain. For SS.13BN rats, n = 8–10. For Dahl SS rats, n = 5–9.

The maximal KCl-induced constriction (Fig. 2) was similarly reduced in +PVAT+ENDO compared with −PVAT+ENDO aortic rings in both SS.13^BN rats (E_max, % increase in force: 94.8 ± 9.0 vs. 130.4 ± 11.5, P = 0.03) and Dahl SS rats (E_max, % increase in force: 96.9 ± 4.6 vs. 130.2 ± 11.1, P = 0.02). The AUC was significantly reduced in +PVAT+ENDO compared with −PVAT+ENDO aortic rings in Dahl SS rats (6,426.0 ± 353.9 vs. 10,019.4 ± 1,210.2, P < 0.05) but not SS.13^BN rats (5,109.6 ± 471.2 vs. 6,785.4 ± 484.0). Statistical analysis of AUC by two-way ANOVA indicated that there was an effect of PVAT to blunt vasoconstriction in both rat strains with the post hoc test showing only a significant difference within the Dahl SS strain. KCl-induced constriction (E_max and AUC) from −PVAT+ENDO aortic rings was similar in both strains.

Fig. 2.

KCl-induced vasoconstriction curves in aortic rings with the PVAT removed (−PVAT) or intact (+PVAT) and the endothelium intact (+ENDO) from SS.13^BN rats (n = 8) (A) and Dahl SS rats (n = 6) (B). C: area under the curve (AUC) analysis to examine total vascular constriction in aortic ring preparations. Number of multiple comparisons = 2. KCl-induced constriction (E_max and AUC) from −PVAT+ENDO aortic rings was similar in both strains. The two-way ANOVA indicated there was an effect of PVAT to blunt vasoconstriction in both rat strains with the post hoc test showing a significant difference only within the Dahl SS strain.*P < 0.05 vs. Dahl SS aortic rings −PVAT+ENDO.

Endothelium dependency of PVAT-mediated buffering of vasoconstriction in SS.13^BN and Dahl SS rat aortas.

Removing the endothelium (−ENDO) enhanced the E_max, AUC, and logEC₅₀ of vasoconstriction to PE in aortic rings +PVAT from SS.13^BN (Fig. 3A, Tables 1–3) and Dahl SS rats (Fig. 3C, Tables 1–3), although this increase was greater in the Dahl SS rat strain (Fig. 3E, Tables 1–3). In aortic rings without PVAT (−PVAT), −ENDO did not alter E_max in either rat strain (Fig. 3, B and D, Table 1), whereas denudation significantly increased the AUC only in the SS.13^BN rats (Fig. 3F, Table 2). These data show that the PVAT-mediated buffering of PE-induced vasoconstriction in aortic rings from both SS.13^BN and Dahl SS rats is dependent on the endothelium and that this endothelium dependency was greater in the Dahl SS rats. It is noted that experiments in Figs. 1 and 3 were conducted at different times, resulting in a quantitatively different value of maximum constriction in the +PVAT+ENDO aortic rings of Dahl SS rats; however, all experiments were conducted in a paired manner of both rat strains with or without PVAT and/or ENDO. Qualitative findings were consistent among all experiments.

Fig. 3.

Phenylephrine (PE)-induced vasoconstriction curves in aortic rings with the PVAT removed (−PVAT) or intact (+PVAT) and the endothelium removed (−ENDO) or intact (+ENDO) in SS.13^BN rats (n = 8–10) (A, B) and Dahl SS rats (n = 5–9) (C, D). AUC examining the effect of endothelial denudation on the vasoconstriction in aortic rings +PVAT (E) and −PVAT (F) in SS.13^BN and Dahl SS rats. Number of multiple comparisons = 2. *P < 0.05 vs. SS.13^BN aortic ring controls; †P < 0.05 vs. Dahl SS aortic ring controls.

NOS dependency of PVAT-mediated buffering of vasoconstriction in SS.13^BN and Dahl SS rat aortas.

To examine the mechanisms whereby the endothelium mediates the PVAT-mediated buffering of PE-induced vasoconstriction in aortas and the greater dependency of this effect in Dahl SS rats, vascular reactivity experiments were conducted in aortic rings +PVAT+ENDO in the presence or absence of the nonselective NOS inhibitor l-NAME. l-NAME enhanced the PE vasoconstriction (E_max, AUC, and logEC₅₀) in aortic rings +PVAT+ENDO from SS.13^BN (Fig. 4A, Tables 1–3) with this increase being even greater in the Dahl SS rats (Fig. 4, C and E, Tables 1–3). Thus, there was not a difference in the effect of l-NAME to augment PE-induced vasoconstriction in aortic rings having either +PVAT+ENDO or −PVAT+ENDO in SS.13^BN rats, but this effect was more pronounced in Dahl SS rats. In aortic rings −PVAT+ENDO, l-NAME induced an increase in AUC and logEC₅₀ but not E_max in the SS.13^BN rats only (Fig. 4, B and D, Tables 1–3).

Fig. 4.

Phenylephrine (PE)-induced vasoconstriction curves in aortic rings with the PVAT removed (−PVAT) or intact (+PVAT) and in the presence or absence of the nonselective NOS inhibitor l-NAME in SS.13^BN rats (n = 8–10) (A, B) and Dahl SS rats (n = 5–9) (C, D). AUC examining the effect of NOS inhibition on the vasoconstriction in aortic rings +PVAT (E) and −PVAT (F) without (−) or with (+) corresponding treatments. Number of multiple comparisons = 2. *P < 0.05 vs. SS.13^BN aortic ring controls; †P < 0.05 vs. Dahl SS aortic ring controls.

To examine whether the PVAT affected the endogenous ability of NOS to control vascular tone, aortic rings were preconstricted with a submaximal dose of PE followed by generation of constriction curves to l-NAME. It was found that the E_max response to NOS inhibition was similar between −PVAT+ENDO and +PVAT+ENDO aortic preparations, respectively, from both SS.13^BN rats (34.9 ± 3.3%, n = 8 vs. 31.8 ± 2.6%, n = 7) and Dahl SS rats (29.5 ± 4.2%, n = 6 vs. 23.6 ± 4.1%, n = 6). Also, the sensitivity to l-NAME (logEC₅₀) was similar in −PVAT+ENDO and +PVAT+ENDO aortic rings, respectively, within SS.13^BN (−2.9 ± 0.1 M vs. −2.8 ± 0.1 M) and Dahl SS (−2.6 ± 0.2 M vs. −2.6 ± 0.1 M) rat strains. However, the endogenous ability of NOS to regulate vascular tone was significantly less in Dahl SS compared with the SS.13^BN rat genotype (P < 0.05).

To determine whether PVAT may also lead to the release of NOS-independent vasodilator prostanoids, PE constriction curves were generated in aortic rings +PVAT+ENDO in the absence or presence of indomethacin, respectively. E_max was similar in aortic rings from SS.13^BN rats (127.9 ± 11.0%, n = 3 vs. 112.8 ± 14.7%, n = 5) and Dahl SS rats (120.2 ± 3.7%, n = 3 vs. 114.7 ± 11.3%, n = 5), as well as sensitivity to indomethacin treatment among aortic rings from SS.13^BN rats (logEC₅₀, −7.3 ± 0.1 M vs. −7.1 ± 0.1 M) or Dahl SS rats (logEC₅₀, −7.2 ± 0.3 M vs. −7.6 ± 0.2 M).

Effect of PVAT on endothelial-dependent vasorelaxation in SS.13^BN and Dahl SS rat aortas.

We assessed whether the action of PVAT in buffering vasoconstriction was also accompanied by augmentation of endothelium-dependent relaxation. ACh-induced endothelial function was similar in the presence and absence of PVAT in aortic rings +ENDO from SS.13^BN rats (E_max, 84.6 ± 3.9% vs. 81.3 ± 3.1%; logEC₅₀, −6.5 ± 0.2 M vs. −6.6 ± 0.1 M, respectively) and Dahl SS rats (E_max, 75.6 ± 2.2% vs. 75.4 ± 3.0%; logEC₅₀, −6.1 ± 0.1 M vs. −6.3 ± 0.1 M, respectively). Statistical analysis by two-way ANOVA indicated that Dahl SS rat strain as a whole showed reduced maximum relaxation and sensitivity to ACh (P < 0.05). Endothelial-independent vasorelaxation induced by SNP was unaltered in the presence and absence of PVAT in SS.13^BN rats (E_max, 98.5 ± 2.5% vs. 97.4 ± 0.6%; logEC₅₀, M: −7.6 ± 0.1 M vs. −7.7 ± 0.1 M), as well as Dahl SS rats (E_max, 106.2 ± 3.3% vs. 98.9 ± 0.7%; logEC₅₀, −7.7 ± 0.2 M vs. −7.7 ± 0.2 M, respectively), and these responses were not different between the rat strains.

Aortic PVAT weight and adipocyte size in SS.13^BN and Dahl SS rats.

PVAT weight was significantly greater in the Dahl SS rats compared with the SS.13^BN control rat strain (Fig. 5A); however, there was no significant difference in adipocyte size between SS.13^BN or Dahl SS rats as quantified in Fig. 5B and depicted in the representative histology images in Fig. 5C.

Fig. 5.

A: perivascular adipose tissue (PVAT) weight normalized to aorta weight (PVAT:aorta weight) in SS.13^BN rats (n = 9) and Dahl SS rats (n = 5). B: quantification of PVAT adipocyte area. C: representative histological images of PVAT adipocytes in SS.13^BN rats and Dahl SS rats. *P < 0.05 vs. SS.13^BN rats.

Leptin levels in PVAT and vascular leptin receptor expression in SS.13^BN and Dahl SS rat aortas.

Leptin levels when normalized to total PVAT protein were not different in the PVAT between rat strains (SS.13^BN: 1,714 ± 224 pg/mg total protein vs. Dahl SS: 1,343 ± 163 pg/mg total protein, P > 0.05). However, when the greater PVAT weight in the Dahl SS rat group was taken into consideration, total tissue leptin levels were found to be elevated in the Dahl SS compared with SS.13^BN rats (Fig. 6A). Whole thoracic aorta (−PVAT+ENDO) levels of the long and short forms of the leptin receptor (Ob-R) were similar between rat strains (Fig. 6B), as pictured in the representative Western blot in Fig. 6C.

Fig. 6.

A: total leptin peptide levels in the PVAT from SS.13^BN rats and Dahl SS rats. These levels were calculated by normalizing leptin levels to milligrams of total PVAT protein and then multiplied by relative PVAT:aorta weight. B: semiquantification of results from Western blot analyses examining expression of the long Ob-R (top) and short Ob-R (bottom) forms of the leptin receptor in aortic vascular tissue. C: representative Western blot. n = 6 rats/group. *P < 0.05 vs. SS.13^BN rats.

Effects of leptin on PE-induced constriction in SS.13^BN and Dahl SS −PVAT+ENDO rat aortic rings.

On the basis of the above findings, we further determined whether −PVAT+ENDO aortic rings from Dahl SS rats compared with those of SS.13^BN rats would have greater leptin-mediated buffering of PE-induced vasoconstriction. Indeed, Fig. 7 shows that leptin preincubation significantly blunted PE-induced vasoconstriction only in aortic rings from Dahl SS rats (E_max, Dahl SS rats: 138.8 ± 7.9% vs. 122.6 ± 5.1%, P < 0.05; SS.13^BN: 143.5 ± 3.5% vs. 135.1 ± 12.6%). The logEC₅₀ responses were similar between untreated and treated aortic rings, respectively, in Dahl SS rats (−7.6 ± 0.1 M vs. −7.6 ± 0.04 M) and SS.13^BN rats (−7.6 ± 0.06 M vs. −7.6 ± 0.1 M). The AUC was similar between untreated and treated aortic rings, respectively, in Dahl SS rats (369.6 ± 28.6 vs. 338.9 ± 12.2) and SS.13^BN rats (373.5 ± 12.8 vs. 364.3 ± 36.2).

Fig. 7.

PE-induced vasoconstriction curves in aortic rings with the PVAT removed (−PVAT) and the endothelium intact (+ENDO) from SS.13^BN rats (A) and Dahl SS rats (B) incubated in the presence or absence of leptin at 20 ng/ml for 20 min. C: maximum constriction (E_max) to −log [4.5, M] of PE. n = 5 rats/group. Number of multiple comparisons = 2. *P < 0.05 vs. Dahl SS aortic ring without leptin.

DISCUSSION

The main finding of our study is that, in contrast to our hypothesis, the PVAT from Dahl SS rats is functionally capable of blunting the response to adrenergic receptor-induced vasoconstriction in aortic tissue. Moreover, the PVAT-buffering effect exceeded that observed in the SS.13^BN rat strain. These data provide evidence that PVAT function is present and enhanced in Dahl SS rats on standard chow diet. As a result, although aortic rings without the PVAT but with an intact endothelium constricted more in response to phenylephrine in the Dahl SS rats than in the SS.13^BN rats, when the PVAT remained intact, there was an equalization of the vasoconstriction between the two rat strains. Mechanistically, in both rat strains, the PVAT function was entirely dependent on the endothelium. In both rat strains having both an intact PVAT and endothelium, treatment of the vascular ring segments with the nonselective NOS inhibitor produced a similar response as with endothelial denudation alone. In aortic rings without PVAT, endothelial denudation or l-NAME treatment, significantly increased vasoconstriction only in the SS.13^BN rat strain. These data provide compelling evidence that PVAT-mediated buffering of vasoconstriction is enhanced in Dahl SS rats, where it works to control vascular tone in a manner that is dependent on signaling to NOS in the endothelium.

In the present study, we compared the PVAT-mediated buffering of vasoconstriction in Dahl SS rats vs. their genetic control counterpart, the SS.13^BN rat. We initially hypothesized that Dahl SS rats may exhibit PVAT dysfunction, that is, a reduced effect of the PVAT to buffer vasoconstriction. The Dahl SS rat strain is well known for its predisposition to develop cardiovascular disease as adults with endothelial dysfunction (3). This occurs in adults, even though they are maintained on a normal-salt chow. Indeed, we previously showed that endothelial function, as assessed by ACh-induced vasorelaxation, was significantly attenuated in Dahl SS rats vs. SS.13^BN rats that were maintained under standard chow diet conditions (38). In the present study and contrary to our hypothesis, we found that the PVAT blunted phenylephrine-induced vasoconstriction in aortic ring preparations from both rat strains compared with their respective aortic rings with the PVAT removed. Interestingly, this response was observed to be greater in the Dahl SS rat strain. As a result, in vascular preparations having both an intact PVAT and an endothelium, the adrenergic vasoconstriction was similar between both rat strains.

Intriguingly, work from Allyn Mark's group demonstrated in situ experiments that there is similar aortic distensibility, as assessed by constructing pressure-volume curves, between Dahl SS rats and their salt-resistant counterparts while on a lower-salt (0.13% NaCl) diet (16). Vascular distensibility allows large arteries like the aorta to accommodate pulsatile cardiac output. With the aorta having native PVAT in vivo, we speculate that the PVAT is an important regulator of aortic response to factors that increase tension of the blood vessel wall.

Studies in Sprague-Dawley (10, 11, 23, 24, 26) and Wistar-Kyoto rats (13–15, 24, 42) have demonstrated that the PVAT attenuates vasoconstriction by stimulating NO release from the endothelium (15). However, this has not been examined in Dahl SS rats. We found that the NOS inhibitor l-NAME significantly enhanced PE-induced vasoconstriction in aortic rings that have both the endothelium and PVAT intact. This occurred not only in the control SS.13^BN rat strain, but also in Dahl SS rats on the normal-chow regimen. NOS inhibition produced a similar effect to that observed with endothelial denudation. Therefore, our data confirm that PVAT blunts vasoconstriction in aortic tissue in the SS.13^BN rat similar to other control strains via activation of NOS in the endothelium. Our findings also show that the Dahl SS rat uses a similar mechanism in that the PVAT activates NOS signaling in the endothelium to mediate enhanced buffering capacity. In contrast, we found that the presence of the PVAT did not alter endothelium-dependent vasorelaxation to ACh, endothelium-independent vasorelaxation to sodium nitroprusside, nor did its presence impact l-NAME-induced vasoconstriction in aortic rings from either rat strain. Thus, we propose that the PVAT-mediated stimulation of NOS in the endothelium is, most likely, through a pathway that is distinct from the ACh-induced activation of endothelial NOS or any alteration of basal endogenous NO bioavailability.

Verlohren et al. (40) found that longitudinal removal of 50% of the PVAT attenuates the PVAT-mediated buffering of vasoconstriction by approximately half in mesenteric arteries. These data suggest that the extent to which the PVAT is able to buffer vasoconstriction is dependent on its mass. Their study has particular relevance to our study, since we found that Dahl SS rats had a greater amount of PVAT weight and PVAT buffering of vasoconstriction than control rats. Furthermore, we found that the total levels of leptin in the PVAT surrounding the thoracic aortas, which was calculated by multiplying leptin per gram of total protein by the PVAT weight, were greater in the Dahl SS vs. SS.13^BN rats. However, when leptin levels were quantified on the basis of grams of total protein, there was no difference in leptin levels in the PVAT. As leptin production from adipocytes is based on size, it follows that we observed no difference between groups in PVAT adipocyte size but found increased PVAT weight. In functional experiments, we found that incubation with rat leptin blunted PE-induced vasoconstriction in aortic rings −PVAT+ENDO from Dahl SS but not SS.13^BN rats. The leptin concentration of 20 ng/ml chosen for use in our study was based on existing literature reporting a 50–100% maximum relaxation in ex vivo vascular preparations of aorta and mesenteric arteries without PVAT from rats (19, 25, 34). This indicates that PVAT-derived leptin may be involved in the greater effect that PVAT has on buffering vasoconstriction in the Dahl SS rat strain. However, the effect of leptin alone, although significant for the higher concentrations of phenylephrine, was considerably smaller than the effect of PVAT in suppressing vasoconstriction. Thus, this leptin treatment did not completely recapitulate the buffering effect seen in +PVAT+ENDO aortic segments. Leptin may act in combination with other factors derived from the PVAT to produce the more pronounced PVAT buffering capacity in Dahl SS rats; thus, further work is needed to decipher these mechanisms.

A study by Galvez-Prieto et al. showed that the periaortic adipose tissue from WKY rats expresses leptin (12). This study went on to demonstrate that exogenous application of leptin elicited vasorelaxation in isolated aortic rings with an intact endothelium without the PVAT. Interestingly, Benkhoff et al. (4) showed that leptin treatment in vivo, which increased plasma leptin levels 5 × to ∼10 ng/ml, restored aortic endothelial function in a hypertensive mouse model that displays endothelial dysfunction. We showed that the total leptin levels in the PVAT from Dahl SS rats are greater than that found in SS.13^BN rats. This was calculated on the basis of the fact that PVAT was greater in the Dahl SS rats. There was not a difference in the level of either the long or short form of the leptin receptor in the aortic vascular tissue. Collectively, these data suggest that the adipocyte hormone leptin is a player in mediating the greater PVAT function present in Dahl SS rats.

Although the Dahl SS rats had greater amounts of aortic PVAT, their body weight was less than that of the SS.13^BN rat strain. The rats in this study were age-matched, resulting in this disparity in body weight. We speculate that this is a result of the genetic difference between the two strains. Previous publications find that the Dahl SS rat is hyperlipidemic (1, 37). This altered metabolic state may promote the greater adiposity around the blood vessels to preserve the vascular function in the face of these increased circulating lipids, without changing adiposity in the other fat pads. Future studies should compare the effects of changes in body weight, such as in response to high-fat diet feeding or caloric restriction, on PVAT function between these rat strains.

Although we have previously found that Dahl SS rats at this age and on standard chow have higher blood pressure than their SS.13^BN controls (38), we did not detect a difference in mean arterial blood pressure between rat strains in the present study. This may be due to the acute methods used to collect blood pressure measurements compared with telemetry in the previous study (38). However, it was verified that endothelium-dependent relaxation was reduced in the Dahl SS rats vs. SS.13^BN rats under the standard chow conditions of this study, supporting the observation that the Dahl SS rat strain is predisposed to cardiovascular disease under normal-salt chow conditions. With that being said, it is interesting to speculate what may happen to PVAT function in Dahl SS rats in the context of hypertension. The Dahl SS rat is known for its defined hypertensive response to a high-salt diet (44). Intriguingly, it was demonstrated in a form of essential hypertension, namely the spontaneously hypertensive rat (SHR), that the aortic PVAT levels of leptin are significantly reduced along with a loss of its anticontractile effect (12). Therefore, we speculate that the hypertension produced by high-salt diet in Dahl SS rats may produce PVAT dysfunction as found in the SHR. Although it is not yet understood how hypertension in the Dahl SS rat affects PVAT-mediated buffering of vasoconstriction, what is known is that high-salt diet-induced hypertension elicits dramatic endothelial dysfunction in their aortas but not in their salt-resistant counterparts (28). Under certain conditions such as a prooxidant environment, NOS becomes uncoupled and generates reactive oxygen species, specifically superoxide, instead of NO (7, 9, 21, 27, 30, 41, 43). Because PVAT function was observed to be dependent on the endothelium in the Dahl SS rat, this knowledge further leads us to propose that PVAT function would be attenuated in Dahl SS rats in the face of a high-salt diet. Future studies will be designed to test this hypothesis.

Dahl SS rats also have a hypertensive response to other dietary stressors. We and others have shown that these animals also become hypertensive with a high-fat, normal-salt diet compared with their genetic controls (5, 32, 37). This indicates that Dahl SS rats are also fat-sensitive. However, paradoxically, Dahl SS rats are protected from aortic endothelial dysfunction on both a short-term (4 wk) and long-term (20 wk) high-fat diet regimens, even though they are hypertensive (5, 38). These studies noted that endothelium-dependent vasorelaxation was increased following a high-fat diet along with elevated blood vessel expression of NOS3 and phosphorylation at the activation site Ser-1177 of NOS3 in Dahl SS rats. PVAT function was not examined in these studies. Because it has been shown that a high-fat diet enlarges adipocyte size and leptin levels in the PVAT of carotid arteries from mice (35), it is attractive to hypothesize that the PVAT plays an important role in the vasoprotective mechanisms observed with high-fat diet in Dahl SS rats.

Perspectives and Significance

The PVAT is a novel regulator of vascular function via blunting of vasoconstriction (33). This study demonstrated that PVAT function is present in Dahl SS and SS.13^BN rats on a normal-salt diet. This function depends on NOS signaling in the endothelium. This function was greater in the Dahl SS rats than that observed in their genetic controls. These data suggest that the PVAT is a valid target of diet-induced alterations in vascular function in the Dahl SS rat strain. As understanding the regulation of vascular function in cardiovascular disease in humans is complicated due, in part, to complex dietary patterns, the Dahl SS rats may provide novel, clinically relevant information on mechanisms of PVAT function in pathological states.

GRANTS

This study was supported by National Institutes of Health (NIH) P01 HL-69999 (J. S. Pollock), NIH R01 HL-60653 (J. S. Pollock), NIH 2T32HL-076146 (F. T. Spradley), American Heart Association Grant 10PRE4010001 (F. T. Spradley), and NIH 1F32HL-116145 (D. H. Ho).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: F.T.S., D.H.H., and J.S.P. conception and design of research; F.T.S. and D.H.H. performed experiments; F.T.S., D.H.H., and J.S.P. analyzed data; F.T.S., D.H.H., and J.S.P. interpreted results of experiments; F.T.S. and D.H.H. prepared figures; F.T.S. drafted manuscript; F.T.S., D.H.H., and J.S.P. edited and revised manuscript; F.T.S., D.H.H., and J.S.P. approved final version of manuscript.