Localization of Atrial Natriuretic Factor Receptors in the Mesenteric Arterial Bed

Preparation of Vascular and Fatty Membranes

Sprague-Dawley rats (300 to 325 g) were decapitated, and their mesenteric arcades were removed by blunt dissection and placed in ice-cold saline. A large piece of fat overlying the superior mesenteric artery was eliminated, because preliminary experiments showed it to have three to four times lower ¹²⁵I-ANF binding than that found in adipose tissue around the mesenteric arteries and arterioles. The superior mesenteric vein and branching veins were eliminated. The vascular preparation consisted of the mesenteric arterial bed, including the superior mesenteric artery, ileal and jejunal arteries, and all branching arterioles. The smallest outer diameter of the mesenteric vessels in the preparation was 90 to 100 μm. Adipose tissue surrounding the mesenteric arteries and arterioles was detached by stripping them with forceps. Vascular and fatty tissues were homogenized in 0.25 mol/L fresh ice-cold sucrose solution with a Polytron homogenizer (setting 8, four times for 10 seconds) and centrifuged at 1075g for 10 minutes at 4°C. The supernatant was kept on ice, and the process was repeated for the pellet. The supernatants of both centrifugations were pooled, filtered through 20-μm nylon mesh, and centrifuged at 100 000g for 60 minutes at 4°C. The pellet was resuspended in 0.05 mol/L Tris-HCl buffer, pH 7.4. Protein concentration was assessed by the Coomassie blue technique described by Spector.³³

Binding Assay

The binding assay was performed as described previously.^{34 35} Aliquots of fatty and vascular membranes (30 to 40 μg of protein) were incubated in duplicate for 90 minutes at 22°C in the presence of increasing concentrations of unlabeled peptides (1 pmol/L to 1 μmol/L). In competition experiments, membranes were incubated with 90 to 105 pmol/L ¹²⁵I-ANF (750 Ci/mmol), 70 to 80 pmol/L ¹²⁵I–[Sar¹,Ile⁸]Ang II (1550 Ci/mmol), or ¹²⁵I–ET-1 (1470 Ci/mmol) in a final volume of 0.25 mL. The reaction was stopped by dilution with 3.5 mL assay buffer and rapid filtration through Whatman GF/C filters, which were then rinsed three times with 3 mL of 0.05 mol/L Tris-HCl at pH 7.4. The assay buffer contained 0.05 mol/L Tris-HCl (pH 7.4), 1 μmol/L aprotinin, 0.1% bacitracin, 5 mmol/L MgCl₂, 0.5 mmol/L phenylmethylsulfonyl fluoride (PMSF), 0.4 μmol/L phosphoramidon, and 0.5% bovine serum albumin (BSA). Filtration and rinsing were performed with a semiautomatic harvester system (Brandel), which processes 30 tubes simultaneously. The filters were counted in an LKB gamma counter with 70% efficiency. ¹²⁵I-ANF, ¹²⁵I–[Sar¹,Ile⁸]Ang II, and ¹²⁵I–ET-1 were prepared by the lactoperoxidase method.³⁶ Nonspecific binding was defined as binding remaining in the presence of 1 μmol/L unlabeled peptides. Three saturation experiments and three to four competitive inhibition assays for each displacing agent were undertaken on three different membrane preparations for each tissue. Ten to 12 rats were used in each experiment, and all binding assays were performed in duplicate the same day the tissues were isolated.

Labeled peptide binding to vascular and fatty membranes was studied over a range of protein concentrations from 5 to 50 μg. All experiments were performed with 20 to 40 μg protein. Total, specific, and nonspecific binding is depicted in all saturation figures. In competition experiments, only specific binding is shown. Specific binding was obtained by subtracting nonspecific from total binding.

Covalent Cross-linking of Adipose and Vascular Tissue ANF Receptors

Aliquots (100 μg) of vascular and fatty membranes were incubated in binding assay buffer for 60 minutes at 22°C, with 200 pmol/L ¹²⁵I-ANF in the absence and presence of 1 μmol/L unlabeled ANF, CNP, or C-ANF. Disuccinimidyl suberate (DSS) in dimethyl sulfoxide (DMSO) was added to a final concentration of 0.5 mmol/L and incubated for 15 minutes at 22°C. The reaction was quenched by the addition of ammonium acetate to a final concentration of 50 mmol/L. Sample buffer (100 μL) containing 1% SDS (wt/vol), 100 mmol/L Tris-HCl (pH 6.8), 20% glycerol, 2% β-mercaptoethanol, and traces of bromophenol blue was then added. The samples were boiled for 5 minutes before being analyzed by electrophoresis.

SDS-PAGE of Adipose and Vascular Tissue ANF Receptors

Electrophoresis was performed with a modification of the method of Schägger and von Jagow³⁷ in 8% unidimensional slab gel in the presence of 0.1% SDS. The gels were stained with Coomassie brilliant blue R-250, destained, and dried. Molecular weights were identified by reference standards. The dried gels were exposed to Kodak Omat XRP6 film for 4 to 6 days at −70°C. The relative density of high- and low-molecular-weight ANF receptor subtypes in the autoradiograms was assessed by densitometry.

Autoradiographic Localization of ¹²⁵I-ANF

¹²⁵I-ANF (24 pmol) in sodium phosphate buffer containing 0.1 g/dL BSA was injected in a volume of 300 μL into the superior mesenteric artery of male Sprague-Dawley rats (100 g body weight) under pentobarbital anesthesia (n=3). In parallel experiments, unlabeled ANF (7 nmol) was mixed with ¹²⁵I-ANF and injected into rats under the same conditions (n=3). Two minutes after injection, the mesenteric vascular bed was perfused with 200 mL of Bouin’s solution for 10 minutes. The animals were then killed, and the entire mesenteric vascular bed was removed, fixed for 24 hours in the same fixative, and embedded in paraffin. Sections (5 μm thick) were cut and stained with hematoxylin and lithium carbonate before being dipped into Ilford K5 emulsion. All sections were then exposed for 4 weeks, developed in D19 solution, and fixed with sodium thiosulfate for 4 minutes.

Autoradiographic Localization of ¹²⁵ET-1

Sprague-Dawley rats (300 to 325 g body weight) were decapitated, and their mesenteric arcades were removed and placed in saline. The superior mesenteric vein and branching veins as well as a large piece of fat overlying the superior mesenteric vein were removed and eliminated. The preparation, corresponding to the mesenteric arterial bed and surrounding adipose tissue, was snapped frozen in isopentane at −40°C. Frozen sections were cut in a cryostat at −30°C. The sections were thaw-mounted onto gelatin-coated slides and kept at −70°C until used. Before the assay, the slides were dried in a dessicator for 4 hours at 4°C. They were then preincubated at room temperature for 15 minutes in 50 mmol/L Tris-HCl buffer (pH 7.4) containing 0.5% BSA and subsequently incubated for 60 minutes in fresh buffer supplemented with 150 pmol/L of ¹²⁵I–ET-1 as well as 5 μmol/L MgCl₂, 100 mmol/L NaCl, 1 μmol/L phosphoramidon, 1 μmol/L aprotinin, 1 μmol/L EDTA, 1 μmol/L PMSF, and 0.05% bacitracin. Nonspecific binding was determined in the presence of an excess of unlabeled ligand (1 μmol/L). After incubation, the sections were transferred through four successive 1-minute washes of ice-cold 50 mmol/L Tris-HCl buffer (pH 7.4) containing 0.5% BSA, fixed (30 minutes) in 2% glutaraldehyde (pH 7.4, 4°C), washed in distilled water, dehydrated in ethanol, and dried overnight at 40°C. Subsequently, the slides were dipped in Ilford K5 clear emulsion, exposed for 5 to 6 days at 4°C, developed in Kodak D19, and stained with hematoxylin and eosin.

Histological Preparation

Male Sprague-Dawley rats (300 to 350 g) were decapitated, their mesenteric artery arcades were excised, and their intestines were removed. The entire mesenteric artery arcades or defatted mesenteric vascular beds, prepared as above, were fixed for 24 hours in Bouin’s solution and embedded in paraffin. Sections (5 μm) were cut and stained with hematoxylin, lithium carbonate, and eosin.

Cyclic Nucleotide Measurement

cGMP production was measured as described previously.³⁴ Briefly, mesenteric perivascular adipose tissue and arteries were prepared as in the binding assay, but instead of 0.9% NaCl solution, Hanks’ balanced salt solution (HBSS containing [mmol/L] NaCl 137, KCl 5.4, KH₂PO₄ 0.44, Na₂HPO₄ 0.33, MgSO₄ 0.04, MgCl₂ 0.50, CaCl₂ 1.25, and NaHCO₃ 4.0, at pH 7.4) was used with neither PMSF nor bacitracin. The tissues were resuspended in final HBSS assay buffer containing 0.2 g/dL BSA, 5.5 mmol/L glucose, and 25 mmol/L HEPES. Fatty and vascular samples were preincubated in assay buffer at 37°C for 15 minutes before isobutylmethylxanthine (IBMX) was added at a final concentration of 0.5 mmol/L. Three minutes later, ANF, CNP, C-ANF, or sodium nitroprusside (SNP) was added to achieve a range of concentrations varying from 0.1 nmol/L to 1 μmol/L in a final assay buffer incubation volume of 500 μL. After incubation for 5 minutes, the reaction was stopped by the addition of 500 μL of ice-cold 10% trichloroacetic acid (TCA). The samples were centrifuged at 1075g at 4°C for 15 minutes, the supernatant was retained, and the pellet was digested with 1 mol/L NaOH for protein determination.³³ TCA was removed by extracting the supernatant four times with 2 mL of water-saturated ether, and the remaining ether was evaporated at room temperature. The aqueous phase was dried with a Speed Vac concentrator (Savant Instruments, Inc). Each sample was resuspended in 300 μL of 50 mmol/L sodium acetate buffer, pH 6.2. A 100-μL aliquot was acetylated, and cGMP concentration was assessed by radioimmunoassay. The fatty and vascular samples were preincubated for 5 minutes with IBMX. Three fresh preparations were obtained for each tissue (five to seven rats for each preparation), and two separate determinations were conducted for each preparation. All experiments were performed the same day the tissues were isolated.

Mesenteric Artery Perfusion

Vasoconstrictor responses to norepinephrine (NE), vasopressin (AVP), ET-1, and Ang II were determined in mesenteric preparations in which the surrounding adipose tissue was either completely removed or left intact. Male Sprague-Dawley rats (300 to 350 g) were used in these experiments. They were anesthetized with pentobarbital sodium (60 mg/kg body wt IP). The superior mesenteric artery was cannulated with PE-90 tubing, flushed with 20 mL Krebs’ solution, and dissected free from the intestine. The whole vascular bed (with or without surrounding fat) was placed in a water-jacketed chamber maintained at 37°C and perfused with Krebs’ solution (mmol/L: NaCl 112, KCl 5.0, NaH₂PO₄ 1.0, MgSO₄ 1.2, CaCl₂ 2.5, NaHCO₃ 2.5, and glucose 11.2), which was kept at 37°C and aerated with a mixture of 5% CO₂/95% O₂ to obtain a pH of 7.4. Intact mesenteric preparations and defatted arteries were perfused at 7.5±0.5 and 15±2 mL/min, respectively, with a peristaltic pump to achieve a constant pressure of ≈20 mm Hg. Perfusion pressure was recorded with a pressure transducer (Statham P231D) connected to a polygraph (Grass 7D). The preparations were allowed to equilibrate for 1 hour before a dose-response curve to each agent was established.

Chemicals

All materials were of the highest reagent grade available. Tricine, Coomassie brilliant blue, lactoperoxidase, bacitracin, PMSF, pepstatin, phosphoramidon, polyethylene glycol, BSA, and gamma globulin were from Sigma Chemical Co; DSS, from Pierce; 2-mercaptoethanol, from J.T. Baker Inc; DMSO and glycerol, from Anachemia; aprotinin, from Miles Laboratories; molecular weight standards, from Pharmacia Fine Chemicals; SDS, acrylamide, bisacrylamide, and tetramethylethylendiamine, from Bio-Rad; ¹²⁵I-sodium, from Amersham Canada Limited; ANF-(99-126), CNP, C-ANF, Ang II, [Sar¹,Ile⁸]Ang II, and ET-1, from Bachem; BQ-123, from the Peptide Institute Inc; and Omat XRP6 film, from Eastman Kodak. Losartan potassium and PD123319 were generous gifts from DuPont Merck Pharmaceuticals Co and Parke-Davis, respectively.

Analysis of Data

Whenever appropriate, the results are expressed as mean±SEM. Binding data were analyzed by processing raw data with the computer-based ebda program (Elsevier, Biosoft). The binding capacity (B_max) and apparent affinity (K_d) of binding sites were then determined by the computer-based ligand program.³⁸

When the Hofstee plot and line of best fit displayed by the ligand program with an assumed one-site model were inadequate, suggesting departure from linearity, a two-site model investigated the same data; these procedures were followed by testing of whether the second model provided a better statistical fit than the first model.

Light Microscopy

Fig 1⇓ is a low-power magnification (×1.5) of two mesenteric preparations before (top panel) and after (bottom panel) the surrounding adipose tissue and mesenteric veins were removed. Two light micrographs of untouched (top panel) and completely defatted (bottom panel) mesenteric arteries are shown in the insets. The walls of both arteries present undamaged intimal, medial, and adventitial layers.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Low-power magnifications (original magnifications ×1.5) of mesenteric arterial bed consisting of superior mesenteric, jejunal, and ileal arteries. A, Intact arcade. B, Defatted preparation (adipose tissue and veins removed). Insets are light micrographs of intact (A) and defatted (B) preparations (original magnifications ×63).

Saturation and Competitive Inhibition of ¹²⁵I-ANF, ¹²⁵I–[Sar¹,Ile⁸]Ang II, and ¹²⁵I–ET-1 Binding to Fatty and Vascular Membranes

¹²⁵I-ANF saturation binding assays in adipose tissue membranes revealed a single class of high-affinity binding sites with a B_max of 420±16 fmol/mg protein and a K_d of 343±16 pmol/L (n=3) (Fig 2a⇓ and 2c⇓). In contrast, ¹²⁵I-ANF saturation assays in vascular membranes indicated that most of the ¹²⁵I-ANF binding was nonspecific (Fig 2b⇓). The scatter of data on poor specific binding was extremely high, rendering Scatchard transformation meaningless and suggesting that ANF receptor expression in the arterial mesenteric vasculature was either extremely low or absent. The pharmacological profile of ligands competing for ¹²⁵I-ANF (Fig 2d⇓) indicated that the majority of receptors present in adipose tissue did not discriminate between the different natriuretic peptides, because they recognized ANF, CNP, and C-ANF with close affinities (424±85, 930±87, and 770±48 pmol/L, respectively). C-ANF, a specific ligand for the ANP-C receptor, potently competed for >98% of ¹²⁵I-ANF binding sites, suggesting that ANP-C receptors were predominant in mesenteric adipocytes.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Pharmacological characterization of atrial natriuretic factor (ANF) receptors in mesenteric adipose and vascular tissues. a and b, Representative ¹²⁵I-ANF saturation binding curves in membranes prepared from mesenteric adipose tissue (a) and arteries (b). ○ indicates total binding; ▿, nonspecific binding; and •, specific binding. c, Scatchard transformation of specific binding data in mesenteric adipose tissue (B_max, 404 fmol/mg protein; K_d, 333 pmol/L) and vascular membranes (○). d, Representative competitive inhibition curves of ¹²⁵I-ANF binding to mesenteric adipose tissue. • indicates ANF; □, C-type natriuretic peptide; and ○, des-[Gln¹⁸,Ser¹⁹,Gly²⁰,Leu²¹,Gly²²]ANF-(4-23).

¹²⁵I–[Sar¹,Ile⁸]Ang II saturation binding curves in adipose tissue (Fig 3a⇓ and 3c⇓) exhibited a single class of high-affinity binding sites with a B_max of 211±4 fmol/mg protein and a K_d of 520±10 pmol/L (n=3). In mesenteric artery membranes, ¹²⁵I–[Sar¹,Ile⁸]Ang II saturation binding assays showed little specific binding, with a B_max 10 times lower than that observed in mesenteric fat (20±5 fmol/mg protein) and a similar K_d (620±40 pmol/L, n=3, Fig 3b⇓ and 3c⇓). The pharmacological profile of the ligand competing for ¹²⁵I–[Sar¹,Ile⁸]Ang II binding was consistent with an AT₁ receptor subtype, with the following rank order of potency: [Sar¹,Ile⁸]Ang II>Ang II>losartan>PD123319 (Fig 3d⇓). Losartan, a specific AT₁ antagonist, inhibited 100% of ¹²⁵I–[Sar¹,Ile⁸]Ang II binding, whereas PD123319, a nonpeptide AT₂ antagonist, did not compete for radioligand binding at concentrations of up to 10 μmol/L.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Pharmacological characterization of angiotensin II (Ang II) receptors in mesenteric adipose and vascular tissues. a and b, Representative ¹²⁵I–[Sar¹,Ile⁸]Ang II saturation binding curves in membranes prepared from mesenteric fat (a) and arteries (b). ○ indicates total binding; ▿, nonspecific binding; and •, specific binding. c, Scatchard transformation of the specific binding data. • indicates fatty membranes (B_max, 210 fmol/mg protein; K_d, 500 pmol/L); ○, vascular membranes (B_max, 15 fmol/mg protein; K_d, 580 pmol/L). d, Representative competitive inhibition curves of ¹²⁵I–[Sar¹,Ile⁸]Ang II binding to mesenteric adipose tissue. • indicates [Sar¹,Ile⁸]Ang II; ○, Ang II; ▿, losartan; and □, PD123319.

¹²⁵I–ET-1 saturation binding experiments in adipose tissue showed a single class of binding sites. Unlike ¹²⁵I–[Sar¹,Ile⁸]Ang II and ¹²⁵I-ANF, ¹²⁵I–ET-1 binding to mesenteric arteries was a saturable process, with saturation being reached between 1 and 2 nmol/L (Fig 4a⇓ and 4b⇓). Scatchard analysis demonstrated a single class of high-affinity binding sites. The binding parameters derived from competitive inhibition with ET-1 showed a B_max of 179±21 and 312±7 fmol/L mg protein and a K_d of 215±45 and 580±111 pmol/L for mesenteric arteries and adipose tissue, respectively. BQ-123–sensitive binding was 79% and 64% for mesenteric arteries and adipose tissue, respectively (Fig 4c⇓ and 4d⇓ and Table⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Pharmacological characterization of endothelin-1 (ET-1) receptors in mesenteric adipose and vascular tissues. a, Representative ¹²⁵I–ET-1 saturation specific-binding curves in membranes prepared from mesenteric adipose tissue (•) and arteries (○). b, Scatchard transformation of specific binding data in mesenteric adipose tissue (•; B_max, 390 fmol/mg protein; K_d, 600 pmol/L) and mesenteric arteries (○; B_max, 205 fmol/mg protein; K_d, 200 pmol/L). c and d, Representative competitive inhibition curves of ¹²⁵I–ET-1 binding to mesenteric arteries (c) and mesenteric adipose tissue (d). ○ indicates ET-1; •, BQ-123.

cGMP Measurements

ANF and CNP stimulated cGMP generation by adipose tissue in a dose-dependent manner (Fig 5a⇓). The threshold concentration was 1 to 10 nmol/L, with maximal stimulation occurring in the presence of 1 μmol/L of either ANF or CNP as 50- and 22-fold increases, respectively, over baseline values. Mesenteric arteries did not respond to either ANF or CNP concentrations of 1 μmol/L (Fig 5b⇓), indicating the absence of guanylate cyclase–coupled receptors. C-ANF (1 μmol/L) did not evoke any cGMP stimulation by either tissue (Fig 5a⇓). SNP stimulated cGMP production by the mesenteric arteries in a dose-dependent manner (Fig 5b⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Graphs showing cGMP stimulation by atrial natriuretic factor (ANF), C-type natriuretic peptide (CNP), and des-[Gln¹⁸,Ser¹⁹,Gly²⁰,Leu²¹,Gly²²]ANF-(4-23) (C-ANF) in mesenteric adipose tissue (a) and by ANF, CNP, C-ANF, and sodium nitroprusside (SNP) in mesenteric arteries (b).

Affinity Cross-linking Studies

To further characterize ANF receptor subtypes, affinity cross-linking studies were performed on fatty and vascular membranes. DSS cross-linking of ¹²⁵I-ANF revealed two receptor species of 130 and 70 kD on adipose tissue (Fig 6⇓, lane 1). The proportion of high- and low-molecular-weight bands assessed by densitometry was ≈3% and 97%, respectively. ANF, C-ANF, or CNP (1 μmol/L each) (Fig 6⇓, lanes 2, 3, and 4, respectively) completely inhibited ¹²⁵I-ANF binding in both high- and low-molecular-weight bands. Vascular tissue did not exhibit any chemical cross-linking of ¹²⁵I-ANF (Fig 6⇓, lanes 5 and 6).

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

Autoradiography by SDS-PAGE of cross-linked atrial natriuretic factor (ANF) receptors in adipose (lanes 1 through 4) and vascular (lanes 5 and 6) tissues in the absence (lanes 1 and 5) and presence (lanes 2 and 6) of 10⁻⁶ mol/L ANF, des-[Gln¹⁸,Ser¹⁹,Gly²⁰,Leu²¹,Gly²²]ANF-(4-23) (lane 3), and C-type natriuretic peptide (lane 4).

Autoradiographic Localization of ¹²⁵I-ANF

To further substantiate our findings in vascular and fatty particulate fractions, we undertook in situ autoradiographic localization of ¹²⁵I-ANF binding sites in the mesenteric vascular bed. Two minutes after injecting ¹²⁵I-ANF, most adipocytes surrounding the mesenteric arteries were labeled, as indicated by dense deposits of overlaid silver grains that followed the contour of the adipocyte plasma membrane (Fig 7a⇓). Mesenteric arterioles also exhibited a significant accumulation of silver grains. Injection of an excess of unlabeled ANF together with ¹²⁵I-ANF resulted in a significant decrease in the number of silver grains over adipocytes (Fig 7b⇓). Reduced radioactivity was also observed in the mesenteric arteries but to a lesser extent, indicating its nonspecific nature.

• Download figure
• Open in new tab
• Download powerpoint

Figure 7.

In situ autoradiographic localization of ¹²⁵I–atrial natriuretic factor binding sites in the mesentery in the absence (a) and presence (b) of unlabeled atrial natriuretic factor (original magnification ×160).

Autoradiographic Localization of ¹²⁵I–ET-1

Fig 8⇓ depicts in vitro autoradiographic localization of ¹²⁵I–ET-1 binding sites in mesenteric artery and surrounding adipose tissue. Dense deposits of silver grains labeled the arteriole wall (Fig 8a⇓) and the contour of adipocyte plasma membranes (Fig 8c⇓). Excess of unlabeled ET-1 together with ¹²⁵I–ET-1 resulted in the almost complete disappearance of silver grains in both arteriole (Fig 8b⇓) and adipose tissue (Fig 8d⇓), indicating a specific binding.

• Download figure
• Open in new tab
• Download powerpoint

Figure 8.

In vitro autoradiographic localization of ¹²⁵I–endothelin-1 binding sites in mesenteric arterioles (a and b) and surrounding adipose tissue (c and d) in the absence (a and c) and presence (b and d) of unlabeled endothelin-1 (original magnification ×400).

Mesenteric Artery Perfusion

In isolated rat mesenteric arteries perfused with Krebs’ solution, mean basal perfusion pressure in intact and defatted preparations was 21±2 and 19±1 mm Hg, respectively. NE, AVP, ET-1, and Ang II were administered in the perfusate of both intact and defatted preparations in amounts varying from 10 to 100 μg. Ang II up to 100 μg did not evoke a significant pressor response in intact (not shown) or defatted mesenteric arteries (Fig 9⇓). NE, AVP, and ET-1 induced a dose-dependent pressor effect in both intact (not shown) and defatted preparations. NE exerted a transient pressor effect, whereas ET-1 elicited a long-lasting vasopressor response (Fig 9⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 9.

Tracings showing representative responses of isolated defatted, perfused mesenteric arteries to several agents. NE indicates norepinephrine; ET-1, endothelin-1; AVP, vasopressin; and ANG II, angiotensin II.