I. Effects on Nonesterified Fatty Acid and Plasma Urea N

Effects of a2-Adrenoceptor Antagonists on Metabolic Processes of Swine: I. Effects on Nonesterified Fatty Acid and Plasma Urea Nitrogen Concentrations in Jugularly Catheterized Pigs R. M. Cleale1, J. M. Ingling, D. J. Search, J. R. Hadcock2, and M. H. Pausch Fort Dodge Animal Health, Cyanamid Agricultural Research Center, Princeton, NJ 08543-0400

ABSTRACT: The presence of a2-adrenoceptors in membranes from omental and s.c. adipose tissue from gilts and barrows was shown in saturation binding assays with [3H]yohimbine. Four trials tested effects of a2-adrenoceptor antagonists (A2AA) on plasma concentrations of NEFA and urea nitrogen (PUN). In Trial 1, barrows were given i.v. injections of saline, 200 mg/kg BW of one of three A2AA (efaroxan, idazoxan, or RX821002), or 25 mg/kg BW of isoproterenol. Concentrations of NEFA were measured in plasma harvested every 15 min from 1 h before to 2 h after treatment. Compared with results for salinetreated pigs, areas under the curve (AUC) for NEFA were increased ( P < .05) by efaroxan, RX821002, and isoproterenol. In Trial 2, barrows received i.v. doses of saline, efaroxan (200 or 400 mg/kg BW), or RX821002 (200 or 400 mg/kg BW). Levels of NEFA were quantified in plasma obtained at 15-min intervals through 2 h after treatment. Among pigs treated with RX821002 at 400 mg/kg BW, mean NEFA AUC was more than three times greater ( P < .05) than that for saline-treated animals. Trial 3 tested whether NEFA responses to A2AA were due to direct effects on a2receptors or involved b-adrenoceptor mediation. Pigs were first treated i.v. with saline or propranolol ( 1 mg/kg BW). One hour later, pigs were treated i.v.

with RX821002 (400 mg/kg BW) or the b-adrenoceptor agonist cimaterol (25 mg/kg BW). Compared to values for pigs treated with saline at both injections, NEFA AUC among pigs treated with saline at the first injection and RX821002 at the second doubled ( P > .05). Plasma NEFA AUC among pigs treated with saline then cimaterol rose nearly fourfold ( P < .05) compared with saline-treated controls. Mean NEFA AUC among propranolol-treated pigs was similar to values for saline-treated pigs, suggesting b-adrenoceptor involvement in the effect of A2AA on NEFA. In Trial 4, pigs were treated s.c. 10 times at 8-h intervals with saline, RX821002 (400 mg/[kg BW·injection]), cimaterol (20 mg/[kg BW·injection]) or recombinant porcine somatotropin (rpST; 1 mg/ [pig·injection]). After the 10th treatment, only cimaterol increased NEFA AUC compared to salinetreated controls ( P < .05). Mean PUN AUC was reduced by RX821002 and rpST compared to controls; PUN among rpST-treated pigs was lower than that among RX821002-treated pigs ( P < .05). In summary, A2AA increase lipolysis in swine by potentiating lipolytic effects of endogenous catecholamines on badrenoceptors. Reduced PUN suggests improved nitrogen efficiency may result from treatment with A2AA.

Key Words: Pigs, Antagonists, Lipolysis, Metabolism 1998 American Society of Animal Science. All rights reserved.

Introduction Adrenoceptors are membrane-bound G protein coupled receptors through which catecholamines act to regulate various aspects of metabolism, including cellular and organ functions. Lipolysis in mammalian adipose tissue is regulated by cell-surface a2- and b-

1To

whom correspondence should be addressed: P.O. Box 400. address: Wyeth-Ayerst Research, CN 8000, Princeton, NJ 08543-8000. Received April 24, 1997. Accepted December 2, 1997. 2Present

J. Anim. Sci. 1998. 76:1838–1848

adrenoceptors. Epinephrine and norepinephrine stimulate a2- and b-adrenoceptors, but with opposing effects on adenylate cyclase activity, intracellular cyclic AMP accumulation, and lipolysis (Fain and Garcia-Sainz, 1983). Stimulation of bAR in human adipocytes results in activation of the adenylate cyclase complex and stimulation of lipolysis, and stimulation of a2-adrenoceptors is accompanied by inhibition of the adenylate cyclase complex and inhibition of lipolysis (Richelsen, 1986). In contrast to results in humans and other species, in vitro (Mersmann, 1984; Hu et al., 1987) and in vivo (Mersmann, 1987) studies provided no evidence for a2-adrenoceptor-mediated inhibition of lipolysis in swine.

1838

PLASMA METABOLITE CONCENTRATIONS IN SWINE

Effects of catecholamines on lipid metabolism are determined by the affinity of adrenoceptors for their cognate agonists and the relative abundance of each receptor type in adipose tissue (Fain and GarciaSainz, 1983; Richelsen, 1986). Predictions that treatment with a2-adrenoceptor antagonists ( A2AA) could stimulate net lipolysis were confirmed by Galitzky et al. (1988); oral administration of the A2AA yohimbine to humans increased concentrations of glycerol and NEFA. These results were confirmed in human adipocytes in situ (Arner et al., 1990). A further effect of catecholamines on metabolism is reflected by the well-characterized anabolic effects of b-adrenoceptor agonists. Treatment of swine with the b-adrenergic agonist cimaterol stimulated skeletal muscle deposition in swine (Jones et al., 1985; Moser et al., 1986; Cromwell et al., 1988). Studies reported herein examined effects of parenteral administration of various A2AA on plasma NEFA and urea nitrogen ( PUN) concentrations in swine.

Materials and Methods Binding Assays. From two barrows and two gilts (BW approximately 110 kg) from an unrelated study, samples of omental and s.c. adipose tissue were harvested immediately after the animals were killed. Tissues were placed in ice-cold sterile physiological saline and transported to the laboratory, where adipose tissue membranes were prepared using procedures described by Bahouth and Malbon (1988). Membranes were suspended in lysing buffer and stored at −80°C. The protein content of membrane preparations was determined by the method of Lowry et al. (1951). Binding assays were performed with [3H]yohimbine (Amersham, Arlington Heights, IL) in 96-well microtiter plates according procedures described by Kobilka et al. (1987) except the binding assay buffer contained 50 mM HEPES, pH 7.4, 10 mM MgCl2, .25% BSA, and protease inhibitors ( 5 mg/mL leupeptin, 5 mg/mL aprotinin, 100 mg/mL bacitracin, and 100 mg/ mL benzamidine). All components were diluted in binding buffer containing protease inhibitors and added to the microtiter plate wells in the following order: binding buffer, non-radiolabeled competitor (10 mM yohimbine), and .1 to 40 nM [3H]yohimbine (75 to 95 Ci/mmol). Binding reactions were initiated by adding 75 mg of membrane protein in a 170 mL volume. Final reaction volume was 200 mL. Incubations were carried out at approximately 20°C for 2 h. Free radioligand was separated from bound by rapid filtration through a glass fiber filter using a cell harvester (Inotech Biosystems International, Lansing, MI). Filter disks were washed four times with cold ( 4 °C ) binding buffer lacking BSA prior to quantifying radioactivity in a liquid scintillation counter (Micromedic Systems, Huntsville, AL; 60% efficiency). Bind-

1839

ing data were subjected to computer analysis assuming a single binding site using a commercial software program (GraphPad Prism version 2.0, GraphPad Software, San Diego, CA) to calculate Kd and Bmax for individual tissues from each animal. Statistical evaluations of estimates of Kd and Bmax were performed by two-way ANOVA (animal sex and tissue depot) using GLM procedures of SAS (1990). General Animal Care. Surgical procedures and experimental protocols were approved by the Cyanamid Agricultural Research Center Animal Care and Use Committee. Following surgery, pigs were at all times individually housed in stainless steel pens on raised slatted floors and fed a ground corn-soybeanbased diet containing 20% CP and 1.15% lysine. Animals had ad libitum access to water from nipple waterers. Animals also had ad libitum access to feed. However, to encourage animals to eat, feed was offered in equal portions at 0630 and 1300. Previous studies (R. Cleale, unpublished data) demonstrated feeding animals at 0630 (at least 1 to 1.5 h before initiation of blood collections) ensured that uniform concentrations of NEFA were measured in plasma samples collected during the pretreatment period on study days. For each experiment, animals were weighed on the day prior to treatment for the purpose of calculating the dose of drug to be administered. All in vivo experiments were performed with conscious swine into which cannulas were surgically placed in jugular veins. Placement and Maintenance of Cannulas. Surgical placement of cannulas in jugular veins occurred at least 7 d prior to experiments and was performed according to the protocol described by Kraft et al. (1985). Between blood collections within each experiment, cannulas were filled with saline containing .15 M sodium citrate to prevent blood coagulation. In the time interval between experiments, catheters were flushed once daily with sodium citrate followed by flushing and filling catheters with heparinized saline (400 IU/mL) to maintain patency. Plasma Metabolite Concentration Assays. In all studies reported here, NEFA concentrations served as the indicator of lipolytic activity. Nonesterified fatty acids produced as a result of lipolysis can be reintroduced into triacylglycerols by the process of reesterification. Glycerol is another analyte useful as an indicator of lipolysis. Unlike NEFA, glycerol produced during lipolysis cannot be reused for lipolysis because adipose tissue lacks the enzymes necessary to phosphorylate glycerol (see review by Bartley, 1980). The rationale for choosing NEFA as the preferred analyte in our studies is that due to reesterification, changes measured in plasma NEFA would be considered indicators of net lipolysis. Plasma NEFA concentrations were determined colorimetrically using a commercial kit (Wako NEFA C; acyl coenzyme A-acyl coenzyme A synthetase method; Wako Chemicals USA, Dallas, TX). This

1840

CLEALE ET AL.

procedure was adapted to a 96-well microtiter plate format and automated for use on a Biomek 1000 system (Beckman Instruments, Fullerton, CA). In each well, 30 mL of standard or sample was mixed with 75 mL of Reagent A and incubated for 20 min at 37°C. Next, 150 mL of Reagent B was added, plates were incubated 20 min at 37°C and 10 min at approximately 20°C, and finally read at 540 nm. The intra- and interassay variability was 3.3 and 2.6%, respectively. Plasma urea nitrogen concentrations were determined colorimetrically using a commercial kit involving phenol/nitroprusside and sodium hypochlorite reagents (procedure no. 640, Sigma Diagnostics, St. Louis, MO). This procedure, in which ammonia derived from urea is quantified, was adapted for use in a 96-well microtiter plate format and automated using the Biomek 1000. Samples and standards were diluted 1:4 with distilled water. To each well, 20 mL of sample was mixed with 50 mL of urease reagent, and the mixture was incubated at approximately 20°C for 20 min. Following incubation, 100 mL of phenol nitroprusside solution and 100 mL of alkaline hypochlorite solution were added, samples were incubated at approximately 20°C an additional 20 min, and absorbance was measured at 540 nm. The intra- and interassay variability was 5.0 and 2.6%, respectively. Trial 1. The objective of this experiment was to evaluate the ability of three putative a2-adrenoceptor antagonists to provoke increased concentrations of NEFA in blood plasma from treated swine. Fourteen cannulated barrows (BW = 72.6 ± 4.7 kg) were assigned to five treatments in a completely random experimental design. Blood samples (4.5 mL) were collected from cannulas into heparinized tubes (Monovette, Sarstedt Inc., Newton, NC) at 15-min intervals from 0800 through 1100 and immediately placed on ice. Each treatment was administered as a single i.v. bolus through the cannula immediately following the 0900 blood collection. Treatments included sterile saline ( 9 mL/pig; negative control), 25 mg/kg BW of the potent bAR agonist isoproterenol (positive control; Sigma Chemical Co., St. Louis, MO) dissolved in saline at a concentration of .225 mg/mL, and three commercially available putative A2AAselective antagonists, each dissolved in saline at a concentration of 1.8 mg/mL and administered to animals at a dose rate of 200 mg/kg BW. Experimental compounds included efaroxan hydrochloride, idazoxan hydrochloride, and the 2-methoxy derivative of idazoxan, which is referred to in the published literature (Stillings et al., 1986) and in the present paper as RX821002. The dose of A2AA was selected based on the report of Lafontan et al. (1992) that showed yohimbine administered to dogs at the rate of 200 mg/kg BW was safe and stimulated plasma NEFA elevation. Two animals received saline, whereas three animals were assigned to each of the other treatments.

Blood samples were centrifuged at 1,000 × g for 20 min. Plasma was removed and analyzed for NEFA concentrations as described above. Areas under the curves ( AUC) for NEFA were estimated for pretreatment and posttreatment periods using the trapezoidal summation method (Abramovitz and Slegun, 1972). In addition, peak NEFA concentration and mean NEFA concentration were calculated. Effects of treatments on response variables were corrected for pretreatment variation by analysis of covariance using the GLM procedures of SAS (1990). Comparison of least squares means were performed using Fisher’s protected LSD procedure (Steel and Torrie, 1980). Trial 2. The objective of this study was to evaluate effects on plasma NEFA of varying doses of the two most active compounds evaluated in Trial 1. Twenty barrows (BW = 39.1 ± 2.1 kg) with indwelling catheters were randomly assigned to five treatments yielding four animals per treatment. Treatments, which were administered as single i.v. infusions through the catheter, included saline; efaroxan HCl (Research Biochemicals International, Natick, MA) dissolved in saline at concentrations of .5 or 2.0 mg/ mL and dosed at rates of 100 or 400 mg/kg BW, respectively; and RX821002 (Research Biochemicals) dissolved in saline at concentrations of .5 or 2.0 mg/ mL and dosed at rates of 100 and 400 mg/kg BW, respectively. Treatments were administered immediately after the 0900 bleeding. Blood samples (4.5 mL) were collected into heparinized tubes at 15-min intervals beginning at 0830 and continuing until 1100 and placed on ice. Blood samples were centrifuged at 1,000 × g for 20 min. Plasma was removed and analyzed for NEFA concentrations. Pretreatment and posttreatment areas under the NEFA curve, peak NEFA concentrations, and mean NEFA concentrations were determined for each animal. Data were statistically analyzed using analysis of covariance to correct posttreatment values for pretreatment variability. Comparisons of least squares means were performed as described for Trial 1. Trial 3. The objective of this study was to evaluate the ability of propranolol, a bAR antagonist, to block the stimulation of plasma NEFA concentrations induced in previous trials by the A2AA RX821002. The experiment was designed to test the hypothesis that RX821002 causes NEFA elevation by potentiating effectiveness of endogenous catecholamines on bAR. If this is true, then blocking bAR should block effects of RX821002 on NEFA concentrations. In this experiment, blood samples (4.5 mL) were collected at 15-min intervals over a 4-h period beginning at 0800 on a single study day. A total of 18 cannulated barrows (BW = 41.1 ± 1.2 kg) were used in this study, which involved five treatment combinations. During the first hour of blood collection samples were collected to establish pretreatment plasma NEFA concentrations. Throughout the experiment all treatments were administered intravenously through


cannulas. Following the 0900 bleeding, animals were administered injections of either saline ( n = 11 pigs) or 1 mg of the bAR antagonist DL-propranolol (Sigma Chemical) per kg BW ( n = 7 pigs) dissolved in saline at a concentration of 8 mg/mL. Blood collections resumed, and following the 1000 bleeding pigs were injected with one of three treatments. From the group of 11 animals that received saline at 0900, four received a second injection of saline (negative control), four received 400 mg RX821002/kg BW dissolved in saline at a concentration of 3 mg/mL, and three received 25 mg cimaterol/kg BW (positive control; American Cyanamid Co., Princeton, NJ) dissolved in saline at a concentration of .2 mg/mL. From the group of seven animals that received propranolol at 0900, three received 400 mg RX821002/kg BW and four received 25 mg cimaterol/kg BW. Blood samples were placed on ice until they were centrifuged (1,000 × g for 20 min). Plasma was removed and NEFA concentrations quantified with the method previously described. Pretreatment (0800 through 1000) and posttreatment (1015 though 1200) area under the NEFA curve, peak NEFA concentration, and mean NEFA concentration were determined for each animal. Data were statistically evaluated by analysis of covariance to correct posttreatment values for pretreatment variability and least squares means compared as described for Trial 1. Trial 4. Objectives of this study were to determine 1 ) whether pigs repeatedly injected with RX821002 become increasingly insensitive to the drug using plasma NEFA concentrations as the response variable and 2 ) whether RX821002 induces changes in PUN concentrations. The experiment used a randomized complete block experimental design in which separate groups of pigs were treated under the same protocol at different times (BW = 55.1 ± 1.3 kg; n = 18 in block 1; n = 16 in block 2 ) and four treatments. Treatments included saline; RX821002 dissolved in saline (9.0 mg/ mL) and administered at a dose rate of 400 mg/kg BW; cimaterol dissolved in saline (.5 mg/mL) and administered at a dose rate of 20 mg/kg BW; and recombinantly derived porcine somatotropin ( rpST; American Cyanamid) dissolved ( 1 mg/mL) in .025 M carbonate/bicarbonate-buffered saline (pH 9.4) and administered at a dose rate of 1 mg/injection. Treatments were administered as s.c. injections in the neck at 8-h intervals on 10 occasions (0800, 1600, and 2400 on d 1 through d 3, then at 0800 on d 4). Preliminary studies (R. Cleale, unpublished data) showed administration of RX821002 via i.v. or s.c. routes resulted in bioequivalent responses on NEFA concentrations in cannulated swine over a period of 2 h following treatment. For ease of drug administration, s.c. injections were used in the present study. Blood samples (4.5 mL) were collected from cannulas at 0730 and 0800 prior to treatments 1, 4, and 7 (0800 treatments on d 1 through d 3). Following these treatments, blood samples were col-

1841

lected at 0830, 0900, and 0930. Before the last treatment injection at 0800 on d 4, blood samples were collected at 15-min intervals beginning at 0700. Following treatment, sampling continued at 15-min intervals until 1000. Immediately following collection, blood samples were placed on ice until they were centrifuged (1,000 × g for 20 min). Plasma was removed, and NEFA and PUN concentrations were quantified with methods previously described. Feed consumption can affect plasma metabolite concentrations. Therefore, the amount of feed consumed daily by each animal over the 4 d of this study was measured. For NEFA and PUN, AUC and mean concentration were determined for each animal observation. Analysis of variance was performed with a GLM for a splitplot design to evaluate effects of treatments, blocks, treatment × block interactions, time, and treatment × time interactions (SAS, 1990). Because treatment × time interactions were significant, an independent ANOVA was performed on data generated in response to each injection evaluated (1, 4, 7, and 10) to evaluate effects of treatments and blocks on response variables. Mean comparisons within each injection were performed using Fisher’s protected LSD procedure (Steel and Torrie, 1980).

Results and Discussion Binding Assays. Computer analysis of saturation curve data yielded apparent equilibrium dissociation constants ( Kd) for [3H]yohimbine ranging from 1.78 to 5.68 that were not statistically different from one another (Figure 1). These values are, in general, similar to Kd estimates obtained in studies evaluating binding of [3H]yohimbine to human adipocyte membranes (Lafontan et al., 1983; Richelsen, 1986). Interactions between animal sex and adipose tissue depot were not statistically significant. Averaged across tissue depots, Bmax was 39% higher ( P = .0754) among gilts (129 fmol/mg protein) than among barrows (92 fmol/mg protein). This apparent effect of gender agrees with data from Richelsen (1986), who demonstrated that s.c. fat from women contained 73% more [3H]yohimbine binding sites per unit of protein than s.c. fat from men. Our data, however, show the number of a2-adrenoceptors in swine adipocyte membranes to be far fewer than the number measured by Lafontan et al. (1983) and Richelsen (1986) in human subcutaneous adipose tissue. Trial 1. Mean plasma NEFA concentrations prior to treatment averaged approximately 90 mEq/L (Table 1). Previous experience in our lab has shown that NEFA concentrations are influenced greatly by feeding schedules. Although pigs in each experiment reported in this paper were fed for ad libitum feed consumption, provision of fresh feed prior to initiation of blood sample collection prompted pigs to consume a

1842

CLEALE ET AL.

Table 1. Effects of intravenous administration of saline, imidazoline a2-adrenoceptor antagonists, or isoproterenol on plasma NEFA concentrations in swine (Trial 1) Treatmenta Item No. of pigs Body weight, kg

Saline

Efaroxan

Idazoxan

RX821002

Isoproterenol

SEM

2 71.6

3 65.6

3 73.6

3 76.5

3 77.1

— 5.1

Pretreatment NEFA observations, 1 h Mean concentration, mEq/L Maximum concentration, mEq/L AUCb

91.6 112.0 90.6

86.1 97.0 85.9

74.1 86.3 74.2

97.6 118.3 98.3

90.4 124.3 88.5

8.6 15.9 8.4

Posttreatment NEFA observationsc, 2 h Mean concentration, mEq/Ld Maximum concentration, mEq/Ld AUCb,d

122.2z 171.0z 215.7z

268.2y 393.6y 476.9xy

176.0yz 172.6z 308.6yz

393.9x 1460.8x 584.5x

32.8 41.5 57.0

296.8xy 399.9y 505.8x

aEfaroxan, idazoxan, and RX821002 were administered as hydrochloride salts at a dosing rate of 200 mg/kg BW; isoproterenol was administered at a rate of 25 mg/kg BW. All treatments were administered intravenously in sterile saline. bArea under the treatment × time curve. cData were analyzed by analysis of covariance to correct for pretreatment variation. Means are least squares means. dF-test for main effect of treatment is significant ( P < .01). x,y,zLeast squares means in the same row lacking a common superscript letter are different by Fisher’s protected LSD (P < .05).

meal and aided in stabilizing baseline concentrations of NEFA. In all studies reported here, it was standard practice to provide fresh feed to animals at least 1 h before blood collections began. Plasma NEFA concentrations in Trial 1 remained in a narrow range during the five bleedings preceding treatment administration (Figure 2). Following treatment, NEFA concentrations among saline-treated control animals remained

Figure 1. [3H]Yohimbine saturation radioligand binding to crude membranes prepared from omental and subcutaneous adipose tissue harvested from gilts (n = 2) and barrows (n = 2) weighing 110 kg. Binding data were fit to an equation that assumes a single binding site using a commercial computer software program (GraphPad Prism version 2.0, GraphPad Software, San Diego, CA) to calculate Kd (nM) and Bmax (fmol/mg protein). Data are expressed as means ± SEM.

virtually unchanged from pretreatment values. Among pigs treated with the b-adrenoceptor agonist isoproterenol, NEFA concentrations peaked at 1,460.8 mEq/L within 15 min after treatment, a level higher ( P < .05) than those in all other treatments. However, NEFA concentrations among isoproterenol-treated animals decreased to near 300 mEq/L by 30 min posttreatment and continued to decline slowly for the remainder of the study period. The short response to isoproterenol is likely due to its short biological halflife, which is estimated to be .05 h (see review by Morgan, 1990). Among the imidazoline A2AA, RX821002 provoked the largest increase in NEFA concentration, reaching

Figure 2. Effects of i.v. treatment with saline (n = 2), imidazoline a2-adrenoceptor antagonists (efaroxan, n = 3; idazoxan, n = 3; or RX821002, n = 3, each at 200 mg/kg BW), or isoproterenol (25 mg/kg BW; n = 3) on plasma NEFA concentrations in swine in Trial 1. Data are expressed as means ± SEM.

1843


Table 2. Effects of intravenous administration of saline or doses of imidazoline a2-adrenoceptor antagonists on plasma NEFA concentrations in swine (Trial 2)a Efaroxan ( mg/kg) Item No. of pigs Total dose administered, mg/pig Body weight, kg

Saline

100

400

RX821002 ( mg/kg) 100

400

SEM

4 0.0 37.7

4 15.3 38.3

4 64.9 40.6

4 15.9 39.9

4 62.6 39.1

— — 2.1

Pretreatment NEFA observations, .5 h Mean concentration, mEq/L Maximum concentration, mEq/L AUCb

119.4 127.4 29.9

139.9 160.8 35.0

90.2 97.4 22.6

164.1 191.0 41.1

143.7 175.1 35.9

32.2 38.1 8.0

Posttreatment NEFA observationsc, 2 h Mean concentration, mEq/Ld Maximum concentration, mEq/Ld AUCb,d

168.1y 218.1y 295.7y

188.2y 261.2y 333.0y

252.9y 408.5y 448.8y

224.4y 414.9y 394.6y

523.3x 766.7x 932.7x

51.0 82.4 90.7

aHydrochloride salts of efaroxan and RX821002 were administered intravenously in sterile saline. bArea under the treatment × time curve. cData were analyzed by analysis of covariance to correct for pretreatment variation. Means are least squares means. dF-test for main effect of treatment is significant ( P < .01). x,yLeast squares means in the same row lacking a common superscript letter are different by Fisher’s protected LSD

a maximum averaging nearly 400 mEq/L. This concentration was more than that among saline-treated controls ( P < .05) but less than ( P < .05) the maximum NEFA concentration stimulated by isoproterenol. The NEFA response to efaroxan treatment was similar to that for RX821002 with respect to mean and maximum concentrations. Idazoxan was without effect on NEFA mean, maximum, and AUC compared to saline-treated control pigs. The greater biological effect of RX821002 in pigs compared to idazoxan was anticipated because it was previously reported that pharmacological potency of RX821002 is at least 10-fold that of idazoxan (Stillings et al., 1986). Other data (Galitzky et al., 1990) showed the rank order of potency to be RX821002 > efaroxan > idazoxan using in vitro binding studies with human fat cells. Area under the treatment × time curve is the most important response variable in these studies because it estimates the total response over the entire study period. Whereas isoproterenol stimulated the highest peak NEFA concentration, mean AUC for animals treated with efaroxan, RX821002, and isoproterenol were not different from one another but were higher than the mean AUC for saline-treated controls ( P < .05). Relative to the maximum response stimulated, responses to efaroxan and RX821002 were more persistent than the response to isoproterenol (Figure 2). Data from this and subsequent studies are especially interesting because increased NEFA concentrations were induced by A2AA administration to alimented swine. Galitzky et al. (1988) and Lafontan et al. (1992) showed that oral administration of the prototype A2AA, yohimbine, to humans fasted overnight resulted in elevated concentrations of NEFA. However, in individuals administered yohimbine after

(P < .05).

a meal, not only was there no increase in NEFA, but NEFA concentrations decreased. Lafontan et al. (1992) attributed the lack of response in fed individuals to elevated insulin concentrations, which are inversely related to NEFA concentrations. Insulin secretion was induced not only as a consequence of food consumption but also by direct stimulation of insulin secretion from pancreatic cells caused by a2adrenoceptor blockade. Trial 2. The mean pretreatment NEFA concentration of all treatment groups was 131 mEq/L and was considered normal for our pigs (Table 2). However, mean pretreatment NEFA concentrations ranged from 90.2 among animals that ultimately received 400 mg efaroxan/kg BW to 164.1 mEq/L among pigs that ultimately were treated with 100 mg RX821002/kg BW. None of the pretreatment variable means (mean concentration, maximum concentration, and AUC) were statistically different from one another. Mean concentrations of NEFA remained essentially unchanged among saline-treated pigs throughout the experiment (Figure 3). Although the magnitude of increase varied depending on treatment, the general shape of NEFA concentration curves following administration of efaroxan or RX821002 was similar, with peak NEFA concentrations noted 30 to 45 min after treatment, followed by decreases in NEFA concentrations over the remainder of the bleeding period. Greatest NEFA concentration increases were observed among pigs treated with 400 mg RX821002/ kg BW, where mean and maximum NEFA concentrations averaged 523.3 and 766.7 mEq/L, respectively. These means were higher than those for any other treatments ( P < .05). No statistically significant increases in mean or maximum NEFA concentrations were noted in response to treatment with either dose of efaroxan or to treatment with 100 mg RX821002/kg

1844

CLEALE ET AL.

Figure 3. Effects of i.v. treatment (n = 4 per treatment) with saline, efaroxan (100 or 400 mg/kg BW), or RX821002 (100 or 400 mg/kg BW) on plasma NEFA concentrations in swine in Trial 2. Data are expressed as means ± SEM.

BW. However, review of Figure 3 and the posttreatment AUC means presented in Table 2 suggest that NEFA concentrations were numerically increased in a dose-dependent manner in response to treatment with RX821002. Trial 3. In the hour following treatment of pigs with saline or propranolol, the concentration of NEFA did not change compared to concentrations noted in the preceding hour (Figure 4). Therefore, for purposes of calculating NEFA variable means for this experiment, the pretreatment period was defined as the 2-h time interval preceding treatment with RX821002

Figure 4. Effects of propranolol (1 mg/kg BW) on RX821002- (400 mg/kg BW) or cimaterol- (25 mg/kg BW) stimulated NEFA concentrations in swine in Trial 3. Data are expressed as means ± SEM.

or cimaterol, and the posttreatment period was defined as the interval following treatment with these compounds. The pretreatment NEFA concentration averaged across all treatment groups was approximately 85 mEq/L (Table 3). No significant pretreatment differences were noted among treatment groups with respect to mean and maximum NEFA concentrations, or AUC. Among pigs that received an initial treatment with saline at 0900 and a second i.v. treatment of saline at 1000, NEFA concentrations averaged 92.3 mEq/L

Table 3. Effects of propranolol on RX821002- or cimaterol-stimulated plasma NEFA concentrations in swine (Trial 3) Saline

Injection 1a: Item

Injection

2a:

Saline

RX821002

Propranolol Cimaterol

RX821002

Cimaterol

SEM

No. of pigs Body weight, kg

4 43.1

4 38.2

3 42.0

3 39.9

4 42.2

— 1.2

Pretreatment NEFA observations, 0800 to 1000 Mean concentration, mEq/L Maximum concentration, mEq/L AUCb

84.5 96.9 85.0

81.9 96.2 81.5

83.2 98.9 81.6

85.9 106.5 83.9

87.9 97.4 87.8

4.8 9.0 5.0

92.3y 108.7y 251.6y

191.3y 359.1y 542.6y

91.6y 132.9y 254.2y

94.0y 116.0y 249.8y

Posttreatment NEFA observationsc, 1015 to 1200 Mean concentration, mEq/Ld Maximum concentration, mEq/Ld AUCbd

422.0x 839.4x 1228.7x

60.6 115.3 177.3

aFor injection 1, either sterile saline or propranolol ( 1 mg/kg BW) solubilized in sterile saline was administered at 0900. For injection 2, either sterile saline, RX821002 (400 mg/kg BW) solubilized in sterile saline, or cimaterol (25 mg/kg BW) solubilized in sterile saline was administered at 1000. bArea under the treatment × time curve. cData were analyzed by analysis of covariance to correct for pretreatment variation. Means are least squares means. dF-test for main effect of treatment is significant ( P < .01). x,yLeast squares means in the same row lacking a common superscript letter are different by Fisher’s protected LSD (P < .05).


(Table 3 ) and did not change during the posttreatment period (Figure 4). Likewise, NEFA concentrations did not change in pigs that initially were treated with 1 mg propranolol/kg BW, followed 1 h later with treatment of either 400 mg RX821002/kg BW or 25 mg cimaterol/kg BW. However, in pigs that received an initial treatment of saline followed by treatment with 25 mg cimaterol/kg BW, NEFA concentration rose to an average maximum of 1,228.7 mEq/L, which was higher ( P < .05) than in pigs on all other treatments. For this same group of animals the average posttreatment NEFA concentration was 422 mEq/L and was higher ( P < .05) than in other treatments. Although the effect was not statistically significant, plasma NEFA concentration was numerically increased in pigs treated first with saline then with RX821002. The maximum plasma NEFA concentration achieved in pigs receiving this treatment averaged 359.1 mEq/L, which was more than three times that of salinetreated controls. Data indicate NEFA concentrations in pigs administered the A2AA RX821002 are not raised as a direct consequence of the drug. Propranolol, a bAR antagonist, completely blocked the elevation of NEFA concentrations induced by administration of RX821002. Therefore, effects of RX821002 on NEFA in swine seem to involve mediation by bAR. One possible effect of RX821002 may be to directly stimulate secretion of catecholamines. In studies with human subjects receiving i.v. infusions of the A2AA ethoxyidazoxan, an analog of RX821002, plasma concentrations of norepinephrine increased following initiation of treatment approximately 50% higher than pretreatment levels (Coupland et al., 1994). Elevated plasma concentrations of norepinephrine were also noted in humans to which yohimbine was orally administered, and the effect was not modified by propranolol treatment (Galitzky et al., 1988; Lafontan et al., 1992). It is interesting that Coupland et al. (1994) noted that not all subjects exhibited robust responses to treatment. We have noted similar variability in pigs, and this may partially explain why statistically significant responses to treatment with A2AA have at times been difficult to demonstrate. Another likely mechanism by which RX821002 increases plasma NEFA concentrations in swine is by blocking the agonist activity of catecholamines on a2adrenoceptors in adipose tissue. With a2-adrenoceptors blocked, the primary agonist effect of catecholamines on adipocytes is via b-adrenoceptors. In agreement with our data, Galitzky et al. (1988) showed oral administration of the A2AA yohimbine to human subjects induced elevated plasma concentrations of glycerol and NEFA, which could be inhibited if badrenoceptors were blocked with propranolol. Arner et al. (1990) showed treatment of human subjects with the A2AA phentolamine resulted in dramatic increases in concentrations of glycerol produced by perfused adipocytes in situ.

1845

It must be acknowledged that definite species differences exist with respect to ability of adipocytes to respond directly to various adrenergic agonists and antagonists. Based on understanding of adrenergic control of lipolysis (Fain and Garcia-Sainz, 1983), administration of A2AA should directly stimulate adenylate cyclase activity and lead to lipolysis. This was shown to be true in isolated fat cells from humans (van de Venter et al., 1992), rats, and dogs (MilavecKrizman and Wagner, 1978). By similar reasoning, it would be expected that administration of a2adrenoceptor agonists should inhibit lipolysis. Support for this hypothesis was provided by Richelsen (1986) and Mauriege et al. (1987), who showed that the a2adrenoceptor agonist clonidine had a pronounced in vitro antilipolytic effect in adipocytes from women. However, in vitro data of Mersmann (1984) and Hu et al. (1987) did not demonstrate such effects of a2adrenoceptor agonists in adipose tissue slices from swine, which led them to conclude that either swine adipose tissue does not have a2-adrenoceptors or that a2-adrenoceptors do not mediate inhibition of lipolysis in swine adipocytes. Trial 4. Over the 4 d of this experiment, feed consumption by pigs averaged 2.0 ± .1 kg/d (as-fed basis). Feed intake was not different among treatment groups. Plasma NEFA concentrations remained relatively constant throughout the experiment among salinetreated pigs, averaging approximately 93 mEq/L (Figure 5a). Following the first injection of RX821002, NEFA AUC was about 3.5 times higher than that for saline-treated controls ( P < .05; Table 4), and NEFA AUC among cimaterol-treated animals was eight times higher than that in saline controls ( P < .05). Following the fourth dose, NEFA AUC among pigs treated with RX821002 and cimaterol were, respectively, two times ( P > .05) and four times ( P < .05) higher than that of saline treated pigs; following the seventh dose, two times ( P > .05) and three times ( P > .05) higher than that of saline-treated pigs; and following the tenth dose, 1.4 times ( P > .05) and 1.9 times ( P < .05) higher than that of saline-treated pigs. A graph of mean NEFA concentrations following treatments 1, 4, 7, and 10 (Figure 5a) presents evidence that pigs became less sensitive to RX821002 and cimaterol with repeated treatment. Spurlock et al. (1994) showed the number of bAR in adipose tissue from pigs fed ractopamine decreased by up to 50% of pretreatment concentrations following initiation of treatment, so reduced sensitivity to cimaterol treatment was not an unexpected event. It is noted, however, that despite the diminishing sensitivity of adipocytes to cimaterol (based on NEFA concentrations), the benefits of long-term feeding of cimaterol (Jones et al., 1985; Moser et al., 1986; Cromwell et al., 1988) or an analog of cimaterol (clenbuterol; Ricks et al., 1984) on adiposity of hog carcasses was reported. On the basis of its presumed mechanism of action

1846

CLEALE ET AL.

Figure 5. Effects of repeated s.c. injections of saline (n = 9), RX821002 (400 mg·kg BW−1·injection−1; n = 8), cimaterol (20 mg·kg BW−1·injection−1; n = 9), or recombinantly derived porcine somatotropin (rpST, 1 mg/ injection; n = 8) on mean posttreatment plasma concentrations of NEFA (4a) and urea nitrogen (4b) in swine in Trial 4. Injections were administered at 8-h intervals. Data are expressed as means ± SEM.

(Etherton et al., 1995), it was not surprising that at no time during the experiment was a measurable effect of treatment with rpST observed on NEFA measurements. The evaluation of PUN concentrations was performed on study animals because chemicals that improve efficiency of protein metabolism are known to typically induce concomitant decreases in PUN levels (Preston, 1968; Davis et al., 1970; Wray-Cahen et al., 1991). It was of interest to obtain an early indication of whether RX821002 had the potential to improve efficiency of protein utilization. Over the period of study, PUN concentrations among saline-treated pigs remained relatively constant, averaging about 17 mg/ dL (Figure 5b). Because diurnal patterns of PUN concentrations do not respond to metabolism-modifying agents as acutely as NEFA concentrations (Wray-

Cahen et al., 1991), there were no measurable effects of treatment on PUN concentrations in the 2-h period following the first treatment. However, by the fourth treatment, drug-related effects were obvious. By far the greatest effect on PUN was induced by treatment with rpST; PUN AUC following treatments 4, 7, and 10 were approximately 35 to 45% lower than for saline-treated pigs ( P < .05). It is because of the documented effect of growth hormone on PUN concentrations in swine (Wray-Cahen et al., 1991) that rpST was included in the present experiment as a positive control to assess effects of cimaterol and RX821002 on PUN concentrations. Following the fourth treatment with RX821002 or cimaterol, means for PUN AUC were reduced, compared to untreated controls, by 30 ( P < .05) and 10% ( P > .05), respectively; following the seventh treatment by 15 ( P > .05) and 17% ( P < .05), respectively; and following the tenth treatment by 21 ( P < .05) and 16% ( P > .05), respectively. In this experiment, each pig received a total dose of cimaterol of approximately 3 mg per 24 h, and the average reduction in PUN AUC was approximately 16%. The effect of feeding even lower doses of cimaterol than those used in the present study on the lean content of hog carcasses has been frequently documented (Jones et al., 1985; Moser et al., 1986; Cromwell et al., 1988). Similarly, treating pigs with growth hormone improves efficiency of protein utilization (Caperna et al., 1991; Wray-Cahen et al., 1991) and greatly increases protein deposition in treated animals (Caperna et al., 1991). Despite fluctuation in PUN concentrations among animals treated with RX821002 in the postinjection time intervals studied (Figure 5b), the reduction of PUN among RX821002-treated animals suggests that this compound could change the efficiency of protein metabolism in swine. Additional experiments using nitrogen balance measurements in treated animals must be performed to determine the validity of this theory.

Implications Data reported herein provide supporting evidence for the role that a2-adrenoceptors play in mediating lipid metabolism. Our data demonstrate that treating swine with a2-adrenoceptor antagonists results in increased concentrations of nonesterified fatty acids. It is possible that a2-adrenoceptor antagonists directly stimulate lipolysis. However, our data suggest that lipolytic effects of a2-adrenoceptor antagonists on adipocytes are not direct. Instead, lipolysis seems to result from enhanced sensitivity of b-adrenoceptors to endogenous catecholamines when a2-adrenoceptors are blocked by a2-adrenoceptor antagonists and(or) stimulation of catecholamine secretion. A second feature of the data reported herein is the possible role a2-adrenoceptor antagonists might play in improving

1847


Table 4. Metabolic responses in pigs treated repeatedly with subcutaneous injections of metabolism-modifying agents (Trial 4)a Treatment Item No. of pigs Drug dose per injection Body weight, kg Area under NEFA concentration curve followingb Injection 1c Injection 4c Injection 7 Injection 10c

Saline

RX821002

Cimaterol

9 0 54.4

8 400 mg/kg 54.1

9 20 mg/kg 54.5

105.1z 108.7y 109.1 153.8y

Area under plasma urea nitrogen concentration curve followingb Injection 1 21.1 Injection 4c 20.4x c Injection 7 22.7x Injection 10c 32.0x

373.4y 220.1xy 240.9 214.2y 21.3 14.3y 19.2xy 25.2y

819.1x 431.8x 316.2 296.5x 22.1 18.4x 18.8y 27.0xy

rpST 8 1 mg/pig 55.4

SEM — — 1.2

101.0z 126.5y 136.5 192.6y

70.1 80.5 65.6 22.7

20.3 13.4y 14.1z 17.3z

1.3 1.1 1.2 1.8

aRX821002 and cimaterol were administered as subcutaneous injections in sterile saline; rpST is recombinantly derived porcine somatotropin and was administered as subcutaneous injections in .025 M carbonate/bicarbonate-buffered saline (pH 9.4). All animals received 10 doses of their respective treatment at 8-h intervals. bTreatments 1, 4, 7, and 10 occurred at 0800. Following injections 1, 4, and 7, plasma was harvested from blood samples collected at 0830, 0900, and 0930; following injection 10, plasma was harvested from blood samples collected at 0815, 0830, 0845, 0900, 0915, 0930, 0945, and 1000. cF-test for main effect of treatment is significant ( P < .05). x,y,zMeans in the same row lacking a common superscript letter are different by Fisher’s protected LSD ( P < .05).

efficiency of nitrogen metabolism. In summary, our data suggest that it may be possible to reduce the fat content and increase the lean content of pigs by treating with a2-adrenoceptor antagonists.

Literature Cited Abramovitz, M., and I. A. Slegun. 1972. Handbook of Mathematical Functions. Dover Publications, New York. Arner, P., E. Kriegholm, P. Engfeldt, and J. Bolinder. 1990. Adrenergic regulation of lipolysis in situ at rest and during exercise. J. Clin. Invest. 85:893−898. Bahouth, S. W., and C. C. Malbon. 1988. Subclassification of badrenergic receptors of rat fat cells: A re-evaluation. Mol. Pharmacol. 34:318−326. Bartley, J. C. 1980. Lipid metabolism and its disorders. In: J. J. Kaneko (Ed.) Clinical Biochemistry of Domestic Animals. pp 54−90. Academic Press, New York. Caperna, T. J., D. R. Komarek, D. Gavelek, and N. C. Steele. 1991. Influence of dietary protein and recombinant porcine somatotropin administration in young pigs: II. Accretion rates of protein, collagen and fat. J. Anim. Sci. 69:4019−4029. Coupland, N. J., J. E. Bailey, S. J. Wilson, W. Z. Potter, and D. J. Nutt. 1994. A pharmaco-dynamic study of the a2-adrenergic receptor antagonist ethoxyidazoxan in healthy volunteers. Clin. Pharmacol. Ther. 56:420−429. Cromwell, G. L., J. D. Kemp, T. S. Stahly, and R. H. Dalrymple. 1988. Effects of dietary level and withdrawal time on the efficacy of cimaterol as a growth repartitioning agent in finishing swine. J. Anim. Sci. 66:2193−2199. Davis, S. L., U. S. Garrigus, and F. C. Hinds. 1970. Metabolic effects of growth hormone and diethylstilbestrol in lambs. II. Effects of daily ovine growth hormone injections on plasma metabolites and nitrogen-retention in fed lambs. J. Anim. Sci. 30:236−240. Etherton, T. D., S. S. Donkin, and D. E. Bauman. 1995. Mechanisms by which porcine somatotropin (pST) decreases adipose tissue growth in growing pigs. In: S. B. Smith and D. R. Smith (Ed.)

The Biology of Fat in Meat Animals. pp 53−69. American Society of Animal Science, Champaign, IL. Fain, J. N., and J. A. Garcia-Sainz. 1983. Adrenergic regulation of adipocyte metabolism. J. Lipid Res. 24:945−966. Galitzky, J., D. Larrouy, M. Berlan, and M. Lafontan. 1990. New tools for human fat cell alpha-2A adrenoceptor characterization. Identification on membranes and on intact cells using the new antagonist [3H]RX821002. J. Pharmacol. Exp. Ther. 252: 312−319. Galitzky, J., M. Taouis, M. Berlan, D. Riviere, M. Garrigues, and M. Lafontan. 1988. a2-Antagonist compounds and lipid mobilization: Evidence for a lipid mobilizing effect of oral yohimbine in healthy male volunteers. Eur. J. Clin. Invest. 18:587−594. Hu, C. Y., J. Novakofski, and H. J. Mersmann. 1987. Hormonal control of porcine adipose tissue fatty acid release and cyclic AMP concentration. J. Anim. Sci. 64:1031−1037. Jones, R. W., R. A. Easter, F. K. McKeith, R. H. Dalrymple, H. M. Maddock, and P. J. Bechtel. 1985. Effect of the b-adrenergic agonist cimaterol (CL 263,780) on the growth and carcass characteristics of finishing swine. J. Anim. Sci. 61:905−913. Kobilka, B. K., H. Matsui, T. S. Kobilka, T. L. Yang-Feng, U. Franke, M. G. Caron, R. J. Lefkowitz, and J. W. Regan. 1987. Cloning, sequencing, and expression of the gene coding for the human platelet a2-adrenergic receptor. Science (Wash., DC) 239:650−656. Kraft, L. A., P. K. Baker, C. A. Ricks, V. A. Lance, W. A. Murphy, and D. H. Coy. 1985. Stimulation of growth hormone release in anesthetized and conscious pigs by synthetic human pancreatic growth hormone-releasing factor [hpGHF(1-29)-NH2]. Domest. Anim. Endocrinol. 2:133−139. Lafontan, M., M. Berlan, J. Galitzky, and J.-L. Montastruc. 1992. Alpha-2 adrenoceptors in lipolysis: a2 antagonists and lipidmobilizing strategies. Am. J. Clin. Nutr. 55:219S−227S. Lafontan, M., M. Berlan, and A. Villeneuve. 1983. Preponderance of a2- over b1-adrenergic receptor sites in human fat cells is not predictive of the lipolytic effect of physiological catecholamines. J. Lipid Res. 24:429−440. Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265−275.

1848

CLEALE ET AL.

Mauriege, P., J. Galitzky, M. Berlan, and M. LaFontan. 1987. Heterogeneous distribution of beta and alpha-2 adrenoceptor binding sites in human fat cells from various fat deposits: functional consequences. Eur. J. Clin. Invest. 17:156−165. Mersmann, H. J. 1984. Absence of a-adrenergic inhibition of lipolysis in swine adipose tissue. Comp. Biochem. Physiol. 79C: 165−170. Mersmann, H. J. 1987. Acute metabolic effects of adrenergic agents in swine. Am. J. Physiol. 252:E85−E95. Milavec-Krizman, M., and H. Wagner. 1978. Effects of the adrenergic agonists isoprenaline and noradrenaline and the a-blocking agents dihydroergotamine and phentolamine on the lipolysis in isolated fat cells of the rat, guinea pig, dog and man. Biochem. Pharmacol. 27:2305−2310. Morgan, D. J. 1990. Clinical pharmacokinetics of b-agonists. Clin. Pharmacokinet. 18:270−294. Moser, R. L., R. H. Dalrymple, S. G. Cornelius, J. E. Pettigrew, and C. E. Allen. 1986. Effect of cimaterol (CL 263,780) as a repartitioning agent in the diet for finishing pigs. J. Anim. Sci. 62: 21−26. Preston, R. L. 1968. Reduction of plasma urea-N by diethylstilbestrol in ruminants. Proc. Soc. Exp. Biol. Med. 129:250−256. Richelsen, B. 1986. Increased a2- but similar b-adrenergic receptor activities in subcutaneous gluteal adipocytes from females compared to males. Eur. J. Clin. Invest. 16:302−309.

Ricks, C. A., P. K. Baker, R. H. Dalrymple, M. E. Doscher, D. L. Ingle, and J. Pankavich. 1984. Use of clenbuterol to alter muscle and fat accretion in swine. Fed. Proc 43:857 (Abstr.). SAS. 1990. SAS Procedures Guide (Version 6). SAS Inst. Inc., Cary, NC. Spurlock, M. E., J. C. Cusumano, S. Q. Ji, D. B. Anderson, C. K. Smith, II, D. L. Hancock, and S. E. Mills. 1994. The effect of ractopamine on the b-adrenoceptor density and affinity in porcine adipose and skeletal muscle tissue. J. Anim. Sci. 72:75−80. Steel, R.G.D., and J. H. Torrie. 1980. Principles and Procedures of Statistics: A Biometrical Approach (2nd Ed.). McGraw-Hill Publishing Co., New York. Stillings, M. R., C. D. England, A. P. Welbourn, and C.F.C. Smith. 1986. Effect of methoxy substitution on the adrenergic activity of three structurally related a2-adrenoceptor antagonists. J. Med. Chem. 29:1780−1783. van de Venter, M., D. Litthauer, and W. Oelofsen. 1992. The effect of high concentrations of a2-adrenoceptor antagonists on the lipolytic responsiveness of isolated human adipocytes. Comp. Biochem. Physiol. 103C:327−332. Wray-Cahen, D., D. A. Ross, D. E. Bauman, and R. D. Boyd. 1991. Metabolic effects of porcine somatotropin: Nitrogen and energy balance and characterization of the temporal pattern of blood metabolites and hormones. J. Anim. Sci. 69:1503−1514.