INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

INCREASED ADRENERGIC TONE resulting from both sympathetic nervous system and adrenomedullary activation is an important part of the physiological stress response. Adrenergic activation, along with a surge of counterregulatory hormones, accounts for the increased glucose production (R_a) in stress (17, 23, 28, 50). Glucose utilization (R_d), however, may increase or decrease depending on the form of stress, level of insulin, and physical activity (50, 52). An acute model of moderate stress induced by the intracerebroventricular (ICV) injection of carbachol, an acetylcholine analog, has been used to study the impact of centrally induced stress on glucose homeostasis (10, 21). We observed a marked increase in tracer-determined R_a in this stress model in normal dogs (31). In addition, however, R_d and metabolic clearance rate (MCR) increased markedly and rapidly (31). The driving mechanism for the increase in MCR could not be determined because it took place in the presence of large surges in counterregulatory hormones.

To unravel the mechanism(s) of the intriguing increase in R_d, we wished to identify the roles of insulin and adrenergic activation. Although the stress model was associated only with a small increase in plasma insulin, it appears that even a small increment in insulin may be important in the control of R_d (29). Limited sensitivity of the insulin immunoassay could also underestimate the contribution of small changes in circulating insulin levels to glucose disposal. Furthermore, it cannot be excluded that some of the stress conditions, as in the case of exercise, may have a synergistic effect with insulin, thus increasing the effect of even small changes in plasma insulin. To identify the role of insulin in this stress model, we used a pancreatic bihormonal clamp that maintains stable insulin and glucagon levels by suppressing endogenous insulin and glucagon secretion with somatostatin and replacing the two hormones with exogenous infusions. It is possible that stress may activate an otherwise dormant neuroendocrine pathway in modulating glucose uptake. The notion that the increased glucose clearance could be linked to neurally mediated autonomic outflow was prompted by our observation that ICV injection of a somatostatin analog, ODT8-SS, before carbachol administration abolished both the stress-induced increment in R_d, and that in epinephrine and norepinephrine (30). The role of the catecholamines in regulation of glucose metabolism appears to be multifaceted, depending on the metabolic milieu. Simulation of the adrenergic stress response with epinephrine infusion resulted in carbohydrate intolerance with inhibition of R_d and MCR (16, 36). However, catecholamines have been shown to increase both basal and insulin-mediated glucose uptake, chronically in the basal state in vivo (26) and acutely in the in vitro rat brown adipocyte (27) and acutely stimulated energy expenditure (6). Efferent sympathetic neural activity is believed to mediate an increase in glucose uptake in brown adipose tissue induced by electrical stimulation of the ventromedial hypothalamus (49).

We also wanted to determine contributions of glucagon and catecholamines to the rapid R_a response to centrally induced stress. During moderate exercise, R_a is governed primarily by the glucagon-to-insulin (glucagon/insulin) ratio (41, 53). In strenuous exercise, however, catecholamines rather than the glucagon/insulin ratio (42) become the main regulators of R_a. In this centrally induced model of stress, it is not known whether the increased R_a is driven by glucagon and/or catecholamines. To explore the overall role of the neural and adrenal catecholamines and to differentiate between the - and -adrenoceptor-mediated catecholamine effects in modulating R_a and R_d, we employed combined and selective - and -blockade in this study. - and -Adrenoceptors may have differential or even opposing effects on MCR. In addition, both - and -adrenergic effects may stimulate R_a through different second messengers. Selective inhibition of one adrenoceptor type with consequential changes in R_a would allow us to determine contributions of each adrenoceptor subtype.

The purposes of the present study were threefold. They were to determine in this stress model 1) the reason and extent to which the increment in R_d and MCR is dependent on rises in insulin; 2) the effects of adrenergic activation on the non-insulin-dependent portion of the stress-induced increase in MCR; and 3) the relative roles of glucagon and catecholamines in the increase in R_a. To achieve these purposes, it was imperative to use the bihormonal pancreatic clamp (PC), to minimize changes in insulin and glucagon secretion due to variations of catecholamine and - and -blockers.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Experimental animals. In normal male mongrel dogs (body wt, 15-25 kg), under general anesthesia (fluothane and nitrous oxide) and assisted ventilation, an ICV stainless 22-gauge guide cannula was implanted into the third ventricle of the brain by using a canine stereotaxic apparatus. The patency of the cerebral cannula was verified by aspiration of a small volume of cerebrospinal fluid. For insulin and glucagon infusion, the portal vein was cannulated with a Silastic catheter (1.0 mm ID), which was inserted via a branch of the splenic vein. A Silastic catheter (1.0 mm ID) was placed into the aortic arch via a carotid artery for arterial blood sampling. Three Silastic catheters (0.76 mm ID) were inserted via an external jugular vein into the superior vena cava just above the right atrium for infusions. All catheters were tunneled subcutaneously and exteriorized from the back of the neck. The catheters were filled with a heparin solution (1,000 U/ml) and flushed weekly to maintain patency. The dogs were housed under controlled-temperature and -light conditions and fed once per day a mixed diet of 300-400 g of dry dog chow (Lab canine diet 5006; Purina Mills, St. Louis, MO) and 400-500 g of meat (Romar Pet Supply, Toronto, ON). Food intake, body weight, temperature, and hematocrit were monitored to ensure that only healthy dogs were used in the experiments. All procedures were performed in accordance with Canadian Council on Animal Care standards and were approved by the Animal Care Committee of the University of Toronto.

Experimental protocol. The dogs were allowed at least 2 wk to recover from surgery before the first experiment. A minimum period of 1 wk was allowed between successive experiments. Food was withdrawn 18 h before each experiment. Each experiment consisted of a tracer equilibration period (160 to 40 min), a basal sampling period (40 to 0 min), and a stress period (0-120 min). A priming (25-µCi) dose of [3-³H]glucose (New England Nuclear, Boston, MA) was given at 160 min, when a constant infusion of [3-³H]glucose (0.25 µCi/min) was initiated and continued throughout the study. Five groups were included in the present study. 1) A previously reported carbachol control group (31) was supplemented with three additional experiments (Con, n = 10). No differences were observed between present and former control subjects in values of hormones and metabolites [insulin, glucagon, norepinephrine, epinephrine, cortisol, free fatty acids (FFA), and glycerol], and glucose turnover (R_a, R_d) under both basal and stress conditions. 2) A group received a PC (n = 7) by means of a peripheral somatostatin infusion at a dose established to inhibit release of endogenous insulin and glucagon (11). Constant basal levels of insulin and glucagon were sustained by intraportal replacement infusions. 3) A group received a PC and combined adrenergic blockade (+; n = 5) through infusions of an -blocker (phentolamine) and a -blocker (propranolol). 4) A group received an -blocker (phentolamine; 5 µg · kg¹ · min¹, Ciba-Geigy Canada, Mississauga, ON) infusion superimposed on a PC (; n = 6). 5) Another group received a -blocker (propranolol, 7 µg · kg¹ · min¹, Ciba-Geigy Canada) infusion and a PC (; n = 5). Somatostatin (0.8 µg · kg¹ · min¹) infusion (in PC, +, , and ) was started at 160 min. Simultaneous intraportal replacement infusions of insulin (1,200-2,100 pmol · kg¹ · min¹) and glucagon (at a fixed rate of 0.65 ng · kg¹ · min¹) were then begun. The rate of insulin infusion was adjusted so that plasma glucose levels monitored every 5-10 min were maintained at preequilibration euglycemia. The last adjustment of the insulin infusion was at least 30 min before the start of the basal sampling period. - and -Blockers were given at doses shown previously to attain sufficient - or -blockade (20, 22, 57). They were initiated at 20 min and continued throughout the experiment. At time 0, 5 µg of carbachol (carbamylcholine; Aldrich) in 50 µl of sterile water were injected ICV through a 24-gauge injection cannula connected to a 100-µl Hamilton microsyringe via polyethylene tubing (Intramedic PE 50; Clay Adams). Arterial blood pressure and heart rate were monitored by using a pressure transducer connected to the carotid catheter and recorded with a physiograph (Gilson).

Processing of blood samples. Blood samples were taken every 5-10 min for plasma glucose levels and every 10-20 min for determination of insulin, glucagon, cortisol, norepinephrine, epinephrine, lactate, glycerol, and FFA. Blood samples for determination of plasma glucose, glucose specific activity, insulin, cortisol, lactate, and glycerol were collected in tubes containing heparin (50 U/ml) and NaF (1-2 mg/tube). Samples for determination of plasma glucagon and FFA concentrations were collected in tubes containing 1:1 (vol/vol) aprotinin (Trasylol, 10,000 kIU/ml, Bayer) and EDTA (25 mg/ml, BDH Chemicals). For assays of norepinephrine and epinephrine, 1-ml blood samples were collected in poly-ethylene tubes containing 2.5 mg glutathione (Boehringer Mannheim) and 10 µl EGTA (Sigma Chemical). All blood samples were stored at 4°C and centrifuged within 60 min. The resultant plasma was stored at 20°C, except the plasma for catecholamine assays, which was deproteinized with 2 M HClO₄ and stored at 70°C. Plasma glucose was measured by a glucose oxidase method by using a Glucose Analyzer II (Beckman Instruments, Fullerton, CA). Plasma glucose specific activity was derived from plasma glucose concentration and [3-³H]glucose radioactivity determined in a liquid scintillation counter after deproteinization of the plasma samples with Ba(OH)₂ and ZnSO₄ and evaporation of the supernatant (15). Insulin and glucagon were analyzed by radioimmunoassays. Catecholamines were measured by a radioenzymatic assay (45) and cortisol by a modified competitive binding assay. Plasma FFA were determined by a radiochemical technique (18). Lactate and glycerol were measured by enzymatic microfluorometric methods (24).

Calculations. Glucose concentration and specific activity data were systematically smoothed by using the optimal segments technique (13). R_a and R_d were calculated with an equation for non-steady-state turnover (46). Glucose MCR was calculated by dividing R_d by the prevailing plasma glucose concentration. MCR represents an estimate of R_d partially corrected for the mass action effect of glucose (33).

Statistical analyses. All values are expressed as means ± SE. Statistical analysis, to assess the stress-induced responses and differences both within and between the experimental groups, was performed by using ANOVA for unbalanced data (Statistical Analysis System; SAS Institute, Carey, NC), with the time effects of treatments involved in the experiments taken into account.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Insulin and glucagon levels. Basal insulin levels were slightly higher (P < 0.05) in the Con (75.6 ± 5.0 pM) and -blockade (79 ± 11.5 pM) vs. the PC (57.0 ± 3.6 pM) and -blockade (59 ± 8.4 pM) groups. Also, basal glucagon levels were greater in + than in the Con and  groups (186 ± 24 vs. 144 ± 9 and 134 ± 14 ng/l, respectively, P < 0.05). Because of such basal differences among groups,  values, representing changes from the basal period, were calculated (Fig. 1). A small rise in insulin levels seen in the Con animals during stress was obviated with the bihormonal PC (P < 0.05, PC vs. Con). Arterial plasma insulin levels were stable in PC (basal vs. stress period: 57 ± 3.6 vs. 58 ± 4.0 pM),  (79.2 ± 11.4 vs. 82.8 ± 8.4 pM), and + (69.6 ± 5.4 vs. 68.4 ± 4.5 pM) groups, although the  group was associated with a small but significant increase in insulin levels during stress (59 ± 8.4 vs. 73 ± 6.6 pM, P < 0.05). The increase in plasma glucagon due to stress observed in the Con group (144 ± 9 to 192 ± 19 ng/l, P < 0.01) was obviated by the PC in the PC (171 ± 16 to 172 ± 15 ng/l), + (186 ± 24 to 164 ± 13 ng/l),  (134 ± 14 to 154 ± 20 ng/l), and  (150 ± 12 to 146 ± 9 ng/l) groups (all P < 0.01 vs. Con).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effects of intracerebroventricular (ICV) carbachol injection (5 µg at time 0)-induced stress on plasma insulin (top) and glucagon increments (bottom) with respect to basal period. Experimental protocol consisted of a basal (before ICV carbachol, 40 to 0 min) and stress period (after ICV carbachol, 0-120 min). Values are means ± SE. Experiments were conducted in 5 groups of dogs: 1) control (Con; ; n = 10); 2) pancreatic clamp (PC; ; n = 7); 3) PC plus combined adrenergic blockade (+; ; n = 5) with infusions of -blocker phentolamine (7 µg · kg1 · min1) and -blocker propranolol (5 µg · kg1 · min1); 4) PC and -blockade (; ; n = 6) with infusion of -blocker phentolamine (7 µg · kg1 · min1); and 5) PC plus -blockade (; ; n = 5) with infusion of -blocker propranolol (5 µg · kg1 · min1).

Basal glucose turnover (Table 1). Basal R_a in the + and  groups was greater (P < 0.05) than in the Con, PC, and  groups. Similarly, basal R_d and MCR in the + and  groups were greater (P < 0.05) than in Con, PC, and , respectively.

View this table: [in this window] [in a new window]   Table 1.   Basal period (40 to 0 min) glucose turnover data (plasma glucose, glucose production, glucose uptake, and metabolic clearance rate of glucose) before ICV carbachol injection (time 0)

Glucose turnover during stress. During stress, plasma glucose levels increased slightly yet significantly in the Con, PC, and  groups (Fig. 2). However, in the + and  groups there was a small but significant fall. In all groups, maximal changes in plasma glucose during stress never exceeded 11% of the respective basal values. On induction of stress with ICV carbachol injection, R_a rapidly reached peak levels in the Con (170 ± 45 µmol · kg¹ · min¹), PC (95.5 ± 10 µmol · kg¹ · min¹), and  (111 ± 24 µmol · kg¹ · min¹) groups within 15 min (Fig. 3A). However, the peak increment in R_a (56 ± 14 µmol · kg¹ · min¹) in the + group was substantively lower (P < 0.05) than those in the Con, PC, and  groups (Fig. 3A). In the  group, the peak of R_a was delayed and rose to 101 ± 31 µmol · kg¹ · min¹ at 50 min of stress. Poststress R_a gradually reverted to respective baseline values in all groups except for +, in which the decline of R_a was much slower. R_d peaked within the first 20 min of stress in all groups except +, in which the maximum increment in R_d was reached at almost 40 min of stress (Fig. 3). The marked peak increment in R_d seen in Con (109 ± 27 µmol · kg¹ · min¹) was significantly attenuated in the PC (peak: 64 ± 15 µmol · kg¹ · min¹),  (peak: 63 ± 15 µmol · kg¹ · min¹), and + (peak: 56 ± 39 µmol · kg¹ · min¹) groups. The stress-induced peak R_d reverted to respective baseline rates at 120 min in all groups except , which displayed the greatest increment in R_d (peak: 129 ± 33 µmol · kg¹ · min¹) and the slowest decline in poststress R_d of all the other groups (P < 0.01, Fig. 3). When both insulin and glucagon levels were clamped, the peak increments in MCR in the PC (1.03 ± 0.35 ml · kg¹ · min¹) and  (1.00 ± 0.23 ml · kg¹ · min¹) groups were 30% lower (P < 0.05) than that observed in the Con group (1.52 ± 0.18 ml · kg¹ · min¹). In comparison with the PC group, in the + group the peak increment in MCR was significantly (P < 0.01) augmented (1.7-fold). The increment in MCR during stress with  peaked at 2.76 ± 0.72 ml · kg¹ · min¹, which was much greater than in all other groups.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Plasma glucose concentrations before and after ICV carbachol-induced stress in Con, PC, , , and + groups. Experimental design and symbols are as outlined in Fig. 1.

View larger version (37K): [in this window] [in a new window]   Fig. 3.   A: effects of ICV carbachol-induced stress on increments in glucose production (Ra; top), utilization (Rd; middle), and metabolic clearance rate (MCR; bottom) in Con, PC, and + groups. , Change. Experimental design and legend are as in Fig. 1. B: effects of ICV carbachol-induced stress on Ra, Rd, and MCR increments in , , and + groups. Experimental design and symbols are as outlined in Fig. 1.

Hormones and metabolites. On ICV administration of carbachol, arterial epinephrine and norepinephrine levels increased (P < 0.05) in all groups (Table 2). The rise in epinephrine was greater (P < 0.01) when -adrenoceptor was blocked with  +  (4-fold) and  (4.4-fold) compared with other groups (1.7-fold in PC, 2.9-fold in , and 2.8 fold in Con). Norepinephrine levels rose (P < 0.05) during the entire stress period from basal values in the Con (1.6-fold), PC (1.2-fold), and  (1.7-fold) groups, whereas the largest increment was obtained when -blockade was applied (1.9-fold in  and 2.8 fold in +). ICV carbachol resulted in rapid two- to threefold elevations (P < 0.01) in arterial plasma ACTH, an index of central mediation of the stress response, in all groups (Table 1). The stress response was also characterized by significant (P < 0.05) cortisol increments in the  (4.4-fold),  (5.7-fold), PC (1.5 fold), and Con (4-fold) groups (Table 1). The increment in the + group (66 ± 17 to 84 ± 10 nM) did not reach significance. The induction of stress was associated with increments of FFA in the Con (898 ± 70 to 1,178 ± 65 µeq/l, P < 0.05) and  (722 ± 91 to 1,064 ± 74 µeq/l, P < 0.05) groups (Fig. 4). The increment in FFA was blunted in the + (580 ± 47 to 681 ± 63 µeq/l) and PC (654 ± 44 to 746 ± 49 µeq/l) groups. During stress, there was a small rise in glycerol (PC, 88 ± 5 to 128 ± 7 µM P < 0.05) (Fig. 5). With , the glycerol profile, parallel to that of FFA, increased from 87 ± 8 to 168 ± 12 µM (P < 0.05). However, -blockade resulted in a fall (656 ± 64 to 440 ± 29 µeq/l, P < 0.05) in FFA levels and prevented a stress-induced rise in glycerol levels (91 ± 8 to 84 ± 4 µM). The observed increment of arterial glycerol in the Con group during stress (100 ± 6 to 187 ± 13.3 µM, P < 0.05) was reduced (P < 0.05) in the PC (88 ± 4.5 to 129 ± 7 µM, P < 0.05) and + (99 ± 16 to 125 ± 13 µM, P < 0.05) groups. Dramatic rises (2- to 3-fold, P < 0.01) in arterial lactate levels were observed during stress (0-120 min) in the PC, , and Con groups (Fig. 6). However, most strikingly,  and  +  completely prevented any stress-induced increment in lactate levels, suggesting a reduction in glycogenolysis due to inhibition of -adrenergic activities.

View this table: [in this window] [in a new window]   Table 2.   Effects of ICV 5 µg of carbachol injected at time 0 on plasma concentrations of epinephrine, norepinephrine, ACTH and cortisol in control, PC-, PC + -blockade, PC + -blockade, and PC +  + -infused normal dogs

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Effects of ICV carbachol injection on free fatty acid (FFA) increments in Con, PC, , , and + groups. Experimental design and symbols are as outlined in Fig. 1.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Effects of ICV carbachol injection on plasma concentration of plasma glycerol levels in Con, PC, , , and + groups. Experimental design and symbols are as outlined in Fig. 1.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Plasma concentration of lactate in control, PC, , , and + groups before and after ICV injection of carbachol. Experimental design and symbols are as outlined in Fig. 1.

Hemodynamic data. -Blockade increased (P < 0.05) basal systolic, diastolic, and mean arterial pressure above those in the other three groups (Table 3). Stress induced increases (P < 0.01) in systolic (152 ± 10 to 176 ± 5 mmHg), diastolic (126 ± 13 to 151 ± 5 mmHg), and mean arterial pressure (135 ± 12 to 159 ± 5 mmHg) in the + group. During stress, systolic, diastolic, and mean arterial pressure were greater (P < 0.05) in + and  groups than in the other two groups. -Blockade during stress resulted in an increased heart rate (110 ± 11 to 193 ± 12 beats/min, P < 0.001).

View this table: [in this window] [in a new window]   Table 3.   Effects of 5 µg of ICV carbachol injected at time 0 on hemodynamic parameters in PC-, PC + -blockade, PC + -blockade-, and PC +  + -infused normal dogs

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cholinergic mechanisms originating in the circumventricular neurons play a physiological role in the central regulation of glucose metabolism (21). ICV administration of carbachol, an acetylcholine analog, induces changes in neuroendocrine activity with increased release of counterregulatory hormones similarly to general stress (9, 30, 31). We hypothesize the involvement of a putative efferent pathway activated by the circumventricular hypothalamic neurons during cholinergic stimulation. Our present study aims at delineating the roles of insulin, glucagon, the catecholamines, and their receptors in the control of R_d, MCR, and R_a during centrally induced stress. Owing to baseline variations associated with all five experimental approaches,  values, reflecting net changes from baseline values during stress, were used for statistical analyses and comparisons. The baseline variations were due to the effects of adrenergic blocker infusions started during the basal period. The elevated basal glucose production, uptake, and clearance rates with - and  + -blockade presumably reflect increased metabolic fuel reliance on glucose because of -blockade-induced inhibition of lipolysis and muscle glycogenolysis. Thus, with the use of  values, the effects of each experimental condition on glucose metabolism during stress are separated from their baseline effects. Analyses with  values allow for quantitative assessment of the stress-induced net changes in glucose turnover against the uneven basal background of each group.

The effects of pancreatic hormonal clamping on glucose turnover during stress. In a previous study, R_d markedly increased during ICV carbachol stress in the normal dog, whereas plasma insulin increased only marginally (31). Because even a small increase in insulin could have a significant impact on whole glucose uptake during exercise, another form of stress (40), the use of a PC, was necessary to identify the contribution of insulin in this form of central stress. The stress-induced increases in glucose uptake and clearance observed in the Con group were only modestly reduced when stable insulin levels were maintained by the PC, indicating that the increases were largely insulin independent.

Effects of adrenergic blockade on R_d and MCR during stress. In our previous experiments (30), ICV injection of a somatostatin analog before ICV carbachol both decreased the release of catecholamines and prevented the rise in glucose uptake. Although an increase in epinephrine directly decreases MCR in dogs and in humans in a dose-dependent fashion (16, 36), some in vitro studies indicate a dual effect of epinephrine, whereby it inhibits R_d at low doses but stimulates R_d at higher doses (5). Also, -adrenergic stimulation with isoproterenol increased glucose uptake in rat epitrochlearis muscle incubated with albumin (60). Long-term exposure to norepinephrine resulted in stimulation of both basal and insulin-stimulated glucose uptake (26) in the rat.

Effect of adrenergic blockade on R_a during stress. Our study outlined the roles of glucagon and catecholamines in control of R_a during stress. In the early phase of stress, glucagon clamp (as seen with PC) suppressed R_a by 44%, but combined glucagon clamp and double adrenergic blockade suppressed R_a by 80% compared with the Con group. The difference between the two protocols indicates the contribution of the catecholamines, presumably via glycogenolysis (47). The greater increment in circulating catecholamine levels observed in the + than in the Con, PC, and either  or  groups is due to a reduction in catecholamine binding and clearance. Plasma catecholamines are only a rough indicator of sympathoadrenergic activation (4) and during blockade cannot be associated with changes in R_a. However, even in the absence of the adrenergic blockade, the peak levels of infused epinephrine lag behind the R_a increments (47). It is surprising that there is a residual increase in R_a, despite the bihormonal clamp and - + -blockade. This may reflect activity of other stress hormones such as vasopressin, which increases in stress (30) and may stimulate R_a (12). It is also conceivable that hepatic autoregulation is triggered by the small decline in glycemia. Cortisol is not an important factor in control of rapid changes in R_a, as it is a slow-acting hormone.

Conclusions. In the ICV carbachol model of stress, the increment in MCR is largely independent of changes in insulin and is enhanced with combined adrenergic blockade. -Blockade enhances stress-induced MCR, presumably through inhibition of lipolysis and muscle glycogenolysis, to a much greater extent than with combined blockade, and it unmasks the full impact of the insulin-independent neuroendocrine stimulation of glucose uptake in vivo. The effect of -blockade in enhancing glucose utilization during stress does not appear to be dependent on its cardiovascular actions. The neuroendocrine mechanism responsible for the insulin-independent stress-induced increments in glucose uptake needs further identification. Both catecholamines and glucagon play an equal role in stimulating the early-phase increment in R_a, whereas the increment in the later phase is mostly mediated by glucagon. However, a small remaining portion of the increment in R_a is not accounted for by either hormone. Combined - and -blockade, but not - or -blockade per se, diminished increments in R_a during stress. This suggests a complementary backup mechanism of the - and -adrenergic receptors in stimulation of glucose production during stress.