Role of Insulin in the Blunted Glucose Metabolic to Epinephrine Diane M. Hargrove,
Nebojsa
Skrepnik,
Response of Septic Rats
Charles H. Lang, Gregory J. Bagby, and John J. Spitzer
Epinephrine produces smaller incremental increases in plasma glucose concentration and rate of glucose appearance (Ra) in septic rats compared with nonseptic animals. In the present study, we investigated the role of insulin in the diminished response of septic rats to epinephrine-induced increases in glucose turnover. Glucose kinetics were assessed by the infusion of [6-3H]-glucose in conscious catheterized rats made septic by subcutaneous injections of live Escherichia co/i. Epinephrine was infused at 1 pg/min/kg for 2 hours in the presence and absence of somatostatin and mannoheptulose (SRIF + MH). In comparison to nonseptic control animals, epinephrine-induced increases in plasma glucose concentration and glucose Ra were blunted by more than 60% in the septic rats. Infusion of SRIF + MH with epinephrine restored the blunted response to normal. During the infusion of epinephrine alone, the plasma insulin concentration in the septic rats was 2.6-fold higher than the nonseptic controls. SRIF + MH lowered the plasma insulin concentrations in both the nonseptic and septic rats to less than 10 pU/mL. SRIF + MH reversed the sepsis-induced hyperglucagonemia, but did not prevent a slight increase in glucagon levels during the epinephrine infusion in the nonseptic rats. In a second study, septic rats infused with SRIF + MH and replacement insulin showed a smaller increase in glucose concentration and glucose production in response to epinephrine than did septic animals administered SRIF + MH and no insulin. These results indicate that insulin plays an important role in the diminished response of septic rats to epinephrine. 0 1990 by W.B. Saunders Company.
I
N VIVO AND IN VITRO studies demonstrate that catecholamines produce profound changes in carbohydrate metabolism.’ The infusion of epinephrine produces hyperglycemia as a result of a transient increase in hepatic glucose production and a more sustained decrease in the metabolic clearance rate of glucose.2*3 The elevated rate of glucose appearance (Ra) is produced by an early increase in hepatic glycogenolysis and a later enhancement of gluconeogenesis.4 This increase in de novo glucose synthesis results from both an increase in gluconeogenic precursor availability and intrahepatic conversion of gluconeogenic precursors to glucose.4.’ In addition to these direct effects, epinephrine also modulates the release of other glucoregulatory hormones that can influence glucose flux. Epinephrine has been shown to stimulate glucagon release3,6,7 and inhibit insulin secretion.’ The contribution of these changes to the effects of epinephrine on glucose metabolism has not been clearly established. A hyperglycemic response to the infusion of epinephrine has been observed both in the absence of an epinephrine-induced increase in plasma glucagon concentration and when glucagon is maintained at basal levels by the infusion of somatostatin (SRIF) and replacement glucagon.3~5,7~9.‘0However, other investigators have found that infusion of SRIF effectively diminishes epinephrine-induced hyperglycemia.“~‘* The blunting of insulin secretion during hyperglycemia produced by epinephrine may also occur. This is likely to play a role in
From the Department of Physiology, Louisiana State University Medical Center, New Orleans, LA. Supported by National Institutes of Health Grant No. GM 32654. D.M.H. was a postdoctoralfellow supported by HL 07098. Dr Hargrove’s current address is Pjzer Central Research, Groton, CT. Address reprint requests to John J. Spitzer, MD, Department of Physiology, LSU Medical Center, 1901 Perdido St, New Orleans, LA 70112. @ 1990 by W.B. Saunders Company. 00260495/90/391 I-001 2$03.00/O
1180
augmenting epinephrine-induced increases in hepatic glucose production and thereby plasma glucose concentration, since insulin is a potent suppressor of hepatic glucose oUtpUt.8,9.‘3 In a previous report we showed that, in comparison to control animals, septic rats are less responsive to epinephrineinduced increases in plasma glucose concentration and glucose Ra over a wide range of epinephrine infusion rates (0.05 to 1 pg/min/kg). I4 The mechanism responsible for this suppressed response in septic rats is not known; however, a differential response of circulating glucoregulatory hormones to the epinephrine infusion was noted in our previous study and may be important. At the highest epinephrine infusion rate (1 pg/min/kg), nonseptic rats exhibited no change in plasma insulin, while glucagon levels increased slightly. In contrast, the infusion of epinephrine in septic animals resulted in increased plasma insulin concentrations and no change in glucagon levels from basal values. Because small increases in insulin are capable of inhibiting hepatic gluconeogenesis and increases in glucagon can stimulate glucose production, both alone and synergistically with epinephrine,* the divergent response of these glucoregulatory hormones between the two groups might explain the observed differences in response to epinephrine. In order to better understand the metabolic responses to epinephrine in sepsis, we infused SRIF and mannoheptulose (MH) to inhibit glucagon and insulin release during the infusion of epinephrine. This allowed us to assess the importance of these glucoregulatory hormones in the diminished epinephrine-induced carbohydrate response of the septic rats. MATERIALS AND METHODS Animal
Preparation
and Induction
of Sepsis
Experiments were performed on male Sprague-Dawley rats (250 to 300 g, Charles River, Wilmington, MA). On the day before the experiment, animals were anesthetized with an intramuscular injection of ketamine and xylazine (90 and 9 mg/kg, respectively), and vascular catheters placed in the jugular vein and carotid artery using aseptic surgical techniques. An additional catheter was implanted
Metabolism, Vol39, No 11 (November), 1990: pp 1180-l
185
1181
METABOLIC RESPONSE TO EPINEPHRINE DURING SEPSIS
subcutaneously on the dorsal surface of the animal for the subsequent administration of live bacteria.14.1s Gram-negative hypermetabolic sepsis was induced by multiple injections of live Escherichiu coli (5 to 12 x 10” organisms per injection, El 1775, American Type Culture Collection, Rockville, MD), as previously described.14,15The injections were administered at 12 noon and 5 PM on the day of surgery and 8 AM the following day. All experiments were started within 1 hour of the final injection of E coli. Experimental Protocol Glucose kinetics were determined by the primed-constant intravenous infusion of [6-3H]-glucose (5 pCi bolus injection followed by infusion at 10 rCi/h). The tracer was infused throughout the experiment. Blood samples (0.35 mL) for the determination of basal glucose kinetics were taken from the carotid artery catheter at 120 and 140 minutes of tracer infusion. After the 140-minute basal sample, a primed infusion of MH (100 mg bolus + 100 mg/h; Sigma, St Louis, MO) and SRIF (150 rg/kg bolus + 2.5 pg/min/ kg; Bachem, Torrance, CA) was initiated to inhibit insulin and glucagon secretion.s MH was infused to inhibit basal, as well as glucose-stimulated insulin secretion, 16since preliminary experiments indicated that SRIF alone was unable to completely prevent glucosestimulated insulin secretion in the rat. Blood samples were taken after 30 and 60 minutes of the infusion of SRIF and MH. After the 60-minute sample, epinephrine (1 .O pg/min/kg) was infused for 2 hours and blood samples were taken at 30, 60, and 120 minutes. Infusion of MH and SRIF was continued throughout the epinephrine infusion period. Three time-matched infusion combinations were administered to nonseptic and septic rats. In group 1, animals were infused with saline (0 to 180 minutes) and epinephrine (60 to 180 minutes) to determine the glucose metabolic response to epinephrine alone. Animals in group 2 were infused with SRIF + MH (0 to 180 minutes) and saline (60 to 180 minutes) to assess the metabolic effects of inhibiting glucagon and insulin secretion. In group 3 rats were infused with both SRIF + MH (0 to 180 minutes) and epinephrine (60 to 180 minutes) to determine whether the metabolic response to epinephrine was altered by preventing changes in insulin and glucagon levels. Saline was infused in the first two protocols to match the volume administered to the rats that received SRIF + MH and epinephrine. Plasma samples for determination of insulin and glucagon were taken during the basal period, at 60 minutes of the SRIF + MH infusion and at the end of epinephrine infusion. Because of the limited blood volume in rats, samples for hormone determinations were taken from a separate group of animals subjected to the same experimental protocol. A second series of experiments were performed in which insulin was infused in combination with SRIF and MH, before the administration of epinephrine, into septic rats in order to prevent the reduction in plasma insulin levels. Regular purified porcine insulin (Squibb-Novo; Princeton, NJ) dissolved in 0.9% saline containing 0.25% human serum albumin (HSA) was infused at a rate of 0.8 mU/min/kg. A group of septic rats also received an equal volume of SRIF + MH dissolved in HSA (0.8 mL/h) without insulin. Blood samples were obtained as described above. Analytical Procedures Plasma glucose and lactate concentrations were determined enzymatically on neutralized supernatants of deproteinized plasma.” Plasma glucose-specific activity was determined as previously described.14 Glucose rates of appearance (Ra) and disappearance (Rd) were calculated using non-steady-state equations.” Plasma immunoreactive insulin and glucagon concentrations were deter-
mined by radioimmunoassay using porcine standards (ICN, Irvine, CA). Statistics Experimental values represent means * SE. To assess the effect of the treatments over the time course of the experiment and to eliminate differences due to different initial basal values, the area under the curve was calculated using the trapezoidal solution” by subtracting basal values for individual animals from the values obtained during the infusion of epinephrine. Data for the incremental area were subjected to one-way ANOVA. When the ANOVA indicated significant effects of treatment (P < .05), a StudentNewman Keuls multiple range test was performed to determine which means were statistically different. RESULTS
The basal metabolic values for nonseptic and septic rats are presented in Table 1. Septic rats had an increased basal plasma lactate concentration (104%) and glucose Ra (6 1%); the plasma glucose concentration was decreased by 9%. The plasma insulin concentration was similar between the two groups, but the glucagon levels were elevated by more than threefold in septic animals. Sepsis also increased rectal temperature by 1°C. Sepsis attenuated the glucose metabolic response to the infusion of epinephrine (Fig 1). When epinephrine was administered alone, the maximum increases in the plasma glucose concentration, glucose Ra and plasma lactate levels were 3.2-, 2.5, and 5.5fold in nonseptic animals, and 2.1-, OS-, and 4.6-fold, respectively, in septic rats. When the response to epinephrine was expressed as the incremental area above basal, the plasma glucose concentration and the glucose Ra areas were significantly lower in the septic rats; whereas the response of plasma lactate did not differ between the two groups (Table 2). These results are consistent with our previous findings.14 Infusion of SRIF + MH resulted in a gradual increase in the plasma glucose concentration from basal in the nonseptic (130% at 3 hours, Fig 1A) and septic rats (68% at 3 hours, Fig 1B). Glucose Ra exhibited a transient increase during the first 30 minutes of the SRIF + MH infusion in the nonseptic rats (Fig lC), but did not change in the septic group (Fig 1D). Plasma lactate concentration was unaffected by the infusion of SRIF + MH in either septic or nonseptic rats (Fig 1E and F). When epinephrine was infused in the presence of SRIF + MH, the blunted response of the septic rats to epinephrineinduced increases in plasma glucose concentration and glucose Ra was abolished (Fig 1, Table 2). In septic rats infused Table 1. Basal Metabolic
Values for Nonaeptic and Septic Rata Nonseotic
Sectic
Glucose concentration (mmol/L) Lactate concentration (mmol/L)
6.4 * 0.2 0.72 + 0.4
5.8 t 0.2, 1.47 t 0.07s
Glucose Ra (~mol/min/kg)
44.7
72.1
Insulin concentration (pU/mL) Glucagon concentration (pg/mL) Body temperature (“C)
2 1.6
31 *3 138 * 14 37.2
* 0.1
* 3.8’
26 + 3 447 ir 47* 38.2
* 0.1.
NOTE. Values are mean + SE of 22 nonseptic and 19 septic rats. lP < .05 compared with nonseptic values.
1182
HARGAOVE ETAL
SEPTIC
‘“1 B I 8 “v
20.
20. I 15.
15.
i/------A ;’
l__-:
10.
E-
L
Nebojsa
Skrepnik,
Response of Septic Rats
Charles H. Lang, Gregory J. Bagby, and John J. Spitzer
Epinephrine produces smaller incremental increases in plasma glucose concentration and rate of glucose appearance (Ra) in septic rats compared with nonseptic animals. In the present study, we investigated the role of insulin in the diminished response of septic rats to epinephrine-induced increases in glucose turnover. Glucose kinetics were assessed by the infusion of [6-3H]-glucose in conscious catheterized rats made septic by subcutaneous injections of live Escherichia co/i. Epinephrine was infused at 1 pg/min/kg for 2 hours in the presence and absence of somatostatin and mannoheptulose (SRIF + MH). In comparison to nonseptic control animals, epinephrine-induced increases in plasma glucose concentration and glucose Ra were blunted by more than 60% in the septic rats. Infusion of SRIF + MH with epinephrine restored the blunted response to normal. During the infusion of epinephrine alone, the plasma insulin concentration in the septic rats was 2.6-fold higher than the nonseptic controls. SRIF + MH lowered the plasma insulin concentrations in both the nonseptic and septic rats to less than 10 pU/mL. SRIF + MH reversed the sepsis-induced hyperglucagonemia, but did not prevent a slight increase in glucagon levels during the epinephrine infusion in the nonseptic rats. In a second study, septic rats infused with SRIF + MH and replacement insulin showed a smaller increase in glucose concentration and glucose production in response to epinephrine than did septic animals administered SRIF + MH and no insulin. These results indicate that insulin plays an important role in the diminished response of septic rats to epinephrine. 0 1990 by W.B. Saunders Company.
I
N VIVO AND IN VITRO studies demonstrate that catecholamines produce profound changes in carbohydrate metabolism.’ The infusion of epinephrine produces hyperglycemia as a result of a transient increase in hepatic glucose production and a more sustained decrease in the metabolic clearance rate of glucose.2*3 The elevated rate of glucose appearance (Ra) is produced by an early increase in hepatic glycogenolysis and a later enhancement of gluconeogenesis.4 This increase in de novo glucose synthesis results from both an increase in gluconeogenic precursor availability and intrahepatic conversion of gluconeogenic precursors to glucose.4.’ In addition to these direct effects, epinephrine also modulates the release of other glucoregulatory hormones that can influence glucose flux. Epinephrine has been shown to stimulate glucagon release3,6,7 and inhibit insulin secretion.’ The contribution of these changes to the effects of epinephrine on glucose metabolism has not been clearly established. A hyperglycemic response to the infusion of epinephrine has been observed both in the absence of an epinephrine-induced increase in plasma glucagon concentration and when glucagon is maintained at basal levels by the infusion of somatostatin (SRIF) and replacement glucagon.3~5,7~9.‘0However, other investigators have found that infusion of SRIF effectively diminishes epinephrine-induced hyperglycemia.“~‘* The blunting of insulin secretion during hyperglycemia produced by epinephrine may also occur. This is likely to play a role in
From the Department of Physiology, Louisiana State University Medical Center, New Orleans, LA. Supported by National Institutes of Health Grant No. GM 32654. D.M.H. was a postdoctoralfellow supported by HL 07098. Dr Hargrove’s current address is Pjzer Central Research, Groton, CT. Address reprint requests to John J. Spitzer, MD, Department of Physiology, LSU Medical Center, 1901 Perdido St, New Orleans, LA 70112. @ 1990 by W.B. Saunders Company. 00260495/90/391 I-001 2$03.00/O
1180
augmenting epinephrine-induced increases in hepatic glucose production and thereby plasma glucose concentration, since insulin is a potent suppressor of hepatic glucose oUtpUt.8,9.‘3 In a previous report we showed that, in comparison to control animals, septic rats are less responsive to epinephrineinduced increases in plasma glucose concentration and glucose Ra over a wide range of epinephrine infusion rates (0.05 to 1 pg/min/kg). I4 The mechanism responsible for this suppressed response in septic rats is not known; however, a differential response of circulating glucoregulatory hormones to the epinephrine infusion was noted in our previous study and may be important. At the highest epinephrine infusion rate (1 pg/min/kg), nonseptic rats exhibited no change in plasma insulin, while glucagon levels increased slightly. In contrast, the infusion of epinephrine in septic animals resulted in increased plasma insulin concentrations and no change in glucagon levels from basal values. Because small increases in insulin are capable of inhibiting hepatic gluconeogenesis and increases in glucagon can stimulate glucose production, both alone and synergistically with epinephrine,* the divergent response of these glucoregulatory hormones between the two groups might explain the observed differences in response to epinephrine. In order to better understand the metabolic responses to epinephrine in sepsis, we infused SRIF and mannoheptulose (MH) to inhibit glucagon and insulin release during the infusion of epinephrine. This allowed us to assess the importance of these glucoregulatory hormones in the diminished epinephrine-induced carbohydrate response of the septic rats. MATERIALS AND METHODS Animal
Preparation
and Induction
of Sepsis
Experiments were performed on male Sprague-Dawley rats (250 to 300 g, Charles River, Wilmington, MA). On the day before the experiment, animals were anesthetized with an intramuscular injection of ketamine and xylazine (90 and 9 mg/kg, respectively), and vascular catheters placed in the jugular vein and carotid artery using aseptic surgical techniques. An additional catheter was implanted
Metabolism, Vol39, No 11 (November), 1990: pp 1180-l
185
1181
METABOLIC RESPONSE TO EPINEPHRINE DURING SEPSIS
subcutaneously on the dorsal surface of the animal for the subsequent administration of live bacteria.14.1s Gram-negative hypermetabolic sepsis was induced by multiple injections of live Escherichiu coli (5 to 12 x 10” organisms per injection, El 1775, American Type Culture Collection, Rockville, MD), as previously described.14,15The injections were administered at 12 noon and 5 PM on the day of surgery and 8 AM the following day. All experiments were started within 1 hour of the final injection of E coli. Experimental Protocol Glucose kinetics were determined by the primed-constant intravenous infusion of [6-3H]-glucose (5 pCi bolus injection followed by infusion at 10 rCi/h). The tracer was infused throughout the experiment. Blood samples (0.35 mL) for the determination of basal glucose kinetics were taken from the carotid artery catheter at 120 and 140 minutes of tracer infusion. After the 140-minute basal sample, a primed infusion of MH (100 mg bolus + 100 mg/h; Sigma, St Louis, MO) and SRIF (150 rg/kg bolus + 2.5 pg/min/ kg; Bachem, Torrance, CA) was initiated to inhibit insulin and glucagon secretion.s MH was infused to inhibit basal, as well as glucose-stimulated insulin secretion, 16since preliminary experiments indicated that SRIF alone was unable to completely prevent glucosestimulated insulin secretion in the rat. Blood samples were taken after 30 and 60 minutes of the infusion of SRIF and MH. After the 60-minute sample, epinephrine (1 .O pg/min/kg) was infused for 2 hours and blood samples were taken at 30, 60, and 120 minutes. Infusion of MH and SRIF was continued throughout the epinephrine infusion period. Three time-matched infusion combinations were administered to nonseptic and septic rats. In group 1, animals were infused with saline (0 to 180 minutes) and epinephrine (60 to 180 minutes) to determine the glucose metabolic response to epinephrine alone. Animals in group 2 were infused with SRIF + MH (0 to 180 minutes) and saline (60 to 180 minutes) to assess the metabolic effects of inhibiting glucagon and insulin secretion. In group 3 rats were infused with both SRIF + MH (0 to 180 minutes) and epinephrine (60 to 180 minutes) to determine whether the metabolic response to epinephrine was altered by preventing changes in insulin and glucagon levels. Saline was infused in the first two protocols to match the volume administered to the rats that received SRIF + MH and epinephrine. Plasma samples for determination of insulin and glucagon were taken during the basal period, at 60 minutes of the SRIF + MH infusion and at the end of epinephrine infusion. Because of the limited blood volume in rats, samples for hormone determinations were taken from a separate group of animals subjected to the same experimental protocol. A second series of experiments were performed in which insulin was infused in combination with SRIF and MH, before the administration of epinephrine, into septic rats in order to prevent the reduction in plasma insulin levels. Regular purified porcine insulin (Squibb-Novo; Princeton, NJ) dissolved in 0.9% saline containing 0.25% human serum albumin (HSA) was infused at a rate of 0.8 mU/min/kg. A group of septic rats also received an equal volume of SRIF + MH dissolved in HSA (0.8 mL/h) without insulin. Blood samples were obtained as described above. Analytical Procedures Plasma glucose and lactate concentrations were determined enzymatically on neutralized supernatants of deproteinized plasma.” Plasma glucose-specific activity was determined as previously described.14 Glucose rates of appearance (Ra) and disappearance (Rd) were calculated using non-steady-state equations.” Plasma immunoreactive insulin and glucagon concentrations were deter-
mined by radioimmunoassay using porcine standards (ICN, Irvine, CA). Statistics Experimental values represent means * SE. To assess the effect of the treatments over the time course of the experiment and to eliminate differences due to different initial basal values, the area under the curve was calculated using the trapezoidal solution” by subtracting basal values for individual animals from the values obtained during the infusion of epinephrine. Data for the incremental area were subjected to one-way ANOVA. When the ANOVA indicated significant effects of treatment (P < .05), a StudentNewman Keuls multiple range test was performed to determine which means were statistically different. RESULTS
The basal metabolic values for nonseptic and septic rats are presented in Table 1. Septic rats had an increased basal plasma lactate concentration (104%) and glucose Ra (6 1%); the plasma glucose concentration was decreased by 9%. The plasma insulin concentration was similar between the two groups, but the glucagon levels were elevated by more than threefold in septic animals. Sepsis also increased rectal temperature by 1°C. Sepsis attenuated the glucose metabolic response to the infusion of epinephrine (Fig 1). When epinephrine was administered alone, the maximum increases in the plasma glucose concentration, glucose Ra and plasma lactate levels were 3.2-, 2.5, and 5.5fold in nonseptic animals, and 2.1-, OS-, and 4.6-fold, respectively, in septic rats. When the response to epinephrine was expressed as the incremental area above basal, the plasma glucose concentration and the glucose Ra areas were significantly lower in the septic rats; whereas the response of plasma lactate did not differ between the two groups (Table 2). These results are consistent with our previous findings.14 Infusion of SRIF + MH resulted in a gradual increase in the plasma glucose concentration from basal in the nonseptic (130% at 3 hours, Fig 1A) and septic rats (68% at 3 hours, Fig 1B). Glucose Ra exhibited a transient increase during the first 30 minutes of the SRIF + MH infusion in the nonseptic rats (Fig lC), but did not change in the septic group (Fig 1D). Plasma lactate concentration was unaffected by the infusion of SRIF + MH in either septic or nonseptic rats (Fig 1E and F). When epinephrine was infused in the presence of SRIF + MH, the blunted response of the septic rats to epinephrineinduced increases in plasma glucose concentration and glucose Ra was abolished (Fig 1, Table 2). In septic rats infused Table 1. Basal Metabolic
Values for Nonaeptic and Septic Rata Nonseotic
Sectic
Glucose concentration (mmol/L) Lactate concentration (mmol/L)
6.4 * 0.2 0.72 + 0.4
5.8 t 0.2, 1.47 t 0.07s
Glucose Ra (~mol/min/kg)
44.7
72.1
Insulin concentration (pU/mL) Glucagon concentration (pg/mL) Body temperature (“C)
2 1.6
31 *3 138 * 14 37.2
* 0.1
* 3.8’
26 + 3 447 ir 47* 38.2
* 0.1.
NOTE. Values are mean + SE of 22 nonseptic and 19 septic rats. lP < .05 compared with nonseptic values.
1182
HARGAOVE ETAL
SEPTIC
‘“1 B I 8 “v
20.
20. I 15.
15.
i/------A ;’
l__-:
10.
E-
L
