Insulin Sensitivity of Isolated Perfused Rat Liver Carl E. Mondon, Ph.D., Constantine B. Dolkas, M.S., Jerrold M. Olefsky, M.D., and Gerald M. Reaven, M.D., Palo Alto and Moffett Field, California
SUMMARY The responsiveness of the isolated perfused rat liver to different metabolic effects of insulin was investigated duriwg recycling perfusion. Infusion of porcine insulin at rates of 6,9,16 and 33 mU/hr. resulted in stable perfusate insulin levels averaging 41, 72,120 and 229 ixU/ral., respectively. Since the portal vein insulin concentration in the intact rat averaged 48 ^U/ml. after a twenty-six-hour fast and 125 ^U/ml. two hours after removal of food, the studies were conducted at insulin levels within the physiological range. The effect of each insulin concentration on the net accumulation of K + , amino acid nitrogen, urea nitrogen and glucose in the perfusing medium was assessed against the net accumulation of perfusate constituents during perfusion of control livers and livers perfused with perfusate insulin levels greater than 500 ^U/ml. The results indicate that essentially maximal suppression of amino acid nitrogen outflow and retention of K+ occurred at insulin concentrations of 72 /xU/ml., with lesser effects being noted at 41 /xU/ml. Inhibition of ureogenesis was demonstrated at insulin levels above 120 /uU/ml. However, significant effects of insulin on suppressing net glucose outflow was not observed until insulin levels had reached 500 /xU/ml. due presumably to the absence of a sustained rate of glycogenolysis by control livers. The observation that perfused livers from normal rats are extremely sensitive to several metabolic effects of insulin at physiological concentrations suggests that this experimental approach can provide useful information as to the role of the liver in the pathogenesis of various insulin resistant states. DIABETES 24:225-29, February, 1975.
Considerable information exists attesting to the fact that isolated perfused livers are responsive to in vitro addition of insulin. Previous reports have demon-
From the Department of Medicine, Stanford University School of Medicine; Veterans Administration Hospital, Palo Alto, California; and NASA-AMES Research Center, Moffett Field, California. Send reprint requests to: Dr. Carl E. Mondon, Veterans Administration Hospital, 3801 Miranda Avenue,.Palo Alto, California 94304. Accepted for publication November 7, 1974. FEBRUARY, 1975
strated an effect of insulin on hepatic potassium retention, 1 " 4 inhibition of glycogenolysis, 4 ' 6 proteolysis, 78 and suppression of the glycogenolytic and gluconeogenic effect of added glucagon or cyclic AMP. 9 ' 1 1 On the other hand, relatively little information exists as to the relationship between insulin concentration and metabolic response. For example, although it has been suggested that the perfused liver responds to specific metabolic effects of insulin at concentrations ranging from 400 to less than 135 ju,U/ml., lDi11 these conclusions are based upon inferences and not measurements of insulin in the perfusing medium. Therefore, in order to define the quantitative nature of the relationship between insulin concentration and hepatic insulin sensitivity we have measured the ability of perfused livers from normal fed rats to respond to the effect of insulin on suppressing the net efflux of K + , glucose, amino acid nitrogen and urea nitrogen over a wide range of constant perfusate insulin levels. METHODS Liver perfusion. Livers from nonfasted male Sprague Dawley rats, weighing 114- 137 gm., were perfused in situ at constant flow rates ranging from 0.9 to 1. 2 ml. per gram liver per minute by previously described technics. 6 ' 7 The perfusate consisted of forty-five parts defibrinated rat blood, obtained from large male rats of the same strain, with the hematocrit adjusted to 20 per cent by addition of fifty-five parts Krebs Ringer bicarbonate (KRB) buffer containing 3gm. bovine albumin (Reheis Chemical Co., Chicago) per 100 ml. Both liver and blood donor animals were anesthetized with sodium thiamylal (62 mg. per 100 gm. body weight) prior to surgery. The perfusing medium volume averaged 33 ml. at the start of perfusion, and was recycled 225
INSULIN SENSITIVITY OF ISOLATED PERFUSED RAT LIVER
through the liver for 120 minutes. Blood samples of 1.0 ml. were drawn from the perfusate reservoir at 0, 30, 60, 90, and 120 minutes. Analytic methods. Blood samples were centrifuged, and serum was stored at -20° C. for subsequent analysis. Serum glucose was determined by electronic measurement of the rate of O2 consumption during the glucose oxidase reaction (Glucose Analyzer, Beckman Instruments, Fullerton, Calif.). Serum K + , amino acid nitrogen (AAN), and urea nitrogen were analyzed by methods'previously described, 67 and insulin concentration was determined by modification of the immunoassay procedure of Desbuquois and Aurbach. 12 Portal vein insulin concentration was assessed against rat insulin standards and perfusate serum insulin was assessed against porcine insulin standards. Microhematocrits were determined in duplicate on all blood samples for calculation of perfusate serum volume at each sampling time. Infusion of insulin. Porcine insulin, diluted in KRB buffer containing 1 per cent bovine albumin, was added continuously to the perfusate reservoir flask for 120 minutes. Although the volumes were equivalent in each experiment, the amount of insulin was varied in order to provide infusion rates corresponding to 6, 9, 16, 33 and > 42 mU insulin per hour. Control livers received a comparable volume of hormone diluent. Calculations. The net output of K + , AAN, urea nitrogen and glucose by the liver was calculated by multiplying the perfusate serum volume by the change in concentration of perfusate constituents from the zero time value. Net changes in serum volume were corrected for sampling losses, but not for infusion of insulin or hormone diluent, since the volume of hormone added (.435 ml. per hour) approximated the anticipated loss of volume due to bile secretion.6 The significance of differences between means was established by the Student / test.
with the tip of the needle lying anterior to the branch of the pyloric vein and blood was drawn into a 5 ml. syringe. These results are seen in table 1 and indicate that both plasma glucose and insulin levels were lower in fasted rats. The glucose levels were similar in the two systems, while the insulin levels were five to eight times greater in the portal vein.
TABLE l Systemic and portal vein glucose and insulin concentrations in rats fed ad libitum and rats fasted twenty-six hours
Group
Systemic Glucose Insulin (mg.%)
AiU/ml.
Portal vein Insulin Glucose AiU/ml. (mg.%)
Fed (6) Fasted (6)
153 ± 3 96 ± 7
24 ± 4 6 ± 2
156 ± 7 96 ± 7
125 ± 25 48 ± 14
Values represent mean ± S.E. of six rats in each group. Food was removed from the cages of rats fed laboratory chow ad libitum at 8 a.m. and blood samples were drawn from fed rats after two hours and from fasted rats after twenty-six hours.
Maximal effect of insulin. Infusion of insulin to six livers at rates ranging from 42 to 90 mU/hr. resulted in a progressive increase in serum insulin concentration to levels ranging from 510 to 1500 /u,U/ml. after 120 minutes' perfusion. Since the metabolic effect of insulin on suppressing the net accumulation of K + , AAN, urea nitrogen and glucose in the perfusing medium was comparable at all these doses, the results of these six studies were averaged and are illustrated in figure 1, along with the alterations in perfusate constituents of nine control livers. The data indicate that the effect of insulin on enhancing net uptake of K + by the liver began by thirty minutes and reached a maximum by sixty minutes. In contrast, the effect of insulin on suppressing the net accumulation of glucose, amino RESULTS acid nitrogen and urea nitrogen in the perfusing medium increased progressively for the duration of Systemic and portal vein serum insulin concentration in normal male rats. To determine the physiologic range of perfusion, and the maximum effect of the hormone insulin concentration in normal male rats fed laborat- occurred after 120 minutes' perfusion. ory chow ad libitum, blood samples were drawn from Effect of varying insulin concentrations on net release of the portal vein and tail vein of six rats two hours after liver K+, AAN, urea nitrogen and glucose. The perfusate removal of food, and from six rats fasted twenty-six insulin levels which resulted from infusion of insulin at hours. Rats were anesthetized with sodium thiamylal rates of 6, 9, 16 and 33 mU/hr. are listed in table 2. and systemic blood was sampled from the tip of the tail The data indicate that infusion of insulin at the above into EDTA-primed capillary tubes. Immediately rates resulted in stabilization of serum insulin concentthereafter, a 20-gauge needle filled with 5 per cent ration at levels averaging 41,72, 120 and 229 /-cU/ml. EDTA was inserted posteriorly into the portal vein between thirty and 120 minutes' perfusion. At these 226
DIABETES, VOL. 2 4 , NO. 2
CARL E. MONDON, PH.D., AND ASSOCIATES Amino Acid N 40-i
80-
I
20-
60-
E ~
0
40-
in
I -20-
20-
to o u
0-
o-
To
Glucose
T
Urea N
5 a
160n
160-
cu en « .c u a5
120-
120
80-
80
40-|
40
,/T
-
z
/
/
0
0 60
120
60
Minutes Control • • Insulin o o
120
The accumulation of serum K + after sixty minutes' perfusion and accumulation of serum AAN, urea nitrogen and glucose after 120 minutes' perfusion in each group was normalized to unit liver weight and these data are listed in table 3. These results indicate that the liver is most sensitive to the effects of insulin on K + uptake and AAN outflow under the conditions of this study. Thus, a diminished uptake of K + and suppression of AAN outflow was seen at perfusate insulin levels of AlfAJ/mi., although the effect on K4- did not attain statistical significance. Essentially, maximal effects of insulin on both K + and AAN seemed to occur at insulin concentrations of 72 /xU/ml. The ability of insulin to suppress urea nitrogen outflow was not seen until perfusate insulin reached a level of 120 jiiU/ml., and significant effects of insulin on inhibition of hepatic glucose release were only obtained at insulin levels in excess of 500 /xU/ml.
Minutes
DISCUSSION
FIG. 1. Effect of serum insulin concentrations greater than 500 /AJ/ml. on the net accumulation of serum K + , amino acid nitrogen, glucose and urea nitrogen in the perfusing medium. Vertical bars represent 1 S.E.
levels, we estimated from the rate of infusion that utilization of insulin by the liver remained constant, and averaged .45, .40, .39 and .47 juU/gm. liver per 1 /LtU/ml. insulin concentration, respectively. These values essentially agreed with results obtained in recent studies demonstrating that insulin removal by the isolated rat liver occurs at uniform rates approximating .35 /xU/gm. liver per minute per 1 ^tU/ml. insulin concentration at perfusate insulin levels ranging from 20 to 500/uU/ml. 1 3 The net accumulation of K + , AAN, urea nitrogen and glucose in the perfusing medium at the above insulin concentrations was compared with the nine control livers and six livers with perfusate insulin levels above 500 /u,U/ml. previously illustrated in figure 1.
In general, previous investigations of the action of insulin on the perfused liver have focused on whether or not insulin could exert a specific metabolic effect. In contrast, in these studies we have attempted to define the quantitative relationship between perfusate insulin concentration and the ability of insulin to modify several facets of hepatic metabolism. The results demonstrate that significant suppression of amino acid outflow occurred at perfusate insulin levels as low as 41 With elevation of perfusate insulin to 12 ., representing a 50 per cent increase of fasting portal vein insulin concentration, suppression of amino acid outflow had essentially reached maximum levels and K + uptake had plateaued at levels comparable to previously reported values averaging 5 to 6 /^moles K/g liver. 1 " 3 Larger quantitative effects of insulin on net K + movement have been reported during suppression of glycogenolysis,4"6 and the greater net uptake of K + obtained at maximal insulin concentrations (table 3)
TABLE 2 Serum insulin concentration during infusion of 6, 9, 16, 33 mU insulin per hour
Infusion rate (mU/hr.) 6 9 16 33
Liver wt. (gm •) 5.21 5.41 5.76 5.55
±: ±: ±: ±:
.29 .25 .19 .10
0' 37 ± 6 61 ± 9 110 ± 13 175 ± 27
30' 39 73 112 199
± ± ± ±
Serum insulin concentration 60' 4 8 3 9
41 ± 4 73 ± 13 113 ± 7 219 ± 10
42 75 127 237
± ± ± ±
X cone. 30'-120'
120'
90' 2 9 5 5
42 66 128 260
± ± ± ±
4 5
6 11
41 72 120 229
Values represent the mean ± S.E. of six experiments at each insulin infusion rate. Porcine insulin was added to each flask prior to start of infusion to raise the initial serum concentration to levels approximating the steady state level. FEBRUARY, 1975
227
INSULIN SENSITIVITY OF ISOLATED PERFUSED RAT LIVER
TABLE 3 Effect of serum insulin concentration on uptake of K+ and release of AAN, urea and glucose by isolated perfused liver Group: mean insulin cone. OiU/ml.) Control (9) 41 (6) 72 (6) 120 (6) 229 (6) >500 (6)
K+ uptake (/nmol/gm .) - 1 . 3 ± .6 2.6 ± 1 .8 3.9 ± .9 4.2 ± .6 5.2 ± .6 7.6 ± .7
Group: mean insulin cone. (MU/ml.)
AAN outflow (/imol/gm .)
Insulin effect (I-C)
14.4 10.6 5.7 6.3 5.8 5.4
-3.8 -8.7 -8.1 -8.6 -9.0
Control 41 72 120 229 >500
(9) (6) (6) (6) (6) (6)
± .8 ± 1. 3 ± .4 ± .5 ± ..7 ± .7
Insulin effect (I-C) 3.9 5.2 5.5 6.5 8.9
± 1.9 ± 1.1 ± .8 ± .8 ± .9
± ± ± ± ±
P
.05
SUMMARY The responsiveness of the isolated perfused rat liver to different metabolic effects of insulin was investigated duriwg recycling perfusion. Infusion of porcine insulin at rates of 6,9,16 and 33 mU/hr. resulted in stable perfusate insulin levels averaging 41, 72,120 and 229 ixU/ral., respectively. Since the portal vein insulin concentration in the intact rat averaged 48 ^U/ml. after a twenty-six-hour fast and 125 ^U/ml. two hours after removal of food, the studies were conducted at insulin levels within the physiological range. The effect of each insulin concentration on the net accumulation of K + , amino acid nitrogen, urea nitrogen and glucose in the perfusing medium was assessed against the net accumulation of perfusate constituents during perfusion of control livers and livers perfused with perfusate insulin levels greater than 500 ^U/ml. The results indicate that essentially maximal suppression of amino acid nitrogen outflow and retention of K+ occurred at insulin concentrations of 72 /xU/ml., with lesser effects being noted at 41 /xU/ml. Inhibition of ureogenesis was demonstrated at insulin levels above 120 /uU/ml. However, significant effects of insulin on suppressing net glucose outflow was not observed until insulin levels had reached 500 /xU/ml. due presumably to the absence of a sustained rate of glycogenolysis by control livers. The observation that perfused livers from normal rats are extremely sensitive to several metabolic effects of insulin at physiological concentrations suggests that this experimental approach can provide useful information as to the role of the liver in the pathogenesis of various insulin resistant states. DIABETES 24:225-29, February, 1975.
Considerable information exists attesting to the fact that isolated perfused livers are responsive to in vitro addition of insulin. Previous reports have demon-
From the Department of Medicine, Stanford University School of Medicine; Veterans Administration Hospital, Palo Alto, California; and NASA-AMES Research Center, Moffett Field, California. Send reprint requests to: Dr. Carl E. Mondon, Veterans Administration Hospital, 3801 Miranda Avenue,.Palo Alto, California 94304. Accepted for publication November 7, 1974. FEBRUARY, 1975
strated an effect of insulin on hepatic potassium retention, 1 " 4 inhibition of glycogenolysis, 4 ' 6 proteolysis, 78 and suppression of the glycogenolytic and gluconeogenic effect of added glucagon or cyclic AMP. 9 ' 1 1 On the other hand, relatively little information exists as to the relationship between insulin concentration and metabolic response. For example, although it has been suggested that the perfused liver responds to specific metabolic effects of insulin at concentrations ranging from 400 to less than 135 ju,U/ml., lDi11 these conclusions are based upon inferences and not measurements of insulin in the perfusing medium. Therefore, in order to define the quantitative nature of the relationship between insulin concentration and hepatic insulin sensitivity we have measured the ability of perfused livers from normal fed rats to respond to the effect of insulin on suppressing the net efflux of K + , glucose, amino acid nitrogen and urea nitrogen over a wide range of constant perfusate insulin levels. METHODS Liver perfusion. Livers from nonfasted male Sprague Dawley rats, weighing 114- 137 gm., were perfused in situ at constant flow rates ranging from 0.9 to 1. 2 ml. per gram liver per minute by previously described technics. 6 ' 7 The perfusate consisted of forty-five parts defibrinated rat blood, obtained from large male rats of the same strain, with the hematocrit adjusted to 20 per cent by addition of fifty-five parts Krebs Ringer bicarbonate (KRB) buffer containing 3gm. bovine albumin (Reheis Chemical Co., Chicago) per 100 ml. Both liver and blood donor animals were anesthetized with sodium thiamylal (62 mg. per 100 gm. body weight) prior to surgery. The perfusing medium volume averaged 33 ml. at the start of perfusion, and was recycled 225
INSULIN SENSITIVITY OF ISOLATED PERFUSED RAT LIVER
through the liver for 120 minutes. Blood samples of 1.0 ml. were drawn from the perfusate reservoir at 0, 30, 60, 90, and 120 minutes. Analytic methods. Blood samples were centrifuged, and serum was stored at -20° C. for subsequent analysis. Serum glucose was determined by electronic measurement of the rate of O2 consumption during the glucose oxidase reaction (Glucose Analyzer, Beckman Instruments, Fullerton, Calif.). Serum K + , amino acid nitrogen (AAN), and urea nitrogen were analyzed by methods'previously described, 67 and insulin concentration was determined by modification of the immunoassay procedure of Desbuquois and Aurbach. 12 Portal vein insulin concentration was assessed against rat insulin standards and perfusate serum insulin was assessed against porcine insulin standards. Microhematocrits were determined in duplicate on all blood samples for calculation of perfusate serum volume at each sampling time. Infusion of insulin. Porcine insulin, diluted in KRB buffer containing 1 per cent bovine albumin, was added continuously to the perfusate reservoir flask for 120 minutes. Although the volumes were equivalent in each experiment, the amount of insulin was varied in order to provide infusion rates corresponding to 6, 9, 16, 33 and > 42 mU insulin per hour. Control livers received a comparable volume of hormone diluent. Calculations. The net output of K + , AAN, urea nitrogen and glucose by the liver was calculated by multiplying the perfusate serum volume by the change in concentration of perfusate constituents from the zero time value. Net changes in serum volume were corrected for sampling losses, but not for infusion of insulin or hormone diluent, since the volume of hormone added (.435 ml. per hour) approximated the anticipated loss of volume due to bile secretion.6 The significance of differences between means was established by the Student / test.
with the tip of the needle lying anterior to the branch of the pyloric vein and blood was drawn into a 5 ml. syringe. These results are seen in table 1 and indicate that both plasma glucose and insulin levels were lower in fasted rats. The glucose levels were similar in the two systems, while the insulin levels were five to eight times greater in the portal vein.
TABLE l Systemic and portal vein glucose and insulin concentrations in rats fed ad libitum and rats fasted twenty-six hours
Group
Systemic Glucose Insulin (mg.%)
AiU/ml.
Portal vein Insulin Glucose AiU/ml. (mg.%)
Fed (6) Fasted (6)
153 ± 3 96 ± 7
24 ± 4 6 ± 2
156 ± 7 96 ± 7
125 ± 25 48 ± 14
Values represent mean ± S.E. of six rats in each group. Food was removed from the cages of rats fed laboratory chow ad libitum at 8 a.m. and blood samples were drawn from fed rats after two hours and from fasted rats after twenty-six hours.
Maximal effect of insulin. Infusion of insulin to six livers at rates ranging from 42 to 90 mU/hr. resulted in a progressive increase in serum insulin concentration to levels ranging from 510 to 1500 /u,U/ml. after 120 minutes' perfusion. Since the metabolic effect of insulin on suppressing the net accumulation of K + , AAN, urea nitrogen and glucose in the perfusing medium was comparable at all these doses, the results of these six studies were averaged and are illustrated in figure 1, along with the alterations in perfusate constituents of nine control livers. The data indicate that the effect of insulin on enhancing net uptake of K + by the liver began by thirty minutes and reached a maximum by sixty minutes. In contrast, the effect of insulin on suppressing the net accumulation of glucose, amino RESULTS acid nitrogen and urea nitrogen in the perfusing medium increased progressively for the duration of Systemic and portal vein serum insulin concentration in normal male rats. To determine the physiologic range of perfusion, and the maximum effect of the hormone insulin concentration in normal male rats fed laborat- occurred after 120 minutes' perfusion. ory chow ad libitum, blood samples were drawn from Effect of varying insulin concentrations on net release of the portal vein and tail vein of six rats two hours after liver K+, AAN, urea nitrogen and glucose. The perfusate removal of food, and from six rats fasted twenty-six insulin levels which resulted from infusion of insulin at hours. Rats were anesthetized with sodium thiamylal rates of 6, 9, 16 and 33 mU/hr. are listed in table 2. and systemic blood was sampled from the tip of the tail The data indicate that infusion of insulin at the above into EDTA-primed capillary tubes. Immediately rates resulted in stabilization of serum insulin concentthereafter, a 20-gauge needle filled with 5 per cent ration at levels averaging 41,72, 120 and 229 /-cU/ml. EDTA was inserted posteriorly into the portal vein between thirty and 120 minutes' perfusion. At these 226
DIABETES, VOL. 2 4 , NO. 2
CARL E. MONDON, PH.D., AND ASSOCIATES Amino Acid N 40-i
80-
I
20-
60-
E ~
0
40-
in
I -20-
20-
to o u
0-
o-
To
Glucose
T
Urea N
5 a
160n
160-
cu en « .c u a5
120-
120
80-
80
40-|
40
,/T
-
z
/
/
0
0 60
120
60
Minutes Control • • Insulin o o
120
The accumulation of serum K + after sixty minutes' perfusion and accumulation of serum AAN, urea nitrogen and glucose after 120 minutes' perfusion in each group was normalized to unit liver weight and these data are listed in table 3. These results indicate that the liver is most sensitive to the effects of insulin on K + uptake and AAN outflow under the conditions of this study. Thus, a diminished uptake of K + and suppression of AAN outflow was seen at perfusate insulin levels of AlfAJ/mi., although the effect on K4- did not attain statistical significance. Essentially, maximal effects of insulin on both K + and AAN seemed to occur at insulin concentrations of 72 /xU/ml. The ability of insulin to suppress urea nitrogen outflow was not seen until perfusate insulin reached a level of 120 jiiU/ml., and significant effects of insulin on inhibition of hepatic glucose release were only obtained at insulin levels in excess of 500 /xU/ml.
Minutes
DISCUSSION
FIG. 1. Effect of serum insulin concentrations greater than 500 /AJ/ml. on the net accumulation of serum K + , amino acid nitrogen, glucose and urea nitrogen in the perfusing medium. Vertical bars represent 1 S.E.
levels, we estimated from the rate of infusion that utilization of insulin by the liver remained constant, and averaged .45, .40, .39 and .47 juU/gm. liver per 1 /LtU/ml. insulin concentration, respectively. These values essentially agreed with results obtained in recent studies demonstrating that insulin removal by the isolated rat liver occurs at uniform rates approximating .35 /xU/gm. liver per minute per 1 ^tU/ml. insulin concentration at perfusate insulin levels ranging from 20 to 500/uU/ml. 1 3 The net accumulation of K + , AAN, urea nitrogen and glucose in the perfusing medium at the above insulin concentrations was compared with the nine control livers and six livers with perfusate insulin levels above 500 /u,U/ml. previously illustrated in figure 1.
In general, previous investigations of the action of insulin on the perfused liver have focused on whether or not insulin could exert a specific metabolic effect. In contrast, in these studies we have attempted to define the quantitative relationship between perfusate insulin concentration and the ability of insulin to modify several facets of hepatic metabolism. The results demonstrate that significant suppression of amino acid outflow occurred at perfusate insulin levels as low as 41 With elevation of perfusate insulin to 12 ., representing a 50 per cent increase of fasting portal vein insulin concentration, suppression of amino acid outflow had essentially reached maximum levels and K + uptake had plateaued at levels comparable to previously reported values averaging 5 to 6 /^moles K/g liver. 1 " 3 Larger quantitative effects of insulin on net K + movement have been reported during suppression of glycogenolysis,4"6 and the greater net uptake of K + obtained at maximal insulin concentrations (table 3)
TABLE 2 Serum insulin concentration during infusion of 6, 9, 16, 33 mU insulin per hour
Infusion rate (mU/hr.) 6 9 16 33
Liver wt. (gm •) 5.21 5.41 5.76 5.55
±: ±: ±: ±:
.29 .25 .19 .10
0' 37 ± 6 61 ± 9 110 ± 13 175 ± 27
30' 39 73 112 199
± ± ± ±
Serum insulin concentration 60' 4 8 3 9
41 ± 4 73 ± 13 113 ± 7 219 ± 10
42 75 127 237
± ± ± ±
X cone. 30'-120'
120'
90' 2 9 5 5
42 66 128 260
± ± ± ±
4 5
6 11
41 72 120 229
Values represent the mean ± S.E. of six experiments at each insulin infusion rate. Porcine insulin was added to each flask prior to start of infusion to raise the initial serum concentration to levels approximating the steady state level. FEBRUARY, 1975
227
INSULIN SENSITIVITY OF ISOLATED PERFUSED RAT LIVER
TABLE 3 Effect of serum insulin concentration on uptake of K+ and release of AAN, urea and glucose by isolated perfused liver Group: mean insulin cone. OiU/ml.) Control (9) 41 (6) 72 (6) 120 (6) 229 (6) >500 (6)
K+ uptake (/nmol/gm .) - 1 . 3 ± .6 2.6 ± 1 .8 3.9 ± .9 4.2 ± .6 5.2 ± .6 7.6 ± .7
Group: mean insulin cone. (MU/ml.)
AAN outflow (/imol/gm .)
Insulin effect (I-C)
14.4 10.6 5.7 6.3 5.8 5.4
-3.8 -8.7 -8.1 -8.6 -9.0
Control 41 72 120 229 >500
(9) (6) (6) (6) (6) (6)
± .8 ± 1. 3 ± .4 ± .5 ± ..7 ± .7
Insulin effect (I-C) 3.9 5.2 5.5 6.5 8.9
± 1.9 ± 1.1 ± .8 ± .8 ± .9
± ± ± ± ±
P
.05
