Glucose and amino acid metabolism effect of insulin and amino acids

in chronic

renal failure:

PIETRO CASTELLINO, ANNA SOLINI, LIVIO LUZI, J. GRANT BARR, DOUGLAS J. SMITH, ALEXANDER PETRIDES, MAURO GIORDANO, CHRIS CARROLL, AND RALPH A. DEFRONZO Divisions of Nephrology and Diabetes, University of Texas Health Science Center and Audi L. Murphy Veterans Administration Hospital, San Antonio, Texas 78284-7886; and Istituto di Medicina Interna e Nefrologia, Universita’ degli Studi di Napoli, Naples, Italy Castellino, Pietro, Anna Solini, Livio Luzi, J. Grant Barr, Douglas J. Smith, Alexander Petrides, Mauro Giordano, Chris Carroll, and Ralph A. DeFronzo. Glucose and amino acid metabolism in chronic renal failure: effect of insulin and amino acids. Am. J. Physiol. 262 (Renal Fluid Electrolyte Physiol. 31): F168-F176, 1992.-The effects of hyperinsulinemia and hyperaminoacidemia on glucose and amino acid metabolism were examined in 16 control and 13 chronic renal failure (CRF) patients under two conditions: 1) euglycemic hyperinsulinemia and 2) amino acid infusion. All studies were performed with continuous indirect calorimetry and [ lJ4C]leucine infusion. In CRF patients insulin-mediated whole body glucose metabolism was reduced by 35% (4.41 * 0.50 vs. 6.76 t 0.73 rng- kg-l. min-‘, P < O.Ol), primarily due to a decrease in nonoxidative glucose disposal (1.70 t 0.70 vs. 4.32 t 0.60 mg kg-’ min-‘, P c 0.01); glucose oxidation was similar in both groups. In the postabsorptive state total leucine turnover (1.56 t 0.06 vs. 1.75 & 0.06), leucine oxidation (0.25 & 0.01 vs. 0.30 k O.Ol), and nonoxidative leucine disposal (1.29 t 0.06 vs. 1.40 & 0.07 prnol. kg-l min-‘) were reduced in CRF vs. control subjects (all P < 0.05). In response to hyperinsulinemia, endogenous leucine flux (index of proteolysis), leucine oxidation, nonoxidative leucine disposal (NOLD) (index of protein synthesis), and net leucine flux into protein were similar in CRF and control subjects. In contrast, the ability of hyperaminoacidemia to enhance NOLD (1.54 & 0.11 vs. 2.10 t 0.10 pmol kg-‘. min-l, P < 0.01) and net leucine balance (0.27 t 0.05 vs. 0.41 t 0.05, P < 0.05) was reduced in CRF patients. In summary, in patients with CRF 1) basal leucine turnover and oxidation are reduced, 2) insulin-mediated suppression of proteolysis and net leucine flux into protein are normal, 3) amino acid-induced stimulation of protein synthesis and net flux of leucine into protein are impaired, and 4) insulin-mediated stimulation of glucose metabolism is reduced because of diminished nonoxidative glucose disposal. These results demonstrate a clear-cut dissociation between the effects of insulin on glucose vs. amino acid-protein metabolism, and an impairment in amino acid-induced stimulation of protein anabolism. l

l

l

l

protein

metabolism

RENAL FAILURE (CRF) is characterized by abnormalities in glucose (7, 9) and amino acid (1, 23) metabolism. The plasma amino acid profile is typically altered in CRF with a reduction in total and essential amino acid levels, a low ratio of essential to nonessential amino acids, and a decreased tyrosine-to-phenylalanine ratio (1, 23). Clinical signs of malnutrition and muscle wasting often are observed in these patients (16). Nonetheless, few studies have attempted to evaluate the changes in amino acid-protein metabolism that contribute to the above-mentioned alterations in plasma amino acid profile and protein anabolism. Even less is known

CHRONIC

F168

0363-6127/92

$2.00 Copyright

about the effect of insulin, the key anabolic hormone (4), on protein metabolism in CRF. This is of particular importance because it is well established that the ability of insulin to stimulate glucose utilization is severely impaired in patients with CRF (7, 9). Similarly, it is not known whether the anabolic effect of amino acid administration (4) is impaired in patients with chronic renal insufficiency. In the present study we have employed the euglycemic insulin clamp and amino acid infusion in combination with indirect calorimetry and [ 14C]leucine to examine the effects of insulin and hyperaminoacidemia on protein metabolism in patients with moderately advanced renal insufficiency. METHODS Subjects There were 16 healthy volunteers and 13 patients with CRF who participated in the study. The controls ranged in age from 31 to 57 yr [47 * 2 (SE) yr], and the CRF patients from 29 to 64 yr [49 & 3 (SE) yr]. All subjects were within 30% of their ideal body weight (113 & 4% in controls and 112 t 5% in CRF) based on the midpoint for medium frame individuals from the Metropolitan Life Insurance Tables, 1959. In control subjects the body weight and height were 75 t 3 kg and 172 & 2 cm, respectively. In the CRF group body weight and height averaged 72 -+ 3 kg and 170 t 2 cm, respectively. There were 11 males and 5 females in the control group and 10 males and 3 females in the CRF group. Except for renal insufficiency, no subject had any evidence of major organ system disease, and there was no family history of diabetes mellitus. Plasma creatinine (cl.2 mg/dl), urea nitrogen (~14 mg/dl), and electrolytes were within normal limits in controls. In CRF patients plasma creatinine averaged 4.0 t 0.4 mg/dl (range 2.8-7.0), urea nitrogen 54 t 9 mg/dl (range 30-94), creatinine clearance 24 t 4 ml/min (range lo-43), arterial pH 7.37 t 0.02 (range 7.34-7.40), arterial bicarbonate 23 & 1 meq/l (range 19-25), total protein 1.0 t 0.4 g/dl, albumin 3.5 t 0.3 g/dl, transferrin 250 k 24 mg/dl, hemoglobin 11 & 1 g/dl, calcium 9.1 * 0.4 mg/dl (range 8.710.0), phosphate 4.6 t 0.4 mg/dl (range 3.9-5.1); plasma sodium, chloride and potassium concentrations were normal. None of the CRF subjects was taking any medication except for multivitamins and bicarbonate. The mean duration of renal failure was 4 & 1 yr. The etiology of the renal disease was nephritis (n = 3), polyglomerulonephritis (n = 3), interstitial cystic kidney disease (n = 5), and idiopathic (n = 2). In both control and CRF subjects body weight was stable for at least 3 mo before study. All participants consumed a weight-maintaining diet containing at least 250 g of carbohydrate/day. Dietary protein intake was estimated by 24-h urinary nitrogen excretion that was obtained on the day preceding the study. The urinary nitrogen excretion averaged 14.4 t 1.4 g/day in controls and 12.7 & 1.2 g/day in CRF patients. The estimated dietary protein

0 1992 the American

Physiological

Society

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

GLUCOSE

AND

AMINO

ACID

intake averaged 90 t 8 and 80 $- 7 g/day in the controls and CRF subjects, respectively. The purpose and potential risks of the study were explained to all subjects, and their voluntary written consent was obtained before their participation. The study protocol was reviewed and approved by the Institutional Review Board of the University of Texas Health Science Center at San Antonio. Experimental

Protocol

All tests were performed in the postabsorptive state beginning at 0800 h after a 12-h overnight fast. A polyethylene catheter was inserted into an antecubital vein for infusion of all test substances.A secondcatheter was placed into a wrist vein for blood sampling.The hand waskept in a heated box at 70°C to ensure arterialization of the venous blood. In each study protocol a primed continuous infusion of [l-14C]leucine (16-22 PCi bolus followed by 0.20-0.25 &i/min) wasadministered in combination with a priming dose of [14C]bicarbonate (4 &i). After 2 h of isotopeequilibration, sampleswere drawn every lo-15 min from 120 to 180 min for the determination of baselineleucine and a-ketoisocaproic (KIC) specific activities and plasmahormoneand substratedeterminations. Continuous indirect calorimetry wasstarted at 120min. Expired air samples were collected at 15min intervals and bubbled through a CO, trapping solution (Hyamine hydroxide-absolute ethanol-O.1% phenolphthalein 3&l). The solution was titrated to trap 1 mmol of CO*/3 ml of the solution. The 14C02radioactivity was subsequently determined using a LS 5000 SE Scintillation Counter (Beckman Instruments, Fullerton, CA), and the expired 14C02specific activity was calculated. Total 14C02was calculated from the product of the CO:! specific activity, and the total 14C02production was determined as describedbelow. After the 180-min equilibration period, one of the following protocols was performed. In study 1, a prime continuous infusion of crystalline porcine insulin (Eli Lilly, Indianapolis, IN) was administeredat the rate of 40 mU rne2.mine1for an additional 180 min to raise and maintain the plasma insulin concentration by ~80 pU/ml above baseline.In patients with CRF, to offset the lower metabolic clearance rate of insulin (7, 9), the rate of insulin infusion was reduced by 20%. The plasma glucose concentration was maintained constant at the basal level by determination of the plasmaglucoseat 5-min intervals and periodic adjustment of a 20% glucoseinfusion aspreviously described(8). The insulin infusion was continued for 180 min. Urine was collected separately during the basal and insulin periods, and the urinary nitrogen concentration was measured. In study 2, a prime continuous infusion of a 10% balanced amino acid solution (Travasol lo%, Travenol, Savage, MD) was begunvia a peripheral vein at a rate of 1.7 ml gkg-‘. min-l and continued for 180min. This rate of infusion was chosento increaseand maintain the total plasmaamino acid concentration -1.5-fold above basal postabsorptive levels. Urine was collected separately during the basal and amino acid infusion periods,and the urinary nitrogen concentration was measured.

METABOLISM

IN

CRF

F169

determined using an amino acid analyzer (System 6300,Beckman). To precipitate plasmaproteins, 2.5 ml of 10% sulfosalicylic acid were addedto 2.5 ml of plasma,and a l-ml aliquot of the supernatant was analyzed in duplicate for plasma amino acid concentration. One milliliter of the remaining supernatant was placedin duplicate on a Dowex 50 G cation exchangeresin column (Bio-Rad Laboratories, Richmond, CA), and the free amino acid fraction was eluted with 4 N NH40H, subsequently dehydrated, and reconstituted in water. Scintillation fluid (10 ml) wasaddedto each vial and 14Cradioactivity wasmeasured in a LS 5000 SE Scintillation Counter. Plasma(x-KIC specific activity was measuredusing a modification of the method previously describedby Nissen et al. (31). Plasma (1 ml) was placed in duplicate on a Dowex 50 G cation exchange resin column (Bio-Rad Laboratories), and the free cw-ketoacidfraction was eluted with 4 ml of 0.01 N HCl in 50-ml culture tubes. Methylene chloride (35 ml) was added, and after shaking vigorously for 1 min the tube was centrifuged for 5 min at 2,000 rpm to extract the free ar-ketoacidfraction from plasma.After decantation of the supernatant the a-ketoacid wasextracted in 350 ~1of 0.2 M NaH2P04 at pH 7. After a brief centrifugation 200 ~1of the supernatant were injected into a high-performance liquid chromatographic system. The system utilizes a Cls reverse-phasecolumn (Waters Nova-Pak, 0.3 x 30 cm) that was eluted with 2% acetonitrile in 0.1 NaH2P04 buffer (pH 7.0) at a rate of 1.4 ml/min. Absorbance of KIC was monitored at 206 nm. Radioactivity eluting with the KIC peak was measuredby scintillation counting. The interassay and intra-assay variations for the determination of [“Cl leucine specificactivity were 4 t 2 and 5 t 2%, respectively. More than 98% of the radioactivity collected in the amino acid fraction was in the leucine peak after separation by ion exchange chromatography. The interassay and intra-assay variations for the determination of [‘“CIKIC specific activity were 5 -t 2 and 5 t 3%. The recovery of [‘“C]KIC was 68 & 4%. Plasma insulin concentration was measuredwith standard radioimmunoassaytechniques.Plasma glucagon concentration was determined by radioimmunoassay using the 30K antibody of Unger as previously described(11). Plasma glucoseconcentration was determined by the glucose oxidase method with a Beckman GlucoseAnalyzer (Beckman Instruments). Urinary nitrogen wasdeterminedby the Kjeldhal method (20). Methods for the determination of plasma catecholamines,growth hormone, and cortisol have beenpublished previously (35). Plasma fatty acid concentrations were measured by the microfluorometric method of Miles et al. (28). Calculations

Protein metabolism. Whole body leucine flux was calculated with a stochastic model for protein metabolism. The analysis assumesnear steady-state conditions. The validity and assumptionsof the modelhave beenpreviously discussedin detail by Waterlow et al. (40) and Golden and Waterlow (17). Briefly, the model generates the following equations in which total leucine turnover or flux equalsQ = S + C = B + I, where S is Respiratory Exchange Measurements the total rate of leucine incorporation into protein (or nonoxiIn all studies,respiratory exchangemeasurementswere per- dative leucine disposal),C is the rate of leucine oxidation, B is formed aspreviously described(38). Briefly, a plastic ventilated the rate of leucine release from protein (endogenousleucine hood wasplaced over the head of the subject and madeairtight appearance),and I is the rate of exogenousleucine input. around the neck. A slight negative pressurewas maintained in The rate of leucine turnover (Q) is calculated as follows: Q the hood to avoid lossof expired air. The carbon dioxide and = F/Leu sp act, where F is the infusion rate of [14C]leucine[in oxygen content of the expired air was continuously measured disintegrations/min (dpm)] and Leu sp act is the specificradioby a Deltatrac Metabolic Monitor (Sensormedic, Anaheim, activity of leucine in the plasma compartment under steadystate conditions. The leucine oxidation rate is calculated as CA) . follows c = O/(K x Leu sp act), where 0 is the rate of Analytical Determination appearanceof 14C02in the expired air (dpm/min) and E( is a Plasmaleucine and CY-KICspecific activities were measured correction factor (0.81) that takes into account the incomplete as previously described(4). Plasma leucine concentration was recovery of labeled 14C02from the bicarbonate pool (21). We Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

F170

GLUCOSE

AND AMINO

ACID METABOLISM

IN CRF

have demonstratedthat this recovery factor is unchangedduring a +lOO pU/ml euglycemic insulin clamp (R. Bonadonna and R. A. DeFronzo, unpublished observations). An estimate of the rate of leucine incorporation into protein (S) can be calculated as follows: S = Q - C. An estimate of the rate of leucine releaseinto the plasmaspacefrom endogenousprotein (B) can be calculated as follows B = Q - I. When subjectsare in the postabsorptive state, leucine intake (I) equalszero and B = Q. During the amino acid infusion, I equals the rate of leucine administration. To calculaterates of leucineturnover and oxidation, we have employedthe plasmaKIC specific activity becauseit has been suggestedthat the plasma cu-KIC specific activity, the transaminated product of leucine, may provide a better estimate of the specific activity in the intracellular mixing pool (25, 34). Glucosemetabolism.During the insulin clamp studies the glucoseinfusion rate was calculated at 20-min intervals, and a small spacecorrection was applied for over or under filling of the glucosespacewhen appropriate (8). For data presentation, the meanof the three 20-min intervals from 120to 180 min is given. Hepatic glucoseproduction was not determined in the present study. However, we have previously shown that both in normal and in uremic subjectsa similar level of hyperinsulinemia suppresses hepatic glucoseproduction by >90% (7,36). Therefore, under the steady-state conditions of euglycemic hyperinsulinemia employed in the present study, glucoseuptake by the entire body can be consideredequal to the rate of exogenousglucoseinfusion (corrected for changesin the glucose spacethat was CO.1 mg kg-l. min-’ in all subjects). Rates of glucoseand lipid oxidation and energy expenditure were calculated from indirect calorimetric measurementsthat were averagedover 5-min intervals during the basalstate and during the insulin clamp as previously described (38). Nonoxidative glucosedisposalwascalculated by subtracting the glucoseoxidation rate from the rate of total body glucoseuptake during the 120-to 180-min interval of the insulin clamp.

the 60- to 180-min interval of the insulin clamp study (P < 0.01 vs. basal). After amino acid infusion, plasma glucagon levels rose significantly and averaged 222 t 33 pg/ml (P < 0.01 vs. basal). In patients with CRF the fasting plasma glucose was similar to controls and averaged 89 t 3 mg/dl. During the insulin clamp study the steady-state plasma glucose concentration averaged 87 t 3 mg/dl with a coefficient of variation of 3 t 2%. After amino acid infusion, no change in plasma glucose concentration was observed (90 i 4 mg/dl). Urinary glucose was absent in both studies. The mean fasting plasma insulin concentration was 12 t 1 pU/ml and rose to 65 t 7 pU/ml during the insulin clamp study [P < 0.001 vs. basal and P = not significant (NS) vs. controls]. In study 2, after amino acid infusion, plasma insulin concentration increased significantly and averaged 25 t 2 pU/ml (P < 0.01 vs. basal and P = NS vs. controls). The fasting plasma glucagon concentration was 235 t 13 pg/ml (P < 0.01 vs. controls) and declined to 202 t 20 pg/ml during the insulin clamp (P c 0.01 vs. basal). In study 2, plasma glucagon levels rose to 360 t 28 pg/ml (P < 0.01 vs. basal and controls) in the 60- to 180-min interval of the amino acid infusion. The fasting plasma growth hormone concentrations were 5.4 t 3.1 rig/ml in CRF patients and 1.6 t 0.6 in controls (P < 0.05). Plasma epinephrine and norepinephrine levels averaged 33 t 5 and 219 t 25 pg/ml in controls and 52 t 16 and 342 t 94 pg/ml in CRF patients (P = NS), respectively. Plasma cortisol concentration averaged 427 t 25 pg/ml in CRF and 560 t 34 pg/ml in controls (P = NS). Basal total plasma calcium concentrations were 9.1 t 0.4 mg/dl in CRF patients and 9.3 t 0.3 mg/dl in controls Statistical Analysis and did not change significantly during study 1 (9.0 t 0.5 and 9.1 t 0.3 mg/dl in CRF patients and controls, All values are expressedas meanst SE. Comparisonsbe- respectively) or in study 2 (8.9 t 0.6 and 9.0 t 0.3 mg/ tween the basaland the insulin (or amino acid) infusion period were performed using the t test for paired data. Intergroup dl). Basal plasma phosphorus was 4.6 t 0.4 mg/dl in CRF patients and 3.5 t 0.3 mg/dl in controls (P C 0.01 comparisonswere performed by one-way analysisof variance. vs. controls). In study 1 plasma phosphorus declined to 4.1 t 0.3 and 3.1 t 0.3 mg/dl in CRF patients and RESULTS controls, respectively. No significant changes were observed in study 2 (4.4 t 0.5 mg/dl in CRF patients and Plasma Substrate and Hormone Concentrations 3.4 t 0.4 mg/dl in controls). In the control group the fasting plasma glucose concentration averaged 88 t 3 mg/dl. During the euglycemic Glucose and Lipid Metabolism (Fig. 1) insulin clamp, the steady-state plasma glucose level was maintained close to baseline (88 t 3 mg/dl) with a In control subjects, the rate of glucose infusion recoefficient of variation of 4 t 2%. During amino acid quired to maintain euglycemia during the 120- to 180infusion the plasma glucose concentration (89 t 2 mg/ min interval of the euglycemic insulin clamp (study 1) dl) did not change significantly from baseline. Urinary :;:i:Cont1 rols El glucose was absent in both studies. Basal plasma free fatty acid concentrations were 382 t 64 and 462 t 85 N CRF v8 controls pmol/l in control and CRF patients, respectively. The mean fasting plasma insulin concentration was 7 t 1 pU/ml and rose to 69 t 3 pU/ml during the 60- to 180min interval of the insulin clamp study (P < 0.001 vs. basal). During amino acid infusion (study 2), the plasma M ox NON OX insulin concentration also rose and averaged 19 t 2 pU/ Fig. 1. Whole body insulin-mediated glucose metabolism (M), glucose ml during the 60- to 180-min time interval (P < 0.01 vs. oxidation (OX), and nonoxidative glucose disposal (NON-OX) in conbasal). Basal plasma glucagon averaged 115 t 9 pg/ml trol and CRF subjects during last hour of euglycemic insulin clamp. All in the basal state and declined to 70 t 8 pg/ml during values represent means t SE. l

:: .‘...... .a.*.*.* .. .* m l .*.*.*. l .*.*.*. :.>:.I. . . . . .. ...‘... . . .

:: :: i

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

GLUCOSE

AND AMINO

ACID METABOLISM

was significantly higher than in CRF patients (6.76 t 0.73 and 4.41 t 0.50 mg* kg-‘. min-‘, P < 0.01). Basal glucose oxidation was similar in the two groups and averaged 1.37 t 0.18 mg* kg-l. min-’ in controls and 1.57 t 0.18 mg kg-‘. min-’ in uremics. During the last hour of the insulin clamp study, glucose oxidation rose significantly and similarly in both groups and averaged 2.78 t 0.29 and 2.61 t 0.37 mg#kg-’ emin-’ in CRF and control subjects, respectively (both P < 0.01 vs. basal and P = NS for CRF vs. controls). The rate of nonoxidative glucose disposal (total glucose uptake minus glucose oxidation) during the 120- to MO-min interval of the insulin clamp study was significantly reduced in CRF patients compared with controls (1.70 t 0.70 vs. 4.32 t 0.60 mgo kg-’ amin-‘, P < 0.01). The basal rate of lipid oxidation was similar in control and CRF subjects (0.62 t 0.09 and 0.65 t 0.13 mgekg-‘*min-‘, respectively) and declined in both groups (to 0.20 t 0.08 and 0.06 t 0.05 rng. kg-‘. min-‘) during the 120- to 180-min period of hyperinsulinemia (P < 0.01 vs. basal, P = NS for CRF vs. controls). Energy expenditure in the postabsorptive state was similar in control and CRF subjects (1.01 t 0.05 and 1.01 t 0.05 kcal/min) and rose slightly to 1.11 t 0.07 and 1.07 t 0.07 kcal/min during hyperinsulinemia in the two groups (both P < 0.05 vs. basal). In the postabsorptive state the nonprotein respiratory quotient (NPRQ) was similar in the two groups and averaged 0.86 t 0.02 and 0.84 t 0.02 in control and CRF patients, respectively. After hyperinsulinemia the NPRQ rose similarly in the two groups and averaged 0.96 t 0.03 and 0.97 t 0.01, respectively (both P < 0.01 vs. basal, P = NS for CRF vs. controls). l

Plasma Amino Acid Concentrations (Table 1) and Plasma Leucine and cu-KIC Specific Activity (Figs. 2 and 3) In the control group insulin (study 1) caused a consistent decrease in most plasma amino acid concentrations, with the notable exceptions of alanine and glycine.

F171

IN CRF

Total plasma amino acid concentration fell from 1,931 t 104 to 1,662 t 48 pmol/l, branched-chain amino acid (BCAA) concentration from 401 t 19 to 207 t 16 pmol/l, plasma leucine from 123 t 6 to 56 t 6, and plasma cyKIC from 32 t 3 to 17 t 2 pmol/l (all P < 0.01 vs. basal). A steady-state plateau for plasma leucine and KIC concentrations was achieved during the last hour of the insulin clamp study (Fig. 2). Plasma leucine and KIC specific activities were constant during the last hour of the equilibrium period and averaged 1.96 t 0.33 and 1.27 t 0.28 dpm/nmol, respectively, and rose to new steadystate plateaus (leucine 2.97 t 0.51, KIC 1.91 t 0.41 dpm/ nmol) during the last hour of insulin infusion (Fig. 2). In patients with CRF the total plasma amino acid concentration in the basal state averaged 1,666 t 87 pmol/l and decreased to 1,368 t 92 ,umol/l during the last hour of insulin infusion (P < 0.001 vs. basal, P = NS vs. controls). In the postabsorptive state BCAA levels (322 t 23 pmol/l) were significantly lower (P < 0.01) than in controls and declined to 216 t 17 pmol/l after hyperinsulinemia (P < 0.01 vs. basal). Basal plasma leucine (93 t 7 pmol/l) and KIC (19 t 2 pmol/l) concentrations were significantly reduced in CRF patients compared with controls (P c 0.01). During the last hour of insulin infusion a lower steady-state plateau in plasma leucine and KIC concentrations was observed in both groups (Fig. 2). Basal leucine and KIC specific activities averaged 4.00 t 0.28 and 2.32 t 0.30 dpm/nmol, respectively, and rose to a new higher constant plateau during the last hour of insulin infusion (Fig. 2). In the control group, after amino acid infusion (study 2), there was a significant rise in total plasma amino acid concentration from 2,064 t 98 to 4,297 t 274 pmol/ 1, BCAA from 381 t 22 to 1,228 t 41 pmol/l, plasma leucine from 122 t 8 to 381 t 22 pmol/l, and plasma aKIC from 34 t 3 to 54 t 3 pmol/l (all P < 0.01 vs. basal). A new steady-state plateau was achieved for plasma leucine and KIC concentrations and specific activities during the last hour of the study (Fig. 3). Plasma leucine and KIC specific activities averaged 2.80 t 0.29 and 1.71

Table 1. Plasma amino acid concentrations in 4 experimental protocols in basal state and during 120- to 180-min time interval Controls study

Tau ASP

Thr

Ser Glut GlY Ala CYS Val Met Ile Leu TYr

Chronic

1

study

Basal

Insulin

52k4 8k2 148t12 112t11 545t33 221*18 265241 66t4 217t8 27t4 61k4 123t6 45t4 41t4 1,931+104 401*19

38*4* 5+3-t 96t7* 74*5* 502t31 193t13 260t37 49+3t 129t6* 16t6* 22t4* 56&6* 26t6* 30+4t 1,662+48* 207k167

Basal 58t3 9k2 138*13 103t8 641t30 251t20 315+41 68t3 192k9 34*3 65k3 124t8 4727 45,t7 2,064+98 381k22

Phe Total amino acids BCAA Values are means t SE in pmol/l. BCAA, branched-chain

2

Amino

Study

acids

63k4t 6+2 303t40* 262t21” 629t33 761t63* 970*88* 52t3* 545t36* 180tl1* 302t18* 381t22” 69k7t 155tl1” 4,297*274* 1,228*41*

Basal 47k6 8t1 88+10$ 84+16$ 564t23 184t19 199+20$ 61t4 182+22$ 18+1$ 47+3$ 93*7$ 39+2 52t2 1,666+87$ 322+23$

Renal Failure

1

study

Insulin

33&6-f6zkl.f 63+10*$ 64*12* 521k27 165&19$ 179+17$ 47f3”f 139+19t 941”$ 24t3* 53t8* 27t2* 38&2* 1,368&92*$ 216&17*

Basal 67k6 9t2 9329% 81+6$ 556k22 257t40 225+_22$ 89t4 135+11$ 25&3$ 56+4$ 98&7$ 29+5$ 38t3 1,758+97 289k26

2

Amino acids 75+8t 6t2 267t25* 247t24* 549t25 881&98* 913+90* 77t3 44Ok59” 167-+9* 287t21* 346t26* 36t6 174t7 4,465+261* 1,073+43

amino acids. * P < 0.01 and t P < 0.05 vs. basal. $ P < 0.01 vs. controls.

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

F172

GLUCOSE

AND

AMINO

ACID

9-Y-P

OL

L -30

BASAL

30

-

150

0

-30

180

0

INSULIN

-

150

BASAL

180 INSULIN

IN

CRF

26 vs. 1,076 t 46 pmol/l), leucine (98 t 7 vs. 346 t 26 pmol/l), and KIC (21 t 3 vs. 32 t 4 pmol/l) rose significantly (all P < 0.01 vs. basal). During the last hour of amino acid infusion a new steady state in plasma leucine and KIC levels and specific activities (Fig. 3) was achieved. Urinary leucine concentration was 40 mmol/ 1 in the two groups in the basal state and after both experimental manipulations. Total Leucine Flux, Leucine Oxidation Leucine Disposal (Figs. 4 and 5)

-&-A-h

zw-0 L---2

ml

METABOLISM

and Nonoxidative

In controls basal leucine turnover, as calculated from the plasma KIC specific activity, averaged 1.75 t 0.06 In patients with CRF basal leucine pmol kg-‘. min. turnover was significantly reduced in comparison with controls and averaged 1.56 t 0.06 prnol. kg-‘. min-’ (P c 0.01 vs. controls). After hyperinsulinemia leucine turnover decreased significantly in both groups (P < 0.01 vs. basal) and averaged 1.06 t 0.07 and 0.98 t 0.04 prnol. kg-‘. min-’ in control and CRF subjects, respectively. The percentage decline was similar in both groups and averaged 39 t 3% in controls and 38 t 4% in CRF (Fig. 4, top left). In study 2, amino acid infusion induced a significant rise in leucine turnover both in controls (3.43 t 0.10 pmol . kg-’ min-‘) and in CRF (2.80 t 0.16 pmol . kg-’ l rein-‘) (both P < 0.01 vs. basal) (Fig. 5, top left). In controls, basal leucine oxidation averaged 0.30 t 0.02 pmol . kg-‘. min-’ and this represented 17 t 1% of the total flux. In patients with CRF basal leucine oxidation was significantly reduced in comparison with controls and averaged 0.25 t 0.01 prnol. kg-’ emin-’ (16 t 1% of the total flux) (P < 0.01 vs. controls). After insulin infusion the leucine oxidation rate declined significantly (P c 0.01 vs. basal) in both groups and averaged 0.17 t 0.02 pmol kg-’ min-’ (17 t 2% of the total flux) in l

21.

20 -y-w,

T -of

i5 --A-Ll

10 -

T 0-0

-w&

P-P-P

T---r P

O-

-30

-9-P

0

L 150

0

Z---T

-30

180

BASAL

INSULIN

TIME

-00

0

BASAL

(minutes)

CONTROLS

150

TIME ---O--

180

INSULIN (minutes)

CRF

Fig. 2. Plasma leucine concentration (top left) and specific activity (top right), and plasma ketoisocaproic (KIC) concentration (bottom left) and specific activity (bottom right) in control and chronic renal failure (CRF) subjects during basal state and euglycemic insulin clamp. Values are means t SE. 400

-

300

-

200

-

4-

r,-Li!Lzs T--f--f--f

a

&-is-&-.6 -*--p-*--q

100

9-T-f-T

35 E E 2r

i

j

l

k

P-P-P-P

g-g-g-g

l-

l

OL

OL II

1

-

-30

0 BASAL

150 AMINO

180

*30

ACIDS

0

’

l

2.0

A

150

180

AMINO

BASAL

60-

’

ACIDS

.-E 1.5 .

c

< Z

1.0.

f,

< Z k

4o

P-9

=2 E

-9-p r--r--r--r

20 -y-r

OL

-f B

‘--T--r

-

0 BASAL

AMINO

TIME -o-

Fig. 3. Plasma (top right), and activity (bottom and amino acid

150

0.0.

’

INS CLAMP

0

L11

-30

0.5.

180

I

’

BASAL

ACIDS

TIME

(minutes)

CONTROLS

’

-30

--e-0

I

L

0

150 AMINO

’

’

BASAL

INS CLAMP

BASAL

INS CLAMP

I

1.5

180 ACIDS

(minutes)

CRF

leucine concentration (top left) and specific activity plasma KIC concentration (bottom left) and specific right), in control and CRF subjects during basal state infusion. Values are means t SE.

+ 0.20 dpm/nmol, respectively, and declined to 1.54 t G.20 and 1.03 t 0.16 dpm/nmol, respectively, during the last hour of amino acid infusion. In CRF patients, after amino acid infusion total plasma amino acid concentration (1,758 t 97 VS. 4,465 t 261 pmol/l), BCAA (289 t

0.2 .E’

f & 1.0 < z

1

5. 0.6

0.0 BASAL

.‘.. : .I... *:*:*:a: *::.*.* . . . . . CONTROLS I3

INS CLAMP

IEI

CRF

Fig. 4. Endogenous leucine flux (top left), leucine oxidation (top right), nonoxidative leucine disposal (bottom Left), and net leucine balance (bottom right) in control and CRF subjects during basal period and last hour of euglycemic insulin clamp (INS CLAMP), as calculated using plasma KIC specific activity. Values are means t SE. # P C 0.05 vs. controls. * P c 0.01 VS. basal.

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

GLU COSE

AND

AMINO

ACID

METABOLISM

IN

F173

CRF

0.05 vs. 0.29 t 0.06 prnol* kg-‘. min-‘,

respectively)

(Fig.

5, bottom right).

lb.E b l.O-r 5 5 0.6-

DISCUSSION

In the present study we have evaluated the effect of insulin on glucose and leucine metabolism in patients o.owith moderately advanced chronic renal insufficiency. In BASAL BASAL AA INF these CRF patients total body glucose disposal during euglycemic hyperinsulinemia was reduced by 35% compared with controls. This degree of impairment of insu,g o.slin-mediated glucose metabolism is consistent with pre...... E ...... ...... ...... ...... in _ ...... vious results (7,9,32,36) obtained in patients with more # ...... ...... ...... -r ...... .*.-.*.*.,.* Z ...... advanced renal insufficiency and indicates that the onset 5 o.o...... ...... ...... ...... of insulin resistance occurs well before the onset of ...... ...... ........... symptoms or signs of the uremic syndrome. We have -O.Spreviously demonstrated that in both uremic and control BASAL AA INF BASAL AA INF subjects muscle is the primary tissue responsible for the removal of the majority, 80-90%, of an infused glucose CRF load under euglycemic hyperinsulinemic conditions (7). Fig. 5. Endogenous leucine flux (top left), leucine. oxiaation (top rzgnt), The glucose that is taken up by muscle can undergo one nonoxidative leucine disposal (bottom left), and net leucine balance of two metabolic fates: 1) oxidation to carbon dioxide (bottom right) in control and CRF subjects in basal state and last hour of amino acid infusion (AA INF) period, as calculated using plasma and water or 2) nonoxidative glucose disposal, which KIC specific activity. Exogenous rate of leucine infusion is shown by primarily represents glycogen formation. In the present open bars (top left). Values are means t SE. # P < 0.01 vs. controls. study basal glucose oxidation in CRF patients was similar * P < 0.01 vs. basal. to controls and increased normally in response to insulin. controls and 0.16 t 0.02 prnol. kg-‘. min-’ (17 t 1% of These results are in agreement with previous data in rats of insuthe total flux) in CRF (P = NS vs. controls) (Fig. 4, top (26), which demonstrated a normal stimulation lin-mediated glucose oxidation and lactate release in right). In study 2, hyperaminoacidemia induced a similar increase in leucine oxidation in controls (1.18 t 0.10 muscle preparations of acutely uremic animals. In conpmol kg-‘. min-’ or 34 t 4% of the total flux) and in trast, the ability of insulin to stimulate nonoxidative CRF (1.27 t 0.08 ~molokg-l~min-l or 45 t 5% of the glucose disposal (total glucose uptake minus glucose oxtotal flux) (both P < 0.01 vs. basal, P = NS CRF vs. idation) in CRF patients was reduced by >50%, and this defect accounted for all of the impairment in whole body controls) (Fig. 4, top right). glucose disposal. These results suggest that a specific Basal nonoxidative leucine disposal (NOLD), an index defect in glycogen formation is responsible for the insulin of protein synthesis, was significantly higher (P c 0.05) resistance of uremia. The normal increase in glucose in control subjects in comparison with CRF and averaged 1.40 t 0.07 and 1.29 t 0.06 prnol. kg-’ min-‘, respec- oxidation in response to insulin also argues against an early defect in insulin action, i.e., insulin receptor or tively. After hyperinsulinemia, NOLD declined to 0.88 t 0.06 pmol . kg-‘. min-’ in controls and to 0.84 t 0.04 glucose transport, because this would be expected to pmol kg-‘. min-1 in CRF (both P < 0.001 vs. basal; P = impair both the glucose oxidative and nonoxidative pathNS, CRF vs. controls) (Fig. 4, bottom left). In study 2, ways to a similar extent. In a previous paper we have hyperaminoacidemia induced a smaller rise in NOLD in shown that in CRF patients the dose-response curve relating the plasma insulin concentration and glucose CRF (1.54 t 0.11 prnol* kg-‘. min-‘) than in controls metabolism is shifted to the right and does not normalize (2.10 t 0.10 pmolkg-’ .min-‘) (both P C 0.01 vs. basal, at supraphysiological insulin levels (36). Similar results P c 0.01 vs. controls) (Fig. 5, bottom left). have been obtained by Schmitz et al. (32). In vitro studies have documented normal insulin receptor binding and Net Flux of Leucine Into Protein receptor phosphorylation in CRF. Therefore, both the in The difference between the NOLD and the endogenous vivo and in vitro data are consistent and argue for a leucine flux represents the net flux of leucine into protein postreceptor defect in insulin action. and as such provides an index of protein anabolism. In Our observation that nonoxidative glucose disposal is the postabsorptive state the net leucine flux into protein impaired in chronically uremic humans is in good agree: was negative, -0.30 t 0.02 and -0.26 t 0.01 prnol. kg-‘. ment with previous reports from in vitro muscle prepamin-l in controls and CRF, respectively. In controls, rations from uremic animals that showed a reduction in after insulin, the net leucine balance became less nega- insulin-stimulated glycogen synthesis (26). Direct meastive, -0.17 t 0.02 pmol kg-’ .min-‘, but a positive value urements of muscle glycogen synthase activity in acutely was never observed. A similar response was observed in uremic animals (26) also have demonstrated impaired CRF, -0.16 t 0.04 prnol. kg-lomin-l (Fig. 4, bottom enzyme activity. It is noteworthy that a similar reduction right). After amino acid infusion, the leucine flux into in nonoxidative glucose disposal (with normal or only protein became markedly positive in both groups and slightly reduced glucose oxidation) has been observed in was significantly higher in controls than in CRF (0.41 t patients with a variety of clinical disorders that are l

:

1

‘1

. .

/,

.

I

,\

l

l

l

l

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

F174

GLUCOSE

AND AMINO

ACID METABOLISM

characterized by insulin resistance, including noninsulindependent diabetes mellitus, obesity, and essential hypertension (2, 6, 12, 24). Whether acquired (i.e., obesity, uremia) or inherited (i.e., type II diabetes, essential hypertension), a defect in glycogen formation appears to be a characteristic finding in all of these varied insulinresistant states. CRF represents a complex metabolic model with concomitant elevation of many hormones (glucagon, growth hormone, cortisol, insulin, catecholamines) and substrates (free fatty acids), all of which have been shown to induce insulin resistance in normal subjects. However, in uremic patients maintained on chronic hemodialysis or continuous ambulatory peritoneal dialysis (CAPD), the hormones remain elevated, yet the insulin resistance is almost completely reversed (5, 7, 9, 32). Moreover, plasma levels of glucagon, cortisol, and catecholamines were not significantly increased in the uremic group. Because elevated plasma growth hormone levels have been shown to impair insulin-mediated glucose disposal, we cannot exclude a role for this hormone in the insulin resistance observed in our uremic subjects. However, because growth hormone is protein anabolic, elevated levels cannot explain the defect in amino acid-induced stimulation of protein synthesis. Because all parameters of leucine turnover (see subsequent discussion) in response to insulin were normal in CRF patients, it also follows that the elevated levels of catecholamines, growth hormone, and glucagon do not exert a deleterious effect on insulin-mediated leucine metabolism. The severe insulin resistance with respect to glucose metabolism does not appear to extend to amino acidprotein metabolism. In the postabsorptive state CRF patients demonstrated the characteristic plasma amino acid profile (1, 23), with a reduction in both total and BCAA amino acid levels. Plasma leucine concentration was reduced by 20% compared with controls. This decrease could not be explained by an increase in plasma leucine oxidation or nonoxidative leucine disposal, i.e., protein synthesis, because both of these metabolic pathways were significantly reduced in the CRF group. Of particular note, the net flux of leucine into protein was less negative in uremic than in control subjects, indicating that in the postabsorptive state patients with moderately advanced renal insufficiency are not excessively catabolic. The reason for the decrease in basal leucine flux and plasma leucine concentration is not readily obvious from the present study. One possible explanation relates to the elevated fasting plasma insulin concentration in the uremic patients. It previously has been shown that the metabolic clearance rate of insulin is prolonged in subjects with chronic renal insufficiency (9), and insulin secretion is commonly increased (9). Even a small increment in the plasma insulin concentration exerts a powerful inhibitory action on protein degradation (13). This would lead to a decline in plasma leucine concentration with secondary decreases in both leucine oxidation and nonoxidative leucine disposal (3). Consistent with this scenario, we observed a significant correlation (r = -0.65, P < 0.05, Fig. 6) between the fasting plasma insulin concentration and the endogenous leucine flux. The decrease in basal leucine oxidation and basal leucine

IN CRF 2.2

43

I.&

1.0’

0

1

1 I

I

I

I

I

I

I

I

I

If

0

2

4

6

8

10

12

14

16

18

FASTING

20

PLASMA INSULIN ( NJ/ml 1

Fig. 6. Correlation between fasting plasma insulin levels and basal leucine flux in normals and CRF subjects (r = -0.67; P < 0.05).

flux observed in the present study is consistent with previous results obtained using various tracers ([ 14C] valine, [ 15N] lysine, [ 13C]leucine, and [ 14C]leucine) in uremic children and adults (3,X& 33). In contrast, Goodship et al. (18) have recently reported normal leucine flux and oxidation in CRF patients despite the presence of elevated fasting plasma insulin levels. Taken collectively, these results are at variance with the concept that CRF is characterized by increased protein catabolism and negative nitrogen balance under postabsorptive conditions. It should be noted, however, that these observations do not exclude a disturbance in amino acidstimulated or insulin-stimulated protei n metabolism, as occurs after t he ingestion of a mixed meal. Moreover, most of the studies in which a significant catabolic effect of uremia was dot umented were performed in animals with acute renal failure, in whom surgical stress and acute metabolic acidosis may have played a major role (19, 26), or in patients with a more advanced degree of renal insufficiency (14, 16). It is well established that insulin is a major protein anabolic hormone (4,13,37). It is possible that the ability of insulin to promote protein anabolism is impaired by the uremic state. To examine this issue, uremic subjects received a euglycemic insulin clamp. In response to a physiological increment in the plasma insulin concentration, the endogenous leucine flux, i.e., protein degradation, declined and the net leucine balance became less negative to a similar extent as control subjects (Fig. 4). These results suggest a normal sensitivity to the antiproteolytic action of insulin. Consistent with this, May et al. (27) observed an increase in the estimated rate of proteolysis in the absence of insulin but documented a normal suppression of protein degradation when insulin was added to an isolated muscle preparation from acutely uremic rats. The present results clearly demonstrate that under basal conditions and in the insulin-stimulated state, CRF patients do not demonstrate any tendency toward a protein catabol ic state. The normal response of leucine turnover to insulin stands in marked contrast to the severe impairment in insulin- mediated glucose dis-

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

GLUCOSE

AND

AMINO

ACID

posal. This dichotomy provides convincing evidence that an insulin receptor defect cannot account for the insulin resistance of uremia. Before it is concluded that protein metabolism is normal in CRF patients, it is important to note that insulin failed to stimulate nonoxidative leucine disposal, i.e., protein synthesis, either in controls or in patients with chronic renal insufficiency. This observation is consistent with previous results reported in normal subjects by us (4), as well as by others (13, 37). To produce a stimulation of protein synthesis, it was necessary to infuse amino acids (4) to mimic the combined hyperaminoacidemia-hyperinsulinemia that occurs after the ingestion of a protein meal. In the present study amino acid infusion caused an approximate twofold increase in plasma insulin and amino acid levels in both control and CRF subjects. This combination of hyperinsulinemia and hyperaminoacidemia stimulated leucine oxidation to a similar extent in both groups, consistent with previous studies demonstrating that the plasma amino acid concentrations (4,30) are a major regulator of their oxidative rate. In contrast to leucine oxidation, the ability of amino acids and insulin to augment nonoxidative leucine disposal (an index of protein synthesis) during amino acid infusion was significantly impaired in patients with CRF (Fig. 5). As a result, net leucine balance into protein was 32% lower in CRF patients compared with controls. Insulin, while maintaining plasma amino acid levels constant, does not augment nonoxidative leucine disposal (4). This suggests that the anabolic response to amino acid-protein administration is impaired by the uremic state. Moreover, because combined hyperaminoacidemia caused a normal inhibition of the endogenous leucine flux (i.e., protein degradation), the impaired anabolic response observed in CRF patients can be attributed entirely to a defect in protein synthesis. This reduced anabolic response to amino acid infusion may have important clinical relevance to the management of patients with moderate-severe chronic renal failure in whom muscle wasting and malnutrition are not uncommon. It is important to note that in the present study CRF patients were maintained on a free protein and caloric intake. None had any clinical evidence of muscle wasting or weakness, suggesting that the impaired stimulation of protein metabolism after amino acid infusion may have been compensated for by maintaining a protein intake that was well above the daily minimal requirement. Consistent with this, the dietary protein intake was estimated to be 80 t 7 g/day in the CRF group. This observation also indicates that the diminished ability of hyperaminoacidemia to stimulate protein synthesis in our CRF patients cannot be explained by dietary protein deficiency. However, if the dietary protein intake is restricted, as has been recommended to slow the progression of CRF (29), the reduced anabolic response to combined hyperinsulinemia-hyperaminoacidemia may lead to negative nitrogen balance and clinically significant muscle wasting and weakness. One previous study (18) has examined the effect of protein feeding on [ f4C]leucine turnover in patients with CRF and failed to observe any difference between CRF and control subjects. However, in both CRF and control individuals, these investigators

METABOLISM

IN

F175

CRF

failed to observe any stimulation of protein synthesis by protein feeding in either group despite the presence of hyperinsulinemia and hyperaminoacidemia, factors that are known to augment protein anabolism. It is likely that the failure to observe any increase in protein synthesis in this study (18) resulted from the long duration of the tracer infusion (10 h) and resultant increase in recycling of the tracer (40). This would be expected to cause an underestimation of protein synthesis and breakdown and an overestimation of the anabolic response (leucine balance) to protein ingestion. Currently, the National Institute of Health is sponsoring a large-scale prospective trial to examine the effect of a low-protein diet on renal function in patients with CRF. In view of the present results, one must be concerned about the impact of such a diet on muscle metabolism and function. This is an area that clearly needs more study before the use of such low-protein diets can be widely applied to treat patients with advanced chronic renal insufficiency. Last, some comment is warranted concerning the contribution of decreased metabolism by renal tissues vs. uremia-related alterations in the metabolic milieu to the observed differences in glucose and protein metabolism between control and CRF patients. We previously have shown that during a euglycemic insulin clamp muscle and splanchnic tissues account for essentially all of the decline in plasma amino acid concentration (7, 9). Similar results were obtained after amino acid infusion (15). These studies indicate that the renal tissues per se do not play a major role in total body amino acid-protein homeostasis. Consistent with this conclusion, renal vein catheterization studies have failed to demonstrate a significant uptake of BCAA, indicating that utilization of these substrates by the kidney is small (10, 39). We thank Eleanor Andjuar and Anna Crowder for assistance in performing the various analytical determinations. Sheri Contero and Stella Merla provided expert secretarial support in preparation of the manuscript. This work was supported in part by General Clinical Research Center Grant RR-01346, by the GRECC, and by Veterans Affairs Medical Research Funds. Address for reprint requests: R. A. DeFronzo, Diabetes and Nephrology Div., Univ. of Texas Health Science Center, 7703 Floyd Curl Dr., San Antonio, TX 78284-7886. Received

26 November

1990; accepted

in final

form

3 September

1991.

REFERENCES 1. Bergstrom, J., P. Furst, L. 0. Noree, and E. Vinnars. Intracellular free amino acids in muscle tissue of patients with chronic uraemia: effect of peritoneal dialysis and infusion of essential amino acids. Clin. Sci. 1MoZ. Med. 54: 51-60, 1978. 2. Bogardus, C., S. Lillioja, K. Stone, and D. Mott. Correlation between muscle glycogen synthase activity and in vivo insulin action in man. J. Clin. Invest. 73: 1185-1190, 1984. 3. Castellino, P., R. A. DeFronzo, and C. Giordano. Glucose metabolism and insulin sensitivity in CAPD patients. Peritoneal Dial. Bull. Suppl. 7: 22, 1987. 4. Castellino, P., L. Luzi, D. C. Simonson, M. Haymond, and R. A. DeFronzo. Effect of insulin and plasma amino acid concentrations on leucine metabolism in man: the role of substrate availability on estimates of whole body protein synthesis. J. Clin. Invest. 80: 1784-1793,1987. 5. Conley, S. B., G. M. Rose, A. M. Robson, and D. M. Bier. Effects of dietary intake on protein turnover in uremic children. Kidney Int. 17: 837-846,198O.

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

F176

GLUCOSE

AND

AMINO

ACID

6. DeFronzo, R. A. The triumvate: beta-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes 37: 667-687, 1988. 7. DeFronzo, R. A., A. Alvestrand, D. Smith, R. Hendler, E. Hendler, and J. Wahren. Insulin resistance in uremia. J. CZin. Invest. 67: 563-568, 1981. 8. DeFronzo, R. A., J. D. Tobin, and R. Andres. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am. J. Physiol. 237 (Endocrinol. Metab. Gastrointest. Physiol. 6): E214-E223, 1979. 9. DeFronzo, R. A., J. D. Tobin, J. W. Rowe, and R. Andres. Glucose intolerance in uremia. Quantification of pancreatic beta cell sensitivity to glucose and tissue sensitivity to insulin. J. CZin. Invest. 62: 425-435, 1978. 10. Dies, F., J. Herrera, M. Matos, E. Avelar, and G. Ramos. Substrate uptake by the dog kidney in vivo. Am. J. Physiol. 218: 405-410,197o. 11. Faloona, G. R., and R. H. Unger. Glucagon. In: Methods for Hormone Radioimmunoassay, edited by B. M. Jaffe and H. R. Bohman. New York: Academic, 1974, p. 317-330. 12. Ferrannini, E., G. Buzzigoli, R. Bonadonna, M. A. Giorico, M. Oleggini, L. Graziadei, R. Pedrinelli, L. Brandi, and S. Bevilacqua. Insulin resistance in essential hypertension. N. EngZ. J. Med. 317: 350-357,1987. 13. Fukagawa, N. K., K. L. Minaker, J. W. Rowe, M. N. Goodman, D. E. Matthews, D. M. Bier, and V. R. Young. Insulin mediated reduction of whole body protein breakdown. Dose-response effects on leucine metabolism in postabsorptive men. J. CZin. Invest. 76: 2306-2311, 1985. 14. Furst, P., A. Alvestrand, and J. Bergstrom. Effects of nutrition and catabolic stress on intracellular amino acid pools in uremia. Am. J. CZin. Nutr. 33: 1387-1395, 1980. 15. Gelfand, F., M. G. Glickman, P. Castellino, and R. A. DeFronzo. Measurement of “C-l-leucine kinetics in splanchnic and leg tissues in humans: effect of amino acid infusion on regional protein turnover. Diabetes 37: 1365-1372, 1988. 16. Giordano, C., N. G. DeSanto, and R. Senatore. Effects of catabolic stress in acute and chronic renal failure. Am. J. CZin. Nutr. 31: 1561-1571,1978. 17. Golden, M. H. N., and J. C. Waterlow. Total protein synthesis in elderly people: a comparison of results with [i5N]glycine and [‘4C]leucine. Clin. Sci. Mol. Med. 53: 277-288, 1977. 18. Goodship, T. H. J., W. E. Mitch, R. A. Hoerr, D. A. Wagner, T. H. Steinmann, and V. R. Young. Adaptation to low-protein diets in renal failure: leucine turnover and nitrogen balance. J. Am. Sot. Nephrol. 1: 66-75, 1990. 19. Harter, H. R., L. E. Karl, S. Klahr, and D. M. Kipnis. Effects of reduced renal mass and dietary protein intake on amino acid release and glucose uptake by rat muscle in vitro. J. CZin. Invest.

METABOLISM

25.

26.

27.

28.

29. 30.

31.

32.

33.

34.

35.

36. 37.

64:513-523,1979. 20. Hawk, P. D. Kjeldahl method. In: Practical Physiological Chemistry (12th ed.). Toronto, Canada: Blakiston, 1947, p. 814-822. 21. Issekutz, B., P. Paul, H. I. Miller, and W. M. Bortz. Oxidation of plasma FFA in lean and obese humans. MetaboZism 17: 62-73, 1968. 22. Jones, M. R., and J. D. Kopple. Valine metabolism in normal and chronically uremic man. Am. J. CZin. Nutr. 31: 1660-1664, 1978. 23. Kopple, J. D., and M. R. Jones. In: Advances in Nephrology, edited by J. Hamburger, J. Crosnier, and J. P. Grunfeld. Chicago, IL: Year Book Medical Publishers, 1979, vol. 8, p. 233-268. 24. Lillioja, S., D. M. Mott, J. K. Zawadzki, A. A. Young, W. G. Abbott, and C. Bogardus. Glucose storage is a major determi-

38.

39.

40.

IN

CRF

nant of in vivo “insulin resistance” in subjects with normal glucose tolerance. J. CZin. Endocrinol. Metab. 62: 922-927, 1986. Matthews, D. E., H. P. Schwartz, R. D. Yang, K. J. Motil, V. R. Young, and D. M. Bier. Relationship of plasma leucine and alpha-ketoisocaproate during a [ l-13C]leucine infusion in man: a method for measuring human intracellular leucine tracer enrichment. Metabolism 31: 1105-1112, 1980. May, R. C., A. S. Clark, A. Goheer, and W. E. Mitch. Specific defects in insulin-mediated glucose metabolism in acute uremia. Kidney Int. 28: 490-497,1985. May, R. C., R. A. Kelly, and W. E. Mitch. Mechanisms for defects in muscle protein metabolism in rats with chronic uremia: the influence of metabolic acidosis. J. CZin. Invest. 79: 1099-1103, 1987. Miles, J. R., J. Glassock, J. Aikens, J. Gerich, and M. Haymond. A microfluorometric method for the determination of free fatty acids in plasma. J. Lipid Res. 24: 96-99, 1983. Mitch, W. E. The influence of diet on the progression of renal insufficiency. Annu. Rev. Med. 35: 249-264,1984. Motil, K. J., D. E. Matthews, D. M. Bier, J. F. Burke, H. N. Munro, and V. R. Young. Whole body leucine and lysine metabolism: response to dietary protein’ intake in young men. Am. J. Physiol. 240 (Endocrinol. Metab. 3): E712-E721, 1981. Nissen, S. L., C. Van Huisen, and M. W. Haymond. Measurements of branched chain amino acids and branched chained alphaketoacids in plasma by high performance liquid chromatography. J. Chromatogr. 232: 170-175,1982. Schmitz, O., K. G. M. M. Alberti, N. J. Christensen, C. Hasling, E. Hjollund, H. Beck-Nielsen, and H. Orskov. Aspects of glucose homeostasis in uremia as assessed by the hyperinsulinemic euglycemic clamp technique. Metabolism 34: 465-473, 1985. Schrieber, M., S. Kalhan, A. McCullough, and S. Savin. Branched chain amino acid metabolism in chronic renal failure and haemodialysis. Proc. Eur. Dialysis Transplant Assoc. 22: 116120,1985. Schwenk, W. F., B. Beaufrere, and M. W. Haymond. Use of reciprocal pool specific activities to model leucine metabolism in humans. Am. J. Physiol. 249 (Endocrinol. Metab. 12): E646-E650, 1985. Simonson, D. C., W. V. Tamborlane, R. A. DeFronzo, and R. S. Sherwin. Intensive insulin therapy reduces counterregulatory hormonal responses to hypoglycemia in patients with type I diabetes. Ann. Intern. Med. 103: 184-190, 1985. Smith, D., and R. A. DeFronzo. Insulin resistance in uremia is mediated by postbinding defects. Kidney Int. 22: 54-62, 1982. Tessari, P., R. Trevisan, S. Inchiostro, G. Biolo, R. Nosadini, S. V. DeKreutzenberg, E. Duner, A. Tiengo, and G. Crepaldi. Dose-response curves of effects of insulin on leucine kinetics in humans. Am. J. Physiol. 251 (Endocrinol. Metab. 14): E334-E342,1986. Thiebaud, D., E. Jacot, R. A. DeFronzo, E. Maeder, E. Jequier, and J. P. Felber. The effect of graded doses of insulin on total glucose uptake, glucose oxidation and glucose storage in man. Diabetes 31: 957-963, 1982. Tizianello, A., G. DeFerrari, G. Garibotto, G. Gurreri, and C. Robaudo. Renal metabolism of amino acids and ammonia in subjects with normal renal function and in patients with chronic renal insufficiency. J. CZin. Invest. 65: 1162-1173, 1980. Waterlow, J. C., P. J. Garlik, and D. J. Millward. Protein Turnover in Mammalian Tissues and in the Whole Body. New York: Elsevier-North Holland, 1978, p. 118-250.

Downloaded from www.physiology.org/journal/ajprenal by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.