Secretion, Degradation, and Elimination of Glucagon-Like

Home , Angiotensin

Secretion, Degradation, and Elimination of Glucagon-Like Peptide 1 and Gastric Inhibitory Polypeptide in Patients with Chronic Renal Insufﬁciency and Healthy Control Subjects Juris J. Meier,1 Michael A. Nauck,1,2 Daniel Kranz,1 Jens J. Holst,3 Carolyn F. Deacon,3 Dirk Gaeckler,4 Wolfgang E. Schmidt,1 and Baptist Gallwitz1

CRI patients vs. healthy control subjects, respectively. ؎ ؎ Glucagon-like peptide 1 (GLP-1) and gastric inhibitory polypeptide (GIP) are important factors in the patho- Plasma half-lives of intact GIP were 6.9 1.4 and 5.0 ؎ ؎ ؍ genesis of type 2 diabetes and have a promising thera- 1.2 min (P 0.31) and 38.1 6.0 and 22.4 3.0 min for ؍ peutic potential. Alterations of their secretion, in vivo the GIP metabolite (P 0.032) for CRI patients vs. degradation, and elimination in patients with chronic healthy control subjects, respectively. Insulin concen- renal insufficiency (CRI) have not yet been character- trations tended to be lower in the patients during all ized. Ten patients with CRI (aged 47 ؎ 15 years, BMI experiments, whereas C-peptide levels tended to be kg/m2, and serum creatinine 2.18 ؎ 0.86 elevated. These data underline the importance of the 2.2 ؎ 24.5 mg/dl) and 10 matched healthy control subjects (aged kidneys for the final elimination of GIP and GLP-1. The years, BMI 24.9 ؎ 3.4 kg/m2, and serum creati- initial dipeptidyl peptidase IV–mediated degradation of 12 ؎ 44 nine 0.89 ؎ 0.10 mg/dl) were included. On separate both hormones is almost unaffected by impairments in occasions, an oral glucose tolerance test (75 g), an renal function. Delayed elimination of GLP-1 and GIP in intravenous infusion of GLP-1 (0.5 pmol ⅐ kg؊1 ⅐ min؊1 renal insufficiency may influence the pharmacokinetics over 30 min), and an intravenous infusion of GIP (1.0 and pharmacodynamics of dipeptidyl peptidase IV–resis- pmol ⅐ kg؊1 ⅐ min؊1 over 30 min) were performed. Venous tant incretin derivatives to be used for the treatment of blood samples were drawn for the determination of patients with type 2 diabetes. Diabetes 53:654–662, glucose (glucose oxidase), insulin, C-peptide, GLP-1 2004 (total and intact), and GIP (total and intact; specific immunoassays). Plasma levels of GIP (3-42) and GLP-1 (9-36 amide) were calculated. Statistics were performed using repeated-measures and one-way ANOVA. nsulin secretion after the ingestion of a mixed meal After the oral glucose load, plasma concentrations of is stimulated not only by the rise in glucose concen- intact GLP-1 and intact GIP reached similar levels in trations but also by the secretion of incretin hor- respectively). The ,0.87 ؍ and P 0.31 ؍ both groups (P concentrations of GIP (3-42) and GLP-1 (9-36 amide) Imones, namely glucagon-like peptide 1 (GLP-1) and were significantly higher in the patients than in the gastric inhibitory polypeptide (GIP; also referred to as respec- glucose-dependent insulinotropic polypeptide), from the ,0.027 ؍ and P 0.0021 ؍ control subjects (P tively). During and after the exogenous infusion, GLP-1 gut (1,2). Both hormones are currently considered for the (9-36 amide) and GIP (3-42) reached higher plasma treatment of type 2 diabetes because of their glucose- concentrations in the CRI patients than in the control lowering activity (3,4). However, the therapeutic use of the .respectively), peptides is still limited by their short in vivo half-lives ,0.0033 ؍ subjects (P < 0.001 and P whereas the plasma levels of intact GLP-1 and GIP were Both hormones are cleaved within minutes at the NH - 2 ,0.27 ؍ and P 0.29 ؍ not different between the groups (P ,(terminus by the enzyme dipeptidyl peptidase IV (DPP IV ؎ ؎ respectively). Plasma half-lives were 3.4 0.6 and 2.3 yielding the fragments GLP-1 (9-36 amide) and GIP (3-42) and 5.3 ؎ 0.8 and (0.13 ؍ min for intact GLP-1 (P 0.4 for (5,6). The cleavage products have lost their insulinotropic (0.029 ؍ min for the GLP-1 metabolite (P 0.4 ؎ 3.3 activity and may even act as partial antagonists at their respective receptors (7–9). Different approaches are currently being evaluated to From the 1Department of Medicine I, St. Josef Hospital, Ruhr University Bochum, Bochum, Germany; the 2Diabeteszentrum Bad Lauterberg, Bad make use of the therapeutic potential of the incretin Lauterberg, Germany; the 3Department of Medical Physiology, The Panum hormones: DPP IV–resistant analogues of GIP and GLP-1 Institute, University of Copenhagen, Copenhagen, Denmark; and the 4Outpa- have been synthesized to extend the in vivo half-life of the tient Center for Diabetology and Nephrology, Bochum, Germany. Address correspondence and reprint requests to Dr. Juris J. Meier, Division peptides (10,11), and inhibitors of the degrading enzyme of Endocrinology and Diabetes, Keck School of Medicine, University of DPP IV have been generated to block the rapid degrada- Southern California, 1333 San Pablo St., Los Angeles, CA 90033. E-mail: [email protected]. tion of endogenous GIP and GLP-1 (11a). Received for publication 22 September 2003 and accepted in revised form Earlier studies already indicated that both incretin hor- 21 November 2003. mones are eliminated by the kidneys (12,13). This was CRI, chronic renal insufficiency; DPP IV, dipeptidyl peptidase IV; GIP, gastric inhibitory polypeptide; GLP-1, glucagon-like peptide 1. supported by elevated plasma concentrations of GIP and © 2004 by the American Diabetes Association. GLP-1 found in patients with uremia (13a,13b). However,

654 DIABETES, VOL. 53, MARCH 2004 J.J. MEIER AND ASSOCIATES

TABLE 1 Characteristics of the subjects/patients Healthy control Parameter subjects Patients with CRI Significance (P value)* Anthropometric data Sex (female/male) 4/6 5/5 0.65 Age (years) 44 Ϯ 12 47 Ϯ 15 0.59 BMI (kg/m2) 24.9 Ϯ 3.4 24.5 Ϯ 2.2 0.80 Waist-to-hip ratio (cm/cm) 0.81 Ϯ 0.09 0.92 Ϯ 0.05 0.31 Hematological, metabolic, and lipid parameters Hemoglobin (g/dl) 14.3 Ϯ 0.44 11.9 Ϯ 0.3 0.00033 Fasting glucose (mmol/l) 5.66 Ϯ 0.56 5.28 Ϯ 0.72 0.24 120-min glucose (mmol/l)† 6.94 Ϯ 1.0 6.17 Ϯ 1.33 0.17 Ϯ Ϯ HbA1c (%) 5.9 0.5 5.6 0.7 0.38 Ϯ Ϯ HOMAB-cell function (% normal)‡ 88 55 78 59 0.70 Ϯ Ϯ HOMAinsulin resistance (fold normal)‡ 2.48 2.08 1.49 1.19 0.21 Total cholesterol (mmol/l) 5.67 Ϯ 1.37 5.05 Ϯ 0.73 0.21 HDL cholesterol (mmol/l) 1.11 Ϯ 0.44 1.27 Ϯ 0.65 0.54 LDL cholesterol (mmol/l) 4.144 Ϯ 1.22 3.21 Ϯ 0.47 0.039 Triglycerides (mmol/l) 1.85 Ϯ 1.2 1.49 Ϯ 0.87 0.46 Hypertension (yes/no) 0/10 9/1 Ͻ0.0001 Blood pressure Systolic (mmHg) 121 Ϯ 12 127 Ϯ 11 0.25 Diastolic (mmHg) 82 Ϯ 784Ϯ 4 0.43 Parameters of kidney function Serum creatinine (mg/dl) 0.89 Ϯ 0.10 2.18 Ϯ 0.86 0.00016 Serum urea (mg/dl) 25.9 Ϯ 3.86 84.7 Ϯ 27.7 Ͻ0.0001 Creatinine clearance (ml/min)§ 107 Ϯ 27 46 Ϯ 24 Ͻ0.0001 Cystatin C (mg/l) 0.71 Ϯ 0.13 1.9 Ϯ 0.46 Ͻ0.0001 Albuminuria (mg/day) 20.7 Ϯ 46 350.2 Ϯ 602 0.12 Proteinuria (mg/day) 150 Ϯ 151 827 Ϯ 128 0.14 Data are means Ϯ SD. *ANOVA or ␹2 test. †120 min after oral glucose ingestion. ‡Calculated according to (24). §Calculated according to (25). because those studies were based on immunoassays that ments of standard hematologic and clinical chemistry parameters. Urine was were unable to discriminate the intact hormone levels collected over 24 h for the determination of albumin and protein by standard methods. Patients with anemia (hemoglobin Ͻ10 g/dl) and elevation in liver from their respective degradation products, it was not enzymes (alanine aminotransferase, aspartate aminotransferase, AP, ␥-glu- possible to take DPP IV–mediated degradation of the tamine transferase) to higher activities than double the respective normal hormones into consideration. The availability of specific value were excluded. The participant characteristics are presented in Table 1. The diagnoses leading to renal insufficiency included immunoglobulin A antibodies raised against the NH2-termini of intact GIP (1-42) and GLP-1 (7-36 amide) now allows determination nephropathy (Berger’s disease) in three cases, amyloidosis in one case, cystic kidney disease in two cases, secondary nephroangiosclerosis in two cases, of their degradation and elimination in more detail (6,20). and hereditary renal dysplasia in two cases. Five CRI patients and four control Type 2 diabetes is often complicated by the develop- subjects who had participated in the screening were excluded because they ment of renal insufficiency (17). This may have an influ- had impaired glucose tolerance. Nine patients with renal insufficiency re- ence on the pharmacokinetic and pharmacodynamic ceived one or more antihypertensive drugs: angiotensin-converting enzyme inhibitors in seven, angiotensin II receptor antagonists in four, ␤-blocking properties of incretin derivatives to be used as antidiabetic agents in three, diuretics in five, calcium antagonists in three, and ␣-blocking drugs. Therefore, we studied the secretion as well as the agents in two cases. In contrast, none of the control subjects received degradation and elimination of GIP and GLP-1 in patients antihypertensive medication. with chronic renal insufficiency (CRI) and in healthy Peptides. Synthetic human GLP-1 was a gift from Restoragen (Lincoln, NE; control subjects. For clearly distinguishing alterations as a lot number [pharmaceutical grade] 0340298). Synthetic human GIP was purchased from PolyPeptide Laboratories (Wolfenbu¨ ttel, Germany; lot num- result of impaired renal function from those secondary to ber [pharmaceutical grade] E-0517). High-performance liquid chromatography diabetes (6,44), only patients with renal insufficiency as a profiles (provided by the manufacturer) showed that the preparation was result of causes other than diabetes were included in this Ͼ95% pure (single peak co-eluting with appropriate standards). The peptides study. were dissolved in 0.9% NaCl/1% human serum albumin (Behring, salt poor, Marburg, Germany), filtered through 0.2-␮m nitrocellulose filters (Sartorius, Go¨ ttingen, Germany) and stored frozen at Ϫ28°C. Samples were analyzed for RESEARCH DESIGN AND METHODS bacterial growth (standard culture techniques) and for pyrogens (laboratory The study protocol was approved by the ethics committee of the Ruhr- of Dr. Balfanz, Mu¨ nster, Germany). No bacterial contamination was detected. University of Bochum on 20 January 2000 (registration number 1,417) before Endotoxin concentrations in samples from the stock solutions were 0.08 IU/ml the study. Written informed consent was obtained from all participants. for both GLP-1 and GIP. Participants. Two groups of subjects/patients were studied: 1) 10 subjects Study design. At a screening visit, venous blood samples for the determina- with normal renal function: subjects were included when they had a serum tion of standard hematologic and clinical chemistry parameters were drawn creatinine concentration Ͻ1.1 mg/dl (normal range, 0.5–1.1 mg/dl) and exhib- and a clinical examination was performed. When the subjects met the ited no other clinical signs of renal insufficiency; and 2) 10 patients with CRI: inclusion criteria, they were recruited for the following tests. All subjects were patients were included when they had a serum creatinine concentration Ͼ1.5 examined on three occasions: 1) oral glucose tolerance test: after basal mg/dl. All patients with impaired or diabetic oral glucose tolerance were venous blood samples were drawn (Ϫ15 and 0 min), 75 g of oral glucose excluded. (OGTT; Roche Diagnostics, Mannheim, Germany) was ingested within 5 min; From all participants, blood was drawn in the fasting state for measure- venous blood samples were drawn after 30, 60, 90, and 120 min; 2) GLP-1

DIABETES, VOL. 53, MARCH 2004 655 RENAL ELIMINATION OF GLP-1 AND GIP

FIG. 1. Plasma concentrations of glucose (A), insulin (B), and C- peptide (C) after oral ingestion of 75 g of glucose (arrows) in 10 patients with CRI (}) and 10 healthy control subjects (F). Data are FIG. 2. Plasma concentrations of total GLP-1 (7-36 amide plus split means ؎ SE. P values were calculated using repeated measures ANOVA products [A], intact GLP-1 (7-36 amide [B]), and the GLP-1 metabolite and denote differences between the groups (A), differences over time (total GLP-1 minus intact GLP-1 [C]) after oral ingestion of 75 g of (B), and differences as a result of the interaction of group and time glucose (arrows) in 10 patients with CRI (ࡗ) and 10 healthy control AB). *P < 0.05 at individual time points (one-way ANOVA). subjects (F). Data are means ؎ SE. P values were calculated using) repeated-measures ANOVA and denote differences between the groups infusion: after basal venous and capillary blood samples were drawn twice, at (A), differences over time (B), and differences as a result of the ϭ t 0 min, synthetic human GLP-1 (7-36 amide) was infused over 30 min with interaction of group and time (AB). *P < 0.05 at individual time points an infusion rate of 0.5 pmol ⅐ kgϪ1 ⅐ minϪ1; venous blood samples were (one-way ANOVA). obtained at 15, 30, 32, 35, 40, 50, 60, 90, 120, 150, and 180 min; and 3) GIP infusion: after basal venous blood samples were drawn twice, at t ϭ 0 min, Munich, Germany). Insulin was measured using an insulin microparticle synthetic human GIP (1-42) was infused over 30 min with an infusion rate of enzyme immunoassay (IMx Insulin; Abbott Laboratories, Wiesbaden, Ger- 1.0 pmol ⅐ kgϪ1 ⅐ minϪ1; venous blood samples were obtained at 15, 30, 32, 35, many). Intra-assay coefficients of variation were ϳ4%. C-peptide was mea- 40, 50, 60, 90, 120, 150, and 180 min. sured using an enzyme-linked immunoabsorbent assay from DAKO The tests were performed in the morning after an overnight fast with the (Cambridgeshire, U.K.). Intra-assay coefficients of variation were 3.3–5.7%, subjects in a sitting position throughout the experiments. Two forearm veins and interassay variation was 4.6–5.7%. were punctured with Teflon cannulas (Moskito 123, 18 gauge; Vygon, Aachen, GLP-1 immunoreactivity was determined using two different assays spe-

Germany) and kept patent using 0.9% NaCl (for blood sampling and for GIP or cific for either the COOH-terminus or the NH2-terminus of the peptide. The GLP-1 administration, respectively). At least 48 h had to pass between the COOH-terminal assay measures the sum of the intact peptide plus the primary tests to avoid carryover effects. metabolite GLP-1 (9-36 amide) using the antiserum 89390 and synthetic GLP-1 Blood specimen. Venous blood was drawn into chilled tubes containing (7-36 amide) as standard. This assay cross-reacts Ͻ0.01% with COOH- EDTA and aprotinin (Trasylol; 20,000 KIU/ml, 200 ␮l/10 ml blood; Bayer AG, terminally truncated fragments and 83% with GLP-1 (9-36 amide). The ϳ ␮ Leverkusen, Germany) and kept on ice. A sample ( 100 l) was stored in NaF detection limit was 3 pmol/l. The NH2-terminal assay measures the concen- (Microvette CB 300; Sarstedt, Nu¨ mbrecht, Germany) for the immediate tration of intact GLP-1 (7-36 amide) using antiserum no. 93242, which measurement of glucose. After centrifugation at 4°C, plasma for hormone cross-reacts ϳ10% with GLP-1 (1-36 amide) and Ͻ0.1% with GLP-1 (8-36 analyses was kept frozen at Ϫ28°C. This procedure has previously been amide) and GLP-1 (9-36 amide). The detection limit was 3 pmol/l. For both shown to prevent in vitro degradation of incretin hormones in human plasma assays, intra-assay and interassay coefficients of variation were Ͻ6 and 15%, samples (14). respectively, at 40 pmol/l (20). Laboratory determinations. Glucose was measured as described (19) using GIP immunoreactivity was determined using two different assays specific a glucose oxidase method with a Glucose Analyzer 2 (Beckman Instruments, for either the COOH-terminus or the NH2-terminus of the peptide as well (6).

656 DIABETES, VOL. 53, MARCH 2004 J.J. MEIER AND ASSOCIATES

FIG. 3. Plasma concentrations of total GIP (1-42 plus split products FIG. 4. Plasma concentrations of glucose (A), insulin (B), and C- [A], intact GIP (1-42 [B]), and the GIP metabolite (total GIP minus peptide (C) during and after the intravenous infusion of GLP-1 (0.5 intact GIP [C]) after oral ingestion of 75 g of glucose (arrows) in 10 pmol ⅐ kg؊1 ⅐ min؊1) over 30 min in 10 patients with CRI (ࡗ) and 10 patients with CRI (ࡗ) and 10 healthy control subjects (F). Data are healthy control subjects (F). Data are means ؎ SE. P values were means ؎ SE. P values were calculated using repeated-measures ANOVA calculated using repeated-measures ANOVA and denote differences and denote differences between the groups (A), differences over time between the groups (A), differences over time (B), and differences as (B), and differences as a result of the interaction of group and time a result of the interaction of group and time (AB). *P < 0.05 at (AB). *P < 0.05 at individual time points (one-way ANOVA). individual time points (one-way ANOVA).

The COOH-terminal assay using antiserum R65 fully reacts with intact GIP Diagnostica, Marburg, Germany), as described (23). Intra-assay variation was (1-42) and the truncated metabolite (3-42) but not with the so-called 8-kDa Ͻ3.3%, and interassay variation was Ͻ4.1%.

GIP, of which the chemical nature and relation to GIP secretion is uncertain. Calculations. Plasma half-lives were calculated by loge linear regression The assay has a detection limit of Ͻ2 pmol/l and an intra-assay variation of analysis of the peptide concentrations in the samples collected after the end ϳ 6%. The NH2-terminal assay measures the concentration of intact GIP (1-42), of the infusion period. For integrated incremental responses of glucose, using antiserum 98171. The cross-reactivity with GIP (3-42) was Ͻ0.1%. The insulin, and C-peptide, the positive or negative area under the curve was lower detection limit of the assay is ϳ5 pmol/l. Intra-assay variation was Ͻ6%, calculated using the trapezoidal method (baseline subtracted). Metabolic and interassay variations were ϳ8 and 12% for 20 and 80 pmol/l standards, clearance rates were calculated from the total amount of peptide infused respectively. For both assays, human GIP (Peninsula Laboratories, Europe) divided by the integrated incremental plasma concentrations. The distribution was used as standard, and radiolabeled GIP was obtained from Amersham space (DS) was calculated using the formula DS ϭ MCR/k, where MCR is the ϭ Pharmacia Biotech (Aylesbury, U.K.). Valine-pyrrolidide (0.01 mmol/l, final metabolic clearance rate and k is the fractional clearance rate ( 0.693/t1/2). concentration) was added to the assay buffers to prevent NH2-terminal The homeostasis model assessment model was used as an estimate of insulin degradation of GIP during the assay incubation. The concentrations of the resistance and B-cell function (24). Creatinine clearance was calculated metabolites GIP (3-42 amide) and GLP-1 (9-36 amide) were calculated as the according to the formula by Cockcroft and Gault (25). concentration differences between the total GIP/GLP-1 and the intact GIP/ Statistical analysis. Results are reported as mean Ϯ SE. All statistical GLP-1. calculations were carried out using repeated-measures ANOVA using Statis- Cystatin C plasma concentrations were determined for the assessment of tica Version 5.0 (Statsoft Europe, Hamburg, Germany). This analysis provides the glomerular filtration rate, because this parameter was postulated to be P values for differences between groups/experiments, differences over time, even more accurate than creatinine clearance (21,22). Cystatin C concentra- and for the interaction of group/experiment with time. When a significant tions were measured using a commercially available assay (Behringwerke interaction of treatment and time was documented (P Ͻ 0.05), values at single

DIABETES, VOL. 53, MARCH 2004 657 RENAL ELIMINATION OF GLP-1 AND GIP

high degree of correlation between the cystatin C concentrations and the creatinine clearance (r ϭ 0.81; P Ͻ 0.0001; details not shown). Urinary albumin and total protein excretion were higher in the patients only by trend but without a statistically significant difference between the groups (Table 1). Hemoglobin concentrations were lower in the patients than in the control subjects (P ϭ 0.00033; Table 1). Oral glucose tolerance test. After the ingestion of 75 g of oral glucose, plasma glucose concentrations increased significantly in both groups (P Ͻ 0.001; Fig. 1A) but without any differences between the patients and the control subjects (P ϭ 0.61). Insulin and C-peptide concentrations rose significantly in both groups (P Ͻ 0.001). There was a trend toward higher rises in C-peptide concentrations but lower insulin levels in the CRI patients (Fig. 1B and C). As a result, the molar C-peptide–to–insulin ratio calculated from the total integrated plasma concentrations after oral glucose ingestion was significantly higher in the patients than in the control subjects (11.2 Ϯ 1.0 vs. 7.5 Ϯ 1.4, respectively; P ϭ 0.045). Plasma concentrations of intact GLP-1 (7-36 amide) only marginally increased after glucose ingestion. There were no differences between the patients with renal insufficiency and the control subjects (P ϭ 0.31; Fig. 2B). In contrast, plasma concentrations of the GLP-1 metabolite (9-36 amide) significantly increased in response to the oral glucose load (P Ͻ 0.0001). The plasma levels of GLP-1 (9-36 amide) in the CRI patients significantly exceeded those reached in the control subjects (P ϭ 0.027). Plasma concentrations of intact GIP (1-42) increased after the oral glucose load (P Ͻ 0.001), but no differences occurred between both groups (P ϭ 0.87; Fig. 3B). There was a marked increase in the GIP metabolite (3-42) in both groups (P Ͻ 0.001). The rise in GIP (3-42) levels after glucose ingestion was greater in the renal insufficiency patients than in the control subjects (P ϭ 0.0021). GLP-1 infusion. During the intravenous infusion of FIG. 5. Plasma concentrations of total GLP-1 (7-36 amide plus split GLP-1, plasma glucose concentrations were lowered sig- products [A], intact GLP-1 (7-36 amide [B]), and the GLP-1 metabolite nificantly in both groups (P Ͻ 0.001; Fig. 4A). This was (total GLP-1 minus intact GLP-1 [C]) during and after the intravenous infusion of GLP-1 (0.5 pmol ⅐ kg؊1 ⅐ min؊1) over 30 min in 10 patients accompanied by a significant rise in insulin and C-peptide with CRI (ࡗ) and 10 healthy control subjects (F). Data are means ؎ secretion (P Ͻ 0.001; Fig. 4). There were no differences in SE. P values were calculated using repeated-measures ANOVA and denote differences between the groups (A), differences over time (B), glucose, insulin, or C-peptide concentrations between the and differences as a result of the interaction of group and time (AB). CRI patients and the control subjects (P ϭ 0.55, P ϭ 0.87, *P < 0.05 at individual time points (one-way ANOVA). and P ϭ 0.96, respectively; Fig. 4). Plasma concentrations of intact GLP-1 (7-36 amide) time points were compared by one-way ANOVA. P Ͻ 0.05 was taken to increased during GLP-1 infusion in both groups (P Ͻ 0.001; indicate significant differences. Fig. 5A). There were no differences in the intact GLP-1 levels between the patients with renal insufficiency and the RESULTS control subjects (P ϭ 0.29; Fig. 5B). In contrast, plasma Patient characteristics. The groups were well matched concentrations of the GLP-1 metabolite (9-36 amide) were for age, sex, and obesity (Table 1). There were no differ- significantly higher in the patients compared with the Ͻ ences in the HbA1c levels or in the fasting glucose concen- control subjects during and after the infusion (P 0.001; trations between both groups (P ϭ 0.38 and P ϭ 0.24, Fig. 5C). respectively). Serum creatinine concentrations and serum The incremental area under the curve as well as the urea concentrations were significantly higher in the CRI plasma half-lives and the metabolic clearance rates for patients compared with the control subjects (P Ͻ 0.001; intact GLP-1 were similar in the patients and the control Table 1). Cystatin C plasma concentrations, used as a subjects (P ϭ 0.67, P ϭ 0.13, and P ϭ 0.92, respectively; marker of the glomerular filtration rate, were significantly Table 2). However, calculation of the same parameters for higher in the CRI patients (P Ͻ 0.0001). Similarly, the the GLP-1 metabolite (9-36 amide) revealed significantly creatinine clearance was significantly reduced in the pa- higher values in the CRI patients than in the control tients with renal insufficiency (P Ͻ 0.0001). There was a subjects (P Ͻ 0.05; Table 2).

658 DIABETES, VOL. 53, MARCH 2004 J.J. MEIER AND ASSOCIATES

TABLE 2 Pharmacokinetic parameters for GLP-1 and GIP (intact and metabolites) during and after intravenous infusion in patients with renal insufﬁciency and control subjects Healthy control Parameter subjects Patients with CRI Signiﬁcance (P value)* Intact GLP-1 (7-36 amide) Area under the curve (pmol ⅐ lϪ1 ⅐ min) 584 Ϯ 97 648 Ϯ 110 0.67 Metabolic clearance rates (l/min) 2.42 Ϯ 0.45 2.35 Ϯ 0.54 0.92 Ϯ Ϯ t1/2 (min) 2.3 0.4 3.4 0.6 0.13 Distribution volume (l) 7.1 Ϯ 1.8 12.6 Ϯ 4.6 0.28 GLP-1 metabolite (9-36 amide) Area under the curve (pmol ⅐ lϪ1 ⅐ min) 2,287 Ϯ 293 5,203 Ϯ 776 0.0019 Metabolic clearance rates (l/min) 0.64 Ϯ 0.16 0.25 Ϯ 0.04 0.041 Ϯ Ϯ t1/2 (min) 3.3 0.4 5.3 0.8 0.029 Distribution volume (l) 2.7 Ϯ 0.4 1.9 Ϯ 0.3 0.12 Intact GIP (1-42) Area under the curve (pmol ⅐ lϪ1 ⅐ min) 1,200 Ϯ 358 1,003 Ϯ 177 0.63 Metabolic clearance rates (l/min) 3.18 Ϯ 0.62 2.02 Ϯ 0.58 0.77 Ϯ Ϯ t1/2 (min) 5.0 1.2 6.9 1.4 0.31 Distribution volume (l) 17.4 Ϯ 4.4 15.9 Ϯ 3.8 0.80 GIP metabolite (3-42) Area under the curve (pmol ⅐ lϪ1 ⅐ min) 1,847 Ϯ 311 2,965 Ϯ 356 0.031 Metabolic clearance rates (l/min) 1.56 Ϯ 0.27 0.93 Ϯ 0.21 0.088 Ϯ Ϯ t1/2 (min) 22.4 3.0 38.1 6.0 0.032 Distribution volume (l) 48.3 Ϯ 9.3 57.0 Ϯ 20.8 0.71 Data are means Ϯ SE. *ANOVA.

GIP infusion. During the intravenous infusion of GIP, and GLP-1. There was only an insignificant trend toward plasma glucose concentrations were only slightly but higher concentrations of both intact hormones in the CRI significantly lowered in both groups (P Ͻ 0.001; Fig. 6A). patients after the oral glucose load as well as during the Insulin and C-peptide concentrations increased during GIP peptide infusions (Figs. 2, 3, 5, and 7). This is in good infusion (P Ͻ 0.001; Fig. 6). There were no differences agreement with previous studies in pigs showing a high ϭ between the CRI patients and the control subjects (P degree of NH2-terminal degradation of both GLP-1 and GIP 0.33 for glucose, P ϭ 0.062 for insulin, and P ϭ 0.62 for in the hepato-portal bed and throughout the extremities C-peptide; Fig. 6). (27), where DPP IV is found in high concentrations asso- Plasma concentrations of intact GIP (1-42) increased ciated with hepatocytes and endothelial cells (28,29). The significantly during the infusion in both groups (P Ͻ 0.001; kidneys were found to extract ϳ70% of intact GLP-1 and Fig. 7B), but the plasma concentrations reached during only 25% of intact GIP (15,27), but because the kidneys and after the infusion period were similar in both groups receive only ϳ25% of the cardiac output, they would be (P ϭ 0.27). Different from the intact peptide, the GIP expected to contribute to the total body DPP IV–mediated metabolite (3-42) reached significantly higher plasma con- degradation of the intact incretin hormones by only 10– centrations in the CRI patients during and after the infu- 20%. This probably explains the modest increases in intact sion period (P ϭ 0.0033; Fig. 7C). GIP and GLP-1 observed in the present experiments. It is The metabolic clearance rates, the plasma half-lives, and interesting that renal extraction of intact GIP can still be the area under the curve for intact GIP (1-42) were similar detected after DPP IV inhibition, suggesting that glomeru- in both groups (Table 2). For the GIP metabolite (3-42), a lar filtration and possibly peritubular uptake are at least in significantly higher area under the curve and a longer part involved in the extraction of GIP (15). Studies in plasma half-life was calculated in the patients (P ϭ 0.031 isolated tubules have also suggested that the renal extrac- and P ϭ 0.032, respectively; Table 2). The same trend was tion of GLP-1 involves both glomerular filtration and obvious for the metabolic clearance rates (P ϭ 0.088; tubular uptake and catabolism (13), whereas detailed Table 2). studies of the renal handling of the related peptide glucagon have pointed to the involvement of peritubular uptake DISCUSSION in addition to glomerular filtration (30,31). The incretin hormones GIP and GLP-1 and their deriva- Although DPP IV–mediated degradation of GIP and tives are currently being discussed as a potential new GLP-1 takes place in different tissues, the present data treatment of type 2 diabetes because of their glucose- demonstrate that the kidneys are the major site of extrac- lowering potential (3,4,26). Because diabetes is frequently tion of GIP (3-42) and GLP-1 (9-36 amide). Accordingly, complicated by the development of renal insufficiency plasma levels of the incretin metabolites were found to be (17), it was important to assess the pharmacokinetic increased in response to oral glucose as well as during properties of GIP and GLP-1 in patients with impaired exogenous infusion in the patients with renal insufficiency renal function. (Figs. 2, 3, 5, and 7). The present data indicate that the kidneys are not the Considering derivatives of incretin hormones as a ther- primary site of the DPP IV–mediated metabolism of GIP apeutic approach for type 2 diabetes or obesity, the

DIABETES, VOL. 53, MARCH 2004 659 RENAL ELIMINATION OF GLP-1 AND GIP

FIG. 6. Plasma concentrations of glucose (A), insulin (B), and C- FIG. 7. Plasma concentrations of total GIP (1-42 plus split products peptide (C) during and after the intravenous infusion of GIP (1.0 pmol [A], intact GIP (1-42 [B]), and the GIP metabolite (total GIP minus ⅐ kg؊1 ⅐ min؊1) over 30 min in 10 patients with CRI (}) and 10 healthy intact GIP [C]) during and after the intravenous infusion of GIP (1.0 ؊1 ؊1 control subjects (F). Data are means ؎ SE. P values were calculated pmol ⅐ kg ⅐ min ) over 30 min in 10 patients with CRI (ࡗ) and 10 using repeated-measures ANOVA and denote differences between the healthy control subjects (F). Data are expressed as means ؎ SE. P groups (A), differences over time (B), and differences as a result of the values were calculated using repeated-measures ANOVA and denote interaction of group and time (AB). *P < 0.05 at individual time points differences between the groups (A), differences over time (B), and (one-way ANOVA). differences as a result of the interaction of group and time (AB). *P < 0.05 at individual time points (one-way ANOVA). consequences resulting from a slowed elimination in patients with impairments in renal function should be taken rises in total GLP-1 levels probably partly reflect the much into account. Although in the present study intact hor- higher clearance rate of the intact hormone versus its mone levels were only marginally affected by impairments metabolite. This finding is in agreement with previous in renal function, it cannot be excluded that elevated studies (18,36). The lower levels of intact hormone may plasma levels of the metabolites GIP (3-42) and GLP-1 also be explained by the high rate of conversion of intact (9-36 amide) may antagonize the respective intact peptides GLP-1 into its metabolite immediately after secretion, at their receptors. However, antagonistic properties of GIP catalyzed by the high concentrations of DPP IV in the (3-42) and GLP-1 (9-36 amide) have been shown only capillaries of the gut mucosa (37). However, there is during the administration of supraphysiological plasma evidence that under physiological circumstances, GLP-1 levels in some (7,32,33) but not all (34,35) studies. In the can interact with sensory nerve fibers before it enters the present experiments, there was only a trend toward lower circulation and is degraded by DPP IV (38). In this way, plasma levels of insulin after oral glucose ingestion as well L-cell secretion may have biological effects, e.g., on insulin as during the infusion of GIP and GLP-1 in the patients secretion, without changes in the circulating levels of (Figs. 1, 4, and 6). intact GLP-1. The low responses of intact GLP-1 plasma concentra- Previous studies based on immunoassays that were tions to oral glucose ingestion compared with the higher unable to distinguish between the intact and degraded

660 DIABETES, VOL. 53, MARCH 2004 J.J. MEIER AND ASSOCIATES forms of GLP-1 and GIP have revealed plasma half-lives (which are likely to be cleared with similar kinetics as the between 3 and 11 min for GLP-1 (39–41) and ϳ20 min for primary metabolite) in patients with renal impairment GIP (42,43). However, on the basis of the same immuno- should be taken into consideration if these analogues are assays that were used in the present study, Vilsbøll et al. used for therapy. (44) recently reported half-lives of ϳ2 min for intact GLP-1 and between 4 and 5 min for the GLP-1 metabolite after the ACKNOWLEDGMENTS intravenous bolus administration of different GLP-1 doses This study was supported by grants from the Deutsche in healthy volunteers. The investigators also found similar Forschungsgemeinschaft (DFG) (Na 203/6-1) and the For- elimination rates in patients with type 2 diabetes and in schungsfo¨ rderung Ruhr-Universita¨t Bochum Medizinische healthy control subjects (44). For GIP, we have previously Fakulta¨t (FoRUM) (F233/00). calculated a plasma half-life of 7.4 Ϯ 0.4 min after intra- We kindly acknowledge the expertise of Dr. Cedrik venous infusion in healthy volunteers (6). Meier for the calculations of the pharmacokinetic param- In the present study, the metabolic clearance rates of eters and statistical analyses. The excellent technical both GIP (3-42) and GLP-1 (9-36 amide) were lower than assistance of Birgit Baller and Lone Bagger is greatly those calculated for the respective intact hormones (Table acknowledged. 2). A similar observation was reported recently for the elimination of GLP-2, which is also subject to DPP IV REFERENCES degradation (45). However, this probably reflects the 1. Creutzfeldt W: The incretin concept today. Diabetologia 16:75–85, 1979 overall greater contribution of DPP IV–mediated degrada- 2. Creutzfeldt W, Nauck M: Gut hormones and diabetes mellitus. Diabetes tion in the elimination of the intact hormones, rather than Metab Rev 8:149–177, 1992 a lower effectiveness of the kidneys. It is interesting that 3. Meier JJ, Gallwitz B, Nauck MA: Glucagon-like peptide 1 and gastric inhibitory polypeptide: potential applications in type 2 diabetes mellitus. despite a high degree of sequence homology between BioDrugs 17:93–102, 2003 GLP-1 and GLP-2, the latter is metabolized much more 4. Drucker DJ: Biological actions and therapeutic potential of the glucagon- slowly (45). like peptides. Gastroenterology 122:531–544, 2002 The secretion of incretin hormones from endocrine K- 5. Kieffer TJ, McIntosh CH, Pederson RA: Degradation of glucose-dependent insulinotropic polypeptide and truncated glucagon-like peptide 1 in vitro and L-cells in the gut is controlled by the ingestion of and in vivo by dipeptidyl peptidase IV. Endocrinology 136:3585–3596, nutrients (46). It is unclear whether increased plasma 1995 concentrations of incretin hormones can inhibit their own 6. Deacon CF, Nauck MA, Meier JJ, Hu¨ cking K, Holst JJ: Degradation of secretion via feedback mechanisms. In the present exper- endogenous and exogenous gastric inhibitory polypeptide (GIP) in healthy and in type 2 diabetic subjects as revealed using a new assay for iments, despite higher rises in the plasma concentrations the intact peptide. J Clin Endocrinol Metab 85:3575–3581, 2000 of GIP (3-42) and GLP-1 (9-36 amide), there was no 7. Gault VA, Parker JC, Harriott P, Flatt PR, O’Harte FP: Evidence that the suppression of incretin secretion, as judged from the major degradation product of glucose-dependent insulinotropic polypep- intact hormone concentrations. However, because only tide, GIP(3-42), is a GIP receptor antagonist in vivo. J Endocrinol 175:525–533, 2002 the plasma levels of metabolites were significantly affected 8. Grandt D, Sieburg B, Sievert J, Schimiczek M, Becker U, Holtmann D, by impaired renal function, it cannot be excluded that Layer P, Reeve JR, Eysselein VE, Goebell H, Mueller M: Is GLP-1 (9-36) increased plasma levels of intact GIP or GLP-1 would lead amide an endogenous antagonist at GLP-1 receptors (Abstract)? Diges- to a feedback suppression of their own secretion. The tion 55:302, 1994 9. Knudsen LB, Pridal L: Glucagon-like peptide-1-(9-36) amide is a major latter hypothesis is supported by the observation of re- metabolite of glucagon-like peptide-1-(7-36) amide after in vivo adminis- duced total hormone levels and higher rises of intact tration to dogs, and it acts as an antagonist on the pancreatic receptor. incretin hormone levels after DPP IV inhibition in dogs Eur J Pharmacol 318:429–435, 1996 (35). 10. Juhl CB, Hollingdal M, Sturis J, Jakobsen G, Agersø H, Veldhuis J, Porksen N, Schmitz O: Bedtime administration of NN2211, a long-acting Disturbances of glucose homeostasis have previously GLP-1 derivative, substantially reduces fasting and postprandial glycemia been characterized in patients with renal insufficiency in type 2 diabetes. Diabetes 51:424–429, 2002 (47). Glucose intolerance frequently develops as a result of 11. Kim JG, Baggio LL, Bridon DP, Castaigne JP, Robitaille MF, Jette L, a high degree of insulin resistance, as well as elevated Benquet C, Drucker DJ: Development and characterization of a glucagon- plasma levels of glucagon, growth hormone, and cat- like peptide 1-albumin conjugate: the ability to activate the glucagon-like peptide 1 receptor in vivo. Diabetes 52:751–759, 2003 echolamines (47). In the present study, patients with 11a. Ahren B, Simonsson E, Larsson H, Landin-Olsson M, Torgeirsson M, impaired or diabetic oral glucose tolerance were excluded. Jansson PA, Sandqvist M, Bavenholm P, Efendic S, Eriksson JW, Dickin- As a result, glucose, insulin, and C-peptide concentrations son S, Holmes D: Inhibition of dipeptidyl peptidase IV improves metabolic were similar after glucose ingestion in both healthy sub- control over a 4-week study period in type 2 diabetes. Diabetes Care 25:869–875, 2002 jects and patients (Fig. 1), and homeostasis model assess- 12. Ruiz Grande C, Pintado J, Alarcon C, Castilla C, Valverde I, Lopez Novoa ment analysis indicated no differences in insulin sensitivity JM: Renal catabolism of human glucagon-like peptides 1 and 2. Can or B-cell function (Table 1). However, the significant J Physiol Pharmacol 68:1568–1573, 1990 difference in the C-peptide–to–insulin ratio calculated 13. Ruiz-Grande C, Alarco´ n C, Alcańtara A, Castila C, Lo´ pez Novoa JM, from the integrated areas under the curves does suggest a Villanueva-Pen˜ acarrillo ML, Valverde I: Renal catabolism of truncated glucagon-like peptide 1. Horm Metab Res 25:612–616, 1993 reduced elimination of C-peptide in CRI patients. 13a. Sirinek KR, O’Dorisio TM, Gaskill HV, Levine BA: Chronic renal failure: In conclusion, the present data provide evidence that effect of hemodialysis on gastrointestinal hormones. Am J Surg 148:732– the kidneys are less important for the initial degradation 735, 1984 (inactivation) of GIP and GLP-1. In contrast, the extraction 13b. Orskov C, Andreasen J, Holst JJ: All products of proglucagon are elevated in plasma from uremic patients. J Clin Endocrinol Metab 74:379–384, of the metabolites GIP (3-42) and GLP-1 (9-36 amide) 1992 primarily depends on renal function. Therefore, altered 14. Deacon CF, Johnsen AH, Holst JJ: Degradation of glucagon-like peptide-1 elimination of DPP-IV–resistant analogues of GLP-1 by human plasma in vitro yields an N-terminally truncated peptide that is

DIABETES, VOL. 53, MARCH 2004 661 RENAL ELIMINATION OF GLP-1 AND GIP

a major endogenous metabolite in vivo. J Clin Endocrinol Metab 80:952– J: High potency antagonists of the pancreatic glucagon-like peptide-1 957, 1995 receptor. J Biol Chem 272:21201–21206, 1997 15. Deacon C, Danielsen P, Klarskov L, Olesen M, Holst JJ: Dipeptidyl 33. Wettergren A, Wojdemann M, Holst JJ: The inhibitory effect of glucagon- peptidase IV inhibition reduces the degradation and clearance of GIP and like peptide-1 (7-36)amide on antral motility is antagonized by its N- potentiates its insulinotropic effects in anesthetized pigs. Diabetes 50: terminally truncated primary metabolite GLP-1 (9-36)amide. Peptides 1588–1597, 2001 19:877–882, 1998 16. Deacon CF, Highes TE, Holst JJ: Dipeptidyl peptidase IV inhibition 34. Vahl TP, Paty BW, Fuller BD, Prigeon RL, D’Alessio DA: Effects of potentiates the insulinotropic effect of glucagon-like peptide 1 in the GLP-1-(7-36)NH2, GLP-1-(7-37), and GLP-1-(9-36)NH2 on intravenous anesthetized pig. Diabetes 47:764–769, 1998 glucose tolerance and glucose-induced insulin secretion in healthy hu- 17. Ritz E, Orth SR: Nephropathy in patients with type 2 diabetes mellitus. mans. J Clin Endocrinol Metab 88:1772–1779, 2003 N Engl J Med 341:1127–1133, 1999 35. Deacon CF, Wamberg S, Bie P, Hughes TE, Holst JJ: Preservation of 18. Vilsbøll T, Krarup T, Deacon CF, Madsbad S, Holst JJ: Reduced postpran- active incretin hormones by inhibition of dipeptidyl peptidase IV sup- dial concentrations of intact biologically active glucagon-like peptide 1 in presses meal-induced incretin secretion in dogs. J Endocrinol 172:355– type 2 diabetic patients. Diabetes 50:609–613, 2001 362, 2002 19. Meier JJ, Hu¨ cking K, Holst JJ, Deacon C, Schmiegel W, Nauck MA: 36. Vilsboll T, Krarup T, Madsbad S, Holst JJ: Both GLP-1 and GIP are Reduced insulinotropic effect of gastric inhibitory polypeptide in ﬁrst- insulinotropic at basal and postprandial glucose levels and contribute degree relatives of patients with type 2 diabetes. Diabetes 50:2497–2504, nearly equally to the incretin effect of a meal in healthy subjects. Regul 2001 Pept 114:115–121, 2003 20. Deacon CF, Nauck MA, Toft-Nielsen M, Pridal L, Willms B, Host JJ: Both 37. Hansen L, Deacon CF, Orskov C, Holst JJ: Glucagon-like peptide-1-(7- subcutaneously and intravenously administered glucagon-like peptide 1 36)amide is transformed to glucagon-like peptide-1-(9-36)amide by didep-

are rapidly degraded from the NH2-terminus in type 2-diabetic patients tidyl peptidase IV in the capillaries supplying the L cells of the porcine and in healthy subjects. Diabetes 44:1126–1131, 1995 intestine. Endocrinology 140:5356–5363, 1999 21. Tan GD, Lewis AV, James TJ, Altmann P, Taylor RP, Levy JC: Clinical 38. Balkan B, Li X: Portal GLP-1 administration in rats augments the insulin usefulness of cystatin C for the estimation of glomerular filtration rate in response to glucose via neuronal mechanisms. Am J Physiol Regul Integr type 1 diabetes: reproducibility and accuracy compared with standard Comp Physiol 279:R1449–R1454, 2000 measures and iohexol clearance. Diabetes Care 25:2004–2009, 2002 39. Ørskov C, Wettergren A, Holst JJ: Biological effects and metabolic rates 22. Laterza OF, Price CP, Scott MG: Cystatin C: an improved estimator of of glucagonlike peptide-1 7-36 amide and glucagonlike peptide-1 7-37 in glomerular filtration rate? Clin Chem 48:699–707, 2002 healthy subjects are indistinguishable. Diabetes 42:658–661, 1993 23. Finney H, Newman DJ, Gruber W, Merle P, Price CP: Initial evaluation of 40. Kreymann B, Williams G, Ghatei MA, Bloom SR: Glucagon-like peptide-1 cystatin C measurement by particle-enhanced immunonephelometry on [7-36]: a physiological incretin in man. Lancet 2:1300–1304, 1987 the Behring nephelometer systems (BNA, BN II). Clin Chem 43:1016– 41. Schjoldager BT, Mortensen PE, Christiansen J, Ørskov C, Holst JJ: GLP-1 1022, 1997 (glucagon-like peptide 1) and truncated GLP-1, fragments of human 24. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner proglucagon, inhibit gastric acid secretion in humans. Dig Dis Sci RC: Homeostasis model assessment: insulin resistance and ␤-cell func- 34:703–708, 1989 tion from fasting plasma glucose and insulin concentrations in man. 42. Sarson DL, Hayter RC, Bloom SR: The pharmacokinetics of porcine Diabetologia 28:412–419, 1985 glucose-dependent insulinotropic polypeptide (GIP) in man. Eur J Clin 25. Cockcroft DW, Gault MH: Prediction of creatinine clearance from serum Invest 12:457–461, 1982 creatinine. Nephron 16:31–41, 1976 43. Nauck M, Schmidt WE, Ebert R, Strietzel J, Cantor P, Hoffmann G, 26. Holst JJ: Glucagon-like peptide 1 (GLP-1): an intestinal hormone, signal- Creutzfeldt W: Insulinotropic properties of synthetic human gastric ling nutritional abundance, with an unusual therapeutic potential. Trends inhibitory polypeptide in man; interactions with glucose, phenylalanine, Endocrinol Metab 10:229–235, 1999 and cholezystokinin-8. J Clin Endocrinol Metab 69:654–662, 1989 27. Deacon CF, Pridal L, Klarskov L, Olesen M, Holst JJ: Glucagon-like 44. Vilsboll T, Agerso H, Krarup T, Holst JJ: Similar elimination rates of peptide 1 undergoes differential tissue-specific metabolism in the anes- glucagon-like peptide-1 in obese type 2 diabetic patients and healthy thetized pig. Am J Physiol Endocrinol Metab 271:E458–E464, 1996 subjects. J Clin Endocrinol Metab 88:220–224, 2003 28. Lojda Z: Studies on dipeptidyl(amino)peptidase IV (glycyl-proline-naph- 45. Hartmann B, Harr MB, Jeppesen PB, Wojdemann M, Deacon CF, thylamidase). II. Blood vessels. Histochemistry 59:153–166, 1979 Mortensen PB, Holst JJ: In vivo and in vitro degradation of glucagon-like 29. Mentlein R: Dipeptidyl-peptidase IV (CD26)—role in the inactivation of peptide-2 in humans. J Clin Endocrinol Metab 85:2884–2888, 2000 regulatory peptides. Regul Pept 85:9–24, 1999 46. Ørskov C, Wettergren A, Holst JJ: Secretion of the incretin hormones 30. Peterson DR, Green EA, Oparil S, Hjelle JT: Transport and hydrolysis of glucagon-like peptide-1 and gastric inhibitory polypeptide correlates with glucagon in the proximal nephron. Am J Physiol 251:F460–F467, 1986 insulin secretion in normal man throughout the day. Scand J Gastroen- 31. Deacon CF, Kelstrup M, Trebbien R, Klarskov L, Olesen M, Holst JJ: terol 31:665–670, 1996 Differential regional metabolism of glucagon in anesthetized pigs. Am J 47. Wright AD, Lowy C, Fraser TR, Spitz IM, Rubenstein AH, Bersohn I: Physiol Endocrinol Metab 285:E551–E560, 2003 Serum-growth hormone and glucose intolerance in renal insufficiency. 32. Montrose Rafizadeh C, Yang H, Rodgers BD, Beday A, Pritchette LA, Eng Lancet 2:798–801, 1968

662 DIABETES, VOL. 53, MARCH 2004