0013-7227/90/1275-2418$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society

Vol. 127, No. 5 Printed in U.S.A.

Hypoglycemic Potency and Metabolic Clearance Rate of Intravenously Administered Human Proinsulin and Metabolites* HARTMUT TILLILt, BRUCE H. FRANK, ALLEN H. PEKAR, CHRISTOPH BROELSCH, ARTHUR H. RUBENSTEIN, AND KENNETH S. POLONSKY$ Departments of Medicine (H.T., A.H.R., K.S.P.) and Surgery (C.B.), University of Chicago, Pritzker School of Medicine, Chicago, Illinois 60637; and Lilly Research Laboratories (B.H.F., A.H.P.), Indianapolis, Indiana 46285

tials to the respective decay curves. The MCR of HPI (3.3 ± 0.1 ml/kg -min) was significantly lower (P < 0.05) than that of des(64,65)HPI (6.4 ± 0.6 ml/kg-min), but was not significantly different from that of des-(31,32)HPI (3.8 ± 0.4 ml/kg-min). The MCR of biosynthetic insulin (17.2 ± 1.8 ml/kg-min), as measured in three of the dogs, was higher than that of HPI or the two metabolites. The blood glucose-lowering ability (defined as nadir glucose/fasting glucose, expressed as a percentage) of des-(64,65)HPI (49.3 ± 5.0%) was significantly greater (P < 0.05) than that of intact HPI (87 ± 2.2%), and the glucoselowering ability of des-(31,32)HPI (75.2 ± 3.8%) was intermediate. In conclusion, HPI metabolites are more biologically active than intact HPI. The extent of in vivo conversion of proinsulin to metabolites may enhance the biological activity of proinsulin and, thus, have physiological, pathophysiological, and therapeutic significance. (Endocrinology 127: 2418-2422, 1990)

ABSTRACT. Since circulating proinsulin has been suggested to be important in the pathogenesis of noninsulin-dependent diabetes, and biosynthetic human proinsulin (HPI) may have a therapeutic role in patients with diabetes mellitus, the biological activity of proinsulin metabolites is of potential significance. Moreover, recent studies have suggested that the majority of circulating proinsulin immunoreactivity consists of metabolites. We, therefore, compared the blood glucose-lowering ability and MCR of the two proinsulin metabolites des-(31,32)HPI and des(64,65)HPI with intact HPI in seven anesthetized dogs after an overnight fast. Intravenous bolus injections of 12.5 jug HPI/kg BW and equimolar amounts of des-(31,32)HPI and des(64,65)HPI were given on three separate occasions. In addition to blood glucose, des-(31,33)HPI, des-(64,65)HPI, and HPI were measured using an insulin RIA and peptide-specific standard curves. Kinetic parameters were derived by fitting two exponen-

I

NTEREST in the study of proinsulin stems from the potential importance of this peptide in the pathophysiology and treatment of diabetes. It has been known for some years that circulating proinsulin concentrations are elevated in noninsulin-dependent diabetes (NIDDM), and it has been proposed that this finding may be used as a marker of /3-cell injury (1). Received March 23, 1990. Address all correspondence and requests for reprints to: Kenneth S. Polonsky, M.D., Department of Medicine, University of Chicago, Box 435, 5841 South Maryland Avenue, Chicago, Illinois 60637. * This work was supported by NIH Grants DK-31842, DK-13941, and DK-20595 (Diabetes Research and Training Center) and a grant from Eli Lilly Co. This study was presented in part (in abstract form) at the 100th Annual National Meeting of the Association of American Physicians, San Diego, CA, May 1-4, 1987 (Clin Res 35:659A, 1987). t On leave from the Division of Gastroenterology and Endocrinology, Department of Internal Medicine, University of Gottingen, Gottingen, Germany, and supported by the Deutsche Forschungsgemeinschaft (Training Grant Ti 154/1-1) and the Deutsche Diabetes-Gesellschaft (Junior Science Award). X Recipient of a Research Career Development Award from the NIH (DK-01234).

The ability to produce biosynthetic human proinsulin (HPI) by application of recombinant DNA technology (2) has raised the possibility that it may be useful in the therapy of diabetes mellitus (3). Proinsulin has a prolonged plasma half-life compared to human insulin (HI) and appears to exert a greater effect on the liver than on peripheral insulin-sensitive tissues (4). A complete understanding of the biological activity of proinsulin immunoreactivity requires knowledge of its metabolism, in view of the recent suggestion that concentrations of circulating proinsulin metabolites may be elevated in NIDDM (5). Previous studies from our laboratory (6) have demonstrated that sc administration of proinsulin to patients with type 1 (insulin-dependent) diabetes can result in the formation of biologically active proinsulin metabolites. After iv infusion, no significant processing of proinsulin to metabolites was detected, and no processing to insulin was observed by either administration route. Data on the biological activity of proinsulin metabolites compared to proinsulin are only available in

2418

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 00:58 For personal use only. No other uses without permission. . All rights reserved.

HUMAN PROINSULIN AND METABOLITES rats after sc injection of the peptides (7), and the in vivo metabolic and kinetic characteristics of proinsulin metabolites have not yet been fully described (8, 9). We, therefore, assessed the acute blood glucose-lowering ability and the MCR of intact human proinsulin, human des-(31,32)proinsulin [des-(31,32)HPI], and human des(64,65)proinsulin [des-(64,65)HPI] after equimolar iv bolus injections of the pure peptides in fasted dogs.

2419

Calculations The individual plasma disappearance curves of proinsulin and metabolites were resolved into the sum of two exponentials by nonlinear least squares regression analysis. The insulin decay curves were analyzed for the first hour after the bolus injection. The MCR of the peptides was calculated as the ratio of the amount of peptide in the bolus to the area under the peripheral concentration curve of the peptide integrated to infinity.

Materials and Methods Derivation of proinsulin metabolites

Statistical analysis

Des-(31,32)HPI and des-(64,65)HPI were derived from biosynthetic human proinsulin (produced by recombinant DNA technology, Eli Lilly Co., Indianapolis, IN) by in vitro partial enzymatic cleavage, as previously described (6, 10). This involved the use of trypsin and carboxypeptidase, followed by isolation using preparative HPLC. Their identity was confirmed by amino acid analyses, peptide mapping, and amino acid sequencing.

All results are expressed as the mean ± SEM. Nonlinear least squares regression analysis of the plasma disappearance curves was performed using the BMDP 3R program (BMDP Statistical Software, Los Angeles, CA). Means were compared by the paired two-tailed t test or two-way analysis of variance, with subsequent Bonferroni (Dunn) t tests where indicated. Differences were regarded as being statistically significant if the corresponding P value was less than or equal to 0.05. Data analysis was performed using the Statistical Analysis System (SAS Version 6 Edition for Personal Computers, SAS Institute, Cary, NC).

Experimental protocol Studies were performed on seven male mongrel dogs (body weight, 19.3 ± 1.2 kg; range, 14.7-22.5). The animals were fed once daily, had free access to drinking water, and were kept on a 12-h light, 12-h dark cycle. After an overnight fast, anesthesia was induced with 200 mg/kg thiamylal sodium (Surital, Parke-Davis, Morris Plains, NJ) and maintained by 2% halothane using a respirator. A sampling catheter was inserted into a leg artery, and an iv catheter for injection of the peptides was placed into the opposite leg. After a baseline period of 20 min, each dog received iv bolus injections of 12.5 Mg/kg biosynthetic human proinsulin or equimolar amounts of des-(31,32)HPI and des-(64,65)HPI on three separate occasions. The proinsulin dose was chosen in order to elicit only a mild hypoglycemic response. In addition, three of the seven dogs received an iv bolus injection of 0.05 U/kg regular biosynthetic HI (1.8 ^gm/kg; Humulin R, Eli Lilly Co.) on another occasion. Arterial blood samples were taken at 10, 5, and 0 min before the bolus injection and 1, 2, 3, 4, 5, 6, 7, 8, 9,10, 11, 14,17, 20, 25, 30, 35, 40, 45, 50, 55, 60, 70, 80, 90,100, 110, 120, 130, 140, 150, 160, 180, 190, 200, 220, and 240 min after the bolus injection. Sample collection and analytical methods Blood samples for proinsulin, metabolite, or insulin determinations were allowed to clot at room temperature, and the serum was stored at -20 C until assayed. Serum insulin immunoreactivity was assayed by a double antibody technique, as previously described (11). The intra- and interassay coefficients of variation were 4% and 6%, respectively. Proinsulin, des(31,32)HPI, and des-(64,65)HPI in serum were measured with the same insulin RIA, using peptide-specific standard curves. Plasma glucose was measured immediately with the Beckman Glucose Analyzer 2 (Beckman Instruments, Fullerton, CA), using the glucose oxidase method.

Results Biological activity The plasma glucose concentrations in the basal period and during the 240 min after the administration of HPI, des-(31,32)HPI, or des-(64,65)HPI are shown in Fig. 1. All three peptides produced a clear-cut hypoglycemic effect, with a nadir 20 min after the injection of HPI and nadirs 30 min after the injection of both des-(31,32)HPI and des-(64,65)HPL After the injection of HPI, glucose decreased from a fasting concentration of 99.2 ± 3.2 to 86.4 ± 3.9 mg/dl at 20 min. After des-(31,32)HPI, glucose decreased slightly more from 103.2 ± 4.2 mg/dl at baseline to 77.4 ± 4.9 mg/dl at 30 min. The greatest hypoglycemic effect was seen after des-(64,65)HPI, where glucose fell from 94.6 ± 2.5 mg/dl fasting to 46.4 ± 4.4 mg/dl at 30 min. For all three peptides, the nadir glucose was

^h-h^kiz4 —•proinsulin —«des (31,32) proinsulin --•des (64,65) proinsulin

Time (mins)

FIG. 1. Arterial plasma glucose concentrations in the basal state and after iv bolus injections of equimolar amounts of HPI, des-(31,32)HPI, and des-(64,65)HPI.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 00:58 For personal use only. No other uses without permission. . All rights reserved.

HUMAN PROINSULIN AND METABOLITES

2420

significantly lower than the fasting glucose (P < 0.05). Thereafter, the glucose concentration increased, but it did not return to baseline after the injection of des(64,65)HPI (78.0 ± 4.9 mg/dl at 240 min vs. 94.6 ± 2.5 mg/dl at baseline; P < 0.05). When the glucose concentration after bolus injection of the peptides was expressed as a percentage of the corresponding mean basal glucose concentration, glucose fell to 87 ± 2.2% of the basal value 20 min after HPI, to 75.2 ± 3.8% of the basal value 30 min after des-(31,32)HPI, and to 49.3 ± 5.0% of the basal value 30 min after des-(64,65)HPI. While the relative decrease after des-(64,65)HPI was significantly greater (P < 0.05) than the decrease after des-(31,32)HPI and HPI, the difference between des-(31,32)HPI and HPI became insignificant after correction for multiple comparisons.

Endo • 1990 Vol 127 • No 5

± 6.2 fmol/ml 240 min after the injection. The mean concentration of des-(64,65)HPI 240 min after injection was significantly lower (P < 0.05) than that of HPI and des-(31,32)HPI. The MCRs of HPI, des-(64,65)HPI and des-(31,32)HPI are given in Fig. 3. The MCR of des(64,65)HPI (6.4 ± 0.6 ml/kg-min) was significantly higher (P < 0.05) than those of des-(31,32)HPI (3.8 ± 0.4 ml/kg-min) and HPI (3.3 ± 0.1 ml/kg-min). The MCR of des-(31,32)HPI was not significantly different from that of HPI. For comparison, the MCR of biosynthetic HI was calculated from data obtained in three of the seven dogs. The MCR of HI was 17.2 ± 1.8 ml/kg-min and was higher than that of the other three peptides. Thus, the MCR of HI was approximately 5 times higher than the MCR of HPI and approximately 2.5 times higher than

thatofdes-(64,65)HPI. Kinetic properties

Discussion

Figure 2 shows the time course of the disappearance of iv injected HPI and metabolites from serum. The concentration of HPI fell from 21,814 ± 1,272 fmol/ml 2 min after the bolus injection to 225.4 ± 20.5 fmol/ml 240 min after the injection, and the concentration of des(31,32)HPI decreased from 24,736 ± 1,646 fmol/ml at 2 min to 151.9 ± 30.4 fmol/ml at 240 min. The mean concentration of des-(64,65)HPI was lower than that of HPI and des-(31,32)HPI at all time points, and fell from 17,329 ± 1,282 fmol/ml 2 min after the injection to 49.4 25000-1

• — • proinsulin 20000-

• • — • des (31,32) proinsulin » - - • ' d e s (64.65) proinsulin

15000-

The physiological importance of proinsulin has long been viewed as minor, although proinsulin is secreted from the pancreatic B-cells along with insulin and Cpeptide and under basal conditions constitutes 10-20% of the measurable circulating immunoreactive insulin in normal man (12). In the preparation of animal insulins for the treatment of diabetes, proinsulin was considered a contaminant to be removed (13, 14). Through recombinant DNA-technology it has now become possible to prepare large quantities of human proinsulin (2), and this has led to a renewed interest in the possible physiological role and clinical usage of this hormone (3, 4, 15-17). Recent studies have revealed that biosynthetic human proinsulin has intrinsic biological activity and is an insulin agonist with lower biological potency than insulin. It has been demonstrated that proinsulin has a longer plasma half-life and a lower MCR than insulin (3, 4). In p

Hypoglycemic potency and metabolic clearance rate of intravenously administered human proinsulin and metabolites.

Since circulating proinsulin has been suggested to be important in the pathogenesis of noninsulin-dependent diabetes, and biosynthetic human proinsuli...
666KB Sizes 0 Downloads 0 Views