Insulin Degradation by Adipose Tissue Is Increased in Human Obesity

White adipose tissue samples from obese and lean patients were used for the estimation of insulin protease and insulin:glutathione transhydrogenase using 125I-labeled insulin. There was no activity detected in the absence of reduced glutathione, which indicates that insulin is cleaved in human adipose tissue through reduction of the disulfide bridge between the chains. Obese patients showed higher transhydrogenase activity (per U tissue protein wt, per U tissue wt, and in the total adipose tissue mass) than the lean group. There is a significant correlation between the activity per U tissue wt, and protein and total activity in the whole adipose tissue with respect to body mass index, with a higher activity in obese patients. The potential of insulin cleavage by adipose tissue in obese patients was a mean 5.6-fold higher than that in controls. The coexistence of high insulinemia and high cleavage capability implies that insulin secretion and turnover are increased in the obese. Thus, white adipose tissue may be crucial in the control of energy availability through modulation of insulin cleavage.


ABSTRACT
White adipose tissue samples from obese and lean patients were used for the estimation ofinsulin protease and insulin:glutathione transhydrogenase using 125 1-labeled insulin.There was no activity detected in the absence of reduced glutathione, which indicates that insulin is cleaved in human adipose "tissue through reduction of the disulfide bridge between the chains.O bese patients showed higher transhydrogenase activity (per U tissue protein wt, per U tissue wt, and in the total adipose tissue mass) than the lean group.There is a significant correlation between the activity per U tissue O BESITY is often characterized by hyperinsulinemia, which has a direct bearing on the increased lipid accrual in obese individuals, insulin resistance, and lateonset diabetes (1).Because hepatic disposal of insulin is decreased in obesity (2), we attempted to investiga te whether this trend was also observed in adipose tissue, as its large mass in the obese may influence peripheral insulin degradation to a greater extent than in individuals with normal fat depots.
The half-life of circulating insulin is short, in the range of minutes, although its functional tumover may be somewhat longer in normal weight humans, beca use it is rapidly taken up and inactivated by liver (3, 4), adipose tissue (5), and muscle (5, 6).Insulin may be inactivated through intern_ alization after interacting with its receptor ( 7), but most of it is cleaved by specific systems through a selective proteolytic pathway (4) or by means of the more commonly accepted reduction of the disulfide bridge between the a-and J3-chains (3,5).This reaction is catalyzed by insulin:glutathione transhydrogenase (EC 1.8.4.2) (3, 5), with glutathione providing the reducing power.
• respect to body mass index, with a higher activity in obese patients.
The potential ofinsulin cleavage by adipose tissue in obese patients was a mean 5.6-fold higher than that in controla.The coexistence of high insulinemia and high cleavage capability implies that insulin secretion and turnover are increased in the o bese.Thus, white adipose tissue may be crucial in the control of energy availability through modulation ofinsulin cleavage.(J Clin Endrocrinol Metab 80:693-695, 1995)

Materials and Methods
Samples of se abdominal white adipose tissue were obtained from patients undergoing laparotomy because of hernia (n = S) or cholecystectomy (n = 3), control group; five men and three women; body mass index (BMI), <27 kg/m 2 ; or obese patients (BMI, >30 kg/m 2 ) subjected to laparotomy ~cause of cholecystectomy (n = 1) or vertical-banded gastroplasty (n = 9; seven women and three men).None of the women included in the experiment had reached menopause; patients receiving medication were not used in the study.Informed consent of the patients and of the Clinical Assays Committee was obtained prior to the undertaking of the experiments.The parameters defining the two groups of patients studied are presented in Table l .Blood was extracted in the morning after an ovemight fast; the •samples were used for the measurement of glycemia (¡:!) and insulinemia (through radioimmunoanalysis; Diagnostic Products Corp., Los Angeles, CA).Pieces of adipose tissue ( -2 g) were dissected from the abdominal se fat masses near the surgical wound; they were immediately frozen in liquid nitro gen after extraction and kept under liquid nitrogen or in deep freeze ( -80 C) until processed.Aliquots of about 0.5 g tissue were weighed and homogenized with a Polytron tissue disrupter (Kinematica, Luzern, Switzerland) in 10 vol chilled 50 mmol/L phosphate buffer, pH 7.5, containing 250 mmol/L sucrose.The homogenates were centrifuged for 10 min at 5000 X g, and the supernatant layer of fat and the pellet were discarded; the clear homogenate was used for the estimation of insulin:glutathione transhydrogenase, u sing 125 !-labeled insulin (Amersham, UK) with the method of Thomas and Varandani (9).A set of determinations was carried out by omitting the reduced glutathione, as this experimental set-up has been used to measure insulin protease (10).Protein in homogenates was estimated with the Lowry method (11).

693
The degree of mterrelationship between data was estimated by establishing the linear correlation coefficient for paired parameters.Comparison between groups was performed with Student's t test.