Indinavir increases glucose production in healthy HIV-negative men

In the absence of HIV infection, changes in adipose tissue and lipid levels, HIV protease inhibitor therapy increases fasting glucose levels, suggestive of hepatic insulin resistance. After 4 weeks of indinavir treatment in nine HIV-negative healthy men, fasting glucose production and glycogenolysis were significantly increased. During the euglycemic hyperinsulinemic clamp, indinavir blunted the ability of insulin to suppress glucose production. Therefore, indinavir worsens hepatic insulin sensitivity, increasing endogenous glucose production. Highly active antiretroviral therapy regimens containing a protease inhibitor (PI) have been associated with insulin resistance, hyperglycemia, and frank diabetes [1]. We previously reported that 4 weeks of treatment with indinavir in HIV-negative healthy men caused a 17% reduction in insulin sensitivity as assessed by euglycemic hyperinsulinemic clamp, in the absence of significant changes in body composition or lipid profiles [2]. However, we also found increases in fasting glucose and fasting insulin, which cannot be caused by the induction of peripheral insulin resistance, as insulin-mediated glucose disposal is minimal in the fasting state [3]. Hepatic glucose production accounts for approximately 80% of the fasting whole body (endogenous) glucose production [4]. An increase in hepatic glucose production would be the likely etiology for the increase in fasting glucose levels. Therefore, we determined the rates of overall endogenous glucose production, gluconeogenesis, and glycogenolysis using stable isotopic tracer methods during indinavir treatment of HIV-negative men. The study was approved by the Committee on Human Research of the University of California, San Francisco. As previously described, individuals were admitted to the General Clinical Research Center for baseline studies, including fasting insulin and glucose levels, euglycemic hyperinsulinemic clamp, and body composition while on a constant metabolic diet [2]. Subjects were then discharged, given indinavir 800 mg every 8 h, and re-admitted after 4 weeks for repeat studies. Nine of the ten originally studied HIV-negative men also participated in the tracer protocol to determine their glucose kinetics; for one subject, isotopes were unavailable to measure these metabolic pathways. On the evening before the euglycemic hyperinsulinemic clamp, catheters were placed in the forearm veins of each arm for the infusion of isotopic solutions and periodic blood sampling. At 04.30 hours the following morning, a continuous infusion of [6,6-2H2] glucose (3.6 mg/kg body mass per hour) and [2-13C] glycerol (15 mg/kg lean body mass per hour) was begun. Stable isotopes were purchased from Cambridge Isotope Laboratories (Andover, MA, USA) and Isotec (Miamisburg, OH, USA). Steady state blood sampling was performed every 10 min between 08.00 and 08.30 hours (fasting measurements) and during the final 30 min of a 180-min euglycemic hyperinsulinemic clamp begun at 09.00 hours [2]. During the clamp, we used the labeled exogenous glucose infusate technique. The rate of appearance of glucose (Ra glucose) and gluconeogenesis were determined using gas chromatography–mass spectrometry with the traditional dilution techniques [5] and by mass isotopic distribution analysis as described previously [6–8]. Glycogenolysis was calculated by subtracting the rate of gluconeogenesis from that of endogenous glucose production. The last dose of indinavir was given to the patient before the start of the clamp as described [2]. Statistical analyses were performed using Sigma Stat version 2.03 (SPSS, Inc., San Rafael, CA, USA). Data are presented as mean ± SEM. A two-tailed P value less than 0.05 was considered to be statistically significant. Fasting glucose and insulin increased, and insulin-mediated glucose disposal decreased significantly by 12% at the end of 4 weeks of indinavir (Table 1). There were no changes in lipid levels, lipoprotein profile, or adipose tissue by computed tomography. Endogenous fasting glucose production (Ra glucose) and glycogenolysis increased significantly (Table 1). There was a trend towards an increase in gluconeogenesis, which was of a similar relative magnitude as the increases in endogenous glucose production and glycogenolysis.Table 1: Glucose metabolism and body composition at pre-treatment and at the end of 4 weeks of indinavir therapy in nine subjects.During the euglycemic hyperinsulinemic clamp, steady-state serum insulin levels (∼500 pmol/l) were achieved after 30 min. Hyperinsulinemia during the clamp was associated with a marked suppression of endogenous glucose production, gluconeogenesis, and glycogenolysis (Table 1) in comparison with rates measured under fasting conditions. This suppressive effect of insulin was evident before as well as during indinavir treatment. However, at the end of 4 weeks of indinavir treatment, there was a significant blunting of the insulin-induced suppression of endogenous glucose production, gluconeogenesis, and glycogenolysis. These results indicate that indinavir has effects on endogenous glucose production that are distinct from its effects on peripheral insulin-mediated glucose metabolism. In the fasting state, indinavir increases overall glucose production through a significant increase in glycogenolysis. This magnitude of increase in glucose production levels could explain the increased fasting glucose levels seen during the original study. Endogenous glucose production is the primary determinant of fasting glucose levels [9]. Both fasting glucose levels and fasting glucose production increased to a similar degree with indinavir treatment, consistent with the primary role of glucose production in determining circulating glucose concentrations in the fasting state. In this study, one subject developed impaired fasting glucose according to the 1997 American Diabetes Association criteria [10], and one subject developed impaired fasting glucose by the 2003 American Diabetes Association criteria [11]. Fasting glucose should be monitored in HIV-infected patients treated with indinavir. In the hyperinsulinemic state, the insulin suppression of glucose production was blunted by indinavir, as evidenced by a less effective suppression of glucose production during the clamp after 4 weeks of indinavir. This blunted suppression of glucose production during hyperinsulinemia was a consequence of proportionally higher rates of both glycogenolysis and gluconeogenesis. These results are consistent with the induction of hepatic insulin resistance. The mechanism for the indinavir-induced increase in endogenous glucose production is unknown. Murata et al. [12] reported that indinavir caused an acute blockade of GLUT4 in vitro. Supporting this hypothesis, we also demonstrated that a single dose of indinavir acutely decreased insulin-mediated glucose disposal [13]. However, GLUT4 transporters are not found in the liver and cannot explain the increase in endogenous glucose production. Increased endogenous glucose production during fasting and hyperinsulinemic states has several implications. The mechanism for impaired glucose tolerance with PI such as indinavir is therefore multifactorial; indinavir induces peripheral insulin resistance [2], impairs pancreatic insulin secretion [14], and increases endogenous glucose production. The assumption that all PI increase endogenous glucose production should be approached with caution. Previously, we showed that indinavir induced peripheral insulin resistance but no changes in lipid levels [2], whereas lopinavir/ritonavir caused hyperlipidemia, but no changes in peripheral insulin sensitivity or fasting glucose or insulin levels [15]. Each PI should thus be examined for its effects on endogenous glucose production, peripheral insulin resistance and insulin secretion. Acknowledgements The authors would like to thank Wendy Lee, Andrew Young, Barbara Chang, Joy Hirai, and the GCRC nursing and dietary staff for their technical assistance. Sponsorship: This work was supported by the National Institutes of Health (DK54615 and DK66999) and the University-wide AIDS Research Program (ID01-SF-014). This study was also partly funded by Merck, Inc., who makes indinavir. Merck played no role in accruing or analysing the data or in the writing of the paper. All studies were conducted in the General Clinical Research Center at San Francisco General Hospital with support by the National Center for Research Resources, National Institutes of Health (RR00083).