Human Serum Attenuates the Activity of Protease Inhibitors toward Wild-Type and Mutant Human Immunodeficiency Virus

Abstract

The potency of therapeutic regimens containing human immunodeficiency virus (HIV) protease inhibitors is related to the ability to maintain concentrations of drug in the plasma of patients that are sufficient for blocking viral replication. The estimation of concentrations required for in vivo activity using in vitro assays is complicated by the fact that extensive binding of many protease inhibitors to serum proteins attenuates their antiviral potency. To provide insight into the relative in vivo potency of current protease inhibitors, we assayed their in vitro activity against wild-type and mutant HIV in the presence of human serum (HS). Using this assay, ABT-378, a new protease inhibitor with trough levels in humans far in excess of the EC50 in the presence of 50% HS, was identified. The antiviral activity of ABT-378 was only modestly attenuated by HS, in contrast to ritonavir, saquinavir, and nelfinavir. Examination of the effect of individual serum components suggested that the activity of ABT-378 is affected predominantly by binding to alpha1-acid glycoprotein (AGP) while the activity of ritonavir is modulated by both AGP and albumin. The method described here may provide insight into the in vivo potency of protease inhibitors and be useful for the preclinical evaluation and selection of new protease inhibitors for clinical studies.

Long-term effects of HIV-1 protease inhibitors on insulin secretion and insulin signaling in INS-1 beta cells

“…In contrast, nelfinavir and saquinavir inhibited the glucose-stimulated insulin secretion at a 20-times lower concentration. This is again in contrast to the inhibitory effects on viral replication, where the free indinavir concentration only needed to be two or three times higher than nelfinavir and saquinavir to attain similar effects (IC 50 0·041, 0·025 and 0·014 µM for indinavir, nelfinavir and saquinavir respectively; Molla et al 1998). Since indinavir has lower serum protein binding and thus results in a higher free indinavir concentration, it was effective at even lower concentrations than nelfinavir and saquinavir if 50% human serum was present (IC 50 0·09, 0·92 and 0·37 µM for indinavir, nelfinavir and saquinavir, respectively; Molla et al 1998).…”

“…This is again in contrast to the inhibitory effects on viral replication, where the free indinavir concentration only needed to be two or three times higher than nelfinavir and saquinavir to attain similar effects (IC 50 0·041, 0·025 and 0·014 µM for indinavir, nelfinavir and saquinavir respectively; Molla et al 1998). Since indinavir has lower serum protein binding and thus results in a higher free indinavir concentration, it was effective at even lower concentrations than nelfinavir and saquinavir if 50% human serum was present (IC 50 0·09, 0·92 and 0·37 µM for indinavir, nelfinavir and saquinavir, respectively; Molla et al 1998). These data for the effects on virus replication and our results, based on a long-term incubation period, therefore suggest that, compared with the effect on virus replication, indinavir causes less impairment of insulin secretion than nelfinavir and saquinavir.…”

“…Thus 2·5 µM ritonavir inhibited glucose-stimulated insulin secretion, whereas amprenavir had no effect even at an eight-times higher concentration. This difference between the two PIs was observed although ritonavir and amprenavir are similarly effective at inhibiting the production of infectious virus from HIV-infected lymphocytes (IC 50 0·068 and 0·078 µM for ritonavir and amprenavir respectively; Molla et al 1998). In the presence of binding proteins (50% human serum), a condition that more likely resembles the situation in our experiments (our medium contained 10% fetal calf serum) ritonavir was even less effective at inhibiting virus replication than amprenavir (IC 50 1·34 and 0·55 µM for ritonavir and amprenavir respectively; Molla et al 1998).…”

“…This difference between the two PIs was observed although ritonavir and amprenavir are similarly effective at inhibiting the production of infectious virus from HIV-infected lymphocytes (IC 50 0·068 and 0·078 µM for ritonavir and amprenavir respectively; Molla et al 1998). In the presence of binding proteins (50% human serum), a condition that more likely resembles the situation in our experiments (our medium contained 10% fetal calf serum) ritonavir was even less effective at inhibiting virus replication than amprenavir (IC 50 1·34 and 0·55 µM for ritonavir and amprenavir respectively; Molla et al 1998). Taken together, this suggests that, at least in vitro, ritonavir inhibits virus replication only at concentrations similar to or higher than amprenavir, but, after a 2 day incubation period, impairs insulin secretion at considerably lower concentrations than amprenavir.…”

Cellular models for the screening and development of anti-hepatitis C virus agents

“…It has been demonstrated that drugs including certain antiviral compounds show markedly reduced activity in vivo or in cellular models because of their natural tendency to bind serum proteins (Molla et al, 1998). To evaluate this possibility, the inhibitory effect of the antiviral candidates should be measured in cells cultured in media containing increasing concentrations of human serum (Huang et al, 2008).…”