High-Affinity Binding of Chemokine Analogs that Display Ligand Bias at the HIV-1 Co-receptor CCR5

Abstract

The chemokine receptor CCR5 is a drug target to prevent transmission of HIV/AIDS. We studied four analogs of the native chemokine RANTES (CCL5) that have anti-HIV potencies of around 25 pM, which is more than four orders-of-magnitude higher than that of RANTES itself. It has been hypothesized that the ultra-high potency of the analogs is due to their ability to bind populations of receptors not accessible to native chemokines. To test this hypothesis, we developed a homogeneous dual-color fluorescence cross-correlation spectroscopy (FCCS) assay for saturation and competition binding experiments. The FCCS assay has the advantage that it does not rely on competition with radioactively labeled native chemokines used in conventional assays. We prepared site-specifically labeled fluorescent analogs using native chemical ligation of synthetic peptides, followed by bioorthogonal fluorescent labeling. We engineered a mammalian cell expression construct to provide fluorescently labeled CCR5, which was purified using a tandem immunoaffinity and size-exclusion chromatography approach to obtain monomeric fluorescent CCR5 in detergent solution. We found subnanomolar binding affinities for the two analogs 5P12-RANTES and 5P14-RANTES, and about twenty-fold reduced affinities for PSC-RANTES and 6P4-RANTES. Using homologous and heterologous competition experiments with unlabeled chemokine analogs, we conclude that the analogs all bind at the same binding site; whereas, the native chemokines (RANTES and MIP1) fail to displace bound fluorescent analogs even at tens of micromolar concentrations. Our results can be rationalized with de novo structural models of the N-terminal tails of the synthetic chemokines that adopt a different binding mode as compared to the parent compound.