Binding modes of CCR5-targetting HIV entry inhibitors: partial and full antagonists

Abstract

Since the CC-chemokine receptor 5 (CCR5) was identified as a major co-receptor for human immunodeficiency virus type 1 (HIV-1) entry into a host cell, CCR5-targetting HIV entry inhibitors have been developed and some of them are currently in clinical trials. Most of these inhibitors also inhibit the physiological chemokine reaction function of CCR5, which is so far considered to be safe to patients based on the observation that individuals that naturally lack CCR5 do not show apparent health problems. Nevertheless, to minimize the toxicity and side effects, it would be ideal to preserve the chemokine receptor activity. In this work, we simulated the flexible docking of two small molecule inhibitors to CCR5 in a solvated phospholipid bilayer environment. One of the inhibitors, aplaviroc has a unique feature of preserving two of the natural chemokine ligands binding to CCR5 and subsequent activation whereas the other one, SCH-C fully blocks chemokine-CCR5 interactions. Our results revealed significantly different binding modes of these two inhibitors although both established extensive interaction networks with CCR5. Comparison of the different binding modes suggests that avoiding the deep insertion of inhibitors into the transmembrane helix bundle may be able to preserve chemokine-CCR5 interactions. These results could help design HIV co-receptor activity-specific inhibitors.

Figure 1

Chemical structures of aplaviroc and SCH-C and their binding constants with CCR5. The labeled atoms were used for superimposition onto retinal to obtain the initial placements in CCR5.

Figure 2

Sequence alignment of CCR5 with bovine rhodopsin (PDB entry: 1U19 and chain A). The residues in 3Å distance to retinal in rhodopsin and the corresponding residues in CCR5 are colored red.

Figure 3

The purple balls show the putative inhibitor-binding pockets of CCR5, detected by the PASS program, which is buried in the interface region between the transmembrane domain and the extracellular β-hairpin loop. Left: the last snapshot structure of CCR5 in the equilibrium simulation. Middle: Side chains of Tyr108, Phe109, Ser180 and Met287 were adjusted for docking aplaviroc. Right: Side chains of Phe112, Phe113, Ile116, Trp248, Tyr251 and Met287 were adjusted for docking SCH-C.

Figure 4

Structures of the CCR5-inhibitor complexes after the 4.11 ns molecular dynamics simulations. The CCR5 proteins are shown in ribbon and the inhibitors are highlighted by the Van der Waals sphere representations. Left: aplaviroc; Right: SCH-C. Aplaviroc bound in a shallow pocket underneath the extracellular β-hairpin loop while SCH-C bound in a pocket spanning into the middle of the transmembrane bundle.

Figure 5

Stereoviews of the interaction networks between CCR5 and the inhibitors, established during the flexible docking simulations. The inhibitors aplaviroc and SCH are shown in sticks and their starting conformations were colored yellow. The protein residues showing important interactions with the inhibitors are shown by lines. For clarity, some residues in 3Å distance to the inhibitors are not shown, see text for the complete lists. The water molecules in 3Å distance to the inhibitors are shown in red balls. The hydrogen bonds are shown in dashed lines. The pictures were made from the same view after superimposing the protein onto the starting conformations. upper: aplaviroc; lower: SCH-C.