Structural comparison of monomeric variants of the chemokine MIP-1 beta having differing ability to bind the receptor CCR5

MIP-1 beta, a member of the chemokine family of proteins, tightly binds the receptor CCR5 as part of its natural function in the immune response, and in doing so also blocks the ability of many strains of HIV to enter the cel l. The sin-le most important MIP-1 beta residue known to contribute to its interaction with the receptor is Phe 13, which when mutated reduces the abi lity of MIP-1 beta to bind to CCR5 by more than 1000-fold. To obtain a stru ctural understanding of the dramatic effect of the absence of Phe13 in MIP- 1 beta, we used multidimensional heteronuclear NMR to determine the three-d imensional structure of the MIP-1 beta F13A variant. We had previously show n that, unlike the wild-type protein which has been shown to be a tight dim er, the F13A mutant is monomeric even at high concentrations [Laurence, J.S ., Blanpain, C., Burgner, J.W., Parmentier, M., and LiWang, P.J. (2000) Bio chemistry 39, 3401-3409], leading to significant changes in the NMR spectra of F13A and the wild-type protein. We have obtained a total of 940 structu ral restraints for MIP-1 beta F13A, and have calculated a family of structu res having a backbone rmsd from the average of 0.55 Angstrom (residues 12-6 7). A structural comparison of the F13A mutant with a fully active monomeri c variant, PSA, shows that despite some differences in the H-1-N-15 HSQC sp ectra the two are nearly identical in NOE distance restraints and in backbo ne conformation. A comparison of F13A with the wild-type protein shows larg ely the same fold, although differences exist in the N-terminal and loop re gions for which the loss of the dimer in F13A can mainly account. A dynamic s comparison confirms greater flexibility in F13A than in the wild-type pro tein in regions of dimer contact in the wildtype protein. In an analysis to determine if the large functional effect resulting from the loss of Phe13 is due to the local side chain change or due to more global structural chan ges, we conclude that local effects predominate. This suggests that a strat egy for designing tight binding anti-CCR5 therapeutics should include a Phe -like component.