of Work: Infection by the HIV virus causes a change in a number of physiological processes, the etiologies of which are poorly understood. The initial step is syncitium formation, currently accepted as involving HIV gp120 and gp41, CD4 and a chemokine receptor. Later developments include the depletion of T cells expressing CD4 and B cells expressing immunoglobulins, and changes in the regulation of cytokines. Crucial to an understanding of these processes and to developing therapeutic strategies related to these changes is the determination of the structural motifs critical to the physiological processes involved in the changes. We are employing the methodologies we have been using for epitope determination (protection assays and surface modification reactions combined with mass spectrometry) to probe receptor-ligand pairs relevant to HIV infection including: a) CD4 and gp120 and; b) the complex between gp120, CD4 and chemokine receptor CXCR4. The ternary complex between gp120, CD4 and a chemokine receptor is now accepted as the crucial interaction involved in cellular infection by the HIV. Recently, the crystal structure of a 1:1:1 complex between mutant gp120, mutant CD4 and an antigen-binding fragment of an antibody has been reported. The gp120 used in this study, however, did not contain the variable loops nor was it fully glycosylated. In view of the highly mutated structure of the gp120 used in the crystal structure, information about complex stoichiometry and sites of interaction in the full length, fully glycosylated gp120 in solution is still uncertain. Even more importantly, the V-3 loop (and its associated glycans) of gp120, which were not present in the gp120 construct used for the crystal structure determination, has been implicated in binding with chemokine receptors and CD4. Our approach to probing the gp120/CD4 interaction site uses chemical modification of the complex in its native state compared to the same modification on the individual non-complexed components using the reaction of phenylglyoxal with Arg residues to determine which Args are surface accessible in the complex. There are two adjacent Arg s on the surface of CD4. Based on the structure of the truncated gp120 and CD4, one of these Rs, R59, is involved in the interaction of CD4 with gp120, while the other is not. The question arose as to whether or not the presence of the glycans in the in vivo system might change the conformation of the CD4 such that both R s might be involved in the interaction or that R59 is less involved. We observed that R59 is totally protected in the complex while R58 is not protected. In solution both R59 and R58 are accessible for modification. Thus, the in vivo complex behaves as predicted from the structure of the truncated complex. This study also shows that phenylglyoxalation of arginine residues is a sensitive measurement of Arg surface accessibility. We are in the process of expressing the third component of the complex CXCR4.
