Cell, Vol. 93, 681–684, May 29, 1998, Copyright 1998 by Cell PressHIV Entry and Its Inhibition MinireviewThedetailsofthisconformational changeareonlybegin-David C. Chan and Peter S. KimWhitehead Institute for Biomedical Research ning to be understood, but at least include exposure ofHoward Hughes Medical Institute the “fusion” peptide at the amino terminus of gp41. ByDepartment of Biology analogy with the spring-loaded model of influenza virusMassachusetts Institute of Technology(Carr and Kim, 1993), this fusion-peptide region isCambridge, Massachusetts 02142thought to insert into the target membrane at an earlystep of the fusion process. The observation that the HIVenvelope complexreadilyundergoes receptor-activatedThe human immunodeficiency virus type 1 (HIV-1) isconformational change suggests that its native state isan enveloped virus, and its envelope protein complexmetastable, again similar to the pH-activated envelopecontrols the key process of viral entry. This envelopeprotein of influenza virus (Carr et al., 1997). That is,proteindeterminesviral tropismandfacilitates themem-the labile native state of the HIV envelope complex isbrane fusion process that allows invasion of the viraltransformed by receptor binding to an energeticallygenome. The envelope protein can also promote themore stable, fusion-active conformation.fusion of infected cells with uninfected neighboringProtein dissection studies demonstrated that the twocells, a phenomenon called syncytium formation that is4,3 hydrophobic repeat regions within gp41form aheli-readily observed in cultured cells and may be responsi-cal trimer of antiparalleldimers (Lu et al., 1995). Crystal-ble for some of the cytopathic effects of advanced HIVlographic analysis confirmed that this gp41 core is ainfection.Herewe reviewrecentinsightsin HIV envelopesix-helix bundle in which the N and C helices are ar-protein structure and function and present our currentranged into three hairpins (Figures 1B and 1C) (Chan etunderstandingof the entryprocess.Wealso review howal., 1997; Weissenhorn et al.,1997). TheN peptides formthese findings leadtonew approaches forinhibiting HIVthree central helices arranged in a trimeric coiled coil.entry and may provide insights into the design of betterThe C peptides form three outer helices that pack in anHIV vaccines.antiparallel manner into highly conserved, hydrophobicThe HIV-1 envelope protein complex is initially pro-grooves on the surface of this coiled coil (Figure 1D).duced as the precursorgp160, which is extensively gly-This structurelikely represents thefusion-activeconfor-cosylated and proteolytically cleaved into two subunitsmation ofgp41(see discussion inChan etal.,1997), andby a cellular convertase (Luciw, 1996). The resultingresembles the proposed fusion-active conformations ofsurface subunit (gp120) and transmembrane subunitthe transmembrane envelope proteins from influenza(gp41) remain noncovalently associated and oligomer-virus and Moloney murine leukemia virus.ize, most likely as trimers, on the surface of the virion.Fusion Mechanismgp120 binds to CD4 and a coreceptor (a seven-trans-Anymodel forHIVentry mustaccountfor theremarkablemembrane protein of the chemokine receptor family),observation that synthetic C peptides inhibit HIV infec-which are present on susceptible cells such as T lym-tion and syncytia formation at nanomolar concentra-phocytesandmacrophages.Subsequently,gp41under-tions (Jiang et al., 1993; Wild et al., 1994). Preincubationgoes conformational changesthat mediatefusion of theof virus with C peptide, followed by its removal, doesviral membranewiththetarget cellmembrane. HIV entrynot block HIV entry; this feature suggests that theseisthereforeacomplexprocessinvolving multipleproteinpeptides do not acton the native conformation ofgp41.interactions, each of which is a potential target for theBased on the structural features of the fusion-activedevelopment of antiviral compounds.gp41 core, these peptide inhibitors likely work by bind-gp41 Structureing to the trimeric coiled coil of gp41, thereby actingThe gp41 molecule is a transmembrane protein withthrough a dominant-negative mechanism (Lu et al.,several importantfeatures withinitsectodomain (Figure1995; Chan et al., 1997; Weissenhorn et al., 1997).1A). First, the amino terminus of gp41, created by pro-A simple dominant-negative model, however, fails toteolytic cleavage of the gp160 precursor, contains aexplain adequately one puzzling feature of C peptidehydrophobic,glycine-rich“fusion” peptidethatisessen-inhibition: How can C peptides act at such low concen-tial for membrane fusion.Second, thereare two regionstrations, given that the N and C peptide regions arewitha4,3 hydrophobic (heptad) repeat, asequencemo-within a single gp41 molecule? That is, how can the Ctif characteristic of coiled coils. Synthetic peptides(seepeptides soeffectivelyinhibit anintramolecular associa-below) derived from these two regions are termed Ntion reaction,in which the two regions of gp41 are pres-(amino-terminal) and C (carboxy-terminal)peptides. Be-ent ina high effective concentration?It seems likelythattween these two heptad repeat regions is a loop regionthe solution to this puzzle is that C peptides must bindcontaining two cysteines.to gp41 prior to formation of the six-helix complex be-A large number of studies support the notion that thecause, oncethisgp41 coreis assembled, it is extremelyenvelopecomplexexists inatleasttwo major conforma-stable (the melting temperature of the gp41 core is intions (see references in Chan et al., 1997). The native,excess of908C) and isunlikelytobedisruptedby exoge-or nonfusogenic, conformation exists on the surfacenous peptides. These observations suggest the exis-of free virions after budding from infected cells. Upontence of a transiently populated, third state of gp41. Cbinding of gp120 to target cell receptors, gp41 under-goes a conformational change to a fusion-active state. peptides act at this intermediate stage, after the nativeCell682Figure 1. HIV gp41 Structure(A) A schematic view of gp41, showing thelocation of the fusion peptide (fp), the twohydrophobic heptad repeats, the transmem-brane segment (tm), and the cytoplasmic re-gion (cyto). Thelocationof various gp41 pep-tides are shown above. The locations of thethree residues that bindthe cavityin the cen-tral coiled coil (D) are indicated by red bars.(B and C) A ribbon diagram of the N36/C34complex (which forms the core of the