Structural Biology of HIV - RCSB · Structural Biology of HIV BrianG.TurnerandMichaelF.Summers*...

REVIEW

Article No. jmbi.1998.2354 available online at http://www.idealibrary.com on J. Mol. Biol. (1999) 285, 1±32

Structural Biology of HIV

Brian G. Turner and Michael F. Summers*

Howard Hughes MedicalInstitute and Department of

The human immunode®ciency virus (HIV) genome encodes a total ofthree structural proteins, two envelope proteins, three enzymes, and six

Chemistry and BiochemistryUniversity of MarylandBaltimore County, 1000 HilltopCircle, BaltimoreMD 21250, USA

E-mail address of the [email protected]

immunode®ciency virus type 1; HTleukemia virus type 1; SIV, simianvirus; EIAV, equine infectious anemMoloney murine leukemia virus; Preverse transcriptase; IN, integrasep24; NC, nucleocapsid protein p7;p17; SU, surface glycoprotein gp12transmembrane protein gp41; RREelement; ER, endoplasmic reticulummagnetic resonance; CypA, cyclophdissociation constant; MHR, majorcapsid; SL3, stem-loop-3 of the HIVFood and Drug Administration; TAresponse element; Cdk9, cyclin-depkinase-9; HA, hemagglutinin; LTR,MPMV, Mason-P®zer monkey viruhelix; SH3, Src-homology-3; PPII, pSLIIB, stem-loop IIB; BIV, bovine imvirus.

0022-2836/99/010001±32 $30.00/0

accessory proteins. Studies over the past ten years have provided high-resolution three-dimensional structural information for all of the viralenzymes, structural proteins and envelope proteins, as well as for threeof the accessory proteins. In some cases it has been possible to solve thestructures of the intact, native proteins, but in most cases structural datawere obtained for isolated protein domains, peptidic fragments, ormutants. Peptide complexes with two regulatory RNA fragments and aprotein complex with an RNA recognition/encapsidation element havealso been structurally characterized. This article summarizes the high-res-olution structural information that is currently available for HIV proteinsand reviews current structure-function and structure-biological relation-ships.

# 1999 Academic Press

Keywords: acquired immunode®ciency syndrome (AIDS); humanimmunode®ciency virus (HIV); protein structure; X-ray crystallography;nuclear magnetic resonance

According to the World Health Organization, 5.8

*Corresponding author

Introduction
million people were infected with the human
Since the identi®cation of Acquired Immunode®-ciency Syndrome (AIDS) in developed countries inthe early 1980s, the AIDS epidemic has resulted ina total 11.7 million deaths, including the deaths of4.0 million women and 2.7 million children.

Abbreviations used: AIDS, acquiredimmunode®ciency syndrome; HIV-1, human

ing author:

LV-1, human T-cellimmunode®ciencyia virus; MoMuLV,

R, protease; RT,; CA, capsid proteinMA, matrix protein0; TM,, Rev response

; NMR, nuclearilin A; Kd,homology region of-1 -RNA; FDA,R, transactivatingendent proteinlong terminal repeat;s; HTH, helix-turn-oly-L-proline type II;

munode®ciency

immunode®ciency virus (HIV) in 1997 alone, and30.6 million people are currently living with HIVinfection. HIV was identi®ed as the causativeagent for AIDS in 1983 (BarreÂ-Sinoussi et al., 1983;Gallo et al., 1984; Levy et al., 1984). The virusinfects CD4� lymphocytes and causes theirdestruction with a half-life of less than two days(Ho et al., 1995; Perelson et al., 1996; Wei et al.,1995).

Efforts to control the AIDS epidemic havefocused heavily on studies of the biology, biochem-istry, and structural biology of HIV and on inter-actions between viral components and new drugcandidates. The reverse transcriptase inhibitor AZT(zidovudine) was ®rst approved by the U.S. Foodand Drug Administration (FDA) for treating AIDSin 1987, and seven FDA-approved reverse tran-scriptase (RT) inhibitors (including ®ve nucleosideand two non-nucleoside inhibitors) are now com-mercially available. Although these drugs delaythe progression of the disease, they do not preventit, as infection readily leads to drug-resistantmutants. More recently, a new class of drugs thattarget the HIV protease (PR) was introduced, andfour different PR inhibitors are currently on themarket. These drugs were developed via structure-based rational drug design strategies, in which

# 1999 Academic Press

mailto:[email protected]

drug candidates were designed, tested, and modi-®ed on the basis of high-resolution three-dimen-

stabilized as a ribonucleoprotein complex with ca.2000 copies of the nucleocapsid protein (NC, p7;

2 Structural Biology of HIV

sional structural information obtained for PR andPR-inhibitor complexes. As for the RT inhibitors,the virus is capable of developing resistance to thePR inhibitors. This is likely due to the low ®delityof RT, which does not have a proofreading func-tion. Although a given cell is believed to beinfected only once by HIV, it has been estimatedthat at least 109 new cells are infected per day inlatent HIV-infected patients, and that pointmutations occur along the entire length of the gen-ome at the rate of 104 to 105 times per day (Cof®n,1995).

Recently developed ``drug cocktails'' containingcombinations of PR and RT inhibitors can reduceviral loads to undetectable levels, and these lowlevels can be maintained for periods of two yearsor more (Gulick et al., 1997). Although there aregrounds for optimism that current drug cocktailsmay keep the virus at bay for extended periods, itappears unlikely that the current repertoire willlead to a cure. The most serious problem is that thevirus apparently can be maintained in reservoirsthat are not susceptible to the current drugs (Chunet al., 1997; Finzi et al., 1997; Wong et al., 1997). Inaddition, the current drug regimes are expensiveand compliance is dif®cult, and it is therefore pru-dent to continue to pursue other viral componentsas potential drug targets.

Morphology of the Mature Virion

HIV is a member of the lentivirus genus, whichincludes retroviruses that possess complex gen-omes and exhibit cone-shaped capsid core par-ticles. Other examples of lentiviruses include thesimian immunode®ciency virus (SIV), visna virus,and equine infectious anemia virus (EIAV). Like allretroviruses, HIV's genome is encoded by RNA,which is reverse-transcribed to viral DNA by theviral reverse transcriptase (RT) upon entering anew host cell.

The general features of the mature HIV virionand ribbon drawings of the structurally character-ized viral proteins are shown in Figure 1. All lenti-viruses are enveloped by a lipid bilayer (yellow)that is derived from the membrane of the host cell.Exposed surface glycoproteins (SU, gp120; cyan)are anchored to the virus via interactions with thetransmembrane protein (TM, gp41; violet). Thelipid bilayer also contains several cellular mem-brane proteins derived from the host cell, includingmajor histocompatibility antigens, actin andubiquitin (Arthur et al., 1992). A matrix shell com-prising approximately 2000 copies of the matrixprotein (MA, p17; green) lines the inner surface ofthe viral membrane, and a conical capsid coreparticle comprising ca. 2000 copies of the capsidprotein (CA, p24; red) is located in the center ofthe virus. The capsid particle encapsidates twocopies of the unspliced viral genome, which is

blue), and also contains three essential virallyencoded enzymes: protease (PR; pink), reversetranscriptase (RT; purple) and integrase (IN; olive).Virus particles also package the accessory proteins,Nef (orange), Vif and Vpr (not shown). Threeadditional accessory proteins that function in thehost cell, Rev, Tat and Vpu, do not appear to bepackaged.

The HIV-1 Replication Cycle

General features of the HIV replication cycle areshown in Figure 2. The early phase begins with therecognition of the target cell by the mature virionand involves all processes leading to and includingintegration of the genomic DNA into the chromo-some of the host cell. The late phase begins withthe regulated expression of the integrated proviralgenome, and involves all processes up to andincluding virus budding and maturation.

Early phase

HIV-1 particles bind speci®cally to cells bearingCD4, a protein that normally functions in immunerecognition. Binding occurs via speci®c interactionsbetween the viral envelope glycoprotein SU(gp120) and the amino-terminal immunoglobulindomain of CD4. These interactions are suf®cientfor binding but not for infection. Unlike otherretroviruses, the primate lentiviruses requireadditional cell-surface proteins to promote fusionof the viral and cellular membranes. For HIV-1,membrane fusion can be triggered by one ofseveral chemokine receptors, including CXCR4 andCCR5 (Chapham & Weiss, 1997; Doranz et al.,1996; Feng et al., 1996; Moore, 1997).

Membrane fusion is followed by a poorly under-stood uncoating event that affords an intracellularreverse transcription complex. Reverse transcrip-tion is catalyzed in the cytosol by reverse transcrip-tase (RT). The accessory protein Vif appears to beimportant during one or more of these earlyevents, perhaps by facilitating the initial stages ofreverse transcription. RT-dependent DNA syn-thesis is also dependent on the viral NC proteins,and is initiated by the binding of a cellulartRNALys primer. Although the process of reversetranscription is complex, the mechanism of RT-dependent DNA synthesis has emerged fromextensive in vitro and in vivo studies and the readeris referred to the literature for details (for example,Goff, 1990; Katz & Skalka, 1994; Skalka & Goff,1993; Telesnitsky & Goff, 1997; Whitcomb &Hughes, 1992).

Once synthesized, the viral DNA is transportedto the nucleus as part of a preintegration complexthat appears to include the IN, MA, RT, and Vprproteins, as well as the cellular host proteinHMG-I(Y) (Miller et al., 1997). The HIV CA pro-teins do not appear to be part of the preintegration

complex, although they contribute to the structureof other retroviral preintegration complexes

Freed & Martin, 1994; Nie et al., 1998; Reil et al.,1998). After active transport to the nucleus, the

Figure 1. Drawing of the mature HIV virion surrounded by ribbon representations of the structurally characterizedviral proteins and protein fragments. The protein structures have been drawn to the same scale. The TM ectodomainshown is that determined for the closely related SIV (see the text for references to the HIV-1 TM fragment structures).

Structural Biology of HIV 3

(Bowerman et al., 1989). Nuclear localization of thepreintegration complex is directed by the accessoryprotein Vpr (Fouchier et al., 1997; Freed et al., 1995;Nie et al., 1998), which does not contain a nuclearlocalization signal but appears to function by con-necting the preintegration complex to the cellularnuclear import machinery (including importin-aand the nucleoporins; Fouchier et al., 1998; Popovet al., 1998; Vodicka et al., 1998). Vpr also interfereswith normal cell cycle control by arresting thegrowth of infected cells in the G2 phase (Jowettet al., 1995; Re et al., 1995; Rogel et al., 1995).Nuclear localization may be facilitated by the MAproteins (Bukrinsky et al., 1993; von Schwedleret al., 1994), although this proposal has been ques-tioned (Fouchier et al., 1997; Freed et al., 1995;

viral DNA is covalently integrated into the hostgenome by the catalytic activity of IN.

Late phase

The late phase of the virus life cycle begins withthe synthesis of unspliced and spliced mRNA tran-scripts, which are transported out of the nucleusfor translation (Figure 2). Initially, short splicedRNA species that encode the regulatory proteinsTat, Rev and Nef are synthesized. Tat is an essen-tial transcriptional activator that binds to a stem-loop element of the nascent RNA transcript (TAR,for trans-activating response element) and recruitsthe cellular proteins cyclin T and cyclin-dependentprotein kinase-9 (Cdk9; previously called TAK or

PITALRE). Recent studies indicate that cyclin Tbinds directly to Tat, enhancing its af®nity and

lyzed to GDP, the complex dissociates, and theamino-terminal nuclear localization signal (NLS) of

Figure 2. General features of the HIV-1 replication cycle. The early phase (upper portion of the diagram) beginswith CD4 recognition and involves events up to and including integration of the proviral DNA, and the late phaseincludes all events from transcription of the integrated DNA to virus budding and maturation.


altering its speci®city for the TAR RNA (Wei et al.,1998). Cdk9 then phosphorylates the RNA poly-merase II transcription complex, stimulating tran-scription elongation (Herrmann & Rice, 1995;Reines et al., 1996; Wei et al., 1998).

Ordinarily, unspliced cellular mRNAs areretained in the nucleus where they can be furtherprocessed or degraded. However, full length andsingly spliced HIV mRNA transcripts that containfunctional introns are needed in the cytoplasm forGag and Gag-Pol synthesis and packaging, andtheir export is mediated by the essential HIV acces-sory protein Rev. Rev binds as an oligomer to therev response element (RRE) of nascent unsplicedmRNAs and recruits the cellular nuclear shuttlingprotein exportin-1 (XPO; Ohno et al., 1998) and thenuclear export factor Ran guanosine triphosphatase(in its GTP-bound form; Pollard & Malim, 1998).This complex is then transported through thenuclear pore to the cytosol where GTP is hydro-

Rev directs its import back into the nucleus(Emerman & Malim, 1998). In this manner, Revfunctions as a switch between the early synthesisof highly spliced mRNAs (encoding Tat, Rev andNef) and the later synthesis of unspliced (encodingthe Gag and Gag-Pol proteins) and singly spliced(encoding Env, Vpu, Vif and Vpr) mRNAs.

The Env precursor polyprotein (gp160) is syn-thesized in the endoplasmic reticulum (ER) usingthe spliced env mRNA gene as the message. Theprotein appears to oligomerize to a trimeric struc-ture in the ER, and is heavily glycosylated (Chanet al., 1997; Earl et al., 1991; Leonard et al., 1990; Luet al., 1995; Wyatt & Sodroski, 1998). Env is post-translationally modi®ed in the ER and Golgiapparatus and is cleaved to produce the non-cova-lently associated (TM-SU)3 trimeric glycoproteincomplex. The heterogeneously glycosylated TM-SUtrimer is then transported to the cell membrane forvirus assembly. Env and CD4 molecules are both

synthesized in the ER, and the premature bindingof CD4 to Env in the ER can inhibit translocation

cleaved by PR to produce the independentenzymes, as well as the MA, CA and NC structural


of Env to the cell membrane or the formation of afully functional TM-SU complex (Hoxie et al.,1986). Thus, CD4 is targeted for removal from theER by the viral accessory protein Vpu, which bindsCD4 molecules and signals their degradation viathe ubiquitin-proteasome pathway (Crise et al.,1990; Margottin et al., 1998; Schubert et al., 1998).Similarly, cell-surface CD4 molecules are targetedfor endosomal degradation by the binding of theaccessory protein Nef, which also binds to theAP-2 adapter complex and stimulates the for-mation of clathrin-coated pits (Foti et al., 1997; Gallet al., 1998; Greenberg et al., 1997; Piguet et al.,1998). The down-regulation of CD4 molecules onthe surface of infected cells may also serve as ameans for avoiding an immune response.

The Gag polyprotein is synthesized in the ribo-somes from the unspliced mRNA. A translationalframeshift results in the generation of smalleramounts of Gag-Pol precursor proteins, whichassociate with the Gag polyprotein at the cellularmembrane (Figure 2). The N-terminally myristoy-lated MA domain of the polyproteins directs bind-ing to the cellular membrane (Bennett et al., 1993;Bryant & Ratner, 1990; Chazal et al., 1994; FaÈckeet al., 1993; Freed et al., 1995, 1990, 1994; Freed &Martin, 1995; Gheysen et al., 1989; GoÈ ttlinger et al.,1989; Shoji et al., 1990; Spearman et al., 1994; Wang& Barklis, 1993; Wang et al., 1993; Wills et al., 1991;Yuan et al., 1993; Zhou et al., 1994) and interactswith the cytoplasmic tail of TM (Dorfman et al.,1994b; Freed & Martin, 1996; Freed et al., 1994;Freed & Martin, 1995; Mammano et al., 1995; Wang& Barklis, 1993; Wang et al., 1993; Yu et al., 1992,1993) . Approximately 1200 to 2000 copies of Gagbud to form an immature particle, which encapsi-dates two copies of the unspliced viral genome.Subsequent to budding, the polyproteins are

proteins. The structural proteins rearrange via aprocess called maturation to form the infectiousvirus particle. Cleavage of Gag appears to occurvia an ordered, sequential cleavage process that iscontrolled by different intrinsic proteolysis rates atthe different cleavage sites (Wiegers et al., 1998).Other factors may also be important for infectivity.For example, HIV-1 requires the packaging of thecellular protein cyclophilin A, whereas HIV-2 andmost other primate lentiviruses do not (Frankeet al., 1994b). In addition, Vif is required for theproduction of infectious virions from some, but notall, cell lines (Courcoul et al., 1995; Sova et al., 1995;von Schwedler et al., 1993). Although virions pack-age 7-100 copies of Vif (Camaur & Trono, 1996;Fouchier et al., 1996; Liu et al., 1995), it is notknown if this packaging is essential (Camaur &Trono, 1996). Recent studies suggest that Vif mayfunction by interacting with cellular factors ratherthan with viral components (Simon et al., 1998;Trono, 1995).

The Envelope Proteins

Infection of the host cell by HIV is initiated byinteractions between SU and cell-surface CD4 mol-ecules. SU binds to CD4 and anchors the virus tothe cell surface, and additional interactions withchemokine receptors trigger a conformationalchange that leads to fusion of the viral and cellularmembranes. The mechanism of fusion is not wellunderstood, but may be similar to fusion processesinduced by conformational changes in the envel-ope protein hemagglutinin (HA) of the in¯uenzavirus. In¯uenza particles are taken into cells byendocytosis, and a drop in pH leads to a large con-formational change in the Env protein that isbelieved to trigger fusion of the viral and endoso-

Figure 3. Potential mechanismsfor CD4 and chemokine receptor-induced fusion of the viral and cel-lular membranes. (a) Spring-loadedmechanism similar to that pro-posed for hemagglutinin, whereconformational changes in the TMectodomain lead to a major displa-cement of the N-terminal fusogenicpeptide toward the cellular mem-brane. (b) Shedding mechanism,where CD4 and chemokine bindingresult in the loss of SU proteins,enabling reorientation of the TMand membrane fusion.

mal membranes (Figure 3(a)). This conformationalchange, referred to as a ``spring-loaded mechan-

vides structural information for the SU-interactingresidues and unambiguous connectivity between

Figure 4. Helix packing of the trimeric coiled-coilstructure of the HIV-1 TM ectodomain. N- and C-term-inal helices are colored blue and purple, respectively(from Chan et al. (1997), with permission).


ism'', involves the conversion of a helix-loop-helixsegment (native, ``spring-loaded'' state) to a single,extended helix in a coiled-coil conformation(``sprung'' state), exposing the fusion peptide (Carr& Kim, 1993). Similarities within the amino acidsequences of HIV-1 TM and in¯uenza HA, andbiochemical and structural studies, have led to thesuggestion that conformational changes in HIV-1TM, perhaps triggered by CD4 and chemokinereceptor binding to SU, may parallel thoseobserved in HA (Chan et al., 1997; Lu et al., 1995;Tan et al., 1997; Weissenhorn et al., 1997). However,HIV differs from in¯uenza in that it fuses directlywith the cell membrane in a pH-independent man-ner. As described below, an alternative hypothesisthat does not invoke such a spring-loaded mechan-ism has recently been proposed (Caffrey et al.,1998; Figure 3(b)).

TM (gp41)

The HIV-1 transmembrane protein consists of anN-terminal ectodomain, a transmembrane domain,and a C-terminal intraviral segment that interactswith MA. Three-dimensional structures of peptidescorresponding to portions of the amino-terminalectodomain of TM from HIV-1 (Chan et al., 1997;Tan et al., 1997; Weissenhorn et al., 1997) and Molo-ney murine leukemia virus (MoMuLV; Fass et al.,1996) have been solved by X-ray crystallography.In addition, the structure of the essentially intactectodomain of simian immunode®ciency virus(SIV; residues 27-149), which lacks the N-terminalfusogenic peptide and contains C86A and C92Amutations, has recently been determined by NMRmethods (Caffrey et al., 1998).

All of the constructs in the X-ray crystallographicstudies lacked residues 35-44, which appear to beimportant for TM-SU interactions. Thus, the X-raystructure determined by Kim and co-workers con-sisted of two peptides, designated N36 (residues546-581) and C34 (residues 628-661; Chan et al.,1997). The X-ray structure of Harrison and co-workers also consisted of two peptides, onecontaining the amino-terminal portion of the TMectodomain, in which the fusogenic segment (resi-dues 1-29) was replaced by a GCN4 segment thatforms a stable trimeric coiled coil, and the othercorresponding to the C-terminal portion of the TMectodomain (Weissenhorn et al., 1997). The X-raystructure by Tan and co-workers was determinedfor a single polypeptide chain in which the 45 resi-dues that separate the N- and C-terminal segmentswere substituted by a six-residue linker peptide.

All four structures exhibit essentially identicalcore conformations, as expected given the highsequence identity between the HIV and SIV pro-teins. Views of the structure by Kim and colleaguesshowing the packing arrangement of the trimerichelical bundle are shown in Figure 4. The SIVNMR structure, which is more complete and pro-

the N- and C-terminal segments, is given inFigure 1.

The ectodomain is a symmetrical trimer, witheach monomer consisting of two antiparallela-helices connected by an extended loop (the X-raystructures lack the loop, as well as portions of theadjacent helices). The N-terminal helices form acentral parallel three-helix coiled coil, and theC-terminal helices pack around the central coiledcoil in an antiparallel arrangement (Figures 1 and4). The structure has the overall shape of a rod,with the C and N termini (and thus the fusogenicpeptides) at one end and the extended loop at theother. Although the extended loop of TM from SIVis somewhat more ¯exible than the helical core(Caffrey et al., 1997), it is well ordered, eventhough the internal two cysteine residues that canform inter- and intramolecular disul®de linkageshave been mutated to alanine residues.

The TM ectodomain structures are similar insome respects to the ``sprung'', fusogenic form ofin¯uenza HA2. Both structures contain a central,three-stranded parallel coiled coil, with each mono-mer consisting of two helices packed in an antipar-allel arrangement. However, unlike for HA2, theN-terminal helices of the HIV and SIV TM ectodo-mains are located within the interior of the trimericcoiled coil, and this would presumably inhibit alarge reorientation of these helices. In addition, theextended loops of SIV and HIV TM contain prolineresidues and exhibit a low propensity to forma-helices. These differences prompted Clore, Gro-nenborn and co-workers to suggest that TM-induced fusion by HIV and SIV may not proceedvia an HA2-like spring-loaded conformationalchange. Instead, they propose that CD4 and che-mokine receptor binding may lead to shedding ofthe SU proteins, which would allow TM to reorientparallel to the viral (and cellular) membranes and

permit insertion of the fusogenic peptide into thecellular membrane (Caffrey et al., 1998; Figure 3(b)).

sections of the inner and outer domains, respect-ively, and the V4 and V5 loops projecting from the

Figure 5. Structure of the HIV-1 SU core, determinedfor the complex with a two-domain fragment of CD4and an antigen-binding fragment of a neutralizing anti-body that blocks chemokine-receptor binding (notshown). The yellow b-strand forms an antiparallelb-sheet with residues in CD4. The domain is probablyoriented with the viral membrane near the N terminusand the cellular membrane near the bridging sheet(from Kwong et al. (1998) with permission).


SU (gp120)

SU is extensively glycosylated (Leonard et al.,1990) and contains ®ve variable regions (V1-V5;Starcich et al., 1986), four of which (V1-V4) formsurface-exposed loops with disul®de bonds at theirbases (Leonard et al., 1990). The resulting surfacevariability is probably important for evading aneffective immune response (Profy et al., 1990).Numerous studies indicate that CD4 bindinginduces a conformational change in SU thatexposes the chemokine receptor binding surfaceand enhances interactions with the chemokinereceptors (for example, Sattentau et al., 1993;Sattentau & Moore, 1991; Thali et al., 1993; Trkolaet al., 1996). Thus, the likely mechanism for bindingand fusion involves initial interactions betweenHIV SU and CD4 molecules, which result in aconformational change in SU that promotes inter-actions between SU and a chemokine receptormolecule. The binding of the chemokine receptorprobably leads to additional conformationalchanges in SU that are transmitted to TM, with theresulting conformational changes in TM (or theTM-SU trimer) leading to membrane fusion.

High-resolution structural information wasobtained recently for the core domain of HIV-1 SUbound to a CD4 fragment (comprising the two N-terminal immunoglobulin domains) and a Fab 17bantibody that is an inhibitor of chemokine receptorbinding (Kwong et al., 1998). Native SU was notamenable to crystallization due to extensive hetero-geneous glycosylation and conformational dis-order, and the core domain of SU that was actuallystudied was partially glycosylated and containedboth N- and C-terminal deletions and truncatedloops. Speci®cally, the sample that crystallized (1)lacked 52 N-terminal and 19 C-terminal residues,(2) contained a Gly-Ala-Gly tripeptide substitutionfor 67 residues of the V1/V2 loop, (3) contained aGly-Ala-Gly substitution for 32 residues of the V3loop, and (4) was stripped of over 90 % of thecarbohydrate. These extensive surface modi®-cations did not signi®cantly affect the ability of SUto bind CD4 or relevant antibodies (Binley et al.,1998).

The SU core structure consists of two majordomains (``inner'' and ``outer'') and a four-stranded ``bridging'' b-sheet (Figure 5). The innerdomain contains two a-helices, a ®ve-strandedb-sandwich, and several loops. The outer domainis a stacked double barrel, with one barrel compris-ing a six-stranded b-sheet which twists to enfoldan a-helix as a seventh barrel stave, and the otherbarrel comprising a seven-stranded antiparallelb-barrel. The orientation shown in Figure 5 revealsan overall heart-shaped appearance, with theN- and C-terminal residues projecting from theupper-left-hand section of the inner domain,the V1/V2 and V3 loops projecting from the lower

upper section of the outer loop. The disul®debridges observed in the X-ray structure are consist-ent with previous studies (Leonard et al., 1990).The domain is probably oriented on the virus withthe N and C termini on the inner domain directedtowards the virus and the CD4 and 17b (and thusthe chemokine receptor) binding sites orientedtowards the host cell (Kwong et al., 1998). A modelof a putative trimeric complex was also derived,based on comparisons of the X-ray structure withepitope maps (Wyatt et al., 1998).

CD4 binds into a carbohydrate-free depressionformed at the interface of the outer and innerdomains and the bridging sheet (Figure 6). Resi-dues implicated in SU-CD4 interactions on thebasis of mutagenesis studies of SU (Cordonnieret al., 1989; Kowalski et al., 1987; Olshevsky et al.,1990) and CD4 (Moebius et al., 1992; Ryu et al.,1994; Sweet et al., 1991) were shown to participatein intermolecular interactions. Interestingly,although complex formation buries signi®cant sur-face areas of SU (802 AÊ 2) and CD4 (742 AÊ 2), a largeproportion of the buried residues do not makeintermolecular contacts but instead line the sur-faces of large, buried cavities. The smaller cavity(152 AÊ 3) is formed when Phe43 of CD4 covers adeep cleft between the inner and outer domains ofSU (Figure 6). Many of the hydrophobic residuesof SU that line this ``Phe43 cavity'' are highly con-served and essential for SU-CD4 interactions, even

though they do not interact directly with CD4 inthe crystal structure. It is possible that this cavity

that CD4 binding alters the positions of the vari-able loops (Wyatt et al., 1995), the X-ray structure

Figure 6. Ribbon diagram showing interactionsbetween the SU core (red) and the N-terminal immuno-globulin domain of CD4 (yellow) in the SU core/CD4fragment/antibody fragment complex. The side-chain ofCD4 residue Phe43, which caps a large hydrophobiccavity, is also shown (from Kwong et al. (1998) with per-mission).


and others result from a CD4-induced confor-mational change (Kwong et al., 1998). The largercavity (279 AÊ 3) is lined primarily by hydrophilicresidues that are not critical for SU-CD4 binding.The non-interacting SU residues within this cavityexhibit sequence variability. Thus, the CD4-bindingsurface of SU consists of a non-interacting, ``vari-ational island'' (or ``anti-hotspot'') surrounded byconserved, CD4-binding residues, and this unusualrecognition topology may serve as the primarymeans for avoiding an immune response (Kwonget al., 1998).

The 17b antibody binds to the bridging sheet onthe opposite side relative to the CD4 binding site.No direct interactions between CD4 and 17b wereobserved, and no SU residues were observed tomake simultaneous contacts with CD4 and 17b.Previous studies indicated that the 17b and chemo-kine receptor binding sites overlap (Wu et al.,1996). Thus, the X-ray structure provides the ®rstglimpse of the CD4-induced conformation of thechemokine receptor epitope. This epitope is prob-ably shielded from the immune system by thenearby V3 variable loop, and possibly also by alarge CD4-induced conformational change (Kwonget al., 1998).

Since the binding of 17b to intact SU is greatlyenhanced by CD4, it was suggested that CD4 bind-ing induces a conformational change in SU thatexposes the 17b epitope. Although it is possible

provides very strong evidence for a much largerconformational change involving the core residuesof SU. In particular, it was suggested that thebridging sheet and Phe43 pocket are probablyunstable in the absence of CD4 and might exist asan equilibrium between the observed structure anda signi®cantly different conformation (Kwong et al.,1998). Such a conformational change might also beimportant for the CD4- and chemokine receptor-induced exposure of the fusion peptides of TM inthe context of the intact trimeric TM-SU complex.

The Structural Proteins

The HIV gag gene encodes a 55 kDa Gag poly-protein that self-assembles at the cell membrane toform the immature virion. Gag can actually formmembrane-enveloped virus-like particles in theabsence of other viral components (Gheysen et al.,1989). Gag is proteolytically cleaved during matu-ration to the MA, CA and NC proteins, in additionto the low molecular mass cleavage products, p1,p2 and p6. A conserved LXXLF sequence withinthe p6 domain of the Gag precursor is importantfor Vpr packaging (Checroune et al., 1995; Kondo& Gottlinger, 1996). Subsequent to processing,however, none of the low molecular mass peptideshas a known function, and isolated p6 has beenshown to lack a stable tertiary structure (Stys et al.,1993; M.F. Summers, unpublished results). Thetransframe protein p6*, which replaces the C-term-inal Gag proteins p1 and p6 in the transframeregion of the Gag-Pol precursor, was also shownby NMR methods to be disordered in solution(Beissinger et al., 1996).

Matrix

The three-dimensional structure of the HIV-1matrix protein was determined initially by NMRmethods (Massiah et al., 1994, 1996; Matthews et al.,1994, 1995). X-ray structures of the matrix proteinsfrom SIV (Rao et al., 1995) and HIV-1 (Hill et al.,1996), and NMR structures of the MA proteinsfrom bovine leukemia virus (Matthews et al., 1996)and human T-cell leukemia virus type-II(Christensen et al., 1996) were determined shortlythereafter. The HIV-1 matrix protein consists of®ve a-helices, two short 310 helical stretches, and athree-strand mixed b-sheet (Figure 1). Helices I-IIIand the 310 helices pack about a central helix (IV)to form a compact globular domain that is cappedby the b-sheet.

Although HIV-1 MA is monomeric in solution,both the HIV-1 and SIV MA proteins crystallize astrimers. Although the physiological relevance oftrimerization has not been established unambigu-ously, several lines of evidence suggest that MAtrimers serve as a fundamental building block forformation of the MA shell within the mature vir-ion. Indeed, the HIV-1 and SIV MA trimers bear a

strong resemblance to one model of particle for-mation that was based on low-resolution electron

or alters its three-dimensional structure. Inaddition, it has been proposed that phosphoryl-


microscopy studies (Nermut et al., 1993; Nermut &Thomas, 1994). The individual proteins of thetrimer are arranged to create a large basic surface,and it has been proposed that this surface interactsdirectly with the acidic inner membrane of thevirus (Figure 7). In this model, the myristoylatedamino-terminal residues are in close proximity tothe viral membrane, and the C-terminal helicesproject away from the membrane and toward thecenter of the virus. Basic residues implicated inmembrane binding and potential nuclear localiz-ation functions are clustered on and near the three-strand b-sheet. Of these, residues that have beenshown to be essential for virus production (Freedet al., 1995) are located on the putative membranebinding surface, whereas the non-essential basicresidues are removed from the membrane (Hillet al., 1996).

Comparison of the monomeric, solution-stateHIV-1 MA structure with the X-ray structurerevealed a ca. 6 AÊ displacement of a short 310 helix(Pro66-Gly71) located at the trimer interface(Massiah et al., 1996). The ®ndings furthersuggested that residues Pro66 and Gly71, whichare highly conserved and bracket the 310 helix,serve as ``hinges'' which allow the 310 helix toundergo this structural reorientation (Massiah et al.,1996). The displacement of the 310 helix may thusre¯ect a physiologically relevant conformationalchange that occurs during virion assembly and dis-assembly.

Although the structure of HIV-1 MA has beenwell characterized, several interesting issuesremain. For example, the structure of the nativemyristoylated form of the protein has not beensolved, and it will be interesting to determine if themyristate packs into the body of the protein and/

Figure 7. Model of the trimeric HIV-1 matrix proteininteracting with a lipid membrane. Essential and non-essential basic residues are colored magenta and green,respectively, and the N-terminal myristoyl groups aredrawn in red (from Hill et al. (1996), with permission).

ation of MA serves to switch the protein fromtargeting the cell membrane to targeting thenucleus (Gallay et al., 1995a,b; Trono & Gallay,1997), although this hypothesis has recently beenquestioned (Freed et al., 1997).

Capsid

The capsid proteins form a cone-shaped,electron-dense structure in the center of the maturevirus that encapsidates the viral RNA, NC pro-teins, and key enzymes. High-resolution three-dimensional structural information is not availablefor any intact retroviral capsid core particles,which are dif®cult to isolate (Vogt, 1997). Limitedproteolysis studies monitored by NMR revealedthat HIV-1 CA contains two domains (Gitti et al.,1996). Mutations and deletions within the C-term-inal oligomerization domain generally impair orabolish viral assembly, whereas mutations in theamino-terminal core domain often give rise toviruses that can assemble and bud but are non-infectious and typically do not form a normalcapsid (Chazal et al., 1994; Dorfman et al., 1994a;Franke et al., 1994a; GoÈ ttlinger et al., 1989; Hong &Boulanger, 1993; Jowett et al., 1992; Reicin et al.,1995; Spearman et al., 1994; Von Poblotzki et al.,1993; Wang & Barklis, 1993).

HIV-1 CA also binds the human cellular prolineisomerase, cyclophilin A (CypA), resulting in thepackaging of �200 copies of CypA into each HIV-1virion (Braaten et al., 1996; Franke et al., 1994b;Luban et al., 1994, 1993; Thali et al., 1994). Mutantvirions that do not package CypA appear normal(Kong et al., 1998), but are poorly infectious.Although the precise function of cyclophilin A isnot yet clear, it appears to perform an essential(yet unknown) role early in the HIV-1 replicationcycle (Steinkasserer et al., 1995; Thali et al., 1994),perhaps by destabilizing the capsid during uncoat-ing or by performing an additional chaperonefunction.

Analytical ultracentrifugation experiments havedemonstrated that HIV-1 CA forms dimers with adissociation constant (Kd) of 18(�1) mM (Gambleet al., 1997; Rose et al., 1992; Yoo et al., 1997), whichis similar to the dissociation constant measured forthe isolated C-terminal oligomerization domain(Kd � 10(�3) mM). This behavior is puzzling inview of the fact that MA tends to form trimers,and it is not immediately obvious how these sub-domains would interact in the context of the intactGag precursor polyprotein. In this regard, it is notknown if other retroviral CA proteins form dimers,or if this property is unique to HIV-1 CA.

Structural studies of the intact HIV-1 capsidprotein have been confounded by the fact that thenative protein can form a complex mixture of oli-gomers in solution, including dimers, tetramers,dodecamers, spheres, ®bers, and tubes (Ehrlichet al., 1992). At present, no high-resolution structur-

al information is available for any intact retroviralcapsid protein. However, the structure of the

proteolytic processing of the MA-CA junctionduring viral maturation (Gitti et al., 1996). The


isolated N-terminal core domain (residues 1-151)was determined by NMR methods (Gitti et al.,1996; Figure 1), and the structures of the coredomain complexed with cyclophilin A (Gambleet al., 1996), and the partial structure determinationof CA (showing the core domain only) bound toan antibody fragment (Momany et al., 1996) weredetermined by X-ray crystallography.

The structures of the HIV-1 N-terminal coredomain are all in good agreement. The CA coredomain consists of seven a-helices, two b-hairpins,and an exposed, partially ordered loop. Thedomain is shaped like an arrowhead, with leadingedge lengths of ca. 31 AÊ , a trailing edge length ofca. 39 AÊ , and a thickness of ca. 16 AÊ . The b-hair-pins and loop project from the trailing edge of thearrowhead and the carboxyl-terminal helix projectsfrom the tip. The structure of the HIV-1 CA coredomain differs signi®cantly from other RNA viruscoat protein structures (Hogle et al., 1985;Rossmann, 1988, 1989; Rossmann et al., 1985;Rossmann & Johnson, 1989; Valegard et al., 1990;Zhao et al., 1996), and also differs signi®cantlyfrom models used for epitope mapping and tostimulate drug design (Argos, 1989; Coates et al.,1987; Langedijk et al., 1990; Robert-Hebmann et al.,1992; Rossmann, 1988).

The CA core domain contains a conservedamino-terminal proline residue which forms a saltbridge with a conserved, buried aspartate residue(Asp51). This salt bridge can not exist in the Gagprecursor polyprotein, and it was proposed thatthe amino-terminal b-hairpin only forms upon

b-hairpin itself may therefore be important fortriggering condensation of the core particle (Gittiet al., 1996; Figure 8). Two groups have obtainedevidence in support of this hypothesis. First, thecrystal structure of the CA core domain complexedwith human cyclophilin A reveals an extensiveCA-CA interface involving the amino-terminalb-hairpin (Gamble et al., 1996). Sundquist and co-workers also showed with an in vitro assemblysystem that, whereas native HIV-1 CA moleculesassemble into tubular structures, CA constructscontaining as few as four MA residues appendedto their amino termini assemble into spheres (vonSchwedler et al., 1998). Tubes could be obtained ifthe appended MA residues were cleaved with theHIV-1 protease. Substitution of Asp51 by Ala alsoprevented tube formation, and this same substi-tution in a viral clone led to the formation of non-infectious particles that lacked the normal conicalcapsid. Krausslich and co-workers also demon-strated that N-terminal extensions to HIV-1 CAprevented tube formation in their in vitro assemblysystem (Gross et al., 1998). These results providestrong support for the hypothesis (Gitti et al., 1996)that the N-terminal b-hairpin, which can only formafter proteolytic cleavage of the MA-CA junction,plays an important role in directing proper capsidassembly.

The binding site for cyclophilin A is located onthe exposed loop and encompasses the essentialproline residue, Pro90. In the free monomericdomain, Pro90 adopts both cis and trans confor-mations due to the ¯exibility of the exposed loop.

Figure 8. Proposed structuralchanges in the core domain of CAthat accompany proteolytic clea-vage of the MA-CA junction. Theamino-terminal b-hairpin, whichforms only after proteolysis, maytrigger CA-CA interactions andpromote condensation of the capsidcore during virus maturation (fromGamble et al. (1996), with per-mission).

The cis conformer is present in low abundance(ca. 14 %), and was not observable in the crystal

logical relevance of this disul®de bond has notbeen established, although the high conservation of

Figure 9. Model of the dimeric HIV-1 capsid protein,constructed from the independently characterizedN-terminal core (purple) and C-terminal oligomerization(cyan) domains. The residues that link the domains aredisordered in the X-ray and NMR structures of theisolated domains, and could allow ca. 90 � reorientationsof the N-terminal domains in the intact protein (fromGamble et al. (1997), with permission).


structures. It is not clear if this conformationalequilibrium is affected by cyclophilin binding, or ifit is even functionally relevant. The conformationalequilibrium affects a few neighboring residues anddoes not appear to have long-range structuralconsequences (Gitti et al., 1996). Using their in vitroassembly assay, Krausslich and co-workers foundthat tube formation by HIV-1 CA could be inhib-ited by the equimolar addition of cyclophilin A.However, when CypA was added at a ratio of 1:10(CypA:CA; equal to that found in mature virions),signi®cantly longer tubes and less aggregationwere obtained. These ®ndings suggest that CypAmay serve as a molecular chaperone, facilitatingcorrect capsid condensation during viral matu-ration (Gross et al., 1998) and perhaps destabilizingthe capsid shell during viral entry and uncoating.Details of the intermolecular interactions have beenobtained from X-ray structural studies of the CAcore-CypA complex (Gamble et al., 1996). Interest-ingly, Pro90 binds to the CypA active site in anunprecedented trans conformation. CypA mayfunction by destabilizing speci®c CA-CA inter-actions, thus promoting disassembly of the viruscore during infectivity (Gamble et al., 1996).

The crystal structure of the carboxyl-terminaloligomerization domain of HIV-1 CA was deter-mined recently for constructs comprising residues146-231 and 151-231. Both constructs crystallize assymmetrical homodimers, with each monomercontaining an extended amino-terminal strandfollowed by four helices (Figure 1; Gamble et al.,1997). The two constructs exhibit different dimeri-zation interfaces with, however, the dimerizationhelices (helix 2) aligned in a parallel fashion in thelonger construct and tilted relative to each other by30 � in the shorter construct. The major homologyregion (MHR), a sequence that is highly conservedin all retroviruses (except the spumaviruses), isremoved from the dimer interface and does notparticipate in intermolecular interactions. Instead,the MHR forms an intricate network of hydrogenbonds between the strand and the ®rst twoa-helices. The N-terminal residues of the C-term-inal domains are oriented in a manner that wouldallow packing interactions between the N-terminalcore domains, and a proposed model of the intactHIV-1 CA dimer is shown in Figure 9.

Equilibrium sedimentation experiments revealedthat the longer construct has a dissociation con-stant of 10(�3) mM, whereas the shorter constructdid not appear to dimerize in solution, even atconcentrations of 100 mM (Gamble et al., 1997).These ®ndings are explained by a higher-resolutionX-ray structure re®nement of the CA C-terminaldomain, which revealed that the additionalN-terminal residues of the longer construct form a310 helix that participates in intermolecular contacts(Worthylake et al., 1998). Interestingly, in both con-structs, conserved residues Cys198 and Cys218form an intramolecular disul®de bond. The physio-

these two cysteine residues among retroviralcapsid proteins suggests that oxidation of thesecysteine residues to the disul®de may be animportant process in the viral replication cycle.

Efforts are currently underway in several labora-tories to determine how the CA molecules assem-ble to form the capsid core. Analyses of electronmicroscopic (EM) images of virus-like particlesformed by Gag precursors (Hockley et al., 1994;Nermut et al., 1993; Nermut & Thomas, 1994), aswell as Gag precursor and CA proteins bound tolipid monolayers (Barklis et al., 1998, 1997), havedemonstrated that these proteins can assemble intofullerene-like lattices containing cages of hexamericrings. Recent cryo-EM studies of hollow cylindersformed by CA (Ehrlich et al., 1992) revealed thatthe cylinders consist of a helical lattice of CAhexamers (W. I. Sundquist, unpublished results). Inaddition, Sundquist and co-workers recently dis-covered that CA-NC fusion proteins can assembleinto cones that are similar in appearance to auth-entic viral cores, and these workers have proposeda geometric model of the capsid core structure(Figure 10; W. I. Sundquist, personal communi-cation).

Nucleocapsid

Most of the known functions of the HIV nucleo-capsid protein involve interactions with nucleicacids. As a domain within the Gag precursor, NC

functions in the recognition and packaging of theviral genome (Aldovini & Young, 1990; Dorfman

but unknown roles in viral assembly and also inthe early stages of the viral infection cycle

Figure 10. Model for the symmetry of the HIV-1 cap-sid core proposed by Sundquist and co-workers. Thecone is formed by a P6 helical array of capsid hexamers,and is capped on each end by pentameric ``defects''(W. I. Sundquist, unpublished results).


et al., 1993; Gorelick et al., 1993, 1990; Julian et al.,1993) and packaging of reverse transcriptionprimer tRNALys3 (De Rocquigny et al., 1992).Subsequent to proteolytic processing, NC forms aribonucleoprotein complex in the mature virion(Aronoff et al., 1993), initiates reverse transcriptionby annealing the tRNALys3 primer to the viral gen-ome (Barat et al., 1993), facilitates elongation ofproviral DNA by reducing reverse transcriptasepausing at stable stem-loop sites (Wu et al., 1996),and stabilizes the ®nal, proviral DNA (Lapadat-Tapolsky et al., 1993). NC is also important forviral particle formation (Franke et al., 1994a; Zhang& Barklis, 1997), and may function through theformation of interprotein NC-NC interactions(Zang et al., 1998).

Except for the spumaviruses, all retroviral NCproteins contain one or two copies of a conserved``CCHC array'' (Cys-X2-Cys-X4-His-X4-Cys; whereX � variable amino acid residue; Henderson et al.,1981), sometimes referred to as ``zinc knuckle'' or``zinc ®nger like'' array (Berg, 1986) that binds zincto form a stable three-dimensional mini-globulardomain (Chance et al., 1992; DeÂmeÂneÂ et al., 1994b;Green & Berg, 1989, 1990; MeÂly et al., 1991, 1996,1993a,b; Morellet et al., 1994, 1992; Omichinski et al.,1991; South et al., 1991, 1990, 1989; South &Summers, 1993; Summers, 1991; Summers et al.,1992, 1990; Surovoy et al., 1993, 1992).

The NC protein from the human imunode®-ciency virus type-1 (HIV-1) contains two CCHCarrays separated by a short ``linker sequence'',RAPRKKG. The amino-terminal domain (F1) isessential for genome recognition, whereas thecarboxyl-terminal domain (F2) plays important

(Dannull et al., 1994; Gorelick et al., 1993;Tanchou et al., 1998). The domains have beensuccessfully targeted in vitro by nitroso-containingantiviral agents that eject zinc from the CCHCzinc knuckles (Rice et al., 1993a,b; Yu et al., 1995),and several new classes of zinc-ejecting agentswith potential chemotherapeutic utility have beenrecently identi®ed (Rice et al., 1993b, 1995; Rice &Turpin, 1996; Rice et al., 1997), two of which arein AIDS clinical trials (Shailer et al., 1997;Vandevelde et al., 1996).

Atomic-level three-dimensional structures ofsynthetic peptides with sequences correspondingto the HIV-1 NC zinc knuckle domains(Omichinski et al., 1991; South et al., 1991;Summers et al., 1990), and structures of intactnucleocapsid proteins from HIV-1 (Morellet et al.,1992; Summers et al., 1992; Figure 1), and theMoloney murine leukemia virus (MoMuLV;DeÂmeÂneÂ et al., 1994b), which contains a singleCCHC zinc knuckle, have been determined bynuclear magnetic resonance (NMR) methods. Inthese cases, the CCHC zinc knuckle domains adoptsimilar three-dimensional folds. The amino-term-inal residues X(i ÿ 1)-C(i)-X(i � 1)-X(i � 2)-C(i � 3)-G(i � 4)-X(i � 5) form a metal-coordinating reverseturn termed a ``rubredoxin knuckle'' (Blake &Summers, 1994; Summers et al., 1990) due to itshigh similarity to metal-coordinating substructuresobserved originally in the iron domain of rubre-doxin (Adman et al., 1975; Watenpaugh et al., 1973;although NH±S hydrogen bonds were notobserved in the MoMuLV structure (DeÂmeÂneÂ et al.,1994b)). Subsequent residues form a loop, whichleads to a carboxyl-terminal 310 helix.

Roques and co-workers were the ®rst to detectweak nuclear Overhauser enhancement (NOE)interactions between the two knuckles (DeÂmeÂneÂet al., 1994a; MeÂly et al., 1996; Morellet et al., 1992),and NMR relaxation studies indicate that these arethe result of a transient interaction between thetwo domains (Lee et al., 1998). The solution beha-vior of NC may be best considered as a rapidequilibrium between conformations with weaklyinteracting and non-interacting knuckle domains,and this inherent conformational ¯exibility may beimportant for adaptive binding of NC to differentnucleic acid targets.

There is now strong evidence that genome recog-nition and packaging is mediated by interactionsbetween the NC zinc knuckles (in the context ofthe Gag precursor) and a �120 nucleotide regionof the unspliced viral RNA known as the -sitelocated between the 50 long terminal repeat (LTR)and the gag initiation codon. Mutations that abol-ish zinc binding result in the production ofnon-infectious virions that lack their genomes(Berkowitz et al., 1995; Dupraz et al., 1990; Gorelicket al., 1993, 1988, 1990; Jentoft et al., 1988; MeÂric &Goff, 1989; MeÂric et al., 1988). Mutations of conser-vatively substituted hydrophobic residues within

the CCHC arrays (MeÂric & Goff, 1989), or the sub-stitution of NC domains from different retroviruses

packs against the F1 knuckle and forms a hydro-e

Figure 12. Portion of the NC-SL3 structure showinginteractions between guanosine-9 of the tetraloop andthe amino-terminal zinc knuckle. The guanosine nucleo-base packs within a hydrophobic cleft formed by conser-vatively substituted amino acid residues and formshydrogen bonds with exposed backbone NH and Oatoms located at the bottom of the cleft.

Structural BiologyofHIV 13

(Berkowitz et al., 1995), can alter RNA packagingspeci®city. The HIV-1 -site contains four stem-loops (SL1-SL4) that are important for ef®cientencapsidation (Clever et al., 1995; Clever &Parslow, 1997; Harrison & Lever, 1992; Hayashiet al., 1992, 1993; McBride & Panganiban, 1996;Sakaguchi et al., 1993).

Initial structural studies of Gag-genome recog-nition focused on interactions between the iso-lated NC protein and the SL3 stem-loop (DeGuzman et al., 1998), a highly conserved element(Hayashi et al., 1993) that is suf®cient to directthe recognition and packaging of heterologousRNAs (Hayashi et al., 1992). The protein interactspredominantly with the loop nucleotides(Figure 11). The F1 and F2 zinc knuckles interactwith G9 and G7, respectively, with the guanosinebases binding within hydrophobic clefts formedby conservatively substituted amino acid resi-dues. The guanosine O6 and H1 atoms formhydrogen bonds with backbone NH and O atomslocated at the bottom of the hydrophobic clefts(Figure 12). These interactions are essentiallyidentical with those observed previously in anisolated zinc knuckle-DNA complex (South &Summers, 1993), suggesting that retroviral zincknuckles function primarily by binding speci®-cally to guanosine bases. The A8 nucleobase

Figure 11. Structure of the complex formed betweenthe HIV-1 nucleocapsid protein and the SL3 stem-looprecognition element of the genomic -RNA packagingsignal. The coloring scheme of the tetraloop bases is asfollows: G6, green; G7, pink; A8, blue; G9, orange. Thecoloring scheme for the NC protein is: N-terminal 310

helix, pink; N-terminal zinc knuckle, cyan; linker, red;C-terminal zinc knuckle, green; cysteine and histidineside-chains, yellow and cyan, respectively.

gen bond with the side-chain N -H proton of thehighly conserved residue, Arg32. In addition,upon binding to SL3, the amino-terminal tail ofNC undergoes a transition from a random coil toa 310 helix, with conserved residue Asn5 formingspeci®c hydrogen bonds to C11 in the RNAmajor groove.

The mechanism of retroviral genome recognitionis complex and dif®cult to study directly. AlthoughSL3 is suf®cient to direct packaging of heter-ologous RNAs (Hayashi et al., 1992), other stem-loops of the -RNA are also important (Cleveret al., 1995; Clever & Parslow, 1997; McBride &Panganiban, 1996), and it is likely that in vivopackaging involves more than one Gag polypro-tein. Since intact Gag forms oligomers, onepotential model for recognition involves the simul-taneous interactions of two or more NC domainswith two or more -RNA stem-loop recognitionelements. In this regard, the inherent ¯exibility ofNC may be important for adaptive binding todifferent RNA targets via different subsets of inter-and intra-molecular interactions (De Guzman et al.,1998).

Very recently, the NC proteins from MasonP®zer monkey virus (MPMV; Gao et al., 1998)and MMTV (M. F. Summers, unpublished results)have been solved. Although the N-terminal zincknuckles adopt structures similar to thosedescribed above, the C-terminal knuckles containadditional structure, with residues that follow theCCHC arrays folded into a reverse turn thatpacks against the knuckle. It is unclear how or ifthese differences affect the nucleic acid interactive

properties of the NC protein from HTLV-1,MPMV and related D-type retroviruses (Gao et al.,

is stabilized by a four-stranded antiparallel b-sheetformed by N- and C-terminal b-strands. The


1998).

Viral Enzymes

Protease

Protease was the ®rst HIV-1 protein to bestructurally characterized (Miller et al., 1989b;Navia et al., 1989; Wlodawer et al., 1989), and sev-eral excellent reviews of the biochemistry andstructural biology of HIV-1 PR have appeared (forexample, that of Swanstrom & Wills (1997). As forother lentiviruses, PR is translated as a gag-polfusion product that is produced from a ribosomalframeshift (Jacks et al., 1988), and is released fromthe Gag-Pol precursor protein by an autocatalyticmechanism (Debouck et al., 1987; Farmerie et al.,1987). From 1989 to 1993, more than 160 crystalstructures of HIV-1 PR and PR-inhibitor complexeswere determined (Wlodawer & Erickson, 1993),most of which were used to guide the develop-ment of protease inhibitors. Indeed, these effortsserve to illustrate the enormous potential andutility of the rational structure-based drug designstrategy.

PR is a symmetrical homodimer that is structu-rally similar to other aspartyl proteases of the pep-sin family (Navia et al., 1989; Wlodawer et al.,1989), as well as those from other retrovirusesincluding RSV (Miller et al., 1989a), HIV-2(Mulichak et al., 1993; Tong et al., 1993), SIV (Roseet al., 1993; Zhao et al., 1993), FIV (Wlodawer et al.,1995) and EIAV (Gustchina et al., 1996). The dimer

Figure 13. HIV-1 protease free (a) and complexed withconformational change in which the ``¯aps'' close down ovepermission).

enzyme active site is formed at the interface of thetwo subunits and contains a catalytic triad (Asp25-Thr26-Gly27) responsible for the cleavage reactionsof PR. Each monomer contains a ``¯ap'' comprisingtwo antiparallel b-strands connected by a b-turn(residues 49 to 52) and situated on top of thecatalytic site (Figure 13). The conformation of the¯ap differs signi®cantly in the PR and PR-inhibitorcomplexes, with some backbone Ca atoms beingdisplaced by up to 7 AÊ (Miller et al., 1989b;Figure 13).

NMR studies con®rmed that the protease ¯apsare indeed ¯exible, and suggest that this ¯exibilitymay be important for enzyme activity (Nicholsonet al., 1995; Yamazaki et al., 1996). Although thenative protein is a symmetric dimer, inhibitorsbind in an asymmetric manner, as must the naturalprotein substrates. In many of the early crystalstructures, a water molecule was observed to par-ticipate in hydrogen-bonding interactions betweenthe ¯aps of the protein and the inhibitors(Wlodawer & Erickson, 1993). This structural watermolecule could be displaced by a carbonyl groupin an appropriately designed inhibitor (Lam et al.,1994). NMR studies con®rmed the displacement ofthe water molecule (Wang et al., 1996), and alsoveri®ed the presence of other structurally import-ant water molecules (Grzesiek et al., 1994).

Reverse transcriptase

Reverse transcription of the HIV-1 genomeoccurs mainly in the cytosol of the cell shortly after

an inhibitor (b). Inhibitor binding induces a signi®cantr the active site (from Wlodawer & Erickson (1993), with

viral entry. DNA synthesis proceeds within areverse transcription complex that is poorly under-

DNA Pol I, suggesting a possible conservativestructural motif in all polymerases (Kohlstaedt

Figure 14. Ribbon drawings of the p51 and p66 pro-teins that form the HIV-1 RT heterodimer. Subdomainsare colored as follows: ®ngers, blue; palm, red; thumb,green, and connection, yellow (from Jacobo-Molina et al.(1993), with permission)


stood but probably comprises several viral proteinsincluding MA, NC and perhaps Nef (Schwartzet al., 1995) and Vif (Goncalves et al., 1996; Sova &Volsky, 1993; von Schwedler et al., 1993). Themechanism of reverse transcription is complex(Gilbona et al., 1979), and the interested reader isreferred to the following reviews (Arts &Wainberg, 1996; Telesnitsky & Goff, 1997; Varmus& Swanstrom, 1984; Whitcomb & Hughes, 1992).

The ®rst drugs approved by the FDA for thetreatment of AIDS function by targeting RT.Nucleoside inhibitors such as AZT and dideoxy-inosine (ddI) and dideoxycytidine (ddC) areincorporated during reverse transcripton and resultin the termination of viral DNA synthesis.A second class of non-nucleoside RT inhibitorsfunctions by binding directly to the enzyme, inhi-biting catalysis without blocking substrate binding.

RT is initially packaged into virions as a Gag-Polprecursor, with proteolytic cleavage initially pro-ducing a homodimer of two p66 molecules. p66contains both a polymerase and an RNase Hdomain. Subsequent proteolytic removal of theRNase H domain of one of the subunits results inthe mature p66-p51 RT heterodimer.

Structural information for HIV-1 RT is availablefrom X-ray crystallographic studies of RT-inhibitor(Kohlstaedt et al., 1992, 1993; Ren et al., 1995;Smerdon et al., 1994) and RT-DNA complexes(Arnold et al., 1992; Jacobo-Molina et al., 1993), aswell as for the unliganded protein (Esnouf et al.,1995; Hsiou et al., 1996; Rodgers et al., 1995;Figure 1). A 3.5 AÊ X-ray structure by Steitz and co-workers of an RT complex with the non-nucleosideinhibitor nevirapine, revealed that the polymerasedomain is composed of four subdomains known asthe ``®ngers'', ``palm'', ``thumb'', and ``connection''(Kohlstaedt et al., 1992). The ``®ngers'' subdomainis composed of mixed b-strands and threea-helices, and the ``palm'' includes ®ve b-strandsthat form hydrogen bonds with four b-strandspositioned at the base of the ``thumb.`` A helicalbundle forms the ``thumb'' subdomain of theenzyme, and the ``connection'' subdomain, whichconnects the polymerase and RNase H domains, iscomposed of a large b-sheet and two a-helices. Thesubdomains of the p66 subunit pack together toform an ``open right-hand'' con®guration, creatinga large cleft in the polymerase site that exposes thethree catalytic residues Asp110, Asp185, andAsp186 (Figure 14, bottom). Despite having anidentical amino acid sequence, the relative subdo-main packing in p51 is dramatically different, withthe ®ngers closed over the palm in a manner thatburies the catalytic residues (Kohlstaedt et al., 1992;Figure 14, top). p51 is catalytically inactive (LeGrice et al., 1991), but interacts with the RNase Hdomain in the heterodimer and is important for theoverall RT structure. The p66 domain of HIV-1 RTis structurally similar to the DNA polymerasedomain of the Klenow fragment of Escherichia coli

et al., 1992).HIV-1 RT is conformationally ¯exible, and this

probably hampered early efforts to obtain the res-olution necessary to precisely de®ne side-chainorientations and stereochemistry. Higher-resolutionstructural information was initially obtained fromcrystallographic studies of recombinant ribonu-clease H (Davies et al., 1991) and ®nger-palm(Unge et al., 1994) domains, and the secondarystructure (Powers et al., 1991) and backbonedynamics (Powers et al., 1992) of the RNase Hdomain were also reported. More recently, high-resolution (to 2.2 AÊ ) structural information hasbeen obtained for non-nucleoside inhibitor com-plexes with intact RT (Ren et al., 1995), as well asfor unliganded RT (Esnouf et al., 1995; Hsiou et al.,1996). Comparison of these structures, includingdifferent crystal forms, revealed that the protein is

indeed conformationally ¯exible, with the orien-tation of the thumb domain being dependent on

Figure 1). Interestingly, although the intact HIV-1integrase appears to function as a tetramer (Jenkins


the hydration state of the crystal lattice rather thanon the binding or nature of non-nucleoside inhibi-tors (Esnouf et al., 1995). Interestingly, the bindingof non-nucleoside inhibitors results in a shift ofthree b-strands that contain active site aspartylresidues to a conformation that is similar to thatobserved in the inactive p51 subunit. Thus, thenon-nucleoside inhibitors appear to function bymimicking protein-protein interactions of the p51subunit that distort the active site of the enzyme(Esnouf et al., 1995).

A 3.0 AÊ crystal structure of an HIV-1 RT-DNAduplex complex (which also contains a boundmonoclonal antibody Fab fragment) revealed thatDNA binding does not signi®cantly alter the struc-ture of the protein (Jacobo-Molina et al., 1993).Interestingly, binding results in a �45 � bend in theDNA that bridges A- and B-form conformations.Of course, the enzyme functions by interacting notwith duplex DNA, but with RNA-RNA and RNA-DNA duplexes, which would likely adopt A-helicalstructures. The majority of the RT-DNA contactsinvolve residues of the p66 ®ngers, palm andthumb subdomains, with the palm and thumb act-ing as a clamp that positions the DNA relative tothe active site residues (Jacobo-Molina et al., 1993).The 30-OH of the primer strand is positioned closeto the polymerase active site, consistent withmechanistic hyphotheses.

Integrase

HIV-1 integrase (IN) is essential for incorpor-ation of the viral DNA into the chromosomal DNAof the target cell. Extensive studies of the mechan-ism of proviral DNA integration are beyond thescope of this review, but have been summarizedrecently (Brown, 1997). Brie¯y, as part of the prein-tegration complex, IN recognizes long terminalrepeats (LTRs) at the 50 and 30 ends of the newlysynthesized viral DNA duplex and cleaves two (orsometimes three) bases from the 30 ends. IN thenligates the 30 ends to the cellular DNA in thenucleus, with reactions appearing to occur prefer-entially at sites with highly bent DNA. The result-ing unligated 30 ends of the cellular DNA aresubsequently extended to ®ll gaps, and additionalprocessing leads to the complete covalent incorpor-ation of the proviral DNA.

HIV-1 IN consists of three separate structuraland functional domains, including an N-terminalzinc binding domain that facilitates oligomeriza-tion, a central catalytic core domain, and a C-term-inal DNA-binding domain. The structure of thecatalytic domain of IN (residues 50-212) was deter-mined by Davies and co-workers using X-ray crys-tallographic methods (Dyda et al., 1994), and NMRmethods were used to determine the solution-statestructures of the N-terminal zinc binding (Cai et al.,1997) and C-terminal DNA-binding domains(Eijkelenboom et al., 1995; Lodi et al., 1995;

et al., 1996; Zheng et al., 1996), each of the isolateddomains forms stable dimers. Peptides that inhibitIN activity may function by binding to exposedhydrophobic residues via coiled-coil interactionsand inhibiting higher-order oligomerization(Sourgen et al., 1996).

The N-terminal domain of IN is composed offour a-helices (per monomer), with residues His12,His16, Cys40, and Cys43 forming a tetrahedralcoordination site that binds one atom of zinc. Themonomer exists in two different conformations,designated D and E, which arise from differentmodes of zinc binding to His12. Thus, in theD-conformation, His12 binds zinc via the Nd1 ringnitrogen atom and in the E-conformation, zinccoordination occurs via the His12-Ne2, and theequilibrium constant of this interconversion is closeto one at room temperature. The D-form predomi-nates at higher temperatures, and this formappears to contain a partially disordered helix-1.The physiological relevance (if any) of this equili-brium is not known, although it results in smalllong-range perturbations of surface residues. Inter-estingly, helices 2 and 3 form a helix-turn-helix(HTH) substructure that is similar to thoseobserved in several classes of DNA binding pro-teins. However, in IN, helix 3 participates inintermolecular protein-protein interactions,whereas the equivalent helices in HTH proteinsbind DNA (Cai et al., 1997).

Attempts to crystallize the native core domain ofIN were confounded by poor solubility. However,a single amino acid residue substitution,Lys185Phe, resulted in a more soluble protein thatretained wild-type levels of catalytic activity andwas amenable to crystallization (Dyda et al., 1994).This mutant core domain crystallized as a dimer,with each monomer containing a ®ve-strandb-sheet and six a-helices (Figure 1). Interestingly,the structure is topologically similar to a variety ofother enzymes that perform similar functions,including the ribonuclease H (RNase H) domain ofHIV-1 reverse transcriptase, the Holiday junctionresolvase, and the core domain of a transposase(Dyda et al., 1994). Three highly conserved residuesthat are essential for catalysis, Asp64, Asp116, andGlu152, form a ``D,D-35-E'' motif located in thecore domain of the enzyme (Dyda et al., 1994).Although there is little sequence similarity betweenthe IN core domain and the RT RNase H domain,two of the catalytic acidic residues (Asp64 andAsp116 of IN and Asp443 and Asp498 of RT)occupy similar surface positions in the enzymeactive sites (Dyda et al., 1994). The structure con-tains an extensive dimer interface that is probablyalso present in the intact, native protein. However,the active sites of the dimer are separated by 35 AÊ ,which is incompatible with the ca. 15 AÊ separationexpected from most models (Dyda et al., 1994).Two active sites could be juxtaposed appropriatelyin a tetramer, with the remaining active sites ser-

ving non-catalytic roles. A model of the tetramergenerated from the independently determined

et al., 1991). Furthermore, clinical studies of long-term (>ten years) HIV-infected individuals posses-

Figure 15. Model of the HIV-1 IN tetramer generatedusing independently solved structures of the N-terminal,core, and C-terminal domains. The four molecules of thetetramer are displayed in different colors (from Cai et al.(1997), with permission).


structures of the IN domains is shown in Figure 15.Alternatively, the dimer interface could be modi-®ed in the native protein (Dyda et al., 1994).

The structure of the C-terminal DNA-bindingdomain of IN was determined independently bytwo groups using NMR methods (Eijkelenboomet al., 1995; Lodi et al., 1995). The two structuresappear to be essentially identical. Each of themonomeric subunits of the symmetrical dimer iscomposed of ®ve antiparallel b-strands arranged ina b-barrel and folded in a way that is topologicallysimilar to the Src-homology-3 (SH3) domain. Anextensive hydrophobic dimer interface is formedby the ``face-to-face'' packing of three b-sheetstrands from each monomer (Figure 1). Dimeriza-tion results in a large saddle-shaped groove thatcontains several positively charged residues,including Lys264, which has been shown to play arole in DNA binding (Puras-Lutzke et al., 1994).This protruding lysine residue may function inconcert with the surrounding positively chargedresidues to bind DNA (Lodi et al., 1995).

Accessory Proteins

Nef

HIV-1 negative factor (Nef) is a 27 kDa, N-term-inally myristoylated regulatory factor of 206 aminoacid residues that is expressed in high concen-trations shortly after viral infection (Goldsmithet al., 1995). This accessory protein is important forachieving and maintaining high viral loads in vivo.Inoculation of Rhesus monkeys with a nef-deletionmutant strain of SIV does not lead to AIDS-likedisease and actually results in long-term immunityagainst pathogenic SIV (Daniel et al., 1992; Kestler

sing apparent deletions within the nef gene shownormal CD4 levels and exhibit no signs of pro-gression to AIDS (Deacon et al., 1995; Kirchoff et al.,1995).

Nef has at least two distinct roles: it enhancesviral replication and stimulates a reduction in thenumber of CD4 receptors on the surface of theinfected cell (Goldsmith et al., 1995). Down-regu-lation of the CD4 receptor proceeds via endocytoticand degradation processes, and appears to involvethe direct interaction of Nef with residues withinthe cytoplasmic tail of CD4 (Salghetti et al., 1995).The reduction in cell-surface CD4 levels appears tobe important for preventing reinfection by buddingvirions. In addition, by eliminating premature Env-CD4 binding, Nef may increase Env incorporationinto virions and promote the release of infectiousparticles (Mangasarian & Trono, 1997).

The structure of the core domain of Nef (resi-dues �71-205) was determined in solution byNMR methods (Grzesiek et al., 1996, 1997;Figure 1), and X-ray structures of Nef-SH3 domaincomplexes (Arold et al., 1997; Lee et al., 1996), anduncomplexed Nef (Arold et al., 1997) were reportedshortly thereafter. All structural studies wereperformed with recombinant constructs that lack acentral, 14 amino acid residue segment responsiblefor aggregation. The structures are all in goodagreement. The global fold of Nef is similar to thatobserved for the family of winged helix-turn-helixDNA binding proteins, and consists of threea-helices, a ®ve-stranded antiparallel b-sheet, a left-handed poly-L-proline type-II helix (PP-II), and a310 helix (Figure 1). The PP-II helix is present at theamino terminus of the protein, and solution NMRstudies indicated that these residues participate inSH3 domain binding (Grzesiek et al., 1996, 1997).The crystal structure of the complex with a mutantFyn kinase SH3 domain provided detailed infor-mation regarding the determinants of recognitionand binding (Figure 16). In particular, the crystalstructure revealed how conserved residue Arg77provides critical hydrophobic and electrostaticintermolecular interactions (Lee et al., 1996). Themore recent crystal structure of HIV-1 Nef boundto the wild-type Fyn kinase SH3 domain (Aroldet al., 1997) is essentially identical with the earlierX-ray and NMR structures. Interestingly, the PP-IIhelix appears partially disordered in the X-raystructure of uncomplexed Nef (Arold et al., 1997),and it was proposed that the left-handed polypro-line helical structure is only fully formed uponcomplexation with the SH3 domain. This ®ndingappears to be at odds with the NMR data, whichdo not provide evidence for non-helical structuresor signi®cantly enhanced dynamics.

The type II polyproline helix is formed byresidues that comprise a consensus sequence forSH3 binding (Pro-X-X-Pro, where X is any aminoacid residue; Figure 16). These residues are essen-tial for the enhancement of viral infectivity. Thus,

in the absence of the polyproline repeat, down-regulation of the CD4 receptor is still observed,

et al., 1989; Malim et al., 1989b, 1990). One Rev pro-tein initially targets a high af®nity stem-loop

Figure 16. Stereoview of the HIV-1 Nef core domain(residues 71-120; N- and C-terminal residues in yellowand green, respectively; the broken line represents a dis-ordered loop) bound to the Fyn(R961) SH3 domain(colored blue). Side-chains are drawn for the criticaltryptophan and isoleucine residues (colored red) of theSH3 domain and the polyproline helix of Nef (from Leeet al. (1996), with permission).

Figure 17. Space-®lling representation of the Revresponse element (RRE; gray) bound to the RNA-bind-ing segment of Rev (green). The Rev peptide forms ana-helix that binds within a widened major groove.


indicating that the polyproline repeat plays no rolein the reduction of CD4 counts. In contrast, anN-terminal membrane-targeting sequence isimportant for both down regulation of the CD4receptor and enhancement of viral infectivity(Goldsmith et al., 1995).

Nef interacts with and binds speci®cally to theSH3 domain of Src family tyrosine kinases (Sakselaet al., 1995) and is also associated with Ser/Thrprotein kinase activity (Sawai et al., 1994). Theinteraction of Nef with Src family tyrosine kinasesis potentially signi®cant, since these proteins aredirectly involved in intracellular signaling path-ways (Bolen, 1993). The Pro-X-X-Pro motif isessential for maximal proliferation of the HIV-1virus in primary cell cultures, and it is thus likelythat interactions of Nef with the Src family of pro-teins may be responsible for the enhancement ofviral replication and infectivity (Saksela et al.,1995).

Rev

As described earlier, Rev participates in thesequence-speci®c transport of unspliced andincompletely spliced viral mRNAs from thenucleus to the cytoplasm. The 116 residue proteinbinds speci®cally to the ``Rev-response element''(RRE) RNA target sequence located within theunspliced viral transcripts of the env gene (Daly

(SLIIB) of the RRE (Kd ca. 4 nM) that contains apurine-rich bubble, and additonal Rev moleculesthen oligomerize on the RRE (Heaphy et al., 1991;Zapp et al., 1991).

Rev contains a basic domain (Arg35 to Arg50)that functions in nuclear localization signaling,binding of Rev to the RRE, and Rev multimeriz-ation at the recognition site (Kjems et al., 1992;Malim et al., 1989a; Malim & Cullen, 1991; Malimet al., 1989b). Residues immediately N- and C-term-inal to this arginine-rich domain contribute to Revoligomerization on the RRE but have no detectablerole in RRE binding (Malim & Cullen, 1991). TheRev activation or effector domain, a leucine-richregion extending from Leu75 to Leu83, functionsas a binding site for necessary Rev-associated fac-tors (Bogerd et al., 1995; Malim & Cullen, 1991).

The three-dimensional structure of a peptide cor-responding to the basic segment of Rev bound tothe stem-loop IIB (SLIIB) of the RRE recognitionsite was determined by Williamson and co-workersusing NMR methods (Battiste et al., 1996). Thestructure was determined to high atomic resolutionwith samples containing isotopically labeled pep-tide and RNA, and built upon previous structuralmodeling of the RNA in a complex containingunlabeled peptide (Battiste et al., 1995). Thepeptide, which contains a non-native aspartic acidresidue at its amino terminus, is largely unstruc-tured in solution (�12 % helical content), but formsan a-helix upon binding to the SLIIB RNA stem.The peptide binds within a signi®cantly widenedmajor groove that contains two purine-purinebase-pairs (G47 �A73 and G48 �G71) separated by anon-stacked bulged uridine base (U72; Figure 17).The phosphodiester backbone is severely distortedat the binding site, resulting in a locally parallel-strand orientation. Arginine and asparagine resi-dues implicated in sequence-speci®c binding onthe basis of chemical modi®cation and in vitroselection experiments interact with nucleotides inthe major groove, and one threonine and severalarginine residues make additional electrostatic orhydrogen-bonding contacts with the phosphodi-ester linkages at the periphery of the widened

major groove. For example, residue Arg38, whichcannot be substituted by lysine, appears to form

under non-denaturing conditions, contains1.6 atoms of zinc per protein and exists in solution


hydrogen bonds to the phosphodiester groups ofU66 and G67. The peptide and RNA structuresappear to be mutually stabilizing, consistent withthe view that the entire complex behaves as asingle folding unit (Tan & Frankel, 1994).

Shortly after the above work appeared, Pateland co-workers reported the NMR structure of aRev basic peptide bound to a 35-nucleotide RNAaptamer (Ye et al., 1996). The Rev-aptamer struc-ture is globally similar to the Rev-RRE. In the apta-mer complex, the a-helical Rev peptide bindswithin a widened major groove using similar sub-sets of hydrogen bonding and electrostatic inter-actions. The composition of the aptamer isdifferent from that of the RRE SLIIB, and the majorgroove of the aptamer is widened by a differentcombination of intranucleotide interactions. BothRNA structures contain purine-purine mismatches(A �A and G �A in the case of the aptamer) and alooped-out nucleotide at the peptide binding site.The A �A mismatch appears to be isosteric with theG �G mismatch in the Rev-RRE complex. Inaddition, the aptamer contains a U*AU triple thatis not present in Rev-RRE.

Tat

Transcription of the integrated proviral DNA isinitiated at the HIV-1 promoter, which is locatedin the U3 region of the 50 long terminal repeat(LTR). The promoter binds RNA polymerase II,as well as numerous other cellular factors, includ-ing NF-kB, Sp1, TBP and others (Jones &Peterlin, 1994). In the early stages of infection,transcription is terminated prematurely due toabortive elongation. As described above, Tatfunctions to enhance transcriptional elongation bybinding to the TAR (trans-activating responseelement) stem-loop site on the nascent RNA tran-script, probably as a Tat-cyclin T complex, andrecruiting Cdk9. This mode of transcriptional acti-vation differs from that of most other well-characterized transcription factors that functionby binding to the duplex DNA template. Cdk9then stimulates transcriptional elongation byphosphorylating the RNA polymerase II tran-scription complex (Herrmann & Rice, 1995;Reines et al., 1996; Wei et al., 1998).

The HIV-1 Tat sequence consists of four differentregions that share homologies with the Tatproteins of other lentiviruses (Dorn et al., 1990),including (from N to C terminus) the cysteine-rich,core, basic, and glutamine-rich segments. The roleof the cysteine-rich segment is unknown. Interest-ingly, some lentiviruses, such as EIAV, encode Tatproteins that do not contain a cysteine-richsequence (Dorn et al., 1990). Early reportssuggested that this domain binds zinc and stimu-lates Tat dimerization (Frankel et al., 1989). How-ever, more recent evidence indicates thatrecombinant and functional Tat protein, isolated

as a mixture of monomer and higher-order aggre-gate species (Slice et al., 1992). Still others havesuggested that Tat functions as a monomer (Rice &Chan, 1991), does not require zinc, and that thecysteine residues may exist as disul®de groups(Koken et al., 1994).

HIV-1 Tat has a strong tendency to oxidize, andthis is the likely reason that neither the intact pro-tein nor the cysteine-rich array have been struct-urally characterized to high atomic resolution. The75 amino acid residue Tat protein from EIAV,which lacks a cysteine-rich segment, has been mod-eled using NMR-derived distance restraints andmolecular dynamics simulations (Willbold et al.,1994). Residues Tyr35-Tyr49 of the core sequencefold to form a hydrophobic core, which appears tobe stabilized mainly by the packing of hydro-phobic side-chains. The ¯anking basic and gluta-mine-rich sequences appear highly ¯exible. Indeed,the entire structure is apparently ¯exible, based onthe lack of slowly exchanging backbone amide pro-tons (Willbold et al., 1994). Similar studies of HIV-1Tat suggest that the core domains adopt aconserved three-dimensional structure (Bayer et al.,1995). The HIV Tat core domain was not de®nedas precisely as the EIAV core, due possibly to thefact that the EIAV core contains a stabilizing disul-®de bond whereas the HIV Tat core domain doesnot.

The basic segment of Tat (R49KKRRQRRR57)is essential for recognition and binding to TARRNA. Frankel and colleagues showed that substi-tution of the basic residues by a stretch of ninearginine residues does not affect in vivo transacti-vation, whereas substitution by a stretch of ninelysine residues leads to a 100-fold decrease inactivity (Calnan et al., 1991). By systematicallysubstituting arginine residues back into the nine-lysine-residue mutant, they found that a singlearginine residue at either position 52 or 53 is suf-®cient to restore wild-type levels of transactiva-tion activity. Ethylation interference experimentssuggested that a single arginine residue bindssimultaneously to two phosphodiester groups ina manner referred to as the ``arginine fork''(Calnan et al., 1991).

Information regarding the structure of the HIV-1Tat basic domain is currently somewhat controver-sial. NMR studies of a biologically active 25-resi-due peptide containing the basic, RNA-bindingsegment of HIV-1 Tat linked to the core domain ofEIAV Tat suggested that the basic RNA-bindingregion forms an a-helix (Mujeeb et al., 1994).Although the basic residues of intact EIAV Tatexhibited weak evidence of a-helix formation(Willbold et al., 1994), the analogous residues of anintact HIV-1 Tat protein appeared disordered(Bayer et al., 1995). However, recent high-resol-ution structural studies of HIV-1 Tat basic peptide-TAR RNA complexes indicate that the basic

domain actually adopts a b-hairpin upon bindingto TAR (Ye et al., 1995; Figure 18).

mide complexes with the TAR RNAs of bovineimmunode®ciency virus (BIV; Ye et al., 1995) and

Figure 18. Space-®lling representation of the bovineimmunode®ciency virus (BIV) Tat peptide bound to theBIV TAR RNA. As for the HIV-RRE, binding occurs in awidened major groove. However, the BIV Tat peptidebinds in a b-hairpin conformation.


Although high-resolution structural informationis not available for HIV-1 Tat, much informationhas been obtained about the structure and inter-actions of its TAR RNA target. NMR-based struc-tural studies of a complex formed between the31-nucleotide TAR RNA and argininamide, theminimalist model of the Tat basic site, providedsupport for the arginine fork model (Puglisi et al.,1992). These studies revealed that the TAR RNAundergoes a major conformational change uponbinding the argininamide ligand, leading to astable RNA fold in which the argininamide side-chain forms hydrogen bonds to the G26 nucleobaseof the G26 �C39 base-pair, and also interacts withphosphodiester groups P22 and P23 (Puglisi et al.,1992). In addition, nucleotides U38, A27 and U23were proposed to form a base-triple upon arginin-amide binding. Formation of the base-triple resultsin increased exposure of the major groove andenhanced accessibility of G26 to the guanidiniumgroup of arginine.

In subsequent NMR studies of TAR-arginin-amide and TAR-Tat peptide complexes, Varaniand co-workers suggested that nucleotides U38,A27 and U23 do not form a base-triple (Aboul-elaet al., 1995). The Tat peptide employed in these stu-dies contained both the basic and core regions ofTat, and although the quality of the data was notsuf®cient to allow a high-resolution structuredetermination of the bound peptide, the studiesdid reveal that core residues participate directly inTat-RNA interactions (Aboul-ela et al., 1995). Thenature of the discrepancies between the TAR-argi-ninamide structures determined by the Varani(Aboul-ela et al., 1995) and Williamson (Puglisiet al., 1992) laboratories is not clear. Although thelatter structure was determined with signi®cantlymore experimentally determined distancerestraints, very recent NMR studies of arginina-

HIV-2 (Brodsky & Williamson, 1997) are consistentwith the originally proposed U38-A27-U23 base-tri-ple. In addition, an isomorphic HIV-2 TAR mutantforms a C38-G27-C23� triple upon argininamidebinding in the expected pH-dependent manner(Brodsky et al., 1998).

The HIV-1 TAR RNA appears to be very sen-sitive to divalent cations. A recent crystal struc-ture revealed that calcium ions can induce aTAR conformation that is similar in somerespects with (but not identical with) that of theargininamide-bound structure (Ippolito & Steitz,1998).

Vpr, Vpu and Vif

No high-resolution structural information isavailable for these accessory proteins.

Future Directions

Although tremendous progress has been madeover the past ten years in characterizing thestructures of proteins from HIV-1, many unan-swered questions remain. Some proteins, such asthe protease, have received detailed structuralcharacterization. However, for most proteins,structures were determined for isolated subdo-mains or deletion mutants that sometimes lackedfunctionally important groups. No atomic-levelstructural information is available for the acces-sory proteins Vpr, Vpu and Vif. In addition,although the accessory proteins Tat and Revhave been modeled on the basis of limited struc-tural data, high-resolution structures of theseproteins have not been reported. The structureof the Gag precursor polyprotein, which couldprovide important insights into the mechanismof virus assembly and maturation, is alsounknown.

In addition to completing structural studies ofisolated viral components, future efforts willneed to focus on functional intermolecular inter-actions. Although the molecular determinants ofseveral key interactions have been addressed,such as those associated with target cell recog-nition and penetration, genome recognition andpackaging, and the regulated transcription ofthe integrated viral DNA, complete understand-ing of the structural biology associated withthese and other processes has not beenrevealed. In addition, understanding the inter-molecular interactions associated with capsidcore assembly and disassembly could lead tothe development of antiviral agents designed toinhibit these processes, as has been achieved forother viruses (e.g. the ``WIN'' compounds thattarget the capsid of rhinovirus; Smith et al.,1986). The recent preparation of core-like par-ticles by Sundquist and co-workers offers theexciting possibility that structural information

for the capsid core may become available in thenot-too-distant future.

proteins on a lipid monolayer. J. Biol. Chem. 273,


Acknowledgments

This work was supported in part by NIH grantsAI30917 and GM42561. B.G.T. is a Meyerhoff Scholar atUMBC.

References

Aboul-ela, F., Karn, J. & Varani, G. (1995). The structureof the human immunode®ciency virus type-1 TARRNA reveals prinicples of RNA recognition by Tatprotein. J. Mol. Biol. 253, 313-332.

Adman, E., Watenpaugh, E. D. & Jensen, L. H. (1975).NHÐS hydrogen bonds in Peptococcus aerogenesferredoxin, Clostridium pasteurianum rubredoxin, andChromatium high potential iron protein. Proc. NatlAcad. Sci. USA, 72, 4854-4858.

Aldovini, A. & Young, R. A. (1990). Mutations of RNAand protein sequences involved in human immuno-de®ciency virus type 1 packaging result inproduction of noninfectious virus. J. Virol. 64, 1920-1926.

Argos, P. (1989). A possible homology between imuno-de®ciency virus p24 core protein and picornaviralVP2 coat protein: prediction of HIV p24 antigenicsites. EMBO J. 8, 779-785.

Arnold, E., Jacobo-Molina, A., Nanni, R. G., Williams,R. L., Lu, X., Ding, J., Clark, A. D., Zhang, A.,Ferris, A. L., Clark, P., Hizi, A. & Hughes, S. H.(1992). Structure of HIV-1 reverse transcriptase/DNA complex at 7 AÊ resolution showing active sitelocations. Nature, 357, 85-89.

Arold, S., Franken, P., Strub, M.-P., Hoh, F., Benichou,S., Benarous, R. & Dumas, C. (1997). The crystalstructure of HIV-1 Nef protein bound to the Fynkinase SH3 domain suggests a role for this complexin altered T cell receptor signaling. Structure, 5,1361-1372.

Aronoff, R., Hajjar, A. M. & Linial, M. L. (1993). Avianretroviral RNA encapsidation: reexamination offunctional 50 RNA sequences and the role of nucleo-capsid Cys-His motifs. J. Virol. 67, 178-188.

Arthur, L. O., Bess, J. W., Jr, Sowder, R. C. I.,Benveniste, R. E., Mann, D. L., Chermann, J.-C. &Henderson, L. E. (1992). Cellular proteins bound toimmunode®ciency virus: implications for pathogen-esis and vaccines. Science, 258, 1935-1938.

Arts, E. J. & Wainberg, M. A. (1996). Human immuno-de®ciency virus type 1 reverse transcriptase andearly events in reverse transcription. Advan. VirusRes. 46, 97-163.

Barat, C., Schatz, O., Le, Grice S. & Darlix, J.-L. (1993).Analysis of the interactions of HIV1 replication pri-mer tRNA-Lys,3 with nucleocapsid protein andreverse transcriptase. J. Mol. Biol. 231, 185-190.

Barklis, E., McDermott, J., Wilkens, S., Schabtach, E.,Schmid, M. F., Fuller, S., Karanjia, S., Love, Z.,Jones, R., Rui, Y., Zhao, X. & Thompson, D. (1997).Structural analysis of membrane-bound retroviruscapsid proteins. EMBO J. 16, 1199-1213.

Barklis, E., McDermott, J., Wilkens, S., Fuller, S. &Thompson, D. (1998). Organization of HIV-1 capsid

7177-7180.BarreÂ-Sinoussi, F., Chermann, J. C., Rey, F., Nugeyre,

M. T., Chamaret, S., Gruest, J., Dauguet, C., Axler-Blin, C., VeÂzinet-Brun, F., Rouzioux, C.,Rozenbaum, W. & Montagnier, L. (1983). Isolationof a T-lymphotropic retrovirus from a patient atrisk for acquired immune de®ciency syndrome(AIDS). Science, 220, 868-871.

Battiste, J. L., Tan, R., Frankel, A. D. & Williamson, J. R.(1995). Assignment and modeling of the Revresponse element RNA bound to a Rev peptideusing 13C-heteronuclear NMR. J. Biomol. NMR, 6,375-389.

Battiste, J. L., Mao, H., Rao, N. S., Tan, R., Muhandiram,D. R., Kay, L. E., Frankel, A. D. & Williamson, J. R.(1996). a-Helix major groove recognition in an HIV-1 Rev peptide-RRE RNA complex. Science, 273,1547-1551.

Bayer, P., Kraft, M., Ejchart, A., Westendorp, M., Frank,R. & Rosch, P. (1995). Structural studies of HIV-1Tat protein. J. Mol. Biol. 247, 529-535.

Beissinger, M., Paulus, C., Bayer, P., Wolf, H., Rosch, P.& Wagner, R. (1996). Sequence-speci®c resonanceassignments of the 1H-NMR spectra and structuralcharacterization in solution of the HIV-1 transframeprotein p6*. Eur. J. Biochem. 237, 383-392.

Bennett, R. P., Nelle, T. D. & Wills, J. W. (1993). Func-tional chimeras of the Rous sarcoma virus andhuman immunode®ciency virus gag proteins.J. Virol. 67, 6487-6498.

Berg, J. M. (1986). Potential metal-binding domains innucleic acid binding proteins. Science, 232, 485-487.

Berkowitz, R. D., Ohagen, A., Hoglund, S. & Goff, S. P.(1995). Retroviral nucleocapsid domains mediatethe speci®c recognition of genomic viral RNAs bychimeric Gag polyproteins during RNA packagingin vivo. J. Virol. 69, 6445-6456.

Binley, J. M., Wyatt, R., Desjardins, E., Kwony, P. D.,Hendrickson, W., Moore, J. P. & Sodroski, J. (1998).Analysis of the interaction of antibodies with a con-served, enzymatically deglycosylated core of theHIV-1 gp120 envelope glycoprotein. AIDS Res.Hum. Retrovir. 14, 191-198.

Blake, P. R. & Summers, M. F. (1994). Probing the unu-sually similar metal coordination sites of retroviralzinc ®ngers and iron-sulfur proteins by nuclearmagnetic resonance. Advan. Biophys. Chem. 4, 1-30.

Bogerd, H. P., Fridell, R. A., Madore, S. & Cullen, B. R.(1995). Identi®cation of a novel cellular cofactor forthe Rev/Rex class of retroviral regulatory proteins.Cell, 82, 485-494.

Bolen, J. B. (1993). Nonreceptor tyrosine protein kinases.Oncogene, 8, 2025-2031.

Bowerman, B., Brown, P. O., Bishop, J. M. & Varmus,H. E. (1989). A nucleoprotein complex mediates theintegration of retroviral DNA. Genes Dev. 3, 469-478.

Braaten, D., Franke, E. K. & Luban, J. (1996). CyclophilinA is required for an early step in the life cycle ofhuman immunode®ciency virus type-1 beforethe initiation of reverse transcription. J. Gen. Virol.3551-3560.

Brodsky, A. S. & Williamson, J. R. (1997). Solution struc-ture of the HIV-2 TAR-arnininamide complex.J. Mol. Biol. 267, 624-639.

Brodsky, A. S., Erlacher, H. A. & Williamson, J. R.(1998). NMR evidence for a base triple in the HIV-2

TAR C-G*C� mutant-argininamide complex. Nucl.Acids Res. 26, 1991-1995.

the different classes of pathogenic human retro-viruses. J. Mol. Biol. 264, 1117-1131.


Brown, P. O. (1997). Integration. In Retroviruses (Cof®n,J. M., Hughes, S. H. & Varmus, H. E., eds), vol. 1,pp. 161-203, Cold Spring Harbor Laboratory Press,Cold Spring Harbor, NY.

Bryant, M. & Ratner, L. (1990). Myristoylation-depen-dent replication and assembly of human immuno-de®ciency virus 1. Proc. Natl Acad. Sci. USA, 87,523-527.

Bukrinsky, M. I., Haggerty, S., Dempsey, M. P.,Sharova, N., Adzhubei, A., Spitz, L., Lewis, P.,Goldfarb, D., Emerman, M. & Stevenson, M. (1993).A nuclear localization signal within HIV-1 matrixprotein that governs infection of non-dividing cells.Nature, 365, 666-669.

Caffrey, M., Cai, M., Kaufman, J., Stahl, S. J., Wing®eld,P. T., Gronenborn, A. M. & Clore, G. M. (1997).Determination of the secondary structure and globaltopology of the 44 kDa ectodomain of gp41 of thesimian immunode®ciency virus by multidimen-sional nuclear magnetic resonance spectroscopy.J. Mol. Biol. 271, 819-826.

Caffrey, M., Cai, M., Kaufman, J., Stahl, S. J., Wing®eld,P. T., Covell, D. G., Gronenborn, A. M. & Clore,G. M. (1998). Three-dimensional solution structureof the 44 kDa ectodomain of SIV gp41. EMBO J. 17,4572-4584.

Cai, M., Zheng, R., Caffrey, M., Craigie, R., Clore, G. M.& Gronenborn, A. M. (1997). Solution structure ofthe N-terminal zinc binding domain of HIV-1 inte-grase. Nature Struct. Biol. 4, 567-577.

Calnan, B. J., Tidor, B., Biancalana, S., Hudson, D. &Frankel, A. D. (1991). Arginine-mediated RNA rec-ognition: the arginine fork. Science, 252, 1167-1171.

Camaur, D. & Trono, D. (1996). Characterization ofhuman immunode®ciency virus type 1 Vif particleincorporation. J. Virol. 70, 6106-6111.

Carr, C. M. & Kim, P. S. (1993). A spring-loaded mech-anism for the conformational change of in¯uenzahemagglutinin. Cell, 73, 823-832.

Chan, D. C., Fass, D., Berger, J. M. & Kim, P. S. (1997).Core structure of gp41 from the HIV envelope gly-coprotein. Cell, 89, 263-273.

Chance, M. R., Sagi, I., Wirt, M. D., Frisbie, S. M.,Scheuring, E., Chen, E., Bess, J. W., Jr, Henderson,L. E., Arthur, L. O., South, T. L., Perez-Alvarado, G.& Summers, M. F. (1992). Extended x-ray absorp-tion ®ne structure studies of a retrovirus: equineinfectious anemia virus cysteine arrays are coordi-nated to zinc. Proc. Natl Acad. Sci. USA, 89, 10041-10045.

Chapham, P. R. & Weiss, R. A. (1997). Spoilt choice ofco-receptors. Nature, 388, 230-231.

Chazal, N., CarrieÈre, C., Gay, B. & Boulanger, P. (1994).Phenotypic characterization of insertion mutants ofthe human immunode®ciency virus type 1 gag pre-cursor expressed in recombinant baculovirus-infected cells. J. Virol. 68, 111-122.

Checroune, F., Yao, X.-J., Gottlinger, H., Bergeron, D. &Cohen, E. A. (1995). Incorporation of Vpr intohuman immunode®ciency virus type 1: role of con-served regions within the P6 domain of Pr55gag.J. Acquired Immune De®c. Syndr. 10, 1-7.

Christensen, A. M., Massiah, M. A., Turner, B. G.,Sundquist, W. I. & Summers, M. F. (1996). Three-dimensional structure of the HTLV-II matrix proteinand comparative analysis of matrix proteins from

Chun, T.-W., Stuyver, L., Mizell, S. B., Ehler, L. A.,Mican, J. A. M., Baseler, M., Lloyd, A. L., Nowak,M. A. & Fauci, A. S. (1997). Presence of an inducibleHIV-1 latent reservoir during highly active antire-troviral therapy. Proc. Natl Acad. Sci. USA, 94,13193-13197.

Clever, J. L. & Parslow, T. G. (1997). Mutant humanimmunode®ciency virus type 1 genomes withdefects in RNA dimerization or encapsidation.J. Virol. 71, 3407-3414.

Clever, J., Sassetti, C. & Parslow, T. G. (1995). RNA sec-ondary structure and binding sites for gag geneproducts in the 50 packaging signal of humanimmunode®ciency virus type 1. J. Virol. 69, 2101-2109.

Coates, A. R. M., Cookson, J., Barton, G. J., Zvelebil,M. J. & Sternberg, M. J. E. (1987). AIDS vaccine pre-dictions. Nature, 326, 549-550.

Cof®n, J. M. (1995). HIV population dynamics in vivo:implications for genetic variation, pathogenesis andtherapy. Science, 267, 483-489.

Cordonnier, A., Montagnier, L. & Emerman, M. (1989).Single amino acid changes in HIV envelope affectviral tropism and receptor binding. Nature, 340,571-574.

Courcoul, M., Patience, C., Rey, F., Blanc, D., Harmache,A., Sire, J., Vigne, R. & Spire, B. (1995). Peripheralblood mononuclear cells produce normal amountsof defective vif- human immunode®ciency virustype 1 particles which are restricted for the prere-trotranscriptional steps. J. Virol. 69, 9358-9365.

Crise, B., Buonocore, L. & Rose, J. K. (1990). CD4 isretained in the endoplasmic reticulum by thehuman immunode®ciency virus type 1 glycoproteinprecursor. J. Virol. 64, 5585-5593.

Daly, T. J., Cook, K. S., Gran, G. S., Maione, T. E. &Rusche, J. R. (1989). Speci®c binding of HIV-1recombinant Rev protein to the Rev-responseelement in vitro. Nature, 342, 816-819.

Daniel, M. D., Kirchoff, F., Czajak, S. C., Sehgal, P. K. &Derosiers, R. C. (1992). Protective effects of a liveattenuated SIV vaccine with a deletion in the nefgene. Science, 258, 1938-1941.

Dannull, J., Surovoy, A., Jung, G. & Moelling, K. (1994).Speci®c binding of HIV-1 nucleocapsid protein toPSI RNA in vitro requires N-terminal zinc ®ngerand ¯anking basic amino acid residues. EMBO J.13, 1525-1533.

Davies, J. F. I., Hostomska, Z., Hostomsky, Z., Jordon, S.& Matthews, D. A. (1991). Crystal structure of theribonuclease H domain of HIV-1 reverse transcrip-tase. Science, 252, 88-95.

Deacon, N. J., Tsykin, A., Solomon, A., Smith, K.,Ludford-Menting, M., Hooker, D. J., McPhee, D. A.,Greenway, A. L., Ellett, A., Chat®eld, C., Lawson,V. A., Crowe, S., Maerz, A., Sonza, S., Learmont, J.,Sullivan, J. S., Cunningham, A., Dwyer, D.,Dowton, D. & Mills, J. (1995). Genomic structure ofan attenuated quasi species of HIV-1 from bloodtransfusion donor and recipients. Science, 270, 988-991.

Debouck, C., Gorniak, J. G., Strickler, J. E., Meek, T. D.,Metcalf, B. W. & Rosenberg, M. (1987). Humanimmunode®ciency virus protease expressed inEscherichia coli exhibits autoprocessing and speci®cmaturation of the gag precursor. Proc. Natl Acad.Sci. USA, 84, 8903-8906.

De Rocquigny, H., Gabus, C., Vincent, A., FournieÂ-Zaluski, M.-C., Roques, B. & Darlix, J.-L. (1992).

Emerman, M. & Malim, M. H. (1998). HIV-1 regulatory/accessory genes: keys to unraveling viral and host


Viral RNA annealing activities of human immuno-de®ciency virus type 1 nucleocapsid protein requireonly peptide domains outside the zinc ®ngers. Proc.Natl Acad. Sci. USA, 89, 6472-6476.

DeÂmeÂneÂ, H., Dong, C. Z., Ottmann, M., Rouyez, M. C.,Jullian, N., Morellet, N., Mely, Y., Darlix, J. L.,FournieÂ-Zaluski, M. C., Saragosti, S. & Roques, B. P.(1994a). 1H NMR structure and biological studies ofthe His23 to Cys mutant nucleocapsid protein ofHIV-1 indicate that the conformation of the ®rstzinc ®nger is critial for virus infectivity. Biochemis-try, 33, 11707-11716.

DeÂmeÂneÂ, H., Jullian, N., Morellet, N., de Rocquigny, H.,Cornille, F., Maigret, B. & Roques, B. P. (1994b).Three-dimensional 1H NMR structure of the nucleo-capsid protein NCp10 of Moloney murine leukemiavirus. J. Biomol. NMR, 4, 153-170.

De Guzman, R. N., Wu, Z. R., Stalling, C. C.,Pappalardo, L., Borer, P. N. & Summers, M. F.(1998). Structure of the HIV-1 nucleocapsid proteinbound to the SL3 -RNA recognition element.Science, 279, 384-388.

Doranz, B. J., Rucker, J., Yi, Y., Smyth, R. J., Samson,M., Peiper, S. C., Parmentier, M., Collman, R. G. &Doms, R. W. (1996). A dual-tropic primary HIV-1isolate that uses fusin and the b-chemokine recep-tors CKR-5, CKR-3, and CKR-2b as fusion cofactors.Cell, 85, 1149-1158.

Dorfman, T., Luban, J., Goff, S. P., Haseltine, W. A. &GoÈ ttlinger, H. G. (1993). Mapping of functionallyimportant residues of a cysteine-histidine box in thehuman immunode®ciency virus type 1 nucleocapsidprotein. J. Virol. 67, 6159-6169.

Dorfman, A. T., Bukovsky, A., Ohagen, A. S., Hoglund,H. & Gottlinger, G. (1994a). Functional domains ofthe capsid protein of human immunode®ciencyvirus type 1. J. Virol. 68, 8180-8187.

Dorfman, T., Mammano, F., Haseltine, W. A. &GoÈ ttlinger, H. G. (1994b). Role of the matrix proteinin the virion association of the human immunode®-ciency virus type 1 envelope glycoprotein. J. Virol.68, 1689-1696.

Dorn, P., DaSilva, L., Martarano, L. & Derse, D. (1990).Equine infectious anemia virus tat: insights into thestructure, function, and evolution of lentivirustrans-activator proteins. J. Virol. 64, 1616-1624.

Dupraz, P., Oertle, S., MeÂric, C., Damay, P. & Spahr, P.-F. (1990). Point mutations in the proximal Cys-Hisbox of Rous sarcoma virus nucleocapsid protein.J. Virol. 64, 4978-4987.

Dyda, F., Hickman, A. B., Jenkins, M. T., Engelman, A.,Craigie, R. & Davies, D. R. (1994). Crystal structureof the catalytic domain of HIV-1 integrase: simi-larity to other polynucleotidyl transferases. Science,266, 1981-1986.

Earl, P. L., Moss, B. & Doms, R. W. (1991). Folding,interaction with GRP78-BiP, assembly, and trans-port of the human immunode®ciency virus type 1envelope protein. J. Virol. 65, 2047-2055.

Ehrlich, L. S., Agresta, B. E. & Carter, C. A. (1992).Assembly of recombinant human immunode®ciencyvirus type 1 capsid protein in vitro. J. Virol. 66,4874-4883.

Eijkelenboom, A. P., Lutzke, R. A., Boelens, R., Plasterk,R. H., Kaptein, R. & Hard, K. (1995). The DNA-binding domain of HIV-1 integrase has an SH3-likefold. Nature Struct. Biol. 2, 807-810.

cell biology. Science, 280, 1880-1884.Esnouf, R., Ren, J., Ross, R., Jones, Y., Stammers, D. &

Stuart, D. (1995). Mechanism of inhibition of HIV-1reverse transcriptase by nonnucleoside inhibitors.Nature Struct. Biol. 2, 303-308.

FaÈcke, M., Janetzko, A., Shoeman, R. L. & KraÈusslich,H.-G. (1993). A large deletion in the matrix domainof the human immunode®ciency virus gag generedirects virus particle assembly from the plasmamembrane to the endoplasmic reticulum. J. Virol.67, 4972-4980.

Farmerie, W. G., Loeb, D. D., Casavant, N. C.,Hutchinson, C. A. I., Edgell, M. H. & Swanstrom,R. (1987). Expression and processing of the AIDSvirus reverse transcriptase in Escherichia coli. Science,236, 305-308.

Fass, D., Harrison, S. C. & Kim, P. S. (1996). Retrovirusenvelope domain at 1.7 AÊ resolution. Nature Struct.Biol. 3, 465-469.

Feng, F., Broder, C. C., Kennedy, P. E. & Berger, E. A.(1996). HIV-1-entry cofactor: functional cDNA clon-ing of a seven-transmembrane, G protein-coupledreceptor. Science, 272, 872-877.

Finzi, D., Hermankova, M., Pierson, T., Carruth, L. M.,Buck, C., Chaisson, R. E., Quinn, T. C., Chadwick,K., Margolick, J., Brookmeyer, R., Gallant, J.,Markowitz, M., Ho, D. D., Richman, D. D. &Siliciano, R. F. (1997). Identi®cation of a reservoirfor HIV-1 in patients on highly active antiretroviraltherapy. Science, 278, 1295-1300.

Foti, M., Mangasarian, A., Piguet, V., Lew, D. P.,Krause, K. H., Trono, D. & Carpentier, J. L. (1997).Nef-mediated clathrin-coated pit formation. J. CellBiol. 139, 37-47.

Fouchier, R. A., Simon, J. H., Jaffe, A. B. & Malim, M. H.(1996). Human immunode®ciency virus type 1 Vifdoes not in¯uence expression or virion incorpor-ation of gag-, pol- and env-encoded proteins.J. Virol. 70, 8263-8269.

Fouchier, R. A. M., Meyer, B. E., Simon, J. H. M.,Fischer, U. & Malim, M. H. (1997). HIV-1 infectionof non-dividing cells: evidence that the amino-term-inal basic region of the viral matrix protein isimportant for Gag processing but not for post-entrynuclear import. EMBO J. 16, 4531-4539.

Fouchier, R. A. M., Meyer, B. E., Simon, J. H., Fischer,U., Albright, A. V., Gonzalez-Siarano, F. & Malim,M. H. (1998). J. Virol. 72, 6004-6013.

Franke, E. K., Yuan, H. E. H., Bossolt, K. L., Goff, S. P.& Luban, J. (1994a). Speci®city and sequencerequirements for interactions between various retro-viral Gag proteins. J. Virol. 68, 5300-5305.

Franke, K. E., En Hui, Yuan H. & Luban, J. (1994b).Speci®c incorporation of cyclophilin A into HIV-1virions. Nature, 24, 359-362.

Frankel, A. D., Biancalana, S. & Hudson, D. (1989).Activity of synthetic peptides from the Tat proteinof human immunode®ciency virus type 1. Proc. NatlAcad. Sci. USA, 86, 7397-7401.

Freed, E. O. & Martin, M. A. (1994). HIV-1 infection ofnon-dividing cells. Nature, 369, 107-108.

Freed, E. O. & Martin, A. M. (1995). Virion incorpor-ation of envelope glycoproteins with long but notshort cytoplasmic tails is blocked by speci®c, singleamino acid substitutions in the human immunode®-ciency virus type 1 matrix. J. Virol. 69, 1984-1989.

Freed, E. O. & Martin, A. M. (1996). Domains of thehuman immonode®ciency virus type 1 matrix and

Goff, S. P. (1990). Retroviral reverse transcriptase: syn-thesis, structure and function. J. Aquired Immune


gp41 cytoplasmic tail required for envelope incor-poration into virions. J. Virol. 70, 341-351.

Freed, E. O., Myers, D. J. & Risser, R. (1990). Character-ization of the fusion domain of the human immuno-de®ciency virus type 1 envelope glycoprotein gp41.Proc. Natl Acad. Sci. USA, 87, 4650-4654.

Freed, E. O., Orenstein, J. M., Buckler-White, A. J. &Martin, M. A. (1994). Single amino acid changes inthe human immunode®ciency virus type 1 matrixprotein block virus particle production. J. Virol. 68,5311-5320.

Freed, E. O., Englund, G. & Martin, A. M. (1995). Roleof the basic domain of human immonde®ciencyvirus type 1 matrix in macrophage infection. J. Virol.69, 3949-3954.

Freed, E. O., Englund, G., Maldarelli, F. & Martin, M. A.(1997). Phosphorylation of residue 131 of HIV-1matrix is not required for macrophage infection.Cell, 88, 171-173.

Le Gall, S., Erdtmann, L., Benichou, S., Berlioz-Torrent,C., Liu, L., Benarous, R., Heard, J. M. & Schwartz,O. (1998). Nef interacts with the mu subunit ofclathrin adaptor complexes and reveals a crypticsorting signal in MHC I molecules. Immunity, 8,483-495.

Gallay, P., Swingler, S., Aiken, C. & Trono, D. (1995a).HIV-1 infection of nondividing cells: C-terminaltyrosine phosphorylation of the viral matrix proteinis a key regulator. Cell, 80, 379-388.

Gallay, P., Swingler, S., Song, J., Bishman, F. & Trono,D. (1995b). HIV nuclear import is governed by thephosphotyrosine-mediated binding of matrix to thecore domain of integrase. Cell, 83, 569-576.

Gallo, R. C., Salahuddin, S. Z., Popovic, M., Shearer,G. M., Kaplan, M., Haynes, B. F., Palker, T. J.,Red®eld, R., Oleske, J., Safai, B., White, G., Foster,P. & Markham, P. D. (1984). Frequent detection andisolation of cytopathic retroviruses (HTLV-III) frompatients with AIDS and at risk for AIDS. Science,224, 500-503.

Gamble, T. R., Vajdos, F., Yoo, S., Worthylake, D. K.,Houseweart, S. M., Sundquist, W. I. & Hill, C. P.(1996). Crystal structure of human cyclophilin Abound to the amino-terminal domain of the HIV-1capsid. Cell, 87, 1285-1294.

Gamble, T. R., Yoo, S., Vajdos, F. F., von Schwedler,U. K., Korthylake, D. K., Wang, H., McCutcheon,J. P., Sundquist, W. I. & Hill, C. P. (1997). Structureof the carboxyl-terminal dimerization domain of theHIV-1 capsid protein. Science, 278, 849-853.

Gao, Y., Kaluarachchi, K. & Griedroc, D. P. (1998). Sol-ution structure and backbone dynamics of Mason-P®zer monkey virus nucleocapsid protein. ProteinSci. 7, 2265-2280.

Gheysen, D., Jacobs, E., de Foresta, F., Thiriart, C.,Francotte, M., Thines, D. & De Wilde, M. (1989).Assembly and release of HIV-1 precursor Pr55gag

virus-like particles from recombinant baculovirus-infected insect cells. Cell, 59, 103-112.

Gilbona, E., Mitra, S. W., Goff, S. & Baltimore, D. (1979).A detailed model of reverse transcription and testsof crucial aspects. Cell, 18, 93-100.

Gitti, R. K., Lee, B. M., Walker, J., Summers, M. F., Yoo,S. & Sundquist, W. I. (1996). Structure of the amino-terminal core domain of the HIV-1 capsid protein.Science, 273, 231-235.

De®c. Syndr. 3, 817-831.Goldsmith, M. A., Warmerdam, M. T., Atchison, R. E.,

Miller, M. D. & Greene, W. C. (1995). Dissociationof the CD4 downregulation and viral infectivityenhancement functions of human immunode®ciencyvirus type 1 Nef. J. Virol. 69, 4112-4121.

Goncalves, J., Korin, Y., Zack, J. & Gabuzda, D. (1996).Role of vif in human immunode®ciency virus type1 reverse transcriptase. J. Virol. 70, 8701-8709.

Gorelick, R. J., Henderson, L. E., Hanser, J. P. & Rein, A.(1988). Point mutants of Moloney murine leukemiavirus that fail to package viral RNA: evidence forspeci®c RNA recognition by a ``zinc ®nger-like''protein sequence. Proc. Natl Acad. Sci. USA, 85,8420-8424.

Gorelick, R. J., Nigida, J., Stephen, M., Bess, J. W.,Arthur, L. O., Henderson, L. E. & Alan, R. (1990).Noninfectious human immunode®ciency virus type1 mutants de®cient in genomic RNA. J. Virol. 64,3207-3211.

Gorelick, R. J., Chabot, D. J., Alan, R., Henderson, L. E.& Arthur, L. O. (1993). The two zinc ®ngers in thehuman immunode®ciency virus type 1 nucleocapsidprotein are not functionally equivalent. J. Virol. 67,4027-4036.

GoÈ ttlinger, H. G., Sodroski, J. G. & Haseltine, W. A.(1989). Role of capsid precursor processing andmyristoylation in morphogenesis and infectivity ofhuman immunode®ciency virus type 1. Proc. NatlAcad. Sci. USA, 86, 5781-5785.

Green, L. M. & Berg, J. M. (1989). A retroviral Cys-Xaa2-Cys-Xaa4-His-Xaa4-Cys peptide binds metalions: spectoscopic studies and a proposed three-dimensional structure. Proc. Natl Acad. Sci. USA, 86,4047-4051.

Green, L. M. & Berg, J. M. (1990). Retroviral nucleocap-sid protein-metal ion interactions: folding andsequence variants. Proc. Natl Acad. Sci. USA, 87,6403-6407.

Greenberg, M. E., Bronson, S., Lock, M., Neumann, M.,Pavlakis, G. N. & Skowronski, J. (1997). EMBO J.16, 6964-6976.

Gross, I., Hohenberg, H., Juckhagel, C. & Krausslich, H.-G. (1998). N-terminal extension of human immuno-de®ciency virus capsid protein converts the in vitroassembly phenotype from tubular to spherical par-ticles. J. Virol. 72, 4798-4810.

Grzesiek, S., Bax, A., Nicholson, L. K., Yamazaki, T.,Wing®eld, P., Stahl, S. J., Eyermann, C. J., Torchia,D. A., Hodge, C. N., Lam, P. Y. S., Jadhav, P. K. &Chang, C.-H. (1994). NMR evidence for the displa-cement of a conserved interior water molecule inHIV protease by a non-peptide cyclic urea-basedinhibitor. J. Am. Chem. Soc. 116, 1581-1582.

Grzesiek, S., Bax, A., Clore, G. M., Gronenborn, A. M.,Hu, J.-S., Kaufman, J., Palmer, I., Stahl, S. J. &Wing®eld, P. T. (1996). The solution structure ofHIV-1 Nef reveals an unexpected fold and permitsdelineation of the binding surface for the SH3domain of Hck tyrosine protein kinase. NatureStruct. Biol. 3, 340-345.

Grzesiek, S., Bax, A., Hu, J.-S., Kaufman, J., Palmer, I.,Stahl, S. J., Tjandra, N. & Wing®eld, P. T. (1997).Re®ned solution structure and backbone dynamicsof HIV-1 Nef. Protein Sci. 6, 1248-1263.

Gulick, R. M., Mellors, J. W., Havlir, D., Eron, J. J.,Gonzalez, C., McMahon, D., Richman, D. D.,

Valentine, F. T., Jonas, L., Meibohm, A., Emini, E. A.& Chodakewitz, J. A. (1997). Treatment with indina-

erization and inhibition mechanisms. Structure, 4,853-860.


vir, zidovudine, and lamivudine in adults withhuman immunode®ciency virus infection and priorantiretroviral therapy. New Eng. J. Med. 337, 734-739.

Gustchina, A., Kervinen, J., Powell, D. J., Zdanov, A.,Kay, J. & Wlodawer, A. (1996). Structure of equineinfectious anemia virus proteinase complexed withan inhibitor. Protein Sci. 5, 1453-1465.

Harrison, G. P. & Lever, A. M. L. (1992). The humanimmunode®ciency virus type 1 packaging signaland major splice donor region have a conservedstable secondary structure. J. Virol. 66, 4144-4153.

Hayashi, T., Shioda, T., Iwakura, Y. & Shibuta, H.(1992). RNA packaging signal of human immuno-de®ciency virus type 1. Virology, 188, 590-599.

Hayashi, T., Ueno, Y. & Okamoto, T. (1993). Elucidationof a conserved RNA stem-loop structure in thepackaging signal of human immunode®ciency virustype 1. FEBS Letters, 327, 213-218.

Heaphy, S., Finch, J. T., Gait, M. J., Karn, J. & Singh, M.(1991). Human immunode®ciency virus type 1 reg-ulator of virion expression, rev, forms nucleoprotein®laments after binding to a purine rich bubblelocated within the rev response region of viralmRNAs. Proc. Natl Acad. Sci. USA, 88, 7366-7370.

Henderson, L. E., Copeland, T. D., Sowder, R. C.,Smythers, G. W. & Oroslzan, S. (1981). Primarystructure of the low molecular weight nucleic acid-binding proteins of murine leukemia viruses. J. Biol.Chem. 256, 8400-8406.

Herrmann, C. H. & Rice, A. P. (1995). Lentivirus Tatproteins speci®cally associate with a cellular proteinkinase, TAK, that hyperphosphorylates the car-boxyl-terminal domain of the large subunit of RNApolymerase II: candidate for a Tat co-factor. J. Virol.69, 1612-1620.

Hill, C. P., Worthylake, D., Bancroft, D. P., Christensen,A. M. & Sundquist, W. I. (1996). Crystal structuresof the trimeric HIV-1 matrix protein: implicationsfor membrane association. Proc. Natl Acad. Sci. USA,93, 3099-3104.

Ho, D. D., Neumann, A. U., Perelson, A. S., Chen, W.,Leonard, J. M. & Markowitz, M. (1995). Rapid turn-over of plasma virions and CD4 lymphocytes inHIV-1 infection. Nature, 373, 123-126.

Hockley, D. J., Nermut, M. V., Grief, C., Jowett, J. B. M.& Jones, I. M. (1994). Comparative morphology ofGag protein structures produced by mutants of thegag gene of human immunode®ciency virus type 1.J. Gen. Virol. 75, 2985-2997.

Hogle, J. M., Chow, M. & Filman, D. J. (1985). Three-dimensional structure of poliovirus at 2.9 AÊ resol-ution. Science, 229, 1358-1365.

Hong, S. S. & Boulanger, P. (1993). Assembly-defectivepoint mutants of the human immunode®ciencyvirus type 1 Gag precursor phenotypicallyexpressed in recombinant baculovirus-infected cells.J. Virol. 67, 2787-2798.

Hoxie, J. A., Alpers, J. D., Rackowski, J. L., Huebner, K.,Haggarty, B. S., Cedarbaum, A. J. & Reed, J. C.(1986). Alterations in T4 (CD4) protein and mRNAsynthesis in cells infected with HIV. Science, 234,1123-1127.

Hsiou, Y., Ding, J., Das, K., Clark, A. D., Jr, Hughes,S. H. & Arnold, E. (1996). Structure of unligandedHIV-1 reverse transcriptase at 2.7 AÊ resolution:implications of conformational changes for polym-

Ippolito, J. A. & Steitz, T. A. (1998). A 1.3 AÊ resolutioncrystal structure of the HIV-1 trans-activationresponse region RNA stem reveals a metal ion-dependent bulge conformation. Proc. Natl Acad. Sci.USA, 95, 9819-9824.

Jacks, T., Power, M. D., Masiarz, F. R., Luciw, P. A.,Barr, P. J. & Varmus, H. E. (1988). Characterizationof ribosomal frameshifting in HIV-1 gag-polexpression. Nature, 331, 280-283.

Jacobo-Molina, A., Ding, J., Nanni, R. G., Clark, A. D.,Jr, Lu, X., Tantillo, C., Williams, R. L., Kamer, G.,Ferris, A. L., Clark, P., Hizi, A., Hughes, S. H. &Arnold, E. (1993). Crystal structure of humanimmunode®ciency virus type 1 reverse transcriptasecomplexed with double-stranded DNA at 3.0 AÊ res-olution shows bent DNA. Proc. Natl Acad. Sci. USA,90, 6320-6324.

Jenkins, T. M., Hickman, A. B., Dyda, F., Ghirlando, R.,Davies, D. R. & Craigie, R. (1996). A soluble activemutant of HIV-1 integrase: involvement of both thecore and carboxyl terminal domains in multimeriza-tion. J. Biol. Chem. 271, 7712-7718.

Jentoft, J. E., Smith, L. M., Fu, X., Johnson, M. & Leis, J.(1988). Conserved cysteine and histidine residues ofthe avian myeloblastosis virus nucleocapsid proteinare essential for viral replication but are not ``zinc-binding ®ngers''. Proc. Natl Acad. Sci. USA, 85,7094-7098.

Jones, K. A. & Peterlin, B. M. (1994). Control of RNAinitiation and elongation at the HIV-1 promoter.Annu. Rev. Biochem. 63, 717-743.

Jowett, J., Hockley, D., Nermut, M. V. & Jones, I. M.(1992). Distinct signals in human immunode®ciencyvirus type 1 Pr55 necessary for RNA binding andparticle formation. J. Gen. Virol. 73, 3079-3086.

Jowett, J. B. M., Planelles, V., Poon, B., Shah, N. P.,Chen, M.-L. & Chen, I. S. Y. (1995). The humanimmunode®ciency virus type 1 vpr gene arrestsinfected T cells in the G2 �M phase of the cellcycle. J. Virol. 69, 6304-6313.

Julian, N., Demene, H., Morellet, N., Maigret, B. &Roques, B. P. (1993). Replacement of His23 by Cysin a zinc ®nger of HIV-1 NCp7 led to a change in1H NMR-derived 3D structure and to a loss of bio-logical activity. FEBS Letters, 331, 43-48.

Katz, R. A. & Skalka, A. M. (1994). The retroviralenzymes. Annu. Rev. Biochem. 63, 133-173.

Kestler, H. W. I., Ringler, D. J., Mori, K., Panicali, D. L.,Sehgal, P. K., Daniel, M. D. & Desrosiers, R. C.(1991). Importance of the nef gene for maintenanceof high virus loads and for development of AIDS.Cell, 65, 651-662.

Kirchoff, F., Greenough, T. C., Brettler, D. B., Sullivan,J. L. & Desrosiers, R. C. (1995). Absence of intactnef sequences in a long-term servivor with nonpro-gressive HIV-1 infection. New Eng. J. Med. 332, 228-232.

Kjems, J., Calnan, B. J., Frankel, A. & Sharp, P. A.(1992). Speci®c binding of a basic peptide fromHIV-1 Rev. EMBO J. 11, 1119-1129.

Kohlstaedt, L. A., Wang, J., Friedman, J. M., Rice, P. A.& Steitz, T. A. (1992). Crystal structure at 3.5 AÊ res-olution of HIV-1 reverse transcriptase complexedwith an inhibitor. Science, 256, 1783-1790.

Kohlstaedt, L. A., Wang, J., Rice, P. A., Friedman, J. M.& Steitz, T. A. (1993). The structure of HIV-1reverse transcriptase. In Reverse Transcriptase

(Skalka, A. M. & Goff, S. P., eds), vol. 1, pp. 223-249, Cold Spring Harbor Laboratory Press, Cold

from clinically healthy individuals. J. Infect. Dis.152, 734-738.


Spring Harbor, NY.Koken, S. E., Greijer, A. E., Verhoef, K., van Wamel, J.,

Burrinskaya, A. G. & Berkhout, B. (1994). Intracellu-lar analysis of in vitro modi®ed HIV Tat protein.J. Biol. Chem. 269, 8366-8375.

Kondo, E. & Gottlinger, H. G. (1996). A conservedLXXLF sequence is the major determinant in p6gagrequired for the incorporation of human immuno-de®ciency virus type 1 Vpr. J. Virol. 70, 159-164.

Kong, L. B., An, D., Ackerson, B., Canon, J., Rey, O.,Chen, I. S. Y., Krogstad, P. & Steward, P. L. (1998).Cryoelectron microscopy examination of humanimmunode®ciency virus type-1 virions withmutations in the cyclophilin A binding loop. J. Virol.72, 4403-4407.

Kowalski, M., Potz, J., Basiripour, L., Dorfman, T., Goh,W. C., Terwilliger, E., Dayton, A., Rosen, C.,Haseltine, W. & Sodroski, J. (1987). Functionalregions of the envelope glycoprotein of humanimmunode®ciency virus type 1. Science, 237, 1351-1355.

Kwong, P. D., Wyatt, R., Robinson, J., Sweet, R. W.,Sodroski, J. & Hendrickson, W. A. (1998). Structureof an HIV gp120 envelope glycoprotein in complexwith the CD4 receptor and a neutralizing humanantibody. Nature, 393, 648-659.

Lam, P. Y. S., Jadhav, P. K., Eyermann, C. J., Hodge,C. N., Ru, Y., Bacheler, L. T., Meek, J. L., Otto, M. J.,Rayner, M. M., Wong, Y. N., Chang, C.-H., Weber,P. C., Jackson, D. A., Sharpe, T. R. & Erickson-Viitanen, S. (1994). Rational design of potent, bioa-vailable, nonpeptide cyclic ureas as HIV proteaseinhibitors. Science, 263, 380-384.

Langedijk, J. P. M., Schalken, J. J., Tersmette, M.,Huisman, J. G. & Meloen, R. H. (1990). Location ofepitopes on the major core protein p24 of humanimmunode®ciency virus. J. Gen. Virol. 71, 2609-2614.

Lapadat-Tapolsky, M., De Rocquigny, H., Van Gent, D.,Roques, B., Plasterk, R. & Darlix, J.-L. (1993). Inter-actions between HIV-1 nucleocapsid protein andviral DNA may have important functions in theviral life cycle. Nucl. Acids Res. 21, 831-839.

Lee, B. M., De Guzman, R. N., Turner, B. G., Tjandra,N. & Summers, M. F. (1998). Dynamical behavior ofthe HIV-1 nucleocapsid protein. J. Mol. Biol. 279,633-649.

Lee, C.-H., Saksela, K., Mirza, U. A., Chait, B. T. &Kuriyan, J. (1996). Crystal structure of the con-served core of HIV-1 Nef complexed with a Srcfamily SH3 domain. Cell, 85, 931-942.

Le, Grice S. F. J., Naas, T., Wohlgensinger, B. & Schatz,O. (1991). Subunit-selective mutagenesis indicatesminimal polymerase activity in heterodimer associ-ated p51 of HIV-1 reverse transcriptase. EMBO J.10, 3905-3911.

Leonard, C. K. Spellman, M. W., Riddle, L., Harris, R. J.,Thomas, J. N. & Gregory, T. J. (1990). Assignmentof intrachain disul®de bonds and characterizationof potential glycosylation sites of the type 1 recom-binant immunode®ciency virus envelope glyco-protein (gp120) expressed in Chinese hamster ovarycell. J. Biol. Chem. 265, 10373-10382.

Levy, J. A., Hoffmann, A. D., Kramer, S. M., Landis,J. A., Shimabukuro, J. M. & Oshiro, L. S. (1984).Recovery of AIDS-associated retroviruses frompatients with AIDS or AIDS-related conditions and

Liu, H., Wu, X., Newman, M., Shaw, G. M., Hahn, B. H.& Kappes, J. C. (1995). The Vif protein of humanand simian immunode®ciency viruses is packagedinto virions and associates with viral core struc-tures. J. Virol. 69, 7630-7638.

Lodi, P. J., Ernst, J. A., Kuszewski, J., Hickman, A. B.,Engelman, A., Craigie, R., Clore, G. M. &Gronenborn, A. M. (1995). Solution structure of theDNA binding domain of HIV-1 integrase. Biochemis-try, 34, 9826-9833.

Lu, M., Blacklow, S. C. & Kim, P. S. (1995). A trimericstructural domain of the HIV-1 transmembrane gly-coprotein. Nature Struct. Biol. 2, 1075-1082.

Luban, J., Bossolt, K. L., Franke, E. K., Kalpana, G. V. &Goff, S. P. (1993). Human immunode®ciency virustype 1 gag protein binds to cyclophilins A and B.Cell, 73, 1067-1078.

Luban, J., Franke, E. K. & Yuan, H. E. (1994). HIV-1uses host-cell proteins to form fully infectious vir-ions. Nature, 7, 37-42.

Malim, M. H. & Cullen, B. R. (1991). HIV-1 structuralgene expression requires the binding of multiplerev monomers to the viral RRE: implications forHIV-1 latency. Cell, 65, 241-248.

Malim, M. H., Bohnlein, S., Hauber, J. & Cullen, B. R.(1989a). Functional dissection of the HIV-1 Revtrans-activator - derivation of a trans-dominantrepressor of Rev function. Cell, 58, 205-214.

Malim, M. H., Hauber, J., Le, S. Y., Maizel, J. V. &Cullen, B. R. (1989b). The HIV-1 rev transactivatoracts through a structured target sequence to activatenuclear export of unspliced viral mRNA. Nature,338, 254-257.

Malim, M. H., Tiley, L. S., McCarn, D. F., Rusche, J. R.,Hauber, J. & Cullen, B. R. (1990). HIV-1 structuralgene expression requires binding of the Rev trans-activator to its RNA target sequence. Cell, 60, 675-683.

Mammano, F., Kondo, E., Sodroski, J., Bukovsky, A. &Gottlinger, H. G. (1995). Rescue of human immuno-de®ciency virus type 1 matrix protein mutants byenvelope glycoproteins with short cytoplasmicdomains. J. Virol. 69, 3824-3830.

Mangasarian, A. & Trono, D. (1997). The multifacetedrole of HIV Nef. Res. Virol. 148, 30-33.

Margottin, F., Bour, S. P., Durand, H., Selig, L.,Benichou, S., Richard, V., Thomas, D., Strebel, K. &Benarous, R. (1998). A novel human WD protein, h-beta TrCp, that interacts with the HIV-1 Vpu con-nects CD4 to the ER degradation pathway throughan F-box motif. Mol. Cell, 1, 565-74.

Massiah, M. A., Starich, M. R., Paschall, C., Summers,M. F., Christensen, A. M. & Sundquist, W. I. (1994).Three-dimensional structure of the human immuno-de®ciency virus type 1 matrix protein. J. Mol. Biol.244, 198-223.

Massiah, M. A., Worthylake, D., Christensen, A. M.,Sundquist, W. I., Hill, C. P. & Summers, M. F.(1996). Comparison of the NMR and X-ray struc-tures of the HIV-1 matrix protein: evidence for con-formational changes during viral assembly. ProteinSci. 5, 2391-2398.

Matthews, S., Barlow, P., Boyd, J., Barton, G., Russell,R., Mills, H., Cunningham, M., Meyers, N., Burns,N., Clark, N., Kingsman, S., Kingsman, A. &Campbell, I. (1994). Structural similarity between

the p17 matrix protein of HIV-1 and interferon-g.Nature, 370, 666-668.

of mutants in conjunction with a high-resolutionCD4 atomic structure. J. Exp. Med. 176, 507-517.

Momany, C., Kovari, L. C., Prongay, A. J., Keller, W.,


Matthews, S., Barlow, P., Clark, N., Kingsman, S.,Kingsman, A. & Campbell, I. (1995). Re®ned sol-ution structure of p17, the HIV matrix protein. Bio-chem. Soc. Trans. 23, 725-728.

Matthews, S., Mikhailov, M., Burny, A. & Roy, P.(1996). The solution structure of the bovine leuke-mia virus matrix protein and similarity with lenti-viral matrix proteins. EMBO J. 15, 3267-3274.

McBride, M. S. & Panganiban, A. T. (1996). The humanimmunode®ciency virus type 1 encapsidation site isa multipartite RNA element composed of functionalhairpin structures. J. Virol. 70, 2963-2973.

MeÂly, Y., Cornille, F., FournieÂ-Zaluski, M.-C., Darlix, J.-L., Roques, B. P. & GeÂrard, D. (1991). Investigationof zinc-binding af®nities of moloney murine leuke-mia virus nucleocapsid protein and its related zinc®nger and modi®ed peptides. Biopolymers, 31, 899-906.

MeÂly, Y., de Rocquigny, H., PieÂmont, E., DeÂmeÂneÂ, H.,Julian, N., FournieÂ-Zaluski, M.-C., Roques, B. P. &GeÂrard, D. (1993a). In¯uence of the N- and C-term-inal chains on the zinc-binding and conformationalproperties of the central zinc-®nger structure ofMoloney murine leukemia virus nucleocapsid pro-tein: a steady-state and time-resolved ¯uorescencestudy. Biochim. Biophys. Acta, 1161, 6-18.

MeÂly, Y., PieÂmont, E., Sorinas-Jimeno, M., DeRocquigny, H., Julian, N., Morellet, N., Roques, B. P.& GeÂrard, D. (1993b). Structural and dynamiccharacterization of the aromatic amino acids of thehuman immunode®ciency virus type 1 nucleocapsidprotein zinc ®ngers and their involvement in heter-ologous tRNAPhe binding: a steady-state and timeresolved ¯uorescence study. Biophys. J. 65, 1513-1522.

MeÂly, Y., De Rocquigny, H., Morellet, N., Roques, B. P.& GeÂrard, D. (1996). Zinc binding to the HIV-1nucleocapsid protein: a thermodynamic investi-gation by ¯uorescence spectroscopy. Biochemistry,35, 5175-5182.

MeÂric, C. & Goff, S. P. (1989). Characterization of Molo-ney murine leukemia virus mutants with single-amino-acid substitutions in the Cys-His box of thenucleocapsid protein. J. Virol. 63, 1558-1568.

MeÂric, C., Gouilloud, E. & Spahr, P.-F. (1988). Mutationsin Rous sarcoma virus nucleocapsid protein p12(NC): deletions of Cys-His Boxes. J. Virol. 62, 3328-3333.

Miller, M., Jaskolski, M., Rao, Mohana J. K., Leis, J. &Wlodawer, A. (1989a). Crystal structure of a retro-viral protease proves relationship to aspartic pro-tease family. Nature, 337, 576-579.

Miller, M., Schneider, J., Sathyanarayana, B. K., Toth,M. V., Marshall, G. R., Clawson, L., Selk, L., Kent,S. B. H. & Wlodawer, A. (1989b). Structure of acomplex of synthetic HIV-1 protease with a sub-strate-based inhibitor at 2.3 AÊ resolution. Science,246, 1149-1152.

Miller, M. D., Farnet, C. M. & Bushman, F. D. (1997).Human immunode®ciency virus type 1 preintegra-tion complexes: studies of organization and compo-sition. J. Virol. 71, 5382-5390.

Moebius, U., Clayton, L., Abraham, S., Harrison, S. &Reinhertz, E. (1992). The human immunode®ciencyvirus gp120 binding site of CD4: delineation byquantitative equilibrium and kinetic binding studies

Gitti, R. K., Lee, B. M., Gorbalenya, A. E., Tong, L.,McClure, J., Ehrlich, L. S., Summers, M. F., Carter,C. & Rossmann, M. G. (1996). Crystal structure ofdimeric HIV-1 capsid protein. Nature Struct. Biol. 9,763-770.

Moore, J. P. (1997). Coreceptors: implications for HIVpathogenesis and therapy. Science, 276, 51-52.

Morellet, N., Jullian, N., De Rocquigny, H., Maigret, B.,Darlix, J.-L. & Roques, B. P. (1992). Determinationof the structure of the nucleocapsid protein NCp7from the human immunode®ciency virus type 1 by1H NMR. EMBO J. 11, 3059-3065.

Morellet, N., de Rocquigny, H., MeÂly, Y., Jullian, N.,DeÂmeÂneÂ, H., Ottman, M., GeÂrard, D., Darlix, J. L.,Fournie-Zaluski, M. C. & Roques, B. P. (1994). Con-formational behaviour of the active and inactiveforms of the nucleocapsid NCp7 of HIV-1 studiedby 1H NMR. J. Mol. Biol. 235, 287-301.

Mujeeb, A., Bishop, K., Peterlin, B. M., Turck, C.,Parslow, T. G. & James, T. L. (1994). NMR structureof a biologically active peptide containing the RNA-binding domain of human immunode®ciency virustype 1 Tat. Proc. Natl Acad. Sci. USA, 91, 8248-8252.

Mulichak, A. M., Hui, J. O., Tomasselli, A. G.,Heinrikson, R. L., Curry, K. A., Tomich, C. S.,Thaisrivongs, S., Sawyer, T. K. & Watenpaugh, K. D.(1993). The crystallographic structure of the pro-tease from human immunode®ciency virus type 2with two synthetic peptidic transition state analoginhibitors. J. Biol. Chem. 268, 13103-13109.

Navia, M. A., Fitzgerald, P. M. D., McKeever, B. M.,Leu, C.-T., Heimbach, J. C., Herber, W. K., Sigal,I. S., Darke, P. L. & Springer, J. P. (1989). Three-dimensional structure of aspartyl protease fromhuman immunode®ciency virus HIV-1. Nature, 337,615-620.

Nermut, M. V. & Thomas, D. (1994). Fullerene-likeorganization of HIV gag-protein shell in virus-likeparticles produced by recombinant baculovirus. Vir-ology, 198, 288-296.

Nermut, M. V., Grief, C., Hashmi, S. & Hockley, D. J.(1993). Further evidence of icosahedral symmetry inhuman and simian immunode®ciency virus. AIDSRes. Hum. Retrovir. 9, 929-938.

Nicholson, L. K., Yamazaki, T., Torchia, D. A., Grzesiek,S., Bax, A., Kaufman, J. D., Stahl, S. J., Wing®eld,P., Lam, P. Y. S., Jadhav, P. K., Hodge, C. N.,Domaille, P. J. & Chong-Hwan, C. (1995). Flexibilityand function in HIV-1 protease. Nature Struct. Biol.2, 274-280.

Nie, Z., Bergeron, D., Subbramanian, R. A., Yao, X.-J.,Checroune, F., Rougeau, N. & Cohen, E. A. (1998).The putative alpha helix 2 of human immunode®-ciency virus type-1 Vpr contains a determinantwhich is responsible for the nuclear translocation ofproviral DNA in growth-arrested cells. J. Virol. 72,4104-4115.

Ohno, M., Fornerod, M. & Mattaj, I. W. (1998). Nucleo-cytoplasmic transport: the last 200 nanometers. Cell,92, 327-336.

Olshevsky, U., Helseth, E., Furman, C., Li, J., Haseltine,W. & Sodroski, J. (1990). Identi®cation of individualhuman immunode®ciency virus type 1 gp120amino acids important for CD4 receptor binding.J. Virol. 64, 5701-5707.

Omichinski, J. G., Clore, G. M., Sakaguchi, K., Appella,E. & Gronenborn, A. M. (1991). Structural character-

& Stammers, D. (1995). High resolution structuresof HIV-1 RT from four RT-inhibitor complexes.


ization of a 39-residue synthetic peptide containingthe two zinc binding domains from the HIV-1 p7nucleocapsid protein by CD and NMR spec-troscopy. FEBS Letters, 292, 25-30.

Perelson, A. S., Neumann, A. U., Markowitz, M.,Leonard, J. M. & Ho, D. D. (1996). HIV-1 dynamicsin vivo: virion clearance rate, infected cell life-span,and viral generation time. Science, 271, 1582-1586.

Piguet, V., Chen, Y. L., Mangasarian, A., Foti, M.,Carpentier, J. L. & Trono, D. (1998). Mechanism ofNef-induced CD4 endocytosis: Nef connects CD4with the mu chain of adaptor complexes. EMBO J.17, 2472-2481.

Pollard, V. W. & Malim, M. H. (1998). Annu. Rev. Micro-biol. in the press.

Popov, S., Rexach, M., Zybarth, G., Reiling, N., Lee,M. A., Ratner, L., Lane, C. M., Moore, M. S., Blobel,G. & Bukrinsky, M. (1998). Viral protein R regulatesnuclear import of the HIV-1 pre-integration com-plex. EMBO J. 17, 909-917.

Powers, R., Clore, G. M., Bax, A., Garrett, D. S., Stahl,S. J., Wing®eld, P. T. & Gronenborn, A. M. (1991).Secondary structure of the ribonuclease H domainof the human immunode®ciency virus reverse tran-scriptase in solution using three-dimensional doubleand triple resonance heteronuclear magnetic reson-ance spectroscopy. J. Mol. Biol. 221, 1081-1090.

Powers, R., Clore, G. M., Stahl, S. J., Wing®eld, P. J. &Gronenborn, A. (1992). Analysis of the backbonedynamics of the ribonuclease H domain of thehuman immunode®ciency virus reverse transcrip-tase using 15N relaxation measurements. Biochemis-try, 31, 9150-9157.

Profy, A. T., Salinas, P. A., Eckler, L. I., Dunlop, N. M.,Nara, P. L. & Putney, S. D. (1990). Epitopes recog-nized by the neutralizing antibodies of an HIV-1-infected individual. J. Immunol. 144, 4641-4647.

Puglisi, J. D., Tan, R., Calnan, B. J., Frankel, A. D. &Williamson, J. R. (1992). Conformation of the TARRNA-arginine complex by NMR spectroscopy.Science, 257, 76-80.

Puras-Lutzke, R. A., Vink, C. & Plasterk, R. H. A.(1994). Characterization of the minimal DNA-bind-ing domain of the HIV integrase protein. Nucl.Acids Res. 22, 4125-4131.

Rao, Z., Belyaev, A. S., Fry, E., Roy, P., Jones, I. M. &Stuart, D. I. (1995). Crystal structure of SIV matrixantigen and implications for virus assembly. Nature,378, 743-747.

Re, F., Braaten, D., Franke, E. K. & Luban, J. (1995).Human immunode®ciency virus type 1 Vpr arreststhe cell cycle in G2 by inhibiting the activation ofp34-cdc2-cyclin B. J. Virol. 69, 6859-6864.

Reicin, A. S., Paik, S., Berkowitz, R. D., Luban, J., Lowy,I. & Goff, S. P. (1995). Linker insertion mutations inthe human immunode®ciency virus type 1 gaggene: effects on virion particle assembly, release,and infectivity. J. Virol. 69, 642-650.

Reil, H., Bukovsky, A. A., Gelderblom, H. R. &Gottlinger, H. G. (1998). Ef®cient HIV-1 replicationcan occur in the absence of the viral matrix protein.EMBO J. 17, 2699-2708.

Reines, D., Conaway, J. W. & Conaway, R. C. (1996).The RNA polymerase II general elongation factors.Biochem. Sci. 9, 351-355.

Ren, J., Esnouf, R., Garman, E., Somers, D., Ross, C.,Kirby, I., Keeling, J., Darby, G., Jones, Y., Stuart, D.

Nature Struct. Biol. 2, 293-302.Rice, A. P. & Chan, F. (1991). Tat protein of human

immunode®ciency virus type 1 is a monomer whenexpressed in mammalian cells. Virology, 185, 451-454.

Rice, W. G. & Turpin, J. A. (1996). Virus-encoded zinc®ngers as targets for antiviral chemotherapy. Revs.Med. Virol. 6, 187-199.

Rice, W. G., Schaeffer, C. A., Graham, L., Bu, M.,McDougal, J. S., Orloff, S. L., Villinger, F., Young,M., Oroszlan, S., Fesen, M. R., Pommier, Y.,Mendeleyev, J. & Kun, E. (1993a). The site of anti-viral action of 3-nitrosobenzamide on the infectivityprocess of human immunode®ciency virus inhuman lymphocytes. Proc. Natl Acad. Sci. USA, 90,9721-9724.

Rice, W. G., Schaeffer, C. A., Harten, B., Villinger, F.,South, T. L., Summers, M. F., Henderson, L. E.,Bess, J. W., Jr, Arthur, L. O., McDougal, J. S., Orloff,S. L., Mendeleyev, J. & Kun, E. (1993b). Inhibitionof HIV-1 infectivity by zinc-ejecting aromatic C-nitroso compounds. Nature, 361, 473-475.

Rice, W. G., Supko, J. G., Malspeis, L., Buckheit, J. R. W.,Clanton, D., Bu, M., Grahm, L., Schaeffer, C. A.,Turpin, J. A., Domagala, J., Gogliotti, R., Bader, J. P.,Halliday, S. M., Coren, L., Sowder, R. C., II, Arthur,L. O. & Henderson, L. E. (1995). Inhibitors of HIVnucleocapsid protein zinc ®ngers as candidates forthe treatment of AIDS. Science, 270, 1194-1197.

Rice, W. G., Turpin, J. A., Clanton, D., Buckheit, R. W.,Jr, Summers, M. F., McDonnell, N., DeGuzman,R. N., Covell, D. G., Wallqvist, A., Zalkow, L.,Bader, J. P., Haugwitz, R. D. & Sausville, E. A.(1997). Azodicarbonamide inhibits HIV-1 replicationby targeting the nucleocapsid protein. Nature Med.3, 341-345.

Robert-Hebmann, V., Emiliani, S., Jean, F., Resnicoff, M.,Traincard, F. & Devaux, C. (1992). Clonal analysisof murine B cell response to the human immunode-®ciency virus type 1 (HIV-1)-gag p17 and p25 anti-gens. Mol. Immunol. 29, 729-738.

Rodgers, D. W., Gamblin, S. J., Harris, B. A., Ray, S.,Culp, J. S., Hellmig, B., Woolf, D. J., Debouck, C. &Harrison, S. C. (1995). The structure of unligandedreverse transcriptase from the human immunode®-ciency virus type 1. Proc. Natl Acad. Sci. USA, 92,1222-1226.

Rogel, M. E., Wu, L. I. & Emerman, M. (1995). Thehuman immunode®ciency virus type 1 vpr geneprevents cell proliferation during chronic infection.J. Virol. 69, 882-888.

Rose, R. B., Rose, J. R., Salto, R., Craik, C. S. & Stroud,R. M. (1993). Structure of the protease from simianimmunode®ciency virus: complex with an irrevers-ible nonpeptide inhibitor. Biochemistry, 32, 12498-12507.

Rose, S., Hensley, P., O'Shannessy, D. J., Culp, J.,Debouck, C. & Chaiken, I. (1992). Characterizationof HIV-1 p24 self-association using analytical af®-nity chromatography. Proteins: Struct. Funct. Genet.13, 112-119.

Rossmann, M. G. (1988). Antiviral agents targeted tointeract with viral capsid proteins and a possibleapplication to human immunode®ciency virus. Proc.Natl Acad. Sci. USA, 85, 4625-4627.

Rossmann, M. G. (1989). The canyon hypothesis. ViralImmunol. 2, 143-161.

Rossmann, M. G. & Johnson, J. E. (1989). IcosahedralRNA virus structure. Annu. Rev. Biochem. 58, 533-

Vif is cell species restricted: a role for Vif in deter-mining virus host range and cross-species trans-


573.Rossmann, M. G., Arnold, E., Erickson, J. W.,

Frankenberger, E. A., Grif®th, J. P., Hecht, H. J.,Johnson, J. E., Kamer, G., Luo, M., Mosser, A. G.,Rueckert, R. R., Sherry, B. & Vriend, G. (1985).Structure of a human common cold virus and func-tional relationship to other picornaviruses. Nature,317, 145-153.

Ryu, S.-E., Truneh, A., Sweet, R. W. & Hendrickson,W. A. (1994). Structure of an HIV and MHC bind-ing fragment from human CD4 as re®ned in twocrystal lattices. Structure, 2, 59-74.

Sakaguchi, K., Zambrano, N., Baldwin, E. T., Shapiro,B. A., Erickson, J. W., Omichinski, J. G., Clore,G. M., Gronenborn, A. M. & Appella, E. (1993). Ide-ni®cation of a binding site for the human immuno-de®ciency virus type 1 nucleocapsid protein. Proc.Natl Acad. Sci. USA, 90, 5219-5223.

Saksela, K., Cheng, G. & Baltimore, D. (1995). Proline-rich (PxxP) motifs in HIV-1 Nef bind to SH3domains of a subset of Src kinases and are requiredfor the enhanced growth of Nef� viruses but notfor down-regulation of CD4. EMBO J. 14, 484-491.

Salghetti, S., Mariani, R. & Skowronski, J. (1995).Human immunode®ciency virus type 1Nef andp56lck protein-tyrosine kinase interact with a com-mon element in CD4 cytoplasmic tail. Proc. NatlAcad. Sci. USA, 92, 349-353.

Sattentau, Q. & Moore, J. (1991). Conformationalchanges induced in the human immunode®ciencyvirus envelope glycoprotein by soluble CD4 bind-ing. J. Expt. Med. 174, 407-415.

Sattentau, Q. J., Moore, J. P., Vignaux, F., Traincard, F.& Poignard, P. (1993). Conformational changesinduced in the envelope glycoproteins of thehuman and simian immunode®ciency viruses bysoluble receptor binding. J. Virol. 67, 7383-7398.

Sawai, E. T., Baur, A., Struble, H., Peterlin, B. M., Levy,J. A. & Cheng-Mayer, C. (1994). Human immunode-®ciency virus type 1 Nef associates with a cellularserine kinase in T lymphocytes. Proc. Natl Acad. Sci.USA, 91, 1539-1543.

Schubert, U., AntoÂn, L. C., Bacik, J., Cox, J. H., Bour, S.,Bennick, J. R., Orlowski, M., Strebel, K. & Yewdell,J. W. (1998). CD4 glycoprotein degradation inducedby human immuno-de®ciency virus type I Vpu pro-tein requires the function of proteosomes and theubiquitin-conjugating pathway. J. Virol. 72, 2280-2288.

Schwartz, O., Marechal, V., Danos, O. & Heard, J.-M.(1995). Human immunode®ciency virus type 1 nefincreases the ef®ciency of reverse transcription inthe infected cell. J. Virol. 69, 4053-4059.

Shailer, P., Brodfuehrer, J., Sedman, A. & Vassos, A.(1997). Single-dose safety, tolerance, and pharmaco-kinetics of CI-1012, a new HIV antiretroviral agent,in healthy volunteers. In Proceedings: Fourth Confer-ence on Retroviruses and Opportunistic Infections,Abstract no. 229.

Shoji, S., Tashiro, A. & Kubota, Y. (1990). Antimyristoy-lation of gag proteins in human T-cell lymphotropicand human imunode®ciency viruses by N-myristoylglycinal diethylacetal. Ann. N. Y. Acad. Sci. 616, 97-115.

Simon, J. H., Miller, D. L., Fouchier, R. A., Soares, M. A.,Peden, K. W. & Malim, M. H. (1998). The regulationof primate immunode®ciency virus infectivity by

mission. EMBO J. 17, 1259-1267.Skalka, A. M. & Goff, S. P. (1993). Reverse Transcriptase,

Cold Spring Harbor Laboratory Press, Cold SpringHarbor, NY.

Slice, L. W., Codner, E., Antelman, D., Holly, M.,Wegrzynski, B., Wang, J., Toome, V., Hsu, M. C. &Nalin, C. M. (1992). Characterization of recombi-nant HIV-1 Tat and its interaction with TAR RNA.Biochemistry, 31, 12062-12068.

Smerdon, S. J., Jager, J., Wang, J., Kohlstaedt, L. A.,Chirino, A. J., Friedman, J. M., Rice, P. A. & Steitz,T. A. (1994). Structure of the binding site for nonnu-cleoside inhibitors of the reverse transcriptase ofhuman immunode®ciency virus type-1. Proc. NatlAcad. Sci. USA, 91, 3911-3915.

Smith, T. J., Kremer, M. J., Luo, M., Vriend, G., Arnold,E., Kamer, G., Rossmann, M. G., McKinlay, M. A.,Diana, G. D. & Otto, M. J. (1986). The site of attach-ment in human rhinovirus 14 for antiviral agentsthat inhibit uncoating. Science, 233, 1286-1293.

Sourgen, F., Maroun, R. G., Frere, V., Bouziane, M.,Aucoair, C., Troalen, F. & Fermandjian, S. (1996). Asynthetic peptide from the human immunode®-ciency virus type-1 integrase exhibits coiled-coilproperties and interferes with the in vitro inte-gration activity of the enzyme. Eur. J. Biochem. 240,765-773.

South, T. L. & Summers, M. F. (1993). Zinc- andsequence-dependent binding to nucleic acids by theN-terminal zinc ®nger of the HIV-1 nucleocapsidprotein: NMR structure of the complex with thePsi-site analog, dACGCC. Protein Sci. 2, 3-19.

South, T. L., Kim, B. & Summers, M. F. (1989). 113CdNMR studies of a 1:1 Cd adduct with an 18-residuezinc ®nger peptide from HIV-1 nucleic acid bindingprotein, p7. J. Am. Chem. Soc. 111, 395-396.

South, T. L., Kim, B., Hare, D. R. & Summers, M. F.(1990). Zinc ®ngers and molecular recognition.Structure and nucleic acid binding studies of anHIV zinc ®nger-like domain. Biochem. Pharm. 40,123-129.

South, T. L., Blake, P. R., Hare, D. R. & Summers, M. F.(1991). C-terminal retroviral-type zinc ®ngerdomain from the HIV-1 nucleocapsid protein isstructurally similar to the N-terminal zinc ®ngerdomain. Biochemistry, 30, 6342-6349.

Sova, P. & Volsky, D. J. (1993). Ef®ciency of viral DNAsynthesis during infection of permissive and non-permissive cells with vif-negative human immuno-de®ciency virus type 1. J. Virol. 67, 6322-6326.

Sova, P., Van Ranst, M., Gupta, P., Balachandran, R.,Chao, W., Itescu, S., McKinley, G. & Volsky, D. J.(1995). Conservation of an intact vif gene in vitroand in vivo. J. Virol. 69, 2557-2564.

Spearman, P., Wang, J.-J., Vander, Heyden N. & Ratner,L. (1994). Identi®cation of human immunode®-ciency virus type 1 gag protein domains essential tomembrane binding and particle assembly. J. Virol.68, 3232-3242.

Starcich, B. R., Hahn, B. H., Shaw, G. M., McNeely,P. D., Modrow, S., Wolf, H., Parks, E. S., Park,W. P., Josephs, S. F., Gallo, R. C., et al. (1986).Identi®cation and characterization of conserved andvariable regions of the envelope gene HTLV-III/LAV, the retrovirus of AIDS. Cell, 45, 637-648.

Steinkasserer, A., Harrison, R., Billich, A.,Hammerschmid, F., Werner, G., Wolff, B., Peichl,

P., Pal®, G., Schnitzel, W. & Mlynar, E., et al. (1995).Mode of action of SDZ NIM 811, a nonimmunosup-

Tong, L., Pav, S., Pargellis, C., Do, F., Lamarre, D. &Anderson, P. C. (1993). Crystal structure of human


pressive cyclosporin A analog with activity againsthuman immunode®ciency virus type 1 (HIV-1):interferrence with early and late events in HIV-1.J. Virol. 69, 814-824.

Stys, D., Blaha, I. & Strop, P. (1993). Structural and func-tional studies in vitro on the p6 protein from theHIV-1 gag open reading frame. Biochim. Biophys.Acta, 1182, 157-161.

Summers, M. F. (1991). Zinc ®nger motif for single-stranded nucleic acids? Investigations by nuclearmagnetic resonance. J. Cell. Biochem. 45, 41-48.

Summers, M. F., South, T. L., Kim, B. & Hare, D. R.(1990). High-resolution structure of an HIV zinc ®n-gerlike domain via a new NMR-based distance geo-metry approach. Biochemistry, 29, 329-340.

Summers, M. F., Henderson, L. E., Chance, M. R., Bess,J. W. J., South, T. L., Blake, P. R., Sagi, I., Perez-Alvarado, G., Sowder, R. C. I., Hare, D. R. &Arthur, L. O. (1992). Nucleocapsid zinc ®ngersdetected in retroviruses: EXAFS studies of intactviruses and the solution-state structure of thenucleocapsid protein from HIV-1. Protein Sci. 1, 563-574.

Surovoy, A., Waidelich, D. & Jung, G. (1992). Nucleo-capsid protein of HIV-1 and its Zn2� complex for-mation analysis with electrospray massspectrometry. FEBS Letters, 311, 259-262.

Surovoy, A., Dannull, J., Moelling, K. & Jung, G. (1993).Conformational and nucleic acid binding studies onthe synthetic nucleocapsid protein of HIV-1. J. Mol.Biol. 229, 94-104.

Swanstrom, R. & Wills, J. W. (1997). Synthesis, assemblyand processing of viral proteins. In Retroviruses(Cof®n, J. M., Hughes, S. H. & Varmus, H. E., eds),pp. 263-334, Cold Spring Harbor Laboratory Press,Cold Spring Harbor, NY.

Sweet, R. W., Truneh, A. & Hendrickson, W. A. (1991).CD4: its structure, role in immune function andAIDS pathogenesis, and potential as a pharmaco-logical target. Curr. Opin. Biotech. 2, 622-633.

Tan, K., Liu, J.-H., Wang, J.-H., Shen, S. & Lu, M. (1997).Atomic structure of a thermostable subdomain ofHIV-1 gp41. Proc. Nat. Acad. Sci. USA, 94, 12303-12308.

Tan, R. & Frankel, A. D. (1994). Costabilization of pep-tide and RNA structure in an HIV Rev peptide-RREcomplex. Biochemistry, 33, 14579-14585.

Tanchou, V., Decimo, D., Pechoux, C., Lener, D.,Rogemond, V., Berthoux, L., Ottmann, M. & Darlix,J.-L. (1998). Role of the N-terminal zinc ®nger ofhuman immunode®ciency virus type 1 nucleocapsidprotein in virus structure and replication. J. Virol.72, 4442-4447.

Telesnitsky, A. & Goff, S. P. (1997). Reverse transcriptaseand the generation of retroviral DNA. In Retro-viruses (Cof®n, J. M., Hughes, S. H. & Varmus,H. E., eds), vol. 1, pp. 121-160, Cold Spring HarborLaboratory Press, Cold Spring Harbor, NY.

Thali, M., Moore, J. P., Furman, C., MacArthur, C., Ho,D. D., Robinson, J. & Sodroski, J. (1993). Character-ization of conserved human immunode®ciencyvirus type 1 gp120 neutralization epitopes exposedupon gp120-CD4 binding. J. Virol. 67, 3978-3988.

Thali, M., Bukovsky, A., Kondo, E., Rosenwirth, B.,Walsh, C. T., Sodroski, J. & Gottlinger, H. G. (1994).Functional association of cyclophilin A with HIV-1virions. Nature, 372, 363-365.

imunode®ciency virus (HIV) type 2 protease incomplex with a reduced amide inhibitor and com-parison with HIV-1 protease. Proc. Natl Acad. Sci.USA, 90, 8387-8391.

Trkola, A., Dragic, T., Arthos, J., Binley, J. M., Olson,W. C., Allaway, G. P., Cheng-Mayer, C., Robinson,J., Maddon, P. J. & Moore, J. P. (1996). CD4-depen-dent, antibody-sensitive interactions between HIV-1and its co-receptor CCR-5. Nature, 384, 184-187.

Trono, D. (1995). HIV accessory proteins: leading rolesfor the supporting cast. Cell, 82, 189-192.

Trono, D. & Gallay, P. (1997). Letter, in response toFreed et al. Cell, 88, 173-174.

Unge, T., Knight, S., Bhikhabhai, R., Lovgren, S., Dauter,Z., Wilson, K. & Strandberg, B. (1994). 2.2 AÊ resol-ution structure of the amino-terminal half of HIV-1reverse transcriptase (®ngers and palm subdo-mains). Curr. Biol. 2, 953-961.

Valegard, K., Liljas, L., Fridborg, K. & Unge, T. (1990).The three-dimensional structure of the bacterialvirus MS2. Nature, 345, 36-41.

Vandevelde, M., Witvrouw, M., Schmit, J.-C., Sprecher,S., DeClercq, E. & Tassignon, J. P. (1996). ADA, apotential anti-HIV drug. AIDS Res. Hum. Retrovir.12, 567-568.

Varmus, H. & Swanstrom, R. (1984). Replication of ret-roviruses. In Molecular Biology of Tumor Viruses.RNA Tumor Viruses (Weiss, R., Teich, N., Varmus,H. & Cof®n, J., eds), vol. 1, pp. 369-512, ColdSpring Harbor Laboratory Press, Cold SprngHarbor, NY.

Vodicka, M. A., Koepp, D. M., Silver, P. A. & Emerman,M. (1998). HIV-1 Vpr interacts with the nucleartransport pathway to promote macrophage infec-tion. Genes Dev. 12, 175-185.

Vogt, V. M. (1997). Retroviral virions and genomes. InRetroviruses (Cof®n, J. M., Hughes, S. H. & Varmus,H. E., eds), vol. 1, pp. 27-69, Cold Spring HarborLaboratory Press, Cold Spring Harbor, NY.

Von Poblotzki, A., Wagner, R., Niedrig, M., Wanner, G.,Wolf, H. & Modrow, S. (1993). Identi®cation of aregion in the Pr55gag polyprotein essential for HIV-1 particle formation. Virology, 193, 981-985.

von Schwedler, U., Song, J., Aiken, C. & Trono, D.(1993). Vif is crucial for human immunode®ciencyvirus type 1 proviral DNA synthesis in infectedcells. J. Virol. 67, 4945-4955.

von Schwedler, U., Kornbluth, R. S. & Trono, D. (1994).The nuclear localization signal of the matrix proteinof human immunode®ciency virus type 1 allowsthe establishment of infection in macrophages andquiescent t lymphocytes. Proc. Natl Acad. Sci. USA,91, 6992-6996.

von Schwedler, U. K., Stemmler, T. L., Klishko, V. Y.,Li, S., Albertine, K. H., Davis, D. R. & Sundquist,W. I. (1998). Proteolytic refolding of the HIV-1 cap-sid protein amino-terminus facilitates viral coreassembly. EMBO J. 17, 1555-1568.

Wang, C.-T. & Barklis, E. (1993). Assembly, processing,and infectivity of human immunode®ciency virustype 1 gag mutants. J. Virol. 67, 4264-4273.

Wang, C.-T., Zhang, Y., McDermott, J. & Barklis, E.(1993). Conditional infectivity of a human immuno-de®ciency virus matrix domain deletion mutant.J. Virol. 67, 7067-7076.

Wang, Y.-X., Freedberg, D. I., Grzesiek, S., Torchia,D. A., Wing®eld, P. T., Kaufman, J. D., Stahl, S. J.,

Chang, C.-H. & Hodge, C. N. Q. (1996). Mappinghydration water molecules in the HIV-1 protease/

Wu, W., Henderson, L. E., Copeland, T. D., Gorelick,R. J., Bosche, W. J., Rein, A. & Levin, J. G. (1996).


DMP323 complex in solution by NMR spec-troscopy. Biochemistry, 35, 12694-12704.

Watenpaugh, K. D., Sieker, L. C., Herriott, J. R. &Jensen, L. H. (1973). Re®nement of the model of aprotein: rubredoxin at 1.5 AÊ resolution. Acta Crystal-log. sect. B, 29, 943-956.

Wei, P., Garber, M. E., Fang, S.-M., Fischer, W. H. &Jones, K. A. (1998). A novel CDK9-associatedC-type cyclin interacts directly with HIV-1 Tat andmediates its high-af®nity, loop-speci®c binding toTAR RNA. Cell, 92, 451-462.

Wei, X., Ghosh, S. K., Taylor, M. E., Johnson, V. A.,Emini, E. A., Deutsch, P., Lifson, J. D., Bonhoeffer,S., Nowak, M. A. & Hahn, B. H., et al. (1995). Viraldynamics in human immunode®ciency virus type 1infection. Nature, 373, 117-122.

Weissenhorn, W., Dessen, A., Harrison, S. C., Skehel, J. J.& Wiley, D. C. (1997). Atomic structure of the ecto-domain from HIV-1 gp41. Nature, 387, 426-430.

Whitcomb, J. M. & Hughes, S. H. (1992). Retroviralreverse transcription and integration: progress andproblems. Annu. Rev. Cell Biol. 8, 275-306.

Wiegers, K., Rutter, G., Kottler, H., Tessmer, U.,Hohenberg, H. & Krausslich, H.-G. (1998). Sequen-tial steps in human immunode®ciency virus particlematuration revealed by alterations of individualGag polyprotein cleavage sites. J. Virol. 72, 2846-2854.

Willbold, D., Rosin-Arbesfeld, R., Sticht, H., Frank, R. &Rosch, P. (1994). Structure of the equine infectiousanemia virus Tat protein. Science, 264, 1584-1587.

Wills, J. W., Craven, R. C., Weldon, R. A., Jr, Nelle, T. D.& Erdie, C. R. (1991). Suppression of retroviral MAdeletions by the amino-terminal membrane-bindingdomain of p60src. J. Virol. 65, 3804-3812.

Wlodawer, A. & Erickson, J. W. (1993). Structure-basedinhibitors of HIV-1 proteinase. Annu. Rev. Biochem.62, 543-585.

Wlodawer, A., Miller, M., Jaskolski, M., Sathyanarayana,B. K., Baldwin, E., Weber, I. T., Selk, L. M.,Clawson, L., Schneider, J. & Kent, B. H. S. (1989).Conserved folding in retroviral proteases: crystalstructure of a synthetic HIV-1 protease. Science, 245,616-621.

Wlodawer, A., Gustchina, A., Reshetnikova, L.,Lubkowski, J., Zdanov, A., Hui, K. Y., Angleton,E. L., Farmerie, W. G., Goodenow, M. M., Bhatt, D.,Zhang, L. & Dunn, B. M. (1995). Structure of aninhibitor complex of the proteinase from felineimmunode®ciency virus. Nature Struct. Biol. 2, 480-488.

Wong, J. K., Hezareh, M., Gunthard, H. F., Havlir, D. V.,Ignacio, C. C., Spina, C. A. & Richman, D. D.(1997). Recovery of replication-competent HIVdespite prolonged suppression of plasma viremia.Science, 278, 1291-1295.

Worthylake, D. K., Wang, H., Yoo, S., Sundquist, W. I.& Hill, C. P. (1998). Structures of the HIV-1 capsidprotein dimerization domain at 2.6 AÊ resolution.Acta Crystallog. sect. D, in the press.

Wu, L., Gerard, N. P., Wyatt, R., Choe, H., Parolin, C.,Ruf®ng, N., Borsetti, A., Cardoso, A. A., Desjardin,E., Newman, W., Garard, C. & Sodroski, J. (1996).CD4-induced interaction of primary HIV-1 gp120glycoproteins with the chemokine receptor CCR5.Nature, 384, 179-183.

Human immunode®ciency virus type 1 nucleocap-sid protein reduces reverse transcriptase pausing ata secondary structure near the murine leukemiavirus polypurine tract. J. Virol. 70, 7132-7142.

Wyatt, R. & Sodroski, J. (1998). The HIV-1 envelope gly-coproteins: fusogens, antigens and immunogens.Science, 280, 1884-1888.

Wyatt, R., Moore, J., Accola, M., Desjardin, E.,Robinson, J. & Sodroski, J. (1995). Involvement ofthe V1/V2 variable loop structure in the exposureof human immunode®ciency virus type 1 gp120epitopes induced by receptor binding. J. Virol. 69,5723-5733.

Wyatt, R., Kwong, P. D., Desjardins, E., Sweet, R. W.,Robinson, J., Hendrickson, W. A. & Sodroski, J. G.(1998). The antigenic structure of the HIV gp120envelope glycoprotein. Nature, 393, 705-711.

Yamazaki, T., Hinck, A. P., Wang, Y.-X., Nicholson,L. K., Torchia, D. A., Wing®eld, P., Stahl, S. J.,Kaufman, J. D., Chang, C.-H., Domaille, P. J. &Lam, P. Y. S. (1996). Three-dimensional solutionstructure of the HIV-1 protease complexed withDMP323, a novel cyclic urea-type inhibitor, deter-mined by nuclear magnetic resonance spectroscopy.Protein Sci. 5, 495-506.

Ye, X., Kumar, R. A. & Patel, D. J. (1995). Molecular rec-ognition in the bovine immunode®ciency virus tatpeptide-TAR RNA complex. Chem. Biol. 2, 827-840.

Ye, X., Gorin, A., Ellington, A. D. & Patel, D. J. (1996).Deep penetration of an a-helix into a widened RNAmajor groove in the HIV-1 rev peptide-RNA apta-mer complex. Nature Struct. Biol. 3, 1026-1033.

Yoo, S., Myszka, D. G., Yeh, C., McMurray, M., Hill,C. P. & Sundquist, W. I. (1997). Molecular recog-nition in the HIV-1 capsid/cyclophilin A complex.J. Mol. Biol. 269, 780-795.

Yu, X., Yuan, X., Matsuda, Z., Lee, T.-H. & Essex, M.(1992). The matrix protein of human immunode®-ciency virus type 1 is required for incorporation ofviral envelope protein into mature virions. J. Virol.66, 4966-4971.

Yu, X., Yuan, X., McLane, M. F., Lee, T.-H. & Essex, M.(1993). Mutations in the cytoplasmic domain ofhuman immunode®ciency virus type 1 transmem-brane protein impair the incorporation of env pro-teins into mature virions. J. Virol. 67, 213-221.

Yu, X., Hathout, Y., Fenselau, C., Sowder, R. C., II,Henderson, L. E., Rice, W. G., Mendeleyev, J. &Kun, E. (1995). Speci®c disul®de formation in theoxidation of HIV-1 zinc ®nger protein nucleocapsidp7. Chem. Res. Toxicol. 8, 586-590.

Yuan, X., Yu, X., Lee, T.-H. & Essex, M. (1993).Mutations in the N-terminal region of humanimmunode®ciency virus type 1 matrix protein blockintracellular transport of the gag precursor. J. Virol.67, 6387-6394.

Zang, Y., Qian, H., Love, Z. & Barklis, E. (1998). Anal-ysis of the assembly function of the human immu-node®ciency virus type 1 gag protein nucleocapsiddomain. J. Virol. 72, 1782-1789.

Zapp, M. L., Hope, T. J., Parslow, T. G. & Green, M. R.(1991). Oligomerization and RNA binding domainsof the type 1 human immunode®ciency virus revprotein: a dual function for an arginine rich bindingmotif. Proc. Natl Acad. Sci. USA, 88, 7734-7738.

Zhang, Y. & Barklis, E. (1997). Effects of nucleocapsidmutations on human immunode®ciency virus

assembly and RNA encapsidation. J. Virol. 71, 6765-6776.

Human rhinovirus 3 at 3.0 AÊ resolution. Structure,4, 1205-1220.


Zhao, B., Winborne, E., Minnich, M. D., Culp, J. S.,Debouck, C. & Abdel, M. S. S. (1993). Three-dimen-sional structure of a simian immunode®ciency virusprotease/inhibitor complex. Implications for thedesign of human immunode®ciency virus type 1and 2 protease inhibitors. Biochemistry, 32, 13054-13060.

Zhao, R., Pevear, D. C., Kremer, M. J., Giranda, V. L.,Kofron, J. A., Kuhn, R. J. & Rossman, M. G. (1996).

Zheng, R., Jenkens, T. M. & Craigie, R. (1996). Zincfolds the N-terminal domain of HIV-1 integrase,promotes multimerization, and enhances catalyticactivity. Proc. Natl Acad. Sci. USA, 93, 13659-13664.

Zhou, W., Parent, L. J., Wills, J. W. & Resh, M. D.(1994). Identi®cation of a membrane-bindingdomain within the amino-terminal region of humanimmunode®ciency virus type 1 gag protein whichinteracts with acidic phospholipids. J. Virol. 68,2556-2569.

Edited by P. E. Wright

(Received 23 September 1998; received in revised form 20 October 1998; accepted 21 October 1998)

Note added in proof: In a major breakthrough, the X-ray structure of a catalytically relevant HIV-1reverse transcriptase complex with DNA and a deoxynucleoside triphosphate has recently beendetermined [Huang, H., Chopra, R., Verdine, G. L., Harrison, S. C. (1998). Structure of a covalentlytrapped catalytic complex of HIV-1 reverse transcriptase: Implications for drug resistance. Science282, 1669-1675].

Structural Biology of HIV - RCSB · Structural Biology of HIV BrianG.TurnerandMichaelF.Summers*...

Documents

Transcript of Structural Biology of HIV - RCSB · Structural Biology of HIV BrianG.TurnerandMichaelF.Summers*...