FEBS Letters 587 (2013) 2943–2951

journal homepage: www.FEBSLetters .org

The V1/V2 loop of HIV-1 gp120 is necessary for Tat binding andconsequent modulation of virus entry

0014-5793/$36.00 � 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.febslet.2013.07.039

⇑ Corresponding author. Address: Lab of Tumor Microenvironment, Institute forCancer Research at Candiolo (IRCC), sp 142 Km 3.95, 10060 Candiolo, Italy. Fax: +39011 9933524.

E-mail address: serena.marchio@ircc.it (S. Marchiò).

Sabrina Cardaci a,b, Marco Soster a,b, Federico Bussolino a,c, Serena Marchiò a,b,⇑a Department of Oncology, University of Torino, Candiolo, Italyb Lab of Tumor Microenvironment, Institute for Cancer Research at Candiolo (IRCC), Candiolo, Italyc Lab of Vascular Oncology, Institute for Cancer Research at Candiolo (IRCC), Candiolo, Italy

a r t i c l e i n f o

Article history:Received 23 April 2013Revised 18 July 2013Accepted 19 July 2013Available online 31 July 2013

Edited by Hans-Dieter Klenk

Keywords:HIV-1gp120TatVirus entryPeptide

a b s t r a c t

Preventing cell entry of human immunodeficiency virus 1 (HIV-1) is of interest for the developmentof innovative therapies. We previously reported a specific interaction between HIV-1 envelope gly-coprotein 120 (gp120) and Tat at the cell surface, which enhances virus attachment and entry. Wealso identified a gp120-mimicking peptide, CT319, that competes with gp120 for Tat binding, thusinhibiting HIV-1 infection. Here we report a molecular dissection of gp120 regions involved in thismechanism. Our findings identify the V1/V2 loop of gp120 as involved in Tat binding, and define thisinteraction as functionally relevant for HIV-1 entry into host cells.

Structured digital abstract:Tat physically interacts with gp120 by pull down (1, 2, 3, 4)

� 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction

Blocking the initial steps of HIV-1 infection is a promising ap-proach for innovative antiretroviral therapies. The classic pathwayof HIV-1 infection starts with a high-affinity binding of the viralenvelope glycoprotein 120 (gp120) to the cell surface receptorCD4 [1–3]; this event induces conformational changes in gp120that expose structural elements for the recognition of the corecep-tor chemokine (CXC-motif) receptor 4 (CXCR4) or chemokine (C–Cmotif) receptor 5 (CCR5) [4–6]. Eventually, these interactions allowthe N-terminal hydrophobic portion of gp41 to insert into the hostcell plasma membrane and to induce membrane fusion [7,8].

Despite the high number of entry inhibitors being investigated,only two have reached the market: the fusion inhibitor Enfuvirtide(T-20, Fuzeon, Roche) [9] and the CCR5 antagonist Maraviroc (Selz-entry/Celsentri, ViiV Helthcare) [10]; and four are currently inphase II/III clinical trials: the CCR5 antagonists Aplaviroc (AK602/GSK-873140, GlaxoSmithKline) [11] and Cenicriviroc (TBR-652,Tobira Theapeutics) [12], and the attachment/post attachmentinhibitors BMS-663068 (Bristol–Myers Squibb) [13] and Ibal-

izumab (TNX-355, TaiMed Biologics) [14]. All these compounds,however, suffer of drawbacks related to drug resistance and tothe fact that most of them are strain-specific, thus limiting theirefficacy in the whole population of HIV-1-infected patients. Thedevelopment of new, broad spectrum entry inhibitors remainstherefore an important medical need.

In this growing field of investigation, we provided insights intoa previously unknown mechanism: a high affinity interaction be-tween two viral proteins, gp120 and Tat, occurs at the surface ofuninfected cell and leads to enhanced HIV-1 infectivity [15]. Tatwas first identified as a transactivator of viral transcription [16];however, this protein can also be secreted [17,18], exploiting sev-eral effects while in the extracellular environment [19,20]. SolubleTat is partially sequestered by heparan sulfate proteoglycans, thusbeing concentrated on the cell surfaces and protected from prote-olytic degradation [19,21,22]. Gp120 constitutes the surface unit ofHIV-1 [23], being present on the viral particles as a heterotrimernon-covalently associated with three gp41 subunits. Gp120 ischaracterized by 5 regions, conserved among different viral strains(C1–C5), and by 5 variable regions (V1–V5) [24]; the presence ofintramolecular disulfide bonds in the latter allows the formationof loop-like structures [25]. We mapped the gp120-interacting sitein Tat, and we described a gp120-mimicking peptide, namelyCT319, that mimics a portion of gp120 and inhibits HIV-1 entrythrough interference with gp120/Tat binding [15].

2944 S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951

Here we provide a molecular dissection of Tat-interacting sitesin gp120, showing that the V1/V2 loop is indispensable for Tatbinding and cell entry, and that a basic stretch in this region isresponsible for Tat affinity and viral infectivity.

2. Materials and methods

2.1. Reagents and cell cultures

The following reagents were provided by the Programme Euro-pean Vaccine against AIDS (EVA)/Centre for AIDS Reagents (CFAR)at the National Institute for Biological Standards and Controls (NIB-SC), supported by the European Community Frame Programme 6/7(EC FP6/7) Europrise Network of Excellence, AIDS Vaccine IntegratedProject (AVIP) and Next Generation HIV-1 Immunogens inducingbroadly reactive Neutralising antibodies (NGIN) Consortia, and theBill and Melinda Gates Global HIV Vaccine Research Cryoreposito-ry-Collaboration for AIDS Vaccine Discovery (GHRC-CAVD) Project,UK: anti-gp120 monoclonal antibody (ICR38/ARP388, donors Dr. J.Cordell and Dr. C. Dean), C8166 (ARP013, donor Dr. G. Farrar), andU937 cells (ARP012, donor Dr. G. Farrar), pCAGGS SF162 gp160 plas-mid, coding the R5-tropic HIV-1 env [26] (ARP2125, donors Dr. L.Stamatatos and Dr. C. Cheng-Mayer). The horseradish peroxidase(HRP)-conjugated antibody was from Jackson Laboratories (WestGrove, PA). Peptides were from New England Peptide (Fitchburg,MA). 293T cells (ATCC CRL-11268) were from LG Promochem (SestoSan Giovanni, Italy). Cells were cultured in Iscove’s modified Dul-becco’s medium (IMDM) supplemented with 10% fetal bovine serum(FBS), 2 mM glutamine, 200 U/ml penicillin and 200 lg/ml strepto-mycin (Lonza, Rockland, ME).

2.2. Deletion mutants of the gp120 coding sequence

Plasmids encoding deletion mutants of the gp120 protein weregenerated by a combination of PCR amplifications, restriction en-zyme digestions and fragment ligations. To delete the V1 loop,PCR amplifications of pEnvHXB [27] were performed using the fol-lowing primer pair combinations (Sigma–Aldrich, Milan, Italy):

A: env16 (50-TCGAGACCGGGCCTTTGT-30)50V1 (50-GTCTGAGTCGCCGGCTCCAGTGCACTTTAAACTAACACA

GAG-3’);B: 30V1 (50-GTCTGAGTCGCCGGCAACTGCTCTTTCAATATCAG-30)env11 (50-TCAGTACAATCTGCTCTG-30).The product of reaction A, digested with Sac I and Nae I, was li-

gated with the product of reaction B, digested with Nae I and Xho I,and subcloned into pEnvHXB. To delete the V2 loop, PCR amplifica-tions were performed with the following primer pairs:

C: env16 (50-TCGAGACCGGGCCTTTGT-30)50V2 (50-GTCTGAGTCGGATCCGGCACCAGAGCAGTTTTTTATCTC

TCC-30);D: 30V2 (50-GTCTGAGTCGGATCCTGTAACACCTCAGTCATTACA

CAG-30)env11 (50-TCAGTACAATCTGCTCTG-30).The product of reaction C, digested with Sac I and Bam HI, was

ligated with the product of reaction D, digested with Bam HI andXho I, and subcloned into pEnvHXB. To delete the V1/V2 loop, PCRamplifications were performed with the following primer pairs:

E: env16 (50-TCGAGACCGGGCCTTTGT-30)50V1V2 (50-GTCTGAGTCGGATCCGGCACCAGAGCAGTTGCCGGCT

CCAGTGCAC TTTAAACTAACACAGAG-30)F: 30V2 (50-GTCTGAGTCGGATCCTGTAACACCTCAGTCATTACA

CAG-30)env11 (50-TCAGTACAATCTGCTCTG-30).The product of reaction E, digested with Sac I and Bam HI, was

ligated with the product of reaction F, digested with Bam HI andXho I, and subcloned into pEnvHXB.

2.3. Point mutants of the gp120 coding sequence

Point mutations were introduced into pEnvHXB using the Quick-Change Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA)with the following primer pairs (Sigma–Aldrich):

K168A: fw K168A (50-AGCATAAGAGGTGCGGTGCAGAAAGAA-30)rv K168A (50-TTCTTTCTGCACCGCACCTCTTATGCT-30);Q170A: fw Q170A (50-AGAGGTAAGGTGGCAAAAGAATATGCA-30)rv Q170A (50-TGCATATTCTTTTGCCACCTTACCTCT-30);K171A: fw K171A (50-GGTAAGGTGCAGGCAGAATATGCATTT-30)rv K171A (50-AAATGCATATTCTGCCTGCACCTTACC-30);K168A/Q170A: fw K168A/Q170A (50-AGCATAAGAGGTGCGGTG

GCAAAAGAA-30)rv K168A/Q170A (50-TTCTTTTGCCACCGCACCTCTTATGCT-30);K168A/K171A: this mutant was obtained using the K168A and

K171A primer pairs consecutively;Q170A/K171A: fw Q170A/K171A (50-GGTAAGGTGGCAGCAGAA

TATGCATTT-30)rv Q170A/K171A (50-AAATGCATATTCTGCTGCCACCTTACC-30);K168A/Q170A/K171A:fw K168A/Q170A/K171A (50-AGAGGTGCGGTGGCAGCAGAATAT

GCA-30)rv K168A/Q170A/K171A (50-TGCATATTCTGCTGCCACCGCACCT

CT-30).

2.4. Production and transduction of HIV-1-derived lentiviral vectors

Gp120-pseudotyped lentiviral vectors (LVs) were produced aspreviously described [15]. Briefly, cotransfection was performedin 293T (packaging) cells using the following plasmidcombinations:

Tat+/gp120-LVs [28]: pRRL.hPGK.GFP.SIN-18 + pCMVDR8.2 +pEnvHXB.

Tat�/gp120-LVs: pRRL.hPGK.GFP.SIN-18 + pMDLg/pRRE +pRSV-Rev [29]+pEnvHXB.

R5-Tat�/gp120-LVs: pRRL.hPGK.GFP.SIN-18 + pMDLg/pRRE +pRSV-Rev + pCAGGS SF162 gp160.

After 16 h, cells were extensively washed to remove free plas-mid DNA. Supernatants were collected 72 h post-transfection andfiltered to remove any free cell and cellular debris. The numberof viral particles in these supernatants was quantified with theHIV-1 p24 ELISA kit (Perkin Elmer, Boston, MA). Viral preparationswere used to transduce 1 � 105 C8166 (host) cells in the presenceof 8 lg/ml polybrene. The multiplicity of infection (MOI) was 5.5and 1.5 for Tat�/gp120-LVs and Tat+/gp120-LVs, respectively.Treatments with recombinant Tat (7 nM) and with synthetic pep-tides were performed as described [15]. Optimization of the pep-tide concentrations and host cell type for the entry assays isdetailed in the Supplementary Material (Supplementary Fig. 1).

2.5. Quantification of virus entry into host cells

LV-transduced C8166 cells were analyzed for the expression ofgreen fluorescent protein (GFP) by cytofluorimetric analysis (FAC-SCalibur, BD Biosciences, Franklin Lakes, NJ) 72 h after infection, aswe previously described [15]. LV entry was also quantified by Real-Time Polymerase Chain Reaction (Real-Time PCR) analysis of intra-cellular HIV-1 transcripts 2 h after infection, adapting previouslypublished protocols [30,31]. Briefly, cells were washed once withphosphate buffered saline (PBS), and lysed in 100 ll of 10 mMTris–HCl (pH 8.3) containing 50 mM KCl, 2.5 mM MgCl2, 0.1 mg/ml gelatin, 0.45% Nonidet P-40, 0.45% Tween 20, and 100 lg/mlproteinase K. After protein digestion (2 h at 56 �C) and inactivationof the proteinase (10 min at 95 �C), 10 ll of cell lysate was directlysubjected to Real-Time PCR using the SYBR� Green PCR Master Mix(Applied Biosystems) with the following primers:

S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951 2945

fw R/U5 (50-GGCTAACTAGGGAACCCACTG-30) 300 nMrv R/U5 (50-CTGCTAGAGATTTTCCACACTGAC -30) 50 nMin a total volume of 25 ll. For each experiment, a reference

curve was prepared with serial dilutions (102 to 106 copies) ofthe pRRL.hPGK.GFP.SIN-18 plasmid, and a negative control wasperformed with water only. Further controls for assay optimizationare detailed in the Supplementary Material (Supplementary Fig. 2).The reaction was carried out with an initial step at 95 �C for10 min, followed by 40 cycles of amplification (15 s at 95 �C and1 min at 60 �C) and a final extension step (1 min at 95 �C and1 min at 55 �C) in a CFX96 Real-Time PCR detection system (Bio-Rad). Data were analyzed following the guidelines of the Real-TimePCR Applications Guide (BioRad).

2.6. Production of recombinant proteins, pull-down and Western blot

Glutathione-S-transferase (GST) and GST-Tat86 proteins wereproduced as described [15]. Briefly, bacteria were pre-grown over-night at 37 �C in Terrific Broth supplemented with 50 lg/ml ampi-cillin, diluted 15-fold in this same medium, and grown at 30 �Cuntil OD590 = 6.0. Protein production was induced with 1 mM iso-propyl b-D-1-thiogalactopyranoside (IPTG) for 2 h at 30 �C. Bacteriawere sonicated in 100 mM NaCl, 50 mM Tris–HCl, 5% glycerol,2 mM DL-dithiothreitol (DTT). Bacterial extracts were incubatedat 4 �C overnight with glutathione-Sepharose resin (GE-Healthcare,Chalfont St. Giles, UK), followed by 4 washes in lysis buffer[150 mM NaCl, 50 mM Tris pH 7.4, 1% Nonidet P40, 1 mM phenylmethyl sulfonyl fluoride (PMSF), and protease inhibitor cocktail(Sigma–Aldrich)]. Gp120s were produced by transient transfectionof 293T cells with 5 lg of the corresponding plasmids using aMammalian Transfection Kit (Promega, Milan, Italy); after 72 h,deriving cell lysates were obtained by extraction in lysis buffer.For the pull-down assays, protein extracts (4 mg) were pre-clearedon GST-resin for 1 h at 4 �C. Unbound proteins were incubated withGST-Tat86-resin for 1 h at 4 �C, washed 3 times with lysis bufferand once with phosphate-buffered saline (PBS). Bound proteinswere separated by SDS–PAGE, transferred onto polyvinylidenefluoride (PVDF) membranes (Millipore, Bedford, MA), and deco-rated with standard procedures. Band intensity was evaluatedusing Quantity One 4.6 Software (Biorad, Milan, Italy).

2.7. Statistical analysis

Two-tailed Student’s t-test was used to evaluate the differenceswithin experimental points (Prism 5, GraphPad Software, La Jolla,CA). Symbols in the graphs indicate the degree of statistical signif-icance: ⁄, § = P < 0.05, ⁄⁄, §§ = P < 0.01, ⁄⁄⁄, §§§ = P < 0.001.

3. Results

We previously described a peptide based on a consensus amonggp120 sequences from several HIV-1 strains, namely CT319 (se-quence: CSFNITTEIRDKVKK). CT319 interferes with gp120/Tatinteraction and inhibits Tat-driven HIV-1 entry at a concentrationas low as 1 nM. We identified two functionally active sites corre-sponding to the N-(CT303, sequence: CSFNIT) and C-terminal(CT304, sequence: RDKVKK) portions of CT319. In binding assaysperformed in vitro, gp120/Tat displacement was not increased byadding concentrations higher than 1 mM for CT303, while CT304showed a complete dose–response effect with an IC50 = 2 lM [15].

3.1. Molecular dissection of active sites in a gp120-mimicking peptide:the N-terminal cysteine and the C-terminal basic stretch are necessaryfor HIV-1-blocking activity.

Although the effect of CT303 in the displacement of the gp120/Tat interaction was weak, the great (�3-log) difference in the effi-

cacy of CT319 compared to CT304 suggested an involvement of theN-terminal portion in the described mechanism. In a first aspect,therefore, this study is based on new peptides in which the N-ter-minal cysteine of CT319 was either deleted (S14K) or replaced bymethylcysteine (Cm15K), serine (S15K), threonine (T15K), ormethionine (M15K). A variant with a supplemental cysteine atthe C-terminal (C16C) was also produced, to trigger cyclizationby the formation of an intramolecular disulfide bond. By mass-spectrometry, we further observed that CT319, when maintainedin non-reducing conditions for several days, spontaneously con-verts into a dimer (DIM), possibly by formation of intermoleculardisulfide bonds (data not shown). We tested the ability of all theseCT319 variants to block cell entry of viral particles that reproducecomplete HIV-1 virions while being replication-defective (Tat+/gp120-LVs). In this system, the first steps of HIV-1 infection areevaluated by counting the numbers of green fluorescent protein(GFP) positive (i.e., LV-transduced) cells [15,27,28]. C8166 lym-pho-monocytes were infected with Tat+/gp120-LVs in the presenceof each synthetic peptide (100 nM). GFP-expressing cells werecounted after 72 h, revealing that only Cm15K and C16C retain asignificant inhibitory effect, while all the other variants of CT319are ineffective (Fig. 1A). We further dissected the role of these pep-tides in Tat-mediated cell entry by using a Tat-defective LV system(Tat�/gp120-LVs). C8166 cells were pretreated with medium onlyor with recombinant Tat as described [15], before being infectedwith Tat�/gp120-LVs in the presence of each synthetic peptide(100 nM). Viral entry was evaluated by Real-Time PCR quantifica-tion of intracellular HIV-1 transcripts after 2 h. As expected, noneof the peptides inhibited, and one (M15K) even enhanced viral en-try in the absence of Tat. When the complete molecular circuit wasrestored by the addition of recombinant Tat, a significant inhibitionin viral entry was again achieved only by incubation with Cm15Kand C16C (Fig. 1B).

We next moved to the investigation of the C-terminal portion ofCT319, where the presence of a positively charged (basic) stretch isin support for an involvement in electrostatic protein–proteininteractions. We obtained basic-to-alanine derivatives (n = 6) ofthe CT304 sequence, and we evaluated the ability of these peptidesto interfere with virus entry. C8166 cells were incubated with Tat+/gp120-LVs in the presence of each synthetic peptide (100 lM), andthe numbers of GFP-positive cells were counted after 72 h. In theseassays, only RDAVKA retained a weak inhibitory activity (�10%),while all the other mutants either had no effect (RDKVKA, RDAVKK,RDAVAK), or even induced an opposite outcome (RDKVAK,RDKVAA) (Fig. 1C). These same peptides (100 lM) were challengedin the cell entry assay with Tat�/gp120-LVs, after pretreatmentwith medium only or with recombinant Tat. In this case, theReal-Time PCR evaluation of intracellular HIV-1 DNA showed thatnone of the peptides was capable of inhibiting viral entry; instead,some of them enhanced this process, either in the absence(RDAVKK, RDAVAK) or in the presence (RDKVAA) of Tat (Fig. 1D).

We also investigated the contribution, if any, of strain-relatedvariants of gp120 proteins in the inhibition of LV entry by the C-terminal basic stretch of CT319. First, we obtained synthetic vari-ants (n = 11) of CT304 that mimic gp120 sequences from differentHIV-1 strains (http://www.hiv.lanl.gov/), and we challenged thesepeptides (100 lM) in the described entry assays with Tat-deficientLVs, either in untreated or Tat-pretreated C8166 cells. Second, inaddition to the HXB2 gp120-pesudotyped LVs (Tat�/gp120-LVs),we also prepared viral particles covered by a R5-tropic envelope,namely the CF162 env (R5-Tat�/gp120-LVs), that were tested insimilar entry experiments. Entry was evaluated both by cytofluori-metric analysis of GFP-positive cells (Fig. 2A) and by Real-Time PCR(Fig. 2B, X4-tropic viruses; Fig. 2C. R5-tropic viruses). In the case ofX4-tropic LVs, as expected none of the peptides exerted an inhibi-tory effect on the entry of Tat-defective gp120-LVs; when cells

Fig. 1. Molecular dissection of active sites in a HIV-1-blocking peptide: both the N- and C-terminal of CT319 contribute to LV entry inhibition. C8166 cells were infected withHIV-1-derived virions in the presence of the indicated CT319 (CSFNITTEIRDKVKK, 100 nM) (A and B) and CT304 (RDKVKK, 100 lM) (C and D) variants. Viral entry wasevaluated by cytofluorimetric analysis of GFP-expressing cells 72 h after infection (A and C) and by Real-Time PCR quantification of HIV-1 intracellular DNA copy numbers 2 hafter infection (B and D). Statistical significance was calculated by a two-tailed Student’s t-test and is expressed as ⁄ = P < 0.05, ⁄⁄ = P < 0.01 and ⁄⁄⁄ = P < 0.001 for theinhibition, and § = P < 0.05, §§ = P < 0.01 and §§§ = P < 0.001 for the increase in virus entry into C8166 cells versus control infections. Values are expressed as percent variationover control infections, and indicate mean ± S.D. of 3 experiments performed in triplicate.

2946 S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951

were pretreated with Tat, entry was generally inhibited, up to�20% and �40% in the two assays, respectively, (Fig. 2A and B).For the R5-tropic LVs, the blocking effect of all the peptide variantsin the presence of Tat was retained, leading to up to �50% inhibi-tion in the Real-Time PCR evaluation. In some instances (i.e.,RDKKKK, RDKKKQ, RDKKKV, RDKKQK, RDKQQK, Fig. 2C) we ob-served a �20% inhibition of viral entry also in the absence of Tat,suggesting that this region of the env protein might participatein other molecular interactions/mechanisms.

Together, these data demonstrate that both the N- and C-termi-nal portions of CT319 are involved in the inhibition of Tat-drivencell entry of HIV-1-derived virions, and that this mechanism isstrain-independent.

3.2. The V1/V2 loop of the gp120 protein is indispensable for Tatbinding and virus entry

CT319 mimics a region spanning the V1/V2 loop and corre-sponding to aa 157–171 of the HXB2 gp120. To investigate the roleof this sequence in the contest of the whole protein, we designeddeletion mutants lacking either the V1 (DV1-gp120) or the V2(DV2-gp120) portions, or the entire V1/V2 (DV1/V2-gp120) loop(Fig. 3A). The corresponding pEnvHXB plasmids were transfectedinto 293T cells, in the same conditions used for producing LV par-ticles. In this way, a correctly processed gp120 protein is present atthe cell surface [27] and can be extracted in its native conforma-tion. Cell lysates were incubated with either GST (control) or

GST-Tat86 (target) [15,32], and separated by SDS–PAGE. Tat-boundgp120 variants were quantified by densitometric analysis of theprotein bands immunostained with an antibody against the C4 re-gion of gp120 (aa 427–436). This assay showed that Tat binding toboth DV1- and DV2-gp120 was as low as �30%, compared to wild-type gp120. Remarkably, DV1/V2-gp120 was almost completelyunable to interact with Tat (Fig. 3B). We also evaluated the capabil-ity of LVs pseudotyped with the same deletion mutants of gp120 toenter into host cells, observing that none of them is capable to en-ter into C8166 cells, both in basal conditions and in the presence ofTat (data not shown).

These findings demonstrate that the V1/V2 loop of gp120 is theregion responsible for specific binding to Tat; however, this regionis involved also in other aspects of cell entry, for which it is non-dispensable.

3.3. The basic stretch in gp120-V1/V2 loop modulates Tat affinity andTat-dependent virus entry

To dissect the molecular interactions involving the V1/V2 loopof gp120, we focused our attention on the CT304-mimicked region,corresponding to aa 166–171 of HXB2 gp120. We transposed allthe combinations of basic-to-alanine mutations into the coding se-quence of gp120 (Fig. 3C), and we performed pull-down assays asdescribed above. A densitometric analysis of GST-Tat86-boundgp120 proteins revealed that 4 mutants, namely K168A, K168A/K171A, Q170A/K171A, and K168A/Q170A/K171A bind Tat with

Fig. 2. Molecular dissection of active sites in a HIV-1-blocking peptide: the inhibitory effect of the basic stretch is not dependent on the viral strain. CT304-like peptides(100 lM) corresponding to gp120 proteins of different HIV-1 strains were challenged in viral entry assays with X4-tropic (A and B) and R5-tropic (C) Tat�/gp120-LVs, inC8166 cells either untreated or Tat-pretreated. Entry was evaluated by cytofluorimetric analysis of GFP-positive cells after 72 h (A) or by Real-Time PCR after 2 h (B and C).Statistical significance was calculated by a two-tailed Student’s t-test and is expressed as described in Fig. 1. Experimental points are mean ± S.D. of 2 experiments performedin triplicate.

S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951 2947

higher—while none with lower—affinity compared to wild-typegp120 (Fig. 3D). We also investigated the capability of LVs pseudo-typed with basic-to-alanine mutants of gp120 to enter into hostcells. For this purpose, we produced both Tat+/gp120-LVs and Tat�/gp120-LVs pseudotyped with each point mutant of gp120. C8166cells were infected with the deriving virions and viral entry wasevaluated by cytofluorimetric analysis of GFP-positive cells

(Fig. 4A and C) and by real-time quantification of intracellularHIV-1 DNA copy numbers (Fig. 4B and D).

In a first set of experiments, we tested the various Tat+/gp120-LVs, observing that none of the mutated LVs has impairedinfectivity; on the contrary, some mutations confer the cognatevirions a weakly enhanced capability of cell entry, which was sta-tistically significant for K171A (both evaluations), K168A/Q170A

Fig. 3. The V1/V2 loop of gp120 is involved in Tat binding and affinity. The indicated deletions (A) and point mutations (C) of the gp120 coding sequence were introduced inpEnvHXB. The deriving plasmids were transfected into 293T cells in the same conditions used for LV production, and all the gp120 protein variants were membrane-extractedto be challenged in pull-down assays with purified GST-Tat86 (B and D). For simplicity of visualization, in D the gp120 mutants are indicated with their corresponding aminoacid sequence (alanine substitutions are underlined), that is, wild-type: RGKVQK; K168A: RGAVQK; Q170A: RGKVAK; K171A: RGKVKA; K168A/Q170A: RGAVAK; K168A/K171A: RGAVKA; Q170A/K171A: RGKVAA; K168A/Q170A/K171A: RGAVAA. The amounts of Tat-bound (pulled-down) gp120 variants, quantified by densitometric analysis,were normalized on their total amounts in cell lysates, and are expressed as fold change over the wild-type protein. Statistical significance was calculated by a two-tailedStudent’s t-test and is expressed as described in Fig. 1. Values indicate mean ± S.D. of 3 experiments.

2948 S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951

(cytofluorimetry), K168A/K171A, Q170A/K171A and K168A/Q170A/K171A (Real-Time PCR) (Fig. 4A and B). Conversely, whenC8166 cells were infected with Tat�/gp120-LVs, the mutations re-sulted in either a decreased (K168A, both evaluations; K168A/K171A, Q170A/K171A, K168A/Q170A/K171A, Real-Time PCR) oran increased (Q170A, Real-Time PCR; K171A, K168A/Q170A bothevaluations) viral entry into host cells (Fig. 4C and D). To isolateTat-specific effects, we subtracted the variations in viral entry ofTat-deficient virions (Tat�/gp120-LVs) from that of the completeHIV-1 particles (Tat+/gp120-LVs) (Fig. 4E and F). This analysis re-vealed that, in the presence of Tat, entry is increased when LVsare pseudotyped with 4 mutant gp120 proteins (K168A, Q170A/K171A, K168A/K171A, K168A/Q170A/K171A, both evaluations),while it is decreased only when they are pseudotyped with theK168A/Q170A mutant protein (both evaluations).

These findings demonstrate that different mechanisms, in-cluded the affinity of gp120 for Tat (compare Fig. 4E and F withFig. 3D), account for Tat- and gp120-mediated virus entry into hostcells.

4. Discussions

We here dissect the role of the gp120 V1/V2 loop in Tat bindingand in the modulation of virus entry. This analysis will be the basisfor a rational evolution of the peptide entry inhibitors that we pre-viously described [15].

The first part of our work is focused on the characterization ofCT319, a peptide mimicking a portion of the gp120 V1/V2 loop.We demonstrate that the N-terminal cysteine is indispensable for

the inhibition of Tat-driven virus entry: only the variants in whichthis residue is preserved retain an inhibitory effect (Fig. 1A and B).In the native HXB2 protein, the cysteine in position 157 is involvedin a disulfide bond thus representing a structural component of theV1 and V2 loops; not surprisingly, the peptide with the highestefficacy in counteracting Tat-mediated infectivity is C16C, the cy-clized version of CT319. We next provide evidence that also theC-terminal basic stretch of this peptide is pivotal for the inhibitionof Tat-driven virus entry: none of the basic-to-alanine variants hasan activity comparable to that of CT304 (Fig. 1C and D). In agree-ment with our previous data [15], we also confirm that the inhib-itory effect of CT304 (and of similar peptides corresponding to thissame region in different gp120s) is independent from the viralstrain, being comparable in assays with both X4- and R5-tropicLVs (Fig. 2). Such peptides or their small-molecule derivatives,therefore, could be evolved in rationally designed, broadly-effec-tive anti-HIV-1 drugs. Intriguingly, peptides from both panels havean agonistic effect either in the presence (S14K, Fig. 1B, RDKVAKand RDKVAA, Fig. 1C and D) or in the absence of Tat (M15K,Fig. 1B, RDAVKK and RDAVAK, Fig. 1D). In both cases, such an out-come may be due to several factors: mutant peptides could induceconformational changes in gp120, interfere with other interactionsin which gp120 is involved, and/or produce off-target effects (com-mon when short peptides are used at relatively high dose).

Further evidence for the importance of the gp120 V1/V2 loop inTat-related HIV-1 infectivity comes from our studies with purifiedproteins and complete virions. We demonstrate that Tat bindingrequires both V1 and V2 loops: indeed, DV1- and DV2-gp120 exhi-bit a binding ability as low as �30% compared to wild-type gp120,

Fig. 4. The basic residues in the V1/V2 loop of gp120 are involved in Tat-mediated virus entry. pEnvHXB plasmids carrying point mutations in the gp120 coding sequencewere used to prepare the cognate LVs. C8166 cells were infected with all the variants of Tat+/gp120-LVs (A and B) and Tat�/gp120-LVs (C and D), and cell entry was evaluatedby cytofluorimetric analysis (A and C) and by Real-Time PCR (B and D). To discriminate Tat-specific effects, the percent variation in cell entry of Tat�/gp120-LVs wassubtracted from that of the corresponding Tat+/gp120-LVs (E and F). For simplicity of visualization, the gp120 mutants are indicated with their corresponding amino acidsequence (alanine substitutions are underlined) as described in Fig. 2. Statistical significance was calculated by a two-tailed Student’s t-test and is expressed as described inFig. 1. Values are expressed as percent variation over infections with wild-type LVs, and indicate mean ± S.D. of 4 experiments performed in triplicate.

S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951 2949

while the fully-deleted DV1/V2-gp120 protein is almost com-pletely unable to interact with Tat (Fig. 3B). In our experiments,the deletion of either part of the gp120 V1/V2 loop led to abroga-tion of LV entry into host cells, as detected by counting the numberof GFP-positive cells (data not shown). Such a result goes beyond

the sole involvement of Tat, and is consistent with previous find-ings indicating a role for the gp120 V1/V2 loop in HIV-1 infectivity.For example, Wyatt et al., show that a deletion mutant lacking theV1 portion of the loop (D136–151) is defective in functions relatedto infection, example, the ability of inducing cell syncytia (14% of

2950 S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951

the wild-type) and of complementing virus entry into Jurkat cells(34%). The latter function is retained at comparable levels (30%)by the fully-deleted variant (D128–194) [33]. This residual infec-tivity of gp120 mutants, in partial discrepancy with our results,may be related to the experimental design, that is the detectionof a reporter activity (an amplification system that can discrimi-nate among little differences) versus our cytofluorimetric evalua-tion of LV-transduced cells. However, other findings are insupport for a dependence on the viral strain and/or the cell typeinvestigated: Cao et al., show that a HIV-1 mutant lacking the V1and V2 variable loops of gp120 replicates in Jurkat lymphocyteswith only a weak delay; they also report the rapid emergence ofrevertants containing a few amino acid changes to counteract theeffect of the deletion [34]. Stamatatos et al., observe that V1/V2loop-deleted gp120 proteins support replication of SF162 virus intoperipheral blood mononuclear cells but not in macrophages [35].All these complex interactions still have to be fully elucidated;however, for the first time we here demonstrate that the deletionof the V1/V2 loop impacts on Tat binding to gp120, and thereforeat least part of these effects may be dependent on this molecularcircuit (e.g. the amount of membrane-bound Tat and its availabil-ity, the presence of particular cell surface proteoglycans and/or ofpeculiar coreceptors, and so on).

A molecular explanation for the role of the V1/V2 loop in HIV-1infectivity has been partially provided in previous studies. Molec-ular determinants in the V1/V2 loop that influence viral infectivityare those involved in heparin binding: a study from Mbemba et al.,demonstrates that gp120 elicits multimolecular complexes, andthat most of these interactions are abolished by heparin [36].Accordingly, heparan sulfate-interacting sites have been identifiedin the coreceptor binding site of gp120 [37] and, recently, four hep-arin-binding domains have been mapped, including lysine 168 inthe V2 region [38]. A mutation of this residue could lead to the on-set of a different interaction network, in which heparan sulfate orother surface proteoglycans are not included as gp120 partners.Such a reorganization could be responsible for changes in virusinfectivity as a possible consequence of different exposure/avail-ability of binding sites. Consistently, our findings demonstrate thatthree out of four gp120 mutants having the lysine 168 convertedinto an alanine (i.e. K168A, K168A/K171A and K168A/Q170A/K171A) have an increased capability to bind recombinant Tat(Fig. 3D).

In an attempt to discriminate between Tat-dependent and Tat-independent effects, we compared the entry of LVs pseudotypedwith the described basic-to-alanine mutant gp120 proteins, eitherin the presence of complete HIV-1 virions or in Tat-defective con-ditions (Fig. 4). These assays demonstrated that the four mutantswith greater affinity to recombinant Tat (i.e. K168A, K168A/K171A, Q170A/K171A and K168A/Q170A/K171A, see Fig. 3D forcomparison) also have an enhanced capability to enter into hostcells, compared to viral particles pseudotyped with the wild-typegp120 protein. The fact that both Tat and gp120 bind to the hepa-ran sulfates may represent one explanation of why these changesin the basic amino acids of gp120 lead to an altered Tat-relatedinfectivity: this can possibly be related to an unbalance of their re-ciprocal molecular interactions. Most remarkably, one of the gp120mutants, that is K168A/Q170A, although retaining an affinity forTat similar to that of the wild-type protein, appears to be pivotalin Tat-mediated cell entry of HIV-1 virions (Fig. 4E and F), againsuggesting that features other than Tat/gp120 affinity, such as ki-netic of interaction, induction of conformational changes, and/orrecruitment of molecular partners are relevant in this context.

The group of Dr. Ensoli, the only other having published a directinteraction between Tat and gp120, has reported results differingfrom ours. In a first set of experiments, they show that immobi-lized Tat binds both retroviral (VSV-G pseudotyped) and lentiviral

(gp120-pesudotyped) virions, thus enhancing viral transduction, asassessed by GFP-expressing cells in cytofluorimetric analysis (sim-ilar to our system) or by p24 quantification using a single-cyclereplication-defective virus [39]. This work shows that Tat-assistedinfection does not require binding to a specific envelope: due to thehigh amounts of Tat used (micromolar range), the observed effectshould therefore be attributed to an electrostatic interaction of Tatwith the viral surface, as previously described for cell entry ofadenoviral vectors [40]. This same feature explains why Tat is effi-cient only when immobilized and not when in solution: in our pre-vious work, we demonstrated that increasing the amount ofsoluble Tat reverts its HIV-1-specific promotion of virus entry, pos-sibly by interfering with other cellular mechanisms [15]. They fur-ther provide some evidence that the cysteine-rich region of Tat isrequired for binding to virus particles. In contrast with their firstset of data, in a recent work they show that Tat and trimericgp120 interact specifically, and that this interaction is responsiblefor the promotion of viral transmission in dendritic cells [41]. Theyconfirm the involvement of the cysteine-rich domain of Tat, andthey state that the V3 loop of gp120 is involved in this binding.Although we do not exclude an interaction of Tat cysteine regionwith gp120 V3 loop (we have not investigated this region of theenvelope protein), these results need further discussion. First, theyfound, by Surface Plasmon Resonance analysis, Kd values for Tat/monomeric DV2 gp120 about 10-fold higher (9.1 � 10�8) thanthose found by our group for Tat/gp120 (8.1 ± 0.3 � 10�9), an indi-cation of lower affinity. Second, again the high amounts of Tat usedmight lead to a switch from the high affinity Tat C-terminal/gp120V2 loop interaction to the lower affinity Tat cysteine domain/gp120 V3 loop interaction. Third, they purify recombinant Tat byheparin-affinity columns, and therefore heparin is present (pre-sumably in high amounts) in all the incubations: as previously dis-cussed, the presence of heparin can unbalance interactions thatinvolve the V2 loop of gp120. On the contrary, our recombinantTat is purified either by ion-exchange columns or GST-taggedand purified by glutathione-Sepharose beads; therefore no heparinor similar contaminants are present in our preparations.

Finally, a brief comment on the methodology that we adoptedin this report is due. To evaluate viral entry, we applied both theLV transduction assay (counting of GFP-positive cells) 72 h afterinfection, and the quantification of intracellular viral DNA (Real-Time PCR) 2 h after infection. Despite being completely differentassays, they gave comparable results in all the experiments (com-pare Fig. 1A and B; Fig. 1C and D; Fig. 4A and B; Fig. 4C and D), withthe DNA quantification being more sensitive and thus leading tohigher variations and to a greater number of statistically significantexperimental points. This is, in our knowledge, the first comparisonof two methods deriving from different investigation fields, that is,gene therapy (LV-transduction) and HIV-1 diagnostic (cDNA copynumber); our results demonstrate that these methods are inter-changeable and could be used indifferently, based on the experi-mental setting being investigated.

In summary, our findings confirm that the V1/V2 loop of gp120is involved in complex interactions as well as in pivotal conforma-tional changes, part of which are related to its interaction with Tat.Our results point to a complex and modulated system, in whichmolecules both on the viral and on the cellular side exploit differ-ent functions and interact with different partners. A completeexplanation of this picture will aid the development of ‘‘smart bul-lets’’ for the therapy of HIV-1.

Authors’ contributions

S.C. performed the experiments, acquired, analyzed and inter-preted the data; M.S. performed the experiments; F.B. critically re-vised the manuscript and contributed to its intellectual content;

S. Cardaci et al. / FEBS Letters 587 (2013) 2943–2951 2951

S.M. designed the experiments and drafted the manuscript. All theauthors approved the final version to be submitted.

Funding

This work was supported by Istituto Superiore di Sanità (to F.B.),Associazione Italiana per la Ricerca sul Cancro (AIRC, to F.B.), AIRC-My First AIRC Grant (AIRC-MFAG, to S.M.), Italian Ministry ofHealth (Ricerca Oncologica 2006, Ricerca Finalizzata 2006–2007to F.B.), Regione Piemonte (Ricerca Sanitaria Finalizzata 2006–2007; Ricerca Industriale e Competitiva 2006, Grant PRESTO; Ric-erca Tecnologie Convergenti 2007, Grant PHOENICS; PiattaformeTecnologiche per le Biotecnologie, Grant DRUIDI to FB), FondazioneCassa di Risparmio di Torino (CRT, to F.B.), and Ministry of Univer-sity and Research (MiUR, PRIN 2007 to F.B.). S.C. is recipient of a‘‘Lagrange’’ fellowship (CRT), M.S. is recipient of a Ph. D. fellowshipsponsored by AIRC-MFAG. These sponsors had no role in study de-sign; in the collection, analysis and interpretation of data; in thewriting of the report; and in the decision to submit the article forpublication.

Acknowledgment

We thank A. Di Nicola and A.K. Lee for technical assistance.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.febslet.2013.07.039.

References

[1] Dalgleish, A.G., Beverley, P.C., Clapham, P.R., Crawford, D.H., Greaves, M.F. andWeiss, R.A. (1984) The CD4 (T4) antigen is an essential component of thereceptor for the AIDS retrovirus. Nature 312, 763–767.

[2] Klatzmann, D., Champagne, E., Chamaret, S., Gruest, J., Guetard, D., Hercend, T.,Gluckman, J.C. and Montagnier, L. (1984) T-lymphocyte T4 molecule behavesas the receptor for human retrovirus LAV. Nature 312, 767–768.

[3] Maddon, P.J., Dalgleish, A.G., McDougal, J.S., Clapham, P.R., Weiss, R.A. andAxel, R. (1986) The T4 gene encodes the AIDS virus receptor and is expressedin the immune system and the brain. Cell 47, 333–348.

[4] Deng, H. et al. (1996) Identification of a major co-receptor for primary isolatesof HIV-1. Nature 381, 661–666.

[5] Dragic, T. et al. (1996) HIV-1 entry into CD4+ cells is mediated by thechemokine receptor CC-CKR-5. Nature 381, 667–673.

[6] Feng, Y., Broder, C.C., Kennedy, P.E. and Berger, E.A. (1996) HIV-1 entrycofactor: functional cDNA cloning of a seven-transmembrane, G protein-coupled receptor. Science 272, 872–877.

[7] Freed, E.O., Myers, D.J. and Risser, R. (1990) Characterization of the fusiondomain of the human immunodeficiency virus type 1 envelope glycoproteingp41. Proc. Natl. Acad. Sci. USA 87, 4650–4654.

[8] Gonzalez-Scarano, F., Waxham, M.N., Ross, A.M. and Hoxie, J.A. (1987)Sequence similarities between human immunodeficiency virus gp41 andparamyxovirus fusion proteins. AIDS Res. Hum. Retroviruses 3, 245–252.

[9] Matthews, T., Salgo, M., Greenberg, M., Chung, J., DeMasi, R. and Bolognesi, D.(2004) Enfuvirtide: the first therapy to inhibit the entry of HIV-1 into host CD4lymphocytes. Nat. Rev. Drug Disc. 3, 215–225.

[10] Fatkenheuer, G. et al. (2008) Subgroup analyses of maraviroc in previouslytreated R5 HIV-1 infection. N. Engl. J. Med. 359, 1442–1455.

[11] Nichols, W.G., Steel, H.M., Bonny, T., Adkison, K., Curtis, L., Millard, J., Kabeya,K. and Clumeck, N. (2008) Hepatotoxicity observed in clinical trials ofaplaviroc (GW873140). Antimicrob. Agents Chemother. 52, 858–865.

[12] Lalezari, J. et al. (2011) Safety, efficacy, and pharmacokinetics of TBR-652, aCCR5/CCR2 antagonist, in HIV-1-infected, treatment-experienced, CCR5antagonist-naive subjects. J. Acquir. Immune Defic. Syndr. 57, 118–125.

[13] Nettles, R.E. et al. (2012) Pharmacodynamics, safety, and pharmacokinetics ofBMS-663068, an oral HIV-1 attachment inhibitor in HIV-1-infected subjects. J.Infect. Dis. 206, 1002–1011.

[14] Bruno, C.J. and Jacobson, J.M. (2010) Ibalizumab: an anti-CD4 monoclonalantibody for the treatment of HIV-1 infection. J Antimicrob. Chemother. 65,1839–1841.

[15] Marchiò, S. et al. (2005) Cell surface-associated Tat modulates HIV-1 infectionand spreading through a specific interaction with gp120 viral envelopeprotein. Blood 105, 2802–2811.

[16] Arya, S.K., Guo, C., Josephs, S.F. and Wong-Staal, F. (1985) trans-Activator geneof human T-lymphotropic virus type III (HTLV-III). Science 229, 69–73.

[17] Chang, H.K., Gallo, R.C. and Ensoli, B. (1995) Regulation of cellular geneexpression and function by the human immunodeficiency virus type 1 TatProtein. J. Biomed. Sci. 2, 189–202.

[18] Albini, A. et al. (1998) HIV-1 Tat protein mimicry of chemokines. Proc. Natl.Acad. Sci. USA 95, 13153–13158.

[19] Ensoli, B., Buonaguro, L., Barillari, G., Fiorelli, V., Gendelman, R., Morgan, R.A.,Wingfield, P. and Gallo, R.C. (1993) Release, uptake, and effects of extracellularhuman immunodeficiency virus type 1 Tat protein on cell growth and viraltransactivation. J. Virol. 67, 277–287.

[20] Albini, A., Barillari, G., Benelli, R., Gallo, R.C. and Ensoli, B. (1995) Angiogenicproperties of human immunodeficiency virus type 1 Tat protein. Proc. Natl.Acad. Sci. USA 92, 4838–4842.

[21] Chang, H.C., Samaniego, F., Nair, B.C., Buonaguro, L. and Ensoli, B. (1997) HIV-1Tat protein exits from cells via a leaderless secretory pathway and binds toextracellular matrix-associated heparan sulfate proteoglycans through itsbasic region. AIDS 11, 1421–1431.

[22] Tyagi, M., Rusnati, M., Presta, M. and Giacca, M. (2001) Internalization of HIV-1Tat requires cell surface heparan sulfate proteoglycans. J. Biol. Chem. 276,3254–3261.

[23] Robey, W.G., Safai, B., Oroszlan, S., Arthur, L.O., Gonda, M.A., Gallo, R.C. andFischinger, P.J. (1985) Characterization of envelope and core structural geneproducts of HTLV-III with sera from AIDS patients. Science 228, 593–595.

[24] Starcich, B.R. et al. (1986) Identification and characterization of conserved andvariable regions in the envelope gene of HTLV-III/LAV, the retrovirus of AIDS.Cell 45, 637–648.

[25] Leonard, C.K., Spellman, M.W., Riddle, L., Harris, R.J., Thomas, J.N. and Gregory,T.J. (1990) Assignment of intrachain disulfide bonds and characterization ofpotential glycosylation sites of the type 1 recombinant humanimmunodeficiency virus envelope glycoprotein (gp120) expressed inChinese hamster ovary cells. J. Biol. Chem. 265, 10373–10382.

[26] Cheng-Mayer, C., Liu, R., Landau, N.R. and Stamatatos, L. (1997) Macrophagetropism of human immunodeficiency virus type 1 and utilization of the CC-CKR5 coreceptor. J. Virol. 71, 1657–1661.

[27] Kowalski, M. et al. (1987) Functional regions of the envelope glycoprotein ofhuman immunodeficiency virus type 1. Science 237, 1351–1355.

[28] Naldini, L., Blomer, U., Gallay, P., Ory, D., Mulligan, R., Gage, F.H., Verma, I.M.and Trono, D. (1996) In vivo gene delivery and stable transduction ofnondividing cells by a lentiviral vector. Science 272, 263–267.

[29] Dull, T., Zufferey, R., Kelly, M., Mandel, R.J., Nguyen, M., Trono, D. and Naldini,L. (1998) A third-generation lentivirus vector with a conditional packagingsystem. J. Virol. 72, 8463–8471.

[30] Malnati, M.S., Scarlatti, G., Gatto, F., Salvatori, F., Cassina, G., Rutigliano, T.,Volpi, R. and Lusso, P. (2008) A universal real-time PCR assay for thequantification of group-M HIV-1 proviral load. Nat. Protoc. 3, 1240–1248.

[31] Weidner, J., Cassens, U., Gohde, W., Sibrowski, W., Odaibo, G., Olaleye, D.,Reichelt, D. and Greve, B. (2011) An improved PCR method for detection ofHIV-1 proviral DNA of a wide range of subtypes and recombinant formscirculating globally. J. Virol. Met. 172, 22–26.

[32] Mitola, S., Soldi, R., Zanon, I., Barra, L., Gutierrez, M.I., Berkhout, B., Giacca, M.and Bussolino, F. (2000) Identification of specific molecular structures ofhuman immunodeficiency virus type 1 Tat relevant for its biological effects onvascular endothelial cells. J. Virol. 74, 344–353.

[33] Wyatt, R., Moore, J., Accola, M., Desjardin, E., Robinson, J. and Sodroski, J.(1995) Involvement of the V1/V2 variable loop structure in the exposure ofhuman immunodeficiency virus type 1 gp120 epitopes induced by receptorbinding. J. Virol. 69, 5723–5733.

[34] Cao, J., Sullivan, N., Desjardin, E., Parolin, C., Robinson, J., Wyatt, R. andSodroski, J. (1997) Replication and neutralization of human immunodeficiencyvirus type 1 lacking the V1 and V2 variable loops of the gp120 envelopeglycoprotein. J. Virol. 71, 9808–9812.

[35] Stamatatos, L., Wiskerchen, M. and Cheng-Mayer, C. (1998) Effect of majordeletions in the V1 and V2 loops of a macrophage-tropic HIV type 1 isolate onviral envelope structure, cell entry, and replication. AIDS Res. Hum.Retroviruses 14, 1129–1139.

[36] Mbemba, E., Benjouad, A., Saffar, L. and Gattegno, L. (1999) Glycans andproteoglycans are involved in the interactions of human immunodeficiencyvirus type 1 envelope glycoprotein and of SDF-1alpha with membrane ligandsof CD4(+) CXCR4(+) cells. Virology 265, 354–364.

[37] Vives, R.R., Imberty, A., Sattentau, Q.J. and Lortat-Jacob, H. (2005) Heparansulfate targets the HIV-1 envelope glycoprotein gp120 coreceptor binding site.J. Biol. Chem. 280, 21353–21357.

[38] Crublet, E., Andrieu, J.P., Vives, R.R. and Lortat-Jacob, H. (2008) The HIV-1envelope glycoprotein gp120 features four heparan sulfate binding domains,including the co-receptor binding site. J. Biol. Chem. 283, 15193–15200.

[39] Nappi, F. et al. (2009) Immobilized HIV-1 Tat protein promotes gene transfervia a transactivation-independent mechanism which requires binding of Tatto viral particles. J. Gene Med. 11, 955–965.

[40] Gratton, J.P., Yu, J., Griffith, J.W., Babbitt, R.W., Scotland, R.S., Hickey, R.,Giordano, F.J. and Sessa, W.C. (2003) Cell-permeable peptides improve cellularuptake and therapeutic gene delivery of replication-deficient viruses in cellsand in vivo. Nat. Med. 9, 357–362.

[41] Monini, P. et al. (2012) HIV-1 tat promotes integrin-mediated HIVtransmission to dendritic cells by binding Env spikes and competesneutralization by anti-HIV antibodies. PLoS One 7, e48781.