Different from the HIV Fusion Inhibitor C34, the Anti-HIV DrugFuzeon (T-20) Inhibits HIV-1 Entry by Targeting Multiple Sites ingp41 and gp120*

Received for publication, September 28, 2004, and in revised form, January 4, 2005Published, JBC Papers in Press, January 7, 2005, DOI 10.1074/jbc.M411141200

Shuwen Liu‡, Hong Lu‡, Jinkui Niu‡, Yujia Xu§, Shuguang Wu¶, and Shibo Jiang‡�

From the ‡Lindsley F. Kimball Research Institute, New York Blood Center, New York, New York 10021, the §Departmentof Chemistry, Hunter College, City University of New York, New York, New York 10021, and the ¶Institute ofPharmaceutical Sciences, Southern Medical University, Guangzhou, Guangdong 510515, China

Fuzeon (also known as T-20 or enfuvirtide), one of theC-peptides derived from the HIV-1 envelope glycopro-tein transmembrane subunit gp41 C-terminal heptad re-peat (CHR) region, is the first member of a new class ofanti-HIV drugs known as HIV fusion inhibitors. It hasbeen widely believed that T-20 shares the same mecha-nism of action with C34, another C-peptide. The C34 isknown to compete with the CHR of gp41 to form a stable6-helix bundle (6-HB) with the gp41 N-terminal heptadrepeat (NHR) and prevent the formation of the fuso-genic gp41 core between viral gp41 NHR and CHR,thereby inhibiting fusion between viral and target cellmembranes. Here we present data to demonstrate that,contrary to this belief, T-20 cannot form stable 6-HBwith N-peptides derived from the NHR region, nor can itinhibit the 6-HB formation of the fusogenic core. In-stead, it may interact with N-peptides to form unstableor insoluble complexes. Our data suggest that T-20 has adifferent mechanism of action from C34. The interactionof T-20 with viral NHR region alone may not prevent theformation of the fusion active gp41 core. We also dem-onstrate that the T-20-mediated anti-HIV activity can besignificantly abrogated by peptides derived from themembrane-spanning domain in gp41 and coreceptorbinding site in gp120. These new findings imply thatT-20 inhibits HIV-1 entry by targeting multiple sites ingp41 and gp120. Further elucidation of the mechanismof action of T-20 will provide new target(s) for develop-ment of novel HIV entry inhibitors.

Human immunodeficiency virus type 1 (HIV-1)1 envelopeglycoprotein (Env), a type I transmembrane protein, plays animportant role in the early stage of HIV entry. Its surface

subunit gp120 is responsible for virus binding to receptor andcoreceptors (1–4), and the transmembrane subunit gp41 medi-ates fusion of the virus with the target cell (5, 6). Like othertype I transmembrane proteins, gp41 consists of cytoplasm(CP), transmembrane (TM), and extracellular domains. Theextracellular domain (ectodomain) contains four major func-tional regions: fusion peptide (FP), N-terminal heptad repeat(NHR or HR1), C-terminal heptad repeat (CHR or HR2), and atryptophan-rich (TR) region.

Starting form the early 1990s, several peptides derived fromthe NHR and CHR regions of gp41, designated N- and C-pep-tides, respectively (Fig. 1A), were discovered to have the viralfusion inhibition activity. The first N-peptide fusion inhibitor,T-21, was identified by Wild et al. (7), which blocked HIV-1 fusionat micromolar concentrations. The first C-peptide fusion inhibi-tor, SJ-2176, was discovered by our group, which inhibited HIV-1replication at nanomolar levels (8, 9). Later, another C-peptidefusion inhibitor, T-20, was identified (10, 11) which overlapsSJ-2176 sequence and also has potent anti-HIV activity. Usingproteolytic dissection strategies, Lu et al. (12, 13) isolated severalpairs of protease-resistant N- and C-peptides from gp41 includ-ing N36 and C34, which was later found to form the stablefusogenic core. The peptide C34 is more potent in inhibitingHIV-1 fusion than both SJ-2176 and T-20. Interestingly, C34overlaps the entire sequence of SJ-2176 with only two additionalresidues at the N and C termini, respectively.

Extensive structural studies on the peptide pair N36 andC34 have revealed the structure of the fusogenic core gp41. TheN- and C-peptides, while stay unstructured on their own, havethe tendency to form a stable six-stranded �-helix bundle (6-HB) when mixed with a 1:1 molar ratio. The crystal structuresof the 6-HB, based on the work of three independent groups,show that three helices from the N-peptides (the N-helices)associate to form the central trimeric coiled-coil and threehelices from C-peptide region (the C-helices) pack obliquely inan anti-parallel configuration into the highly conserved hydro-phobic grooves on the surface of the central coiled-coil (14–16).In each of the grooves, there is a highly conserved hydrophobicdeep cavity, formed by the cavity-forming sequence (residues565–581) in the NHR region (Fig. 1), which is critical for viralfusion and stability of the 6-HB (17–19). The 6-HB is extremelythermostable (12, 13) and can be recognized by a conformation-specific monoclonal antibody (mAb), NC-1 (20, 21). In the 6-HBthe residues at the “e” and “g” positions in the �-helical wheelsof the N-helices can interact with those at the “a” and “d”positions in the C-helices during the process of 6-HB formation(5) (Fig. 1, B and C).

Based on the structure of the 6-HB, a model was proposed toelucidate the mechanism of gp41-mediated membrane fusion

* This work was supported by Grant AI46221 from the NationalInstitutes of Health (to S. J.) and Grants from the 863 Project Founda-tion of China (2001AA214201) and Natural Science Foundation ofChina (301400220) (to S. W.). The costs of publication of this articlewere defrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact.

� To whom correspondence should be addressed: The Lindsley F. KimballResearch Institute, NY Blood Center, New York, NY 10021. Tel.: 1-212-570-3058; Fax: 1-212-570-3099; E-mail: sjiang@nybloodcenter.org.

1 The abbreviations used are: HIV-1, human immunodeficiency virustype 1; Env, envelope glycoprotein; NHR, N-terminal heptad repeat;CHR, C-terminal heptad repeat; FP, fusion peptide; 6-HB, six-helicalbundle; FITC, fluorescein isothiocyanate; N-PAGE, native polyacryl-amide gel electrophoresis; FN-PAGE, fluorescence native polyacryl-amide gel electrophoresis; FLISA, fluorescence-linked immunosorbentassay; CD, circular dichroism; PBS, phosphate-buffered saline; mAb,monoclonal antibody; ELISA, enzyme-linked immunosorbent assay;FMOC, N-(9-fluorenyl)methoxycarbonyl.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 280, No. 12, Issue of March 25, pp. 11259–11273, 2005© 2005 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 11259

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

and the anti-HIV-1 activity of the C-peptides (6). After gp120binds to CD4 and a coreceptor (e.g. CCR5 or CXCR4), gp41changes its conformation to a prefusion (intermediate) state by

inserting its fusion peptide into the target cell membrane. Thenthe gp41 N- and C-helices associate to form the 6-HB bringingthe viral and target cell membranes into close proximity and

FIG. 1. HIV-1 gp41 molecule and its critical sequences in forming 6-HB fusion core. A, schematic diagram of the HIV-1 gp41 molecule. Theresidue numbers of each region correspond to their positions in gp160 of HIV-1HXB2. B, model of 6-HB formed by the N- and C-helices. The internaltrimer is formed through the interaction of the residues located at the a and d positions (in blue) in three N-helices and 6-HB is formed through theinteraction of the residues located at the e and g positions (in red) in the N-helices and the a and d positions (in green) in the C-helices. C, interactionbetween the gp41 N- and C-helices and the critical sequences in the NHR and CHR regions. The lines between the NHR and CHR regions indicate theinteraction between the residues located at e and g positions in NHR and a and d positions in CHR, respectively. Interaction between the sequencesspanning residues 530–540 and residues 666–673 is unknown since no crystal structure containing these sequences is available. The critical sequencesin NHR and CHR include: (i) GIV (residues 547–549, red) is a determinant of resistance to T-20 in NHR region (36), which is presented in the peptidesN36-F10, N36, and N46, but not in the peptide T-21; (ii) LLQLTVWGIKQLQARIL (residues 565–581, green) is the cavity-forming sequence in the NHRregion (14, 31); (iii) WMEWDREI (residues 628–635, orange) is the cavity binding sequence in the CHR region (14, 17); and (iv) WASLWNWF (residues666–673, pink) is a partial tryptophan-rich sequence (11). The N-peptide N36-F10, N36, and T-21 span the N-terminal, middle, and C-terminal portionsof the NHR region, respectively. N46 covers both the N-terminal and middle portions of the NHR region. N46, N36, and T-21, but not N36-F10, containthe entire cavity-forming sequence. The C-peptide C34 covers the major portion of the CHR region, including the cavity binding sequence. T-20 overlapsthe C-terminal portion of the CHR region and part of the tryptophan-rich sequence, but not the cavity binding sequence.

T-20 Targets Multiple Sites in gp120/gp4111260

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

result in the fusion of HIV-1 with the target cell. It was, then,proposed that C-peptide (e.g. C34) may form a heterogeneous6-HB with the NHR region of gp41, thus, block the formation offusion-active core of gp41 and inhibit the fusion between theviral and target cell membranes. The understanding of theaction of the fusion inhibitor has opened a new avenue toidentify antiviral peptides against other viruses with type Itransmembrane envelope glycoprotein, such as simian immu-nodeficiency viruses (SIV) (22), Sendai virus (23), feline immu-nodeficiency virus (FIV) (24), respiratory syncytial virus (RSV),measles virus (25), Ebola virus (26), Nipah and Hendra viruses(27), and most recently, the severe acute respiratory syndrome-associated coronavirus (SARS-CoV) (28–30).

T-20 (brand name: Fuzeon), developed by Trimeris, Inc. andHoffmann-La Roche, Ltd., as the first member of a new class ofanti-HIV drugs, the HIV fusion inhibitors, was recently li-censed by the United States Food and Drug Administration(FDA)2 for treatment of HIV-infected individuals, includingthose who failed to respond to prior antiretroviral therapy(31–33). Because the major sequence of T-20 is derived from theCHR region and partially overlaps C34 sequence (Fig. 1, A andC), it has been proposed that T-20 acts in the same way as C34to inhibit viral fusion by blocking the formation of the fusogeniccore of gp41 (6, 33–35).

However, new evidence on the fusion inhibitor T-20 raiseddoubts about the generality of the above model. Although par-tially overlapping with the C34 peptide, T-20 does not containthe cavity binding sequence (residues 628–635) (Fig. 1C),which is important for gp41-mediated membrane fusion andthe stability of the gp41 core conformation (17–19). There hasbeen no report on the direct interaction of T-20 with the NHRsequence to form 6-HB. T-20-induced resistant mutations inpatients who used T-20 are located not only in gp41, but also ingp120. HIV-1 isolates resistant to T-20 are sensitive to C34-containing peptides, such as T-649 (36). In addition, the viralsensitivity to T-20 is modulated by coreceptor specificity de-fined by the V3 loop of gp120 (37). These findings suggest themechanism of action of T-20 may be different from that of C34.

Here we present data to show that T-20, unlike C34, does notform the stable 6-HB with the N-peptides containing the entirecavity-forming sequence. Instead, it may associate with N-peptides to form unstable or insoluble complexes. In addition,T-20-mediated anti-HIV activity could be abrogated by pep-tides derived from the membrane-spanning domain in gp41and the coreceptor binding site in gp120. These suggest thatT-20 may target multiple sites in gp41 and gp120, a mechanismof action different from that of C34 and other C-peptides. Fur-ther elucidating the mechanism of action of T-20 will providenew target(s) for development of novel HIV entry inhibitors.

EXPERIMENTAL PROCEDURES

Peptides—Peptides T-20, C34, T-21, N36, N46, N36-F10, N36-SIM,N36-HA (a hemagglutinin epitope GGGYPYDVPDYAGPG was taggedat the C terminus of N36), T-20-ANAA, T-20-ANAA-octyl, C34-biotin(biotin was conjugated to the N terminus of C34), N36-F, C34-F, andT-20-F (fluorescein isothiocyanate (FITC) was added to the N termini ofN36, C34 and T-20, respectively) were synthesized by a standard solid-phase FMOC method at the MicroChemistry Laboratory, the New YorkBlood Center. The peptides were purified to homogeneity (�95% purity)by high-performance liquid chromatography (HPLC) and identified bylaser desorption mass spectrometry (PerSeptive Biosystems, Framing-ham, MA). 5-Helix and 5-Helix (D4) were kindly provided by Dr.Michael Root (Thomas Jefferson University, Philadelphia, PA).HIV-1IIIB chronically infected H9 (H9/HIV-1IIIB) cells, MT-2 cells, andHIV-1 MN Env (15-mer) peptides (a complete set) were obtained from

the National Institutes of Health AIDS Research and ReferenceReagent Program, contributed by Drs. Robert Gallo and DouglasRichman, and DAIDS, NIAID, respectively.

Native Polyacrylamide Gel Electrophoresis (N-PAGE)—N-PAGE wascarried out to determine the 6-HB formation between the N- and C-peptides as described previously (38). Briefly, an N-peptide (N36, T-21,or N36-F10) was incubated with a C-peptide (C34, T-20 or T-20 deriv-atives) (the final concentration of each peptide was 40 �M) at 37 °C for30 min. The sample was mixed with Tris-glycine native sample buffer(Invitrogen, Carlsbad, CA) at a ratio of 1:1 and was then loaded onto a10 � 1.0-cm precast Tris-glycine gels (18%, Invitrogen, Carlsbad, CA) at25 �l/per well. Gel electrophoresis was carried out with 125 V constantvoltage at room temperature for 2 h. The gel was then stained withCoomassie Blue and imaged with a FluorChem 8800 Imaging System

2 U. S. FDA, March 13, 2003. FDA approves first drug in new class ofHIV treatments for HIV-infected adults and children with advanceddisease. FDA News.

FIG. 2. Analysis of the secondary structures of N36, C34, T-20,and the mixtures of N36�C34 and N36�T-20 by CD spectroscopy.A, CD spectra for N36, C34, and their complex. B, temperature scan-ning at 222 nm for the N36�C34 complex. Inserted is the first derivativeof the curve against temperature, which was used to determine the Tmvalue. C, CD spectra for N36, T-20, and their complex. Final concen-tration of each peptide is 10 �M.

T-20 Targets Multiple Sites in gp120/gp41 11261

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

(Alpha Innotech Corp., San Leandro, CA).Fluorescence Native Polyacrylamide Gel Electrophoresis (FN-PAGE)—

FN-PAGE was performed under the same conditions using the samereagents as those for N-PAGE described above except that one of thetwo peptides was replaced by FITC-conjugated peptide (N36-F, C34-F,or T-20-F) (38). Immediately after electrophoresis, fluorescence bandsin the gel were imaged by the FluorChem 8800 Imaging System usinga transillumination UV light source with excitation wavelength at 302nm and a fluorescence filter with emission wavelength at 520 nm. Thegel was then stained with Coomassie Blue and imaged again with aFluorChem 8800 Imaging System.

Fluorescence-linked Immunosorbent Assay (FLISA)—C34-F wasused in FLISA for detection of the 6-HB (39). Briefly, the mixture of anN-peptide (N36 or N46) and C34-F at final concentration of 1 �M inDulbecco’s PBS was added to the wells of 96-well polystyrene platesprecoated with mAb NC-1 (8 �g/ml) in the presence or absence of apeptide competitor. After incubation at room temperature for 30 min,the plate was washed extensively, followed by addition of 150 �l of PBS.The fluorescence intensity was measured by the Ultra 384 reader(Tecan U.S., Inc., Durham, NC) using fluorescence filters with excita-tion wavelength at 485 nm and emission wavelength at 535 nm. Thepercentage of inhibition of the 6-HB formation by a competitive peptidewas calculated using the equation: % Inhibition of the fluorescent 6-HB

formation � [1 � (E � N)/(P � N)] � 100, where E represents fluores-cence intensity in the presence of a competitive peptide (unconjugatedC34 or T-20), P represents the fluorescence intensity value in theabsence of a competitive peptide, while N is the background correspond-ing to the fluorescent intensity of the wells where only C34-F, butneither N-peptide nor competitor, was added. IC50 (the concentrationfor 50% inhibition) values were calculated using the computer programCalcusyn (21), kindly provided by Dr. T. C. Chou (Sloan-Kettering Can-cer Center, New York).

Fluorescence Polarization—Binding of C34-F or T-20-F to 5-Helixwas determined by a fluorescence polarization assay. In brief, C34-F orT-20-F at 25 nM was added to wells of black 96-well microplates (Corn-ing, Inc., Corning, NY), followed by addition of 5-Helix or its mutant5-Helix (D4) at varying concentrations ranging from 1.56 nM to 200 nM

in Dulbecco’s PBS containing 0.1 mg/ml bovine � globulin (BGG), whichminimizes nonspecific interaction and enhances fluorescent signals.After incubation at room temperature for 1 h, the fluorescence polar-ization values, expressed in millipolarization units (mP), were meas-ured in Ultra 384 reader (Tecan U.S., Inc.) using filters with excitationand emission wavelengths of 485 and 530 nm, respectively. The mPvalues were calculated using the equation: mP � 1000 � [(IS � ISB) �(IP � IPB)]/[(IS � ISB) � (IP � IPB)], where IS is the parallel emissionintensity measurement and IP is the perpendicular emission intensity

FIG. 3. Determination of the 6-HB formation between N36 and C34 and between N36 and T-20 by FN-PAGE. A, FN-PAGE of N36 withC34-F or T-20-F. B, gel in A stained by Coomassie Blue. C, FN-PAGE of N36-F with C34 or T-20. D, gel in C stained by Coomassie Blue. Finalconcentration of each peptide is 40 �M.

T-20 Targets Multiple Sites in gp120/gp4111262

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

sample measurement, whereas ISB and IPB are the corresponding meas-urements of the background signals in buffer.

For competition experiment, equimolar C34-F and 5-Helix (25 nM)were incubated in the presence of competitive peptides at the gradedconcentration. The mP values were measured and the percent inhibi-tion of the 6-HB formation by a competitor was calculated using the

following equation: % Inhibition � [1 � (E � N)/(P � N)] � 100. Erepresents mP value in the presence of a competitive peptide while Prepresents mP value in the absence of a competitor. N corresponds tothe wells where only C34-F, but neither 5-Helix nor competitive pep-tide, was added.

Enzyme-linked Immunosorbent Assay (ELISA)—An ELISA was per-

FIG. 4. Inhibition of the 6-HB formation between N36 and C34-F. A, 6-HB formation between N36 and C34-F measured by FLISA. B,inhibition of the 6-HB formation between N36 and C34-F detected by FLISA. C, inhibition of the 6-HB formation between N36 and C34-Fdetermined by FN-PAGE: lane 1, C34-F; lane 2, N36 � C34-F; lane 3, N36 � C34 � C34-F; and lane 4, N36 � T-20 � C34-F. D, gel in C was stainedwith Coomassie Blue. E, inhibition of 6-HB formation by C34 at increasing concentrations: lanes 1–4, N36 � C34 at 0, 20, 40, and 80 �M,respectively, � C34-F; lane 5, C34-F only; lane 6, T-20 only; and lane 7, N36 � C34. F, gel in E was stained with Coomassie Blue. G, effect of T-20at increasing concentrations on 6-HB formation: lanes 1–4, N36 � T-20 at 0, 20, 40, and 80 �M, respectively, � C34-F; lane 5, C34-F only; and lane6, T-20 only. H, gel in G was stained with Coomassie Blue.

T-20 Targets Multiple Sites in gp120/gp41 11263

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

formed to compare the relative binding affinities of N36 and N36-SIMwith C34 to form 6-HB. Briefly, wells of polystyrene plates were coatedwith 10 �g/ml of mouse anti-HA mAb (Sigma) in 0.1 M Tris-HCl buffer(pH 8.8) and kept overnight at 4 °C, then blocked with 1% nonfat milk.Fifty microliters of N36 or N36-SIM were mixed with 25 �l of C34-biotin(4 �M) and incubated at 37 °C for 30 min, followed by addition of 25 �lof N36-HA (1 �M). After 30 min of incubation at 37 °C, the mixture wastransferred to the coated wells and incubated at 37 °C for 1 h. Thenstreptavidin-labeled horseradish peroxidase (SA-HRP) (Zymed Labo-ratories Inc., S. San Francisco, CA), and the substrate, 3,3�,5,5�-tetramethylbenzidine (TMB) (Sigma) were added sequentially. Ab-sorbance at 450 nm (A450) was determined spectrophotometrically byan ELISA reader (Ultra 384 model, Tecan US, Research Triangle

Park, NC). The percent inhibition of N36-HA/C34-biotin 6-HB forma-tion by N36 and N36-SIM, respectively, was calculated as describedpreviously (20), and IC50 values were calculated using the computerprogram Calcusyn (21).

Circular Dichroism (CD) Spectroscopy—An N-peptide (N36, N46,T-21, or N36-F10) was incubated with a C-peptide (C34 or T-20) at37 °C for 30 min (the final concentrations of N-peptides and C-peptideswere 10 �M in 50 mM sodium phosphate and 150 mM NaCl, pH 7.2). Theisolated N- and C-peptides were also tested. CD spectra of these pep-tides and peptide mixtures were acquired on Jasco spectropolarimeter(Model J-715, Jasco Inc., Japan) at room temperature using a 5.0-nmbandwidth, 0.1-nm resolution, 0.1-cm path length, 4.0-s response time,and a 50-nm/min scanning speed. The spectra were corrected by the

FIG. 5. Interaction of T-20 and C34, respectively, with T-21. A, analysis of the secondary structure of T-21, C34, and the mixture ofT-21�C34; B, analysis of the secondary structure of T-21, T-20, and the mixture of T-21�T-20; C, analysis of 6-HB formation between T-21 and C34and between T-21 and T-20, respectively, by N-PAGE; D, inhibition of the 6-HB formation between N36 and C34-F by T-20 as determined byFN-PAGE; E, gel in D was stained with Coomassie Blue.

FIG. 6. Interaction of C-peptides with 5-Helix to form the 6-HB and inhibition of 6-HB formation between C34-F and 5-Helix. A,binding of C34-F and T-20-F to 5-Helix and 5-Helix (D4), respectively, to form 6-HB was measured by fluorescence polarization assay. B, inhibitionof interaction between C34-F and 5-Helix to form the 6-HB was quantitated based on the reduction of the polarization values.

T-20 Targets Multiple Sites in gp120/gp4111264

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

subtraction of a blank corresponding to the solvent. Thermal denatur-ation was monitored at 222 nm by applying a thermal gradient of5 °C/min. The melting curve was smoothed, and the midpoint of thethermal unfolding transition (Tm) values was calculated using Jascosoftware utilities as described previously (40).

HIV-1-mediated Cell-Cell Fusion—A dye transfer assay was used fordetection of HIV-1-mediated cell fusion as described previously (8, 41).H9/HIV-1IIIB cells were labeled with a fluorescent reagent, Calcein AM(Molecular Probes, Inc., Eugene, Oregon) and then incubated with MT-2cells (ratio � 1:10) in 96-well plates at 37 °C for 2 h in the presence orabsence of peptides tested. The fused and unfused Calcein-labeled HIV-1-infected cells were counted under an inverted fluorescence microscope(Zeiss, Germany) with an eyepiece micrometer disc. Four fields per wellwere counted. The percentage of inhibition of cell fusion by peptides wascalculated as described previously (41), and IC50 values were calculated(42) using the computer program Calcusyn.

RESULTS

Unlike C34, T-20 Does Not Form Stable 6-HB with N36—Toinvestigate the interaction of T-20 and C34 with N36, respec-tively, the CD spectrum of equimolar mixtures of C34 � N36and of T-20 � N36 were recorded. In agreement with theresults reported by Lu et al. (13), C34 and N36 separately didnot adapt to any stable conformation, characterized by the CDspectra distinctive of random coils, while the equimolar mix-ture of the two peptides indicated the formation of a helicalcomplex, characterized by the saddle-shaped negative peak inthe far UV region of the CD spectrum and the significantincrease of molar ellipticity (�) at 222 nm (Fig. 2A). This helicalbundle is relatively stable in phosphate buffer with a Tm about66 °C (Fig. 2B). The CD spectrum of T-20 was rather similar tothat of C34, indicating mainly random coil conformation. How-ever, the spectrum of the T-20 � N36 mixture (equimolar) wasdistinctively different from that of C34 � N36. The saddle-shaped negative peak in the far UV region was clearly lacking.The �222 was just the sum of the spectra of N36 and T-20,indicating no increase in helical content (Fig. 2C). The CD datademonstrated that the addition of T-20 failed to induce anyhelical structure of N36.

The lack of interaction between T-20 and N36 is also consist-ent with the results of the FN-PAGE experiments (Fig. 3).When the fluorescence-labeled C34 (C34-F) and N36 weremixed with equal molar ratio, two fluorescence bands wererevealed (Fig. 3A, lane 4). The lower band with low fluorescenceintensity was the free C34-F, and the upper band with highfluorescence intensity corresponded to the 6-HB, confirmed byWestern blot using the 6-HB-specific mAb NC-1 and FLISAwith the materials eluted from the upper band as describedpreviously (38). While in the mixture of fluorescence-labeledT-20 (T-20-F) and N36, the only band observable was that offree T-20-F (Fig. 3A, lane 5) with the same intensity as theband of free T-20-F at the same concentration (Fig. 3A, lane 3).The N-peptide N36 was not observable because it did not havefluorescent label and also because it carries net positivecharges. The same experiments were also carried out usingfluorescence-labeled N36 (N36-F) and unconjugated C34 orT-20. Again, the 6-HB formed between C34 and N36-F waseasily identifiable (Fig. 3C, lane 4), while no band was observedfor the mixture of N36-F and T-20 (Fig. 3C, lane 5). The freeN36-F (Fig. 3C, lane 1) did not enter the gel because of its netpositive charges, which causes it to migrate upward toward thecathode (the negative end), and only appeared as a faint bandstacked on the top of the gel. When the same gel was stained byCoomassie Blue (Fig. 3D), the bands of C34, T-20, and theN36-F�C34 complex were observable but not the band of N36-F.These results suggest that T-20, unlike C34, does not interactwith N36 to form 6-HB.

In the next experiment, the effects of T-20 on the 6-HBformation between N36 and C34-F were investigated using a

FLISA established previously (39). Only the N36�C34-F com-plex, but not the isolated C34-F or N36, could be captured bythe mAb NC-1, which specifically recognized the conformationof the 6-HB (Fig. 4A) (20). As expected, addition of the uncon-jugated C34 reduced the fluorescent intensity of the C34-F�N366-HB in a dose-dependent manner (Fig. 4B) because of thecompetitive replacement of C34-F by the unconjugated C34. Inthe contrast, the addition of T-20 did not affect the formation ofC34-F�N36 6-HB and the T-20 appeared to be unable to com-pete with C34-F to form the 6-HB with N36 (Fig. 4B).

The inhibitory activity of T-20 and C34 on formation of the6-HB was further studied by FN-PAGE. As shown in Fig. 4C,the mixture of N36�C34-F (lane 2) showed a lower band ofC34-F and an upper band corresponding to the 6-HB, while thesample containing only C34-F (lane 1) showed only a singlefluorescence band located at the lower portion of the gel. Whenunconjugated C34 was added to the N36�C34-F mixture

FIG. 7. Effect of T-20-resistant mutation of the GIV motif inN36 on the six-helix bundle formation. A, interaction of N36 andN36-SIM with C34 to form the 6-HB, respectively, as determined byFN-PAGE. B, interaction of N36 and N36-SIM with C34 to form the6-HB, respectively, as measure by FLISA. C, inhibition of N36-HA�C34-biotin 6-HB formation by N36 and N36-SIM, respectively, as measuredby a competition ELISA.

T-20 Targets Multiple Sites in gp120/gp41 11265

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

(lane 3), the fluorescence intensity of the C34-F band increasedwhile that of band of the complex decreased. The position of the6-HB complex band moved up, suggesting that the 6-HB isformed by N36 and a mixture of C34 and C34-F. No effects wereobserved when T-20 was added to the N36�C34-F mixture. Theintensity and position of both the bands of N36�C34-F and freeC34-F remained unaffected (lane 4).

The Coomassie Blue staining of the same gel confirmed thesame findings (Fig. 4D). When C34 was added to the N36�C34-Fmixture (lane 3), the band corresponding to the 6-HB formedbetween N36 and the mixture of C34�C34-F was smeared andlocated at a little higher position than that of N36�C34-F 6-HB(lane 2), consistent with the lower mobility of the free, uncon-jugated C34 compared to the isolated FITC-conjugated C34(the free, unconjugated C34 was located just above that of thefree C34-F in lane 3). As the C34-F in the original N36�C34-F

complex was replaced by unconjugated C34, the intensity of theC34-F band was also increased compared to that in lane 2.When free T-20 was added to the N36�C34-F mixture (lane 4),similar as in FN-PAGE, the density and position of the bandscorresponding to N36�C34-F 6-HB and the free C34-F remainedthe same. The band corresponding to the free T-20 was re-vealed clearly by the isolated third band in the middle betweenthe upper band of the 6-HB and the lower band of C34-F(Fig. 4D).

The gel electrophoresis experiments also showed that theunconjugated C34 inhibited the fluorescent N36�C34-F 6-HBformation in a dose-dependent manner. With the increasingconcentrations of C34 added to the N36�C34-F mixture, thefluorescence intensity of the lower band corresponding toC34-F increased and that of the upper band decreased. Theposition of the upper band moved up gradually, suggesting the

FIG. 8. Analysis of the secondary structure of N36-F10, C34, T-20, and the mixtures of N36-F10�C34 and N36-F10�T-20 by CDspectroscopy and temperature scan. A, CD spectra for N36-F10, C34, and their complex. B, temperature scanning at 222 nm for N36-F10�C34complex. C, CD spectra for N36-F10, T-20, and their complex. D, temperature scanning at 222 nm for N36-F�T-20 complex. Final concentration ofeach peptide is 10 �M.

TABLE IN- and C-peptide sequences and their inhibitory activity on HIV-1-mdeiated cell-cell fusion

Peptide Residue no.a Sequence IC50b

nM

N36 546–581 SGIVQQQNNLLRAIEAQQHLLQLTVWGIKQLQARIL 584 � 46N36-SIM 546–581 SSIMQQQNNLLRAIEAQQHLLQLTVWGIKQLQARIL 538 � 30N46 536–581 TLTVQARQLLSGIVQQQNNLLRAIEAQQHLLQLTVWGIKQLQARIL 313 � 7N36-F10 536–569 TLTVQARQLLSGIVQQQNNLLRAIEAQQHLLQLT 2200 � 90T-21 553–590 NNLLRAIEAQQHLLQLTVWGIKQLQARILAVERYLKDQ 247 � 14C34 628–661 WMEWDREINNYTSLIHSLIEESQNQQEKNEQELL 1.0 � 0.2T-20 638–673 YTSLIHSLIEESQNQQEKNEQELLELDKWASLWNWF 3.0 � 0.4T-20-ANAA YTSLIHSLIEESQNQQEKNEQELLELDKWASLANAA 1330 � 57T-20-ANAA-octyl YTSLIHSLIEESQNQQEKNEQELLELDKWASLANAA-octyl 21.7 � 2.7

a The residue numbers of each region correspond to their positions in gp160 of HIV-1HXB2.b IC50 means the concentration of a peptide causing 50% inhibition of HIV-1-mediated cell-cell fusion.

T-20 Targets Multiple Sites in gp120/gp4111266

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6-HBs contain more C34 and less C34-F when higher concen-trations of C34 were added (Fig. 4E). When the gel was thenstained with Coomassie Blue, the density of the band corre-sponding to the isolated C34 (just above the band of C34-F)increased with the increasing concentration of C34 added,while the upper 6-HB bands moved up gradually (Fig. 4F). Thebands located between the 6-HBs formed by N36�C34-F (lane 1)and N36�C34 (lane 7) were the 6-HBs containing differentratios of C34:C34-F. When the increasing concentrations of T-20were added to the N36�C34-F mixture, the intensity and positionof the bands corresponding to the 6-HB and the isolated C34-Fdid not change (Fig. 4G), while the intensity of the middle bandcorresponding to the isolated T-20 increased gradually (Fig. 4H).These confirm that T-20, even at high concentrations, cannotcompete with C34 for binding to N36 to form the 6-HB.

Unlike C34, T-20 Does Not Form Stable 6-HB with T-21,Another N-peptide Containing Full-length Cavity-forming Se-quence, but Not the T-20-resistant Determinant—Early evi-dence that shows that T-20 may target the gp41 NHR region isfrom a CD analysis of the interaction between T-20 and T-21,an N-peptide that is derived from the C-terminal portion of theNHR region and contains the entire cavity-forming sequence(Fig. 1), because addition of T-20 to T-21 results in the decreaseof the �-helicity of T-21 (7, 11, 43). But unlike N36, T-21 doesnot contain the GIV motif, the determinant of T-20 resistance(36, 44). We used similar approaches to compare the secondarystructures formed by T-21�C34 and T-21�T-20. T-21, like N36,interacted with C34 to form an �-helical oligomer (Fig. 5A).However, T-21 could not form any �-helical complex with T-20.In contrast, T-20 distorted the �-helical structure (Fig. 5B),consistent with the observation by Wild et al. (11, 43). Theseresults suggest that T-20 may interact with T-21 to change its�-helicity, but this interaction did not result in the formation of�-helical oligomeric structure.

The interaction of T-20 and C34 with T-21 was also examinedusing N-PAGE and FN-PAGE. As shown in Fig. 6C, T-20 andC34 each gave a band located at the lower portion of the gel.The T-21�C34 mixture gave a lower band corresponding to C34and an upper band corresponding to the 6-HB. However, theT-21�T-20 mixture exhibited only the lower T-20 band, but noupper 6-HB band (Fig. 5C). With the increasing concentrationof T-20, the intensity and the position of either the upper bandcorresponding to the 6-HB formed by T-21�C34-F or the lowerone corresponding to the isolated C34-F did not change, whilethe intensity of the middle band corresponding to the isolatedT-20 increased gradually (Fig. 5, D and E). These results sug-gest that T-20, unlike C34, did not interact with T-21 to formstable 6-HB, nor inhibit interaction between C34-F and T-21.

Unlike C34, T-20 May Weakly Bind to 5-Helix, but Does NotBlock C34 Interacting with 5-Helix to Form 6-HB—Using pro-tein design strategies, Root et al. (45) identified a polypeptide,denoted 5-Helix, which is composed of three N-peptides (N-40,residues 543–582) and two C-peptides (C-38, residues 625–662) connected by -GGSGG- linkers. Since 5-Helix contains fiveof the six �-helical coils and has one of the three groovesexposed because of the missed C-helix, it can attract a CHRdomain of the viral gp41 or a C-peptide, such as C34, to form astable six-helix bundle. Binding of C34-F to 5-Helix to form6-HB was investigated using the fluorescence polarization as-say. As shown in Fig. 6A, the polarization values (mP) in-creased with increasing concentration of the 5-Helix until itreached a plateau when the concentration ratio of 5-Helix:C34-F became about 1:1. Under the same conditions, C34-Fcould not bind to 5-Helix (D4), a 5-Helix mutant with Aspreplacement of the four highly conserved residues at the “e”position (Val549, Leu556, Gln563, and Val570) in the third N40

segment, which is responsible for C-peptide binding. This con-firms that C34-F can specifically interact with 5-Helix to form6-HB. T-20-F did not have significant interaction with 5-Helixat 25 nM. When the concentration of 5-Helix was increased to200 �M (the ratio of 5-Helix:T-20-F is about 8:1), T-20-F showedcertain binding activity at a level about 60% of the maximuminteraction between 5-Helix and C34 at 1:1 ratio. This suggeststhat T-20 may bind to 5-Helix, but has much lower bindingaffinity than C34. In a competition experiment, the unconju-gated free C34 could compete with C34-F to bind 5-Helix,resulting in decrease of the polarization values. In contrast, thefree, unconjugated T-20, which has similar potent anti-HIV-1activity to C34, could not block C34-F binding to 5-Helix (Fig.6B). This further confirmed that T-20 cannot interact with any

FIG. 9. Interaction of C34 and T-20 with N36-F10 determinedby N-PAGE and FN-PAGE. A, binding of C34 and T-20 to N36-F10determined by FN-PAGE; B, dose-dependent binding of T-20-F to N36-F10; C, interaction of T-20 and its WNWF motif mutant (T-20-ANAA)with N36-F10, respectively, as shown by N-PAGE.

T-20 Targets Multiple Sites in gp120/gp41 11267

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

polypeptides or constructs with a full-length cavity-formingsequence to form stable 6-HB.

The Determinant of T-20 Resistance in the NHR Region IsNot the Critical Site Involved in the Interaction Between theNHR and CHR Regions to Form 6-HB—One major evidencesupporting the hypothesis that T-20 targets the NHR region isthat the determinant of T-20 resistance in the HIV-1 variantsis located in the gp41 NHR region (residues 547–549: GIV) (36,44). However, clinical data showed that the subtype differencein T-20 susceptibility is unrelated to NHR genetic variation(46). Changing the sequences other than the gp41 NHR region,such as gp120 V3 loop, can also modulate T-20 sensitivity (47,48). Sensitivity of T-20 also correlates with envelope/coreceptoraffinity, receptor density, and fusion kinetics (49). Further-more, T-20-resistant HIV-1 variants remain fully sensitive tothe C-peptide T-649 (residues 628–663), which covers the en-tire sequence of C34 (residues 628–661) (36). To investigatewhether the GIV motif in the gp41 NHR region is important forthe interaction between the NHR and CHR to form the 6-HB,we synthesized an N36 analogous peptide, designated N36-SIM, in which the wild-type GIV motif was replaced by T-20-resistant motif SIM (36) and compared the ability of N36 andN36-SIM to form the 6-HB with C34. Both wild-type N36 andN36-SIM had similar activity to interact with C34 to form the6-HB as determined by FN-PAGE (Fig. 7A) and FLISA (Fig.7B), and to block the 6-HB formation between N36-HA andC34-biotin (IC50 values of 2.67 � 0.26 and 2.74 � 0.19 �M,respectively) as assessed by ELISA (Fig. 7C). In addition, bothN36 and N-36-SIM had similar inhibitory activity on HIV-1Env-mediated cell fusion with IC50 values of 0.584 and 0.538�M, respectively (Table I). This is in a good agreement with the

result obtained from the experiment on T-21, which does notcontain the GIV motif, but can interact with C34 to form stable6-HB (Fig. 5). These results suggest that the GIV motif may beimportant for T-20 binding and for T-20-mediated anti-HIV-1activity, but may not be critical for the gp41 NHR and CHRassociation to form the 6-HB and for C34-mediated HIV-1fusion inhibitory activity.

T-20 Interacts with N36-F10, an N-peptide Containing theGIV Motif and Partial Cavity-forming Sequence—To furtherinvestigate the molecular interaction of T-20, a peptide, N36-F10, was synthesized which covers the region of residues 536–569, about 10 amino acids preceding the sequence of N36 (res-idues 546–581) and contains the GIV motif, the determinant ofT-20 resistance. Unlike N36, it does not contain the full-lengthsequence required for forming the hydrophobic cavity (Fig. 1C).Peptide C34 appeared to be able to interact with N36-F10 butform only an unstable complex with the helical content muchlower than that of N36�C34 6-HB (Fig. 8A). The Tm of thisN36-F10�C34 complex was only around 36 °C (Fig. 8B) whilethe Tm of N36�C34 is 66 °C (Fig. 2B) (50). Peptide T-20 couldalso interact with N36-F10 to form a complex with low �-helixcontent (Fig. 8C) and low thermal stability (Fig. 8D, Tm �48 °C).

The interaction was confirmed by FN-PAGE. The N36-F10�C34-F complex band was not visible because it may bedamaged during electrophoresis because of its unstable prop-erty (e.g. Tm � 36 °C). Mixing N36-F10 and T-20-F induced aband of the complex (Fig. 9A). With increasing concentrationsof N36-F10, the complex band was stronger, the while theT-20-F band got weaker (Fig. 9B), indicating that there is aninteraction between T-20-F and N36-F10.

FIG. 10. Analysis of the secondary structures of N46, C34, T-20, the mixtures of N46�C34 and N46�T-20 by CD spectroscopy, and theblocking activity of N46 on C34 and T-20-mediated inhibition of cell-cell fusion. A, CD spectra for N46, C34, and their complex. B, CDspectra for N46, T-20, and their complex. C, temperature scanning at 222 nm for N46�C34 complex. Inserted is the first derivative of the curveagainst temperature for Tm determination. Final concentration of each peptide is 10 �M. D, blockage of C34 and T-20 mediated inhibition of cell-cellfusion by N46.

T-20 Targets Multiple Sites in gp120/gp4111268

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Is the interaction of T-20 to N36-F10 enough to explain theinhibitory activity of T-20? T-20 contains a consecutive trypto-phan-rich (TR) region at the C terminus (Fig. 1), and the 3tryptophan residues (Trp666, Trp670, and Trp672) are critical toits inhibitory activity. T-20-ANAA, a mutant T-20 in which thelast four WNWF residues were changed to ANAA, shows muchless anti-HIV activity (51). Using HIV-1-mediated cell-cell fu-sion assay, we confirmed that the inhibitory activity of T-20-ANAA is about 450-fold less than that of wild-type T-20 (TableI). In N-PAGE, T-20-ANAA could interact with N36-F10 toform a complex, which means that the WNWF is not critical forthe binding of T-20 to N36-F10 (Fig. 9C). These results indicatethat although T-20 can bind to gp41 NHR, this binding may notaccount for its potent anti-HIV activity.

T-20 May Interact with N46, Which Overlaps N36 and N36-F10; but Unlike C34, T-20 Cannot Form Stable 6-HB withN46—Although we have compared the ability of C34 and T-20to interact with N36 and N36-F10 to form the 6-HB, one mayargue that neither of these N-peptides contains the full bindingregion for C34 and T-20. Therefore, we synthesized a longer

N-peptide, N46 (residues 536–581), which covers the sequencesof both N36 and N36-F10, and compared the ability of C34 andT-20 to form 6-HB with N46 and to block 6-HB formationbetween N46 and C34-F. CD analysis indicated that C34 couldinteract with N46 to form the �-helical bundle, similar to thatformed by C34 and N36 (Fig. 10A). However, the mixture ofT-20 and N46 did not result in the increase of the �-helicity ofN46, but rather caused a decrease in it (Fig. 10B), in agreementwith the results obtained from a mixture of T-20 and T-21 (Fig.5B). These confirm that T-20, unlike C34, cannot form the�-helical oligomer with the peptides derived from the gp41NHR region, but may interact with the N-peptides to formunstable or insoluble complexes. Noticeably, the N46�C34 com-plex had higher Tm value (88 °C, Fig. 10C) than that of theN36�C34 complex (66 °C, Fig. 2B), suggesting that the �-helicaloligomer formed by C34 with a longer N-peptide is more stablethan that with a shorter N-peptide.

In cell-cell fusion assay, N46 was more effective than N36and N36-F10 in inhibiting HIV-1 Env-mediated membranefusion with an IC50 value of 313 nM (Table I). At the concen-

FIG. 11. Interaction of C34 and T-20with N46 determined by FN-PAGEand FLISA. A, interaction of N46 (50, 25,12.5, 0 �M) with C34-F (50 �M) or T-20-F(50 �M), respectively, as determined byFN-PAGE. B, inhibitory activity of C34and T-20 on the N46�C34-F 6-HB forma-tion as measured by FLISA.

T-20 Targets Multiple Sites in gp120/gp41 11269

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

tration of 200 �M, N46 completely blocks C34-mediated inhibi-tion of HIV-1 fusion, but had no effect on T-20-mediated anti-HIV-1 activity (Fig. 10D).

As determined by FN-PAGE, two fluorescence bands wererevealed in the lanes where mixtures of N46 at graded concen-trations and C34-F at 50 �M were loaded (Fig. 11A, left panel).The lower band corresponds to the isolated C34-F. The fluores-cence intensity of the upper bands decreased with the decreas-ing concentration of N46, suggesting that the upper band is thecomplex formed by N46 and C34-F, like the one formed by N36and C34-F (Fig. 3A). In lanes where the mixtures of N46 withT-20-F were loaded (Fig. 11A, right panel), there was no visibleband corresponding to the complex formed by N46 and T-20-F,except a fluorescence band located at the lower portion of thegel corresponding to the isolated T-20-F. However, the fluores-cence density of this band increased with decreasing concen-trations of N46, and some weak fluorescence bands were re-vealed on the top of the gels, suggesting that a certain amountof T-20-F may interact with N46 to form a high order of oli-

gomers or aggregates that could not enter into the gel.In FLISA, the N46 could interact with C34-F to form a

complex, which could be captured by the conformation-specificmAb NC-1, suggesting that the complex is 6-HB. Formation ofthe 6-HB by N46 and C34-F could be significantly blocked byC34, but not by T-20 (Fig. 11B).

All the above results suggest that C34 can interact with N46to form a stable and soluble 6-HB, while T-20 may bind to N46to form non-�-helical, unstable, or insoluble complexes or ag-gregates, and T-20 cannot block the 6-HB formation betweenN46 and C34.

T-20-mediated Inhibitory Activity Is Abrogated by PeptidesDerived from the Coreceptor Binding Region in gp120 and theMembrane-spanning Domain in gp41—In addition to a targetsite within the NHR region, T-20 may also bind to a non-specified region at the C terminus of the gp41 ectodomain closeto the viral membrane, the membrane-proximal domain (52,53). Other groups have shown that the sensitivity of HIV-1 toT-20 relates with envelope/coreceptor affinity, receptor density,and fusion kinetics (47, 49). These suggest that T-20 may targetmultiple sites in both gp41 and gp120. We thus used a completeset of HIV-1MN Env (15-mer) peptides to investigate the possi-ble binding regions of T-20, assuming that the peptides derivedfrom the T-20 binding sites in gp41/gp120 can interact withT-20 and block its anti-HIV-1 activity. The peptide C34 andPBS were used as controls. The complete set (212 peptides)covers the whole gp120 and gp41 sequence at a 4 amino acidshift for each peptide. Using HIV-1-mediated cell-cell fusionassay, the Env peptides themselves at final concentration of12.5 �g/ml in PBS (about 7 �M) have no inhibitory activity (Fig.12A). T-20 and C34 at 100 nM final concentration can com-pletely inhibit the cell-cell fusion. We mixed the Env peptides(12.5 �g/ml) with T-20 or C34 (100 nM) at 37 °C for 30 min, thendetected their inhibitory activity. None of the peptides couldblock the anti-HIV-1 activity of C34 (Fig. 12B), but the peptides6312, 6313, 6314 derived from gp120 and 6378, 6379, 6380derived from gp41 completely abrogated T-20-mediated inhibi-tion of the HIV-1-induced cell-cell fusion (Fig. 12C). The se-quence and the blocking activity of these peptides expressed asEC50 (effective concentration of a peptide causing 50% abroga-tion of the anti-HIV-1 activity mediated by T-20) are shown inTable II.

The overlapping peptides 6312, 6313, 6314 share a sequenceof amino acid 417–423, QCKIKQI, which overlaps the se-quences of the �-19 to �-20 strands in the gp120 bridging sheet(residues 412–427). This region is centrally located within thecoreceptor binding site (2), suggesting that T-20 may interactwith this region and block HIV-1 binding to the coreceptor. Theoverlapping peptides 6378, 6379, 6380 share a common se-quence of amino acid 684–691, IFIMIVG, a part of the gp41membrane-spanning domain, which is thought to be the possi-ble second site in gp41 for T-20 binding. The blocking activity ofthe peptides 6378, 6379, and 6380 cannot be ascribed to non-specific hydrophobic interaction since other peptides derivedfrom the transmembrane region of HIV-1 gp41 (e.g. peptides6381 and 6382) and those derived from the FP regions (e.g.6337 and 6338) did not block T-20-mediated inhibition of cell-cell fusion (Fig. 12C). Furthermore, the scrambled version ofthe peptide 6380 did not inhibit the anti-HIV-1 activity of T-20(Table II). These suggest that the peptides 6378, 6379, and6380 specifically interact with T-20 to block its inhibitory ac-tivity on HIV-1 infection.

The gp120 Coreceptor Binding Region and gp41 Membrane-spanning Region Interact at Different Sites in T-20—Shai andco-workers reported that C-terminal octylation of SIV gp41-derived T-20-ANAA mutant can rescue this inactive mutant to

FIG. 12. Effect of short peptides derived from the gp120 andgp41 on the anti-HIV-1 activity mediated by C-peptides. A, effectof the short peptides alone. None of the short peptides dissolved in PBSthemselves showed inhibitory activity on HIV-1-mediated cell-cell fu-sion. B, effect of the short peptides on the inhibitory activity of C34; andC, effect of the short peptides on the inhibitory activity of T-20.

T-20 Targets Multiple Sites in gp120/gp4111270

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

an inhibitory potency similar to that of the wild-type T-20 (62).In the present study, we found that C-terminal octylation canalso rescue a low inhibitory activity of HIV-1IIIB gp41-derivedT-20-ANAA mutant (Fig. 13A and Table I). Using N-PAGE,T-20-ANAA-octyl has the similar binding ability with N36-F10as wild-type T-20 and T-20-ANAA (data not shown). Using theblocking assay, gp120 coreceptor binding region-derived pep-tide 6312 at 1 �M can abrogate both T-20 and T-20-ANAA-ocytl(100 nM)-mediated inhibitory activity on cell-cell fusion, whilegp41 membrane-spanning domain-derived peptide 6380 at 5�M can only block the wild-type T-20 anti-HIV-1 activity (Fig.13B). This indicates that the membrane-spanning domain-de-rived peptides may interact with the tryptophan-rich sequenceat the C-terminal region of T-20, while the interaction site ofthe gp120 coreceptor binding region-derived peptide is beyondthis tryptophan-rich region. It seems that the binding to gp120coreceptor binding region contributes more to the inhibitoryactivity of T-20 because: 1) N36-F10 cannot block T-20 activitycompletely at 0.8 �M and only has marginal blocking activity onT-20-ANAA-ocytl anti-HIV activity (Fig. 13B); and 2) the pep-tides derived from the gp120 coreceptor binding region aremuch more potent than those from the gp41 membrane-span-ning domain in blocking the anti-HIV-1 activity of T-20(Table II).

DISCUSSION

T-20 has been widely used as a proof-of-concept for anti-HIVdrugs, albeit the concept of inhibitor reaction was deduced fromthe work using another C-peptide C34 (6, 12, 14, 33–35). Ourstudies clearly demonstrated that, lacking the N-terminal cav-ity binding sequence, T-20 cannot interact with the cavity-forming sequence in the gp41 NHR region to form a stable6-HB. Thus it is difficult to envision that T-20 would act in thesame way as C34 by interrupting the formation of the gp41core. Instead, our data suggest that T-20 may interact with the

N-terminal portion of NHR region containing the highly re-served GIV motif, the membrane-spanning domain in gp41,and the coreceptor binding site on gp120. These interactions,acting together or alone, cause inhibition of viral fusion.

The assumption of the same mechanism of action of T-20 andC34 mainly come from the study that a contiguous 3-amino acidsequence (residues 547–549: GIV) within the NHR region wasidentified to be critical for T-20-mediated anti-HIV activity. Sub-stitutions at two positions within the GIV sequence, Gly to Ser orAsp and Val to Met, resulted in virus resistance to T-20 (36, 54,55). However, we found that the GIV motif may not be involvedin the interaction between the NHR and CHR to form 6-HB since:1) N36-SIM, the N36 analog with mutations in the GIV motif,has similar activity as the wild-type N36 to interact with C34 toform the 6-HB and to block the 6-HB formation between N36-HAand C34-biotin (Fig. 7), and 2) T-21 which does not contain theGIV motif can form 6-HB with C34 (Fig. 5). In addition, T-20-resistant HIV-1 strains with SIM mutations are sensitive toT-649, a C34-containing peptides (36), suggesting that GIV motifis important for T-20-mediated, but not for C34-mediated anti-HIV-1 activity. Our binding study confirmed that T-20 can inter-act with the peptide N36-F10, which contains GIV motif and apartial cavity-forming sequence, however the Tm of the formedcomplex is 48 °C, much lower than that found for the 6-HBbetween N36 and C34 (Tm value � 66 °C). C34 can also interactwith N36-F10 with only marginal stability (Tm value is about36 °C), consistent with the report by Dwyer et al. (19). Therefore,binding of T-20 to the N-terminal portion of the NHR may not bestrong enough to compete with the binding of viral CHR to NHRregion. At least, interaction of T-20 with this region may notcontribute to the major activity of T-20 to inhibit HIV-1-mediatedmembrane fusion.

T-20 binding to the N36-F10 region may interfere with thefusogenic activity mediated by the fusion peptide (FP) adjacent

FIG. 13. HIV-1 fusion inhibitory activity mediated by peptides T-20, T-20-ANAA, and T-20-ANAAoctyl in the presence or absenceof the peptides N36, N36-F, 6312, and 6380. A, inhibitory activity of T-20, T-20 mutant (T-20-ANAA), and its octyl-derivative againstHIV-1-mediated membrane fusion; B, blocking activity of the peptides N36 (0.2 �M), N36-F10 (0.8 �M), 6312 (1 �M), and 6380 (5 �M) on T-20 (20nM) and octylated T-20 mutant (100 nM)-mediated anti-HIV-1 activity. PBS was used as control.

TABLE IIBlocking activity of peptides derived from the HIV-1MN gp120/gp41 on inhibition of HIV-1 fusion mediated by T-20 at100 nM

Peptide Residue no.a Sequence EC50b

�M

6311 409–423 GSNNNITLQCKIKQI 1.581 � 0.0826312 413–427 NITLQCKIKQIINMW 0.136 � 0.0156313 417–431 QCKIKQIINMWQEVG 0.111 � 0.0186378 676–690 TNWLWYIKIFIMIVG 6.231 � 0.2386379 680–694 WYIKIFIMIVGGLVG 7.415 � 0.3126380 684–698 IFIMIVGGLVGLRIV 4.726 � 0.2986380-Sc VIGIVGLVGMIRILF � 100

a The residue numbers of each region correspond to their positions in gp160 of HIV-1HXB2.b EC50 means the effective concentration of a peptide causing 50% abrogation of the T-20-mediated inhibition of HIV-1 fusion.c 6380-S, scrambled peptide of 6380.

T-20 Targets Multiple Sites in gp120/gp41 11271

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

to the NHR of gp41. FP alone can insert into and fuse biologicalmembranes (56). Extension of FP to include partial sequence ofNHR region results in significant enhancement of FP-mediatedmembrane fusion, suggesting that NHR has a synergistic rolein FP-mediated fusogenic activity (57). Mobley et al. (58) dem-onstrated that T-20 can block hemolysis and cell aggregationinduced by FP and the N-peptide fusion inhibitor T-21, imply-ing the interaction of T-20 with the FP. It is plausible that T-20inhibits the viral fusion by affecting the membrane perturba-tions associated with the interactions of FP and NHR, whichunderlie the merging of the viral envelope with the targetcell membrane.

In addition to the interaction with the NHR of gp41, T-20may also target the membrane-spanning domain in gp41 andblock the late step of membrane fusion. Peptides derived fromthe membrane-spanning domain could effectively abrogateT-20-mediated inhibitory activity on cell-cell fusion. However,one may raise doubts about the approach because none of thepeptides derived from the gp41 NHR region abrogates C34-mediated anti-HIV-1 activity. This is because the peptides usedin this study for Pepscan experiment are too short (15-mer) toform multiturn �-helices required for interaction with C34. Weand others (12, 59) have previously demonstrated that N-pep-tides with long sequence (e.g. N36 and N46) can block C34-mediated anti-HIV-1 activity, and the C-peptides with shortsequence (e.g. �26 amino acid residues) lack inhibitory activityon HIV-1-mediated membrane fusion, suggesting that a certainlength of sequence is required for the peptides derived from thegp41 NHR and CHR regions to interact with their counterpartsfor blocking the activity of the corresponding N- and C-pep-tides. The binding site of T-20 in the gp41 membrane-spanningdomain consists of a short sequence with seven residues, IF-IMIVG, most of which are hydrophobic. T-20 may bind to thisfragment through its hydrophobic residues located in the tryp-tophan-rich domain. It has been proven that the hydrophobicresidues in the tryptophan-rich sequence are critical for theanti-HIV-1 activity of T-20 (11). Removal or replacement of 4residues, WNWF, at the C terminus of T-20 almost completelyeliminated its anti-HIV-1 activity while deletion of 3 residuesfrom the N terminus of T-20 has no effect on its fusion inhibi-tory activity (11, 51). The tryptophan-rich domain in gp41 isalso important for Env-mediated fusion (57) because it maybind to the membrane surface (60) and participate in oligomer-ization of gp41 to form fusion pore in the membrane (53). It hasbeen reported that the T-20 analog with mutations in thetryptophan-rich domain, T-20-ANAA, has no HIV-1 fusion in-hibitory activity, but the membrane-anchoring T-20-ANAA hassimilar anti-HIV-1 activity as the wild-type T-20 (61). C-termi-nal octylation rescue the inactive ANAA mutant of T-20-likepeptide derived from the SIV gp41 (62). These suggest thatinteraction of T-20 and the target membrane is necessary forT-20-mediated membrane fusion inhibitory activity. The datafrom our study confirm these observations and suggest thatT-20 may also bind to viral membrane by interacting with themembrane-spanning domain at the late step of membrane fu-sion (post-lipid mixing) to prevent fusion pore formation. SinceT-20 can bind to the N-terminal fragment of the NHR regionand other sites in gp41, it is understandable that T-20 cancapture gp41 in immunoprecipitation experiments (63).

T-20 may interact with the coreceptor binding site on gp120and block interaction of gp120-CD4 complex with the corecep-tor. In the present study, we found that the peptide 6311–6313,which share the sequence of residues 417–423 in the HIV-1gp120 were able to abrogate T-20-mediated anti-HIV activity.Interestingly, this sequence overlaps the �19 and �20 frag-ments (residues 412–427) in the bridging sheet, a coreceptor

binding site in gp120 (2). It suggests that the coreceptor bind-ing site in gp120 may be another target site for T-20 andinteraction of T-20 with this site may account for part of theT-20-mediated HIV-1 fusion inhibitory activity. Our finding isconsistent with the mounting new evidence suggesting thatT-20 may bind to gp120. Reeves et al. (49, 64) showed that theT-20 sensitivity correlated with gp120/coreceptor affinity andwas affected by mutations in the coreceptor binding site ongp120. Most recently, Yuan et al. (65) reported that T-20 boundto gp120 of the X4 but not R5 virus in a CD4-induced, V3loop-dependent manner and binding of T-20 to gp120 blockedthe interaction between the gp120�CD4 complex and theCXCR4 coreceptor, indicating that T-20 preferably interactwith CXCR4 binding site on gp120. Alam et al. (66) usingsurface plasmon resonance (SPR) binding assays found thatT-20 can bind to CD4-induced gp120, and the binding region iscentrally located within the HIV-1 coreceptor binding site.

The interaction with the gp120 coreceptor binding region andgp41 membrane-spanning domain may be mediated by differ-ent regions in T-20. Peptides derived from the gp41 membrane-spanning domain can significantly block inhibition of cell-cellfusion mediated by T-20, but not by T-20-ANAA-ocytl. Thissuggests that the gp41 membrane-spanning domain may inter-act with the C-terminal tryptophan-rich region of T-20. Incontrast, the peptides derived from the HIV-1 gp120 coreceptorbinding region can effectively abrogate the anti-HIV-1 activitymediated by both T-20 and T-20-ANAA-ocytl, indicating thatthe interaction site of the gp120 coreceptor binding region-derived peptides is located outside the T-20 tryptophan-richregion. Because the peptide T-20-ANAA can bind to N36-F10 aswild-type T-20, the binding site for the gp41 NHR region ingp41 may be also located beyond the C-terminal tryptophan-rich domain. Furthermore, since peptides derived from gp120coreceptor binding region are much more potent in blockingT-20 activity than those from the gp41 membrane spanningdomain and the peptide N36-F10, binding of T-20 to the gp120coreceptor binding region seems to contribute more for theinhibitory activity of T-20.

The present findings indicate that T-20 targets multiple sitesin gp41 and gp120 and its anti-HIV activity may be a combi-natory effect mediated by different mechanisms of action. Fur-ther clarification of these mechanisms may provide new targetsfor development of novel anti-HIV drugs and vaccines. SinceC34 has more potent anti-HIV activity than T-20 and a differ-ent mechanism of action from T-20, C34 or its analogs shouldalso be developed for clinical application and used for treat-ment of patients infected by HIV-1 strains resistant to T-20.

Acknowledgments—We thank Dr. Michael Root at Thomas JeffersonUniversity for providing 5-Helix and 5-Helix (D4).

REFERENCES

1. Sattentau, Q. J., Dalgleish, A. G., Weiss, R. A., and Beverley, P. C. (1986)Science 234, 1120–1123

2. Kwong, P. D., Wyatt, R., Robinson, J., Sweet, R. W., Sodroski, J., and Hen-drickson, W. A. (1998) Nature 393, 648–659

3. Rizzuto, C. D., Wyatt, R., Hernandez-Ramos, N., Sun, Y., Kwong, P. D.,Hendrickson, W. A., and Sodroski, J. (1998) Science 280, 1949–1953

4. Feng, Y., Broder, C. C., Kennedy, P. E., and Berger, E. A. (1996) Science 272,872–877

5. Moore, J. P., Jameson, B. A., Weiss, R. A., and Sattentau, Q. J. (1993) in ViralFusion Mechanisms (Bentz, J., ed) pp. 233–289, CRC Press, Boca Raton, FL

6. Chan, D. C., and Kim, P. S. (1998) Cell 93, 681–6847. Wild, C., Oas, T., McDanal, C., Bolognesi, D., and Matthews, T. (1992) Proc.

Natl. Acad. Sci. U. S. A. 89, 10537–105418. Jiang, S., Lin, K., Strick, N., and Neurath, A. R. (1993) Nature 365, 1139. Jiang, S., Lin, K., Strick, N., and Neurath, A. R. (1993) Biochem. Biophys. Res.

Commun. 195, 533–53810. Wild, C., Greenwell, T., and Matthews, T. (1993) AIDS Res. Hum. Retroviruses

9, 1051–105311. Wild, C. T., Shugars, D. C., Greenwell, T. K., McDanal, C. B., and Matthews,

T. J. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 9770–977412. Lu, M., Blacklow, S. C., and Kim, P. S. (1995) Nat. Struct. Biol. 2, 1075–108213. Lu, M., and Kim, P. S. (1997) J. Biomol. Struct. Dyn. 15, 465–471

T-20 Targets Multiple Sites in gp120/gp4111272

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

14. Chan, D. C., Fass, D., Berger, J. M., and Kim, P. S. (1997) Cell 89, 263–27315. Weissenhorn, W., Dessen, A., Harrison, S. C., Skehel, J. J., and Wiley, D. C.

(1997) Nature 387, 426–42816. Tan, K., Liu, J., Wang, J., Shen, S., and Liu, M. (1997) Proc. Natl. Acad. Sci.

U. S. A. 94, 12303–1230817. Chan, D. C., Chutkowski, C. T., and Kim, P. S. (1998) Proc. Natl. Acad. Sci.

U. S. A. 95, 15613–1561718. Ji, H., Shu, W., Burling, T., Jiang, S., and Lu, M. (1999) J. Virol. 73,

8578–858619. Dwyer, J. J., Hasan, A., Wilson, K. L., White, J. M., Matthews, T. J., and

Delmedico, M. K. (2003) Biochemistry 42, 4945–495320. Jiang, S., Lin, K., and Lu, M. (1998) J. Virol. 72, 10213–1021721. Jiang, S., Lin, K., Zhang, L., and Debnath, A. K. (1999) J. Virol. Methods 80,

85–9622. Blacklow, S. C., Lu, M., and Kim, P. S. (1995) Biochemistry 34, 114955–11496223. Rapaport, D., Ovadia, M., and Shai, Y. (1995) EMBO J. 14, 5524–553124. Lombardi, S., Massi, C., Indino, E., La Rosa, C., Mazzetti, P., Falcone, M. L.,

Rovero, P., Fissi, A., Pieroni, O., Bandecchi, P., Esposito, F., Tozzini, F.,Bendinelli, M., and Garzelli, C. (1996) Virology 220, 274–284

25. Lambert, D. M., Barney, S., Lambert, A. L., Guthrie, K., Medinas, R., Davis,D. E., Bucy, T., Erickson, J., Merutka, G., and Petteway, Jr, S. R. (1996)Proc. Natl. Acad. Sci. U. S. A. 93, 2186–2191

26. Weissenhorn, W., Carfi, A., Lee, K. H., Skehel, J. J., and Wiley, D. C. (1998)Mol. Cell 2, 605–616

27. Xu, Y., Gao, S., Cole, D. K., Zhu, J., Su, N., Wang, H., Gao, G. F., and Rao, Z.(2004) Biochem. Biophys. Res. Commun. 315, 664–670

28. Liu, S., Xiao, G., Chen, Y., He, Y., Niu, J., Escalante, C., Xiong, H., Farmar, J.,Debnath, A. K., Tien, P., and Jiang, S. (2004) Lancet 363, 938–947

29. Tripet, B., Howard, M. W., Jobling, M., Holmes, R. K., Holmes, K. V., andHodges, R. S. (2004) J. Biol. Chem. 279, 20836–20849

30. Xu, Y., Lou, Z., Liu, Y., Pang, H., Tien, P., Gao, G. F., and Rao, Z. (2004) J. Biol.Chem. 279, 49414–49419

31. Cooper, D. A., and Lange, J. M. (2004) Lancet Infect. Dis. 4, 426–43632. Lalezari, J. P., Henry, K., O’Hearn, M., Montaner, J. S., Piliero, P. J., Trottier,

B., Walmsley, S., Cohen, C., Kuritzkes, D. R., Eron, J. J., Jr., Chung, J.,DeMasi, R., Donatacci, L., Drobnes, C., Delehanty, J., and Salgo, M. (2003)N. Engl. J. Med. 348, 2175–2185

33. Kilby, J. M., and Eron, J. J. (2003) N. Engl. J. Med. 348, 2228–223834. Kilby, J. M., Hopkins, S., Venetta, T. M., DiMassimo, B., Cloud, G. A., Lee,

J. Y., Alldredge, L., Hunter, E., Lambert, D., Bolognesi, D., Matthews, T.,Johnson, M. R., Nowak, M. A., Shaw, G. M., and Saag, M. S. (1998) Nat.Med. 4, 1302–1307

35. Moore, J. P., and Doms, R. W. (2003) Proc. Natl. Acad. Sci. U. S. A. 100,10598–10602

36. Rimsky, L. T., Shugars, D. C., and Matthews, T. J. (1998) J. Virol. 72, 986–99337. Xu, Y., Zhang, X., Matsuoka, M., and Hattori, T. (2000) FEBS Lett. 487,

185–18838. Liu, S., Zhao, Q., and Jiang, S. (2003) Peptide 24, 1303–131339. Liu, S., Boyer-Chatenet, L., Lu, H., and Jiang, S. (2003) J. Biomol. Screen. 8,

685–69340. Jones, D. K., Badii, R., Rosell, F. I., and Lloyd, E. (1998) Biochem. J. 330,

983–98841. Jiang, S., Radigan, L., and Zhang, L. (2000) Proc. SPIE 3926, 212–219

42. Chou, T. C., and Talalay, P. (1984) Adv. Enzyme Regul. 22, 27–5543. Wild, C., Dubay, J. W., Greenwell, T., Baird Jr., T., Oas, T. G., McDanal, C.,

Hunter, E., and Matthews, T. (1994) Proc. Natl. Acad. Sci. U. S. A. 91,12676–12680

44. Wei, X., Decker, J. M., Liu, H., Zhang, Z., Arani, R. B., Kilby, J. M., Saag, M. S.,Wu, X., Shaw, G. M., and Kappes, J. C. (2002) Antimicrob. Agents Che-mother. 46, 1896–1905

45. Root, M. J., Kay, M. S., and Kim, P. S. (2001) Science 291, 884–88846. Xu, L., Hue, S., Taylor, S., Ratcliffe, D., Workman, J. A., Jackson, S., Cane,

P. A., and Pillay, D. (2002) AIDS 16, 1684–168647. Derdeyn, C. A., Decker, J. M., Sfakianos, J. N., Wu, X., O’Brien, W. A., Ratner,

L., Kappes, J. C., Shaw, G. M., and Hunter, E. (2000) J. Virol. 74,8358–8367

48. Derdeyn, C. A., Decker, J. M., Sfakianos, J. N., Zhang, Z., O’Brien, W. A.,Ratner, L., Shaw, G. M., and Hunter, E. (2001) J. Virol. 75, 8605–8614

49. Reeves, J. D., Gallo, S. A., Ahmad, N., Miamidian, J. L., Harvey, P. E.,Sharron, M., Pohlmann, S., Sfakianos, J. N., Derdeyn, C. A., Blumenthal,R., Hunter, E., and Doms, R. W. (2002) Proc. Natl. Acad. Sci. U. S. A. 99,16249–16252

50. Ernst, J. T., Kutzki, O., Debnath, A. K., Jiang, S., Lu, H., and Hamilton, A. D.(2002) Angew. Chem. Int. Ed. Engl. 41, 278–281

51. Lawless, M. K., Barney, S., Guthrie, K. I., Bucy, T. B., Petteway, Jr. S. R., andG. Merutka, G. (1996) Biochemistry 35, 13697–13708

52. Ryu, J. R., Jin, B. S., Suh, M. J., Yoo, Y. S., Yoon, S. H., Woo, E. R., and Yu,Y. G. (1999) Biochem. Biophys. Res. Commun. 265, 625–629

53. Kliger, Y., Gallo, S. A., Peisajovich, S. G., Munoz-Barroso, I., Avkin, S., Blu-menthal, R., and Shai, Y. (2001) J. Biol. Chem. 276, 1391–1397

54. Greenberg, M. L., and Cammack, N. (2004) J. Antimicrob. Chemother. 54,333–340

55. Sista, P. R., Melby, T., Davison, D., Jin, L., Mosier, S., Mink, M., Nelson, E. L.,DeMasi, R., Cammack, N., Salgo, M. P., Matthews, T. J., and Greenberg,M. L. (2004) AIDS 18, 1787–1794

56. Rafalski, M., Lear, J. D., and DeGrado, W. F. (1990) Biochemistry 29,7917–7922

57. Salzwedel, K., West, J. T., and Hunter, E. (1999) J. Virol. 73, 2469–248058. Mobley, P. W., Pilpa, R., Brown, C., Waring, A. J., and Gordon, L. M. (2001)

AIDS Res. Hum. Retroviruses 17, 311–32759. Jiang, S., and Lin, K. (1995) Peptide Res. 8, 345–34860. Schibli, D. J., Montelaro, R. C., and Vogel, H. J. (2001) Biochemistry 40,

9570–957861. Hildinger, M., Dittmar, M. T., Schult-Dietrich, P., Fehse, B., Schnierle, B. S.,

Thaler, S., Stiegler, G., Welker, R., and von Laer, D. (2001) J. Virol. 75,3038–3042

62. Peisajovich, S. G., Gallo, S. A., Blumenthal, R., and Shai, Y. (2003) J. Biol.Chem. 278, 21012–21017

63. Furuta, R., Wild, C. T., Weng, Y., and Weiss, C. D. (1998) Nat. Struct. Biol. 5,276–279

64. Reeves, J. D., Miamidian, J. L., Biscone, M. J., Lee, F. H., Ahmad, N., Pierson,T. C., and Doms, R. W. (2004) J. Virol. 78, 5476–5485

65. Yuan, W., Craig, S., Si, Z., Farzan, M., and Sodroski, J. (2004) J. Virol. 78,5448–5457

66. Alam, S. M., Paleos, C. A., Liao, H. X., Scearce, R., Robinson, J., and Haynes,B. F. (2004) AIDS Res. Hum. Retroviruses 20, 836–845

T-20 Targets Multiple Sites in gp120/gp41 11273

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Shuwen Liu, Hong Lu, Jinkui Niu, Yujia Xu, Shuguang Wu and Shibo JiangInhibits HIV-1 Entry by Targeting Multiple Sites in gp41 and gp120

Different from the HIV Fusion Inhibitor C34, the Anti-HIV Drug Fuzeon (T-20)

doi: 10.1074/jbc.M411141200 originally published online January 7, 20052005, 280:11259-11273.J. Biol. Chem. 

  10.1074/jbc.M411141200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/280/12/11259.full.html#ref-list-1

This article cites 65 references, 26 of which can be accessed free at

by guest on May 15, 2018

http://ww

w.jbc.org/

Dow

nloaded from