The synthetic peptide derived from the NH2-terminal extracellular region of an orphan G protein-coupled receptor, GPR1, preferentially inhibits infection of X4 HIV-1.

Several G protein-coupled receptors (GPCRs) serve as co-receptors for entry of human immunodeficiency virus type 1 (HIV-1) into target cells. Here we report that a synthetic peptide derived from the NH2-terminal extracellular region of an orphan GPCR, GPR1 (GPR1ntP-(1-27); MEDLEETLFEEFENYSYDLDYYSLESC), inhibited infection of not only an HIV-1 variant that uses GPR1 as a co-receptor, but also X4, R5, and R5X4 viruses. Among these HIV-1 strains tested, viruses that can utilize CXCR4 as their co-receptors were preferentially inhibited. Inhibition of early steps in X4 virus replication was also detected in the primary human peripheral blood lymphocytes. GPR1ntP-(1-27) directly interacted with recombinant X4 envelope glycoprotein (rgp120). This interaction was neither inhibited nor enhanced by the soluble CD4 (sCD4) but inhibited by the anti-third variable (V3) loop-specific monoclonal antibody and heparin known to bind to the V3 loop. Although the conformational changes in gp120, including the V3 loop, have been reported to be required for its interaction with a co-receptor after binding of gp120 to CD4, it has also been reported that the V3 loop is already exposed on the surface of virions before interaction with CD4. We found that GPR1ntP-(1-27) blocked binding of virus to the cells, and this peptide equally bound to rgp120 in the presence or absence of sCD4. Because we detected the binding of GPR1ntP-(1-27) to the highly purified virions even in the absence of sCD4, GPR1ntP-(1-27) probably recognized the V3 loop exposed on the virions, and this interaction was responsible for the anti-HIV-1 activity of GPR1ntP-(1-27).

HIV-1 1 is the causative agent of AIDS (1,2). CD4 glycoprotein as a receptor and GPCRs as co-receptors are required for the entry of HIV-1, and the cell tropism of HIV-1 is mainly determined by its ability to use co-receptors expressed on the target cells (3). The major co-receptor for R5 viruses is CCR5, of which ligands are CC chemokines RANTES (regulated on activation normal T-cell expressed and secreted), MIP1␣, and MIP1␤. On the other hand, the major co-receptor for X4 viruses is CXCR4, of which ligand is the CXC chemokine SDF-1. Dualtropic viruses (R5X4 viruses) can efficiently utilize both of these co-receptors (4 -9).
In addition, primary human brain-derived fibroblast-like cells (BT-cells) are isolated from microvessel segments derived from autopsy of human brain tissue, are positive for CD4 and highly susceptible to HIV-1 variants, but are resistant to R5, X4, and R5X4 HIV-1 strains (10,11). BT-cells are thought to be originated from smooth muscle cells or pericytes. We have previously shown that the substitution of Pro to Ser at the Gly-Pro-Gly-Arg sequence in the V3 loop of envelope glycoprotein (gp120) is responsible for the BT-cell tropism of the HIV-1 variants (10,12). We have recently identified orphan GPCRs, GPR1 and RDC1, as co-receptors for the HIV-1 variants in vitro (13,14). Not only HIV-1 variants, but also several strains of HIV-2 and simian immunodeficiency virus have been shown to utilize GPR1 and RDC1 as their co-receptors (13)(14)(15).
During the course of our previous experiments to generate anti-GPR1 antibody (␣-GPR1) by immunizing rabbits with a synthetic peptide derived from the NH 2 -terminal extracellular region (NH 2 -ECR) of GPR1 (GPR1ntP-(1-27)), we found unexpectedly that not only ␣-GPR1 but also GPR1ntP-  was able to block GPR1-mediated infection of BT-cells with HIV-1 variants. Because it has been demonstrated that the NH 2 -ECRs of the co-receptors play an important role in the interaction with gp120 and a subsequent infection steps (16 -21), we speculated that GPR1ntP-(1-27) might inhibit interaction of gp120 (HIV-1 variants) with its co-receptor, GPR1, on target cell membrane.
In the present study, we tested anti-HIV-1 activity of GPR1ntP-  by several assays, including focal infectivity assay, HIV-1 p24 Gag detection assay, PCR assay, and syncytium formation assay. These assays all demonstrated an inhibitory activity of GPR1ntP-(1-27) against HIV-1 infection. Surprisingly, anti-HIV-1 activity of GPR1ntP-(1-27) was detected not only in HIV-1 variants that utilize GPR1 as a co-receptor, but also in genetically diverse isolates, including X4, R5, and R5X4 viruses. Our findings suggest that synthetic peptides derived from the NH 2 -ECR of GPR1 are novel candidates for the development of GPCR-based and peptide-based agents to inhibit HIV-1 infection.
HIV-1 Strains-The GUN-1WT strain is a clinical isolate and can infect both T-cell lines and macrophages, but not BT-cells (10,12,27). In contrast, the variant of GUN-1WT strain can infect the BT-cells and T-cell lines (10,12). Similar to the variant of GUN-1WT strain, GUN-1Ser strain, which had been prepared by the ligation and transfection of cloned HIV-1 (GUN-1WT) DNA fragments containing a mutation of proline to serine made by the site-directed mutagenesis at the GPGR sequence in the V3 loop, can infect BT-cells as well as T-cell lines (10,12). The GUN-1WT, GUN-1Ser, and IIIB (28) strains of HIV-1 were produced by persistently infected MOLT-4 cells. The BaL (29) and SF162 (30) strains of HIV-1 were produced by infected PBLs. The primary HIV-1 strains, GUN-4 and GUN-14, were isolated from Japanese hemophilia patients as described previously (12). To prepare virus samples, culture supernatants containing HIV-1 were harvested, centrifuged at a low speed, and passed through 0.45-m pore size cellulose acetate syringe filters (Gellman Science, Ann Arbor, MI). The virus samples were stored at Ϫ80°C until use.
Synthetic Peptides-All synthetic peptides used in this study were purchased from Sawady Technology (Tokyo, Japan). The purity of the peptides was more than 95%. The amino acid sequences of the synthetic peptides used in this study are shown in Table I.
HIV-1 Infectivity Assay-To determine HIV-1 infectivity, a focal infectivity assay was performed as described previously (33)(34)(35)(36)(37). Namely, NP-2/CD4 cells expressing one of GPCRs were seeded at 2 ϫ 10 4 per well into 24-well tissue culture plates (Corning, Acton, MA), infected with 200 l of the virus (1,000 -2,500 focus-forming units per ml) in the presence or absence of one of the synthetic peptides. After 2 h at 37°C, the cells were washed, fresh culture medium was added, and the cultures were incubated for 48 h at 37°C. The cells were fixed with methanol for 10 min at room temperature, rinsed with PBS, and then incubated with anti-HIV-1 sera from AIDS patients diluted 1:300 with 3% bovine serum albumin (w/v) in PBS (blocking buffer A) for 2 h at room temperature. After being washed with PBS, cells were incubated with horseradish peroxidase-labeled antibody to human IgG (Dako, Glostrup, Denmark) diluted 1:200 with blocking buffer A for 1 h at room temperature, washed with PBS, and rinsed with Tris-buffered saline. To develop the color, cells were reacted in the dark with a solution of aminoethyl carbazol (Sigma) and H 2 O 2 in acetate buffer (pH 5.0) for 20 min at room temperature, rinsed with water, air dried, and examined under a microscope. To determine the inhibitory activity of peptide to HIV-1 infection (inhibition (%)), the following formula was used: [(the number of foci with no peptide Ϫ the number of foci with peptide)/the number of foci with no peptide] ϫ 100%.
To examine the effect of GPR1ntP-(1-27) on the infectivity of HIV-1 primary isolates, the level of p24 Gag protein in culture supernatants was determined by enzyme-linked immunosorbent assay (ELISA) as described elsewhere (38). Azido-3Ј-deoxythymidine was purchased from Sigma.
PCR Assay to Detect Reverse-transcribed HIV-1 DNA-The virus preparation (IIIB strain) was treated with RNase-free DNase I (10 units/ml, Roche Applied Science) for 30 min at 37°C to remove contaminating viral DNA or cellular DNA. DNase I-treated IIIB virus was incubated with one of the synthetic peptides for 1 h at 37°C, and inoculated onto MOLT-4 cells (5 ϫ 10 5 ) or PBLs (5 ϫ 10 5 ) for 2 h at 37°C. The cells were washed, and fresh medium was added. After incubation for 20 h, the cells were washed and lysed with 10 mM Tris-HCl (pH 8.3) containing 1 mM EDTA, 0.45% Nonidet P-40 (Sigma), 0.45% Tween 20 (Sigma), and 0.2 mg/ml proteinase K (Sigma). The cell lysates were incubated for 2 h at 52°C, heated for 10 min at 96°C to inactivate proteinase K, and used as templates for the subsequent PCR analysis to detect the formation of reverse-transcribed HIV-1 DNA within the cells. PCR was performed with HIV-1 gag-specific primers, SK38 and SK39 (39) (nucleotide sequence: SK38, 5Ј-AAGGGGAAGT-GACATAGCAG-3Ј; SK39, 3Ј-GGACCAACAAGGTTTCTGTC-5Ј) in a PerkinElmer Life Sciences Cycler under the following conditions: 1 cycle at 95°C for 9 min, 30 cycles at 94°C for 1 min, 60°C for 45 s, 72°C for 1 min, and one cycle at 72°C for 5 min. The human ␤-globin gene primers, KM29 and KM38 (nucleotide sequence: KM29, 5Ј-GGTTGGC-CAATCTACTCCCAGG-3Ј; KM38, 5Ј-TGGTCTCCTTAAACCTGTC-TTG-3Ј) (40) were used as an internal control. DNA amplification was performed in the following conditions: 1 cycle at 95°C for 9 min, 30 cycles at 94°C for 1 min, 60°C for 45 s, 72°C for 1 min, and one cycle at 72°C for 5 min. The PCR products were visualized by electrophoresis through 2% agarose gels containing 0.5 g/ml ethidium bromide.
HIV-1 Cell Binding Assay-HIV-1-cell binding assay was essentially done according to the protocol described by Valenzuela et al. (41). MOLT-4 cells (6 ϫ 10 6 ) were incubated with HIV-1 (IIIB strain) (200 ng of p24 Gag ) for 1 h at 37°C in RPMI/FCS (final volume, 0.5 ml). When indicated, viral preparation was incubated with GPR1ntP-(1-27), GPR1ntP-(Y/A), 0.5␤, or heparin, while the cells were incubated with anti-CD4 mAb, NU-TH/I, before binding. After incubation of the cells with HIV-1, the cells were washed once with PBS containing 5 mM EDTA and twice with RPMI/FCS. Then, the cells were lysed with 150 l of lysis buffer (20 mM Tris-HCl (pH 7.6), 150 mM NaCl, 5 mM MgCl 2 , 0.2 mM phenylmethylsulfonyl fluoride, 5 g/ml of aprotinin, 0.5% Triton X-100, and 7 mM 2-mercaptoethanol), and the cell lysates were centrifuged at 1000 ϫ g for 2 min at 4°C. Then, the concentration of HIV-1 p24 Gag in the supernatants was determined by ELISA. The amount of p24 Gag in these cell extracts was expressed as the total amount of viral p24 Gag associated with the cells as described by Valenzuela et al. (41), because this value showed both viral p24 Gag bound to the surface of cells and viral p24 Gag that has entered the cells. To detect only HIV-1 that entered MOLT-4 cells, extracellular bound virus was removed. For this, cells were treated with 2.5 mg/ml of trypsin in PBS containing 5 mM EDTA for 40 s at room temperature after the cells were washed with PBS containing 5 mM EDTA. Then, the cells were washed twice with RPMI/FCS, lysed, and the concentration of intracellular p24 Gag was determined by ELISA. As compared with a T-cell line, CEM cells, which have been reported to be treated with trypsin for 5 min (41), MOLT-4 cells were treated with it for 40 s because they were highly sensitive to trypsin. Syncytium Formation Assay-MOLT-4/IIIB cells and C8166 cells were suspended in RPMI medium at 1 ϫ 10 5 cells per ml and 5 ϫ 10 5 cells per ml, respectively. Before addition of 50 l of C8166 cell suspension to each well of a U-bottom 96-well plate (Corning), MOLT-4/IIIB cells (50 l per well) were incubated with 100 l of RPMI medium containing one of the synthetic peptides or the medium alone for 1 h at 37°C. After incubation at 37°C for 17 h, the number of syncytia per well was counted by light microscope observation (42). The inhibitory activity of peptide to syncytium formation (inhibition (%)) was calculated using the following formula: [(the number of syncytia with no peptide Ϫ the number of syncytia with peptide)/the number of syncytia with no peptide] ϫ 100%.
Association of GPR1ntP-  with HIV-1 Virions-To concentrate virus particles (IIIB strain), the culture supernatant (10 ml) of MOLT-4/IIIB cells was layered onto a 20% (w/v) sucrose cushion (2.5 ml) and centrifuged at 100,000 ϫ g for 2 h at 4°C using a Hitachi SRP28 rotor (Hitachi Koki Co., Tokyo, Japan). The viral pellet was resuspended in 100 l of TNE buffer (20 mM Tris-HCl (pH 7.5), 100 mM NaCl, 1 mM EDTA), layered onto the top of a continuous sucrose gradient (20 -60% (w/v) sucrose in TNE) (43,44), and centrifuged at 100,000 ϫ g for 16 h at 4°C. Fractions (700 l) were collected from the top of the gradient, and the density of each fraction was determined. An aliquot from each fraction (50 l) was mixed with 50 l of TNE buffer containing 2% (v/v) Nonidet P-40 and 0.2% (w/v) bovine serum albumin, the mixture was incubated for 1 h at 56°C, and the HIV-1 p24 Gag protein was detected by dot blotting as described below. The fractions that contained HIV-1 virions were pooled, concentrated again by centrifugation at 100,000 ϫ g for 2 h at 4°C using an RP55S rotor (Hitachi Koki Co.), suspended in TNE buffer, and used for detection of the association of synthetic peptides (GPR1ntP-  or X4ntP-(1-28)) with the HIV-1 virions. The viral suspension (100 l) was incubated with one of peptides (5 g) for 1 h at 37°C, subjected to sucrose density gradient sedimentation, and then fractionated. The dot blotting was carried out to detect the presence of the HIV-1 p24 Gag protein and peptides in each fraction using anti-p24 Gag mAb and anti-peptide antibodies, respectively.
Dot Blotting-The HIV-1 lysates prepared after sucrose density gradient ultracentrifugation were blotted onto nitrocellulose membranes (Schleicher & Schuell, Dassel, Germany) using a Minifol-Microsample Filtration Manifold (Schleicher & Schuell). They were then blocked with 5% nonfat dry milk (w/v) in PBS containing 0.1% Tween 20 (v/v) (PBST) (blocking buffer B) for 2 h at 37°C. The membranes were then probed with ␣-p24 Gag mouse mAb diluted 1:1000 with blocking buffer B, or anti-peptide antibodies (␣-GPR1 or ␣-CXCR4; 2 g/ml in blocking buffer B) for 1 h at room temperature. After being washed three times with PBST, the membranes were incubated with a horseradish peroxidase-labeled antibody to mouse or rabbit IgG (DAKO) diluted 1:1000 with blocking buffer B for 1 h at room temperature, and then washed three times. Immunoreactive spots were detected by the enhanced chemiluminescence (ECL) system (Amersham Biosciences) according to the manufacturer's instruction.
Detection of Binding of rgp120 to GPR1 Peptides by ELISA-The recombinant envelope glycoprotein (rgp120) of HIV-1 (IIIB strain) produced in a baculovirus expression system, Ͼ90% pure, was obtained from the AIDS Reagent Project of the United Kingdom Medical Research Council (Potters Bar, United Kingdom). ELISA to detect peptide-rgp120 or sCD4-rgp120 interaction was based on the protocol as described (45), with a few modifications. Wells of a 96-well microtiter plate (Corning) were coated with GPR1ntP-(1-27) or its fragment or mutant peptides (0.015-5 M in PBS) for overnight at 4°C, washed twice with PBS, and blocked with blocking buffer A for 2 h at room temperature. After being washed with PBS, rgp120 (0.016 M) in PBS containing 1% (v/v) Triton X-100 was added, and the mixture was incubated for 2 h at room temperature and washed with PBST. To detect bound rgp120, anti-HIV-1 sera from AIDS patients diluted 1:2000 with 3% bovine serum albumin (w/v) in PBST (blocking buffer C) was added, and the mixture was incubated for 2 h at room temperature and washed with PBST. Then horseradish peroxidase-labeled antibody to human IgG diluted 1:5000 with blocking buffer C was added, and the mixture was incubated for 1 h at room temperature, washed with PBST, and rinsed with PBS. To develop color, 3,3Ј,5,5Ј-tetramethylbenzidine solution (Bio-Rad) as a substrate was added for 20 min at room temperature, and 1 N H 2 SO 4 was added to stop the reaction. Then the optical densities (A) at 450 nm were determined.
For detection of interaction of sCD4 with GPR1ntP-(1-27), microtiter wells were coated with GPR1ntP-(1-27) as described above. Then, sCD4 (0.016 M), which was obtained from the AIDS Reagent Project of the United Kingdom Medical Research Council, was added for 2 h at room temperature. Bound sCD4 was detected by ELISA as described above using antibody against sCD4 and horseradish peroxidase-labeled antibody to sheep IgG (Dako).
To examine the effect of ␣-GPR1 antibody on binding of GPR1ntP-(1-27) to rgp120, microtiter wells were coated with GPR1ntP-(1-27) as described above. Then purified ␣-GPR1 antibody or normal rabbit IgG (NRG) was added for 2 h at room temperature prior to addition of rgp120 (0.016 M), and then bound rgp120 was detected as described above. For detection of binding of GPR1ntP-(1-27) to rgp120-sCD4, rgp120 (0.016 M) was incubated with sCD4 (0.016 M) for 1 h at 37°C before incubation with GPR1ntP-(1-27) and then bound rgp120 was detected as described above. For detection of the binding of rgp120 to GPR1ntP-  or sCD4 in the presence of heparin (Wako), microtiter wells were coated with either GPR1ntP-(1-27) (1.0 M) or sCD4 (0.016 M) as described, and rgp120 (0.016 M), which had been preincubated with or without heparin (37°C for 1 h) was added. Bound rgp120 was detected as described above.
To examine the effect of GPR1ntP-(1-27) on binding of V3-specific neutralizing mouse mAb (0.5␤) to rgp120, a microtiter plate was coated with rgp120 (0.016 M) at 4°C for overnight, blocked, and treated with or without different concentrations of GPR1ntP-(1-27) or X4ntP-(1-28) (5-100 M) at room temperature for 2 h. After being washed with PBS, mAb 0.5␤ (ascites, 1:500 dilution) was added, and the plate was incubated at room temperature for 2 h, and washed. Captured 0.5␤ was detected as described above using a horseradish peroxidase-labeled antibody to mouse IgG (Dako).

Effects of Synthetic Peptides Derived from the NH 2 -ECR of
GPCRs on Cell-free HIV-1 Infection-We have previously established human glioma-derived, new HIV-1 infection indicator cell lines, which are transduced with both CD4 and co-receptors (NP-2/CD4/GPCRs), and showed that infectivity assay using these cell lines were highly reproducible and sensitive (13-15, 22, 23). As shown in Fig. 1, detection of HIV-1 antigen-positive cells by immunostaining after infection with several HIV-1 strains is dependent on both CD4 and the types of co-receptors expressed on NP-2 cells.
It is known that the CD4-binding domain (CD4-bd) and the V3 loop within gp120 play an important role for its interaction with target cells at the entry step of virus infection. To determine which region in rgp120 is involved in its binding to GPR1ntP-(1-27), we firstly examined the effect of sCD4 on the binding of rgp120 to GPR1ntP-(1-27) (Fig. 7C). When rgp120 had been incubated with sCD4 before addition to GPR1ntP-(1-27), sCD4 did not affect the binding of rgp120 to GPR1ntP- , suggesting that the CD4-bd in rgp120 might not be involved in the primary binding site for GPR1ntP- .
Because it has been reported that the soluble polyanion such as heparin binds to rgp120 at the V3 loop region (48 -50), we next examined the effect of heparin on binding of rgp120 to GPR1ntP-(1-27) (Fig. 7D). The rgp120 was incubated with different concentrations of heparin (37°C for 1 h) before addition to microtiter wells, which had been coated with either GPR1ntP-(1-27) or sCD4, and bound rgp120 was detected by ELISA. We found that heparin could inhibit association of -(1-27). NP-2/ CD4 cells expressing one of the GPCRs were infected with HIV-1 in the presence or absence of synthetic peptides derived from the NH 2 -ECR of GPCRs, and HIV-1antigen-positive cells were detected by focal infectivity assay as described under "Experimental Procedures." The target cell and inoculated HIV-1 strain are indicated at the top of graph and within the graph, respectively. The average number of foci in triplicate wells formed in the presence of peptides was counted, and the inhibition (%) was determined by the comparison with the average number of foci formed in the absence of peptides (ϳ500 foci/well). Four independent experiments gave similar results.  50 and IC 90 of the GPR1ntP-(1-27) for each HIV-1 strain were determined by focal infectivity assay using NP-2/CD4/GPCR cells in comparison with the number of foci induced by each HIV-1 strain without a peptide. As for GUN-4, GUN-14, and SF162 strains, IC 50 and IC 90 were determined by measuring the concentrations of p24 in the culture supernatants by ELISA.

FIG. 2. Inhibition of cell-free HIV-1 infection by GPR1ntP
b The primary HIV-1 strains isolated from Japanese hemophilia patients. c ND, not determined.
Anti-HIV-1 Activity of GPR1 NH 2 -terminal Region Peptide rgp120 with GPR1ntP-(1-27) in a concentration-dependent manner. In contrast, heparin had little effect on the binding of sCD4 to rgp120, which is in consistent with a previous report (49).
To examine whether the V3 loop in rgp120 is involved in binding to GPR1ntP-(1-27), we carried out binding assays using the mAb, 0.5␤, which has been reported to recognize the V3 loop of the IIIB strain (31). Microplate wells were coated with rgp120, incubated with 0.5␤, and bound 0.5␤ was detected by ELISA (Fig. 7E). Incubation of rgp120 with GPR1ntP-(1-27) before addition of 0.5␤ resulted in inhibition of 0.5␤ binding to rgp120 in a concentration-dependent manner, whereas X4ntP-(1-28) did not have much of an effect at any concentrations tested. Taken together, these results strongly suggested that the V3 loop plays an important role for binding of GPR1ntP-(1-27) to rgp120.
To determine which region of GPR1ntP-(1-27) was necessary for rgp120 binding, we next examined the binding activities of the fragment peptides to rgp120 (Fig. 7F). Among the fragment peptides, a small peptide, GPR1ntP-(15-27) significantly bound to rgp120, with a lower affinity compared with GPR1ntP- , which is consistent with our finding that GPR1ntP- (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27) inhibited HIV-1 infection at higher concentrations (Table III). In contrast, any apparent interaction of neither GPR1ntP-(1-13) nor GPR1ntP- (8 -20) with rgp120 was detected. The tyrosine mutant GPR1ntP-(Y/A) completely lost its ability to bind to rgp120, indicating that tyrosine residues in GPR1ntP-  play an important role in its binding to rgp120 as well as in its inhibition of HIV-1 infection (Table III). (1-27)-It has been reported that the V3 loop is exposed on the surface of intact virions (52,53). In addition, it has also been reported that anti-V3 loop mAbs inhibit the binding of HIV-1 to cells (41), suggesting that the V3 loop exposed on the surface of virion plays an important role for its binding to the target cells. Since our sucrose density gradient sedimentation assay indicated that GPR1ntP-  bound to intact virions (Fig. 6), and the ELISA assay revealed that there was a specific interaction between the V3 loop and GPR1ntP-(1-27) (Fig. 7), we examined whether GPR1ntP-(1-27) affected the binding of HIV-1 (IIIB strain) to MOLT-4 cells. For this, the cells were incubated with HIV-1 at 37°C for 1 h, washed three times, and lysed, and the total amount of viral p24 core protein associated with the cells was determined by ELISA. The effects of anti-human CD4 mAb (NU-TH/I), anti-V3 loop mAb (0.5␤), and heparin on the binding of HIV-1 to MOLT-4 cells at a concentration of 10 g/ml were firstly examined. At this concentration, we confirmed that these anti-HIV-1 reagents inhibited infection of MOLT-4 cells with HIV-1 (IIIB strain) by more than 95% (data not shown). Even when MOLT-4 cells were treated with NU-TH/I before inoculation of virus, 1670 pg of p24 Gag (58.8% of control) was still detected to be bound to the cells (Fig. 8A). To distinguish whether this p24 Gag was derived from the extracellular viruses bound to the surface of the cells or from intracellular viruses that entered the cells, the cells were treated with trypsin to remove the former fraction of viruses (Fig. 8B). The result showed that viral p24 Gag protein was hardly detectable after trypsin treatment, indicating that it was derived from the viruses bound to the surface of NU-TH/I-treated MOLT-4 cells.

Inhibition of HIV-1 Binding to MOLT-4 Cells by GPR1ntP-
In agreement with a previous report (41), our findings suggested that HIV-1 was associated with not only CD4, but also with sites other than CD4 on the surface of MOLT-4 cells.
In contrast to anti-CD4 mAb (NU-TH/I), incubation of viruses with 0.5␤ or heparin before inoculation of viruses blocked the binding of HIV-1 to MOLT-4 cells almost completely (Fig.  8A), suggesting that the V3 loop contributed for the binding of HIV-1 to MOLT-4 cells. Interestingly, the binding of HIV-1 to MOLT-4 cells was also inhibited to a similar extent by GPR1ntP-(1-27) at a concentration of 20 M: PCR assay showed the same concentration of GPR1ntP-(1-27) similarly inhibited the entry of HIV-1 (IIIB strain) to MOLT-4 cells (Fig.  4, C and D). A mutant peptide substituting all tyrosine residues in GPR1ntP-  to alanine, GPR1ntP-(Y/A), did not affect the binding of HIV-1 to cells at the same concentration. As expected, intracellular viral p24 Gag was hardly detectable (Fig.  8B) when viruses were treated with 0.5␤, heparin, or GPR1ntP-  prior to inoculation. GPR1ntP-(Y/A) again had hardly any effect on HIV-1 entry. Thus, these findings strongly suggested that the inhibition of binding of HIV-1 to MOLT-4 cells by GPR1ntP-(1-27) might be due to its interaction with the V3 loop exposed on the surface of intact virions.

DISCUSSION
In this study, we found that the 27-amino acid-long synthetic peptide derived from the NH 2 -ECR of GPR1 inhibits HIV-1 infection of human glioma-derived indicator cells (NP-2/CD4/ GPCRs), a T-cell line, and more importantly, primary PBLs, which are natural targets for HIV-1 infection. The inhibitory effect of GPR1ntP-(1-27) was observed not only in GUN-1Ser strain, which utilizes GPR1 as a co-receptor, but also in diverse HIV-1 strains, including X4 (IIIB), R5X4 (GUN-1WT, GUN-4, and , and R5 (BaL and SF162) (Figs. 2 and 3 and Table II). Among these HIV-1 strains tested, R5 viruses were relatively resistant to GPR1ntP-(1-27) as compared with the other strains.
Anti-HIV-1 Activity of GPR1 NH 2 -terminal Region Peptide cause it has been reported that 0.5␤ recognizes 24 amino acids (308 -331) of the V3 loop (31) and does not block binding of gp120 to CD4 (54), our data strongly suggested that the V3 loop plays an important role for binding of GPR1ntP-(1-27) to rgp120.
GPR1ntP-(1-27) inhibited binding of 0.5␤ to rgp120 by 50% at 25 M, although the peptide at higher concentrations up to 100 M showed only a limited decrease in rgp120 binding (Fig.  7E). As for the affinity of 0.5␤ for rgp120, an A 450 value of binding of GPR1ntP-(1-27) at 0.6 M to rgp120 was 0.32 (Fig.  7A), whereas almost a similar A 450 value was obtained when 0.027 M of 0.5␤ was used (data not shown), suggesting that the affinity of 0.5␤ for rgp120 is 22 times as high as that of GPR1ntP- . Thus, the incubation of 0.5␤ with rgp120 may displace GPR1ntP-(1-27) from rgp120 through a competitive Formation of HIV-1 DNA in MOLT-4 cells was detected by PCR as described above. MOLT-4 cells were infected with heat-inactivated (56°C, 30 min) HIV-1 (lane 9). Lane 10 is a mock-infected control. PCR amplification was performed without template DNA (lane 11). ␤-Globin DNA was amplified as a control to confirm the efficiency of the amplification of each sample. D, the intensity of DNA signal in each lane was measured and the relative intensity of gag signal was determined as described above. E, inhibition of HIV-1 infection to primary human PBLs by GPR1ntP- . Primary PBLs prepared from two independent donors (PBL#1 and PBL#2) were infected with HIV-1 (IIIB strain) in the presence or absence of GPR1ntP- . Formation of reversetranscribed HIV-1 gag DNA in PBLs was detected by PCR as described above. F, the intensity of DNA band in each lane was measured, and the relative intensity of gag signal was determined as described above. Three independent experiments gave similar results. -(1-27). A, detection of reverse-transcribed HIV-1 (IIIB strain) DNA by PCR. Human T-cell line, MOLT-4 cells, were inoculated with serially diluted HIV-1, lysed 20 h after inoculation, and examined by PCR using the primers specific for the HIV-1 gag region (lanes 1-3). Amplification of ␤-globin was performed as a control. B, the intensity of signal in each lane was measured by a densitometer. Then the relative intensity of gag signal was determined by dividing the intensity of viral gag DNA by that of ␤-globin DNA and plotted. C, dose-dependent inhibition of HIV-1 infection by GPR1ntP-(1-27). MOLT-4 cells were infected with HIV-1 in the presence of one of the synthetic peptides derived from the NH 2 -ECR of GPCRs (lanes 1-8). 5. Inhibition of syncytium formation by GPR1ntP-(1-27).

FIG. 4. Inhibition of HIV-1 infectivity to T-cell line and primary PBLs by GPR1ntP
HIV-1-negative C8166 (2.5 ϫ 10 4 ) cells were incubated with MOLT-4 (5 ϫ 10 3 ) cells chronically infected with HIV-1 (IIIB strain) in the presence or absence of one of the synthetic peptides for 17 h at 37°C in 96-well plates in 200 l of culture medium. At that time, the number of syncytia was counted by microscopic examination. The inhibitory effect of each peptide (%) was calculated in comparison with the average number of syncytia (ϳ200) of triplicate control wells (no peptide added). Experiments were repeated three times to confirm results.

TABLE III Effects of fragment and mutant GPR1 peptides on HIV-1 infection
a Amino acid sequence of each peptide is shown in Table I. b NP-2/CD4/CXCR4, NP-2/CD4/GPR1, and NP-2/CD4/CCR5 cells were used for infection with IIIB, GUN-1Ser, and BaL strains, respectively.
c IC 50 values (M) of peptides were determined by focal infectivity assay; the number of foci induced by each HIV-1 strain in the presence and absence of a peptide was counted. action of 0.5␤ for rgp120, and this may explain a slight and limited decrease in 0.5␤ binding to rgp120 even in the presence of high concentrations of GPR1ntP- .
The conformational changes in gp120, including the V3 loop, have been reported to be required for its interaction with a co-receptor after binding of gp120 to CD4 (21). The result of ELISA showed that GPR1ntP-(1-27) binds equally to rgp120 in the presence or absence of sCD4 (Fig. 7C), suggesting that GPR1ntP-(1-27) does not probably recognize the conformational changes in rgp120, especially in the V3 loop, induced by sCD4. This finding also suggested that the HIV-1 entry step at which gp120 interacts with a co-receptor after binding of envelope glycoprotein to CD4 is not a primary target for the anti-HIV-1 action of GPR1ntP- . Rather, we speculated that this peptide blocks an HIV-1 infection step before binding of gp120 to CD4 as described below.
It has also been reported that anti-V3 loop mAbs strongly bind to the intact virus particles, indicating that the V3 loop is already exposed on the surface of virions before interaction with CD4 receptor (52,53). Indeed, we found that GPR1ntP-  binds to the highly purified virions (IIIB strain) even in the absence of sCD4 (Fig. 6), suggesting that GPR1ntP-  directly binds to the V3 loop exposed on the surface of intact HIV-1 virions under the physiological conditions. In contrast to the mAbs against the V3 loop, it has also been reported that none of four mAbs against the CD4-binding domain (CD4-bd) bind to intact virions, including IIIB strain, indicating that the CD4-bd is not exposed on the surface of intact virions (52).
It has been reported that there are CD4-dependent and CD4-independent sites for HIV-1 binding on the surface of CD4 ϩ cells (41). The presence of cell surface molecule(s) other than CD4 on MOLT-4 cells responsible for HIV-1 binding was also suggested by our data, i.e. detection of viral p24 Gag bound to MOLT-4 cells even when the cells had been treated with anti-CD4 mAb before inoculation of virus (Fig. 8). In contrast to anti-CD4 mAb, anti-V3 loop mAb, heparin, and GPR1ntP- , expected to bind to the V3 loop exposed on the virions, almost completely inhibited the binding of HIV-1 to MOLT-4 cells (Fig. 8). Because CD4-bd on the surface of HIV-1 virions has not been reported to be exposed as described above, our results suggested that the binding of virus to the cells before interaction with CD4 is mediated by the V3 loop exposed on the surface of virions and that this initial CD4-independent virus binding step is an important target for GPR1ntP-(1-27) to inhibit HIV-1 infection.
Although GPR1 serves as a specific co-receptor for the HIV-1 variants, GPR1ntP-(1-27) blocked infection of diverse HIV-1 strains with different co-receptor usage. It may be explained in two ways. Firstly, the V3 loop is known to have the highest density of positively charged amino acids among all regions of gp120: it contains five to nine basic residues, and their distribution pattern within the V3 loop is well conserved among different strains (51,55). Because GPR1ntP-  shows the highest ratio of acidic amino acids (net negative charge is Ϫ10) among the peptides used in this study (Table I), electrostatic interaction might be responsible for GPR1ntP-  to bind to basic residues within V3 loop, and to inhibit HIV-1 infection. Because the charge of amino acids in the V3 loop of R5 viruses is known to be more acidic than that of X4 or R5X4 viruses (51,55), the relative resistance of R5 viruses (BaL and SF162) to GPR1ntP-(1-27) might be explained by the electrostatic repulsion between GPR1ntP-(1-27) and the V3 loop of the R5 viruses. Secondly, as the presence of a similarity in the conformation or antigenicity of the V3 loop has been suggested (52,56), GPR1ntP-(1-27) would bind to this conserved conformation. Namely, an mAb, 447-52D, has been reported to bind to the V3 loop exposed on the surface of different HIV-1 strains even in the absence of the linear epitope (52,56). Thus, recognition of the conserved conformation in the V3 loop by GPR1ntP-(1-27) might be important for its inhibitory activity to diverse HIV-1 strains.
The exact stoichiometry of GPR1ntP-(1-27) binding to target gp120 molecules in HIV-1 virions remains to be determined. It has been reported that gp120 is shed from infected cells and/or from virions soon after budding (57,58). Schneider et al. (59) reported that virus-free culture fluid contains at least 100-fold more gp120 than virions pelleted from the same volume of culture supernatant. Thus, not only virion-associated gp120 but also free gp120 might also bind to GPR1ntP-(1-27) and affect its anti-HIV-1 activity.
In this study, we showed that the mutant peptide (GPR1ntP-(Y/A)) substituting tyrosine to alanine does not inhibit HIV-1 infection (Table III) and has no activity for binding to rgp120 (Fig. 7F). Thus, tyrosine residues in GPR1ntP-(1-27) could contribute for interaction with rgp120 and play a crucial role in inhibition of HIV-1 infection. Farzan et al. (61) have shown that tyrosine residues from NH 2 -ECR of CCR5 contribute to the co-receptor activity through their sulfation, a post-translational modification resulting in the addition of a negatively charged sulfate. Although the sulfation of tyrosine residues in NH 2 -ECR of GPR1 remains to be elucidated, it is possible that the introduction of sulfated tyrosine residue to GPR1ntP-(1-27) will enhance its anti-HIV-1 activity as a result of an addition of negative charge to GPR1ntP- . With regard to GPR1ntP-(1-27) analogs, we previously reported that a hexapeptide derived from GPR1ntP-(1-27) (amino acid residues from 17 to 22: YDLDYY) has a weak anti-HIV-1 activity at concentrations around 500 M (62). The introduction of sulfation to tyrosine residues of this hexapeptide, however, did not significantly improve its inhibitory activity as compared with a nonsulfated hexapeptide (62). Because GPR1ntP- (15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27) binds to rgp120 less efficiently than GPR1ntP-(1-27), the amino acid residues of GPR1ntP-(1-27) from 1 to 14 (MEDLEETL-FEEFEN), which contain no tyrosine but several acidic amino acids, will also be necessary for the strong inhibitory activity of GPR1ntP-(1-27) against HIV-1 infection. Thus, the addition of acidic amino acid residues to sulfated tyrosine may be important to develop GPR1ntP-(1-27) analogs with enhanced anti-HIV-1 activities. This possibility is currently under investigation.
The entry stages, including initial attachment in HIV-1 replication cycle, are effective targets for the development of new antiretroviral therapies, because de novo infection of HIV-1 to humans can be prevented. This has already been shown by the clinical application of several antiretroviral agents. For instance, PRO 542 is a fusion protein of the gp120-binding domain of CD4 with immunoglobulin constant domains, and is expected to block the interaction between gp120 and the CD4 receptor (63); or T20 is a peptide (36 amino acids) that inhibits the HIV-1 fusion step through binding to the viral envelope protein gp41 (64,65). With a view to the antiviral potency of GPR1ntP- , the direct use of this peptide as a therapeutic agent will remain uncertain until studies of toxicity and pharmacokinetics are carried out. A critical concern with peptidebased antiviral therapy is its potential to elicit an antibody response to a peptide and make it ineffective (66). Because the amino acid sequence of GPR1ntP-(1-27) is derived from human, the antigenicity of GPR1ntP-(1-27) may be little or not at all in treated patients. The elucidation of the in vitro antiviral mechanism of GPR1ntP-  shown in this study will disclose valuable insights in the development of a new class of GPCR-based and peptide-based HIV-1 inhibitors.