3-Hydroxyphthalic Anhydride-Modified Chicken Ovalbumin Exhibits Potent and Broad Anti-HIV-1 Activity: a Potential Microbicide for Preventing Sexual Transmission of HIV-1

Abstract

Heterosexual transmission is the primary route by which women acquire human immunodeficiency virus (HIV)/AIDS. Thus, development of woman-controlled topical microbicides for prevention of sexual transmission of HIV is urgently needed. Here we report that 3-hydroxyphthalic anhydride-modified chicken ovalbumin (HP-OVA) exhibits potent antiviral activity against a broad spectrum of human immunodeficiency virus type 1 (HIV-1) isolates with different genotypes and biotypes. Its antiviral activity is correlated with the percentages of the chemically modified and unmodified lysines and arginines in OVA. HP-OVA inhibits HIV-1 fusion and entry through multiple mechanisms of action, including (i) blocking gp120 binding to CD4 and (ii) interfering with gp41 six-helix bundle formation. Because of the widespread availability and established safety profile of OVA, HP-OVA has good potential to be developed as an effective, safe, and affordable microbicide for prevention of HIV sexual transmission.

Since the first cases of AIDS were reported in 1981, more than 60 million people have been infected by human immunodeficiency virus (HIV) and around 3 million people die of the disease annually (70). Heterosexual transmission, the primary route by which women acquire HIV/AIDS, is the most rapidly growing risk factor in developed countries and is responsible for most HIV infections in developing countries (13). In the absence of an effective HIV vaccine in the foreseeable future, development of an effective, safe, and affordable microbicide to prevent the sexual transmission of HIV is urgently needed (64, 66, 69).

The narrow-spectrum microbicides previously studied are surfactants that are commonly used as contraceptive agents, such as nonoxynol-9 and glyminox (Savvy) (14, 18, 19). These microbicides displayed potent nonspecific virucidal activity in vitro, but they failed to protect against, or even enhanced, human immunodeficiency virus type 1 (HIV-1) infection because of their vaginal toxicity (71, 72). The extended-spectrum microbicides, represented by cellulose sulfate (Ushercell) (73), carrageenan (Carraguard) (58), and naphthalene sulfonate (PRO 2000) (31, 42, 67), consist of polyanionic polymers as active agents. However, none of these microbicides has demonstrated effectiveness in preventing the sexual transmission of HIV (59, 65, 68, 73). One of the main reasons contributing to the failure of these microbicides in clinical trials may be their low efficacy against primary R5 HIV-1 isolates, the most commonly sexually transmitted viruses (8, 63). In contrast, we found that cellulose acetate phthalate (CAP), an “inactive” pharmaceutical excipient, exhibited potent antiviral activity against a broad spectrum of primary HIV-1 isolates, including the R5 and X4 viruses (15, 37, 39). At the same time, however, CAP was found to be unstable in aqueous solution, making it difficult to formulate a vaginal gel.

In the mid-1990s, our group demonstrated that bovine ß-lactoglobulin, a protein present in milk and whey, modified by 3-hydroxyphthalic anhydride (3HP-ß-LG) displayed broad antiviral activities against infection by HIV-1, HIV-2, and simian immunodeficiency virus (SIV), as well as other viral pathogens causing sexually transmitted diseases (STD), such as herpes simplex virus type 1 (HSV-1) and HSV-2 (46-49). 3HP-ß-LG is highly stable in aqueous solution for long-term storage at room temperature and elevated temperatures (47). However, the outbreak of bovine spongiform encephalopathy (BSE) in Europe raised a safety concern with regard to developing bovine proteins for medical use, resulting in discontinuation of further development of 3HP-ß-LG as a microbicide.

In the present study, we substituted bovine ß-lactoglobulin with chicken ovalbumin (OVA) to develop a new microbicide, since OVA is one of most abundant food proteins consumed by people all over the world and, therefore, generally recognized as safe. We found that, like 3HP-ß-LG, 3-hydroxyphthalic anhydride-modified OVA (HP-OVA) was highly effective in inhibiting infection by a broad spectrum of primary HIV-1 isolates of different clades, including both R5 and X4 viruses, by blocking HIV-1 entry through multiple mechanisms of action. This suggests that HP-OVA can be developed further as an effective, safe, and affordable microbicide for preventing HIV sexual transmission.

MATERIALS AND METHODS

Reagents.

CHO-WT cells, MT-2 cells, TZM-b1 cells, HIV-1_IIIB-infected H9 (H9/HIV-1_IIIB) cells, laboratory-adapted and primary HIV-1 strains, zidovudine (AZT)- and T20-resistant HIV-1 strains, HIV-2_ROD, simian-human immunodeficiency virus (SHIV), SIV_mac251 32H, anti-p24 monoclonal antibody (MAb) (183-12H-5C), AZT, AMD-3100, maraviroc, T20, gp120 from HIV-1_BaL, and HIV Ig were obtained from the National Institutes of Health AIDS Research and Reference Reagent Program. VK2/E6E7 cells, Ect1/E6E7 cells, and End1/E6E7 cells were purchased from the American Type Culture Collection (ATCC) (Manassas, VA). Lymphoid cell line CEMx174 5.25M7, expressing CD4 and both coreceptors CCR5 and CXCR4 (20), was kindly provided by C. Cheng-Mayer. Chicken ovalbumin (OVA; lyophilized powder), 3-hydroxyphthalic anhydride (HP), mouse anti-CD4 monoclonal antibody Q4120, 2,4,6-trinitrobenzenesulfonic acid (TNBS), gelatin, bovine serum albumin (BSA), phytohemagglutinin (PHA), interleukin-2 (IL-2), and XTT [2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-5-(phenylamino) carbonyl-2H-tetrazolium hydroxide] were purchased from Sigma (St. Louis, MO). p-Hydroxyphenylglyoxal (p-HPG) was purchased from Fisher Scientific (Valley Park, VA). Calcein-acetoxymethyl ester (calcein-AM) was purchased from Molecular Probes (Eugene, OR). Recombinant soluble CD4 (sCD4), biotinylated sCD4, and gp120 were obtained from Immunodiagnostics (Woburn, MA). Peptides N36, C34, and fluorescein isothiocyanate (FITC)-conjugated C34 (C34-FITC) (3, 41) were synthesized using a standard solid-phase 9-fluorenylmethoxy carbonyl (Fmoc) method at the MicroChemistry Laboratory of the New York Blood Center and were purified by high-performance liquid chromatography (HPLC). ADS-J1 was purchased from ComGenex (Budapest, Hungary). Mouse MAb NC-1, specific for the gp41 six-helix bundle, was prepared and characterized as previously described (26). 3HP-ß-LG was prepared in our lab as previously described (46, 49).

Chemical modification of OVA.

The modified protein HP-OVA was prepared using a previously described method (46, 49). Briefly, OVA (final concentration, 20 mg/ml in 0.1 M phosphate) was treated with HP (1.19 M in dimethylformamide) by the addition of five aliquots at 12-min intervals. The reaction conditions for HP-OVA modification were changed by adding different concentrations of HP from 2.5 mM to 60 mM or modulating the reaction pH values from 3.0 to 10.0. The mixture was kept for 1 h at room temperature and then extensively dialyzed against phosphate-buffered saline (PBS) and filtered through 0.45-μm syringe filters (Gelman Sciences, Ann Arbor, MI).

Protein concentrations were determined using a bicinchoninic acid (BCA) protein assay reagent kit (Pierce, Rockford, IL). To quantitate the lysine residues in OVA and HP-OVA, TNBS treatment was applied as previously described (12). Briefly, 25 μl HP-OVA or unmodified OVA (90 μM) was incubated with 25 μl 0.1 M Na₂B₄O₇ for 5 min at room temperature. Then, 10 μl TNBS was added at room temperature. After 5 min, 100 μl stop solution (0.1 M NaH₂PO₄ and 1.5 mM Na₂SO₃) was added to terminate the reaction. The absorbance at 420 nm (A₄₂₀) was measured using an enzyme-linked immunosorbent assay (ELISA) reader (Ultra 384; Tecan, Research Triangle Park, NC). The number of arginine residues in the protein preparations was determined using the p-HPG method as previously described (10, 77). In brief, 90 μl HP-OVA or OVA (90 μM) in 0.1 M sodium phosphate (pH 9.0) was incubated with 10 μl 50 mM p-HPG for 90 min at room temperature in the dark. The absorbance at 340 nm (A₃₄₀) was measured. The concentration of lysine or arginine was determined from calibration curves relating A₄₂₀ or A₃₄₀ values to lysine or arginine concentrations in standards, respectively.

HIV-1 Env-mediated cell-cell fusion assay.

Inhibition of fusion between MT-2 cells and CHO-WT cells, which were stably transfected with HIV-1 Env gp160, was determined using a syncytium formation assay as previously described (35). Briefly, CHO-WT cells at 2 × 10⁶/ml were incubated with MT-2 cells at 1 × 10⁶/ml in the absence or presence of HP-OVA at graded concentrations at 37°C for 48 h. The syncytia were counted under an inverted microscope. A dye transfer assay was performed to determine the ability of HP-OVA to inhibit fusion between MT-2 cells and H9/HIV-1_IIIB cells as previously described (24, 30). Briefly, 2 × 10⁵ H9/HIV-1_IIIB cells/ml were labeled with 1 mM calcein-AM, a fluorescent reagent, at 37°C for 30 min and then incubated with 1 × 10⁶ MT-2 cells/ml at 37°C for another 2 h in the presence or absence of HP-OVA. The fused and unfused calcein-labeled HIV-1-infected cells were counted under an inverted fluorescence microscope (Zeiss, Germany). In both assays, ADS-J1, a small-molecule anti-HIV-1 compound that inhibits HIV-1 entry by blocking the gp41 six-helix bundle formation (75), and unmodified OVA were used as positive and negative controls, respectively. The 50% inhibitory concentration (IC₅₀) values were calculated using CalcuSyn software (5), kindly provided by T. C. Chou of the Sloan-Kettering Cancer Center (New York, NY).

Cell-to-cell transmission of HIV-1 R5 virus.

Peripheral blood mononuclear cells (PBMCs) were isolated from the blood of healthy donors at the New York Blood Center by standard density gradient centrifugation using Histopaque-1077 (Sigma) and were then plated in 75-cm² plastic flasks and incubated at 37°C for 2 h. The nonadherent cells were collected and resuspended at 5 × 10⁶/ml in RPMI 1640 medium containing 10% fetal bovine serum (FBS), 5 μg/ml of phytohemagglutinin (PHA), and 100 U/ml of interleukin-2, followed by incubation at 37°C for 3 days. The PHA/IL-2-stimulated PBMCs were infected with HIV-1_BaL (R5 virus) at a multiplicity of infection of 0.01 for 7 days as previously described (39). After three washes with culture medium to remove free viral particles, 50 μl of HIV-1-infected PBMCs (1 × 10⁵/ml) was incubated with 50 μl of HP-OVA at graded concentrations at 37°C for 30 min. Then, 100 μl of CEMx174 5.25M7 cells (2 × 10⁵/ml) was added and cocultured at 37°C for 3 days. The cells were collected and lysed for analysis of luciferase activity using a luciferase assay kit (Promega, Madison, WI). The luciferase activity was measured with an Ultra 384 luminometer (Tecan).

Measurement of HIV-1, HIV-2, SHIV, and SIV infectivities.

The inhibitory activities of HP-OVA against infection with laboratory-adapted HIV-1 (IIIB, MN, and RF) and AZT-resistant strains were determined as previously described (27, 30). In brief, MT-2 cells (1 × 10⁵/ml) were infected with an HIV strain at 100 50% tissue culture infective doses (TCID₅₀) in the presence or absence of HP-OVA at graded concentrations at 37°C overnight. Then, the culture supernatants were supplied with fresh medium. On the fourth day postinfection, 100 μl of culture supernatants was collected and mixed with equal volumes of 5% Triton X-100. Then, those virus lysates were assayed for p24 antigen by ELISA (30). Briefly, wells of 96-well polystyrene plates (Immulon 1B; Dynex Technology, Chantilly, VA) were coated with 5 μg/ml HIV Ig in 0.85 M carbonate-bicarbonate buffer (pH 9.6) at 4°C overnight, followed by washing with PBS-T buffer (0.01 M PBS containing 0.05% Tween 20) and blocking with PBS containing 1% dry fat-free milk (Bio-Rad Inc., Hercules, CA). Virus lysates were added to the wells, and the solution was incubated at 37°C for 1 h. After extensive washes, anti-p24 MAb (183-12H-5C), biotin-labeled goat anti-mouse IgG (Santa Cruz Biotechnology, Santa Cruz, CA), streptavidin-labeled horseradish peroxidase (SA-HRP) (Zymed, South San Francisco, CA), and 3,3′,5,5′-tetramethylbenzidine (TMB) (Sigma) were added sequentially. Reactions were terminated by addition of 1 N H₂SO₄. The absorbance at 450 nm was measured with an ELISA reader (Tecan).

For assessment of the inhibitory activity of HP-OVA against infection by T20-resistant strains, HIV-2_ROD, SHIV_SF33A, SHIV_SF162P3, and SIV_mac251 32H, the viruses at 100 TCID₅₀ were incubated with HP-OVA at graded concentrations at 37°C for 30 min prior to the addition of the mixtures to TZM-bl cells (1 × 10⁵/ml). At 72 h postinfection, the cells were collected, washed, and lysed with the lysing reagent included in the luciferase kit (Promega, Madison, WI). Aliquots of cell lysates were transferred to 96-well, flat-bottomed luminometer plates (Costar), followed by the addition of luciferase substrate (Promega). The luciferase activity was measured with an Ultra 384 luminometer (Tecan).

The inhibitory activity of HP-OVA against infection by HIV-1_BaL and primary HIV-1 isolates was determined as previously described (30). Briefly, PBMCs, which were isolated and stimulated with PHA and IL-2 as described above, were infected with the corresponding primary HIV-1 isolates at 100 TCID₅₀ in the absence or presence of HP-OVA at graded concentrations. Culture media were changed every 3 days. The supernatants were collected 7 days postinfection and tested for p24 antigen by ELISA as described above. The percentages of inhibition as determined by p24 production and IC₅₀s were calculated as described above.

Time-of-addition assay.

For testing the laboratory-adapted X4 virus, HIV-1_IIIB at 100 TCID₅₀ was incubated with MT-2 cells (1 × 10⁵/ml) at 37°C for 0, 0.5, 1, 2, 4, 6, and 8 h before the addition of HP-OVA (0.2 μM). AZT (0.1 μM), 3HP-ß-LG (0.2 μM), AMD-3100 (0.2 μM), and T20 (0.5 μM) were used as controls. After another 2 h of culture, the cells were washed to remove the free virus and compounds, and fresh medium was added. On the fourth day postinfection, 100 μl of culture supernatants was collected, mixed with equal volumes of 5% Triton X-100, and assayed for p24 antigen as described above (30, 37, 78).

For testing the laboratory-adapted R5 virus, HIV-1_BaL at 100 TCID₅₀ was incubated with TZM-bl cells (1 × 10⁵/ml) at 37°C for 0, 0.5, 1, 2, 4, 6, and 8 h before the addition of 4.5 μM HP-OVA. AZT at 2 μM, 3HP-ß-LG at 4.5 μM, maraviroc at 0.2 μM, and T20 at 2 μM were used as controls. After culture for an additional 2 h, the cells were washed to remove the free virus and compounds and supplied with fresh medium. On the third day postinfection, the cells were collected, washed, and lysed with the lysing reagent included in the luciferase kit (Promega). Aliquots of cell lysates were transferred to 96-well, flat-bottomed luminometer plates (Costar), followed by the addition of luciferase substrate (Promega). The luciferase activity was measured with an Ultra 384 luminometer (Tecan).

Inhibition of binding of soluble CD4 and Env gp120 from HIV-1_IIIB and HIV-1_BaL.

The interaction between sCD4 and gp120 was determined as described before (49, 53). Briefly, wells of 96-well polystyrene plates were coated with 5 μg/ml gp120 from HIV-1_IIIB or HIV-1_BaL in 0.1 M Tris buffer (pH 8.8) at 4°C overnight, followed by washing with TS buffer (0.14 M NaCl, 0.01 M Tris, pH 7.0). Then, the wells were blocked for 1 h at room temperature with 1 mg/ml bovine serum albumin (BSA) and 0.1 mg/ml gelatin in TS buffer. Biotinylated sCD4 (1 μg/ml) was preincubated with HP-OVA at various concentrations in PBS containing 100 μg/ml BSA for 18 h at 4°C. The mixture, SA-HRP, TMB, and 1 N H₂SO₄ were added sequentially. The A₄₅₀ was measured using an ELISA reader, and the IC₅₀s were calculated as described above.

Inhibition of binding of soluble CD4 and mouse anti-CD4 monoclonal antibody Q4120 (49).

Wells of 96-well polystyrene plates were coated with 5 μg/ml sCD4 in 0.1 M Tris buffer (pH 8.8) at 4°C overnight, followed by washing with TS buffer, and blocked for 1 h at room temperature as described above. Q4120 (1 μg/ml) was preincubated with HP-OVA at various concentrations in PBS containing 100 μg/ml BSA at room temperature overnight. The mixture, biotin-labeled goat anti-mouse IgG (Sigma), SA-HRP, TMB, and 1 N H₂SO₄ were added sequentially. The A₄₅₀ was measured using an ELISA reader, and the IC₅₀s were calculated as described above.

SPR assay.

The interaction between HP-OVA and gp120 or sCD4 was characterized by employing a surface plasmon resonance (SPR) assay as described previously (21, 40, 75). Briefly, gp120 derived from HIV-1_IIIB (2 μg/ml), HIV-1_BaL (2 μg/ml), or sCD4 (2 μg/ml) was immobilized onto a CM3 sensor chip by use of an amine coupling protocol, and the unreacted sites were blocked with 1 M Tris-HCl (pH 8.5). The kinetics of the binding of HP-OVA to the immobilized gp120 or sCD4 was determined by dose-dependent binding of HP-OVA to gp120 or sCD4. The association reaction was initiated by injecting various concentrations of HP-OVA (5, 2.5, 1.25, 0.625, 0.313, and 0.156 μM) at a flow rate of 30 μl/min. The dissociation reaction was carried out by washing with HBS-EP (10 mM HEPES, 150 mM NaCl, 3 mM EDTA, 0.005% P20). At the end of each cycle, the sensor chip surface was regenerated by washing the surface with 10 mM glycine-HCl (pH 3.0) for 30 s, followed by washing with 0.05% sodium dodecyl sulfate (SDS) for 30 s. The resulting sensograms (plots of changes in response units [RU] on the surface as a function of time) were analyzed using BIAeval 3.0 software. The curves were fitted using a 1:1 binding model.

Inhibition of gp41 six-helix bundle formation.

A sandwich ELISA was used as previously described (30, 41) to determine the inhibition of the gp41 six-helix bundle formation in the presence of HP-OVA at graded concentrations. Briefly, 2 μM peptide N36 was preincubated with HP-OVA at various concentrations at 37°C for 30 min, followed by the addition of 2 μM C34. After incubation at 37°C for 30 min, the mixtures were added to wells of a 96-well polystyrene plate (Corning, NY), which were precoated with IgG (2 μg/ml) purified from rabbit antisera directed against the N36/C34 mixture. Then, MAb NC-1, biotin-labeled goat anti-mouse IgG (Sigma), SA-HRP, TMB, and 1 N H₂SO₄ were added sequentially. The A₄₅₀ was measured using an ELISA reader, and the IC₅₀s were calculated as described above.

Fluorescence native polyacrylamide gel electrophoresis (FN-PAGE) was carried out as previously described (30). Briefly, peptide N36 (40 μM in PBS) was incubated with HP-OVA at 15, 7.5, 3.75, and 1.875 μM at 37°C for 30 min, and the FITC-conjugated peptide C34 (C34-FITC; 40 μM in PBS) was added. After incubation at 37°C for another 30 min, the mixture was diluted in Tris-glycine native sample buffer (Invitrogen, Carlsbad, CA) and then loaded onto 10- by 1.0-cm precast Tris-glycine gels (18%; Invitrogen, Carlsbad, CA) at 25 μl/well. Gel electrophoresis was carried out at a 125-V constant voltage at room temperature for 2 h in Tris-glycine native running buffer. Immediately after electrophoresis, fluorescence bands in the gel were imaged by using a FluorChem 8800 imaging system (Alpha Innotech Corp., San Leandro, CA) with a transillumination UV light source at an excitation wavelength of 302 nm and a fluorescence filter at an emission wavelength of 520 nm.

Cytotoxicity assay.

The in vitro cytotoxicities of HP-OVA to virus target cells (MT-2 cells, TZM-b1 cells, and PHA/IL-2-stimulated PBMCs) and normal human tissue cells (VK2/E6E7, Ect1/E6E7, and End1/E6E7 cells) were measured using the XTT colorimetric assay as previously described (30). Briefly, 100 μl of HP-OVA at graded concentrations was added to equal volumes of cells (5 × 10⁵/ml) in wells of 96-well plates. After incubation at 37°C for 4 days, 50 μl of XTT solution (1 mg/ml) containing 0.02 μM phenazine methosulfate (PMS) was added. After 4 h, the absorbance at 450 nm was measured with an ELISA reader. The 50% cytotoxicity concentrations (CC₅₀) were calculated using CalcuSyn software (6).

Assay for spontaneous and PHA-stimulated PBMC proliferation.

The proliferation of PBMCs was determined as described by Lewis et al. and Boscolo et al. (1, 34). In brief, PBMCs (5 × 10⁵/ml) were isolated from blood of healthy donors as described above and were not stimulated or were stimulated with PHA (20 μg/ml) in the presence or absence of HP-OVA at various concentrations. After culture at 37°C for 72 h, the cell proliferation was measured by XTT colorimetric assay as described above. The absorbance at 450 nm was measured with an ELISA reader.

IFN-γ ELISPOT assay.

An enzyme-linked immunospot (ELISPOT) assay was performed using an ELISPOT Pro kit from Mabtech (Mariemont, OH) following the manufacturer's protocols. Briefly, wells of ELISPOT plates were coated with 15 μg/ml of an anti-gamma interferon (anti-IFN-γ) MAb (1-D1K) at 4°C overnight and blocked with RPMI 1640 containing 10% FBS at room temperature for 2 h. PBMCs (2 × 10⁵/ml) isolated as described above were stimulated with PHA (5 μg/ml) or not stimulated and were then added to the anti-IFN-γ MAb-coated wells, respectively, in the presence or absence of HP-OVA at graded concentrations at 37°C for 48 h. After extensive washes with PBS between incubations, biotinylated anti-human IFN-γ MAb (7-B6-1-biotin, 1 μg/ml), SA-HRP, and substrate solution were added. The spots of IFN-γ-producing cells were counted with an ELISPOT reader system (Carl Zeiss, Germany).

RESULTS

The percentages of the HP-modified and unmodified lysine and arginine residues were correlated with the antiviral activity of HP-OVA.

OVA has a molecular mass of about 43 kDa and consists of 385 amino acid residues, including 20 lysine residues and 15 arginine residues. Our previous studies have shown that the number of lysine residues in ß-LG modified by acid anhydride is a key determinant of the antiviral activity of 3HP-ß-LG (49). Using similar approaches, we treated OVA with 3-hydroxyphthalic anhydride (HP) at final concentrations of 2.5 to 60 mM in 0.1 M phosphate (pH 8.0) and thus determined the percentages of the HP-modified and unmodified positively charged residues in HP-OVA. As shown in Table 1, with increasing concentrations of HP, more lysine and arginine residues were modified, leading in turn to stronger anti-HIV-1 activity. Then, we determined the potential effect of pH on the modification of the positively charged residues in OVA by using a fixed concentration of HP (40 mM) under variable pH levels (3.0 to 10.0) of the reaction system. As shown in Table 2, the percentage of modified lysine and arginine residues and the resultant anti-HIV-1 activity increased with increasing pH from 3.0 to 7.0, reaching a plateau at pH 7.0. To avoid the borderline effect, we selected a higher pH, i.e., pH 8.0, as the optimal condition for preparation of HP-OVA for the subsequent studies. Under this condition, the resultant HP-OVA had an average molecular mass of 44.58 kDa, as determined by SDS-PAGE, with 99.86% and 89.26% of the lysine and arginine residues modified by HP, respectively.

TABLE 1.

Effects of concentration of HP in the reaction system on the percentage of modified residues and anti-HIV-1 activity of HP-OVA^a

Concn of HP (mM)	% of modified residuesb		Anti-HIV-1IIIB activity		Anti-HIV-1BaL activity
	Lysine	Arginine	IC50 (μM)	IC90 (μM)	IC50 (μM)	IC90 (μM)
2.5	63.07	43.93	3.570 ± 1.170	>4.48	>4.48	>4.48
5.0	89.52	58.63	0.251 ± 0.073	1.033 ± 0.160	0.207 ± 0.034	2.618 ± 1.946
10.0	96.44	59.12	0.099 ± 0.025	0.364 ± 0.100	0.106 ± 0.053	0.988 ± 0.836
20.0	98.86	71.12	0.031 ± 0.001	0.095 ± 0.025	0.019 ± 0.003	0.231 ± 0.034
40.0	99.86	89.26	0.008 ± 0.000	0.027 ± 0.003	0.007 ± 0.001	0.044 ± 0.019
60.0	99.32	92.36	0.006 ± 0.001	0.019 ± 0.005	0.006 ± 0.002	0.043 ± 0.026

^aEach sample was tested in triplicate, and each experiment was repeated twice. IC₅₀ and IC₉₀ data are presented as means ± standard deviations.

^bThe data are averages from two assays.

TABLE 2.

Effects of pH value of the reaction system on the percentage of modified residues and anti-HIV-1 activity of HP-OVA^a

pH value of reaction system	% of modified residuesb		Anti-HIV-1IIIB activity		Anti-HIV-1BaL activity
	Lysine	Arginine	IC50 (μM)	IC90 (μM)	IC50 (μM)	IC90 (μM)
3.0	0.00	22.72	>4.48	>4.48	>4.48	>4.48
4.0	13.41	47.98	>4.48	>4.48	>4.48	>4.48
5.0	51.16	62.11	0.352 ± 0.040	1.187 ± 0.160	0.292 ± 0.060	>4.48
6.0	96.35	78.78	0.074 ± 0.026	0.208 ± 0.033	0.066 ± 0.022	1.712 ± 0.969
7.0	98.61	79.67	0.013 ± 0.002	0.036 ± 0.007	0.016 ± 0.003	0.139 ± 0.134
8.0	98.09	78.78	0.010 ± 0.002	0.025 ± 0.003	0.014 ± 0.001	0.069 ± 0.005
9.0	98.42	89.76	0.009 ± 0.001	0.029 ± 0.011	0.008 ± 0.002	0.044 ± 0.009
10.0	98.27	89.18	0.010 ± 0.001	0.023 ± 0.000	0.009 ± 0.000	0.045 ± 0.005

^aTwo independent experiments were performed in triplicate. IC₅₀ and IC₉₀ data are presented as means ± standard deviations.

^bThe data are averages from two assays.

HP-OVA inhibited infection by different laboratory-adapted and primary HIV-1 strains as well as HIV-2, SHIV, and SIV.

We determined the inhibitory activities of HP-OVA against infection by a number of laboratory-adapted HIV-1 strains and a panel of primary HIV-1 isolates in MT-2 cells, TZM-bl cells, and PBMCs as described in Materials and Methods. The unmodified OVA was used as a control. As shown in Table 3, HP-OVA showed highly potent inhibitory activities against laboratory-adapted HIV-1 X4 and R5 strains, including IIIB, MN, RF, and BaL, with IC₅₀s at a low nanomolar level. Similarly, HP-OVA could significantly inhibit infection by the representative primary HIV-1 isolates of clades A to G and group O (R5 or X4R5), with IC₅₀s ranging from 0.011 to 0.578 μM. Notably, HP-OVA was also effective in inhibiting infection by AZT-R, an HIV-1 strain resistant to reverse transcriptase inhibitors, such as AZT, and by NL4-3_(36G)V38A, NL4-3_{(36G)V38E/N42T}, and NL4-3_{(36G)V38E/N42S}, which are resistant to the peptide HIV fusion inhibitors T20 and T1249 (9). Interestingly, HP-OVA also had potent inhibitory effects on infection with HIV-2_ROD, SHIV_SF33A, SHIV_SF162P3, and SIV_mac251 32H viruses, with IC₅₀s in the range of 0.033 to 1.420 μM. The control protein, unmodified OVA, had no inhibitory activity against any of the above-mentioned viruses, even at concentrations up to 25 μM (data not shown). Taken together, these results suggest that HP-OVA has a broad spectrum of antiviral activity.

TABLE 3.

Inhibitory activities of HP-OVA on infection by laboratory-adapted and primary HIV-1 strains and HIV-2, SHIV, and SIV strains^a

Virus strain	HP-OVA		T20b
	IC50 (μM)	IC90 (μM)	IC50 (μM)	IC90 (μM)
Laboratory-adapted HIV-1 strains
IIIB (clade B, X4)	0.008 ± 0.000	0.027 ± 0.003	0.097 ± 0.005	0.293 ± 0.042
MN (clade B, X4)	0.019 ± 0.006	0.077 ± 0.030	0.047 ± 0.008	0.148 ± 0.025
RF (clade B, X4R5)	0.009 ± 0.001	0.026 ± 0.003	0.074 ± 0.008	0.263 ± 0.122
BaL (clade B, R5)	0.007 ± 0.001	0.044 ± 0.019	0.097 ± 0.044	0.209 ± 0.077
Primary HIV-1 strains
UG94103 (clade A, X4R5)	0.108 ± 0.031	0.289 ± 0.046	0.023 ± 0.003	0.069 ± 0.015
92US657 (clade B, R5)	0.067 ± 0.022	1.826 ± 1.440	0.120 ± 0.032	0.210 ± 0.052
93IN101 (clade C, R5)	0.044 ± 0.013	0.226 ± 0.200	NDd	ND
92TH009 (clade A/E, R5)	0.011 ± 0.004	0.032 ± 0.007	0.097 ± 0.014	0.135 ± 0.008
BR93020 (clade F, X4R5)	0.376 ± 0.210	12.596 ± 18.198	ND	ND
Ru570 (clade G, R5)	0.578 ± 0.168	6.247 ± 2.549	0.127 ± 0.010	0.236 ± 0.017
BCF02 (clade O, R5)	0.529 ± 0.201	4.131 ± 4.057	ND	ND
Drug-resistant HIV-1 strainsc
AZT-resistant strain	0.072 ± 0.012	0.192 ± 0.030	0.056 ± 0.008	0.116 ± 0.005
NL4-3D36G	0.063 ± 0.018	0.450 ± 0.030	0.036 ± 0.020	0.312 ± 0.112
NL4-3(36G)V38A	0.053 ± 0.025	0.443 ± 0.054	>2.000	>2.000
NL4-3(36G)V38E/N42T	0.081 ± 0.014	0.498 ± 0.170	>2.000	>2.000
NL4-3(36G)V38E/N42S	0.001 ± 0.000	0.003 ± 0.000	>2.000	>2.000
HIV-2ROD	1.420 ± 0.190	4.751 ± 0.650	1.991 ± 0.184	10.94 ± 0.507
SHIV
SF33A	0.033 ± 0.008	0.188 ± 0.023	0.541 ± 0.149	10.99 ± 3.779
SF162P3	0.450 ± 0.110	2.778 ± 0.420	0.205 ± 0.109	21.12 ± 10.39
SIVmac251 32H	0.083 ± 0.007	1.985 ± 0.320	3.051 ± 0.677	13.01 ± 3.916

^aThe measurements were performed in triplicate, and each experiment was repeated at least twice. Data are presented as means ± standard deviations.

^bT20 was used as a control.

^cNL4-3_D36G is a T20-sensitive strain which was the parent strain used for the generation of the T20-resistant mutants NL4-3_(36G)V38A, NL4-3_{(36G)V38E/N42T}, and NL4-3_{(36G)V38E/N42S}. NL4-3_{(36G)V38E/N42S} is also resistant to T1249 (9).

^dND, not done.

HP-OVA inhibited HIV-1 entry by time-of-addition assay.

To determine whether HP-OVA is an HIV entry inhibitor, we first used a time-of-addition assay to evaluate the inhibitory activity of HP-OVA against X4 and R5 viruses when it was added to cells at different intervals postinfection. As shown in Fig. 1, the nucleoside reverse transcriptase inhibitor AZT exhibited similar anti-HIV-1 activities (against both the X4 virus HIV-1_IIIB and the R5 virus HIV-1_BaL) when it was added to cells before viral infection and 8 h postinfection, while the HIV entry inhibitors T20 (against both HIV-1_IIIB and HIV-1_BaL), AMD-3100 (against HIV-1_IIIB), and maraviroc (against HIV-1_BaL) exhibited significantly decreased inhibitory activities when they were added 0.5 to 2 h postinfection. HP-OVA and 3-HP-β-LG showed inhibitory profiles similar to those of HIV entry inhibitors, suggesting that HP-OVA is also an HIV entry inhibitor.

FIG. 1.

Inhibition of HIV-1 entry by HP-OVA as determined by time-of-addition assay. (A) HP-OVA (0.2 μM) was added to MT-2 cells at different intervals after infection with HIV-1_IIIB. 3HP-ß-LG (0.2 μM), AZT (0.1 μM), AMD-3100 (0.2 μM), and T20 (0.5 μM) were included as controls. (B) HP-OVA (4.5 μM) was added to TZM-bl cells at different intervals after infection with HIV-1_BaL. 3HP-ß-LG (4.5 μM), AZT (2 μM), maraviroc (0.2 μM), and T20 (2 μM) were included as controls. These assays were done in triplicate, and the experiment was repeated at least twice. Error bars indicate standard deviations.

HP-OVA inhibited HIV-1 entry by blocking membrane fusion.

Viral envelope glycoprotein (Env)-mediated membrane fusion is a critical step for HIV entry into a target cell. Therefore, it is essential to determine whether HP-OVA can inhibit HIV-1 Env-mediated cell-cell fusion. Here we used noninfectious CHO-WT cells expressing HIV-1 Env and infectious HIV-1_IIIB-infected H9 cells (H9/HIV-1_IIIB cells) as the effector cells and MT-2 cells expressing CD4 and CXCR4 as the target cells to determine the inhibitory activity of HP-OVA against HIV-1 Env-induced cell-cell fusion. As shown in Fig. 2, HP-OVA inhibited fusion of both CHO-WT cells and infectious H9/HIV-1_IIIB cells with MT-2 cells, with IC₅₀s of about 0.113 and 0.057 μM, respectively. ADS-J1, the small-molecule HIV entry inhibitor targeting gp41, exhibited a similar dose-dependent curve. These results suggest that, like ADS-J1, HP-OVA inhibits HIV-1 entry by blocking HIV-1 Env-mediated membrane fusion.

FIG. 2.

Inhibitory activity of HP-OVA against HIV-1-mediated cell-cell fusion. Inhibition of fusion between CHO-WT cells (A) or calcein-labeled HIV-1_IIIB infected H9 (H9/HIV-1_IIIB) cells (B) and MT-2 cells was assessed by a dye transfer assay as described in Materials and Methods. Each sample was tested in quadruplicate (A) or triplicate (B), and the data are presented as means ± standard deviations.

HP-OVA inhibited cell-to-cell transmission of HIV-1 R5 virus.

To determine whether HP-OVA could inhibit cell-to-cell transmission of an HIV-1 R5 virus, PBMCs that were infected with HIV-1_BaL and washed with medium to remove free viral particles were cocultured with CEMx174 5.25M7 cells in the presence of HP-OVA at graded concentrations. After 3 days, the level of luciferase activity, representing HIV-1 infectivity in CEMx174 5.25M7 cells, was measured. As shown in Fig. 3, HP-OVA significantly blocked the transmission of HIV-1_BaL from PBMCs to CEMx174 5.25M7 cells, with an IC₅₀ of 0.319 μM, in a range similar to that for the inhibition of syncytium formation (cell-cell fusion) mediated by HIV-1_IIIB (X4 virus) (Fig. 2). These results suggest that HP-OVA can also prevent cell-to-cell transmission of an HIV-1 R5 virus.

FIG. 3.

HP-OVA-mediated inhibition of transmission of HIV-1_BaL from PBMCs to CEMx174 5.25M7 cells. All samples were tested in triplicate, and the data are presented as means ± standard deviations.

HP-OVA blocked the interaction between gp120 and CD4.

The first step of HIV-1 entry into a CD4⁺ target cell occurs when the Env surface subunit gp120 binds to CD4, which is the HIV-1 receptor on the host cell. Our previous study has shown that 3HP-β-LG interferes with the binding of gp120 to CD4 (46). Using similar approaches, we determined the potential effect of HP-OVA on the interaction between gp120 and soluble CD4 (sCD4). As shown in Fig. 4A and B, HP-OVA significantly inhibited the binding of gp120 from HIV-1_IIIB and HIV-1_BaL with sCD4 in a dose-dependent manner, with IC₅₀s of 0.030 μM and 0.060 μM, respectively. To determine whether HP-OVA interacts with gp120 or CD4 to block the association, we measured its effect on the binding of Q4120, a CD4-specific monoclonal antibody, to sCD4. We found that HP-OVA remarkably inhibited the binding of Q4120 to sCD4, with an IC₅₀ of 0.348 μM (Fig. 4C). To further elucidate the binding target(s) of HP-OVA, the interaction of HP-OVA with gp120 or sCD4 was examined using a surface plasmon resonance (SPR) assay. In this experiment, gp120 from HIV-1_IIIB, HIV-1_BaL, or sCD4 was immobilized onto a CM3 sensor chip and graded concentrations of HP-OVA were injected onto the immobilized surface. The binding data were analyzed using BIAeval 3.0 software. As shown in Fig. 5, HP-OVA could bind to both sCD4 and gp120 but with different binding affinities. When fitted with the 1:1 binding model, HP-OVA bound to sCD4 with a binding affinity (K_D) of 5.72 × 10⁻⁸ M, an association constant (K_a) of 3.81 × 10³ M⁻¹ s⁻¹, and a dissociation constant (K_d) of 2.18 × 10⁻⁴ s⁻¹. Under similar conditions, HP-OVA bound to gp120 from HIV-1_IIIB with a K_D of 2.86 × 10⁻⁷ M, a K_a of 1.23 × 10³ Ms⁻¹, and a K_d of 3.52 × 10⁻⁴ s⁻¹ and bound to gp120 from HIV-1_BaL with a K_D of 3.28 × 10⁻⁸ M, a K_a of 5.09 × 10³ Ms⁻¹, and a K_d of 1.67 × 10⁻⁴ s⁻¹. These results indicated that HP-OVA may bind to both gp120 and CD4 and then interfere with their interaction, resulting in the inhibition of HIV-1 entry.

FIG. 4.

Ability of HP-OVA to inhibit the interaction between sCD4 and CD4-binding proteins. The inhibitory activities of sCD4 binding to gp120 from HIV-1_IIIB (A), gp120 from HIV-1_BaL (B), and anti-CD4 MAb Q4120 (C) were determined by ELISA. Each sample was tested in triplicate, and the data are presented as means ± standard deviations.

FIG. 5.

Binding of HP-OVA to sCD4 and gp120 as assessed by SPR assay. (A) Dose-dependent binding of HP-OVA to sCD4 (A), to gp120 from HIV-1_IIIB (B), and to gp120 from HIV-1_BaL (C) was observed. sCD4 or gp120 was immobilized onto a CM3 sensor chip. HP-OVA at various concentrations (0.16 to 5.00 μM) was injected onto the surface for 200 s at a flow rate of 30 μl/min and then washed with HBS-EP. The results from the association and dissociation reactions were determined. The response units were recorded against the flow time (s).

HP-OVA interfered with gp41 six-helix bundle formation.

The HIV-1 gp41 six-helix bundle formation is a critical conformational change during HIV-1 fusion with the target cells. To investigate whether HP-OVA can block gp41 six-helix bundle formation, we used a model system mimicking gp41 core formation by mixing the gp41 N and C peptides at equal molar concentrations (38, 41). The gp41 core structure can be detected by a sandwich ELISA using a conformation-specific MAb, NC-1, as reported earlier (26, 29). Indeed, HP-OVA inhibited gp41 six-helix bundle formation in a dose-dependent manner, with an IC₅₀ of 0.103 ± 0.016 μM (Fig. 6A). The inhibitory activity of HP-OVA against six-helix bundle formation was further determined using a convenient biophysical method, FN-PAGE, by which N36 and C34-FITC peptides can form a visible band of gp41 core structure. As shown in Fig. 6B, C34-FITC showed a clear band at the lower position (lane 1). When N36 and C34-FITC were mixed together, two bands were revealed (Fig. 6B, lane 2). The major band at the upper position corresponds to the gp41 six-helix bundle formed by N36 and C34-FITC, as confirmed by Western blotting with the MAb NC-1 (38). When HP-OVA at different concentrations (15, 7.5, 3.75, and 1.875 μM) (Fig. 6B, lanes 3 to 6) was preincubated with N36 before the addition of C34-FITC, the intensities of the upper bands decreased while those of the lower bands increased in a dose-dependent manner. The unmodified OVA at 15 μM (Fig. 6B, lane 7) showed no inhibitory activity against gp41 six-helix bundle formation. These results confirm that the gp41 six-helix bundle formation can be inhibited by HP-OVA.

FIG. 6.

Inhibition of gp41 six-helix bundle (6-HB) formation by HP-OVA. (A) The inhibitory activity was determined by ELISA. Each sample was tested in triplicate, and the data are presented as means ± standard deviations. (B) The inhibitory activity was determined by FN-PAGE. The peptide N36 was incubated with HP-OVA at 37°C for 30 min before addition of the peptide C34-FITC at graded concentrations (15, 7.5, 3.75, and 1.875 μM for lanes 3 to 6, respectively). After incubation for another 30 min, the mixtures were analyzed by FN-PAGE.

HP-OVA had very low in vitro cytotoxicity to HIV target cells and reproductive tract epithelial cells.

The efficacy of a microbicide depends on the balance between its specific activity and its safety. Therefore, we evaluated the cytotoxicity of HP-OVA to the vaginal and cervical epithelial cell lines VK2/E6E7, Ect1/E6E7, End1/E6E7 and the cells that were used for testing the anti-HIV-1 activity of HP-OVA, including MT-2 cells, TZM-b1 cells, and PBMCs. By the XTT colorimetric assay results, HP-OVA had very low or no in vitro cytotoxicity to these cells even at the concentration of HP-OVA more than 1,000 times its 90% inhibitory concentration (IC₉₀) against HIV-1_IIIB infection (Table 4).

TABLE 4.

In vitro cytotoxicity of HP-OVA^a

Cell type	CC50 (μM)	CC90 (μM)b
MT-2 cells	86.33 ± 6.29	>179.20
TZM-bl cells	>179.20	>179.20
PBMCs	146.34 ± 8.44	>179.20
VK2/E6E7 cells	115.85 ± 3.61	>179.20
Ect1/E6E7 cells	176.13 ± 13.69	>179.20
End1/E6E7 cells	68.52 ± 11.92	153.42 ± 14.20

^aThe assay was performed in triplicate. Data are presented as means ± standard deviations.

^bCC₉₀, 90% cytotoxicity concentration.

HP-OVA did not affect spontaneous or PHA-stimulated PBMC proliferation.

Subsequently, we determined the potential effect of HP-OVA on spontaneous and PHA-stimulated PBMC proliferation. As shown in Fig. 7, neither HP-OVA nor the unmodified OVA at concentrations as high as 100 μM could significantly affect spontaneous or PHA-stimulated PBMC proliferation. The proliferation stimulation indexes (the ratio of PHA-stimulated PBMC proliferation to spontaneous PBMC proliferation) for cultures containing HP-OVA and OVA were similar to those for the PBS control.

FIG. 7.

Effects of HP-OVA and OVA on spontaneous (A) and PHA-stimulated (B) PBMC proliferation. Absorbance data are expressed as ratios in relation to results for the control (without HP-OVA and OVA). Each sample was tested in triplicate, and the data are presented as means ± standard deviations.

HP-OVA had no significant effect on the production of IFN-γ by PBMCs stimulated with PHA or not stimulated.

IFN-γ is one of the important cytokines produced by CD4⁺ T lymphocytes (1, 76). To explore the possible effect of HP-OVA on the function of human T lymphocytes, we determined whether HP-OVA affected the secretion of IFN-γ by human PBMCs stimulated with PHA or not stimulated. Unmodified OVA was used as a control. As shown in Fig. 8, neither HP-OVA nor OVA could significantly stimulate human PBMCs to secrete IFN-γ, while the positive control, PHA, elicited a high level of expression of IFN-γ. Also, HP-OVA and OVA had no significant effect on IFN-γ secretion by PHA-stimulated PBMCs. These results suggest that HP-OVA may have no harmful or deleterious impact on the function of immune cells.

FIG. 8.

Effects of HP-OVA on IFN-γ secretion by human PBMCs stimulated with PHA or not stimulated. (A) IFN-γ secretion of human PBMCs in the presence of HP-OVA at graded concentrations and 5 μg/ml of PHA as a control. (B) Effects of HP-OVA on IFN-γ secretion by human PBMCs stimulated with PHA (5 μg/ml). The spots of IFN-γ-producing cells in PBMCs were counted and are expressed as the number of spot-forming cells (SFCs) per 10⁶ PBMCs. Each sample was tested in triplicate, and the data are presented as means ± standard deviations.

DISCUSSION

Site-specific chemical modification of amino acid residues in a protein may result in change of the structure and/or function of the protein (56). Our group previously discovered a series of chemically modified bovine milk proteins with anti-HIV-1 activity (46). One of the most efficient antiviral agents is 3-hydroxyphthalic anhydride-modified ß-lactoglobulin (3HP-ß-LG), which is a promising candidate microbicide to prevent the sexual transmission of HIV-1 because of its broad antiviral activity against HIV-1, HIV-2, SIV, and HSV and against Chlamydia trachomatis (25, 28, 33, 47, 49, 51). Most importantly, 3HP-ß-LG is highly soluble and stable for long-term storage. A 3HP-ß-LG solution was stored at 4°C for more than 10 years and still exhibited potent anti-HIV-1 activity (our unpublished data). Therefore, 3HP-ß-LG is an ideal candidate for the development of an anti-HIV microbicide. However, the further development of 3HP-ß-LG was discontinued because the medical use of bovine products and their derivatives, like 3HP-ß-LG, was not recommended by the World Health Organization after the epidemic of BSE.

These circumstances resulted in screening proteins that have been used widely by humans and proven safe in order to find an alternative molecule to replace ß-LG for generating a new microbicide. After extensive screening, we found that several HP-modified proteins not of bovine origin, including OVA, rabbit serum albumin, porcine serum albumin, gelatin from cold-water-fish skin, and gelatin from porcine skin, exhibited inhibitory activity against infection by HIV-1 X4 and R5 viruses (our unpublished data). We selected HP-OVA for further studies because it had relatively high potency in inhibiting HIV-1 infection and because chicken OVA is widely available and inexpensive. It is the main protein in egg white, making up 60 to 65% of the total protein (22, 54). Some positively charged amino acids (lysine and arginine) in OVA can be modified by acid anhydrides. Here we found that the percentages of the HP-modified and unmodified lysine and arginine residues in OVA were dependent on the concentration of HP and pH of the reaction system and were correlated with the anti-HIV-1 activity of HP-OVA (Tables 1 and 2). These results are consistent with those from other studies of 3HP-ß-LG (46, 49), suggesting that the modified amino acid residues play an important role in mediating antiviral activity. Notably, treatment of HP-OVA with trypsin, which predominantly cleaves peptide chains at the carboxyl side of the amino acids lysine and arginine (55), did not affect the anti-HIV-1 activity of HP-OVA (data not shown), indicating that the protein containing the HP-modified lysine and arginine residues becomes resistant to trypsin.

Like 3HP-ß-LG, HP-OVA effectively inhibited infection by laboratory-adapted HIV-1 strains, including X4 and R5 viruses, with IC₅₀s in the nanomolar range. HP-OVA is highly effective in inhibiting infection by primary R5 viruses with distinct genotypes and phenotypes (Table 3). HP-OVA is more effective against the predominant HIV-1 subtypes A, B, and C. In particular, HP-OVA is highly potent against the primary HIV-1 subtype isolated in Thailand, 92TH009 (subtype A/E, R5), with an IC₅₀ of about 10 nM, and it has been reported that the HIV-1 subtype A/E R5 virus was preferentially sexually transmitted (36). HP-OVA has antiviral efficacies against both X4 and R5 strains comparable to that of T20, an FDA-approved peptidic HIV entry inhibitor targeting gp41, but it is much more effective than T20 against the T20-resistant HIV-1 strains (Table 3). Furthermore, HP-OVA exhibited higher anti-HIV-1 potency than the polymeric microbicide candidates naphthalene sulfonate (PRO 2000) and cellulose acetate phthalate (CAP), which inhibit HIV-1 infection at low micromolar levels. Cellulose sulfate and carrageenan, polyanionic microbicides that have failed to demonstrate protective effects against HIV in clinical trials, exhibited very low in vitro efficacy against primary R5 HIV-1 isolates (8, 63) or were even found to enhance HIV-1 infection (68, 73).

Besides the polyanionic polymers, several nonnucleoside reverse transcriptase inhibitors (NNRTIs), including TMC-120 (23, 74), UC781 (61, 74), and tenofovir (2, 43), are being developed as microbicides. However, their future clinical use has been called into question as they may be ineffective in preventing sexual transmission of the drug-resistant HIV-1 strains from HIV/AIDS patients who have been treated with reverse transcriptase inhibitors (73). Notably, HP-OVA was shown to be effective against the HIV-1 variants resistant to AZT and other reverse transcriptase inhibitors (Table 3), suggesting that HP-OVA is capable of preventing the sexual transmission of HIV-1 strains that are resistant to the currently used antiretroviral therapeutics. Therefore, HP-OVA may be used in combination with an NNRTI-based microbicide for preventing sexual transmission of HIV because the combination may have synergistic antiviral activity against a broad spectrum of HIV-1 strains, including those resistant to NNRTIs, and reduce the potential toxic effects due to dose reduction.

HP-OVA is also effective in inhibiting HIV-2 infection, suggesting that this microbicide candidate may be applicable in West Africa, where HIV-2 is prominent (7). Our studies also showed that HP-OVA could potently inhibit infection by SHIV_SF162 (R5), SHIV_SF33A (X4), and SIV, with IC₅₀s ranging from 0.033 to 0.450 μM (Table 3). Since both SHIV and SIV can be used for infection of rhesus macaques (16, 17, 45), HP-OVA will be tested in a nonhuman primate model for evaluation of its in vivo efficacy against SHIV or SIV infection through vaginal challenge.

The failure of nonoxynol-9 (19) and cellulose sulfate (4, 32) in phase III clinical trials warned us that the safety evaluation of a microbicide candidate should be carried out as early as possible. Here we first assessed the potential cytotoxicity of HP-OVA to three well-characterized cell lines derived from the human vaginal and cervical epithelium (11) and three human immune T-cell lines as well as PBMCs, which were used for evaluation of the in vitro anti-HIV-1 activity of HP-OVA. The results showed that HP-OVA had low cytotoxicity to all tested cells (Table 4). Its selectivity index (CC₅₀/IC₅₀) ranged from 253 to 13,066, indicating that HP-OVA appears to be safe in vitro. More-extensive animal studies to evaluate its in vivo toxicity will be carried out in the future.

A microbicide capable of inhibiting HIV infection by targeting the entry step has an obvious advantage because it can block HIV transmission at the beginning of viral infection. By using time-of-addition, cell-cell fusion, and cell-to-cell transmission assays, we demonstrated that HP-OVA, like 3HP-ß-LG, is an HIV entry/fusion inhibitor since it exhibited significantly decreased inhibitory activity when it was added after HIV-1 infection (Fig. 1) and it showed potent inhibitory activity against cell-cell fusion mediated by an X4 virus (Fig. 2) and cell-to-cell transmission of an R5 virus (Fig. 3). Further studies suggested that, similarly to 3HP-ß-LG (49), HP-OVA could block the binding of the HIV-1 Env surface subunit gp120 (from both X4 and R5 viruses) or an anti-CD4 antibody to sCD4, the primary receptor for HIV (Fig. 4). Using an SPR assay, we demonstrated that HP-OVA could bind to both CD4 and gp120, resulting in inhibition of the interaction between gp120 and CD4 or between an anti-CD4 antibody and CD4 (Fig. 5). Furthermore, we demonstrated that HP-OVA could block the formation of the fusion-active gp41 six-helix bundle (Fig. 6). These results suggest that HP-OVA inhibits HIV-1 entry/fusion through multiple mechanisms of action by interacting with both gp120 and gp41 as well as CD4 via the negatively charged residues of HP-OVA, when the positively charged side chains of lysine and arginine residues were converted to negatively charged side chains after modification by 3-hydroxyphthalic anhydride. Consistently, several negatively charged polymeric microbicide candidates, such as cellulose acetate phthalate (CAP), carrageenan, cellulose sulfate, naphthalene sulfonate, and dextran sulfate, can also interact with gp120/gp41 and CD4 to block HIV-1 entry/fusion (50, 52, 53, 57, 60). The binding site of the ionic polymers on CD4 is closely associated with the gp120 binding region (57). In general, these microbicides are more effective against X4 viruses than against R5 viruses, possibly because X4 viruses have more positively charged residues in the V3 loop of gp120 than R5 viruses. However, this is not the case for HP-OVA, since it is almost equally effective against both X4 and R5 viruses (Tables 1 to 3).

Since HP-OVA can bind to sCD4 and the CD4 molecule on T lymphocytes plays an important role in T-cell activation (44), one of the concerns is whether HP-OVA affects the function of CD4⁺ T cells or induces T-cell anergy, a status of the lymphocyte that is functionally inactivated following antigen stimulation (62). Our studies demonstrated that proliferation of T lymphocytes in human PBMCs stimulated with PHA was not significantly affected by HP-OVA at concentrations as high as 100 μM (Fig. 7). Further studies showed that HP-OVA had no significant effect on the production of IFN-γ by PBMCs with or without PHA stimulation (Fig. 8). These results suggest that HP-OVA may not have deleterious effects on the function of CD4⁺ T cells, especially for those circulating in the human body; however, we cannot exclude the possibility that long-term use of CD4 blockers topically may suppress the function of CD4⁺ immune cells located in the vaginal mucosa. Therefore, long-term observation of the potential harmful effect of HP-OVA on the mucosal immune system is warranted.

In conclusion, because of its broad anti-HIV activity and low cytotoxicity as well as the benefits of low cost and the wide availability of OVA, its source material, HP-OVA promises to be an excellent candidate for development as an effective, safe, and affordable microbicide to prevent the sexual transmission of HIV in both developing and developed countries.