Exceptionally broad HIV-1 neutralization via bispecific antibody–mediated prepositioning

Significance

A target for preventing and treating HIV-1 infection is the conserved N-heptad repeat (NHR) region of the viral fusion protein gp41, but antibodies against this site have shown limited potency. We now show that a bispecific antibody, designed to preposition an NHR-targeting antibody at the site of viral entry by binding to the HIV-1 coreceptor CCR5, dramatically improves neutralization over NHR-binding alone. This approach achieves complete neutralization breadth (IC₈₀ < 5 μg/mL) against a large panel of HIV-1 strains. Since almost all initial HIV-1 infections are via CCR5-tropic viruses, this strategy is expected to be effective as a prophylactic agent and supports the concept of targeting the NHR as a viable therapeutic approach.

Abstract

Antibodies that recognize the conserved prehairpin intermediate (PHI) of class I viral membrane-fusion proteins typically show limited neutralization and have not been considered promising therapeutic agents. We previously developed a bispecific antibody (bsAb), iMab/D5_AR, directed toward both the gp41 N-heptad repeat (NHR) that is exposed within the HIV-1 PHI and toward CD4, the HIV-1 receptor on T cells. CD4-binding led to prepositioning of the bsAb at the site of viral fusion, enhancing its neutralization potency and achieving 95% breadth (IC80 < 5 μg/mL) against a panel of 119 pseudotyped, multiclade HIV-1 viruses. In the current study, we engineered a bsAb against NHR that also targets CCR5, one of two HIV-1 coreceptors on T cells. This optimized bsAb design further improves neutralization potency and achieves 100% neutralization breadth against the 119-member pseudotyped virus panel, including those resistant to CD4-binding iMab/D5_AR. Considering that nearly all initial HIV-1 infections occur via CCR5-tropic viruses, we expect our redesigned bsAb targeting CCR5 to be an effective prophylactic agent. These findings further support the rationale for pursuing the NHR as a therapeutic target for HIV-1 and lay the groundwork for a new class of engineered broadly neutralizing antibodies.

HIV-1 infection begins when the virus fuses with the host cell membrane, a process triggered by the binding of the viral envelope glycoprotein (Env) to the host receptor CD4 on T cells and either the CCR5 or CXCR4 coreceptor (1, 2). The HIV-1 Env complex is a trimer composed of gp120 and gp41 subunits (Fig. 1A). During the fusion process, this trimer undergoes a series of structural changes, starting from its native prefusion state, progressing through a prehairpin intermediate (PHI), and eventually forming a trimer-of-hairpins conformation (Fig. 1B) (1–6).

Fig. 1.

Structure of the HIV-1 envelope glycoprotein (Env) and its conformational changes during viral membrane fusion. (A) Schematic of the HIV-1 Env, which is cleaved into gp120 and gp41. The numbers above each domain indicate sequence identity, the percentage of identical amino acids based on data from the Los Alamos 2021 HIV database compromising >8,000 HIV-1 Env sequences. (B) Modeling of HIV-1 viral membrane fusion. The HIV-1 Env is a trimer of gp120/gp41 heterodimers in the native, prefusion conformation (Left; PDB: 7SKA). Upon binding to the HIV-1 receptor CD4 (PDB: 1WIP), gp120 undergoes structural rearrangements and subsequently binds to a coreceptor, either CCR5 (PDB: 4MBS) or CXCR4 (PDB: 3OE0). This interaction triggers a dynamic conformational change in the Env, transitioning to the prehairpin intermediate (center), during which the gp41 N-heptad repeat (NHR, PDB: 8VWE), C-heptad repeat (CHR), and membrane-proximal external region (MPER, PDB: 5T85) are exposed. The NHR and CHR then collapse into a stable trimer-of-hairpins (Right; PDB: 1AIK), bringing the viral and cell membranes into proximity to facilitate membrane fusion. The modeling incorporated available structural data. Blender was used for 3D visualization.

The N-heptad repeat (NHR) of gp41 becomes transiently exposed in the PHI, providing a target for therapeutic binding. This region is recognized by enfuvirtide (T-20), an FDA-approved peptide drug that binds to a conserved hydrophobic groove within the NHR (7, 8). Antibodies targeting NHR have been developed, including D5 (9) and its potency-improved variants, D5_AR (10) and D5_FI (11), which bind to a hydrophobic pocket located C-terminal to the enfuvirtide binding site. However, antibodies targeting the NHR are only weakly neutralizing.

We previously showed that the neutralization by the NHR-targeting antibody D5 could be enhanced in cells expressing human FcRI (12). However, FcγRI is not expressed on CD4+ T cells, which are the major target cells of HIV-1. More recently, we showed that using a bispecific antibody (bsAb), called iMab/D5_AR (13), to preposition D5_AR at the site of viral fusion by binding to CD4, increases its local effective concentration on both FcγRI-expressing and nonexpressing cells, thereby enhancing its ability to prevent viral entry. Notably, the resistance profile of D5_AR differed depending on the prepositioning strategy: mutations conferring resistance to iMab/D5_AR were distinct from those observed when neutralization was enhanced by FcγRI (13).

These findings prompted us to explore alternative prepositioning strategies. Here, we engineered a bsAb, D5_AR/P140, that prepositions D5_AR by targeting CCR5, one of the two HIV-1 coreceptors expressed on CD4+ T cells. D5_AR/P140 showed a distinct neutralization profile compared to iMab/D5_AR, demonstrating increased neutralization potency and breadth, achieving complete neutralization breadth (IC80 < 5 μg/mL) against 119 pseudotyped multiclade HIV-1 strains. Overall, our findings further support the HIV-1 NHR as a promising prophylactic and therapeutic target and establish a foundation for a new class of engineered broadly neutralizing antibodies (bnAbs).

Results

Engineering and Evaluation of a bsAb for Prepositioning on Host Cells.

To generate a bsAb that prepositions an NHR-targeting antibody, we created hybrids between D5_AR and P140 (also known as PRO140 or leronlimab), which competes with gp120 for CCR5 binding (14, 15) (Fig. 2A). While P140 has previously been incorporated into bsAbs (16–18), our study pairs P140 with D5_AR, an antibody that selectively binds the highly conserved NHR, which is specifically and transiently exposed only during the PHI stage of the fusion process. We used the same bsAb construct described previously to generate iMab/D5_AR (13), incorporating a CrossMab design, where one IgG antibody arm includes a heavy and light chain cross-over in the Fab region, and knob-in-hole mutations promote generation of highly pure heterodimeric bsAbs. We included H435R and Y436F mutations in the current design as well to eliminate protein A binding. The resulting heterodimeric bsAb, D5_AR/P140, incorporates the H435R and Y436F mutations in the anti-CCR5 heavy chain (hole arm) and the knob in the anti-NHR heavy chain (knob arm) (Fig. 2A). As controls, we inserted the Fab arm of mAb114, an antibody that targets an unrelated Ebola glycoprotein, into either the knob or hole arm (labeled as “c”), generating the constructs c/P140 and D5_AR/c, respectively. We then evaluated D5_AR/P140 and the control antibodies D5_AR/c and c/P140 for their ability to bind both CCR5 and the NHR mimetic simultaneously on host cells using flow cytometry. TZM-bl cells, which express CCR5, were incubated with the bispecific or control antibodies, followed by biotinylated NHR hydrophobic pocket mimetic IQN17. After washing, we assessed surface binding using streptavidin-APC and FITC-conjugated anti-human IgG Fab fragments. Only D5_AR/P140 demonstrated simultaneous binding to both the cells and the NHR mimetic, while the control antibodies did not (Fig. 2B). These findings indicate that the bsAb functions as designed, simultaneously engaging both CCR5 and the NHR, and is a promising tool for prepositioning at sites of viral infection.

Fig. 2.

Enhancing HIV-1 neutralization by prepositioning an NHR-targeting antibody on host cells via bispecific antibody. (A) Schematic of bsAbs created by fusing antibody fragments from the anti-NHR antibody D5_AR and the anti-CCR5 antibody P140. The Ebola glycoprotein-targeting antibody mAb114 is used as a negative control and abbreviated as c. (B) Flow cytometry analysis of bsAb binding to CCR5-expressing TZM-bl cells and to biotinylated NHR. The x-axis represents binding to CCR5, detected using an FITC-labeled goat anti-human IgG Fab fragment; the y-axis shows binding to biotinylated NHR hydrophobic pocket mimetic IQN17, detected with APC-labeled streptavidin. Control Fab configurations are labeled as (c/) when mAb114 Fab is on the knob arm and (/c) when on the hole arm. Each sample was measured with 10,000 cells and analyzed in FlowJo. (C) Summary of neutralization potency, as measured in the luciferase-based reporter assay: maximum percent inhibition (Left), IC₅₀ (middle), and IC₈₀ (Right) for D5_AR/P140, its individual control arms, and their combination, tested against 26 HIV-1 pseudoviruses spanning multiple clades and neutralization tiers (Dataset S1). Each dot reflects the average of biological duplicates and technical replicates for each pseudovirus (n = 26). Maximum inhibition values were determined at the highest antibody concentration (50 μg/mL for D5_AR/P140, and 50 μg/mL each for D5_AR/c and c/P140 in combination). IC₅₀ and IC₈₀ values were calculated using GraphPad Prism. Error bars represent the median ± interquartile range. mAb114 Fab was used as a negative control. Statistical significance was determined using a Wilcoxon matched-pairs signed-rank test (****P < 0.0001). The black dotted line marks the quantification limit (50 μg/mL) for bsAb.

Bispecific Antibody-Mediated Prepositioning on CCR5 Enhances HIV-1 Neutralization by D5_AR.

To evaluate the neutralization potency of D5_AR/P140, we performed a luciferase-based reporter assay (19, 20) alongside control antibodies using a panel of 26 HIV-1 pseudotyped viruses. This previously described panel includes strains from various clades and neutralization tiers to ensure a representative assessment (13). The two control antibodies, D5_AR/c and c/P140, which each contain a single functional Fab arm targeting either the NHR or CCR5, along with their combination, exhibited limited to moderate neutralizing activity across all tested pseudoviruses. In contrast, D5_AR/140 showed potent neutralization against all 26 pseudoviruses in the panel (Fig. 2C and Dataset S1).

The virus panel includes R5- (CCR5), X4- (CXCR4), and dual-tropic R5X4-tropic pseudoviruses. The mono-specific CCR5-targeting antibody P140 only modestly neutralized R5-tropic pseudoviruses, while the X4-tropic HXB2 and R5X4-tropic WEAUd15.410.5017 pseudoviruses were resistant to P140 (Fig. 2C and Dataset S1). However, since the TZM-bl cells express CCR5, D5_AR/P140 bsAb potently neutralized both pseudoviruses, with IC₈₀ values of 0.075 μg/mL for HXB2 and 0.15 μg/mL for WEAUd15.410.5017, indicating that prepositioning of D5_AR via P140 is critical to maximizing its activity.

D5_AR/P140 Exhibits Potent and Exceptionally Broad Neutralization.

To further test the neutralization potency and breadth of D5_AR/P140, we expanded the virus panel to one with 119 multiclade HIV-1 pseudoviruses (21). This panel represents a wide spectrum of genetic diversity, is almost entirely composed of tier 2/3 viruses, includes representation of transmitted/founder viruses, and has been used as a standard reference panel for evaluating the breadth and potency of novel bnAbs (22–24). We compared the activity of D5_AR/P140 to that of 13 antibodies: iMab/D5_AR, which prepositions D5_AR at the site of viral fusion by binding to CD4, and 10E8v4, which targets the MPER of gp41 and for which neutralization data were obtained from our previous study (13), plus an additional 11 bnAbs, whose data we obtained from the CATNAP database (https://www.hiv.lanl.gov/content/index) of published antibody neutralization data. These antibodies target diverse regions of the HIV-1 Env trimer that are accessible in its native prefusion form (25, 26), including the CD4-binding site, V1/V2 and V3 loops of gp120, the gp120/gp41 interface, and the MPER of gp41.

Compared to iMab/D5_AR, which achieved a median IC₈₀ value of 0.52 μg/mL, D5_AR/P140 was 6.4 fold more potent, achieving neutralization with a median IC₈₀ value of 0.082 μg/mL (Fig. 3A and Dataset S2). To evaluate the clinical relevance of D5_AR/P140 neutralization potency, we compared its IC₈₀ values against the 119-virus pseudovirus panel using thresholds established in the Antibody Mediated Prevention (AMP) trials. The AMP trials evaluated the patient-derived CD4-binding site bnAb VRC01 for pre-exposure prophylaxis and demonstrated that in vitro neutralization potency correlates with HIV-1 prevention efficacy in humans (27). In those trials, clinical prevention efficacies were estimated to be 0%, 75%, and >80% for IC₈₀ values >5 μg/mL, <1 μg/mL, and <0.3 μg/mL, respectively. Based on this scoring framework, D5_AR/P140 achieves exceptionally broad neutralization with predicted clinical prevention efficacies that exceed those of previously characterized bnAbs (Fig. 3B and Dataset S2).

Fig. 3.

Broad and potent HIV-1 neutralization by an NHR-targeting antibody via bispecific antibody-mediated prepositioning. (A) Neutralization potencies of D5_AR/P140, iMab/D5_AR, and 12 broadly neutralizing antibodies (bnAbs) against a panel of 119 HIV-1 pseudoviruses (Dataset S2). Antibodies are color-coded by the epitope they target: CD4-binding site (CD4bs), V1/V2 loop of gp120, high-mannose V3 loop, gp120/gp41 interface, membrane-proximal external region (MPER), and gp41 NHR. The black dotted line denotes the assay’s quantification limit (50 μg/mL). Neutralization data for D5_AR/P140 are from this study. All other neutralization data are from external sources (shown with gray background): iMab/D5_AR and 10E8v4 are from a previous study (13), and 11 bnAbs (VRC07 through 4E10 along the x-axis) were obtained from curated datasets in the Los Alamos HIV database via CATNAP (http://hiv.lanl.gov/catnap). Each dot (n = 119) indicates the mean neutralization value from technical duplicates for a given pseudotyped virus. The error bars show the median with interquartile range. (B) Plot of neutralization potency versus breadth for the same panel of 119 HIV-1 pseudoviruses. In vitro neutralization shown with predicted prevention efficacies of 0 to 75% (Left), 75% (Middle), and >80% (Right), corresponding to IC₈₀ thresholds of <5 μg/mL, <1 μg/mL, and <0.3 μg/mL, respectively, as defined in ref. 27. Breadth for each curve is defined as the percentage of viruses neutralized below the respective IC₈₀ cutoff; viruses above the cutoff are considered nonpreventative.

Prepositioning Strategy Affects the Neutralization Profile of D5_AR.

Next, we evaluated the neutralization efficacy of D5_AR/P140 against five pseudoviruses and viruses resistant to iMab/D5_AR (Fig. 4A). Under the selective pressure of iMab/D5_AR, we previously identified four of these resistance mutations (at residues H564, L568, K574, and Q577 in the BF520 (CCR5-tropism) envelope) through in vitro pseudovirus-based deep mutational scanning (DMS) and in vivo viral envelope sequencing (at residue Q575 in replication-competent (rc) NL4-3 with CXCR4-tropism) (13). All five mutations are in the hydrophobic pocket of the NHR, the binding site of the D5_AR antibody. These residues are highly conserved, as shown by analysis of over 8,000 Env sequences in the 2021 Los Alamos HIV database, which includes the majority of published HIV-1 Env sequences (Fig. 4B). We also tested a sixth virus, MVP-5180, a group O (outlier) HIV-1 isolate with R5X4 tropism with deviations (R574 and R577) from the conserved D5 epitope (K574 and Q577). When compared to the two control antibodies, D5_AR/c and c/P140, D5_AR/P140 demonstrated enhanced neutralization of all six viruses (Fig. 4C). Notably, like the X4-tropic virus HXB2 (Fig. 2C), the X4-tropic rcNL4-3 Q575R mutant was resistant to P140 but was potently neutralized by D5_AR/P140. This comparison highlights the consistent resistance of X4-tropic viruses to P140 and demonstrates that the D5_AR/P140 bsAb can neutralize viruses regardless of coreceptor usage.

Fig. 4.

D5_AR/P140 neutralizes HIV-1 variants with mutations in the NHR. (A) Map (PDB: 2CMR) of all NHR sites where mutations confer resistance to iMab/D5_AR-mediated neutralization (gray). Blue and red indicate the heavy chain (HC) and light chain (LC) of NHR-targeting antibody D5, respectively. (B) Table showing the sequence identity of the NHR region, based on data from the 2021 Los Alamos HIV database (over 8,000 Env sequences), highlighting positions where mutations are associated with resistance to iMab/D5_AR. (C) Neutralization IC₅₀ by D5_AR/P140 and its control antibodies against HIV-1 variants bearing mutations in the NHR hydrophobic pocket. Mutations at residues 564, 568, 574, and 577 are in the BF520 envelope (CCR5-tropic), a mutation at residue 575 is found in the replication-competent NL4-3 strain (CXCR4-tropic), and the HIV-1 isolate MVP-5180 (CCR5/CXCR4-tropic) contains deviations at residues 574 and 577 (R574 and R577) from the known D5 epitope residues (K574 and Q577). IC₅₀ values were calculated using GraphPad Prism. mAb114 Fab was used as a control Fab. The black dotted line indicates limit of quantification (LOQ) of bsAb. (D) Maximum percent inhibition of iMab/D5_AR, and D5_AR/P140 at 50 μg/mL tested against the HIV-1 variants bearing mutations in the NHR hydrophobic pocket. Data for iMab/D5_AR were obtained from a previous study (13).

Finally, we compared the neutralization of D5_AR/P140 with that of iMab/D5_AR (12). iMab/D5_AR showed weak to modest maximum percent inhibition neutralization against all mutant viruses and the outlier virus (Fig. 4D). In contrast, D5_AR/P140 exhibited potent neutralization across the panel.

Discussion

The gp41 NHR is only transiently exposed during the PHI, limiting the ability of NHR-targeting antibodies to bind to intact virions (3). Our previous design to preposition NHR-targeting D5_AR to sites of viral fusion in the PHI utilized CD4-binding iMab (iMab/D5_AR bsAb). Our optimized bsAb design, D5_AR/P140, prepositions the antibody to the transiently exposed NHR by binding to CCR5 (Fig. 5), and shows distinct and improved neutralization, achieving complete neutralization (100% breadth with predicted clinical prevention efficacies ranging from >0 to >80%) against a large pseudotyped virus panel and maintaining neutralization against viruses resistant to iMab/D5_AR. Along with our previous work describing the potentiation of neutralization by FcγRI (12, 28), these results show that prepositioning D5_AR via either CD4 or CCR5 enhances its neutralization potency and results in distinct neutralization profiles. To our knowledge, D5_AR/P140 is the most broadly neutralizing engineered HIV-1 antibody reported to date.

Fig. 5.

Modeling of the hypothesized mechanism for bispecific antibody-mediated prepositioning. It is hypothesized that one arm of the D5_AR/P140 bispecific antibody binds to CCR5, the HIV-1 coreceptor on the cell surface. This binding prepositions the gp41 NHR-targeting arm to rapidly engage the transiently exposed NHR during the prehairpin intermediate, thereby enhancing neutralization. The modeling incorporated available structural data and known interaction sites. The bispecific antibody was modeled based on IgG (PDB: 1HZH), with one arm superimposed onto the NHR-targeting antibody D5_AR bound to the NHR (PDB: 7VWE). While the structure of CCR5 (PDB: 4MBS) is available, only the epitope of the CCR5-targeting antibody P140 was used for positioning. Blender was used for 3D visualization.

Although 10E8-based bsAbs (10E8/iMab) targeting the MPER exhibit resistance to pseudoviruses with mutations in the 10E8 epitope (16), our case differs. Despite mutations in the highly conserved D5_AR epitope, the neutralization profile varied depending on whether prepositioning occurred through CD4 or CCR5. Notably, 10E8 is synergistic with both NHR-targeting enfuvirtide and D5_AR (28), and viruses resistant to NHR-targeting antibody can become sensitive to enfuvirtide (9), suggesting that 10E8, D5_AR, and enfuvirtide could be effectively used in combination therapies.

Lenacapavir, with its 100% protection against HIV-1 infection as a long-acting prophylactic agent (29), has transformed the landscape of HIV-1 prevention strategies. However, for treatment, resistance emerged rapidly during combination therapy in some patients with multidrug-resistant HIV-1, suggesting that lenacapavir has a low genetic barrier to resistance (30–32). Our bsAb may complement its efficacy, particularly in reducing mother-to-child transmission (MTCT). While MTCT has been dramatically reduced through effective public health interventions, an ongoing epidemic of pediatric HIV-1 infection persists, with an estimated 150,000 new pediatric infections occurring globally each year (33). Although the placenta acts as a partial barrier to HIV-1, it does not fully prevent in utero transmission, with the highest risk occurring during the final 14 d before delivery (34). Antiretroviral (ARV) prophylaxis can reduce MTCT, although challenges remain, including maternal adherence and drug-resistant virus strains (33, 35). Antibody-based prophylaxis to prevent HIV-1 MTCT is being explored (36), and the bsAb described here, D5_AR/P140, could be an attractive candidate for consideration in this context. It could be administered to pregnant women, taking advantage of Fc-neonatal Fc receptor (FcRn)-mediated placental transfer efficiency (37), or directly to infants, as passive immunization of antibodies represents a well-tolerated and safe modality in neonates (38).

The higher manufacturing cost of bsAbs compared to monospecific antibodies remains an important consideration, especially in resource-limited settings. Nonetheless, bsAbs targeting HIV-1 are currently in clinical development. These include 10E8/iMab (NCT05890963), which targets the MPER of gp41 and CD4, and CAP256J3LS (NCT06585891), which targets both the V2 apex and the CD4-binding site of gp120. The bsAb described here, D5_AR/P140, is still in an early-stage format, and next-generation versions incorporating features such as Fc engineering to increase its binding to FcRn for half-life extension (37) or the use of common light chains (39) are expected to improve pharmacokinetics and manufacturability further.

We acknowledge potential limitations of our approach. Although D5_AR can neutralize viruses with CXCR4-tropism, our bsAb utilizing P140 will not target CCR5-negative cells, which CXCR4-tropic viruses can still infect. However, maraviroc, an FDA-approved allosteric CCR5 inhibitor (40), highlights the clinical relevance of CCR5 as a therapeutic target for HIV-1, and additionally, almost all initial HIV-1 infection occurs via CCR5-tropic viruses (41–43). Indeed, homozygous deletion of CCR5 (Δ32) has been reported to confer resistance to HIV-1 acquisition (44). As well, the CCR5-targeting antibody used in our design, P140 (also known as leronlimab), has demonstrated a reduction in plasma HIV-1 levels in a recent phase 2b/3 clinical trial (45). Therefore, by prepositioning D5_AR via CCR5, D5_AR/P140 could be effective as a prophylactic agent.

Overall, our findings demonstrate that targeting the PHI with the engineered bsAb D5_AR/P140 results in potent and exceptionally broad neutralization. These results further support the NHR as a viable therapeutic target and define a new class of engineered bnAbs against HIV-1.

Methods

Bispecific Antibody Expression and Purification.

Genes for antibody heavy and light chains were synthesized by Integrated DNA Technologies. These genes were inserted into linearized bispecific IgG1 plasmids, prepared as previously described (13), using the In-Fusion HD Cloning Kit (Clontech). To produce the bsAb, plasmids encoding the heavy chain with a knob, the light chain, the CrossMab heavy chain with a hole, and the CrossMab light chain were cotransfected into Expi293F cells (Thermo Fisher Scientific) at a 1:1:1:1 ratio using FectoPRO transfection reagent (Polyplus). Cells were cultured at 37 °C with 8% CO₂ and shaking at 120 rpm. After 4 to 5 d, cells were collected by centrifugation at 4,000×g for 10 min and the supernatant was filtered through a 0.22-μm membrane. The bsAb was initially purified using a 5 mL MabSelect Sure PRISM™ column (Cytiva) on an ÄKTA Pure FPLC system, followed by size-exclusion chromatography on a Superdex 200 Increase 10/300 GL column (GE Healthcare) for further purification.

HIV-1 Virus Production.

HIV-1 pseudotyped lentiviruses were generated by cotransfecting HEK293T cells with 10 μg of HIV-1 Env plasmids and 20 μg of the pSG3ΔEnv backbone plasmid (catalog number 11051, NIH AIDS Reagent Program) using BioT transfection reagent (Bioland Scientific LLC), according to the manufacturer’s protocol. 16 h posttransfection, the medium was exchanged with fresh medium. Viral supernatants were collected 2 d later, centrifuged at 300×g for 5 min, filtered through a 0.45-μm membrane, and stored at −80 °C for later use.

Replication-competent HIV-1 was produced and titrated as previously described (13). Briefly, full-length HIV-1 plasmids were transfected into HEK293T cells using FuGENE (Promega), following the manufacturer’s guidelines. Media replacement and virus collection were carried out as described above.

Cell Culture.

TZM-bl cells (sourced from the NIH AIDS Reagent Program, originally provided by John C. Kappes and Xiaoyun Wu) and HEK293T cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum, 1% penicillin-streptomycin (Corning), and 1% L-glutamine (Corning). Cells were incubated at 37 °C with 5% CO₂ in a humidified atmosphere.

Viral Neutralization Assay.

TZM-bl cells, which express an HIV-1 Tat-regulated luciferase reporter gene, were utilized to quantify viral infection and antibody neutralization as described previously (19, 20). Briefly, 5,000 TZM-bl cells were seeded overnight in white-walled 96-well plates and incubated at 37 °C with 5% CO₂ in a humidified environment. The following day, the medium was removed, and 50 μL of bsAb at various concentrations was added to the cells for a 1-h incubation. Subsequently, 50 μL of a mixture containing HIV-1 pseudotyped lentivirus and DEAE-dextran (10 μg/mL) was added. After 48 h, the cells were lysed, and BriteLite Plus reagent (Perkin Elmer) was applied to measure luciferase activity. Relative luminescence units (RLU) were read using a Synergy HTX multimode reader (BioTek). Percent infection was calculated using the formula: [% infection = (RLU of test well – RLU of background)/(RLU of virus-only well – RLU of background) × 100]. Wells containing only cells served as the background control.

Flow Cytometry.

TZM-bl cells expressing CCR5 were incubated with 10 nM bsAb in flow cytometry buffer (1% [w/v] BSA in PBS with 0.05% [w/v] sodium azide) for 1 h at 4 °C. After four washes with flow cytometry buffer, the cells were incubated with 10 nM biotinylated IQN17 NHR hydrophobic pocket mimetic (46) for 1 h at 4 °C. Following additional washing, cells were stained with streptavidin-APC (1:200, BioLegend) and fluorescein isothiocyanate (FITC)-conjugated Fab fragment goat anti-human IgG (1:50, Jackson ImmunoResearch) for 1 h at 4 °C. The cells were washed four times before analysis and sorting on an Accuri™ C6 Plus flow cytometer (BD Biosciences). Data from 10,000 cells per sample were collected for analysis using FlowJo software.

Supplementary Material

Dataset S01 (XLSX)

Dataset S02 (XLSX)

Data, Materials, and Software Availability

Excel data have been deposited in CATNAP (N/A). Previously published data were used for this work (13). All other data are included in the manuscript and/or supporting information.