A high-throughput approach to evaluating NCp7 RNA binding activity for HIV-1 drug discovery.

5’ biotinylated U3 III+IV RNA (5’-/5BiotinTEG/GGGAGGCGUGGCCUGGGCGGGACUGGGGAGUGG −3’) was purchased from IDT (Coralville, Iowa) and resuspended in RNase- and DNase-free water (Qiagen) to a concentration of 100μM [ 20 ]. RNA was split into 10μL aliquots and stored at −80°C protected from light. RNA was diluted to working concentrations immediately before use. NCp7 cDNA was cloned into the pGEX6P-2 vector (IDT) containing a GST tag and the plasmid was transformed into BL21 (DE3) bacteria. Starter cultures were grown in Terrific Broth (TB, 25mL) overnight at 37°C while shaking. 1L TB cultures were inoculated and grown to log phase while shaking at 37°C and protein expression was induced via isopropyl β-D-1-thiogalactopyranoside (IPTG, 1mM final concentration) for 4 hours. Cultures were then pelleted and flash frozen for later use. For protein extraction, pellets were resuspended in lysis buffer (10mM Tris HCl pH 8.0, 150mM NaCl, 0.3% lysozyme, 10mM MgCl2, 5 mM DTT, 1.5% Sarkosyl, 0.5 mM phenylmethylsulfonyl fluoride (PMSF), cOmplete Mini Protease inhibitor Cocktail, RNase A, and DNase I). Protein was purified using glutathione sepharose (GE Healthcare) and the eluted protein was desalted twice and exchanged to P Buffer (5mM NaH2PO4, 5mM Na2HPO4, 15mM NaCl) using PD10 columns. Purified protein was aliquoted, flash-frozen in liquid nitrogen, and stored at −80°C. Induction efficiency and protein purity were determined by SDS-PAGE and Coomassie staining, and purified protein concentration was determined using a Pierce BCA Protein Assay kit (ThermoFisher). BRACO-19 was purchased as a solid from MedChem Express (CAS No. : 351351-75-2) and resuspended in DMSO. A concentration curve was run in triplicate starting at 45.45μM followed by ten serial two-fold dilutions and a DMSO-only control. Triplicate values were averaged and normalized to the percentage of the DMSO-only signal. The NIH Chemical Collection (NIHCC) was purchased from Evotec (700 compounds at 10mM concentration in 10μL). For screening, all 700 compounds (see supplement ) were diluted to 500μM with DMSO in a 384-well plate. 2μL of each compound were transferred to assay plates via an Integra Viaflo pipet. The total volume for the assay was 22μL and final compound concentration was 45.45μM. For concentration response curves, hit compounds were run in triplicate with the same starting concentration as screening followed by ten serial two-fold dilutions and a DMSO-only control. Triplicate values were averaged and normalized to the percentage of the DMSO-only signal. Assays were conducted in white 384-well OptiPlate microplates (Revvity). All components were diluted in assay buffer (25mM HEPES, 0.04% casein, 1mg/mL Dextran-500, 0.2% Triton X-100, 0.05% ProClin 300, 10mM NaCl) supplemented with an additional 250mM NaCl. For assay optimization, separate 12-step two-fold serial dilutions were performed for NCp7 and RNA starting at a working concentration of 800nM. 5μL of each were added in a 2-dimensional array along with 5μL of streptavidin donor beads (Revvity, 8μg/mL), and 5μL glutathione acceptor beads (Revvity, 2μg/mL) for a final volume of 20μL and starting concentrations of 200nM NCp7 and RNA. Plates were loaded in the following order: Donor beads, RNA, NCp7, and acceptor beads. Plates were then covered and placed on a benchtop shaker at 400RPM for 1 minute to ensure thorough mixing, followed by a one-hour incubation period protected from light. Plates were read on an EnSpire multimode microplate reader (Revvity) using a standard 384-well low volume Alpha Assay software protocol. For screening, final concentrations of 25nM NCp7, 12.5nM RNA, 4μg/mL donor beads, and 1μg/mL Acceptor beads were found to be optimal. The presence of up to 9% DMSO was well tolerated by this assay. 2μL of compound (0.5mM concentration) was loaded into each well using the Integra Viaflo pipet, along with a 2μL DMSO negative control. The remaining assay components were combined in a master mix in the order described above and incubated on ice for 15 minutes. 20μL of the master mix was added to every well using the Integra Viaflo multichannel pipet, and the plates were covered, mixed, incubated, and scanned as previously mentioned. Data analysis was performed in Microsoft Excel and GraphPad Prism 10. For concentration-response curves, EC 50 calculations were performed using an asymmetric (five-parameter) nonlinear regression with S constrained to 0.5 and HillSlope constrained to one. IC 50 calculations were performed using a 4-parameter nonlinear regression with HillSlope constrained to −1. Curves were defined by the initial value of the parameters. Graphs were generated using GraphPad Prism 10. Data points consist of three technical replicates, lines represent nonlinear curve, and error bars represent ± standard error of the mean.

AlphaLISA is a proximity-based bead assay pioneered by PerkinElmer (now Revvity) that relies on singlet oxygen transfer between a donor bead and an acceptor bead to produce fluorophore emission [ 25 ]. While the proprietary donor and acceptor beads are available only through Revvity, they offer a large toolbox of beads compatible with a variety of protein tags and antibodies. This approach offers a no-wash alternative to traditional ELISA with a wider dynamic range, making it a simple and robust assay suitable for high-throughput screening. The assay is also well suited to other macromolecular interactions including RNA protein interactions [ 26 , 27 ]. In the present study, the third and fourth segments of the G4-rich U3 RNA were synthesized to generate the RNA probe, which was tagged with a 5’ biotin molecule to bind the streptavidin donor beads ( Figure 1B ) [ 20 ]. A recombinant NCp7-GST fusion protein was expressed in E. coli with the GST tag occurring on the N-terminus of NCp7 allowing coupling to glutathione acceptor beads. When acceptor-bound NCp7 binds to donor-bound U3 RNA, the donor and acceptor beads are in close proximity so that when excited with 680nm laser irradiation, singlet oxygen transfer occurs and produces chemiluminescent emission at 615nm ( Figure 1C ) [ 25 ]. This 615nm emission is quantified as a direct measurement of binding activity between NCp7 and U3 RNA. This assay can be run in low-volume with limited incubation time, making it efficient for high-throughput drug discovery. To validate the sensitivity and specificity of our NCp7 AlphaLISA assay, we tested a known inhibitor of NCp7 G4 binding, BRACO-19 ( Figure 1 D ) [ 20 , 28 ]. To generate an IC 50 for this known inhibitor, we ran eleven 2-fold serial dilutions of BRACO-19 with a DMSO-only control. The concentration response assay was run in triplicate and results were normalized to the percentage of DMSO-only signal. We found that BRACO-19 has an IC 50 of 1.29μM in our assay, indicating strong inhibition of NCp7-U3 binding. This result reinforces the specificity of this assay in targeting G4 RNA binding activity and demonstrates sensitivity to a known inhibitor of this interaction. To demonstrate the efficacy of this tool for high throughput applications, we conducted a drug-repurposing screen using the NIH Chemical Collection. Of the seven hundred compounds screened, we identified seventeen hit compounds (hit rate = 2.4%) that significantly altered NCp7 RNA binding with a Z-score of more than three standard deviations from the mean ( Figure 2 ). Of these seventeen hits, ten compounds showed inhibitory effects on NCp7 RNA binding with alpha counts more than three standard deviations below the mean ( Figure 2B ; Ursodeoxycholic acid, Danazol, Chloramphenicol, Homoveratrylamine, Mafenide acetate, Glycopyrronium bromide, Tranilast, Terbutaline sulfate, Clofazimine, and Disulfiram). We also noted seven compounds that seemed to enhance NCp7 RNA binding activity with increased alpha counts more than three standard deviations from the mean ( Figure 2B ; Orlistat, S(−)-Timolol maleate, TFMPP hydrochloride, Mirtazapine, Hexamethylene bisacetamide, Moxifloxacin hydrochloride, and 5-Methoxytryptamine). To determine the efficacy and potency of the hit compounds, a concentration-response study was conducted. An 11-step concentration response assay was run in triplicate for all seventeen compounds and normalized to a DMSO-only control. For the ten compounds showing a decreased signal in the AlphaScreen, a four-variable nonlinear regression analysis was performed with a HillSlope of −1 ( Figure 3A ). A hit was considered validated if there was >50% inhibition at the 45μM concentration. Of the initial ten hits evaluated, seven compounds were validated for NCp7 inhibition (Ursodeoxycholic acid, Danazol, Chloramphenicol, Homoveratrylamine, Mafenide acetate, Glycopyrronium bromide, and Terbutaline sulfate). IC 50 values were calculated for the seven validated hits and reported in Figure 3A . For hits showing an increase in AlphaScreen signal, an asymmetrical nonlinear regression (five parameters) was performed where S was constrained to 0.5 and HillSlope to 1 ( Figure 3B ). A hit was considered validated if there was >50% increase in normalized alpha count at the 45μM concentration. Of the seven initial hits identified, only two compounds were confirmed to enhance NCp7 activity (Timolol maleate and Orlistat). EC 50 values were calculated for these two validated hits and reported in Figure 3B . Of the 7 validated inhibitors of NCp7, the most potent compound was Danazol, a well-tolerated synthetic steroid derivative primarily used in the treatment of endometriosis, with the an IC 50 of 1.19μM [ 29 ]. Notably, Danazol was the only compound that had a lower IC 50 than that of the known inhibitor BRACO-19. Ursodeoxycholic acid, a bile acid often used in the treatment of gallstones, also showed strong inhibition of NCp7 RNA binding with an IC 50 of 2.46μM [ 30 ]. Chloramphenicol also has a low IC 50 of 3.36μM and works by binding in the peptidyl transferase center of the bacterial ribosome to inhibit peptide bond formation but this antimicrobial agent is seldom used today due to the development of more recent antimicrobials with less adverse side effects [ 31 , 32 ]. The remaining four compounds were less potent, all coming in with IC 50 values above 10μM. Homoveratrylamine has an IC 50 of 15.71μM but has no established clinical use. Another antimicrobial, Mafenide acetate, demonstrated an IC 50 of 16.06μM and is used topically in the treatment of burn wounds [ 33 ]. Glycopyrronium bromide, an anticholinergic drug primarily used for the treatment of chronic obstructive pulmonary disease, had an IC 50 of 11.09 μM [ 34 ]. Finally, turbutaline sulfate, a synthetic beta-2 agonist used to treat bronchial spasms associated with common pulmonary disorders such as asthma and emphysema, showed the least potent inhibition of NCp7 with an IC 50 of 22.66μM [ 35 , 36 ]. These results suggest that existing pharmaceutical agents may have a beneficial impact in HIV-1 treatment, but further research is needed to understand the role these drugs may play in disrupting HIV-1 replication and infection. Additionally, an analysis of documented use in individuals with HIV-1 could provide insight into potential synergistic effects with ART in viral suppression. Of the two validated enhancers of NCp7 RNA binding activity, timolol maleate has a slightly lower EC 50 than Orlistat at 25.05μM compared to 25.67μM. Timolol maleate is a beta blocker used in the treatment of glaucoma, while Orlistat is a lipase inhibitor used to treat obesity [ 37 , 38 ]. Interestingly, Orlistat has been shown to cause a resurgence in viral load in patients with HIV-1, and it has been speculated that there may be a drug-drug interaction between antiretrovirals used to treat HIV-1 and Orlistat [ 39 , 40 ]. Our findings suggest that Orlistat may also have a direct effect on NCp7 activity, which could potentially influence viral replication, but this requires further investigation. There is no reported literature on timolol maleate use in HIV-1 infected individuals, so more information is needed to determine if it may influence viral load. NCp7 is a promising broad-spectrum target for the treatment of HIV-1 and is the focus of ongoing drug discovery initiatives. We have developed a novel NCp7 RNA binding assay to advance these initiatives and demonstrated the efficacy of this high-throughput AlphaScreen with a drug repurposing screen. Our results highlighted seven compounds that showed inhibition of NCp7 RNA binding, and two compounds which enhanced this activity, of which one is already known to increase viral load. This assay is also a useful tool to assess zinc finger binding of RNA and has potential to be used as a counter screen in additional drug discovery endeavors.

The epidemic caused by human immunodeficiency virus, type 1 (HIV-1) is an ongoing battle in healthcare [ 1 ]. Current treatment for HIV-1 is focused on suppressing HIV-1 replication through antiretroviral therapy (ART), which requires lifelong treatment and monitoring to maintain an undetectable viral load [ 2 ]. While the introduction of ART has significantly lowered morbidity and mortality of HIV-1 infection, infected individuals continue to live with a chronic illness and face a resurgence in viral load should they deviate from the strict treatment regimens [ 3 ]. Many challenges remain in the treatment of HIV-1 infection as current therapies are accompanied by side-effects, drug resistance, decreased lifespan, and continued immune dysfunction even with viral suppression [ 3 – 5 ]. Due to the limitations of current ART, drug discovery efforts to identify novel HIV-1 therapeutics are essential. As a retrovirus, HIV-1 relies on reverse transcription (RT) of its RNA genome for replication. Two copies of this RNA genome, along with the necessary reverse transcriptase and integrase enzymes, are encapsidated in one viral particle. When a virus invades the cytoplasm of a host cell, it begins RT to synthesize a new double-stranded DNA (dsDNA) genome, which is then transported to the nucleus and integrated into the host genome [ 6 ]. The viral genome is flanked on either side by long terminal repeats (LTRs) that facilitate integration into the host genome [ 7 ]. The 5’ LTR serves as the promoter for all HIV-1 genes and contains the U3, R, and U5 regions, with U3 also occurring in the 3’ LTR [ 8 ]. Most available ARTs target individual stages of the viral reproductive cycle such as integration or reverse transcription, and therapies are often given in combination to increase efficacy. Recently, there has been a push for a new generation of broad spectrum antiretrovirals that are less susceptible to resistance [ 9 , 10 ]. The nucleocapsid protein 7 (NCp7) is a promising therapeutic target given its key roles in integration, reverse transcription, and viral assembly and genome encapsidation [ 11 – 13 ]. In addition to the roles NCp7 plays in the viral lifecycle, it also interacts with cellular components within the host cell to increase viral production and infectivity [ 14 , 15 ]. Unique to HIV, NCp7 is a 72 amino acid zinc finger protein with two CCHC motifs that are derived from the C-terminus of the Pr55gag polyprotein ( Figure 1A ) [ 16 , 17 ]. NCp7 acts as a nucleic acid chaperone protein and promotes dimerization of RNA, encapsidation, and initiation of primer tRNA annealing [ 18 , 19 ]. NCp7 has a high binding affinity for guanine-rich nucleic acids, particularly G quadruplexes (G4s) like those found in the U3 region of the HIV-1 LTR and further promotes transcription by resolving these secondary structures ( Figure 1B ) [ 20 – 24 ]. Given its role as a nucleic acid chaperone protein and involvement in key viral processes, NCp7 seems an ideal target for HIV-1 drug discovery. Here we present a high-throughput AlphaLISA assay to assess NCp7 RNA binding activity in the presence of chemical compounds, which will aid in the discovery of novel HIV-1 therapeutics.