A nanoluciferase SARS-CoV-2 for rapid neutralization testing and screening of anti-infective drugs for COVID-19

The study addresses the urgent need for rapid, reliable serological assays and antiviral screening tools for COVID-19. While RT-PCR detects acute infection, measuring neutralizing antibodies is critical for determining immune status, identifying convalescent plasma donors, and evaluating vaccines. The gold-standard PRNT directly measures neutralization but is low-throughput and slow, limiting large-scale deployment. The authors propose developing a genetically stable nanoluciferase (Nluc) reporter SARS-CoV-2 to enable rapid neutralization testing and high-throughput antiviral screening, aiming to overcome PRNT’s throughput and turnaround limitations.

The paper situates SARS-CoV-2 among human coronaviruses (SARS-CoV, MERS-CoV, and endemic OC43, 229E, NL63, HKU1) and notes existing serological platforms (lateral flow, ELISA, microsphere assays, pseudovirus assays). It underscores PRNT as the gold standard despite low throughput. Prior successes with reporter viruses in other systems (e.g., dengue, Zika) and recent fluorescent SARS-CoV-2 reporters are referenced to justify using luciferase-based reporters for sensitivity and throughput.

- Construction of reporter virus: Using an infectious cDNA clone of SARS-CoV-2 (2019-nCoV/USA_WA1/2020), nanoluciferase (Nluc) was inserted at ORF7. Seven cDNA fragments spanning the genome were ligated to generate a full-length Nluc cDNA with a T7 promoter for in vitro transcription. RNA transcripts were electroporated into Vero E6 cells to recover SARS-CoV-2-Nluc. Plaque morphology and replication kinetics were compared to wild-type recombinant SARS-CoV-2. - Stability and genetic characterization: SARS-CoV-2-Nluc was serially passaged five times (P5) in Vero E6 and also on A549-hACE2 cells. RT-PCR across the insertion junction, restriction enzyme digestion (BsrGI and PacI), and Sanger sequencing were used to confirm retention of the Nluc insert and to identify mutations across passages. - Entry receptor validation: Time-course Nluc signals post-infection established early robust readout. Antibody blocking assays used anti-hACE2 versus anti-DPP4 controls to assess impact on Nluc signal at 6 h post-infection. Infection efficiency was compared between A549 and A549-hACE2 cells across MOIs with luciferase readout at 24 h. - Rapid neutralization assay (Nluc-NT): In 96-well format, human serum samples were incubated with SARS-CoV-2-Nluc (1 h, 37 °C), then added to Vero E6 cells at MOI 0.5. Nluc signals were measured at 4 h post-infection to calculate NT50 via a four-parameter logistic model (Prism 8). Turnaround time was 5 h total. A panel of 21 RT-PCR-confirmed COVID-19 sera and 9 pre-pandemic negative sera were tested. Conventional PRNT50 was performed on the same sera for comparison and correlation analysis. - Antiviral assay development: Cytotoxicity (3-day CC50) and antiviral potency (48 h EC50) were assessed for chloroquine and remdesivir in Vero E6 and A549-hACE2 cells to optimize conditions (12,000 cells/well; MOI 0.025; phenol-red free medium with 2% FBS; 96-well plates). Luciferase signals were normalized to DMSO controls; EC50 estimated via nonlinear regression (Prism 8). A549-hACE2 was selected for broader screening due to higher biological relevance and remdesivir sensitivity. - Drug screening: A panel of clinically approved and investigational antivirals/anti-infectives was evaluated in A549-hACE2 cells, including nucleoside/nucleotide analogs, HIV antiretrovirals (PIs, NRTIs, NNRTIs, INSTIs), HCV antivirals, and other anti-infectives. EC50, CC50, and selectivity index (SI=CC50/EC50) were determined. Considerations for protein binding and clinically achievable exposures were discussed. - Additional experimental details: Cell culture conditions (DMEM with 10% FBS, supplements), generation of A549-hACE2, RNA extraction and sequencing workflows, immunofluorescence for ACE2, institutional approvals for human sera, PRNT procedures, and statistical analyses (ANOVA; P<0.05). All virus work performed in BSL-3 at UTMB.

- Reporter virus construction and stability: SARS-CoV-2-Nluc produced slightly larger plaques but replicated similarly to wild-type in Vero E6. After five passages, plaque morphology, replication kinetics, and Nluc insertion were retained; sequencing identified five nucleotide changes causing amino acid substitutions but no loss of reporter. - Entry receptor confirmation: Early Nluc signals (>10-fold above background at 1 h p.i.) allowed entry studies. Anti-hACE2 antibodies dose-dependently inhibited Nluc signals at 6 h p.i., whereas anti-DPP4 did not. A549-hACE2 cells yielded ~100-fold higher Nluc signals than parental A549 at 24 h across MOIs, supporting hACE2 as the SARS-CoV-2 receptor. - Rapid neutralization assay performance: In a 5-hour workflow (1 h serum-virus incubation + 4 h infection), all 21 COVID-19-positive sera had NT50 of 66–7237; all 9 negatives had NT50 <20. PRNT50 on the same samples were 80–3200 for positives and <20 for negatives. Nluc-NT50 correlated with PRNT50 with r ≈ 0.8380, with Nluc-NT generally yielding higher titers (~3-fold on average), indicating greater sensitivity and dynamic range. - Antiviral assay optimization and benchmark compounds: Chloroquine CC50 >50 µM in both Vero and A549-hACE2; remdesivir CC50 ≈ 50 µM (Vero) and 2.5 µM (A549-hACE2). Remdesivir EC50 was ~1.1 µM in Vero E6 and 0.115 µM in A549-hACE2 (>10-fold potency increase), underscoring cell-type effects. Chloroquine EC50 values were in the low micromolar range and more consistent across cell types. - Drug screening results (A549-hACE2): • Nucleos(t)ide analogs: Remdesivir active (EC50 0.869 µM; CC50 3.27 µM; SI 284 in table summary context; GS-441524 EC50 0.869 µM; CC50 >50 µM; SI >57). Others (e.g., sofosbuvir, favipiravir, ribavirin, entecavir, cidofovir) inactive at ≤10 µM. • HIV antiretrovirals: Nelfinavir active (EC50 0.77 µM; CC50 12.0 µM; SI ~15.7) but projected protein-adjusted potency exceeds typical clinical exposures; lopinavir, saquinavir, tipranavir weak (EC50 ~8–9 µM; SI 3–4); others (darunavir, ritonavir, atazanavir, indinavir, amprenavir) inactive at ≤10 µM. • HCV antivirals: Multiple agents (e.g., ledipasvir, velpatasvir, dasabuvir, sofosbuvir) inactive at ≤10 µM. • Other anti-infectives: Rupintrivir active (EC50 1.87 µM; SI >26); an agent reported with EC50 ~2.7 µM and SI ~10 (text refers to cobicistat as CYP3A inhibitor); oseltamivir carboxylate and baloxavir inactive at ≤10 µM; niclosamide showed low- SI activity (EC50 0.715 µM; SI ~1.8); presatovir active (EC50 2.53 µM; SI >135). Overall, remdesivir, chloroquine, and rupintrivir were among the more notable actives in vitro. - Cell-type dependence: Remdesivir potency varied markedly by cell type due to metabolic activation differences (A549-hACE2 > Vero E6; primary HAE cells reported even more potent), highlighting assay cell selection’s impact on EC50 interpretation.

The study delivers a practical solution to the need for rapid, scalable serology and antiviral testing by engineering a stable nanoluciferase SARS-CoV-2. The rapid neutralization assay shortens turnaround from days (PRNT) to hours, with strong correlation to PRNT and higher sensitivity, facilitating epidemiologic studies, vaccine evaluation, and donor selection for convalescent plasma. The high-throughput antiviral platform, particularly in A549-hACE2 cells, enables efficient screening and emphasizes the importance of biologically relevant cell models, as demonstrated by remdesivir’s cell-dependent potency. The findings refute in silico-derived expectations for several HIV/HCV antivirals (e.g., sofosbuvir, velpatasvir, tenofovir) by showing no measurable in vitro activity against SARS-CoV-2 at tested concentrations. Active hits such as nelfinavir and rupintrivir warrant mechanistic and pharmacological evaluation, though considerations like protein binding and clinically achievable exposures temper translational potential (e.g., nelfinavir’s protein-adjusted potency likely exceeding plasma levels). The platform’s sensitivity and adaptability to higher-density formats (384/1536-well) position it as a valuable tool for SARS-CoV-2 research and drug discovery.

A genetically stable nanoluciferase SARS-CoV-2 was developed that replicates comparably to wild-type virus and provides a robust, rapid neutralization assay with strong concordance to PRNT and superior sensitivity. The reporter virus also supports high-throughput antiviral screening, identifying remdesivir, nelfinavir, and rupintrivir as in vitro inhibitors among a broader panel where many clinically used antivirals were inactive against SARS-CoV-2. The work underscores the importance of using physiologically relevant cell systems for antiviral evaluation. Future research should clarify mechanisms of action (e.g., rupintrivir’s targeting of coronavirus proteases), assess in vivo efficacy and pharmacokinetics/pharmacodynamics, and expand screening to additional compound classes using this scalable platform.

- In vitro findings may not directly translate to clinical efficacy due to factors such as drug metabolism, tissue distribution, protein binding, and achievable exposures; for example, protein-adjusted potency of nelfinavir may exceed clinically attainable plasma levels. - Assay outcomes are cell-type dependent, particularly for prodrugs requiring metabolic activation (e.g., remdesivir), complicating cross-study comparisons and necessitating validation in primary human airway models. - The Nluc reporter’s enzymatic amplification could increase sensitivity relative to PRNT and may yield higher NT50 values, affecting direct comparability of titers. - Work requires BSL-3 containment for authentic virus, limiting accessibility compared to pseudovirus systems. - The screening set, while clinically relevant, was not exhaustive and dosing up to 10 µM may miss weaker actives or those requiring prolonged exposure or specific activation pathways.