Replication-competent adenovirus reporters utilizing endogenous viral expression architecture

Abstract

Adenoviruses are double-stranded DNA viruses widely used as platforms for vaccines, oncolytics, and gene delivery. However, tools for studying adenoviral gene expression in real time during infection remain limited. Here, we describe a set of fluorescent and bioluminescent reporter viruses built using the modular AdenoBuilder reverse genetics system and informed by high-resolution maps of Ad5 transcription. These reporters utilize endogenous early and late transcriptional units, enabling visualization of viral gene expression without exogenous promoters or splicing elements. These model viruses replicate with kinetics nearly indistinguishable from wild-type virus and have enabled real-time fluorescent imaging as well as longitudinal bioluminescent sampling from the same infected samples. Together, this next generation of adenovirus reporters provides a modular and tractable platform for studying viral gene regulation and replication dynamics in real time, with broad applications for basic research and high-throughput screening.

Importance: This research provides powerful new tools to rapidly study adenovirus gene expression and replication. By integrating fluorescent and secreted luciferase reporters into native viral regulatory elements, we enable real-time, non-destructive tracking of early and late stage infection in living cells. These modular reporters are compatible with a wide range of genetic and chemical perturbations, allowing researchers to investigate the function of specific viral genes, host interactions, and the impact of host genes and antiviral compounds. Importantly, the high-throughput nature of these systems overcomes limitations of traditional plaque assays to quantify viral replication dynamics. Our work will allow for the rapid creation of both novel infectious and replication-incompetent viral vectors.

Keywords: adenovirus; bioluminescent virus; fluorescent reporter virus; infectious unit quantification; live-cell imaging; reporter virus; viral replication kinetics.

Fig 1

Construction of reporter viruses using a modified AdenoBuilder method. (A) Ad5 transcriptome map overlaid with AdenoBuilder plasmid blocks. Early genes are shown in gray, late genes are shown in black. (B) Schematic of the deletion and frame-shifting insertion in plasmids B6 and B6/7 of the Adenobuilder system. (C) Visual representation of the cloning method used to insert either mNeonGreen or secreted nanoLuciferase into the E3 early region. (D) Visual representation of the cloning method used to insert P2A-mScarlet into the Fiber late region.

Fig 2

Reporter virus growth kinetics mimic that of wild-type AdV. (A) Viral titers measured in HEK-293 cells by conventional plaque assay (PFU) quantification or fluorescent plaque assay (fPFU) for E3mNG reporter virus compared with WT AdV. (B) In-cell western images of fluorescent reporters compared with WT AdV. Virus-encoded mNG expression (Fluo) overlaps with antibody staining for late viral proteins (IFA) at 48 hpi. (C) Titers of WT AdV and fluorescent reporter viruses measured by quantifying the number of cells stained with late viral proteins (IFA) or the number of cells expressing mNG green fluorescence (Fluo). (D) Representative western blot of A549 cells infected with either Mock (M), WT AdV, E3mNG, or E3gL5s for 16, 24, or 48 hours. Cell lysate was probed for Ad5 capsid proteins to assess late expression and DBP and E1A to assess early expression. GAPDH was used as a loading control. (E) qPCR analysis of viral genome replication in A549 cells. Cells were infected with either WT AdV, E3mNG, or E3gL5 and collected at either 2, 16, 24, or 48 hpi. Values were internally normalized to cellular Tubulin. (F) In-cell western quantification of infectious viral titer produced in A549 cells. Cells were infected with either WT AdV, E3mNG, or E3gL5s and collected at either 16, 24, or 48 hpi. The collected virus was then titered via in-cell western in A549 cells by measuring late viral protein signal. Statistics were determined by an unpaired two-tailed t test; *P < 0.05; ns, not significant; nd, not detected.

Fig 3

Imaging of fluorescent reporters recapitulates the viral replication cycle of AdV. (A) Fluorescence microscopy analysis of A549 cells infected with either WT AdV, E3mNG, or E3gL5s at 16, 24, or 48 hpi. Early reporter expresses mNG (green) and late reporter expresses mScarlet (red). (B) Fluorescence microscopy analysis of A549 cells infected with E3gL5s, treated with DMSO, Hydroxyurea (1 mM), or Flavopiridol (300 nM), and imaged at 24 hpi. (C) CellCyte analysis of early reporter green fluorescent positive A549 cells following E3gL5s infection for 48 hours after treatment with DMSO (green), Hydroxyurea (HydU, blue), or Flavopiridol (Flav, purple) at 6 hpi. The shaded area represents the SEM of six replicates. (D) CellCyte analysis of late reporter red fluorescent A549 cells following E3gL5s infection for 48 hours after treatment with DMSO (red), Hydroxyurea (HydU, blue), or Flavopiridol (Flav, purple) at 6 hpi. The shaded area represents the SEM of six replicates. For all imaging, the white scale bar represents 100 µm.

Fig 4

Optimized luminescence-based assay to assess the growth kinetics of secreted luciferase reporter virus. (A) IFA-FFU quantification of E3nLuc infection compared to that of wild-type Ad5. Cells were infected with either WT AdV or E3nLuc and collected at 16, 24, or 48 hpi. The collected virus was titered by IFA for late viral protein signal. (B) Comparison of luciferase expression in A549 cells infected with E3nLuc, in which supernatant was collected at either 8 or 24 hpi. Cells were infected at MOIs of 1, 3, 10, 30, and 100. Linear regression with shaded 95% confidence intervals, line formula, and R² value is shown. (C) Analysis of secreted luciferase from E3nLuc infection over time. Supernatant was collected every 4 hours until 48 hpi. Best-fit line of nonlinear regression with an R² value is shown. (D) Comparison of direct vs. reinfection secreted luciferase assays. Cells were infected with E3nLuc and collected at either 16, 24, or 48 hpi. Samples were then immediately read for Luciferase activity (Supernatant), freeze-thawed three times, and then read for luciferase activity (Lysate), or equal volumes were used to reinfect naïve A549 cells for 6 hours at a 1:10 dilution (Reinfection) before being collected and read for luciferase enzyme activity. All samples were normalized to the average of their respective 48 hpi sample set to 100.

Fig 5

Construction of fluorescent-tagged mutant viruses with endogenous processing. (A) Schematic of the construction of △E1B55K. E1B19K 5’ splice site is underlined. (B) Schematic of the construction of △E4-6-mNG, △E4-3-mT2, and △E4-3-mT2-6-mNG. The mNeonGreen tag was inserted in orf6 of △E4(mcs), while the mTurquoise2 tag was inserted in orf3. (C) Representative western blot of HeLa cells infected with either Mock (M), WT AdV (MOI 10), △E1B55k (MOI 10), or △E4-6-mNG (MOI 100) for 16, 24, or 48 hours. Cell lysate was probed for AdV5 capsid proteins to assess late expression and DBP and E1A to assess early expression. Cell lysate was also probed for E1B55k, E4orf3, and E4orf6 to confirm knockouts. GAPDH was used as a loading control. (D) Flow cytometric analysis of HeLa cells uninfected or infected with △E4-3-mT2, △E4-6-mNG, or △E4-3-mT2-6-mNG (MOI 100) and analyzed for mNeonGreen and mTurquoise2 expression at 48 hpi. The percentage of positive cells for each fluorophore is present in each gated quadrant.

Fig 6

Recombinant AdV vector system for tetracycline-inducible transgene expression. (A) Schematic of recombinant AdV design with E1 Tet-On cassette and E3 reverse Tetracycline Transactivator (rtTA3) cassette. (B) Flow cytometric analysis of HeLa cells infected with TETeGFP at either MOI 10 or 100. Doxycycline (1 µg/mL) (+Dox) or DMSO (−Dox) treatment was added at 2 hpi, and cells were analyzed for eGFP expression at 48 hpi.