Intact Viral Particle Counts Measured by Flow Virometry Provide Insight into the Infectivity and Genome Packaging Efficiency of Moloney Murine Leukemia Virus

Abstract

Murine leukemia viruses (MLVs) have long been used as a research model to further our understanding of retroviruses. These simple gammaretroviruses have been studied extensively in various facets of science for nearly half a century, yet we have surprisingly little quantitative information about some of the basic features of these viral particles. These include parameters such as the genome packaging efficiency and the number of particles required for a productive infection. The reason for this knowledge gap relies primarily on the technical challenge of accurately measuring intact viral particles from infected cell supernatants. Virus-infected cells are well known to release soluble viral proteins, defective viruses, and extracellular vesicles (EVs) harboring viral proteins that may mimic viruses, all of which can skew virus titer quantifications. Flow virometry, also known as nanoscale flow cytometry or simply small-particle flow cytometry, is an emerging analytical method enabling high-throughput single-virus phenotypic characterizations. By utilizing the viral envelope glycoprotein (Env) and monodisperse light scattering characteristics as discerning parameters of intact virus particles, here, we analyzed the basic properties of Moloney MLV (M-MLV). We show that <24% of the total p30 capsid protein measured in infected cell supernatants is associated with intact viruses. We calculate that about one in five M-MLV particles contains a viral RNA genome pair and that individual intact particle infectivity is about 0.4%. These findings provide new insights into the characteristics of an extensively studied prototypical retrovirus while highlighting the benefits of flow virometry for the field of virology.IMPORTANCE Gammaretroviruses, or, more specifically, murine leukemia viruses (MLVs), have been a longstanding model for studying retroviruses. Although being extensively analyzed and dissected for decades, several facets of MLV biology are still poorly understood. One of the primary challenges has been enumerating total intact virus particles in a sample. While several analytical methods can precisely measure virus protein amounts, MLVs are known to induce the secretion of soluble and vesicle-associated viral proteins that can skew these measurements. With recent technological advances in flow cytometry, it is now possible to analyze viruses down to 90 nm in diameter with an approach called flow virometry. The technique has the added benefit of being able to discriminate viruses from extracellular vesicles and free viral proteins in order to confidently provide an intact viral particle count. Here, we used flow virometry to provide new insights into the basic characteristics of Moloney MLV.

Keywords: flow virometry; genome packaging; intact particle count; murine leukemia virus.

FIG 1

Viral titer measurement by an ELISA. (A) The protein content in viral supernatants was assessed using a commercially available ELISA to quantify both p30 and Env (GFP) levels. (B) The concentrations in panel A were converted to protein molecules using the corresponding molecular masses (Table 1). (C) Numbers of protein molecules in panel B were converted to an inferred virus titer based on the stoichiometry outlined in Table 1. Averages of 2,187 p30 molecules and 300 Env molecules per virion are assumed based on the literature. (D) Ratio of p30 to Env protein molecules for each viral stock.

FIG 2

Measurement of viral gRNA and of the infectious titer. (A) The efficiency of RNA column extraction and reverse transcription was monitored using an RNA standard. (B) Using the efficiencies from panel A, the absolute number of viral genomes was determined using a ddPCR strategy targeting the packaging signal (Psi) or Env sequences. (C) Transducing units (TU) were measured for each viral stock clone. Virus was analyzed from 10 independent cell clones producing MLV-sfGFP. Each data point is representative of results from two independent experiments. P values were calculated by paired Student’s t test. **, P ≤ 0.01.

FIG 3

Comparative uptake of sfGFP and Env-GFP by MLV and EVs. (A and B) SDS-PAGE analysis of sorted NIH 3T3 cells transduced with a retroviral vector expressing sfGFP or Env-GFP (A) or 10 infected cell clones producing replicative MLV-sfGFP (B). (C) Viral supernatants from the cell clones were analyzed by SDS-PAGE and probed for p30 and Env-GFP contents. C1 to C10 depict unique chronically infected MLV-sfGFP cell clones isolated by single-cell sorting. (D) Supernatants from each of the 10 clonal producer cells were analyzed by FVM. The gated region highlights GFP-positive particles. Each dot plot is representative of data from two independent experiments. (E) Histograms illustrating the GFP fluorescence intensity profiles from each of the 10 clones analyzed in panel D. a.u., arbitrary units; MFI, mean fluorescence intensity; CV, coefficient of variation.

FIG 4

Enumeration and analysis of MLV and EV particles by FVM. (A) Supernatants produced from uninfected NIH 3T3 cells (Control), sorted sfGFP- and Env-GFP-transduced cells, and MLV-sfGFP producer clone 6 were analyzed by FVM. The gated region highlights GFP-positive particles. (B and C) The particles gated in panel A were analyzed for the number of GFP⁺ (B) particles and the GFP mean fluorescence intensity (MFI) (C). (D) The producer cells from panel A were analyzed by flow cytometry for differences in GFP mean fluorescence intensities. Dashed lines indicate background levels established with the control sample. Virus was analyzed from 10 independent clones producing MLV-sfGFP; each data point is representative of results from two independent experiments.

FIG 5

Viral particle quantification using antibody staining. (A) FVM analysis comparing fluorescent virus (MLV-sfGFP) (clone 6) to nonfluorescent virus (MLV-DsfGFP). (B) Both viral supernatants were labeled with a fluorescent phycoerythrin (PE) antibody targeting an exposed epitope on GFP. (C) Quantification of virus stocks from three independent experiments. P values were calculated by paired Student’s t test. n.s., not statistically significant (P > 0.05).

FIG 6

MLV particle infectivity and viral gRNA packaging efficiency determined by FVM. (A) All 10 MLV-sfGFP stocks were quantified by FVM and nanoparticle tracking analysis (NTA). Absolute viral counts obtained by each method are compared. Virus counts for the ELISA reflect the total number of viruses obtained if all the p30 was associated with intact virus particles. (B) Zoomed-in view from panel A to compare results obtained by FVM, NTA, and genome pair analysis. (C) Relationship of virus-associated to free viral protein (capsid p30 or Env) determined by using the information from panel A, Fig. 1B, and Table 1. (D) Viral gRNA packaging efficiency calculated from the information in panel A and Fig. 2B. (E) Number of virions required for a productive infection using information from panel A and Fig. 2C. (F) Number of viral gRNA-containing virions required for a productive infection. Each data point is representative of results from two independent experiments. P values were calculated by paired Student’s t test. **, P ≤ 0.01; ****, P ≤ 0.0001; n.s., not statistically significant (P > 0.05).

FIG 7

Impact of Env insertions on MLV stability. (A) FVM analysis of fluorescent MLV-sfGFP and nonfluorescent WT MLV and MLV-V5 viruses. (B) Viral supernatants were labeled with an antibody targeting an exposed epitope on Env (top, anti-GFP-PE; bottom, anti-V5-PE). (C) The viral protein concentration as determined by an ELISA was converted to an inferred virus titer based on the stoichiometry outlined in Table 1, as described in the legend of Fig. 1. (D) Physical titer quantification of virus stocks based on SSC analysis and PE staining. (E) Relationship between virus-associated and free capsid p30 protein based on data from panels C and D. The data represent results of four technical replicates from one experiment. **, P ≤ 0.01; n.s., not statistically significant (P > 0.05).