Fig 1.
RNA-Flow cytometry using the PrimeFlow method affords robust and sensitive analysis of endogenous γHV genes at the single-cell level.
Viral RNA analysis in γHV68-infected fibroblasts by qRT-PCR (A, C) or by flow cytometric analysis using PrimeFlow at 16 hpi (B, D, E). Samples were either mock, TMER-TKO, or WT γHV68-infected, with infections done using 5 plaque forming units/cell. qRT-PCR standardized to 18s RNA, at the indicated times. All flow cytometric events gated on a generous FSC x SSC gate, followed by singlet discrimination. PrimeFlow analysis quantified probe fluorescence for (B) TMERs, (D) ORF73, or (E) ORF18, relative to side-scatter area (SSC-A). Probe fluorescence is indicated, with all probes detected using either AlexaFluor (AF) 647 or 488 conjugates. Data representative of 2 independent experiments, each done with biological replicates.
Fig 2.
Heterogeneous gene expression in a γHV68 latently infected B cell line.
Viral, host RNA analysis during γHV68 latency and reactivation in A20.γHV68 (HE2.1) cells by qRT-PCR (A) and flow cytometric analysis using PrimeFlow (B-G), comparing untreated or TPA-stimulated samples at 24 hrs post-treatment. Analysis includes A20, virus-negative cells and A20.γHV68 cells. (A) qRT-PCR analysis of TMER-5 expression relative to 18s RNA in A20 and A20.γHV68 (HE2.1) cells, untreated or stimulated with TPA for 24 hrs. (B) PrimeFlow detection of TMER expression in A20.γHV68 (HE2.1) cells, comparing either samples that were unstained (solid gray) or stained for the TMERs (open black line). (C) Analysis of TMER expression in multiple conditions, comparing untreated and stimulated A20 and A20.γHV68 cells, with gates defining the frequency of events that expressed either intermediate (mid) or high levels of TMERs. Data depict lymphocytes that were singlets, defined by sequential removal of doublets. (D) Quantification of the frequency of TMERhigh cells in stimulated A20.γHV68 cells. (E) Histogram overlays of HygroGFP and RCA protein expression in A20.γHV68 cells comparing TMERmid cells from untreated cultures (top, gray), TMERmid cells from TPA-stimulated cultures (middle, blue), with TMERhigh cells from stimulated cultures (bottom, red). (F) Quantification of the frequencies of HygroGFP+ (left) and RCA protein+ (right) cells as a function of TMER expression and treatment condition. (G) Flow cytometric analysis on an imaging flow cytometer, with each row showing an individual cell and representative images of brightfield (BF), RCA protein (RCA), DAPI, and TMER localization in TMERmid cells from untreated cultures (left) or TMERhigh RCA+ cells from stimulated (right) A20.γHV68 cells. Data are from two independent experiments, with biological replicates within each experiment for all A20.γHV68 cultures. Graphs depict the mean ± SEM, with each symbol identifying data from a single replicate. Statistical analysis was done using an unpaired t test (D) or one-way ANOVA, subjected to Tukey’s multiple comparison test (A, F), with statistically significant differences as indicated, * p<0.05, **** p<0.0001.
Fig 3.
Single-cell analysis of KSHV PAN RNA expression in the BCBL-1 B cell lymphoma cell line.
(A) Flow cytometric analysis of PAN RNA expression in multiple conditions, from cells incubated with no probe (left), or cells subjected to hybridization using a probe for PAN RNA, comparing virus-negative BL41 cells (second from left) with KSHV+ BCBL-1 cells that were either untreated or stimulated (with TPA and sodium butyrate (NaB)) for 72 hours. Data depict lymphocytes that were singlets, defined by sequential removal of doublets. Representative images were defined as samples that were closest to the median frequency. Quantification of (B) the frequency of PAN RNA+ cells and (C) PAN RNA median fluorescence within PAN RNA+ cells, comparing untreated or stimulated BCBL-1 cells. D, E) Flow cytometric analysis of untreated BCBL-1 cells, using histogram overlays, to compare cell size (forward scatter, FSC), granularity (side scatter, SSC) and ORF73 expression in cells that were either PAN RNA negative [–] (blue line) or PAN RNA positive [+] (red line). Data show (D) histogram overlays of these populations with fluorescence quantification provided in panel E. ORF73 analysis includes samples in which there was no ORF73 probe (i.e. “No Probe”, in solid gray), to define background fluorescence. (F, G) Flow cytometric analysis of stimulated BCBL-1 cells, using histogram overlays, to compare cell size (forward scatter, FSC), granularity (side scatter, SSC) and ORF73 expression in cells that are either PAN RNA negative [–] (blue) or PAN RNA positive [+] (red). Data show (F) histogram overlays of these populations with fluorescence quantification provided in panel G, with ORF73 analysis including a “No Probe” sample (gray) to define background fluorescence. Due to variable baseline fluorescence values for SSC and ORF73 between experiments, values were internally standardized to fluorescent intensities within the PAN RNA negative population for each experiment, with data depicting mean ± SEM. Symbols in panels B and C indicate values from individual samples. Data are from two independent experiments, with biological replicates within each experiment, with total number of biological replicates as follows: No Probe (n = 2), BL41 control (n = 3), BCBL-1 untreated (n = 6), BCBL-1 stimulated (n = 6). Statistical analysis was done using an unpaired t test with statistically significant differences as indicated, * p<0.05, ** p<0.01, **** p<0.0001.
Fig 4.
Single-cell analysis of EBV EBER expression in the Mutu I B cell lymphoma cell line.
(A) Flow cytometric analysis using the PrimeFlow method, to quantify EBER expression in multiple conditions, from cells incubated with no probe (left), or cells subjected to hybridization using a probe for the EBERs, comparing virus-negative BL41 cells (second from left) with Mutu I EBV+ cells that were either untreated (with DMSO) or stimulated (with TPA in DMSO) for 48 hours. Data depict lymphocytes that were singlets, defined by sequential removal of doublets. Representative images were defined as samples that were closest to the median frequency, with data depicting mean ± SEM. (B) Quantitation of median EBER fluorescence within EBER+ Mutu I cells in untreated and stimulated cultures. Horizontal dashed line indicates the background fluorescent signal from BL41 controls. Data are from two independent experiments, with biological replicates within each experiment, with total biological replicates as follows: No Probe (n = 2), BL41 Control untreated (n = 6), Mutu I untreated (n = 6), Mutu I stimulated (n = 6). Statistical analysis was done using an unpaired t test with statistically significant differences as indicated, * p<0.05.
Fig 5.
Actin mRNA degradation identifies virally-infected cells experiencing virus-induced host shutoff.
Actin mRNA analysis by qRT-PCR (A,D) or by flow cytometric analysis using PrimeFlow (B,C,E-G), comparing cells with variable infection status. (A) qRT-PCR analysis of beta-actin (Actb) mRNA expression relative to 18s RNA in mock, WT or TMER-TKO infected 3T12 fibroblasts at 18 hpi. (B) PrimeFlow analysis of actin mRNA in 3T12 fibroblasts, either unstained (“No probe”), mock-infected or infected with WT γHV68 or TMER-TKO at 18 hpi. (C) PrimeFlow analysis of actin mRNA and TMER expression in WT γHV68 infected fibroblasts at 18 hpi. (D) qRT-PCR analysis of beta-actin (Actb) mRNA expression relative to 18s RNA in A20, virus-negative cells and A20.γHV68 (HE2.1) cells, untreated or stimulated with TPA for 24 hrs. (E) PrimeFlow analysis of actin mRNA and TMER expression in untreated and stimulated A20.γHV68 cells, with the frequency of TMERhigh actin mRNAlow cells indicated, based on the gated events. (F,G) Actin mRNA analysis by PrimeFlow using either (F) histogram overlays or (G) quantifying frequencies, comparing A20.γHV68 cells that were either untreated or stimulated, further stratified by whether the cells were TMERmid or TMERhigh (using the gating strategy defined in Fig 2C). All flow cytometry data depict single cells, defined by sequential removal of doublets. Data are from two-three independent experiments, with biological replicates within each experiment. Graphs depict the mean ± SEM, with each symbol identifying data from a single replicate. Statistical analysis was done using one-way ANOVA, subjected to Tukey’s multiple comparison test (A, D, G), with statistically significant differences as indicated, ** p<0.01, *** p<0.001, **** p<0.0001.
Fig 6.
Heterogeneous viral gene expression at the single-cell level during lytic replication.
Viral, host RNA flow cytometric analysis in γHV68-infected fibroblasts at 16 hpi defined by the PrimeFlow method, comparing (A) X-shift clustering analysis and (B-E) biaxial gating analysis for the indicated features. (A) Automated, clustering analysis using the X-Shift algorithm on 10,000 events total, compiled from mock- and γHV68-infected fibroblasts at 16 hrs pi (1,000 events randomly imported per sample, mock infected n = 4, γHV68-infected n = 6) identifies multiple clusters of cells with differential gene expression (7 clusters, colored distinctly, “Cluster ID”), with these clusters then depicted for expression of TMERs, ORF73, Actin mRNA, and RCA. Range of expression is identified for each parameter. (B) Analysis of TMER and ORF73 co-expression in mock (left), WT γHV68-infected (middle), and TMER-TKO-infected (right) samples, with gates depicting populations with different gene expression profiles, defined relative to mock and TMER-TKO infected samples. (C) Color-coded populations from WT-infected sample in panel B, with each color indicating a different gene expression profile. (D) Histogram overlays of the five populations identified in panel C for the indicated parameters. (E) Quantitation of gene expression among the five populations identified in panel C, using the same color-coding strategy. Data are from three independent experiments, with each experiment containing biological replicates. Flow cytometry data shows single cells that are DNA+ (DAPI+). Statistical significance tested by one-way ANOVA, comparing the mean of TMERhigh ORF73high cells to all other means, followed by Dunnett’s multiple testing correction. Significance identified as *** p<0.001, **** p<0.0001.
Fig 7.
Measurement of viral gene expression heterogeneity by multiple parameters.
3T12 cells were infected with WT γHV68 at an MOI = 5 and analyzed at 16 hpi. (A) Histogram overlays comparing expression of the indicated parameters between 5 different populations of cells stratified by TMER and ORF73 expression, as defined on the leftmost panel, with data representative of three biological replicates. (B) Histogram overlays comparing expression of the indicated parameter in mock (black) or WT γHV68 -infected (red) samples, comparing three biological replicates with events defined as DAPI+ singlets. (C) Viral DNA as a function of viral gene expression defined by TMER and 73 expression. NIH 3T12 cells were either mock or virus infected (MOI = 5), harvested at 16 hpi, collected in bulk (mock and infected, no sort) or subject to FACS purification based on expression of TMER and 73 expression to isolate TMER- 73-, TMER+ 73- and TMER high 73 high populations. DNA was harvested from all samples, subjected to qPCR with primers to the gB gene of gHV68. DNA content was normalized relative to cellular DNA, defined by parallel amplification of the NFAT5 gene. Data depict mean ± SEM, with symbols indicating individual samples. Statistical analysis was performed using one-way ANOVA subjected to Tukey's multiple comparisons test, comparing viral DNA content between sorted cell populations, with statistical significance as identified, *** p<0.001. (D) Viral gene expression during primary lytic replication with gHV68 is relatively comparable between cells in different stages of the cell cycle stages. Data are from imaging flow cytometry-based studies, depicting the frequency of cells stratified by expression of TMERs and ORF73 (as defined in Fig 6D) present among total DAPI+ cells (top row) compared to cells in either the G0/G1, S or G2/M phase of the cell cycle. Data depict results from three independent experiments with 8 total replicates, with columns plotting mean ± SEM. Flow cytometry data depict single cells, defined by sequential removal of doublets according to SSC-A x SSC-H and FSC-A x FSC-H.
Fig 8.
Lytic replication is characterized by heterogeneous TMER localization and variable penetrance of actin RNA degradation.
Viral, host RNA flow cytometric analysis in γHV68-infected fibroblasts at 16 hpi defined by the PrimeFlow method. (A) The frequency of γHV68-infected fibroblasts with TMERs primarily in the nucleus was quantified by ImageStream, with data showing the frequency of cells in which TMER:DAPI colocalization (i.e. similarity score) was >1. (B) Images showing brightfield (BF), TMER, RCA protein (RCA), DAPI, ORF73 and actin mRNA localization, comparing cells with nuclear TMER localization (left) versus cytoplasmic TMER localization (right). (C) Analysis of cell subpopulations stratified by TMER and ORF73 expression (defined in Fig 6C), subjected to the tSNE dimensionality reduction algorithm. Data and cell populations are derived from the dataset presented in Fig 6, showing all DNA+ (DAPI+) single cells (FSC-A, SSC-A) subjected to the tSNE algorithm. The tSNE algorithm provides each cell with a unique coordinate according to its expression of Actin mRNA, RCA, ORF73, and TMERs, displayed on a two-dimensional plot (tSNE1 versus tSNE2). Visualization grid of tSNE plots, with plots arranged according to marker expression (rows) relative to phenotype of the cellular population examined (columns). (D) Biaxial analysis of actin RNA versus RCA protein, among the five populations identified in Fig 6. Flow cytometry data shows single cells that are DNA+ (DAPI+). Data in panels A-B from two independent experiments, in panel C-D from three independent experiments.
Fig 9.
Lytic viral gene expression is influenced by target cell type, changes over time, is enhanced by conditions which promote lytic replication.
Fibroblasts (3T12 cells or mouse embryonic fibroblasts, MEFs) were infected with WT or C-RTA γHV68 at 5 or 100 PFU/cell, and harvested at either 6 or 16 hpi as indicated. (A-C) Analysis of viral gene expression by PrimeFlow in either 3T12 or MEFs following WT γHV68 infection (MOI = 5) and harvested at 16 hpi. (A) Analysis of viral gene expression by PrimeFlow, comparing gene expression by histogram overlays between 3T12, WT MEFs and IFNAR KO MEFs. (B) The frequency of events stratified based on expression of ORF73 and TMER, or (C) based on all 16 possible combinations of gene expression with the frequency of events in each population plotted as a heatmap. Data are from three independent experiments harvested at the indicated time, subjected to fluorescent barcoding, pooled together and subjected to PrimeFlow analysis in a single analysis. (D-F) 3T12 fibroblasts were infected with WT or C-RTA γHV68 at 5 or 100 PFU/cell, and harvested at either 6 or 16 hpi as indicated. (D) Analysis of viral gene expression by PrimeFlow, comparing gene expression by histogram overlays between different virus infection conditions and times post-infection, as indicated. (E) The frequency of events stratified based on expression of ORF73 and TMER, or (F) based on all 16 possible combinations of gene expression with the frequency of events in each population plotted as a heatmap. Data are from three independent experiments harvested at the indicated time, subjected to fluorescent barcoding, pooled together and subjected to PrimeFlow analysis in a single analysis. Data depict mean values, with SEM plotted in panels B, D, and heatmap values colored based on the categorical key provided in the figure. Flow cytometry data depict single cells, defined by sequential removal of doublets according to SSC-A x SSC-H and FSC-A x FSC-H, after which files were subjected to debarcoding to identify each of the independent samples in the analysis.