Fig 1.
Viral mRNA comprises 60% of the cytoplasmic mRNA at 6 hours post-infection.
(A) Schematic of experimental design. HeLa cells were infected with VSV at a MOI of 10 and cytoplasmic extracts were prepared at 2 and 6 hpi for mRNA isolation and polysome profiling. Messenger RNA was isolated from fractions corresponding to 80S monosomes, or polysomes containing 3 or more ribosomes, and used for deep sequencing. (B) Polysome analysis of uninfected (black) or VSV (red) infected HeLa cells. Cytoplasmic extracts were sedimented through a 10–50% sucrose gradient and 0.5 ml fractions were collected while continuously monitoring absorbance at λ = 254nm. (C) Distribution of fragments mapping to the concatenated hg38 (human) and VSV genomes for cytoplasmic, monosome, and polysome samples at 2 and 6 hpi. Trimming and mapping was performed in CLC Genomics Workbench. (D) Distribution of reads among the 5 viral genes at 2 and 6 hpi. Expression level is presented as Transcripts per Kilobase Million (TPM) to normalize for gene length and library size, error bars denote standard deviation from two biological replicates.
Fig 2.
The relative abundance of individual cellular mRNAs in the cytoplasm and on polysomes decreases between 2 and 6 hpi.
(A) Scatter plots of Transcripts per Kilobase Million (TPM) for exonic regions at 2 hpi. The TPM in uninfected cells for a given mRNA is graphed on the abscissa, and the TPM for a given cellular mRNA in either uninfected (black circles), or VSV infected cells (gray circles) is graphed on the ordinate. The viral mRNAs are indicated by the red triangles. (B) Density plots of the log2 fold change in TPM for cellular mRNAs between uninfected or VSV infected cells at 2 hpi. C, M, and P denote “cytoplasm”, “monosome”, and “polysome”, respectively. (C) Scatter plots of TPMs for individual mRNAs at 6 hpi, presented as in A. (D) Density plots of the log2 fold change in TPM at 6 hpi, presented as in B.
Fig 3.
Changes in monosome and polysome association of cellular mRNAs following VSV infection.
(A) Density plots show the log2 fold change for any given mRNA in both monosome and polysome fractions at 2 hpi. The region of the density within the 95% confidence interval of the mean is shaded gray. Magenta lines denote ± 2 standard deviations of the mean. Regions with increased association relative to the mean are shaded blue, and regions with decreased association are shaded green. (B) Analysis at 6 hpi, presented as in A. Downstream analyses were performed on genes in the green or blue regions. (C) Plots of log2 fold changes in association at 2 hpi plotted against cytoplasmic abundance in infected cells. Magenta lines denote ± 2 standard deviations of the mean log2 fold change. Genes outside the 95% confidence interval are denoted by blue (increased) or green (decreased) dots. (D) The log2 fold changes in association at 6 hpi as presented in C. (E) Gene ontology analysis for mRNAs with increased polysome association, analysis was performed using GOseq in R. The pie charts shown represent the distribution of mRNA among the 20 most significant GO Terms. (F) Gene ontology analysis for mRNAs with decreased polysome association as determined using GOseq in R.
Fig 4.
Polysome-associated cellular mRNAs are longer and more AU-rich.
(A) Analysis of cellular mRNAs with high cytoplasmic abundance (purple) or low cytoplasmic abundance (orange) as compared to mRNAs with cytoplasmic abundance within 2 standard deviations of the mean abundance (gray) in uninfected cells. Cytoplasmic abundance by TPM is from the data set published with this paper. ***p< 2.2 x 10−16; **p< 5.0 x 10−5; *p< 0.05; all others p> 0.05 as determined by the Wilcoxon rank sum test compared to mRNAs with relative abundance levels within the 95% confidence interval of the mean. Hinges correspond to the 25th-75th percentiles, and whiskers denote 1.5 times the inter-quartile range. (B) Analysis as in A for cytoplasmic abundance in infected cells. (C-F) mRNA characteristics for mRNAs with increased polysome association (blue) or decreased polysome association (green) at 6 hpi, as defined in Fig 3. Data for RNA half-life and poly(A) tail length were from [54, 55]. Analysis was performed as in A.
Fig 5.
Effect of suppressing specific host gene function on viral gene expression.
(A) GFP expression of a rVSV-eGFP reporter virus in uninfected, vehicle-treated, or 2.5 μM 17-DMAG treated HeLa cells. A representative histogram of GFP intensity is shown to the right, and the mean fluorescence intensity (MFI) of live cells is shown below. MFI is normalized to vehicle, and error bars denote the standard deviation from the mean of 3 independent replicates. (B) VSV gene expression in eIF3a depleted HeLa cells. A representative histogram of fluorescence intensity is shown to the right and the MFI below. Error bars denote standard deviation from the mean from three independent replicates. Luciferase expression driven by rVSV-LUC (Firefly Luciferase) or a cellular reporter driving expression of Renilla Luciferase (pRL-CMV) in eIF3a depleted cells. Luciferase expression is presented as the percent of a non-targeting siRNA control, and error bars denote the standard deviation from 3 independent replicates. Metabolic labeling of eIF3a depleted HeLa cells infected with wild-type VSV. The position of viral proteins is noted to the right. Presented is a representative gel from two independent replicates. (C) VSV gene expression and replication in DDIT4 depleted cells. A representative histogram of eGFP expression is shown along with the MFI of cells normalized to a non-targeting siRNA control. Error bars denote the standard deviation from the mean of three independent replicates. For metabolic labeling a representative gel of two independent replicates is presented. Kinetics of viral replication were measured by titration of yields at various times post infection of siRNA treated HeLa cells. Error bars denote the standard deviation from the mean of 3 independent replicates. (D) Immunoprecipitation of cellular proteins synthesized post-infection with VSVWT. Shown is a representative gel from two independent replicates. A quantitative analysis of the HSP70 and YBX1 bands is shown in the bottom two panels, error bars denote the standard deviation from two independent replicates.
Fig 6.
Altered distribution of cellular mRNA within polysomes following infection.
(A) The monosome or polysome association of cellular mRNAs at 6 hpi, as determined by RNAseq. Genes picked for validation by polysome profiling and qPCR are highlighted in orange. (B-E) The distribution of mRNA in polysome profiles from uninfected or infected HeLa cells at 6 hpi. A representative polysome trace from infected cells at 6 hpi is shown in light gray, and the mRNA polysome distribution in uninfected cells is shown in black, and infected cells in red. The RNA distribution is presented as the fraction of the total recovered. Error bars denote the standard deviation from three independent replicates. (B) Distribution of VSV N and G mRNA. (C) RNAs with increased polysome association by RNAseq. (D) RNAs with unchanged polysome association following infection. (E) mRNAs with decreased polysome association.
Fig 7.
Transcripts from viral mutants defective in cap methylation are translated efficiently in infected cells.
(A) Polysome distribution of VSV N and G mRNAs at 6 hpi with wild-type VSV. Results of qPCR for each individual polysome fraction are presented as a fraction of the total recovered, in red. Error bars denote the standard deviation from two independent replicates, and a representative polysome profile from infected cells is shown in gray. (B) Polysome distribution of VSV N and G mRNAs at 6 hpi with a VSV mutant deficient in guanine N-7 cap methylation, LG1670A. (C) Polysome distribution of VSV N and G mRNAs at 6 hpi with a VSV mutant, LG4A, defective in both guanine-N-7 and ribose-2′-O cap methylation. (D) Western blot showing depletion of eIF4E by PPMO. “Scr” and “4E” denote “scramble” or “eIF4E” PPMO treated cells, respectively. (E-F) Rates of viral protein synthesis in PPMO treated VSVWT or VSVG1670A-infected cells following a 10-minute pulse with [35S] methionine. A quantitative analysis is presented as a ratio of the rate of synthesis in eIF4E-depleted cells to scramble PMO treated cells. Error bars denote the standard deviation from the mean of two independent replicates.