Viral mRNA detection by single-colour RNA FISH.

(A) A schematic diagram of mRNA labelling for the diffraction-limited imaging. Each mRNA was hybridised with a tile of oligonucleotide, each labelled with a single fluorophore at the 3′ end. (B) A representative field-of-view of the microscope image in which NP mRNAs were stained with Quasar 670. White, smFISH signal; blue, nucleus stained with DAPI. The red square indicates the area for the sequential close-up of the image. The signal intensity is shown on the same grey scale across the three panels, indicated on the right. (C) Spot detection in 3D. Spots are projected to either the XZ or YZ plane to indicate the distribution of spots in a typical 3D cell volume. The colour of each spot indicates the Z position of that spot. Spot projection to the XY plane at the bottom (magenta) highlights the XY positions of spots relative to the nucleus boundary (blue), that is, nucleus projection on the XY plane. The grid interval on the X and Y axis is ca. 18.3 µm. (D) Total count of mRNAs in each cell. The sample size (n) indicates the number of cells segmented in the microscope images. (E) Intracellular distribution of viral mRNAs. Spots in the 3D volume were projected on the XY plane, together with the cytoplasmic boundary (black solid line) and the nucleus (blue solid line). The grid interval indicates ca. 18.3 µm. (F) Spot segmentation within cells. The panel on the left shows the entire field-of-view (measures ∼294 µm), and the panel on the right shows the close-up of the area indicated in the red square (∼73 µm). Spots within the 2D nuclear projection are indicated in blue (nuclear); spots within the expanded nuclear boundary twofold from the centre of mass of the nucleus in 2D are indicated in orange (perinuclear); and spots outside this expansion are indicated in green (peripheral). (G) Distribution of spots on the Z axis in the three segments (nuclear, perinuclear and peripheral) presented in panel F. The red dashed line shows the median Z coordinates of the peripheral spots, used to indicate the cell baseline. The number in the histogram for nuclear spots (0.731) indicates the fraction of spots that were above the cell base line in the field-of-view presented in panel F. (H) Fraction of spots above the cell baseline in the nuclear segment. The black dots indicate the fraction obtained in each field-of- view, and the blue bar indicates the mean. (I) Fraction of spots within the 2D boundary of the nucleus. The sample size is the same as in panel D.

Viral mRNA detection using multiplex RNA FISH.

(A) An overview of the assay. The viral mRNAs were hybridised with the encoding probes, and then the amplification probes were annealed (not shown in the diagram for simplicity). A flow cell was assembled in which the readout probes were hybridised and the cells were imaged in the three sequential rounds of hybridisation and imaging. Subsequently fluorescent signals were decoded according to the encodings listed on the right. (B) A representative mapping of identified viral mRNAs on the segmented cell and nuclear boundary in the entire field-of-view (left) and the close-up in the red square (right). The mesh grids indicate ∼34-µm interval on the left, and ∼17-µm interval on the right. (C) Total count of mRNAs in 204 single cells (top row) and the nuclear fraction (bottom row) for each segment. (D) Heatmap representation of mRNA counts in single cells. The viral segments are in rows, cells in columns. The cells were clustered column-wise to highlight the inequality in the abundance of transcripts carried per cell. The colour code for the absolute count is indicated in the vertical bar on the right. (E) A pair-wise analysis of the transcript count between segments.

Cell-to-cell heterogeneity for time-course inference.

(A) A theoretical diagram of how viral mRNAs emerge in single cells in the population. The font runners designate ones in which the viral transcription was initiated earlier than others, and the slow starters are ones that started transcription lately. The bottom plot shows the theoretical relationship between the nuclear ratio and the abundance of transcripts in which the front runners and slow starters (and many more along the spectrum in between) compose the clusters as indicated. (B) Scatter plots of nuclear fraction against the total count in each cell, derived from the dataset presented in Fig. 2C. (C) Time-course of the distribution of NP mRNA count per cell, measured by smFISH, at the indicated time points (top row); and the change in the nuclear fraction size, plotted against the total number of transcripts (bottom row). The sample size (n) indicates the number of cells observed at each time point.

Kinetic model for estimating the nuclear export rate.

(A) A schematic diagram representing the production of viral transcripts in the nucleus and the export to the cytoplasm. The two spherical compartments are for the nucleus and the cytoplasm. Viral transcripts (magenta) are produced at a rate constant k in the nucleus, and exported at a rate constant ε to the cytoplasm. (B) The relationship between the nuclear fraction and the total count in the kinetic model, according to the formula presented in panel A. The parameter α determines the curvature of the plot. (C) The parameter fit to the observed datapoints. The parameter α was estimated by fitting the equation to the observed datapoints presented in Fig. 3B. The colour representation in each line is the same as the ones used in panel B, according to the parameter α. (D) Estimated nuclear export rate. The parameter ε was estimated from the three independent assays (indicated in black for each segment); the blue bar indicates the mean of the three measurements.

Independent validation of the total mRNA count and the nuclear ratio of each segment by smFISH.

(A) A schematic diagram of in-house probe synthesis. The pool of single-stranded DNA templates were PCR-amplified, T7-transcribed, and reverse-transcribed using the fluorescent-labelled primer. The number of distinct probes contained in the pool is indicated for each segment. For HA, NA, M and NS segments, two additional fluorescent probes bind to each probe. As a result, the number of fluorophores bound per mRNA molecule is theoretically three times as many as the probes for those segments. (B) Total count of mRNAs per cell for each segment. The sample size (n) indicates the number of cells analysed. (C) The nuclear ratio of mRNAs in each cell. The sample size is the same as the one indicated in panel B.

Independent validation of the nuclear ratio against the total count of mRNAs for each segment by smFISH.

The nuclear ratio of MRNAs in each cell was plotted against the total number of MRNAs in that cell. The original data were derived from Supplementary figure 1B and C.

Exclusive detection of unspliced NS transcripts.

(A) Probe-binding sites on the NS segment. The exon probes, used in Fig. 1, target the second exon of NS2 (spliced form), and the intron probes are against the region that is spliced out. Thus, exon probes detect both NS1 and NS2, while intron probes exclusively detect NS1. The black line represents the NS gene. The protein coding region is indicated with the wide horizontal bar. (B) The total spot count (left column) detected using either exon probes or intron probes. The spot nuclear ratio in each cell is indicated on the right. Note that the plots of the exon probes (orange) are the reproduction from Fig. 1 and are presented for comparison with those of the intron probes (magenta).