Demonstration that pre-mRNA molecules dispersed in the nucleus are capable of being spliced. A. Upper panels show composite images of cells in which the gene containing array 2 intron within an intron was induced for a brief period (2 hour). Most molecules are present within the nuclei. Lower panels show images from the same batch of cells as above, but in which induction was followed by a period of suppression (2 hour). Raw images are shown on the left and overlays with colored balls identifying the RNA species are presented on the right. B. Percentage of three different RNA species in individual cells as a function of time after the addition of doxycycline. Doxycycline turns off new RNA synthesis within several minutes. Even though the proportion of spliced mRNAs continues to increase after 3 hour, the overall number of RNAs declines due to degradation.

Engagement of an intron’s polypyrimidine tract in a strong double-stranded structure increases its tendency to splice post-transcriptionally. A. Sequences upstream of the branch point sites (red) in the array 1 construct and c-Fos intron 1 were altered so that they would form a 13-nucleotide stem with their respective polypyrimidine tracts. B. Percentage of pre-mRNAs in the nucleus for three array 1 constructs: unmodified array 1; array 1 in which an upstream sequence forms a hairpin stem with the polypyrimidine tract; and array 1 in which the polypyrimidine tract is modified by replacing two thymidine residues with adenosines. C. Distribution of c-Fos pre-mRNA and its spliced products resulting from different sequence modifications. The upper panels show merged raw images with green representing the signal from intron probes, and red representing the signal from exon probes. The lower panels identify the molecules on DIC images. D. Number of pre-mRNA (excluding the clusters at the gene loci) as a percentage of total cellular RNA in cell lines of different constructs. Representative images corresponding to the insertions of the arrays into intron 1 are shown in Fig. S3.

nPTB pre-mRNA molecules containing intron 9 are released into the nucleoplasm more frequently from its gene locus than other introns in the same gene. A. Non-neuronal cells express high levels of PTB (blue ovals), which bind to pyrimidine-rich sequences in and around exon 10 and cause the skipping of exon 10, resulting in an mRNA species that contains a premature termination codon and is a substrate of non sense mediated decay. In neuronal cells, there is little PTB, and only constitutive splicing can occur, which produces larger amounts of functional mRNA (Fig. S6.). B. Representative images from HeLa cells showing the locations of pre-mRNAs containing the indicated intron and 3’ terminal exon 13. The panels on the right show the location of each type of molecular species. The average number of molecules per cell are provided in Table 1. C. Kinetics of splicing and degradation of pre-mRNAs containing intron 9 and exon 13 after the addition of actinomycin D to the culture medium. The yellow line is an exponential decay curve fitted to the observed decrease in the pre-mRNA population. The red line is a fitted curve based on modeling of reaction: pre-mRNA → mRNA → Φ.

Imaging the intracellular distribution of single-molecules of pre-mRNAs and their spliced products expressed from a pair of reporter genes integrated into the genome of CHO cell lines. A. Schematic depiction of two reporter genes. Array 1 contains 32-repeats of a 49 nucleotide sequence, array 2 contains 96 repeats of a 50 nucleotide sequence, and array 3 contains 32 repeats of a 80 nucleotide sequence. B. Images of cells from clones expressing the reporter gene in two fluorescence channels. The targets of the probes are indicated on the top of the panels, and the array within the intron is shown on the left. All images in this report are maximum intensity merges of 3-dimensional optical slices. In the composite images, red represents either the 3’-UTR or the coding sequence, and green represents the introns. C. Identification of RNA species using our image-processing algorithm. Circles of different colors are drawn around each detected spot. D. Percentage of three different species in the nuclei of individual cells from two cell lines. Examples of regions from which the counts were obtained are indicated by blue circles in C. Error bars represent 95% confidence intervals. Images from 15–50 cells were analyzed to obtain the data in this report. The probability of obtaining the indicated (or higher) percentage of unspliced pre-mRNA of array 2 by chance is < 10⁻⁶. Scale bars are 5 µm in all figures in this report.

Introns containing array 1 and 2 are spliced co-transcriptionally and post-transcriptionally irrespective of the order of their appearance in the gene. A. Schematic representation of two genes that contain the two introns in a different order. Both array 1 and array 2 have 32 repeats each, and they are embedded within the same splice sites. B. Raw, composite, and interpreted images of a cell expressing the construct array2-array1 in the three different channels that correspond to the coding sequence, array 1, and array 2. The composite shows the GFP coding sequence in red, array 1 in blue, and array 2 in green. The color key on the right lists each of the seven combinations of spliced and unspliced RNA species that can occur. Images for the construct array1–array2 are not shown as it yielded similar results (Table S1).

Sex-specific alternative splicing of Sxl pre-mRNA occurs post-transcriptionally in female Drosophila cells. A. Sxl protein (green oval), present only in females, binds to polypyrimidine tracts in introns 2 and 3 and results in a splicing pattern in which intervening exon 3, which contains a stop codon (red triangle) is not included in the final spliced product. Males have no Sxl protein and constitutive splicing yields a transcript possessing a premature termination codon. B. Images of spots produced by probes against the indicated components of Sxl transcripts in female (Kc167) and male (S2) cells. When the centers of spots in all three channels were within the molecular co-localization distance (Fig. S2), they were interpreted to be originating from a single pre-mRNA molecule containing intron 2, intron 3 and exon 8. The locations of the identified molecules are shown on DIC images. C. Frequency with which the given numbers of pre-mRNA molecules were found in a group of 50 cells. After a 30 minute exposure to actinomycin D no pre-mRNAs could be detected in the cells.