LSE Logo MBoC Logo

Splicing Speckles Are Not Reservoirs of RNA Polymerase II, but Contain an Inactive Form, Phosphorylated on Serine2 Residues of the C-Terminal Domain

Published Online:


“Splicing speckles” are major nuclear domains rich in components of the splicing machinery and polyA+ RNA. Although speckles contain little detectable transcriptional activity, they are found preferentially associated with specific mRNA-coding genes and gene-rich R bands, and they accumulate some unspliced pre-mRNAs. RNA polymerase II transcribes mRNAs and is required for splicing, with some reports suggesting that the inactive complexes are stored in splicing speckles. Using ultrathin cryosections to improve optical resolution and preserve nuclear structure, we find that all forms of polymerase II are present, but not enriched, within speckles. Inhibition of polymerase activity shows that speckles do not act as major storage sites for inactive polymerase II complexes but that they contain a stable pool of polymerase II phosphorylated on serine2 residues of the C-terminal domain, which is transcriptionally inactive and may have roles in spliceosome assembly or posttranscriptional splicing of pre-mRNAs. Paraspeckle domains lie adjacent to speckles, but little is known about their protein content or putative roles in the expression of the speckle-associated genes. We find that paraspeckles are transcriptionally inactive but contain polymerase II, which remains stably associated upon transcriptional inhibition, when paraspeckles reorganize around nucleoli in the form of caps.


The cell nucleus is a complex organelle that harbors chromosomes organized into territories as well as many subcompartments rich in protein complexes with roles in DNA and RNA metabolism. “Splicing speckles,” or interchromatin granule clusters at the ultrastructural level, are major nuclear domains that are rich in components of the splicing machinery and polyA+ RNA (Lamond and Spector, 2003), but they also contain other nuclear proteins with roles in RNA metabolism and transcription, including RNA polymerase II (pol II) and CDK9 (Mintz et al., 1999; Herrmann and Mancini, 2001). Speckles are often found spatially associated with “paraspeckle” domains, which are marked by paraspeckle proteins (PSPs) 1 and 2 (Fox et al., 2002). The protein content of paraspeckles suggests that they are involved in RNA synthesis or processing; PSPs interact with the splicing and transcription factor-associated proteins p54nrb/NonO and PSF, respectively, which in turn interact with pol II (Emili et al., 2002; Kameoka et al., 2004; Myojin et al., 2004; Fox et al., 2005).

The discovery that the association of splicing factors and polyA+ RNA with speckles is highly dynamic has suggested that these domains are metastable compartments (Kruhlak et al., 2000; Phair and Misteli, 2000; Molenaar et al., 2004). However, speckles remain stable in unfixed permeabilized cells for many hours and resist biochemical purification (Mintz et al., 1999), such that some of their components are likely to be structural. Although speckles are mostly devoid of chromatin and are not major sites of transcription, as measured after incorporation of labeled nucleotides into nascent RNAs (Puvion and Puvion-Dutilleul, 1996; Iborra et al., 1998; Cmarko et al., 1999), specific mRNA-coding genes and gene-rich R bands have been found preferentially associated with them (Clemson and Lawrence, 1996; Johnson et al., 2000; Shopland et al., 2002, 2003; Moen et al., 2004). That unspliced mRNAs are present and further accumulate upon splicing inhibition (Johnson et al., 2000) suggests a role for speckles in gene expression and possibly posttranscriptional splicing of pre-mRNAs for which splicing is not completed cotranscriptionally.

Messenger RNAs and some small structural RNAs are transcribed by pol II. Its largest subunit (RPB1) contains an unusual C-terminal domain (CTD) with multiple tandem heptad repeats of consensus sequence Tyr1Ser2Pro3Thr4Ser5Pro6Ser7 (52 repeats in mammals), which is required for RNA processing (Maniatis and Reed, 2002; Zorio and Bentley, 2004) and for recruitment of splicing factors to sites of transcription (Misteli and Spector, 1999). Phosphorylation of the CTD correlates with steps of the transcription cycle; hypophosphorylated pol II is thought to bind to promoters, becoming phosphorylated on Ser5 residues during transcription initiation and on Ser2 once elongation begins (Kobor and Greenblatt, 2002). Several studies have reported enrichment of Ser2P pol II in splicing speckles relative to the nucleoplasm (Bregman et al., 1994, 1995; Bisotto et al., 1995; Mortillaro et al., 1996; Zeng et al., 1997; Patturajan et al., 1998; Mintz and Spector, 2000), proposing a model in which speckles serve as reservoirs of polymerase complexes that are mobilized to supply sites of active transcription. In contrast, other reports indicate a homogeneous distribution of Ser2P pol II throughout the nucleoplasm (Grande et al., 1997; Zeng et al., 1997). These different results may depend on experimental differences such as the antibodies used and cell preparation conditions (Guillot et al., 2004). Although, live cell experiments using enhanced green fluorescent protein (EGFP)-tagged RPB1 reveal a homogeneous distribution of pol II throughout the nucleoplasm (Sugaya et al., 2000; Kimura et al., 2002), these experiments do not exclude the possibility that some forms of the enzyme might be enriched, or cycle through splicing speckles.

Here, we have examined the role of splicing speckles as reservoirs of pol II by systematic localization of the differently phosphorylated forms of pol II in the nucleus of control and transcriptionally inhibited HeLa cells. We show that splicing speckles are not major sites of pol II storage, although they contain a small population of Ser2P pol II complexes that are stably bound but transcriptionally inactive. We also characterize the pol II content and transcriptional activity of the speckle-associated paraspeckle domains.


Cell Culture, Bromo-UTP–labeled RNA (Br-RNA) Labeling, Fixation and Ultrathin (Tokuyasu) Cryosectioning

HeLa cells were grown in 5% CO2 at 37°C in DMEM, supplemented with 10% heat-inactivated fetal calf serum, 2 mM l-glutamine, 100 IU/ml penicillin, 100 μg/ml streptomycin, and 20 mM HEPES (all from Invitrogen, Paisley, United Kingdom). α-Amanitin (α-am) (50 μg/ml; Sigma Chemical, Poole, Dorset, United Kingdom) was added to cultures for 5 h before fixation or whole cell extract preparation.

To label sites of transcription, cells were permeabilized with saponin in a “physiological buffer” containing bovine serum albumin (BSA), and incubated (15 min) in 0.1 M Br-UTP and the other NTPs before fixation (Pombo et al., 1999b). Cells were fixed in 4% freshly depolymerized paraformaldehyde in 250 mM HEPES (pH 7.6; 10 min; 4°C), refixed in 8% paraformaldehyde in 250 mM HEPES (pH 7.6; 2 h; 4°C), scrapped, and pelleted during the latter fixation (Pombo et al., 1999a). Pellets were embedded (2 h) in 2.1 M sucrose in phosphate-buffered saline (PBS), transferred onto a copper block, frozen, and stored in liquid nitrogen. Cryosections (100–150 nm in thickness) were cut with a glass knife using an ultracryomicrotome (UltraCut UCT52; Leica, Milton Keynes, United Kingdom), captured on drops of 2.1 M sucrose in PBS, transferred to 0.17-mm-thick coverslips for immunofluorescence, and used immediately or stored at –20°C.


Primary antibodies to the largest subunit of pol II were as follows. H224 was raised against recombinant protein corresponding to amino acids 1-224 at the amino terminus (rabbit IgG; 1:100 [Western blot; WB] and 1:10 [immunofluorescence, IF]; Santa Cruz Biotechnology, Santa Cruz, CA). ARNA-3 recognizes an epitope delineated by amino acids 794-822 outside the CTD (mouse IgG; 1:500 [WB] and 1:50 [IF]; Research Diagnostics, Flanders, NJ]. 8WG16 preferentially recognizes nonphosphorylated heptads of the CTD (mouse IgG; 1:100 [WB and IF]; BioDesign International, Saco, ME; Patturajan et al., 1998; Cho et al., 2001). Although some 8WG16 batches have been reported to contain significant cross-reactivity to faster migrating bands (Bregman et al., 1995), the one used here shows only a relatively minor reaction with an additional band (Figure 1B, arrow). Rabbit anti-Ser5P recognizes phosphorylated Ser5 peptides (rabbit IgG; 1:3000 [WB] and 1:1000 [IF]; Schroeder et al., 2000]. 4H8 was raised against phosphorylated Ser5 (mouse IgG; 1:2 × 105 [WB] and 1:500 [IF]; Upstate Biotechnology, Lake Placid, NY]. H5 recognizes phosphorylated Ser2 (mouse IgM; 1:500 [WB] and 1:2000 [IF]; Covance, Princeton, NJ; Patturajan et al., 1998; Komarnitsky et al., 2000). H5 cross-reactivities with Ser5P peptides and phosphorylated SR proteins have been described in highly purified systems (Cho et al., 2001; Doyle et al., 2002; Licatalosi et al., 2002; Jones et al., 2004) but do not seem to contribute to the staining observed in whole cells or extracts (see Discussion). Other available Ser2P antibodies available so far are not specific for pol II (CC-3; Lavoie et al., 1999) or also recognize Ser5P epitopes (mara-3 and B3; Patturajan et al., 1998; Cho et al., 2001).

Primary antibodies to splicing factors were as follows: SC35 recognizes a phosphorylated epitope within the serine/arginine (SR) domain of the spliceosome assembly factor, a member of the SR family of non-small nuclear ribonucleoprotein-splicing proteins (mouse IgG1; 1:1500; Sigma Chemical; Fu and Maniatis, 1990); Sm recognizes Sm protein, a spliceosomal factor (human autoimmune antibodies; 1:2000; ANA-CDC).

Other primary antibodies were mouse anti-BrdU (1:1000; Caltag Laboratories, Burlingame, CA) and sheep anti-bromodeoxyuridine (BrdU) (1:200; BioDesign International); and rabbit anti-PSP1 (1:150; Fox et al., 2002). Secondary antibodies conjugated with fluorescein isothiocyanate (FITC) or Cy3 were obtained from Jackson ImmunoResearch Laboratories (West Grove, PA) (multiple labeling grade), with AlexaFluor488 or AlexaFluor555 from Molecular Probes (Eugene, OR), and with horseradish peroxidase (HRP) from Promega. For double-labeling using an IgM (H5) and an IgG (SC35; Figure 3, J and K), secondary antibodies raised in donkey against μ- or γ-immunoglobulin chains (Jackson ImmunoResearch Laboratories) were used, respectively. Control experiments in the absence of each primary antibody showed no detectable cross-reaction (our unpublished data).

Western Blotting

Total HeLa cell protein extract was prepared by harvesting HeLa cells in SDS sample buffer (10% Ficoll, 2.5 mM EDTA, 25 mM Na3PO4, 0.5% SDS, 0.1% bromphenol blue, 10% β-mercaptoethanol, 1 mM NaF, and 1 mM phenylmethylsulfonyl fluoride) and by passing three times through a 21-gauge needle. Denatured total protein extracts (2 × 105–3 × 105 cells/well) were separated by SDS-PAGE on 4–15% Tris-HCl gradient gels (Criterion precast gel system; Bio-Rad, Hertsfordshire, United Kingdom) and transferred to nitrocellulose membranes (Bio-Rad). Membranes were blocked (1 h) in 5% milk in Tris-buffered saline/Tween 20 (TBS-T) buffer (10 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 0.05% Tween 20), incubated (2 h) in primary antibodies, washed (30 min), incubated (1 h) in HRP-conjugated secondary antibodies, and washed (30 min) all in 5% milk in TBS-T. Blots were washed in TBS-T, before detection of the HRP signal with ECL detection reagent (GE Healthcare, Little Chalfont, Buckinghamshire, United Kingdom), and exposed to Hyperfilm ECL (GE Healthcare). Scanned images were contrast stretched in Adobe Photoshop (Adobe Systems, Edinburgh, United Kingdom).

Immunofluorescence Labeling

Cryosections were rinsed (3 times) in PBS, incubated (30 min) in 20 mM glycine in PBS, rinsed (3 times) in PBS, treated (10 min) with 0.1% Triton X-100 in PBS, blocked (1 h) with PBS+ (PBS supplemented with 1% BSA, 0.2% fish skin gelatin, and 0.1% casein; pH 7.6), incubated (2 h) with primary antibodies (in PBS+), washed (3 times; 1 h) in PBS+, incubated (1 h) with secondary antibodies against mouse, rabbit, or human IgGs in PBS+, rinsed (3 times; 30 min) in PBS+, rinsed (3 times) in PBS, counterstained (45 min) with TOTO-3 (2 μM; Molecular Probes) in 0.05% Tween 20 in PBS, and rinsed successively in 0.05% Tween 20 in PBS and then PBS, before coverslips were mounted in VectaShield (Vector Laboratories, Peterborough, United Kingdom). The long incubation times used allow for antibody accessibility throughout the thickness of cryosections, providing the highest sensitivity (Branco et al., 2005). Negative controls included omission of primary antibodies and treatment of cryosections with alkaline phosphatase (AP) (2 h; 37°C) (0.5 U/μl, calf intestinal; New England Biolabs, Hitchin, Hertsfordshire, United Kingdom) before immunolabeling, to test the specificity of anti-pol II antibodies to phosphorylated epitopes.

Double labeling experiments were generally carried out simultaneously by combining both primary antibodies and both secondary antibodies, except when testing for the presence of pol II or Br-RNA within paraspeckles. In this case, to prevent false negative results because of antibody blocking, sections were first incubated with primary antibody against pol II or Br-RNA, washed, incubated with secondary antibody, washed (30 min) and incubated with primary antibodies anti-PSP1, and corresponding secondary antibodies.

Figure 1.

Figure 1. Characterization of antibodies against different forms of RNA polymerase II. (A) Schematic diagram for the binding sites of different antibodies against total and phosphorylated forms of the largest subunit of pol II (RPB1; n, number of CTD repeats). (B) Western blot analyses of HeLa whole cell extracts with antibodies against different forms of the largest subunit of pol II. Arrows indicate minor cross-reactivities. (C–L) The different forms of RPB1 are found in discrete sites throughout the nucleoplasm (dotted line), mostly outside nucleoli (arrows), with a minor fraction of unphosphorylated pol II found in the cytoplasm (arrowheads). Insets represent nucleic acid staining with TOTO-3. Total pol II was detected with antibodies H224 (C) and ARNA3 (D) against non-CTD epitopes of RPB1 in cryosections from HeLa cells. Pol IIA was detected with 8WG16 (E and F); pretreatment of sections with AP increases 8WG16 immunolabeling intensity by 1.5 times (F). Phosphorylated Ser5 pol II was detected with rabbit anti-Ser5P (G) and 4H8 (I) antibodies; absence of signal after pretreatment of sections with AP (H and J) shows that both antibodies bind specifically to phosphorylated epitopes. Phosphorylated Ser2 pol II was detected with H5 (K); H5 immunolabeling signal is also sensitive to AP (L). Bar, 2 μm.

Fluorescence In Situ Hybridization to PolyA+ RNA

PolyA+ RNA was detected with an FITC-labeled poly(dT) probe (Sigma Chemical) in hybridization mixture (Sigma Chemical) to give 1 μM final concentration. RNA-fluorescence in situ hybridization was performed with molecular biology grade reagents (RNAse- and DNAse-free). After immunolabeling, sections were refixed (2 h) with 4% paraformaldehyde in PBS and washed (16 h) in PBS. Sections were incubated (15 min) with 2 mM vanadyl ribonucleoside complex (Sigma Chemical) in PBS (15 min), in 20 mM glycine in DEPC-treated PBS (10 min) with 0.1% Triton X-100, in DEPC-phosphate-buffered saline, and then washed (4 times) in 2× SSC. Hybridization mixture (5 μl) containing FITC-poly(dT) was denatured (10 min; 70°C), applied on Hybrislips (Molecular Probes) before covering with coverslips (10 mm in diameter) containing sections, sealed with rubber cement (Marabuwerke, Tamm, Germany), and hybridized for >60 h at 42°C. Coverslips were then washed in 2× SSC (2 times; 1 h; 20°C), in 1× SSC (2 times; 15 min), and in 0.5× SSC (15 min), and incubated (30 min) with casein blocking solution (containing 2.6% NaCl, 0.5% BSA, and 0.1% fish skin gelatin, pH 7.5–8.0; Vector Laboratories). The FITC-poly(dT) signal was amplified (2 h) with AlexaFluor488 rabbit anti-FITC (1:500; Molecular Probes), washed (3 times; 1 h), incubated (1 h) with rabbit anti-AlexaFluor488 (1:1000; Molecular Probes), washed (3 times; 1 h), and incubated (1 h) with AlexaFluor488 goat anti-rabbit (Molecular Probes), all in casein-blocking solution. Cryosections were counterstained with TOTO-3 as described above. Control experiments in the absence of poly(dT) probe in the hybridization mixture or pretreatment (1 h; 37°C) with 250 μg/ml RNAse A in 2× SSC showed no detectable signal (our unpublished data).


Images were acquired on a confocal laser scanning microscope (Leica TCS SP2, 100× objective, numerical aperture 1.4; Leica Camera, Milton Keynes, United Kingdom), equipped with argon (488-nm) and HeNe (543-, 633-nm) lasers, and pinhole equivalent to 1 Airy disk. Images from the different channels were collected sequentially to prevent bleedthrough. For comparison of different treatments, images were collected on the same day using the same settings, and without saturation of the intensity signal. For an unbiased representation of the whole nuclear volume, nuclear profiles were selected randomly on the TOTO-3 channel, regardless of whether sections contained splicing speckles (Supplemental Figure S2, A and C). Raw TIFF images were merged and analyzed in Adobe Photoshop (Adobe Systems) without further thresholding or filtering (e.g., no background subtraction).

Quantitative and Statistical Analyses

The quantification procedure used for splicing speckles is summarized in Supplemental Figure S2, D. Masks were drawn in the merged images, around the nucleoplasm (excluding nucleoli) in the TOTO-3 channel, around speckles in the “splicing speckle” channel, and in areas of the image outside cells (background). The areas of nucleoplasm and speckles were recorded (in pixels) for each nuclear profile in the image, and the average intensities per pixel measured within the “speckle,” “nucleoplasm,” and “background” masks using the Histogram function for the speckle marker (Figure 2) or pol II (Figure 3) channels.

Speckle volume was calculated as a percentage of the nucleoplasm after calculation of the percentage of nucleoplasmic area occupied by speckles (i.e., the ratio between area speckles and area of nucleoplasm), because the thickness of sections is significantly thinner than the size of the speckles and is equal throughout the nuclear section. Volume data were compiled from single- and double-labeling experiments, because data comparisons showed that the presence of antibodies to pol II during immunolabeling of speckles has no statistically significant effect on measurements of volume (ANOVA with blocking for different experiments; our unpublished data). The effect of α-amanitin on speckle volume was analyzed by ANOVA test with blocking (independent labeling experiments were considered as different blocks).

Figure 2.

Figure 2. Effects of transcriptional inhibition on the volume of speckles and content in SC35, Sm antigen, and polyA+ RNA relative to the nucleoplasm. Cryosections of HeLa cells grown in the absence (A, C, and D) or presence (B–D) of α-am were indirectly immunolabeled with antibodies against SC35 or Sm antigen, or polyA+ RNA detected after in situ hybridization with a poly(dT) probe. Nuclear perimeters identified in TOTO-3 images (insets) were drawn as white dotted lines. (A) SC35 is concentrated in euchromatic domains and throughout the nucleoplasm. (B) Transcriptional inhibition by α-am causes speckles to round up and increase accumulation of SC35. Bar, 2 μm. (C and D) The np fraction of Sm antigen, SC35, or polyA+ RNA present in speckles was determined after measuring the fluorescence intensity (C) and volume (D) in speckles relative to nucleoplasm in control and α-amanitin–treated cells. Total average intensity levels of SC35, Sm, and polyA+ throughout the nucleoplasm did not change with α-amanitin relative to untreated samples (p > 0.7). α-Amanitin treatment causes statistically significant enrichment of speckle markers in speckles (except for Sm antigen; t test *p ≤ 0.05 or **p ≤ 0.01) and increase in speckle volumes (ANOVA with blocking; ***p ≤ 0.003). Numbers of nuclear profiles measured are shown in parentheses. Error bars represent standard deviations.

Average “background” intensity (per pixel) was subtracted from the intensities in the nucleoplasm and in speckles for each image. The total nucleoplasmic (np) levels of speckle markers or polymerase II was normalized relative to untreated controls (Supplemental Figure S2, D.3). For SC35, Sm, and polyA+ RNA, nucleoplasmic intensities were measured in images from single-labeling experiments (Figure 2C), to avoid bias because of antibody steric hindrance detected in double-labeling experiments (our unpublished data). Results for total pol II levels represented in Figure 3 (red bars) come from double-labeling experiments but were confirmed in singly labeled cryosections (our unpublished data), with no significant differences.

The content of speckle markers or pol II in speckles was represented as a proportion of total nucleoplasmic content. For each nuclear profile, average intensities (per pixel) in the nucleoplasm and in speckles were subtracted from the image background and multiplied by the number of pixels in nucleoplasmic and speckle areas, respectively. The relative intensity present in speckles was calculated as a percentage of total intensity in the nucleoplasm and averaged for all sections analyzed (Supplemental Figure S2, D). This value represents an average for the cell population and corrects for the contribution of sections representing different regions of the nucleus. To normalize the marker content in speckles in α-amanitin–treated samples relative to untreated controls, the average content was calculated as for untreated samples, and this value was corrected with the normalized average intensity (Supplemental Figure S2, D.4).

The effects of α-amantin on total nucleoplasmic levels and on speckle content were determined using a t test (assumptions 2-tailed distribution and 2-sample unequal variance).

For paraspeckles, masks were drawn around the domains, before scoring for the presence of pol II or Br-RNA inside, over or outside the edge of masks. The numbers of paraspeckles analyzed ranged from 12 to 29 and originated from eight to 14 different nuclear profiles. The differences in pol II or Br-RNA localization in paraspeckles were analyzed using the chi-square statistical test after merging the values on the “edge” and “outside” speckles into one category.


Specificity of Pol II Antibodies and Distribution of Different Forms of Pol II in the Nucleus of HeLa Cells

The eukaryotic nucleus contains inactive and active pol II complexes that become phosphorylated on Ser2 and Ser5 residues at specific steps of the transcription cycle (Kobor and Greenblatt, 2002). Active complexes account for only approximately one-quarter of nuclear pol II (Jackson et al., 1998; Kimura et al., 2002), most of which is thought to be hyperphosphorylated. The presence of hyperphosphorylated forms of pol II in a nuclear compartment can be an indicator of transcriptional activity but requires confirmation by direct labeling of nascent transcripts, because not all hyperphosphorylated complexes in the cell are associated with DNA (Warren et al., 1992; Jackson et al., 1998; Kimura et al., 2002). Conversely, the presence of hypophosphorylated pol II in a compartment may suggest a role in storage of inactive pol II complexes or could simply reflect some level of permeability. A role in storage can be put in evidence by changing the equilibrium between active and inactive complexes, for example, after transcription inhibition with the drug α-amanitin (Carmo-Fonseca et al., 1992; Spector et al., 1993).

We first tested the specificity of antibodies that recognize different forms of the largest subunit of pol II, RPB1, in Western blots and their labeling properties in fixed HeLa cells (Figure 1; see Materials and Methods and supplemental material for a detailed description of the properties of each antibody). Six antibodies were highly specific for RPB1 in Western blots, three antibodies with specificity for phosphorylated epitopes showed full sensitivity to AP, and all gave consistent results by in situ immunofluorescence.

SDS-PAGE separates RPB1 molecules into two main bands, IIA and IIO, which correspond to hypo- and hyperphosphorylated forms of the subunit, respectively, and an intermediate smear with increasing levels of phosphorylation (Figure 1B). Total pol II can be recognized by antibodies H224 and ARNA3, which bind to non-CTD domains of RPB1 and therefore label both pol IIA and IIO bands in Western blots of whole HeLa cell extracts (Figure 1, A and B). The IIA form can be detected with antibody 8WG16 raised against unphosphorylated CTD (Figure 1B), which also detects a minor fraction of slower migrating molecules, but little of the IIO band (Figure 1B). For simplicity, hereafter, we refer to the pol II molecules detected by 8WG16 as pol IIA. Pol II forms phosphorylated on Ser5 can be detected with rabbit anti-Ser5P and 4H8 antibodies; both antibodies label the IIO form and a smear of faster migrating forms, but not the IIA band (Figure 1B). Pol II complexes phosphorylated on Ser2 can be detected with antibody H5 (Patturajan et al., 1998), which exclusively detects the IIO band in whole cell extracts, but not the IIA form or the intermediate smear, which is detected by antibodies to Ser5P (Figure 1B; see Discussion for comments on previously described cross-reactions of H5). These results suggest that most “hypophosphorylated” IIA molecules have little phosphorylation on Ser2 or Ser5 residues.

Figure 3.

Figure 3. RNA polymerase II is not recruited to splicing speckles upon transcriptional inhibition. Cryosections of HeLa cells grown in the absence (A, D, G, and J) or presence (B, E, H, and K) of α-am were indirectly immunolabeled with antibodies against different forms of pol II and splicing speckle markers (SC35 or polyA+ RNA). Nuclear perimeters were identified in the TOTO-3 images and drawn as white dotted lines or shown as an inset. Higher magnification images of speckles are shown on the right-hand side of lower magnification images. Bar, 2 μm. The np fraction of pol II present in speckles is a ratio of total fluorescence intensity in speckles relative to total intensity throughout the nucleoplasm (C, F, I, and L). The effects α-amanitin on the total amounts of pol II in the nucleoplasm or in speckles are statistically significant except for the amount of Ser2P pol II (L) in speckles, which remains constant (t test; **p ≤ 0.01 or ***p ≤ 0.001); numbers of nuclear profiles measured are shown in parentheses). Error bars represent standard deviations. (A–C) Total pol II (H224; red) and SC35 (green) in control (A) and α-amanitin–treated (B) cells. A small fraction of total pol II is found in speckles; upon transcriptional inhibition, total pol II levels decrease proportionally in both np and speckles (C). (D–F) Pol IIA (8WG16; red) and polyA+ RNA (green). A small fraction of pol IIA is found within speckles; upon transcriptional inhibition, levels decrease in nucleoplasm and speckles (F), and a small number of brighter IIA sites are seen around nucleoli (E; solid line). (G–I) Phosphorylated Ser5 pol II (rabbit anti-Ser5P; red) and SC35 (green). A small fraction of phospho-Ser5 pol II is found in speckles, mostly at their periphery; upon transcriptional inhibition, levels decrease in both nucleoplasm and speckles (I). (J–L) Phosphorylated Ser2 pol II (H5; red) and SC35 (green). A small fraction of phospho-Ser2 pol II is found in speckles; upon transcriptional inhibition, levels decrease throughout the nucleoplasm, whereas levels within speckles remain constant (L).

To characterize the in situ labeling patterns of each antibody in HeLa cells, we used ultrathin cryosections (100–150 nm in thickness) of sucrose-embedded cells because they maximize antibody accessibility to nuclei fixed under conditions that preserve ultrastructure (Figure S2, B; Guillot et al., 2004; Branco et al., 2005), and they also improve the axial resolution of confocal microscopes (Pombo et al., 1999a). All pol II forms studied here are found in discrete sites outside nucleoli, with a small number of brighter sites observed in some nuclear sections (Figure 1, C–G, I, and K; note that ∼150-nm-thick cryosections correspond to ∼1/50 of a nucleus with 5- to 10-μm diameter). EGFP-RPB1 and other subunits of pol II show similar distributions (Jones et al., 2000; Sugaya et al., 2000). Antibodies against the N terminus or nonphosphorylated CTD also detect pol II in the cytoplasm, where RPB1 is likely to be translated (Figure 1, C–F, arrowheads). All detectable immunoreactivity of antibodies to Ser5P or Ser2P pol II is sensitive to AP (Figure 1, H, J, and L). Although all pol II forms have a homogeneous distribution in the nucleus of HeLa cells, without obvious concentration in speckles, it remained possible that some forms of pol II are specifically stored in these compartments.

Inhibition of Transcriptional Activity Induces Minor Changes in the Volume and Splicing Factor Enrichment of “Splicing Speckles”

To ask whether splicing speckles act as storage sites of pol II, we first characterized speckles in our imaging conditions in control and α-amanitin–treated cells to reveal the extent of recruitment because of transcriptional inhibition (Figure 2). Comparisons of three markers of speckles, Sm antigen, SC35, and polyA+ RNA, show they colocalize in large interchromatin domains, as expected (Supplemental Figure S1). As shown previously, splicing speckles become more prominent after α-amanitin treatment (Figure 2, A and B; Carmo-Fonseca et al., 1992; Spector et al., 1993). To compare the amount of splicing factors and polyA+ RNA concentrated in speckles relative to the nucleoplasm in control and α-amanitin–treated cells, we developed a simple stereological procedure that involves computing the proportion of intensity signal contained in speckles relative to the total signal present throughout the nucleoplasm (see Materials and Methods and Supplemental Figure S2). Speckles contain approximately one-fifth of the nucleoplasm content in SC35, Sm antigen, and polyA+ RNA (25, 18, and 17%, respectively; Figure 2C; see also Fay et al., 1997; Wei et al., 1999). The total volume occupied by speckles is relatively small (5, 8, and 4% of the nucleoplasm for speckles marked by SC35, Sm antigen, and polyA+ RNA, respectively; Figure 2D), showing that the splicing factor and polyA+ enrichment in speckles is only two- to fourfold relative to the whole nucleoplasm. The small differences between speckle markers are likely to reflect difficulties with identifying smaller speckles with SC35 and polyA+ RNA, which give more discrete patterns.

After α-amanitin treatment in vivo, speckles round up and seem more enriched in SC35 (Figure 2, A and B; Spector et al., 1993). Unlike what might be predicted from visual inspection of images, intensity measurements show that the recruitment of SC35 into splicing speckles after transcriptional inhibition is relatively small (increasing from 25 to 39% of total nucleoplasmic intensity after treatment with α-amanitin), but statistically highly significant (t test; p = 0.01). Only a marginal, but statistically significant increase, is observed for polyA+ RNA, with no significant effect detected in the case of Sm antigen (Figure 2C). The volume of speckles increases slightly to 6, 10, and 6% of the nucleoplasm volume for SC35, Sm antigen, and polyA+, respectively (Figure 2D); the increase in volume relative to untreated cells for the three speckle markers is statistically highly significant (ANOVA test; p ≤ 0.003).

Figure 4.

Figure 4. Western blot analyses confirm reduction pol II levels after treatment with α-amanitin. HeLa cells were grown ±50 μg/ml α-amanitin for 5 h, before whole cell extracts were prepared and analyzed by Western blot with antibodies against total pol II (H224) and Ser2P pol II (H5). Western with H224 reveals a decrease in the levels of hyperphosphorylated (IIO band) and hypophosphorylated (IIA band) forms of pol II. Western blots with H5 confirm a marked reduction in the levels of Ser2P pol II.

Splicing Speckles Contain but Are Not Enriched in Any of the Different Phosphorylated Forms of Pol II

We next tested whether any of the different pol II forms are present in splicing speckles, using SC35 or polyA+ RNA as markers, depending on the origin of the pol II antibodies available (Figure 3; analyses using Sm antigen gave similar results; our unpublished data). All forms of pol II are found within speckles (Figure 3, A, D, G, and J) but in relative amounts that are proportional to speckle volume (7, 7, 6, and 4% for total pol II, IIA, Ser5P, and Ser2P, respectively; Figure 3, C, F, I, and L), indicating that pol II is neither enriched nor depleted in speckles relative to the whole nucleoplasm. Biochemical purification of speckles is consistent with this result, because pol II copurifies with splicing speckles, but no enrichment is detected relative to whole nuclear extracts (Mintz et al., 1999).

Splicing Speckles Do Not Act as Nuclear Reservoirs of Pol II but Contain a Small Pool of Stably Bound Ser2P Pol II

To reveal whether splicing speckles act as nuclear reservoirs of inactive pol II complexes, we inhibited pol II activity with α-amanitin and measured whether the speckle content in pol II increased. Cryosections from cells grown in the absence or presence of α-amanitin were immunolabeled in parallel and imaged under the same, nonsaturating conditions. After α-amanitin treatment, the total nucleoplasmic levels of total pol II, pol IIA, and Ser5P pol II decrease to 33, 34, and 26%, respectively, relative to untreated cells (Figure 3, A–I). The levels within speckles also decrease to approximately one-third (from ∼6 to 7 to ∼1 to 2%), implying that pol II complexes in speckles are not stably bound but in transit through these domains. The reduction in total pol II levels was confirmed by Western blot analyses for total pol II (Figure 4). Previous analyses of pol II levels by Western blotting also show a decrease in pol II levels induced by α-amanitin (Bregman et al., 1995; Nguyen et al., 1996). Although Nguyen et al. (1996) reported complete degradation of pol II after 4 h in 100 μg/ml α-amanitin (or 24 h in 20 μg/ml), our results are consistent with those of Bregman et al. (1995) in which some Ser5P pol II is detected after 8 h in 10 μg/ml α-amanitin (and after 3 h in 100 μM 5,6-dichloro-1-β-d-ribofuranosylbenzimidazole [DRB]). Such discrepancies are likely to result from differences in sensitivity of Western blotting protocols used in the different laboratories. We also confirmed the presence of pol II and reduction in levels after α-amanitin treatment on equatorial optical sections from whole cells after immunolabeling with the same panel of antibodies (Guillot, Martin, Antunes, Bentley, and Pombo, unpublished data; also see Bregman et al., 1995). Alternative inhibitors of pol II activity, such as DRB, also cause a reduction, albeit less pronounced, of pol II levels in the nucleoplasm, and do not cause detectable accumulation of pol II forms detected with antibodies against total pol II in speckles (Guillot and Pombo, unpublished data; also see Bregman et al., 1995). However, one study claims that DRB has no effect on pol II levels (Nguyen et al., 1996).

Careful analysis of pol IIA distribution in α-amanitin–treated cells also revealed the presence of a small number of intensely labeled sites containing this form of pol II around nucleoli (Figure 3E, solid line; also see Figure 6D), reminiscent of perinucleolar caps such as those seen with coilin, fibrillarin, or paraspeckle protein (Fox et al., 2002; Ospina and Matera, 2002). The perinucleolar pol IIA sites detected by 8WG16 do not colocalize with polyA+ RNA (Figure 3E), coilin, or fibrillarin (our unpublished data), but they do colocalize with paraspeckle protein (see below).

In contrast to the other forms of pol II, the amount of Ser2P pol II in speckles remained constant (∼5% of nucleoplasmic intensity) after α-amanitin treatment, whereas the reduction detected in the nucleoplasm was more prominent (to ∼18% relative to untreated cells; Figure 3, J–L; also confirmed by Western blot, Figure 4, and by immunofluorescence analysis of whole cells; Guillot and Pombo, unpublished data). Although visual inspection of images has previously suggested that Ser2P pol II is recruited from the nucleoplasm to speckles upon transcriptional inhibition (Bregman et al., 1995; Mintz and Spector, 2000), the intensity measurements presented here dismiss this conclusion. The general decrease in Ser2 phosphorylation in euchromatic areas is expected as phosphorylation on Ser2 correlates with transcription elongation, which is inhibited within 1 h of the α-amanitin treatment used here (measured after Br-UTP incorporation; our unpublished data). The lack of sensitivity to α-amanitin and the absence of little detectable chromatin in splicing speckles suggest that the pol II molecules present in speckles are not engaged in transcription. The use of other inhibitors of pol II activity, such as DRB, also revealed a stable pool of Ser2P pol II in speckles (Guillot and Pombo, unpublished; also see Bregman et al., 1995).

To confirm that Ser2P pol II complexes present in speckles are transcriptionally inactive, we localized sites of transcription relative to speckles after incorporation of Br-UTP into newly made RNA. We labeled nascent RNAs in cells permeabilized with a mild detergent in conditions that preserve cellular ultrastructure and the activity of all three nuclear polymerases (I, II, and III), while preventing any detectable movement of completed RNAs down the transport pathway (Iborra et al., 1996; Pombo et al., 1999b). Br-RNA was detected at the edges of speckles (Figure 5; Puvion and Puvion-Dutilleul, 1996; Iborra et al., 1998; Cmarko et al., 1999), unlike what is observed for Ser2P pol II (Figure 3K).

The presence of Br-RNA at the periphery of speckles is consistent with identification of active genes in this region. Although previous analyses of whole cells reported 15–45% of Br-RNA signal in the nucleoplasm is present within nuclear speckles (Fay et al., 1997; Wei et al., 1999), our higher resolution approach finds only ∼5% of Br-RNA at the edge of speckles (marked by SC35 or polyA+ RNA).

Figure 5.

Figure 5. Splicing speckles are not transcriptionally active. Newly labeled Br-RNA (red) was found over the edges of, but not within splicing speckles marked by SC35 (green). Higher magnification view of speckle inside white box is shown on the right hand side. Bar, 2 μm.

Paraspeckles Contain Pol IIA but Little Ser2P Form or Newly Made Br-RNA

We next investigated whether the splicing-associated, paraspeckle domains contained pol II or could be involved in the transcriptional events that take place at the speckle periphery. Because paraspeckles are too small to act as main reservoirs of pol II, we scored the number of parapeckles that contained pol II (Figure 6), instead of measuring the relative content of pol II. Total pol II marked by non-CTD antibodies is found within most paraspeckles (77%) or at their periphery (23%; Figure 6, A and G). The presence of pol II within these domains was confirmed with antibody 8WG16 to hypophosphorylated forms (with or without pretreatment with AP to remove phospho groups; Figure 6, B and G); 8WG16 labels 95% of paraspeckles, probably because of the increased sensitivity afforded by the detection of multiple heptad repeats in the CTD.

Because paraspeckles relocate to the periphery of the nucleolus after transcriptional inhibition with actinomycin D and DRB (Fox et al., 2002, 2005), in the form of cap structures such as those observed here for pol IIA after α-amanitin treatment (Figure 2E), we investigated whether pol II remained stably associated with paraspeckles in these conditions. We found pol IIA colocalized in the same nucleolar caps as PSP1 (Figure 6D), but not coilin (our unpublished data), indicating that the association between nonphosphorylated pol II and PSP1 in paraspeckles is likely to be stable (Figure 6D).

Ser5P and Ser2P pol II were both found within paraspeckles but only in 31 and 11% of the domains, respectively (Figure 6, C, E, and G); most of the remaining paraspeckles were surrounded by these forms of pol II. To test whether the pol II complexes present within paraspeckles are transcriptionally active, we labeled nascent transcripts after incorporation of Br-UTP. Most paraspeckles do not contain but are surrounded by Br-RNA (Figure 5, F and G; Fox et al., 2002). These results indicate that although paraspeckles may be involved in pol II metabolism, they are unlikely to be sites of transcription. Chi-square statistical test showed that differences in the frequency of colocalization of pol II forms or Br-RNA within paraspeckles are highly significant (p < 0.0001). The differences in localization between Ser5P and Ser2P forms and Br-RNA also indicate that the presence of phosphorylated forms of pol II within a nuclear domain does not necessarily imply that the domain is transcriptionally active, as seen previously for Cajal bodies (Schul et al., 1998; Xie and Pombo, 2005).

Figure 6.

Figure 6. Paraspeckles are not transcriptionally active but contain pol II. Paraspeckles contain total pol II (A; detected with ARNA-3), pol IIA (B; detected with 8WG16), and Ser5P pol II (C; detected with 4H8). Higher magnification views of paraspeckles inside white box are shown below each low-power image. Pol IIA colocalizes in PSP1 nucleolar caps formed after α-am treatment (arrow in D). Little detectable Ser2P pol II or Br-RNA is found within paraspeckles (E and F). The magnified views in the last row show two additional examples of paraspeckle domains in which Ser2P or Br-RNA were found inside or at the edge of paraspeckles, respectively. Bar, 2 μm. (G) Quantitative analysis of the number of paraspeckles containing pol II or Br-RNA outside (red), at the edge of (orange), or inside (green) the domains. Numbers of paraspeckles analyzed are shown in parentheses. The chi-square statistical test was performed after merging the values on the”edge“and”outside“speckles into one category; differences were found to be highly significant (p < 0.0001).


RNA Polymerase II Is Present but Not Enriched within Splicing Speckles

Splicing speckles are major nuclear domains that occupy 5–10% of the nucleoplasmic volume and are enriched in splicing factors and polyA+ RNA by a factor of 2 to 4 relative to the nucleoplasm (Figure 2; Fay et al., 1997; Wei et al., 1999). We find that all forms of pol II studied here are present in splicing speckles (Figure 3), but in contrast with previous reports, we do not detect an enrichment of Ser2P pol II, nor of any other pol II form, in these ultrastructurally preserved samples (Bregman et al., 1995; Mintz and Spector, 2000; Guillot et al., 2004). A small pool of hyperphosphorylated pol II is also detected in speckles after biochemical purification of these compartments (Mintz et al., 1999).

Transcription Occurs at the Edge of but Not within Splicing Speckles

Here, we confirm that the interior of speckles is transcriptionally inactive, because little nucleotide incorporation into RNA is detected (also see Puvion and Puvion-Dutilleul, 1996; Iborra et al., 1998; Cmarko et al., 1999). However, a growing number of genes have been found associated with speckles (Shopland et al., 2003; Moen et al., 2004). The resolution of confocal microscopes is limited by the diffraction of light; on the z-axis, it is at best 500 nm and in the size range of most speckles. As suggested from the three- to ninefold overestimation of transcriptional activity in speckles after whole cell measurements of Br-RNA (Fay et al., 1997; Wei et al., 1999) relative to higher resolution achieved after the use of ultrathin sections (Figure 5; Puvion and Puvion-Dutilleul, 1996; Iborra et al., 1998; Cmarko et al., 1999), some of the associations detected between specific genes and speckles in confocal micrographs of whole cells are likely to be too distant to be biologically meaningful. Nevertheless, the accumulation of transcripts in speckles and their enrichment after splice-site mutation provides compelling evidence that some pre-mRNAs cycle through speckles to complete splicing or as a quality control mechanism for unprocessable transcripts (Johnson et al., 2000). Transient association of polyA+ RNAs with speckles has also been recently observed in vivo using fluorescent 2′-O-methyl oligoribonucleotide probes (Molenaar et al., 2004).

Inhibition of Transcription Does Not Result in Pol II Recruitment to Speckles but Reveals a Stably Bound Pool of Ser2P Pol II Complexes

Because the nuclear dynamics of phosphorylated forms cannot so far be assessed by live cell imaging, we set out to analyze the stability of pol II association with speckles after inhibition of pol II transcription. If speckles were storage sites for RNA pol II, transcriptional inhibition should result in an enrichment of pol II in these domains. We find the opposite is true (Figure 3); total pol II levels throughout the nucleoplasm and within speckles decrease proportionally, arguing against a function of splicing speckles as a pool of inactive pol II complexes. These results are consistent with the “sponge-like” properties of splicing speckles recently characterized in the Xenopus oocyte nucleus (Handwerger et al., 2005), and with the idea that pol II complexes cruise through these permeable nuclear compartments.

The hyperphosphorylated Ser2P pol II showed a different behavior. Although no enrichment was detected, the levels of Ser2P pol II complexes in speckles remained constant after transcriptional inhibition, whereas levels in the nucleoplasm were greatly diminished. This suggests that Ser2P complexes in speckles represent a stable component of these domains that do not freely exchange with the nucleoplasm.

Cross-reactivity of H5 with Ser5P CTD peptides and phosphorylated SR proteins has been identified in highly purified systems (Cho et al., 2001; Doyle et al., 2002; Licatalosi et al., 2002; Jones et al., 2004). However, a number of arguments indicate that these cross-reactivities have little contribution to the staining detected in situ after immunofluorescence. First, Ser5P pol II is significantly reduced in speckles after α-amanitin, unlike the epitope detected by H5, which remains constant (Figure 3). Second, H5 does not bind to Cajal bodies, although these domains are rich in the Ser5P form (Schul et al., 1998; Doyle et al., 2002; Xie and Pombo, 2005). It is more difficult to rule out that some of the H5 staining in speckles is because of cross-reactivity to hyperphosphorylated SR proteins. However, it is generally thought that hyperphosphorylated SR proteins reside outside speckles (Gui et al., 1994; Colwill et al., 1996; Neugebauer and Roth, 1997). Finally, the presence of hyperphosphorylated RPB1 and other pol II subunits in speckles is also detected by mass spectrometry after biochemical purification of these compartments (Mintz et al., 1999).

Because phosphorylated pol II is required for splicing (Hirose et al., 1999) and promotes assembly of spliceosomes (Zeng and Berget, 2000), an intriguing possibility for the role of Ser2P pol II in speckles is that the CTD may serve as a platform for posttranscriptional splicing of long transcripts known to traffic through speckles and accumulate in these domains when they contain a splice-site mutation (Johnson et al., 2000). Although CTD truncation has no effect on splicing of injected β-globin pre-mRNA in Xenopus oocytes (Bird et al., 2004), it remains possible that the specific pre-mRNAs found to accumulate in speckles in mammalian cells require a phosphorylated CTD for posttranscriptional splicing. Because splicing speckles are present in cells that contain a CTD-truncated RPB1, the CTD itself is unlikely to be a structural component of these domains (Misteli and Spector, 1999). Three other arguments show that Ser2P modification is not exclusively associated with transcriptionally active complexes and may have roles other than in transcription. First, Ser2P complexes are present in mitotic cells when pol II is inactive (Bregman et al., 1994), with potential roles in inactivation of transcription initiation and splicing factor recruiting to the nucleus upon nuclear envelope formation. Second, ∼10% of Ser2P pol II in the cell is not resistant to treatment with the detergent sarkosyl and therefore is probably transcriptionally inactive (Jackson et al., 1998). Third, Ser2P complexes within speckles are insensitive to α-amanitin (Figure 3K) and DRB (Bregman et al., 1995).

Paraspeckles Are Not Sites of Transcription but Are Stably Associated with Pol IIA

Paraspeckles are marked by PSP1 and PSP2 proteins, which interact with CTD-binding proteins, such as p54nrb and PSF, and are found at the periphery of splicing speckles (Emili et al., 2002; Fox et al., 2002, 2005; Kameoka et al., 2004; Myojin et al., 2004), where active genes are also associated. We tested whether paraspeckles contain pol II or are sites of transcription and found that they contain the Ser5P form of pol II but little of the Ser2P form or newly made Br-RNA (Figure 6). Furthermore, upon transcriptional inhibition, pol II and p54nrb remain stably bound to paraspeckles when they relocate to the periphery of nucleoli (Figure 6D; Fox et al., 2005). The presence of pol II in paraspeckles and the cycling of these compartments between splicing speckles and nucleoli suggests a role in pol II recycling, but it remains unclear why some pol II complexes would be transported to and from the periphery of nucleoli.

In summary, our results show that splicing speckles and paraspeckles are likely to be involved in pol II metabolism but are not major sites of gene transcription or polymerase II storage. It remains to be determined whether the Ser2P pol II complexes present in speckles are required for posttranscriptional splicing that may occur in these domains.


This article was published online ahead of print in MBC in Press ( on February 8, 2006.

Abbreviations used: AP, alkaline phosphatase; Br-RNA, bromo-UTP–labeled RNA; CTD, C-terminal domain; DRB, 5,6-dichloro-1-β-d-ribofuranosylbenzimidazole; EGFP, enhanced green fluorescent protein; IF, immunofluorescence; pol II, RNA polymerase IIA; pol II, hypophosphorylated polymerase II; pol IIO, hyperphosphorylated polymerase II; PSP, paraspeckle protein; Ser2P, phosphorylated serine2 residues of the CTD; Ser5P, phosphorylated serine5 residues of the CTD; SR, serine/arginine; WB, Western blot.

The online version of this article contains supplemental material at MBC Online (

† Present address: Tenovus Research Laboratory, University of Southampton, Southampton General Hospital, Southampton SO15 5PA, United Kingdom

‡ Present address: Institute of Reproductive and Developmental Biology, Imperial College London, London W12 0NN, United Kingdom.


Monitoring Editor: A. Gregory Matera


We thank Drs. Archa Fox and Angus Lamond for PSP1 antibodies; Dr. André Möller, Julie K. Stock, and Miguel R. Branco for comments on the manuscript; Dr. Francisco Ramirez for performing the ANOVA analysis; and the Medical Research Council (United Kingdom) for support (to S.Q.X., S. M., P. V. G., and A. P.). D.L.B. is supported by National Institutes of Health Grant GM-063873.


  • Bird, G., Zorio, D.A.R., and Bentley, D. L. (2004). RNA polymerase II carboxy-terminal domain phosphorylation is required for cotranscriptional pre-mRNA splicing and 3′-end formation. Mol. Cell. Biol. 24, 8963–8969. Crossref, MedlineGoogle Scholar
  • Bisotto, S., Lauriault, P., Duval, M., and Vincent, M. (1995). Colocalization of a high molecular mass phosphoprotein of the nuclear matrix (p255) with spliceosomes. J. Cell Sci. 108, 1873–1882. MedlineGoogle Scholar
  • Branco, M. R., Xie, S. Q., Martin, S., and Pombo, A. (2005). Correlative microscopy using Tokuyasu cryosections: applications for immunolabeling and in situ hybridization. In: Cell Imaging, ed. D. Stephens: Scion Publishing, 201–217. Google Scholar
  • Bregman, D. B., Du, L., Li, Y., Ribisi, S., and Warren, S. L. (1994). Cytostellin distributes to nuclear regions enriched with splicing factors. J. Cell Sci. 107, 387–396. MedlineGoogle Scholar
  • Bregman, D. B., Du, L., Van Der Zee, S., and Warren, S. L. (1995). Transcription-dependent redistribution of the large subunit of RNA polymerase II to discrete nuclear domains. J. Cell Biol. 129, 287–298. Crossref, MedlineGoogle Scholar
  • Carmo-Fonseca, M., Pepperkok, R., Carvalho, M. T., and Lamond, A. I. (1992). Transcription-dependent colocalization of the U1, U2, U4/U6, and U5 snRNPs in coiled bodies. J. Cell Biol. 117, 1–14. Crossref, MedlineGoogle Scholar
  • Cho, E. J., Kobor, M. S., Kim, M., Greenblatt, J., and Buratowski, S. (2001). Opposing effects of Ctk1 kinase and Fcp1 phosphatase at Ser 2 of the RNA polymerase II C-terminal domain. Genes Dev. 15, 3319–3329. Crossref, MedlineGoogle Scholar
  • Clemson, C. M., and Lawrence, J. B. (1996). Multifunctional compartments in the nucleus: insights from DNA and RNA localization. J. Cell. Biochem. 62, 181–190. Crossref, MedlineGoogle Scholar
  • Cmarko, D., Verschure, P. J., Martin, T. E., Dahmus, M. E., Krause, S., Fu, X. D., van Driel, R., and Fakan, S. (1999). Ultrastructural analysis of transcription and splicing in the cell nucleus after bromo-UTP microinjection. Mol. Biol. Cell 10, 211–223. LinkGoogle Scholar
  • Colwill, K., Pawson, T., Andrews, B., Prasad, J., Manley, J. L., Bell, J. C., and Duncan, P. I. (1996). The Clk/Sty protein kinase phosphorylates SR splicing factors and regulates their intranuclear distribution. EMBO J. 15, 265–275. Crossref, MedlineGoogle Scholar
  • Doyle, O., Corden, J. L., Murphy, C., and Gall, J. G. (2002). The distribution of RNA polymerase II largest subunit (RPB1) in the Xenopus germinal vesicle. J. Struct. Biol. 140, 154–166. Crossref, MedlineGoogle Scholar
  • Emili, A., Shales, M., McCracken, S., Xie, W., Tucker, P. W., Kobayashi, R., Blencowe, B. J., and Ingles, C. J. (2002). Splicing and transcription-associated proteins PSF and p54nrb/nonO bind to the RNA polymerase II CTD. RNA 8, 1102–1111. Crossref, MedlineGoogle Scholar
  • Fay, F. S., Taneja, K. L., Shenoy, S., Lifshitz, L., and Singer, R. H. (1997). Quantitative digital analysis of diffuse and concentrated nuclear distributions of nascent transcripts, SC35 and poly(A). Exp. Cell Res. 231, 27–37. Crossref, MedlineGoogle Scholar
  • Fox, A. H., Bond, C. S., and Lamond, A. I. (2005). P54nrb forms a heterodimer with PSP1 that localizes to paraspeckles in a RNA-dependent manner. Mol. Biol. Cell 16, 4304–5315. LinkGoogle Scholar
  • Fox, A. H., Lam, Y. W., Leung, A. K., Lyon, C. E., Andersen, J., Mann, M., and Lamond, A. I. (2002). Paraspeckles. A novel nuclear domain. Curr. Biol. 12, 13–25. Crossref, MedlineGoogle Scholar
  • Fu, X. D., and Maniatis, T. (1990). Factor required for mammalian spliceosome assembly is localized to discrete regions in the nucleus. Nature 343, 437–441. Crossref, MedlineGoogle Scholar
  • Grande, M. A., van der Kraan, I., de Jong, L., and van Driel, R. (1997). Nuclear distribution of transcription factors in relation to sites of transcription and RNA polymerase II. J. Cell Sci. 110, 1781–1791. MedlineGoogle Scholar
  • Gui, J. F., Lane, W. S., and Fu, X. D. (1994). A serine kinase regulates intracellular localization of splicing factors in the cell cycle. Nature 369, 678–682. Crossref, MedlineGoogle Scholar
  • Guillot, P. V., Xie, S. Q., Hollinshead, M., and Pombo, A. (2004). Fixation-induced redistribution of hyperphosphorylated RNA polymerase II in the nucleus of human cells. Exp. Cell Res. 295, 460–468. Crossref, MedlineGoogle Scholar
  • Handwerger, K. E., Cordero, J. A., and Gall, J. G. (2005). Cajal bodies, nucleoli, and speckles in the Xenopus oocyte nucleus have a low-density, sponge-like structure. Mol. Biol. Cell 16, 202–211. LinkGoogle Scholar
  • Herrmann, C. H., and Mancini, M. A. (2001). The Cdk9 and cyclin T subunits of TAK/P-TEFb localize to splicing factor-rich nuclear speckle regions. J. Cell Sci. 114, 1491–1503. MedlineGoogle Scholar
  • Hirose, Y., Tacke, R., and Manley, J. L. (1999). Phosphorylated RNA polymerase II stimulates pre-mRNA splicing. Genes Dev. 13, 1234–1239. Crossref, MedlineGoogle Scholar
  • Iborra, F. J., Jackson, D. A., and Cook, P. R. (1998). The path of transcripts from extra-nucleolar synthetic sites to nuclear pores: transcripts in transit are concentrated in discrete structures containing SR proteins. J. Cell Sci. 111, 2269–2282. MedlineGoogle Scholar
  • Iborra, F. J., Pombo, A., Jackson, D. A., and Cook, P. R. (1996). Active RNA polymerases are localized within discrete transcription “factories'in human nuclei. J. Cell Sci. 109, 1427–1436. MedlineGoogle Scholar
  • Jackson, D. A., Iborra, F. J., Manders, E. M., and Cook, P. R. (1998). Numbers and organization of RNA polymerases, nascent transcripts, and transcription units in HeLa nuclei. Mol. Biol. Cell 9, 1523–1536. LinkGoogle Scholar
  • Johnson, C., Primorac, D., McKinstry, M., McNeil, J., Rowe, D., and Lawrence, J. B. (2000). Tracking COL1A1 RNA in osteogenesis imperfecta: splice-defective transcripts initiate transport from the gene but are retained within the SC35 domain. J. Cell Biol. 150, 417–432. Crossref, MedlineGoogle Scholar
  • Jones, E., Kimura, H., Vigneron, M., Wang, Z., Roeder, R. G., and Cook, P. R. (2000). Isolation and characterization of monoclonal antibodies directed against subunits of human RNA polymerases I, II, and III. Exp. Cell Res. 254, 163–172. Crossref, MedlineGoogle Scholar
  • Jones, J. C., Phatnani, H. P., Haystead, T. A., MacDonald, J. A., Alam, S. M., and Greenleaf, A. L. (2004). C-terminal repeat domain kinase I phosphorylates Ser2 and Ser5 of RNA polymerase II C-terminal domain repeats. J. Biol. Chem. 279, 24957–24964. Crossref, MedlineGoogle Scholar
  • Kameoka, S., Duque, P., and Konarska, M. M. (2004). p54(nrb) associates with the 5′ splice site within large transcription/splicing complexes. EMBO J. 23, 1782–1791. Crossref, MedlineGoogle Scholar
  • Kimura, H., Sugaya, K., and Cook, P. R. (2002). The transcription cycle of RNA polymerase II in living cells. J. Cell Biol. 159, 777–782. Crossref, MedlineGoogle Scholar
  • Kobor, M. S., and Greenblatt, J. (2002). Regulation of transcription elongation by phosphorylation. Biochim. Biophys. Acta 1577, 261–275. Crossref, MedlineGoogle Scholar
  • Komarnitsky, P., Cho, E. J., and Buratowski, S. (2000). Different phosphorylated forms of RNA polymerase II and associated mRNA processing factors during transcription. Genes Dev. 14, 2452–2460. Crossref, MedlineGoogle Scholar
  • Kruhlak, M. J., Lever, M. A., Fischle, W., Verdin, E., Bazett-Jones, D. P., and Hendzel, M. J. (2000). Reduced mobility of the alternate splicing factor (ASF) through the nucleoplasm and steady state speckle compartments. J. Cell Biol. 150, 41–51. Crossref, MedlineGoogle Scholar
  • Lamond, A. I., and Spector, D. L. (2003). Nuclear speckles: a model for nuclear organelles. Nat. Rev. Mol. Cell. Biol. 4, 605–612. Crossref, MedlineGoogle Scholar
  • Lavoie, S. B., Albert, A. L., Thibodeau, A., and Vincent, M. (1999). Heat shock-induced alterations in phosphorylation of the largest subunit of RNA polymerase II as revealed by monoclonal antibodies CC-3 and MPM-2. Biochem. Cell Biol. 77, 367–374. Google Scholar
  • Licatalosi, D. D., Geiger, G., Minet, M., Schroeder, S., Cilli, K., McNeil, J. B., and Bentley, D. L. (2002). Functional interaction of yeast pre-mRNA 3′ end processing factors with RNA polymerase II. Mol. Cell 9, 1101–1111. Crossref, MedlineGoogle Scholar
  • Maniatis, T., and Reed, R. (2002). An extensive network of coupling among gene expression machines. Nature 416, 499–506. Crossref, MedlineGoogle Scholar
  • Mintz, P. J., Patterson, S. D., Neuwald, A. F., Spahr, C. S., and Spector, D. L. (1999). Purification and biochemical characterization of interchromatin granule clusters. EMBO J. 18, 4308–4320. Crossref, MedlineGoogle Scholar
  • Mintz, P. J., and Spector, D. L. (2000). Compartmentalization of RNA processing factors within nuclear speckles. J. Struct. Biol. 129, 241–251. Crossref, MedlineGoogle Scholar
  • Misteli, T., and Spector, D. L. (1999). RNA polymerase II targets pre-mRNA splicing factors to transcription sites in vivo. Mol. Cell 3, 697–705. Crossref, MedlineGoogle Scholar
  • Moen, P. T., Jr., Johnson, C. V., Byron, M., Shopland, L. S., de la Serna, I. L., Imbalzano, A. N., and Lawrence, J. B. (2004). Repositioning of muscle-specific genes relative to the periphery of SC-35 domains during skeletal myogenesis. Mol. Biol. Cell 15, 197–206. LinkGoogle Scholar
  • Molenaar, C., Abdulle, A., Gena, A., Tanke, H. J., and Dirks, R. W. (2004). Poly(A)+ RNAs roam the cell nucleus and pass through speckle domains in transcriptionally active and inactive cells. J. Cell Biol. 165, 191–202. Crossref, MedlineGoogle Scholar
  • Mortillaro, M. J., Blencowe, B. J., Wei, X., Nakayasu, H., Du, L., Warren, S. L., Sharp, P. A., and Berezney, R. (1996). A hyperphosphorylated form of the large subunit of RNA polymerase II is associated with splicing complexes and the nuclear matrix. Proc. Natl. Acad. Sci. USA 93, 8253–8257. Crossref, MedlineGoogle Scholar
  • Myojin, R., Kuwahara, S., Yasaki, T., Matsunaga, T., Sakurai, T., Kimura, M., Uesugi, S., and Kurihara, Y. (2004). Expression and functional significance of mouse paraspeckle protein 1 on spermatogenesis. Biol. Reprod. 71, 926–932. Crossref, MedlineGoogle Scholar
  • Neugebauer, K. M., and Roth, M. B. (1997). Distribution of pre-mRNA splicing factors at sites of RNA polymerase II transcription. Genes Dev. 11, 1148–1159. Crossref, MedlineGoogle Scholar
  • Nguyen, V. T., Giannoni, F., Dubois, M. F., Seo, S. J., Vigneron, M., Kedinger, C., and Bensaude, O. (1996). In vivo degradation of RNA polymerase II largest subunit triggered by alpha-amanitin. Nucleic Acids Res. 24, 2924–2929. Crossref, MedlineGoogle Scholar
  • Ospina, J. K., and Matera, A. G. (2002). Proteomics: the nucleolus weighs in. Curr. Biol. 12, R29–R31. Crossref, MedlineGoogle Scholar
  • Patturajan, M., Schulte, R. J., Sefton, B. M., Berezney, R., Vincent, M., Bensaude, O., Warren, S. L., and Corden, J. L. (1998). Growth-related changes in phosphorylation of yeast RNA polymerase II. J. Biol. Chem. 273, 4689–4694. Crossref, MedlineGoogle Scholar
  • Phair, R. D., and Misteli, T. (2000). High mobility of proteins in the mammalian cell nucleus. Nature 404, 604–609. Crossref, MedlineGoogle Scholar
  • Pombo, A., Hollinshead, M., and Cook, P. R. (1999a). Bridging the resolution gap: imaging the same transcription factories in cryosections by light and electron microscopy. J. Histochem. Cytochem. 47, 471–480. Crossref, MedlineGoogle Scholar
  • Pombo, A., Jackson, D. A., Hollinshead, M., Wang, Z., Roeder, R. G., and Cook, P. R. (1999b). Regional specialization in human nuclei: visualization of discrete sites of transcription by RNA polymerase III. EMBO J. 18, 2241–2253. Crossref, MedlineGoogle Scholar
  • Puvion, E., and Puvion-Dutilleul, F. (1996). Ultrastructure of the nucleus in relation to transcription and splicing: roles of perichromatin fibrils and interchromatin granules. Exp. Cell Res. 229, 217–225. Crossref, MedlineGoogle Scholar
  • Schroeder, S. C., Schwer, B., Shuman, S., and Bentley, D. (2000). Dynamic association of capping enzymes with transcribing RNA polymerase II. Genes Dev. 14, 2435–2440. Crossref, MedlineGoogle Scholar
  • Schul, W., van Driel, R., and de Jong, L. (1998). Coiled bodies and U2 snRNA genes adjacent to coiled bodies are enriched in factors required for snRNA transcription. Mol. Biol. Cell 9, 1025–1036. LinkGoogle Scholar
  • Shopland, L. S., Johnson, C. V., Byron, M., McNeil, J., and Lawrence, J. B. (2003). Clustering of multiple specific genes and gene-rich R-bands around SC-35 domains: evidence for local euchromatic neighborhoods. J. Cell Biol. 162, 981–990. Crossref, MedlineGoogle Scholar
  • Shopland, L. S., Johnson, C. V., and Lawrence, J. B. (2002). Evidence that all SC-35 domains contain mRNAs and that transcripts can be structurally constrained within these domains. J. Struct. Biol. 140, 131–139. Crossref, MedlineGoogle Scholar
  • Spector, D. L., O'Keefe, R. T., and Jimenez-Garcia, L. F. (1993). Dynamics of transcription and pre-mRNA splicing within the mammalian cell nucleus. Cold Spring Harb. Symp. Quant. Biol. 58, 799–805. Crossref, MedlineGoogle Scholar
  • Sugaya, K., Vigneron, M., and Cook, P. R. (2000). Mammalian cell lines expressing functional RNA polymerase II tagged with the green fluorescent protein. J. Cell Sci. 113, 2679–2683. MedlineGoogle Scholar
  • Warren, S. L., Landolfi, A. S., Curtis, C., and Morrow, J. S. (1992). Cytostellin: a novel, highly conserved protein that undergoes continuous redistribution during the cell cycle. J. Cell Sci. 103, 381–388. MedlineGoogle Scholar
  • Wei, X., Somanathan, S., Samarabandu, J., and Berezney, R. (1999). Three-dimensional visualization of transcription sites and their association with splicing factor-rich nuclear speckles. J. Cell Biol. 146, 543–558. Crossref, MedlineGoogle Scholar
  • Xie, S. Q., and Pombo, A. (2005). Distribution of different phosphorylated forms of RNA polymerase II in relation to Cajal and PML bodies in human cells: an ultrastructural study. Histochem. Cell. Biol. 125, 21–31. MedlineGoogle Scholar
  • Zeng, C., and Berget, S. M. (2000). Participation of the C-terminal domain of RNA polymerase II in exon definition during pre-mRNA splicing. Mol. Cell. Biol. 20, 8290–8301. Crossref, MedlineGoogle Scholar
  • Zeng, C., Kim, E., Warren, S. L., and Berget, S. M. (1997). Dynamic relocation of transcription and splicing factors dependent upon transcriptional activity. EMBO J. 16, 1401–1412. Crossref, MedlineGoogle Scholar
  • Zorio, D. A., and Bentley, D. L. (2004). The link between mRNA processing and transcription: communication works both ways. Exp. Cell Res. 296, 91–97. Crossref, MedlineGoogle Scholar