RNA Polymerase II Localizes at Sites of Human Cytomegalovirus Immediate-early RNA Synthesis and Processing

Abstract

Pre-mRNA synthesis in eukaryotic cells is preceded by the formation of a transcription initiation complex and binding of unphosphorylated RNA polymerase II (Pol II) at the promoter region of a gene. Transcription initiation and elongation are accompanied by the hyperphosphorylation of the carboxy-terminal domain (CTD) of Pol II large subunit. Recent biochemical studies provided evidence that RNA processing factors, including those required for splicing, associate with hyperphosphorylated CTDs forming “transcription factories.” To directly visualize the existence of such factories, we simultaneously detected human cytomegalovirus immediate-early (IE) DNA and RNA with splicing factors and Pol II in rat 9G cells inducible for IE gene expression. Combined in situ hybridization and immunocytochemistry revealed that, after induction, both splicing factors and Pol II are present at the sites of IE mRNA synthesis and of IE mRNA processing that extend from the transcribing gene. Noninduced cells revealed no such associations. When IE mRNA-synthesizing cells were treated with a transcription inhibitor, these associations disappeared within 30 min. Our results show that the association of Pol II and splicing factors with IE DNA is dependent on its transcriptional activity and furthermore suggest that splicing factors are still associated with Pol II during active splicing.

Keywords

in situ hybridization immunocytochemistry transcription splicing monoclonal antibody CC-3

SYNTHESIS OF mRNAs is initiated by binding of a hypophosphorylated form of RNA polymerase II (Pol IIA) to the promoter region. This interaction is mediated by several transcription factors that assemble on the promoter and on Pol II, including the carboxyterminal domain (CTD) of the largest subunit. Concomitant with transcription initiation, Pol II is hyperphosphorylated at the CTD, is released from the promoter protein complex, and forms an elongation complex (Laybourn and Dahmus 1990; Dahmus 1996). It was recently shown that splicing factors of the Sm small nuclear ribonucleoprotein (snRNP) and non-snRNP Ser-Arg (SR) family can associate with the hyperphosphorylated CTD of RNA Pol II (Pol IIO), forming a transcription/RNA processing complex (Vincent et al. 1996; Yuryev et al. 1996; Kim et al. 1997). In addition to these factors, cleavage-stimulatory factor (CstF) and cleavage/polyadenylation-specificity factor (CPSF), both involved in 3′-end processing of transcripts, and 5′-capping enzymes were found physically associated with Pol IIO (McCracken et al. 1997a, b; Cho et al. 1997). Together, these findings suggest that elongating Pol II has an important function in recruiting all components required for RNA maturation to transcription sites to form an integrated RNA synthesis and processing factory (Greenleaf 1993; McCracken et al. 1997a; for review see Steinmetz 1997). It has been shown that, with a few exceptions, RNA processing is completed at the transcription site (Beyer and Osheim 1988; Baurén and Wieslander 1994; Wuarin and Schibler 1994).

To understand how the different components required for gene transcription and RNA processing are integrated in the cell nucleus, immunofluorescence microscopic studies have been performed using specific antibodies raised against splicing factors and Pol II. These studies revealed that splicing factors are located in 20-50 speckles and in many small spots throughout the nucleoplasm, excluding nucleoli (for review see Spector 1993). These small spots represent the highest concentrations of splicing factors and coincide with nascent transcripts (Neugebauer and Roth 1997). Speckles are believed to represent storage sites of splicing factors and to have functions in the assembly or disassembly of splicing complexes (Jackson et al. 1993; Wansink et al. 1993; Fakan 1994). In vitro and in vivo studies showed that splicing factor localization is highly dynamic. Speckles were shown to change shape continuously in living cells expressing a splicing factor/GFP fusion protein (Misteli et al. 1997). Furthermore, it was shown that splicing factors are recruited to transcriptionally inducible genes after induction (Baurén et al. 1996; Dirks et al. 1997; Misteli et al. 1997). Because speckles are also enriched for poly A sequences and because some actively transcribing genes are found preferentially associated with these domains, it cannot be excluded that they also have a function in RNA processing (Carter et al. 1991,1993; Xing et al. 1993,1995; Moen et al. 1995; Ishov et al. 1997).

Interestingly, Pol IIO is also found in speckles and in small foci dispersed throughout the nucleoplasm excluding nucleoli (Bregman et al. 1995; Mortillaro et al. 1996). Recently, it was shown that this localization pattern can vary among cells depending on their transcriptional activity (Zeng et al. 1997). When the transcriptional activity of cells was monitored by the abundance of Br-UTP incorporation in nascent transcripts, Pol IIO was not present in speckles in transcriptionally active cells, whereas Pol IIO was concentrated in speckles in cells showing low transcriptional activity. In addition, in exponentially growing HeLa cells, Pol IIO was present in a meshwork throughout the nucleoplasm but not in speckles (Grande et al. 1997). These observations agree with the hypothesis that active transcription and RNA processing occur at sites throughout the nucleoplasm and that speckle domains represent storage sites from which factors are recruited when needed (Jiménez-García and Spector 1993; Spector et al. 1993; Misteli and Spector 1996).

The high degree of overlap between sites of newly synthesized RNA and those of splicing factor and Pol IIO localization suggests further that transcription and RNA processing are spatially linked in vivo (see also Weeks et al. 1993). In this study we used combinations of RNA fluorescence in situ hybridization (FISH) and immunocytochemistry to resolve the spatial and temporal relationship of Pol II distribution with human cytomegalovirus immediate-early (HCMV-IE) DNA, RNA, and splicing factor localization in rat 9G cells. This rat fibroblast cell line contains a tandem repeat of approximately 50 copies of the HCMV-IE gene that are transcriptionally inactive under normal culturing conditions but can be induced for transcription by treatment of the cells with the protein synthesis inhibitor cycloheximide. We show that on induction with cycloheximide, both splicing factors and Pol IIO are recruited to sites of IE gene transcription. After recruitment, they co-localized not only with nascent IE transcripts but also with (pre-)mRNA that accumulated in tracks or dots extending from the transcribing gene. These data suggest that splicing factors remain associated with Pol IIO when Pol IIO is released from the gene after transcription is completed. Furthermore, splicing factors may remain associated with Pol IIO when these factors are actively engaged in the splicing of accumulated HCMV-IE RNA near the IE gene cluster.

Materials and Methods

Cell Culture

Rat 9G cells harboring a tandem repeat of the HCMV-IE gene in one of the rat chromosomes were a generous gift of Rene Boom (Boom et al. 1986). Initially, the copy number of the IE gene was estimated to be 10 (Boom et al. 1986), but we have recently shown by fiber FISH that the number of copies is closer to 50 (van de Corput et al. 1998). Cells were grown in DMEM without the pH indicator Phenol Red supplemented with 10% fetal calf serum (Life Technologies; Gaithersburg, MD), antibiotics and glutamine. Cells were seeded on microscope slides and grown to subconfluency. To induce HCMV-IE gene expression, 50 μg/ml cycloheximide (Sigma; St Louis, MO) was added to the culture medium. After 4-6 hr of incubation, approximately 20% of the cells showed HCMV-IE mRNA-positive hybridization signals. This induction appeared to be S-phase-dependent (Boom et al. 1986; Dirks and Raap 1995), although the mechanism by which IE gene transcription is induced is not yet completely understood.

For some experiments, cells were also incubated for 5 min to 3 hr with 25 μg/ml 5,6-dichloro-1-β-D-ribofuranosylbenzimidazole (DRB; Sigma) or with 0.5 μg/ml actinomycin D (Sigma) to inhibit RNA Pol II transcription.

In Situ Hybridization

Slides with rat 9G cells were fixed, pretreated, and hybridized with a probe as described previously (Dirks et al. 1993). Briefly, cells were washed in PBS for 1 min and fixed with 3.7% formaldehyde (Merck; Darmstadt, Germany) containing 5% acetic acid in PBS for 15 min at room temperature (RT). Cells were then washed in PBS and stored in 70% ethanol at 4C. Before hybridization, cells were washed in deionized water and pretreated with 0.1% (w/v) pepsin (Sigma) in 0.01 M HCl for 1 min at 37C. Cells were then washed in deionized water, dehydrated in a graded ethanol series, and air-dried to prevent uncontrolled dilution of probes. Alternatively, cells were fixed in 2% or 3.7% formaldehyde in PBS for 15 min and either treated with pepsin or left untreated.

Plasmid pSS containing the 5.0-KB SphI-SalI genomic fragment of the immediate-early region of HCMV (Boom et al. 1986) and a PCR-generated IE intron-specific probe (Dirks et al. 1993) were labeled with digoxigenin-, biotin-, or Cy3-dUTP by nick translation. An HCMV-IE DNA-specific sense RNA probe was transcribed in the presence of digoxigenin-11-dUTP from an exon 4-specific PCR fragment containing the T7 RNA polymerase promoter sequences. For this purpose a protocol delivered with the in vitro transcription kit (Boehringer; Mannheim, Germany) was followed. After ethanol precipitation, probes were dissolved in 60% deionized formamide, 2 × SSC (0.3 M NaCl, 0.03 M Nacitrate), 10 mM EDTA, 25 mM NaH₂PO₄, pH 7.4, 10% dextran sulfate, and 250 ng/μl sheared herring sperm DNA at a final concentration of 5 ng/μl. A 50-mer oligonucleotide (dT) probe was labeled with Cy3-dUTP using terminal transferase and dissolved in 10% formamide, 2 × SSC at a concentration of 1 ng/μl.

Ten μl of probe was applied to a slide and covered with an 18 × 18-mm coverslip. Probe and target sequences were denatured simultaneously by placing the slides on an 80C metal plate for 3 min. Hybridizations were done at 37C overnight in a moist chamber.

Posthybridization Washes and Immunocytochemical Detection

After hybridization, cells were washed three times in 50% formamide, 2 × SSC for 10 min each at 37C, and 5 min in Tris-buffered saline (TBS: 150 mM NaCl, 100 mM Tris-HCl, pH 7.4) at RT. Cells hybridized with the oligo (dT) 50 probe were washed three times in 2 × SSC for 3 min each at RT.

For the detection of digoxigenin-labeled probes, cells were incubated with monoclonal antibody (MAb) antidigoxin (Sigma; dilution1:1000) followed by an incubation with a secondary anti-mouse antibody conjugated with FITC (Sigma), Cy3 (Jackson Immunoresearch; West Grove, PA), or Texas Red (Vector Labs; Burlingame, CA) or with a sheep anti-digoxigenin antibody conjugated with FITC (Boehringer Mannheim). Biotin-labeled probes were detected with streptavidin Cy3-(Jackson Immunoresearch), streptavidin-Texas Red (Vector), or avidin-Cy5 (Jackson Immunoresearch).

For the detection of splicing factors, MAb anti-m3G (Oncogene Science; Cambridge, MA), which reacts with the 2,2,7-trimethyl guanosine cap of snRNAs (Reuter et al. 1984), was used. The hyperphosphorylated form of RNA Pol II was detected with MAb CC-3 (Vincent et al. 1996) and the hypophosphorylated form of RNA Pol II with MAb 8WG16 (Promega; Madison, WI). All antibody incubations were done in TBS containing 0.5% (w/v) blocking reagent (Boehringer Mannheim) for 30 min at RT. For double and triple labeling experiments, the different probes and antibodies were mixed in their respective incubation solutions and incubated simultaneously.

Finally, cells were mounted in Vectashield (Vector) containing 4′,6′-diamidino-2-phenyl indole (DAPI) as a DNA counterstain.

Microscopy

Images were collected with an epifluorescence microscope (DM; Leica, Oberkochen, Germany) equipped with a 100 W mercury arc lamp, a triple excitation filter for red, green and blue excitation (Omega; Brattleboro, VT), a filter for Cy5 excitation, and PI Fluotar × 100, NA 1.30-0.60, PI APO x63, NA 1.40 objectives. Different fluorochromes could be selectively excited and recorded without image shifts by inserting appropriate filters in the excitation way. Digital images were captured with a cooled CCD camera (Photometrics; Tucson, AZ) and processed on a Macintosh computer using SCIL-image (Multihouse; Amsterdam, The Netherlands).

Results

Hypo- and Hyperphosphorylated Pol II Have Different Localization Patterns in Rat 9G Cells

Initially, MAb CC-3 was shown to react with a phosphodependent epitope on a 255-kD nuclear matrix protein and possibly with an unidentified 180-kD protein. The 255-kD protein proved to be identical to the phosphorylated form of the largest RNA Pol II subunit (Vincent et al. 1996). Incubation of rat 9G cells with MAb CC-3 and then with rabbit anti-mouse FITC resulted in staining of 15-30 speckle domains and many dispersed small foci in the nucleus, excluding nucleoli (Figures 1A and 1B). This staining pattern was observed in approximately 95% of over 100 cells that were examined in three independent experiments. No significant differences in staining patterns were observed between cells treated with the protein synthesis inhibitor cycloheximide and nontreated cells. This observation complements other studies showing that the localization of splicing factors and the level of transcription are not altered by cycloheximide treatment of cells (O'Keefe et al. 1994; and unpublished observations). Cells were fixed in formaldehyde/acetic acid because this fixative provided optimal FISH results (Dirks et al. 1993). When we compared staining patterns of Pol IIO between cells fixed in formaldehyde/acetic acid or formaldehyde only, no difference in staining pattern was observed. In addition, a short pepsin pretreatment of cells did not appear to have any influence on the appearance of the speckle domains. The staining pattern of speckle domains and small foci was identical to the pattern we have previously observed when rat 9G cells were stained for splicing factors (Dirks et al. 1997). However, compared to other cell types fixed in the same manner as for the rat 9G cells, the speckle domains in rat 9G cells showed a less extended network (Dirks 1996; and unpublished results).

Because it has been suggested that staining of speckles for hyperphosphorylated Pol II is correlated with low transcriptional activity (Zeng et al. 1997), we investigated the transcriptional activity of rat 9G cells by incubating them with Br-UTP according to the method described by Wansink et al. (1993). The result showed that approximately 80% of the 100 cells examined have many transcription foci dispersed throughout the nucleoplasm and are apparently transcriptionally active (not shown).

Next, rat 9G cells were stained with MAb 8WG16 (Thompson et al. 1989), which primarily reacts with hypophosphorylated Pol II (Bregman et al. 1995). Figure 1C shows that Pol IIA is localized in many small foci distributed throughout the nucleoplasm. In addition to these foci, one to three larger intensely stained foci were observed in about 20% of the cells. No staining of speckle domains was observed. In addition, the larger intensely stained foci were not found associated with speckle domains when cells were double stained for Pol IIA and poly A sequences (Figures 1C–1E). This staining pattern appeared to be very similar to the one previously described for MDCK cells (Bregman et al. 1995), although the function of the few large foci with high concentrations of Pol IIA is still unknown.

Hyperphosphorylated Pol II Associates with IE Gene Clusters after Induction

Double labeling experiments in which Pol IIO staining patterns were compared with those of Br-UTP incorporation revealed a high degree of overlap between the two (Grande et al. 1997; Zeng et al. 1997). These results suggest that most of the Pol IIO observed as small foci is actively engaged in transcription. To investigate whether the localization of Pol II at the site of an integrated HCMV-IE gene cluster is dependent on the transcriptional activity of the gene, we performed double labeling experiments. For this purpose, untreated and cycloheximide-treated cells were first hybridized with a probe detecting IE mRNA and then stained for Pol IIO. These experiments were repeated three times and for each experiment at least 100 cells were examined. In more than 95% of the cells showing induction or IE gene expression, a co-localization between Pol IIO and IE mRNA was observed at the IE transcription site (Figures 2A–2C). Because these induced cells did not show any cytoplasmic staining for Pol IIO, it is most unlikely that the observed co-localization is due to crossreactivity of MAb CC-3 with IE mRNA. Noninduced cells revealed no co-localization between Pol IIO and IE DNA (Figures 2D–2F). In less than 1% of the cells, an association of IE DNA with Pol IIO was observed, which could be explained as a chance process when a random localization of the IE gene cluster in the cell nucleus is considered. Because not all speckles are present in the same focal plane as that of the IE gene cluster, some speckles are not visible in the figures shown.

Next, cells were first induced for IE gene expression, then treated with the transcription inhibitor DRB and finally heat-denatured before hybridization. As illustrated in Figures 2G–2I, a co-localization of Pol IIO with the IE gene was no longer observed. Analysis of 50 induced cells that were incubated with DRB for 30 min revealed that in more than 90% of the induced cells there was no association between Pol IIO and IE DNA. Cells induced for IE gene expression could still be recognized by the presence of IE mRNA in the cytoplasm after DRB treatment. When the heatdenaturation step of DRB-treated cells before hybridization was omitted, no nuclear signals were observed, whereas cytoplasmic RNA signals were still present. This confirms that the nuclear signals observed after DRB treatment and denaturation of the cells are indeed showing hybridization to IE DNA.

Figure 1

Localization pattern of RNA Pol IIO (A,B) and RNA Pol IIA (C-E) in rat 9G cells. Cells were fixed in formaldehyde/acetic acid, pretreated with pepsin, and incubated with MAb CC-3 for detection of Pol IIO and with MAb 8WG16 for detection of Pol IIA. Pol IIO is present in irregularly shaped speckles and in many small spots throughout the nucleoplasm, excluding nucleoli (A). (B) The DAPI image of the cell in A. The MAb 8WG16 also stains many small spots throughout the nucleoplasm and one to three larger foci (C, arrows). Because the same cells were also hybridized with a probe for poly A sequences, it appears that these larger intensely stained foci do not co-localize with speckle domains containing poly A sequences (D,E). (D) Staining of poly A sequences in speckles in addition to a more diffuse staining of the nucleoplasm and cytoplasm of the cell. (E) An overlay of images C and D. Bars = 10 μm.

Figure 2

Combined detection of IE RNA/DNA and Pol IIO in cycloheximide-induced (A-C), noninduced (D-F), and in cyclohexamide-induced DRB-treated rat 9G cells (G-I). Cells were first hybridized with a probe detecting IE RNA/DNA and then incubated with MAb CC-3. Cycloheximide-induced cells revealed a clear co-localization of nuclear IE RNA (green spot in A) and Pol IIO (B) at the IE transcription site (yellow spot in C). Because of the short exposure time (1 sec) and being out of focus, cytoplasmic IE mRNA is not visible in A. In noninduced cells, IE DNA (green spot in D) does not co-localize with a speckle containing Pol IIO (E), as shown in the overlay of the two (F). In addition, when cells are first induced for IE expression and then treated with the transcription inhibitor DRB, a co-localization between IE DNA (bright green spot in G) and Pol IIO (H) is no longer observed (I). Cell nuclei are counterstained with DAPI (blue staining). Bar = 10 μm.

Because DRB inhibits phosphorylation of Pol IIO by inhibiting a CTD kinase (Payne and Dahmus 1993; Dubois et al. 1994b; Yankulov et al. 1995; Marshall et al. 1996), treatment of cells with DRB may have a direct influence on the redistribution of Pol IIO. Therefore, cycloheximide-induced cells were also incubated in the presence of actinomycin D, a DNA intercalator and potent inhibitor of transcription (Perry and Kelly 1970). It should be noted, however, that at the concentration used in this study, actinomycin D may inhibit dephosphorylation of Pol IIO (Dubois et al. 1994a). Again, no co-localization of Pol IIO and IE DNA was observed, suggesting loss of Pol IIO at the IE integration site (result not shown).

Recently, we have shown that the association of splicing factors with the IE gene cluster is dependent on its transcriptional activity (Dirks et al. 1997). From the experiments described above, we can conclude that Pol IIO behaves in a similar dynamic manner as splicing factors, suggesting that both are coordinately recruited to sites of active gene transcription.

Next, we performed double labeling experiments combining the detection of IE mRNA with that of Pol IIA. The results showed that Pol IIA was present in small foci throughout the nucleoplasm but did not co-localize with IE mRNA (not shown). Often, a small Pol IIA-positive spot was found associated with an IE transcription domain. However, we cannot conclude whether this represents a functional interaction because of the abundance and widespread distribution of Pol IIA in the cell nucleus. In addition, it cannot be excluded that the number of Pol IIA molecules associated with IE mRNA is below the detection threshold of the technique used because we were unable to detect convincingly the association of Pol IIA with the 50 copies of the IE gene. Nevertheless, our results, which suggest that Pol IIA is not associated with accumulated RNA near the IE gene cluster, are in agreement with biochemical studies showing that Pol IIA is present in transcription initiation complexes only and is not involved in transcription elongation.

Hyperphosphorylated Pol II Co-localizes with Splicing Factors and IE Pre-mRNA Beyond the Dimensions of the Gene Cluster

The association of Pol IIO with sites of IE gene expression was analyzed in more detail using probes that are specific for IE mRNA, DNA, and intron sequences. Double labeling of induced cells with a RNA probe specific for IE DNA and MAb CC-3 revealed that Pol IIO not only co-localizes with the IE gene cluster but also accumulates near this cluster (Figures 3A–3C). This accumulation domain often had the shape of an irregular dot and occasionally of an elongated track. To determine whether Pol IIO co-localizes with IE mRNA in these domains, we performed triple labeling experiments in which IE DNA, IE mRNA, and Pol IIO were visualized simultaneously. Figures 3D–3I show that Pol IIO co-localizes completely with IE mRNA that accumulated in dot- (Figures 3D–3F) or track-(Figures 3G–3I) like domains near the transcribing gene cluster. This complete co-localization was observed in all cells analyzed showing IE gene expression. The gene cluster was always observed as a small spot at one site or in the middle of a dot of IE RNA or at one site of an elongated track. To prove that splicing factors are also present in these domains, cells were first hybridized with the pSS probe and then incubated with MAb anti-m3G. In accordance with our previous observations (Dirks et al. 1997), splicing factors co-localized completely with domains of IE mRNA (Figures 3J–3L). As a control, combined detection of IE DNA/RNA with Pol IIO or splicing factors was performed with different combinations of fluorochromes (AMCA, FITC, Texas Red, Cy3, and Cy5). These experiments showed that the observed co-localization of IE RNA with Pol IIO and splicing factors was not the result of bleeding-through phenomena of the microscopic filters. Previously, we suggested that the IE mRNA dot- and track-like domains would represent accumulation sites of non- or partially spliced IE transcripts in which the splicing process is completed and from which fully spliced transcripts are transported to the cytoplasm (Dirks et al. 1995). To prove that pre-mRNA transcripts are present in these domains we initially designed oligonucleotide probes spanning an intron/exon border. However, the results obtained with such probes proved difficult to interpret. The hybridization signals were mostly too weak to draw firm conclusions. Therefore, we designed a PCR probe specific for the first of three introns present in IE pre-mRNA and applied it in double label experiments with MAb C-33. Figures 3M–3O show that Pol IIO and IE intron sequences co-localize in domains at sites of IE transcription.

Figure 3

Combined detection of IE RNA and/or DNA with Pol IIO (A-H, M-O) and splicing factors (J-L) in cycloheximide-induced cells. Cells hybridized with an IE DNA-specific probe and stained with CC-3 show that Pol IIO (B) accumulates near the IE gene cluster (A,C). The blue cytoplasmic counterstain in C was obtained by a long exposure of cytoplasmic autofluorescence to UV. Cells double-hybridized with an IE DNA- and RNA-specific probe and stained with MAb CC-3 show that IE mRNA (blue signal in D and G) accumulates beyond the dimensions of the IE gene cluster (small green spots indicated by arrows in D and G). Pol IIO (E,H) co-localizes completely with the dot- (F) and track-like (I) accumulations of IE RNA. Cells hybridized with a probe for IE RNA and stained with MAb anti-m3G show that splicing factors (K) also co-localize with track-like domains of IE RNA (J). The orange signal in L shows the co-localization. Cells first hybridized with an IE intron-specific probe and then incubated with MAb CC-3 revealed an IE (pre-)mRNA accumulation site (red dot in M) and speckles containing Pol IIO (N). The overlay of the two images shows that Pol IIO co-localizes with the site at which IE intron sequences accumulated (yellow dot in O). Bars = 10 μm.

Discussion

Previous immunocytochemical studies using MAb CC-3 demonstrated that Pol IIO is localized in speckle domains and in addition is more diffuse throughout the nucleoplasm, except for nucleoli, in a variety of cell types (Bisotto et al. 1995). Furthermore, it was shown that Pol IIO co-localizes with splicing factors in these speckle domains (Bisotto et al. 1995). The results of our double and triple labeling experiments in which we combined the detection of IE mRNA, IE DNA, and splicing factors with that of Pol II show that there exists a clear temporal and spatial association between these components. In a previous study we have shown that splicing factors are recruited to sites of induced IE gene transcription and accumulate with IE mRNA in dot- and track-like domains near the IE gene (Dirks et al. 1997). We explained these results by arguing that not all introns are removed co-transcriptionally due to the high transcriptional activity of the IE gene cluster. Consequently, non- or partially spliced IE transcripts accumulate, together with splicing factors, in domains near the gene where the splicing reaction is completed. The observation that Pol IIO also co-localizes completely with splicing factors and IE mRNA in these domains suggests that Pol IIO does not disperse immediately in the nucleoplasm after being released from the transcribing gene but is retained at this site. A plausible explanation for this could be that Pol IIO is still physically associated with splicing factors after being released from the IE genes and when the splicing factors are actively engaged in the splicing of IE RNA transcripts that accumulated near the transcribing gene cluster. Alternatively or additionally, Pol IIO and splicing factors may be recruited as a complex to IE pre-mRNAs from other sites in the nucleus to participate in the splicing reaction only, which may be interpreted as the formation of a speckle at the IE gene cluster. Although both possibilities are consistent with biochemical studies showing that Pol IIO and splicing factors can exist as a complex even in the absence of pre-mRNA (Kim et al. 1997), we consider the latter possibility less likely. Experiments in which the in vivo processing efficiency of intron-containing transcripts generated by an artificial hybrid gene containing an RNA Pol III promoter were studied revealed that these transcripts are not spliced and not polyadenylated (Sisodia et al. 1987). These data suggest that splicing and polyadenylation are coupled to transcription by RNA Pol II. Interestingly, it was suggested then that this coupling is achieved by the 5′m⁷G cap present on Pol II-derived transcripts only. We now know, however, that splicing factors and factors required for polyadenylation can be linked to the CTD of Pol IIO (see Introduction). The results of the study by Sisodia et al. also imply that splicing factors are not recruited by nascent pre-mRNAs but preferentially by the hyperphosphorylated CTD of the elongating Pol II. Consistent with this implication are the observations that antibodies against Pol II inhibit splicing in vitro (Chabot et al. 1995; Yuryev et al. 1996) and that overexpression of CTD polypeptides inhibits splicing of β-globin transcripts in vivo (Du and Warren 1997).

In agreement with previous data (Dirks et al. 1997), our study shows that the IE gene cluster is not preferentially associated with speckles containing splicing factors and Pol IIO. This implies that, after induction of gene expression, the IE gene has to move to a speckle or, alternatively, that Pol II and splicing factors move to the expression site. Although we cannot exclude the first possibility, we consider it more likely that splicing factors move to sites of active gene expression because transcriptionally inactive IE gene clusters are often found at a relatively big distance from speckles (Dirks et al. 1997). Furthermore, in vivo studies strongly suggested a dynamic movement of splicing factors to transcriptionally activated IE genes (Misteli et al. 1997).

Our results taken together with biochemical data support a model in which splicing factors are recruited by the hyperphosphorylated CTD of elongating Pol IIO, after which they associate with this domain and participate in co-transcriptional splicing. When the transcription rate is high and partially or nonspliced transcripts accumulate near the transcribing gene, Pol IIO will dissociate from the gene after transcription but will remain associated with splicing factors while the splicing reaction is being completed. Finally, Pol IIO and splicing factors may return to speckles as a complex where they are recycled and stored.

We can only speculate about the mechanism by which Pol II and splicing factors are recruited from speckles to sites of active gene expression. It is possible that they are recruited as a complex, in which case Pol IIO must be dephosphorylated before association with a promoter region and splicing factors must be released. Alternatively, Pol IIO is converted to Pol IIA just before or after being released from the speckle, in which case Pol II and splicing factors are recruited separately. In this respect, it should be noted that associations of splicing factors with the hypophosphorylated form of RNA Pol II have not been observed (Mortillaro et al. 1996; Kim et al. 1997).

References

Baurén

Jiang

W-Q

Bernholm

Wieslander

(1996) Demonstration of a dynamic, transcription-dependent organization of pre-mRNA splicing factors in polytene nuclei. J Cell Biol 133:929–941.

Baurén

Wieslander

(1994) Splicing of Balbiani ring 1 gene pre-mRNA occurs simultaneously with transcription. Cell 76:183–192.

Beyer

Osheim

(1988) Splice site selection, rate of splicing, and alternative splicing on nascent transcripts. Genes Dev 2:754–765.

Bisotto

Lauriault

Duval

Vincent

(1995) Co-localization of a high molecular mass phosphoprotein of the nuclear matrix (p255) with spliceosomes. J Cell Sci 108:1873–1882.

Boom

Geelen

Sol

Raap

Minnaar

Klaver

van der Noordana

(1986) Establishment of a rat cell line inducible for the expression of human cytomegalovirus immediate early gene products by protein synthesis inhibitors. J Virol 58:851–859.

Bregman

van der Zee

Warren

(1995) Transcription-dependent redistribution of the large subunit of RNA polymerase II to discrete nuclear domains. J Cell Biol 129:287–298.

Carter

Bowman

Carrington

Fogarty

McNeil

Fay

Lawrence

(1993) A three-dimensional view of precursor messenger RNA metabolism within the mammalian nucleus. Science 259:1330–1335.

Carter

Taneja

Lawrence

(1991) Discrete nuclear domains of poly(A) RNA and their relationship to the functional organization of the nucleus. J Cell Biol 115:1191–1202.

Chabot

Bisotto

Vincent

(1995) The nuclear matrix phosphoprotein p255 associates with splicing complexes as part of the [U4/U6.U5] tri-snRNP particle. Nucleic Acids Res 23:3206–3213.

10.

Cho

E-J

Takagi

Moore

Buratowski

(1997) mRNA capping enzyme is recruited to the transcription complex by phosphorylation of the RNA polymerase II carboxyterminal domain. Genes Dev 11:3319–3326.

11.

Dahmus

(1996) Reversible phosphorylation of the C-terminal domain of RNA polymerase II. J Biol Chem 271:19009–19012.

12.

Dirks

(1996) RNA molecules lighting up under the microscope. Histochem Cell Biol 106:151–166.

13.

Dirks

Daniël

Raap

(1995) RNAs radiate from gene to cytoplasm as revealed by fluorescence in situ hybridization. J Cell Sci 108:2565–2572.

14.

Dirks

de Pauw

ESD

Raap

(1997) Splicing factors associate with nuclear HCMV-IE transcripts after transcriptional activation of the gene, but dissociate upon transcription inhibition: evidence for a dynamic organization of splicing factors. J Cell Sci 110:515–522.

15.

Dirks

Raap

(1995) Cell-cycle-dependent gene expression studied by two-colour fluorescent detection of a mRNA and histone mRNA. Histochem Cell Biol 104:391–395.

16.

Dirks

van de Rijke

Fujishita

van der Ploeg

Raap

(1993) Methodologies for specific intron and exon localization in cultured cells by haptenized and fluorochromized probes. J Cell Sci 104:1187–1197.

17.

Warren

(1997) A functional interaction between the carboxy-terminal domain of RNA polymerase II and pre-mRNA splicing. J Cell Biol 136:5–18.

18.

Dubois

M-F

Bellier

Seo

S-J

Bensaude

(1994a) Phosphorylation of the RNA polymerase II largest subunit during heat shock and inhibition of transcription in HeLa cells. J Cell Physiol 158:417–426.

19.

Dubois

M-F

Nguyen

Bellier

Bensaude

(1994b) Inhibitors of transcription such as 5,6-dichloro-1-β-D-ribofuranosylbenz-imidazole and isoquinoline sulfonamide derivatives (H-8 and H-7) promote dephosphorylation of the carboxyl-terminal domain of RNA polymerase II largest subunit. J Biol Chem 269:13331–13336.

20.

Fakan

(1994) Perichromatin fibrils are in situ forms of nascent transcripts. Trends Cell Biol 4:86–90.

21.

Grande

van der Kraan

de Jong

van Driel

(1997) Nuclear distribution of transcription factors in relation to sites of transcription and RNA polymerase II. J Cell Sci 110:1781–1791.

22.

Greenleaf

(1993) Positive patches and negative noodles: linking RNA processing to transcription?. Trends Biochem Sci 18:117–119.

23.

Ishov

Stenberg

Maul

(1997) Human cytomegalovirus immediate early interaction with host nuclear structures: definition of an immediate transcript environment. J Cell Biol 138:5–16.

24.

Jackson

Hassan

Errington

Cook

(1993) Visualization of focal sites of transcription within human nuclei. EMBO J 12:1059–1065.

25.

Jiménez-García

Spector

(1993) In vivo evidence that transcription and splicing are coordinated by a recruiting mechanism. Cell 73:47–59.

26.

Kim

Bregman

Warren

(1997) Splicing factors associate with hyperphosphorylated RNA polymerase II in the absence of pre-mRNA. J Cell Biol 136:19–28.

27.

Laybourn

Dahmus

(1990) Phosphorylation of RNA polymerase IIA occurs subsequent to interaction with the promoter and before the initiation of transcription. J Biol Chem 265:13165–13173.

28.

Marshall

Peng

Xie

Price

(1996) Control of RNA polymerase II elongation potential by a novel carboxyl-terminal domain kinase. J Biol Chem 271:27176–27183.

29.

McCracken

Fong

Rosonina

Yankulov

Brothers

Siderovski

Hessel

Foster

Amgen

EST

Program Shuman

Bentley

(1997a) 5′-Capping enzymes are targeted to premRNA by binding to the phosphorylated carboxy-terminal domain of RNA polymerase II. Genes Dev 11:3306–3318.

30.

McCracken

Fong

Yankulov

Ballantyne

Pan

Greenblatt

Patterson

Wickens

Bentley

(1997b) The C-terminal domain of RNA polymerase II couples mRNA processing to transcription. Nature 385:357–361.

31.

Misteli

Céceres

Spector

(1997) The dynamics of a premRNA splicing factor in living cells. Nature 387:523–527.

32.

Misteli

Spector

(1996) Serine/threonine phosphatase 1 modulates the subnuclear distribution of pre-mRNA splicing factors. Mol Biol Cell 7:1559–1572.

33.

Moen

Smith

Lawrence

(1995) Compartmentalization of specific pre-mRNA metabolism: an emerging view. Hum Mol Genet 4:1779–1789.

34.

Mortillaro

Blencowe

Wei

Nakayasu

Warren

Sharp

Berezney

(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.

35.

Neugebauer

Roth

(1997) Distribution of pre-mRNA splicing factors at sites of RNA polymerase II transcription. Genes Dev 11:1148–1159.

36.

O'Keefe

Mayeda

Sadowski

Krainer

Spector

(1994) Disruption of pre-mRNA splicing in vivo results in reorganization of splicing factors. J Cell Biol 124:249–260.

37.

Payne

Dahmus

(1993) Partial purification and characterization of two distinct protein kinases that differentially phosphorylate the carboxy-terminal domain of RNA polymerase subunit IIa. J Biol Chem 268:80–87.

38.

Perry

Kelly

(1970) Inhibition of RNA synthesis by actinomycin D: characteristic dose-response of different RNA species. J Cell Physiol 76:127–139.

39.

Reuter

Appel

Bringmann

Rinke

Lührmann

(1984) 5′-Terminal caps of snRNAs are reactive with antibodies specific for 2,2,7-trimethylguanosine in whole cells and nuclear matrices. Exp Cell Res 154:548–560.

40.

Sisodia

Sollner-Webb

Cleveland

(1987) Specificity of RNA maturation pathways: RNAs transcribed by RNA polymerase III are not substrates for splicing or polyadenylation. Mol Cell Biol 7:3602–3612.

41.

Spector

(1993) Macromolecular domains within the cell nucleus. Annu Rev Cell Biol 9:265–315.

42.

Spector

O'Keefe

Jiménez-García

(1993) Dynamics of transcription and pre-mRNA splicing within the mammalian cell nucleus. Cold Spring Harbor Symp Quant Biol 58:799–805.

43.

Steinmetz

(1997) Pre-mRNA processing and the CTD of RNA polymerase II: the tail that wags the dog?. Cell 89:491–494.

44.

Thompson

Steinberg

Aronson

Burgess

(1989) Inhibition of in vivo and in vitro transcription by monoclonal antibodies prepared against wheat germ RNA polymerase II that react with the heptapeptide repeat of eukaryotic RNA polymerase II. J Biol Chem 264:11511–11520.

45.

van de Corput

MPC

Dirks

van Gijlswijk

RPM

van de Rijke

Raap

(1998) Fluorescence in situ hybridization using horseradish peroxidase labeled oligonucleotides and tyramide signal amplification for sensitive DNA and mRNA detection. Histochem Cell Biol 110:431–437.

46.

Vincent

Lauriault

Dubois

M-F

Lavoie

Bensaude

Chabot

(1996) The nuclear matrix protein p255 is a highly phosphorylated form of RNA polymerase II largest subunit which associates with spliceosomes. Nucleic Acids Res 24:4649–4652.

47.

Wansink

Schul

van der Kraan

van Steensel

van Driel

de Jong

(1993) Fluorescent labeling of nascent RNA reveals transcription by RNA polymerase II in domains scattered throughout the nucleus. J Cell Biol 122:283–293.

48.

Weeks

Hardin

Shen

Lee

Greenleaf

(1993) Locus-specific variation in phosphorylation state of RNA polymerase II in vivo: correlations with gene activity and transcript processing. Genes Dev 7:2329–2344.

49.

Wuarin

Schibler

(1994) Physical isolation of nascent RNA chains transcribed by RNA polymerase II: evidence for cotranscriptional splicing. Mol Cell Biol 14:7219–7225.

50.

Xing

Johnson

Dobner

Lawrence

(1993) Higher level organization of individual gene transcription and RNA splicing. Science 259:1326–1330.

51.

Xing

Johnson

Moen

McNeil

Lawrence

(1995) Nonrandom gene organization: structural arrangements of specific pre-mRNA transcription and splicing with SC-35 domains. J Cell Biol 131:1635–1647.

52.

Yankulov

Yamashita

Roy

Egly

J-M

Bentley

(1995) The transcriptional elongation inhibitor 5,6-dichloro-1-β-D-ribofuranosylbenzimidazole inhibits transcription factor IIH-associated protein kinase. J Biol Chem 270:23922–23925.

53.

Yuryev

Patturajan

Litingtung

Joshi

Gentile

Gebara

Corden

(1996) The C-terminal domain of the largest subunit of RNA polymerase II interacts with a novel set of serine/arginine-rich proteins. Proc Natl Acad Sci USA 93:6975–6980.

54.

Zeng

Kim

Warren

Berget

(1997) Dynamic relocation of transcription and splicing factors dependent upon transcriptional activity. EMBO J 16:1401–1412.