Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Transcription of the human U2 snRNA genes continues beyond the 3′ box in vivo

Transcription of the human U2 snRNA genes continues beyond the 3′ box in vivo The EMBO Journal Vol.18 No.10 pp.2867–2877, 1999 Transcription of the human U2 snRNA genes continues beyond the 3 box in vivo Vegvar et al., 1986; Ach and Weiner, 1987; reviewed in Paula Cuello, Diana C.Boyd, Michael J.Dye, Hernandez, 1992). Mutation of this sequence causes the Nick J.Proudfoot and Shona Murphy accumulation of transcripts that are not processed accur- Chemical Pathology Unit, Sir William Dunn School of Pathology, ately and are therefore longer than the mature snRNA. University of Oxford, South Parks Road, Oxford OX1 3RE, UK Interestingly, the 3 box functions efficiently only in the Corresponding author context of a pol II-dependent snRNA promoter. Although e-mail: [email protected] these genes are transcribed by the same polymerase as mRNA genes, their promoters are organized differently. The 3 box of the human class II snRNA genes is Class II snRNA genes have characteristic short, compact, required for proper 3 processing of transcripts, but TATA-less promoters containing a distal sequence element how it functions is unclear. Several lines of evidence (DSE) and a proximal sequence element (PSE) (reviewed suggest that termination of transcription occurs at the by Hernandez, 1992; Lobo and Hernandez, 1994). Replace- 3 box and the terminated transcript is then a substrate ment of a U1 or U2 promoter with an mRNA promoter for processing. However, using nuclear run-on analysis inhibits accurate 3 end formation of the transcripts of endogenous genes, we demonstrate that transcription (Hernandez and Weiner, 1986; Neuman de Vegvar et al., continues for at least 250 nucleotides beyond the 1986). 3 box of the U2 genes. Although in vivo footprinting How the 3 box functions is unknown, but several analysis of both the U1 and U2 genes detects no findings suggest that it may act as a transcription termin- protein–DNA contacts directly over the 3 box, a series ator: transcription of the human U1 genes terminates very of G residues immediately downstream from the close to the 3 box (Kunkel and Pederson, 1985); only 3 box of the U1 gene are clearly protected from very small amounts of long snRNA gene transcripts methylation by dimethylsulfate. In conjunction with containing a copy of the 3 box have been detected the 3 box of the U1 gene, this in vivo footprinted (Hernandez, 1985; Kunkel and Pederson, 1985; Ach and region causes termination of transcription of transi- Weiner, 1987; Lobo and Marzluff, 1987) and these do not ently transfected U2 constructs, whereas a 3 box alone appear to be precursors for processed RNA either in HeLa does not. Taken together, these results indicate that the cells (Lobo and Marzluff, 1987) or in Xenopus oocytes 3 box is not an efficient transcriptional terminator (Ciliberto et al., 1986). Thus, an attractive hypothesis is but may act as a processing element that is functional that transcription terminates just upstream of the 3 box in the nascent RNA. and this produces a transcript that is a substrate for further Keywords: pol II/RNA processing/snRNA/termination/ processing, in analogy to the mechanism of 3 end transcription formation of vertebrate pre-rRNA (Kuhn and Grummt, 1989). However, a more direct role for this element in processing of snRNAs has not been ruled out and, in yeast, the 3 ends of U1, U2 and U5 RNAs are produced Introduction by nuclease cleavage of longer precursors (Chanfreau et al., 1997; Elela and Ares, 1998; Seipelt et al., 1999). The vertebrate class II snRNA genes encode short non- In order to understand the exact role of the 3 box, we polyadenylated nuclear RNAs, including the U1, U2, have undertaken a study of transcriptional termination of U4 and U5 snRNAs that are required for pre-mRNA the U2 snRNA genes using nuclear run-on analysis. We processing. These genes are transcribed by RNA poly- find that, in contrast to the U1 gene, transcription of the merase II (pol II) to give 3-extended precursors that U2 gene continues for at least 250 bp after the 3 box. undergo processing themselves to produce the mature Transcripts derived from the region downstream of the 3 RNA components of snRNPs. Precursors with up to 16 box are contiguous with RNA from upstream and are nucleotide 3 end extensions can be detected, and these detected at very early points in a run-on time course, undergo cap trimethylation, internal base modification and strongly suggesting that the profile we see reflects the 3 end trimming after association with U snRNP proteins in vivo polymerase loading. in the cytoplasm (recently reviewed by Huang et al., A complementary in vivo footprinting analysis of the 1997). The processed RNA and associated proteins are 3 box region of the U1 and U2 genes also gave surprising then transported back into the nucleus where assembly of results. No protection is detected over the 3 box of either the functional snRNP is completed (reviewed in Terns gene, whereas a clear area of protection from methylation et al., 1993; Nagai and Mattaj, 1994). 3 End formation by dimethylsulfate (DMS) is apparent immediately down- of mature snRNAs requires a cis-acting sequence, known as the 3 box, located 9–19 nucleotides downstream from stream from the 3 box of the U1 gene. the coding region of these snRNA genes (Hernandez, These results suggest that termination of transcription 1985; Yuo et al., 1985; Ciliberto et al., 1986; Neuman de may be occurring close to the 3 box in the U1 genes due © European Molecular Biology Organization 2867 P.Cuello et al. to protein(s) binding to this DNA element. We provide direct evidence for this by showing that this region terminates transcription from a U2 promoter when placed downstream of a 3 box. In contrast, a 3 box alone does not function as an efficient terminator. Taken together, these results indicate that most nascent snRNA gene transcripts contain a full copy of the 3 box and transcription terminates beyond this element. We consider it likely, therefore, that the 3 box is an RNA element that is required, in conjunction with a promoter- dependent activity, for rapid cleavage to produce the 3 extended RNAs that are detectable in vivo. Results Transcription continues beyond the 3 box of the U2 genes A precise determination of the site of termination of transcription of the snRNA genes is crucial to an under- standing of the role of the 3 box in this process. We have therefore undertaken a detailed study of the polymerase loading on the human U2 genes using 47–50mer oligo- nucleotide probes in a nuclear run-on analysis. There are ~10 U2 genes in the human haploid genome at 17q21–22 (Hammarstrom et al., 1984; Lindgren et al., 1985a; Pavelitz et al., 1995). These are tandemly repeated and the sequence of each 5.8–6.1 kb repeat unit is highly Fig. 1. Transcription continues beyond the 3 box of the U2 genes. conserved (Van Arsdell and Weiner, 1984; Westin et al., (A) A diagram of the structure of the U2 gene is shown with the 1984; Pavelitz et al., 1995), allowing us to analyse the relative positions of the probes marked below. The numbers noted next to the probes indicate the start or end of the probes used relative to the distribution of polymerases over all of these genes using site of initiation (US and D) or relative to the end of the 3 box (E, J one set of probes. The position of each probe relative to and Z). The results of a run-on analysis and hybridization of the U2 gene and the results of a representative run-on synthetically produced RNA to the probes are shown below each analysis are shown in Figure 1A. A 47mer oligonucleotide probe. (B) A graphic representation is shown of the results of the run-on analysis in (A) as a percentage of the signal over probe C after complementary to probe A was included on the filters as subtraction of α-amanitin-resistant transcription and correction for the a negative control (AS) and a 50mer oligonucleotide hybridization efficiency of each probe. complementary to the pol III-transcribed 7SK RNA (7SK) was included as a positive control. The hybridization efficiency of the probes US to Z was determined using a stretch which is not as highly conserved between labelled synthetic RNA (see Materials and methods), and individuals as the other sequences in the U2 genes (Liao the nuclear run-on results were normalized to these values and Weiner, 1995). These results suggest that a large after subtraction of α-amanitin-resistant transcription (see proportion of transcription termination occurs downstream Figure 3). of the 3 box in the U2 genes, perhaps mainly in the No signal was detected over probe B in either the run- region between J and Z. on or the synthetic RNA control, possibly due to the relatively high A/T content of this sequence, precluding Transcripts downstream of the 3 box are analysis of this region. Probe B was included in subsequent contiguous with the transcripts from the coding run-on assays as an additional background control. The region and are detected early in a run-on time profile of polymerase density over the U2 genes determined course by several such experiments is shown in Figure 1B as a Although the pattern of run-on transcripts over the U2 percentage of the level of signal over probe C. gene suggests that a significant number of polymerases As expected, the polymerase density is low upstream continue transcription beyond the 3 box, we cannot of the site of initiation of the U2 genes (probe US, 2.65%), exclude the possibility that transcription has reinitiated indicating that transcription from upstream genes does not downstream of the coding region. We have therefore used continue into these promoters. However, the pattern of an RNA probe complementary to region A–D of the U2 polymerase density over the U2 genes is strikingly different gene to hybrid-select transcripts contiguous with the from that seen on the U1 genes where transcription was coding RNA. A biotin-labelled selection probe was hybrid- found to terminate close to the 3 box (Kunkel and ized to the run-on RNA, and avidin-coupled magnetic Pederson, 1985). Instead, a high level of polymerase beads were used to separate the probe and associated loading appears to continue until at least 250 bp past the RNA from the mix (see Materials and methods). The 3 box (probe J, 95%). The polymerase density over selected RNA was then hydrolysed and hybridized to region Z, ~600 bp downstream of the 3 box, has dropped filters containing probes US to Z, AS and 7SK (Figure to 18%. We have not scored the polymerase density 2A). Importantly, transcripts from the 7SK gene have not between the regions J and Z since this contains a (CT) been selected, indicating that the selection is specific. 2868 Termination of transcription of snRNA genes Fig. 2. Transcripts downstream of the U2 3 box are contiguous with the transcripts from the coding region. (A) The results of hybridization of RNA that has been hybrid-selected after the run-on are shown below the profile from an unselected run-on (a lower exposure of the same experiment shown in Figure 1A). (B) A graphic representation of the results from an unselected run-on analysis (black bars) is shown compared with the results with RNA hybrid-selected after the run-on (cross-hatched bars), corrected as in Figure 1. Fig. 3. Transcripts downstream of the U2 3 box are detected early in Figure 2B shows the pattern of hybridization from this a run-on time course. (A) The results of adding α-amanitin (2 μg/ml) analysis (cross-hatched bars) compared with that seen to the run-on reaction at different times are shown. The time (in with an unselected run-on analysis performed at the same minutes) is shown at the left of the figure. The result of hybridization to the 7SK probe is shown for each time point at the right. time (black bars). (B) A graphic representation of the results corrected for the 7SK After hybrid selection, the signal over US relative to C signal is shown for the 2 min (black bars), 5 min (cross-hatched bars) has fallen from 2.6% to 1%, while the relative signals and 10 min (empty bars) time points. over probes A–G have remained at approximately the same levels. At least 50% of the transcripts complementary to probes H–I are also selected and are therefore contiguous as an internal control since this should be unaffected by with transcripts from the coding region. Some of the the levels of α-amanitin used. When α-amanitin is added transcripts detected over probes J and Z are also contiguous before the addition of the radioactive nucleotide, a signal with the coding RNA. The drop in signals over probes is detected over probe C, indicating that transcripts pro- H–Z after hybrid selection indicates that some of the duced by either pol I or pol III are hybridizing to this apparent transcription from this region may be the result probe. Accordingly, these background signals have been of cross-hybridization or reinitiation. Alternatively, post- subtracted from the signals over probe C for all other run- transcriptional nuclease digestion (specific or non-specific) on assays. Signals are visible over at least some of the may be responsible. For example, some of the transcripts probes downstream of the 3 box as early as 1 min. detected over probes H–Z in the unselected run-on However, the signals at this time point are too low to be analysis may have been cleaved specifically in a 3 box- quantitated accurately. Therefore, only the results for the dependent reaction. 2, 5 and 10 min time points are depicted graphically in The above results indicate that transcription is not Figure 3B after normalizing to the level of the 7SK signal. terminating efficiently at the 3 box in our run-on assay. The signals over probes A–Z increase from 1 to 10 min, However, transcription potentially can extend 1000 bp indicating that transcription is continuing during this time. in the 15 min of a run-on reaction (Weber et al., 1977). After 2 min, the signals are quite evenly distributed over In addition, polymerases that are normally paused may probes A–J. Although all the signals have increased resume transcription under run-on conditions (see, for ~10-fold from the 2 to 10 min time points (relative to example, Strobl and Eick, 1992). We have therefore carried 7SK), the pattern has not changed markedly, apart from a out a time course of the run-on reaction to investigate the relative increase over probe C. This may suggest that possibility that polymerases are not transcribing the region polymerases are stacking over probe C or that some of downstream of the 3 box in intact cells (Figure 3). this signal is due to cross-hybridization to non-U2 gene α-Amanitin was added to 2 μg/ml, either before or 1, transcripts. Importantly, there is no obvious increase in 2, 5 or 10 min after the addition of nucleotides, to the the signals over the downstream probes relative to probes resuspended nuclei (see Materials and methods) (Figure A–D with time, and the absolute signal over probe A 3A). The probe for the pol III-transcribed 7SK gene serves increases with time. Reinitiation of transcription by pol II is 2869 P.Cuello et al. are not identical in the seven of the 15–30 U1 genes sequenced so far, although they are highly related for up to at least 70 bp past this element (Manser and Gesteland, 1982; Lund and Dahlberg, 1984). We have therefore used a mixture of two longer (80mer) oligonucleotide probes (C and D) complementary to the region from 8 to 88 bp downstream of the coding region of the HSD2 U1 gene (Manser and Gesteland, 1982) and the HU1-1 U1 gene (Lund and Dahlberg, 1984) to increase the possibility of detecting transcripts from this region. Oligos C and D include the complement to the 3 box which spans the sequence from 11 to 24 bp downstream of the coding region. The hybridization efficiency of the probes A–D was determined using labelled synthetic RNAs (Figure 4A) and the nuclear run-on results were normalized to these values after subtraction of α-amanitin-resistant transcrip- tion. The position of each probe relative to the U1 gene and the results of a representative run-on analysis are shown in Figure 4A. The 7SK and AS oligonucleotides have also been included as positive and negative controls, respectively. The profile of polymerase density over the U1 genes determined by three such experiments is shown in Figure 4B as a percentage of the level of signal over Fig. 4. Termination of transcription appears to occur close to the probe A. 3 box of the U1 genes in our assay system. (A) A diagram of the Transcription appears to be much higher over the coding structure of the U1 gene is shown with the relative positions of the region of the U1 genes than directly downstream, where probes marked below. The numbers noted next to the probes indicate it has dropped to 10% of the value for probe A. This the start or end of the probes used relative to the site of initiation (A and B) or relative to the end of the 3 box (C and D). Oligos C suggests that termination of transcription is occuring close and D are complementary to transcripts from the HSD2 and HU1-1 U1 to the 3 box in these genes. The signal detected over alleles, respectively (see text). The results of a run-on analysis and probes C/D may indicate that transcription continues just hybridization of synthetically produced RNA to the probes are shown into this region or that some transcripts corresponding to below each probe. (B) A graphic representation is shown of the results the whole length of the probes are made. Alternatively, of the run-on analysis in (A) as a percentage of the signal over probe A after subtraction of α-amanitin-resistant transcription and the probes we have used may only detect a small proportion correction for the hybridization efficiency of each probe. of the transcripts produced by the U1 genes due to sequence divergence downstream of the 3 box. However, very inefficient in nuclear run-ons (Bentley and Groudine, probes C and D together should detect transcripts from at 1986), and transcripts of mRNA genes are elongated by least the seven sequenced U1 genes (see Manser and ~100 nucleotides in 15 min under our run-on conditions Gesteland, 1982; Lund and Dahlberg, 1984). We therefore (Dye and Proudfoot, 1999). It is therefore likely that think it likely that our results are due to termination of transcription is still running-on over the 47 bp region of transcription of the U1 genes close to the 3 box in our probe A within the 10 min of the time course. assay system and that they confirm the findings of Kunkel These results strongly suggest that polymerases are not and Pederson (1985). artefactually reading through a termination signal under These results are consistent with the notion that the the run-on conditions. 3 box is a termination signal. However, the 3 box of the U2 gene would, therefore, be a very inefficient transcrip- Termination of transcription appears to occur tion termination signal in comparison. close to the 3 box of the U1 genes in our assay system In vivo footprinting detects DNA–protein As mentioned above, Kunkel and Pederson (1985) found interactions downstream of the 3 box of the U1 that transcription of the human U1 genes terminates close gene to the 3 box. Our assay conditions are not identical to The function of the 3 box requires that transcription is theirs and they used longer M13 probes to detect run-on initiated specifically from a class II snRNA promoter transcripts. We obtained essentially identical results to (Hernandez and Weiner, 1986; Neuman de Vegvar et al., those shown above for the U2 genes using the buffer 1986), indicating a strong link between initiation of conditions described by Kunkel and Pederson (1985), transcription and RNA processing, as has recently been although the signals were lower over all probes (data not shown for processing of mRNA (reviewed by Corden shown). We have also carried out an analysis of the and Patturajan, 1997; Neugebauer and Roth, 1997; U1 genes using oligonucleotide probes (Figure 4). To Steinmetz, 1997). investigate the polymerase density across these genes, we In our studies of the DNA–protein interactions required have used two 50mer oligonucleotides complementary to for initiation of transcription of snRNA genes, we have sequences within the coding region (A and B). The carried out an extensive in vivo footprinting analysis of sequences downstream from the 3 box of the U1 genes the human U2 genes (D.C.Boyd, I.Greger and S.Murphy, 2870 Termination of transcription of snRNA genes Fig. 5. In vivo footprinting detects DNA–protein interactions downstream of the 3 box of the U1 gene. (A) A diagram of the structure of the U1 and U2 genes is shown with the relative positions of the in vivo footprinting primers indicated by black arrows. (B) The results of in vivo footprinting over the region of the 3 box of the U1 and U2 genes is shown. The white arrows indicate a reduction in the relative intensity of bands (Gs) in the in vivo treated DNA compared with in vitro modified DNA, and the size of the arrow indicates the magnitude of the reduction. The black arrow indicates a moderate increase in the intensity of a band in the in vivo treated DNA. (C) The sequence of the 3 box region of the HU1-1 U1 gene (Lund and Dahlberg, 1984) and the pTP18 U2 genes (Pavelitz et al., 1995) are shown. The numbers under the U1 sequence indicate the distance from the 3 box. G residues that were protected from methylation by DMS in vivo are indicated in bold. The region downstream of the U1 3 box that is footprinted is indicated by a bracket and a G residue that is hypersensitive after DMS treatment in vivo is marked by a black arrow. unpublished results). Since footprints are easily detected in all of the U1 genes sequenced so far (see Manser and over all known promoter elements of these genes, it is Gesteland, 1982; Lund and Dahlberg, 1984). likely that a large number of the genes are active at any Although these results do not rule out that protein(s) one time and that DNA–protein interactions required for bind to the 3 box and directly mediate termination, they termination of transcription may also be detected. We are consistent with the hypothesis that termination of have therefore analysed the in vivo DNA–protein inter- transcription occurs downstream of this element in both actions over the 3 box region of the U1 and U2 genes the U1 and U2 genes. In addition, they raise the possibility using DMS footprinting. that protein(s) binding just downstream of the 3 box The results of this analysis are shown in Figure 5A mediate termination of transcription of the U1 gene. and B and summarized graphically in Figure 5C. For the U1 gene, the upper strand of HU1-1 (Lund and Dahlberg, The 3 box is required for high levels of correctly 1984) has been analysed and for the U2 gene the lower processed U2 RNA strand has been analysed (see Figure 5A). No methylation To test the function of the 3 box and downstream protection is detected over the 3 box of either the U1 or sequences in processing and termination, we constructed U2 genes, although a G residue just upstream from the a series of U2 genes, marked by the addition of a linker U2 3 box was partially protected from methylation to the coding region, where the 3 box is either present (marked in Figure 5B and C). However, a region of very or absent or has been replaced by U1 sequences (Figure clear protection from methylation is apparent immediately 6A). These constructs were transfected into HeLa cells, downstream from the 3 box of the U1 gene, close to a and the 5 and 3 ends of the transcribed RNAs mapped T-rich stretch of sequence (Figure 5B and C), suggesting by S1 and RNase protection analysis (Figure 6B and C). that a protein is binding here in vivo. Importantly, the The gene for the pol III-dependent VAI RNA was included central 11 nucleotides of the footprint are highly conserved as a co-transfection control. The probes used for these 2871 P.Cuello et al. Fig. 6. The 3 box is required for high levels of correctly processed U2 RNA. (A) The structure of marked U2 genes with changes to the 3 box region is represented. WT is identical to the endogenous U2 gene apart from the insertion of a linker in the coding region. P– is transcriptionally inactive due to a mutation in the promoter. Δ3 BOX has had the 3 box deleted. In U1, the 3 box is replaced by the last 26 bp of the U1 gene (small cross-hatched box) and the 3 box (large cross-hatched box), and U1DS contains the 63 bp just downstream of the U1 3 box in addition (grey box). In DS, the U2 3 box has been replaced by the sequences downstream from the U1 3 box. The relative positions of the S1 probe and riboprobes is noted below the WT structure. The sizes of the specific U2 RNase protection products for the constructs are noted below the WT structure. The size of the additional product from U1 and DS is indicated below the U1 structure. The size of the readthrough products is indicated below each structure. The results of quantitation of S1 (mean and standard deviation) and RNase protection analysis (average of two experiments) are noted on the right of the figure. (B) The results of S1 analysis of RNA transcribed from the constructs in (A) is shown. The construct is noted above each lane, the positions of the S1 products is noted at the left and the position of the probes is noted at the right. (C) The results of RNase protection analysis of RNA transcribed from the constructs in (A) is shown. The construct is noted above each lane. The positions of the protected products are noted on the left. analyses are marked on Figure 6A, and the quantitation 3 box alone (U1DS, lane 5, 67.5%), indicating that of the results is shown to the right of the figure. Since this downstream, in vivo-footprinted region has little effect the 5 S1 probe is complementary to a common region of on the steady-state RNA. Hernandez (1985) also saw little the constructs, the relative amount of steady-state RNA effect of this downstream region on the formation of can be readily assessed (Figure 6B). A high level of 3 ends of transcripts from a U1 promoter. Addition of transcription is detected from the ‘wild-type’ marked U2 the U1 downstream region alone to the Δ3 BOX template gene (WT, lane 1), as seen previously (Murphy, 1997), only further decreased the level of steady-state RNA and mutation of the essential PSE element in the promoter detected (DS, lane 6, 16%). (P–, lane 2) abolishes transcription as expected. Mutation The RNase protection analysis shown in Figure 6C of the 3 box causes a significant reduction in the amount demonstrates the effect of the changes on the 3 ends of steady-state RNA (Δ3 BOX, lane 3, 27% of WT). of transcripts. In this analysis, the endogenous U2 is This is in agreement with previous studies on human complementary to a region in all the probes used (see snRNA genes showing that the 3 box is essential for the Figure 6A) and a protection product of 158 nucleotides accumulation of high steady-state levels of transcripts (EnU2) is therefore present in all lanes. The VAI RNA (Hernandez, 1985; Yuo et al., 1985; Neuman de Vegvar gives an 89 nucleotide product that serves as an internal et al., 1986; Ach and Weiner, 1987; reviewed in Hernandez, transfection control. The major product of the WT con- 1992). Addition of the 3 box of the U1 gene to the Δ3 struct is a band of the expected size for mature marked BOX construct increases the amount of steady-state RNA U2 RNA (WT, lane 1, U2M). There are also distinct (U1, lane 4, 65%), confirming that the 3 boxes are products several nucleotides longer that may correspond largely interchangeable. Replacement of the U2 3 box to the 3-extended precursors of U2 RNA previously and downstream region with the equivalent region of the described (Wieben et al., 1985; Yuo et al., 1985). The U1 gene has essentially the same effect as replacing the same pattern is evident for the endogenous U2 RNA also 2872 Termination of transcription of snRNA genes detected by this analysis (EnU2). The much longer product visible in lane 1 corresponds to small amounts of read- through transcripts mismatching with the end of the probe (RT, 0.3% of total transcripts) (see Figure 6A). Low levels of readthrough transcripts from the endogenous genes are also detected in longer exposures of RNase protection analyses (data not shown). Apart from the endogenous U2 products, none of these bands is detected when the promoter of the U2 gene is disabled (P–, lane 2). When the 3 box is deleted, the amount of mature marked U2 and the potential precursors decreases (Δ3 BOX, lane 3). However, the readthrough transcripts increase to 2% of the total. In the U1 construct, the U2 3 box has been replaced by the last 26 bp of the U1 gene followed by the 3 box (see Materials and methods). Thus, the 3 end directed by the U1 sequences should be distinct from that of the WT construct, and products of the right size (257 nucleotides) are detected for the U1 and U1DS constructs (lanes 4 and 5, U1 3 BOX). Again, potential precursors extended by a few nucleotides at the 3 end are detected. Perhaps surprisingly, an approximately equal level of products equivalent to the mature U2 3 end and associated precursors is also apparent (see below). Approximately the same level of readthrough transcripts is detected for U1 as for the WT construct (lane 4, 0.4%). However, the level of readthrough appears to have decreased when the region downstream of the U1 3 box Fig. 7. The region downstream from the 3 box of the U1 contains a transcription terminator. (A) The structure of the marked U2 gene is present (U1DS, lane 5, 0.1%). Addition of this construct is shown and the relative positions of the probes used for the downstream region alone to the Δ3 box construct has run-on analysis are indicated. (B) The results of run-on analysis of the very little effect on any of the products (DS, lane 6). constructs in Figure 5A are shown. The hybridizations on the left of The results of this analysis confirm that the U2 3 box the figure were carried out with RNA that had been hybrid-selected is necessary for the accumulation of high levels of properly using a probe complementary to sequences in the ‘coding’ RNA from the marked template (see text). The hybridizations on the right were processed transcripts from the marked U2 gene. However, carried out with the RNA in the supernatant from the selection that it appears that proper 3 end formation is still occurring, contains any RNA that has not hybridized to the probe. The probes albeit at a reduced level, in the complete absence of the present on the filters are noted above each slot and the constructs 3 box. This has also been noted in previous studies transfected are noted on the left. The quantitation of the relative levels (Neuman de Vegvar et al., 1986; Ach and Weiner, 1987) of hybridization over the U2 MARK and VEC probes is expressed as a percentage of the ratio for the WT construct. and suggests that this process can occur by more than one pathway. The RNA produced from these constructs should interact with the appropriate proteins to make U2 snRNP, run-on analysis on the transfected U2 templates (Figure including the Sm-binding proteins. It is possible, therefore, 7). The VAI gene was again included as a transfection that longer transcripts, produced in the absence of the U2 control and the probe for 7SK was used to control the 3 box, with or without the addition of the U1 3 box efficiency of the run-on assay. U2 MARK is a 50mer downstream, can be trimmed after complexing with pro- oligonucleotide complementary to the linker in the marked teins in the cytoplasm. However, the pattern of bands is U2 gene plus some flanking sequence (see Materials identical in the absence and presence of the 3 box, and methods) used to detect transcription of the coding suggesting that these ‘precursors’ are generated in the sequence. The probe VEC is an 80mer oligonucleotide same way in both cases. Thus, the slightly 3-extended complementary to the region of the vector downstream transcripts noted above U2M in lane 1 (WT construct) from the U2 3 box which detects readthrough transcription may not represent specific precursors formed by a 3 box- (Figure 7A). Co is an 80mer oligonucleotide complement- dependent mechanism. Alternatively, other sequences ary to a region of the vector upstream from the U2 within the gene may be able to direct a low level of proper promoter and serves as a background control. After 3 end formation in the absence of the 3 box. carrying out the run-on reaction with nuclei from trans- Since the transcripts in this analysis are variable in fected cells, RNA was hybrid-selected using a probe sequence, any differences noted may simply reflect complementary to part of the coding region of the marked changes in RNA stability. However, the drop in detectable U2 gene (see Materials and methods). The selected RNA readthrough products from the U1DS construct is was then hydrolysed and hybridized to nitrocellulose filters consistent with the possibility that the downstream in vivo- (Figure 7B). This procedure ensures that signals are due footprinted region of the U1 gene is capable of terminating to RNA contiguous with the coding region and greatly transcription. reduces the level of background, allowing an accurate quantitation of the low signals over the short U2 MARK The region downstream from the 3 box of the U1 contains a transcription terminator probe. In order to determine unequivocally whether the 3 box RNA from cells transfected with the WT construct is causing termination directly, we have carried out hybridizes to both the U2 MARK and VEC probes, 2873 P.Cuello et al. indicating that transcription is continuing beyond the Termination may be effected directly by proteins binding 3 box in the transfected gene as it does in the endogenous to the sequence within the in vivo-footprinted region, U2 genes. Importantly, no signal is detected over any U2 perhaps in conjunction with a high proportion of T template probe when the PSE is mutated (Figure 7B, P–). residues. Interestingly, the results of our nuclear run-on Neither VAI nor 7SK transcripts have been selected by analysis indicate that the human U2 genes do not have an the probe and remain in the supernatant after the selection efficient termination element within the 250 bp down- procedure, indicating that the selection is specific. Deletion stream of the 3 box. Our results also suggest that of the U2 3 box has little effect on the level of readthrough termination is coupled to recognition of the 3 box. (Δ3 BOX, 89% of the level seen with WT). Introduction of the 3 box of the U1 gene similarly has little effect Why does the U1 gene have a terminator close to (U1, 91.2%). However, the presence of both the U1 the 3 box? 3 box and the immediate downstream region significantly It is not clear why termination of transcription of the U1 reduces the level of readthrough (U1DS, 24.7%). The genes should occur very close to the 3 box while addition of the U1 downstream region alone appears to transcription of the U2 genes continues further down- have a less pronounced effect on the level of readthrough stream. The U1 gene repeat unit is 45 kb long and (DS, 60.5%). The levels of VAI transcripts do not appear repeated 15–30 times on chromosome 1 (Lindgren et al., to vary drastically relative to the coding region transcripts 1985b). This is an extraordinarily large repeat unit con- of the different transfected genes, suggesting that changes sidering that the size of the RNA-coding region of the U1 in the levels of steady-state RNA (Figure 6) are due gene is 200 bp. Unlike the U2 repeat units which are mainly to post-transcriptional effects. arranged tandemly, the U1 genes are loosely clustered and These results lend further support to the hypothesis organized in an irregular and highly polymorphic manner that the in vivo-footprinted element located immediately (Manser and Gesteland, 1982; Lund and Dahlberg, 1984; downstream of the 3 box of the U1 genes is a transcrip- Bernstein et al., 1985). In addition, there are multiple tional terminator. This would satisfactorily explain the genes present in the RNU1 locus. These genes include difference in the run-on profiles of the human U1 and U2 tRNA genes (Van der Drift et al., 1994, 1995) and several genes. The U1 and U2 3 boxes alone appear to have putative neuroblastoma suppressor genes including human little effect on the level of readthrough transcription, Kruppel-related 3 (HKR3) (Maris et al., 1997) and human indicating that these are not efficient termination signals Elk-related kinase (ERK) (Saito et al., 1995). Since not by themselves. However, the putative U1 terminator also all the genes in this region have been identified, or their does not appear to function efficiently in the absence of precise locations determined, it is unclear whether any of the 3 box, suggesting that both elements are necessary these genes lie close to the U1 genes. Perhaps termination for termination of transcription. of transcription of the U1 genes is more critical than for the U2 genes, in order to prevent polymerases reading into downstream genes and causing inhibition of initiation Discussion by transcriptional interference [see Greger et al. (1998) Previous studies on snRNA gene transcription have not for a discussion of this phenomenon]. resolved whether the 3 box functions in the nascent RNA as a processing element or as a DNA element to terminate How might initiation, 3 box function and transcription. The 3 box is necessary for efficient 3 end termination be linked? formation of snRNA gene transcripts (Hernandez, 1985; It is possible that only a fraction of the nascent transcripts Yuo et al., 1985; Ciliberto et al., 1986; Neuman de Vegvar from the human U2 genes are terminated at the 3 box et al., 1986; Ach and Weiner, 1987; reviewed in Hernandez, and processed properly. However, this seems unlikely as 1992), and the function of this element cannot be separated these genes encode very abundant RNAs. In common readily from the process of transcription. Notably, tran- with Yuo et al. (1985) and Neuman de Vegvar et al. scripts containing a copy of the 3 box do not undergo (1986), we favour the hypothesis that the 3 box is an processing when reinjected into Xenopus oocytes RNA processing element present in the nascent RNA and (Ciliberto et al., 1986) and although U1 transcripts that that the 3–16 nucleotide extended precursors detected include the 3 box are detected in HeLa cells, these do in vivo are produced directly by an RNA processing event not seem to be processing substrates (Lobo and Marzluff, mediated by the 3 box. Yuo et al. (1985) have speculated 1987). In addition, recognition of the U1 3 box in vitro further that such a processing reaction may involve appears to be co-transcriptional and only occurs if tran- snRNAs in analogy to 3 end processing of histone scription is initiated from a class II snRNA promoter mRNAs (see below). (Gunderson et al., 1990). Although these findings do not The 3 ends of all other pol II transcripts (mRNAs) rule out that the 3 box is a processing signal, they support are produced by cleavage of nascent RNA. The replication- the notion that this element is a transcription terminator. activated histone mRNAs are produced from precursors The demonstration by Kunkel and Pederson (1985) that by a cleavage event that requires base pairing between transcription terminates very close to the 3 box in the the nascent RNA and the U7 snRNA (reviewed by U1 genes provided the most compelling argument to date Marzluff, 1992). In the case of the remaining mRNAs, that the 3 box is a termination element. However, the cleavage of the nascent RNA and subsequent results presented here show that the 3 box is not sufficient polyadenylation of the resulting 3 end are directed by for efficient termination of transcription. Instead, we the polyadenylation signal in the transcribed RNA implicate a novel element immediately downstream of the (reviewed by Colgan and Manley, 1997). Processing at 3 box in termination of transcription of the U1 gene. the poly(A) site is also linked intimately to both initiation 2874 Termination of transcription of snRNA genes and termination of transcription. Factors required for polyadenylation signals are not recognized efficiently in polyadenylation associate with the C-terminal domain transcripts initiated from the promoters of the U1 or U2 (CTD) of pol II before the poly(A) signal is transcribed, genes (Neuman de Vegvar et al., 1986; Dahlberg and possibly at initiation (Dantonel et al., 1997), and removal Schenborn, 1988; Hernandez and Lucito, 1988; Lobo of the CTD dramatically decreases the efficiency of and Hernandez, 1989). However, recognition of poly(A) polyadenylation in vivo (McCracken et al., 1997). In signals can be uncoupled readily from transcription, addition, termination of transcription of vertebrate mRNA whereas recognition of the 3 box cannot (Lobo and genes requires that splicing of the terminal intron and Marzluff, 1987; Ciliberto et al., 1989; Gunderson et al., polyadenylation take place, even though termination can 1990), as noted above. This suggests that a promoter- occur hundreds of base pairs downstream from the coding dependent activity is also essential for any processing region (Dye and Proudfoot, 1999, and references therein). event directed by the 3 box. Exactly how termination occurs in these genes is not well Termination of transcription of both mRNA genes and understood, but Birse et al. (1998) have shown recently snRNA genes may also occur by a common mechanism that the function of cleavage factors in the polyadenylation involving cleavage of the transcript/release of processing apparatus is required for termination of transcription of factors from the polymerase. Termination of transcription mRNA genes in Saccharomyces cerevisiae. Processing of of the U2 gene would, then, resemble the process in those the RNA at the poly(A) site is likely to influence directly mRNA genes where termination occurs hundreds of base the association of processing factors with the CTD of the pairs downstream from the poly(A) site. In the U1 genes transcribing polymerase. Thus, transcribing a poly(A) instead, factors bound to the DNA just downstream of the signal may convert the polymerase to a more easily processing element terminate the ‘destabilized’ polymerase terminated form as suggested by McCracken et al. (1997) much sooner. and Dye and Proudfoot (1999). In the case of the closely spaced genes encoding the mammalian complement pro- Materials and methods teins, C2 and factor B, however, it has been shown that binding sites for the protein Maz are required in addition Nuclear run-ons and hybrid selection to a functional poly(A) site for efficient termination to The protocol of Ashe et al. (1997) was used with the following changes: ~50 μl of nuclei were collected from two 120 mm dishes of HeLa cells occur (Ashfield et al., 1994). Thus, a combination of for each reaction and 7 μlof [α- P]UTP (3000 Ci/mmol) was added to processing and binding of factors to elements in the DNA each reaction mix. A 1 μg aliquot of each 47–50mer oligonucleotide or can work together to terminate pol II relatively close to 1.6 μg of each 80mer oligonucleotide was denatured and fixed to the the poly(A) site. nitrocellulose filter after slot-blotting. After hybridization, filters were washed three times in 1.0 SSPE/0.1% (w/v) SDS at 37°C (sometimes Efficient recognition of the 3 box in vivo and in vitro with 1 μg/ml RNase A added) and exposed to Kodak AR film at –70°C requires that transcription is initiated from a class II or to Molecular Dynamics Phosphor storage screens for quantitation in snRNA gene promoter, indicating that the pre-initiation a PhosphorImager. complex formed on these genes influences the downstream For run-on analysis of the endogenous U2 genes, the oligos are RNA processing event(s) (Hernandez and Weiner, 1986; complementary to the U2 gene sequence from pTP18 (Pavelitz et al., 1995; DDBJ/EMBL/GenBank accession No. U57614) taking the first Neuman de Vegvar et al., 1986; Gunderson et al., 1990). base pair of the U2 coding sequence as 1: US, –50 to –1; A, 1–47; B, Thus, the initial 3 end processing event of transcripts 48–94; C, 95–141; D, 142–188; E, 189–238; F, 239–288; G, 289–338; from both mRNA and snRNA genes appears to be influ- H, 339–388; I, 399–438; J, 439–489; Z, 789–838. AS is complementary enced directly by events at initiation. The mechanism of to A. The 7SK oligonucloeotide is complementary to sequences 61–110 of the coding region for 7SK (Murphy et al., 1986). co-ordinate promoter/3 box-directed processing of snRNA For run-on analysis of the U1 genes, the oligos A, B and C are gene transcripts may be analogous to the mechanism of complementary to nucleotides 6–55, 101–150 and 171–250, respectively, polymerase-assisted 3 end cleavage and polyadenylation of HSD2 (Manser and Gesteland, 1982) taking the first base of the U1 of mRNA genes. As mentioned above, the structure of coding region as 1. Oligo D is complementary to nucleotides 171–250 the promoters of class II snRNA genes is different from of HU1-1 (Lund and Dahlberg, 1984; DDBJ/EMBL/GenBank accession No. J00318). that of mRNA genes. The essential PSE element is For run-on analysis of transfected genes, the nuclei were collected both necessary and sufficient for initiation and 3 box from two 120 mm dishes of HeLa cells 48 h after transfection with recognition in vivo (Hernandez and Lucito, 1988; Neuman 12.5 μg of the test construct and 1.25 μg of VAI using Lipofectamine de Vegvar and Dahlberg, 1989; Parry et al., 1989). The PSE (Gibco-BRL) according to the manufacturer’s instructions. U2 MARK is recognized by a multisubunit factor known variously as is 5-GAT- ACTACACTTGATCCTCTAGAGCTGGTACCACTAGAGG- ATCTTAGCCA-3. VEC and Co are complementary to bp 61–140 and PTF, SNAPc and PBP (Henry et al., 1998, and references 2792 to 2871 of pGEM 4, respectively, taking the first base pair of therein), which is critical for the formation of a specific the SP6-transcribed region as 1. The VAI oligonucleotide is complement- pre-initiation complex containing TBP and TFIIB (Bernues ary to bp 1–80 of the VAI-coding region. Synthetic RNAs were made et al., 1993; Sadowski et al., 1993). This specialized by T7 or SP6 from sequences cloned into pGEM vectors. Hybrid selection probes were made in the same way with the inclusion of biotin- complex may directly recruit a specialized polymerase 16-UTP (Boehringer Mannheim) as described by Dye and Proudfoot associated with processing factors. Alternatively, the poly- (1999). Hybrid selection was carried out as described by Dye and merase may become associated with specific processing Proudfoot (1999). factors after recruitment, as appears to happen in the interaction of the CTD of pol II with splicing and Constructs The marked U2 gene is described by Murphy (1997) and extends from polyadenylation factors. An obvious scenario is one where the StuI site at –556 of dl-556 (Ares et al., 1985) to the NaeI site at snRNA processing factors rather than mRNA processing 93 downstream of the end of the RNA coding region. This was cloned factors interact with pol II at initiation of transcription of into pGEM4 using the EcoRI and HincII sites in the polylinker. P– was snRNA genes, as already suggested by Neuman de Vegvar made by PCR to replace the sequence between –60 and –49 by et al. (1986). This would account for the finding that TCCCACGGCCGT. The Δ3 BOX construct was made by PCR to 2875 P.Cuello et al. replace the 3 box from 19 to 41 with a C residue to create a unique Bernues,J., Simmen,K.A., Lewis,J.D., Gunderson,S.I., Moncollin,M., MluI site. The U1 sequence from 138 to 187 and the sequence from 192 Egly,J.-M. and Mattaj,I.W. (1993) Common and unique transcription to 250 of pHU1-1 (Lund and Dahlberg, 1984) downstream of the start factor requirements of human U1 and U6 snRNA genes. EMBO J., of transcription were cloned into this site to create U1 and DS, 12, 3575–3585. respectively. For U1DS, MluI–NaeI was replaced with sequences Birse,C.E., Minvielle-Sebastia,L., Lee,B.A., Keller,W. and Proudfoot,N.J. from 138 to 250 of HU1-1. The pGEM 4-based template for the hybrid (1998) Coupling termination of transcription to messenger RNA selection probe for endogenous U2 (Figure 2) was made by PCR from maturation in yeast. Science, 280, 298–301. pTP18 (Pavelitz et al., 1995) using oligos AS and D. For hybrid selection Chanfreau,G., Abou Elela,S., Ares,M. and Guthrie,C. (1997) Alternative of marked U2 transcripts, a fragment containing the sequence from the 3-end processing of U5 snRNA by RNase III. Genes Dev., 11, DdeI site at bp 20 of the coding region to the end of oligonucleotide D 2741–2751. was prepared by PCR of the marked U2 template and cloned into Ciliberto,G., Dathan,N., Frank,R., Philipson,L. and Mattaj,I.W. (1986) pGEM3. Each U2 construct was recloned into the EcoRI site of pGEM4 Formation of the 3 end on U snRNAs requires at least three sequence to place the polylinker region between the T7 promoter and the 3 end elements. EMBO J., 5, 2931–2937. of the U2 sequence. Colgan,D.F. and Manley,J.L. (1997) Mechanism and regulation of mRNA polyadenylation. Genes Dev., 11, 2755–2766. Steady-state RNA analysis Corden,J.L. and Patturajan,M. (1997) A CTD function linking For each 90 cm dish of HeLa cells, a 5 μg aliquot of each U2 construct transcription to splicing. Trends Biochem. Sci., 22, 413–416. was co-transfected with 500 ng of VAI using Lipofectamine (Gibco- Dahlberg,J.E. and Schenborn,E.T. (1988) The human U1 snRNA BRL) as recommended by the manufacturers, and cytoplasmic RNA promoter and enhancer do not direct synthesis of messenger RNA. was collected as described by Whitelaw et al. (1989). RNase protection Nucleic Acids Res., 16, 5827–5840. was carried out as described by Eggemont and Proudfoot (1993). Dantonel,J.C., Murthy,K.G.K., Manley,J.L. and Tora,L. (1997) Riboprobes for analysis of U2 transcripts were made by T7 after Transcription factor TFIID recruits factor CPSF for formation of the digestion with FokI that cuts 42 bp upstream from the start of the U2 3 end of mRNA. Nature, 389, 399–402. coding region. The VAI riboprobe was made by SP6 after digesting the Dye,M.J. and Proudfoot,N.J. (1999) Terminal exon definition occurs co- VAI gene in pGEM4 with BamHI. S1 analysis was carried out as transcriptionally and promotes termination of RNA polymerase II. described by Murphy (1997) using the oligonucleotides described therein. Mol. Cell, 3, 371–378. The products of S1 analysis are 45 and 63 nucleotides for the VAI and Eggermont,J. and Proudfoot,N.J. (1993) Poly(A) signals and U2 RNAs, respectively. transcriptional pause sites combine to prevent interference between RNA polymerase promoters. EMBO J., 12, 2539–2548. In vivo footprinting Elela,S.A. and Ares,J.M. (1998) Depletion of yeast RNase III blocks Cells or purified naked DNA were treated first with DMS (Aldrich) and correct U2 3 end formation and results in polyadenylated but then with piperidine (Aldrich) to prepare in vivo or in vitro DNA functional U2 RNA. EMBO J., 17, 3738–3746. samples, respectively, as described by Greger et al. (1998). Ligation- Greger,I.H., Demarchi,F., Giacca,M. and Proudfoot,N.J. (1998) mediated PCR was carried out on these samples using the following Transcriptional interference perturbs the binding of Sp1 to the HIV-1 nested primers for U1: U1.DS1, 5-CTCGCTTTTTCTCCTATGGC-3; promoter. Nucleic Acids Res., 26, 1294–1300. U1.DS2, 5-CAGGCGACATGTTACTTCCTATTCCGC-3; U1.DS3, 5- Gunderson,S.I., Knuth,M.W. and Burgess,R.R. (1990) The human U1 GGCGACATGTTACTTCCTATTCCGCAGCCCTC-3; and for U2: snRNA promoter correctly initiates transcription in vitro and is U2.US1, 5-GGATTTTTGGAGCAGGGAGA-3; U2.US2, 5-AATA- activated by PSE1. Genes Dev., 4, 2048–2060. GGAGCTTGCTCCGTCCACTCC-3; U2.US3, 5-AGGAGCTTGCTC- Hammarstrom,K., Westin,G., Bark,C., Zabielski,J. and Pettersson,U. CGTCCACTCCACGCATC-3. (1984) Genes and pseudogenes for human U2 RNA. Implications for the mechanism of pseudogene formation. J. Mol. Biol., 179, 157–169. Henry,R.W., Mittal.,V., Ma.,B., Kobayashi,R. and Hernandez,N. (1998) Acknowledgements SNAP19 mediates the assembly of a functional core promoter complex We would like to thank Andre Furger and David Bentley for helpful (SNAPc) shared by RNA polymerases II and III. Genes Dev., 12, suggestions, and Andre Furger and Alexandra Moreira for critical reading 2664–2672. of the manuscript. We would especially like to thank Charlie Birse for Hernandez,N. (1985) Formation of the 3 end of U1 snRNA is directed encouraging us to try using oligonucleotide probes for the run-ons and by a conserved sequence located downstream of the coding region. Ingo Greger for setting up in vivo footprinting in the laboratory. P.C. EMBO J., 4, 1827–1837. was supported by a BBSRC graduate studentship, M.D. was supported Hernandez,N. (1992) Transcription of vertebrate snRNA genes and by a Glaxo graduate studentship, N.J.P. is supported by Wellcome related genes. In McKnight,S.L. and Yamamoto,K.R. (eds), Programme Grant No. 032743, and D.C.B. and S.M. are supported by Transcriptional Regulation. Cold Spring Harbor Laboratory Press, MRC Senior Fellowship No. G117/309. Cold Spring Harbor, NY, Vol. 1, pp. 281–313. Hernandez,N. and Lucito,R. (1988) Elements required for transcription initiation of the human U2 snRNA coincide with elements required References for snRNA 3 end formation. EMBO J., 7, 3125–3134. Hernandez,N. and Weiner,A.M. (1986) Formation of the 3 end of U1 Ach,R.A. and Weiner,A.M. (1987) The highly conserved U small nuclear snRNA requires compatible snRNA promoter elements. Cell, 47, RNA 3-end formation signal is quite tolerant to mutation. Mol. Cell. 249–258. Biol., 7, 2070–2079. Huang,Q., Jacobson,M.R. and Pederson,T. (1997) 3 Processing of Ares,M.,Jr, Mangin,M. and Weiner,A.M. (1985) Orientation-dependent human pre-U2 small nuclear RNA: a base-pairing interaction between transcriptional activator upstream of a human U2 snRNA gene. Mol. the 3 extension of the precursor and an internal region. Mol. Cell. Cell. Biol., 5, 1560–1570. Biol., 17, 7178–7185. Ashe,H.L., Monks,J., Wijgerde,M., Fraser,P. and Proudfoot,N.J. (1997) Kuhn,A. and Grummt,I. (1989) 3-End formation of mouse pre-rRNA Intergenic transcripts and transinduction of the human β-globin locus. involves both transcription termination and a specific processing Genes Dev., 11, 2494–2509. reaction. Genes Dev., 3, 224–231. Ashfield,R., Patel,A.J., Bossone,S.A., Brown,H., Campbell,R.D., Kunkel,G.R. and Pederson,T. (1985) Transcription boundaries of U1 Marcu,K.B. and Proudfoot,N.J. (1994) MAZ-dependent termination small nuclear RNA. Mol. Cell. Biol., 5, 2332–2340. between closely spaced human complement genes. EMBO J., 13, Liao,D. and Weiner,A.M. (1995) Concerted evolution of the tandemly 5656–5667. repeated genes encoding primate U2 small nuclear RNA (the RNU2 Bentley,D.L. and Groudine,M. (1986) A block to elongation is largely locus) does not prevent rapid diversification of the (CT)n·(GA)n responsible for decreased transcription of c-myc in differentiated HL60 microsatellite embedded within the U2 repeat unit. Genomics, 30, cells. Nature, 321, 702–706. 583–593. Bernstein,L.B., Manser,T. and Weiner,A.M. (1985) Human U1 small Lindgren,V., Ares,M., Weiner,A.M. and Franke,U. (1985a) Human genes nuclear RNA genes: extensive conservation of flanking sequences for U2 small nuclear RNA map to a major adenovirus 12 modification suggests cycles of gene amplification and transportation. Mol. Cell. Biol., 5, 2159–2171. site on chromosome 17. Nature, 314, 115–116. 2876 Termination of transcription of snRNA genes Lindgren,V., Bernstein,L.B., Weiner,A.M. and Franke,U. (1985b) Human Van Arsdell,S.W. and Weiner,A.M. (1984) Human genes for U2 small U1 small nuclear RNA pseudogenes do not map to the site of the U1 nuclear RNA are tandemly repeated. Mol. Cell. Biol., 4, 492–499. genes in 1p36 but are clustered in 1q12–q22. Mol. Cell. Biol., 5, Van der Drift,P., Chan,A., van Roy,N., Laureys,G., Westerveld,A., 2172–2180. Speleman,F. and Versteeg,R. (1994) A multimegabase cluster of Lobo,S.M. and Marzluff,W.F. (1987) Synthesis of U1 RNA in isolated snRNA and tRNA genes on chromosome 1p36 harbours an adenovirus/ mouse cell nuclei: initiation and 3 end formation. Mol. Cell. Biol., SV40 hybrid virus integration site. Hum. Mol. Genet., 3, 2131–2136. 7, 4290–4296. Van der Drift,P., Chan,A., Laurets,G., van Roy,N., Sickmann,G., den Dunnen,J., Westerveld,A., Speleman,F. and Versteeg,R. (1995) Lobo,S.M. and Hernandez,N. (1989)A7bp mutation converts a human Balanced translocation in a neuroblastoma patient disrupts a cluster RNA polymerase II snRNA promoter into an RNA polymerase III of small nuclear RNA U1 and tRNA genes in chromosomal band promoter. Cell, 58, 55–67. 1p36. Genes Chromosomes Cancer, 14, 35–42. Lobo,S.M. and Hernandez,N. (1994) Transcription of snRNA genes by Weber,J., Jelinek,W. and Darnell,J.E.,Jr (1977) The definition of a large RNA polymerase II and III. In Conaway,R.C. and Conaway,J.W. (eds), viral transcription unit late in AD2 infection of HeLa cells: mapping Transcription, Mechanisms and Regulation. Raven Press, New York, of nascent RNA molecules labelled in isolated nuclei. Cell, 10, NY, Vol. 3, pp. 127–159. 611–616. Lund,E. and Dahlberg,J.E. (1984) True genes for human U1 small Westin,G., Zabielski,J., Hammarstrom,K., Monstein,H.J., Bark,C. and nuclear RNA: copy number, polymorphism and methylation. J. Biol. Pettersson,U. (1984) Clustered genes for human U2 RNA. Proc. Natl Chem., 259, 2013–2021. Acad. Sci., USA, 81, 3811–3815. Manser,T. and Gesteland,R.F. (1982) Human U1 loci: genes for human Whitelaw,E., Hogben,P., Hanscombe,O. and Proudfoot,N.J. (1989) U1 RNA have dramatically similar genomic environments. Cell, 29, Transcriptional promiscuity of the human α-globin gene. Mol. Cell. 257–264. Biol., 9, 241–251. Maris,J.M., Jensen,J., Sulman,E.P., Beltinger,C.P., Allen,C., Biegel,J.A., Wieben,E.D., Nenninger,J.M. and Pederson,T. (1985) Ribonucleoprotein Brodeur,G.M. and White,P.S. (1997) Human Kruppel-related 3 organisation of eukaryotic RNA. XXXII. U2 small nuclear RNA (HKR3): a candidate for the 1p36 neuroblastoma tumour suppressor precursors and their accurate 3 processing in vitro as ribonucleoprotein gene? Eur. J. Cancer, 33, 1991–1996. particles. J. Mol. Biol., 183, 69–78. Marzluff,W.F. (1992) Histone 3 ends: essential and regulatory functions. Yuo,C., Ares,M. and Weiner,A.M. (1985) Sequences required for 3 end Gene Expression, 2, 93–97. formation of human U2 small nuclear RNA. Cell, 42, 193–202. McCracken,S.N., Fong,N., Yankulov,K., Ballantyne,S., Pan,G.H., Greenblatt,J., Patterson,S.D., Wickens,M. and Bentley,D. (1997) The Received February 10, 1999; revised and accepted March 29, 1999 C-terminal domain of RNA polymerase II couples messenger RNA processing to transcription. Nature, 385, 357–361. Murphy,S. (1997) Differential in vivo activation of the class II and class III snRNA genes by the POU-specific domain of Oct-1. Nucleic Acids Res., 25, 2068–2076. Murphy,S., Tripodi,M. and Melli,M. (1986) A sequence upstream from the coding region is required for transcription of the 7SK RNA genes. Nucleic Acids Res., 14, 9243–9260. Nagai,K. and Mattaj,I.W. (1994) RNA–protein interactions in the splicing snRNPs. In Nagai,K. and Mattaj,I.W. (eds), Frontiers in Molecular Biology. IRL Press at Oxford University Press, Oxford, UK, pp. 150–177. Neugebauer,K.M. and Roth,M.B. (1997) Transcription units as RNA processing units. Genes Dev., 11, 3279–3285. Neuman de Vegvar,H.E., Lund,E. and Dahlberg,J.E. (1986) 3 End formation of U1 snRNA precursors is coupled to transcription from snRNA promoters. Cell, 47, 259–266. Neuman de Vegvar,H.E. and Dahlberg,J.E. (1989) Initiation and termination of human U1 transcription requires the concerted action of multiple flanking sequence elements. Nucleic Acids Res., 17, 9305–9318. Parry,H.D., Tebb,G. and Mattaj,I.W. (1989) The Xenopus U2 gene PSE is a single, compact, element required for transcription initiation and 3 end formation. Nucleic Acids Res., 17, 3633–3644. Pavelitz,T., Rusche,L., Matera,A.G., Scharf,J.M. and Weiner,A.M. (1995) Concerted evolution of the tandem array encoding primate U2 snRNA occurs in situ, without changing the cytological context of the RNU2 locus. EMBO J., 14, 169–177. Sadowski,C.L., Henry,R.W., Lobo,S.M. and Hernandez,N. (1993) Targeting TBP to a non-TATA box cis-regulatory element: a TBP- containing complex activates transcription from snRNA promoters through the PSE. Genes Dev., 7, 1535–1548. Saito,T., Seki,N., Matsuda,Y., Kitahara,M., Murata,M., Kanda,N., Nomura,N., Yamamoto,T. and Hori,T.A. (1995) Identification of the human ERK gene as a putative receptor tyrosine kinase and its chromosomal localization to 1p36.1: a comparative mapping of human, mouse and rat chromosomes. Genomics, 26, 382–384. Seipelt,R.L., Zheng,B., Asuru,A. and Rymond,B.C. (1999) U1 snRNA is cleaved by RNase III and processed through an Sm site-dependent pathway. Nucleic Acids Res., 27, 587–595. Steinmetz,E.J. (1997) Pre-mRNA processing and the CTD of RNA polymerase II: the tail that wags the dog? Cell, 89, 491–494. Strobl,L.J. and Eick,D. (1992) Hold back of RNA polymerase II at the transcription start site mediates down-regulation of c-myc in vivo. EMBO J., 11, 3307–3314. Terns,M.P., Dahlberg,J.E. and Lund,E. (1993) Multiple cis-acting signals for export of pre-U1 snRNA from the nucleus. Genes Dev., 7, 1898–1908. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

Transcription of the human U2 snRNA genes continues beyond the 3′ box in vivo

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/transcription-of-the-human-u2-snrna-genes-continues-beyond-the-3-box-SmgtpmvgMB

References (82)

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 1999
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1093/emboj/18.10.2867
Publisher site
See Article on Publisher Site

Abstract

The EMBO Journal Vol.18 No.10 pp.2867–2877, 1999 Transcription of the human U2 snRNA genes continues beyond the 3 box in vivo Vegvar et al., 1986; Ach and Weiner, 1987; reviewed in Paula Cuello, Diana C.Boyd, Michael J.Dye, Hernandez, 1992). Mutation of this sequence causes the Nick J.Proudfoot and Shona Murphy accumulation of transcripts that are not processed accur- Chemical Pathology Unit, Sir William Dunn School of Pathology, ately and are therefore longer than the mature snRNA. University of Oxford, South Parks Road, Oxford OX1 3RE, UK Interestingly, the 3 box functions efficiently only in the Corresponding author context of a pol II-dependent snRNA promoter. Although e-mail: [email protected] these genes are transcribed by the same polymerase as mRNA genes, their promoters are organized differently. The 3 box of the human class II snRNA genes is Class II snRNA genes have characteristic short, compact, required for proper 3 processing of transcripts, but TATA-less promoters containing a distal sequence element how it functions is unclear. Several lines of evidence (DSE) and a proximal sequence element (PSE) (reviewed suggest that termination of transcription occurs at the by Hernandez, 1992; Lobo and Hernandez, 1994). Replace- 3 box and the terminated transcript is then a substrate ment of a U1 or U2 promoter with an mRNA promoter for processing. However, using nuclear run-on analysis inhibits accurate 3 end formation of the transcripts of endogenous genes, we demonstrate that transcription (Hernandez and Weiner, 1986; Neuman de Vegvar et al., continues for at least 250 nucleotides beyond the 1986). 3 box of the U2 genes. Although in vivo footprinting How the 3 box functions is unknown, but several analysis of both the U1 and U2 genes detects no findings suggest that it may act as a transcription termin- protein–DNA contacts directly over the 3 box, a series ator: transcription of the human U1 genes terminates very of G residues immediately downstream from the close to the 3 box (Kunkel and Pederson, 1985); only 3 box of the U1 gene are clearly protected from very small amounts of long snRNA gene transcripts methylation by dimethylsulfate. In conjunction with containing a copy of the 3 box have been detected the 3 box of the U1 gene, this in vivo footprinted (Hernandez, 1985; Kunkel and Pederson, 1985; Ach and region causes termination of transcription of transi- Weiner, 1987; Lobo and Marzluff, 1987) and these do not ently transfected U2 constructs, whereas a 3 box alone appear to be precursors for processed RNA either in HeLa does not. Taken together, these results indicate that the cells (Lobo and Marzluff, 1987) or in Xenopus oocytes 3 box is not an efficient transcriptional terminator (Ciliberto et al., 1986). Thus, an attractive hypothesis is but may act as a processing element that is functional that transcription terminates just upstream of the 3 box in the nascent RNA. and this produces a transcript that is a substrate for further Keywords: pol II/RNA processing/snRNA/termination/ processing, in analogy to the mechanism of 3 end transcription formation of vertebrate pre-rRNA (Kuhn and Grummt, 1989). However, a more direct role for this element in processing of snRNAs has not been ruled out and, in yeast, the 3 ends of U1, U2 and U5 RNAs are produced Introduction by nuclease cleavage of longer precursors (Chanfreau et al., 1997; Elela and Ares, 1998; Seipelt et al., 1999). The vertebrate class II snRNA genes encode short non- In order to understand the exact role of the 3 box, we polyadenylated nuclear RNAs, including the U1, U2, have undertaken a study of transcriptional termination of U4 and U5 snRNAs that are required for pre-mRNA the U2 snRNA genes using nuclear run-on analysis. We processing. These genes are transcribed by RNA poly- find that, in contrast to the U1 gene, transcription of the merase II (pol II) to give 3-extended precursors that U2 gene continues for at least 250 bp after the 3 box. undergo processing themselves to produce the mature Transcripts derived from the region downstream of the 3 RNA components of snRNPs. Precursors with up to 16 box are contiguous with RNA from upstream and are nucleotide 3 end extensions can be detected, and these detected at very early points in a run-on time course, undergo cap trimethylation, internal base modification and strongly suggesting that the profile we see reflects the 3 end trimming after association with U snRNP proteins in vivo polymerase loading. in the cytoplasm (recently reviewed by Huang et al., A complementary in vivo footprinting analysis of the 1997). The processed RNA and associated proteins are 3 box region of the U1 and U2 genes also gave surprising then transported back into the nucleus where assembly of results. No protection is detected over the 3 box of either the functional snRNP is completed (reviewed in Terns gene, whereas a clear area of protection from methylation et al., 1993; Nagai and Mattaj, 1994). 3 End formation by dimethylsulfate (DMS) is apparent immediately down- of mature snRNAs requires a cis-acting sequence, known as the 3 box, located 9–19 nucleotides downstream from stream from the 3 box of the U1 gene. the coding region of these snRNA genes (Hernandez, These results suggest that termination of transcription 1985; Yuo et al., 1985; Ciliberto et al., 1986; Neuman de may be occurring close to the 3 box in the U1 genes due © European Molecular Biology Organization 2867 P.Cuello et al. to protein(s) binding to this DNA element. We provide direct evidence for this by showing that this region terminates transcription from a U2 promoter when placed downstream of a 3 box. In contrast, a 3 box alone does not function as an efficient terminator. Taken together, these results indicate that most nascent snRNA gene transcripts contain a full copy of the 3 box and transcription terminates beyond this element. We consider it likely, therefore, that the 3 box is an RNA element that is required, in conjunction with a promoter- dependent activity, for rapid cleavage to produce the 3 extended RNAs that are detectable in vivo. Results Transcription continues beyond the 3 box of the U2 genes A precise determination of the site of termination of transcription of the snRNA genes is crucial to an under- standing of the role of the 3 box in this process. We have therefore undertaken a detailed study of the polymerase loading on the human U2 genes using 47–50mer oligo- nucleotide probes in a nuclear run-on analysis. There are ~10 U2 genes in the human haploid genome at 17q21–22 (Hammarstrom et al., 1984; Lindgren et al., 1985a; Pavelitz et al., 1995). These are tandemly repeated and the sequence of each 5.8–6.1 kb repeat unit is highly Fig. 1. Transcription continues beyond the 3 box of the U2 genes. conserved (Van Arsdell and Weiner, 1984; Westin et al., (A) A diagram of the structure of the U2 gene is shown with the 1984; Pavelitz et al., 1995), allowing us to analyse the relative positions of the probes marked below. The numbers noted next to the probes indicate the start or end of the probes used relative to the distribution of polymerases over all of these genes using site of initiation (US and D) or relative to the end of the 3 box (E, J one set of probes. The position of each probe relative to and Z). The results of a run-on analysis and hybridization of the U2 gene and the results of a representative run-on synthetically produced RNA to the probes are shown below each analysis are shown in Figure 1A. A 47mer oligonucleotide probe. (B) A graphic representation is shown of the results of the run-on analysis in (A) as a percentage of the signal over probe C after complementary to probe A was included on the filters as subtraction of α-amanitin-resistant transcription and correction for the a negative control (AS) and a 50mer oligonucleotide hybridization efficiency of each probe. complementary to the pol III-transcribed 7SK RNA (7SK) was included as a positive control. The hybridization efficiency of the probes US to Z was determined using a stretch which is not as highly conserved between labelled synthetic RNA (see Materials and methods), and individuals as the other sequences in the U2 genes (Liao the nuclear run-on results were normalized to these values and Weiner, 1995). These results suggest that a large after subtraction of α-amanitin-resistant transcription (see proportion of transcription termination occurs downstream Figure 3). of the 3 box in the U2 genes, perhaps mainly in the No signal was detected over probe B in either the run- region between J and Z. on or the synthetic RNA control, possibly due to the relatively high A/T content of this sequence, precluding Transcripts downstream of the 3 box are analysis of this region. Probe B was included in subsequent contiguous with the transcripts from the coding run-on assays as an additional background control. The region and are detected early in a run-on time profile of polymerase density over the U2 genes determined course by several such experiments is shown in Figure 1B as a Although the pattern of run-on transcripts over the U2 percentage of the level of signal over probe C. gene suggests that a significant number of polymerases As expected, the polymerase density is low upstream continue transcription beyond the 3 box, we cannot of the site of initiation of the U2 genes (probe US, 2.65%), exclude the possibility that transcription has reinitiated indicating that transcription from upstream genes does not downstream of the coding region. We have therefore used continue into these promoters. However, the pattern of an RNA probe complementary to region A–D of the U2 polymerase density over the U2 genes is strikingly different gene to hybrid-select transcripts contiguous with the from that seen on the U1 genes where transcription was coding RNA. A biotin-labelled selection probe was hybrid- found to terminate close to the 3 box (Kunkel and ized to the run-on RNA, and avidin-coupled magnetic Pederson, 1985). Instead, a high level of polymerase beads were used to separate the probe and associated loading appears to continue until at least 250 bp past the RNA from the mix (see Materials and methods). The 3 box (probe J, 95%). The polymerase density over selected RNA was then hydrolysed and hybridized to region Z, ~600 bp downstream of the 3 box, has dropped filters containing probes US to Z, AS and 7SK (Figure to 18%. We have not scored the polymerase density 2A). Importantly, transcripts from the 7SK gene have not between the regions J and Z since this contains a (CT) been selected, indicating that the selection is specific. 2868 Termination of transcription of snRNA genes Fig. 2. Transcripts downstream of the U2 3 box are contiguous with the transcripts from the coding region. (A) The results of hybridization of RNA that has been hybrid-selected after the run-on are shown below the profile from an unselected run-on (a lower exposure of the same experiment shown in Figure 1A). (B) A graphic representation of the results from an unselected run-on analysis (black bars) is shown compared with the results with RNA hybrid-selected after the run-on (cross-hatched bars), corrected as in Figure 1. Fig. 3. Transcripts downstream of the U2 3 box are detected early in Figure 2B shows the pattern of hybridization from this a run-on time course. (A) The results of adding α-amanitin (2 μg/ml) analysis (cross-hatched bars) compared with that seen to the run-on reaction at different times are shown. The time (in with an unselected run-on analysis performed at the same minutes) is shown at the left of the figure. The result of hybridization to the 7SK probe is shown for each time point at the right. time (black bars). (B) A graphic representation of the results corrected for the 7SK After hybrid selection, the signal over US relative to C signal is shown for the 2 min (black bars), 5 min (cross-hatched bars) has fallen from 2.6% to 1%, while the relative signals and 10 min (empty bars) time points. over probes A–G have remained at approximately the same levels. At least 50% of the transcripts complementary to probes H–I are also selected and are therefore contiguous as an internal control since this should be unaffected by with transcripts from the coding region. Some of the the levels of α-amanitin used. When α-amanitin is added transcripts detected over probes J and Z are also contiguous before the addition of the radioactive nucleotide, a signal with the coding RNA. The drop in signals over probes is detected over probe C, indicating that transcripts pro- H–Z after hybrid selection indicates that some of the duced by either pol I or pol III are hybridizing to this apparent transcription from this region may be the result probe. Accordingly, these background signals have been of cross-hybridization or reinitiation. Alternatively, post- subtracted from the signals over probe C for all other run- transcriptional nuclease digestion (specific or non-specific) on assays. Signals are visible over at least some of the may be responsible. For example, some of the transcripts probes downstream of the 3 box as early as 1 min. detected over probes H–Z in the unselected run-on However, the signals at this time point are too low to be analysis may have been cleaved specifically in a 3 box- quantitated accurately. Therefore, only the results for the dependent reaction. 2, 5 and 10 min time points are depicted graphically in The above results indicate that transcription is not Figure 3B after normalizing to the level of the 7SK signal. terminating efficiently at the 3 box in our run-on assay. The signals over probes A–Z increase from 1 to 10 min, However, transcription potentially can extend 1000 bp indicating that transcription is continuing during this time. in the 15 min of a run-on reaction (Weber et al., 1977). After 2 min, the signals are quite evenly distributed over In addition, polymerases that are normally paused may probes A–J. Although all the signals have increased resume transcription under run-on conditions (see, for ~10-fold from the 2 to 10 min time points (relative to example, Strobl and Eick, 1992). We have therefore carried 7SK), the pattern has not changed markedly, apart from a out a time course of the run-on reaction to investigate the relative increase over probe C. This may suggest that possibility that polymerases are not transcribing the region polymerases are stacking over probe C or that some of downstream of the 3 box in intact cells (Figure 3). this signal is due to cross-hybridization to non-U2 gene α-Amanitin was added to 2 μg/ml, either before or 1, transcripts. Importantly, there is no obvious increase in 2, 5 or 10 min after the addition of nucleotides, to the the signals over the downstream probes relative to probes resuspended nuclei (see Materials and methods) (Figure A–D with time, and the absolute signal over probe A 3A). The probe for the pol III-transcribed 7SK gene serves increases with time. Reinitiation of transcription by pol II is 2869 P.Cuello et al. are not identical in the seven of the 15–30 U1 genes sequenced so far, although they are highly related for up to at least 70 bp past this element (Manser and Gesteland, 1982; Lund and Dahlberg, 1984). We have therefore used a mixture of two longer (80mer) oligonucleotide probes (C and D) complementary to the region from 8 to 88 bp downstream of the coding region of the HSD2 U1 gene (Manser and Gesteland, 1982) and the HU1-1 U1 gene (Lund and Dahlberg, 1984) to increase the possibility of detecting transcripts from this region. Oligos C and D include the complement to the 3 box which spans the sequence from 11 to 24 bp downstream of the coding region. The hybridization efficiency of the probes A–D was determined using labelled synthetic RNAs (Figure 4A) and the nuclear run-on results were normalized to these values after subtraction of α-amanitin-resistant transcrip- tion. The position of each probe relative to the U1 gene and the results of a representative run-on analysis are shown in Figure 4A. The 7SK and AS oligonucleotides have also been included as positive and negative controls, respectively. The profile of polymerase density over the U1 genes determined by three such experiments is shown in Figure 4B as a percentage of the level of signal over Fig. 4. Termination of transcription appears to occur close to the probe A. 3 box of the U1 genes in our assay system. (A) A diagram of the Transcription appears to be much higher over the coding structure of the U1 gene is shown with the relative positions of the region of the U1 genes than directly downstream, where probes marked below. The numbers noted next to the probes indicate it has dropped to 10% of the value for probe A. This the start or end of the probes used relative to the site of initiation (A and B) or relative to the end of the 3 box (C and D). Oligos C suggests that termination of transcription is occuring close and D are complementary to transcripts from the HSD2 and HU1-1 U1 to the 3 box in these genes. The signal detected over alleles, respectively (see text). The results of a run-on analysis and probes C/D may indicate that transcription continues just hybridization of synthetically produced RNA to the probes are shown into this region or that some transcripts corresponding to below each probe. (B) A graphic representation is shown of the results the whole length of the probes are made. Alternatively, of the run-on analysis in (A) as a percentage of the signal over probe A after subtraction of α-amanitin-resistant transcription and the probes we have used may only detect a small proportion correction for the hybridization efficiency of each probe. of the transcripts produced by the U1 genes due to sequence divergence downstream of the 3 box. However, very inefficient in nuclear run-ons (Bentley and Groudine, probes C and D together should detect transcripts from at 1986), and transcripts of mRNA genes are elongated by least the seven sequenced U1 genes (see Manser and ~100 nucleotides in 15 min under our run-on conditions Gesteland, 1982; Lund and Dahlberg, 1984). We therefore (Dye and Proudfoot, 1999). It is therefore likely that think it likely that our results are due to termination of transcription is still running-on over the 47 bp region of transcription of the U1 genes close to the 3 box in our probe A within the 10 min of the time course. assay system and that they confirm the findings of Kunkel These results strongly suggest that polymerases are not and Pederson (1985). artefactually reading through a termination signal under These results are consistent with the notion that the the run-on conditions. 3 box is a termination signal. However, the 3 box of the U2 gene would, therefore, be a very inefficient transcrip- Termination of transcription appears to occur tion termination signal in comparison. close to the 3 box of the U1 genes in our assay system In vivo footprinting detects DNA–protein As mentioned above, Kunkel and Pederson (1985) found interactions downstream of the 3 box of the U1 that transcription of the human U1 genes terminates close gene to the 3 box. Our assay conditions are not identical to The function of the 3 box requires that transcription is theirs and they used longer M13 probes to detect run-on initiated specifically from a class II snRNA promoter transcripts. We obtained essentially identical results to (Hernandez and Weiner, 1986; Neuman de Vegvar et al., those shown above for the U2 genes using the buffer 1986), indicating a strong link between initiation of conditions described by Kunkel and Pederson (1985), transcription and RNA processing, as has recently been although the signals were lower over all probes (data not shown for processing of mRNA (reviewed by Corden shown). We have also carried out an analysis of the and Patturajan, 1997; Neugebauer and Roth, 1997; U1 genes using oligonucleotide probes (Figure 4). To Steinmetz, 1997). investigate the polymerase density across these genes, we In our studies of the DNA–protein interactions required have used two 50mer oligonucleotides complementary to for initiation of transcription of snRNA genes, we have sequences within the coding region (A and B). The carried out an extensive in vivo footprinting analysis of sequences downstream from the 3 box of the U1 genes the human U2 genes (D.C.Boyd, I.Greger and S.Murphy, 2870 Termination of transcription of snRNA genes Fig. 5. In vivo footprinting detects DNA–protein interactions downstream of the 3 box of the U1 gene. (A) A diagram of the structure of the U1 and U2 genes is shown with the relative positions of the in vivo footprinting primers indicated by black arrows. (B) The results of in vivo footprinting over the region of the 3 box of the U1 and U2 genes is shown. The white arrows indicate a reduction in the relative intensity of bands (Gs) in the in vivo treated DNA compared with in vitro modified DNA, and the size of the arrow indicates the magnitude of the reduction. The black arrow indicates a moderate increase in the intensity of a band in the in vivo treated DNA. (C) The sequence of the 3 box region of the HU1-1 U1 gene (Lund and Dahlberg, 1984) and the pTP18 U2 genes (Pavelitz et al., 1995) are shown. The numbers under the U1 sequence indicate the distance from the 3 box. G residues that were protected from methylation by DMS in vivo are indicated in bold. The region downstream of the U1 3 box that is footprinted is indicated by a bracket and a G residue that is hypersensitive after DMS treatment in vivo is marked by a black arrow. unpublished results). Since footprints are easily detected in all of the U1 genes sequenced so far (see Manser and over all known promoter elements of these genes, it is Gesteland, 1982; Lund and Dahlberg, 1984). likely that a large number of the genes are active at any Although these results do not rule out that protein(s) one time and that DNA–protein interactions required for bind to the 3 box and directly mediate termination, they termination of transcription may also be detected. We are consistent with the hypothesis that termination of have therefore analysed the in vivo DNA–protein inter- transcription occurs downstream of this element in both actions over the 3 box region of the U1 and U2 genes the U1 and U2 genes. In addition, they raise the possibility using DMS footprinting. that protein(s) binding just downstream of the 3 box The results of this analysis are shown in Figure 5A mediate termination of transcription of the U1 gene. and B and summarized graphically in Figure 5C. For the U1 gene, the upper strand of HU1-1 (Lund and Dahlberg, The 3 box is required for high levels of correctly 1984) has been analysed and for the U2 gene the lower processed U2 RNA strand has been analysed (see Figure 5A). No methylation To test the function of the 3 box and downstream protection is detected over the 3 box of either the U1 or sequences in processing and termination, we constructed U2 genes, although a G residue just upstream from the a series of U2 genes, marked by the addition of a linker U2 3 box was partially protected from methylation to the coding region, where the 3 box is either present (marked in Figure 5B and C). However, a region of very or absent or has been replaced by U1 sequences (Figure clear protection from methylation is apparent immediately 6A). These constructs were transfected into HeLa cells, downstream from the 3 box of the U1 gene, close to a and the 5 and 3 ends of the transcribed RNAs mapped T-rich stretch of sequence (Figure 5B and C), suggesting by S1 and RNase protection analysis (Figure 6B and C). that a protein is binding here in vivo. Importantly, the The gene for the pol III-dependent VAI RNA was included central 11 nucleotides of the footprint are highly conserved as a co-transfection control. The probes used for these 2871 P.Cuello et al. Fig. 6. The 3 box is required for high levels of correctly processed U2 RNA. (A) The structure of marked U2 genes with changes to the 3 box region is represented. WT is identical to the endogenous U2 gene apart from the insertion of a linker in the coding region. P– is transcriptionally inactive due to a mutation in the promoter. Δ3 BOX has had the 3 box deleted. In U1, the 3 box is replaced by the last 26 bp of the U1 gene (small cross-hatched box) and the 3 box (large cross-hatched box), and U1DS contains the 63 bp just downstream of the U1 3 box in addition (grey box). In DS, the U2 3 box has been replaced by the sequences downstream from the U1 3 box. The relative positions of the S1 probe and riboprobes is noted below the WT structure. The sizes of the specific U2 RNase protection products for the constructs are noted below the WT structure. The size of the additional product from U1 and DS is indicated below the U1 structure. The size of the readthrough products is indicated below each structure. The results of quantitation of S1 (mean and standard deviation) and RNase protection analysis (average of two experiments) are noted on the right of the figure. (B) The results of S1 analysis of RNA transcribed from the constructs in (A) is shown. The construct is noted above each lane, the positions of the S1 products is noted at the left and the position of the probes is noted at the right. (C) The results of RNase protection analysis of RNA transcribed from the constructs in (A) is shown. The construct is noted above each lane. The positions of the protected products are noted on the left. analyses are marked on Figure 6A, and the quantitation 3 box alone (U1DS, lane 5, 67.5%), indicating that of the results is shown to the right of the figure. Since this downstream, in vivo-footprinted region has little effect the 5 S1 probe is complementary to a common region of on the steady-state RNA. Hernandez (1985) also saw little the constructs, the relative amount of steady-state RNA effect of this downstream region on the formation of can be readily assessed (Figure 6B). A high level of 3 ends of transcripts from a U1 promoter. Addition of transcription is detected from the ‘wild-type’ marked U2 the U1 downstream region alone to the Δ3 BOX template gene (WT, lane 1), as seen previously (Murphy, 1997), only further decreased the level of steady-state RNA and mutation of the essential PSE element in the promoter detected (DS, lane 6, 16%). (P–, lane 2) abolishes transcription as expected. Mutation The RNase protection analysis shown in Figure 6C of the 3 box causes a significant reduction in the amount demonstrates the effect of the changes on the 3 ends of steady-state RNA (Δ3 BOX, lane 3, 27% of WT). of transcripts. In this analysis, the endogenous U2 is This is in agreement with previous studies on human complementary to a region in all the probes used (see snRNA genes showing that the 3 box is essential for the Figure 6A) and a protection product of 158 nucleotides accumulation of high steady-state levels of transcripts (EnU2) is therefore present in all lanes. The VAI RNA (Hernandez, 1985; Yuo et al., 1985; Neuman de Vegvar gives an 89 nucleotide product that serves as an internal et al., 1986; Ach and Weiner, 1987; reviewed in Hernandez, transfection control. The major product of the WT con- 1992). Addition of the 3 box of the U1 gene to the Δ3 struct is a band of the expected size for mature marked BOX construct increases the amount of steady-state RNA U2 RNA (WT, lane 1, U2M). There are also distinct (U1, lane 4, 65%), confirming that the 3 boxes are products several nucleotides longer that may correspond largely interchangeable. Replacement of the U2 3 box to the 3-extended precursors of U2 RNA previously and downstream region with the equivalent region of the described (Wieben et al., 1985; Yuo et al., 1985). The U1 gene has essentially the same effect as replacing the same pattern is evident for the endogenous U2 RNA also 2872 Termination of transcription of snRNA genes detected by this analysis (EnU2). The much longer product visible in lane 1 corresponds to small amounts of read- through transcripts mismatching with the end of the probe (RT, 0.3% of total transcripts) (see Figure 6A). Low levels of readthrough transcripts from the endogenous genes are also detected in longer exposures of RNase protection analyses (data not shown). Apart from the endogenous U2 products, none of these bands is detected when the promoter of the U2 gene is disabled (P–, lane 2). When the 3 box is deleted, the amount of mature marked U2 and the potential precursors decreases (Δ3 BOX, lane 3). However, the readthrough transcripts increase to 2% of the total. In the U1 construct, the U2 3 box has been replaced by the last 26 bp of the U1 gene followed by the 3 box (see Materials and methods). Thus, the 3 end directed by the U1 sequences should be distinct from that of the WT construct, and products of the right size (257 nucleotides) are detected for the U1 and U1DS constructs (lanes 4 and 5, U1 3 BOX). Again, potential precursors extended by a few nucleotides at the 3 end are detected. Perhaps surprisingly, an approximately equal level of products equivalent to the mature U2 3 end and associated precursors is also apparent (see below). Approximately the same level of readthrough transcripts is detected for U1 as for the WT construct (lane 4, 0.4%). However, the level of readthrough appears to have decreased when the region downstream of the U1 3 box Fig. 7. The region downstream from the 3 box of the U1 contains a transcription terminator. (A) The structure of the marked U2 gene is present (U1DS, lane 5, 0.1%). Addition of this construct is shown and the relative positions of the probes used for the downstream region alone to the Δ3 box construct has run-on analysis are indicated. (B) The results of run-on analysis of the very little effect on any of the products (DS, lane 6). constructs in Figure 5A are shown. The hybridizations on the left of The results of this analysis confirm that the U2 3 box the figure were carried out with RNA that had been hybrid-selected is necessary for the accumulation of high levels of properly using a probe complementary to sequences in the ‘coding’ RNA from the marked template (see text). The hybridizations on the right were processed transcripts from the marked U2 gene. However, carried out with the RNA in the supernatant from the selection that it appears that proper 3 end formation is still occurring, contains any RNA that has not hybridized to the probe. The probes albeit at a reduced level, in the complete absence of the present on the filters are noted above each slot and the constructs 3 box. This has also been noted in previous studies transfected are noted on the left. The quantitation of the relative levels (Neuman de Vegvar et al., 1986; Ach and Weiner, 1987) of hybridization over the U2 MARK and VEC probes is expressed as a percentage of the ratio for the WT construct. and suggests that this process can occur by more than one pathway. The RNA produced from these constructs should interact with the appropriate proteins to make U2 snRNP, run-on analysis on the transfected U2 templates (Figure including the Sm-binding proteins. It is possible, therefore, 7). The VAI gene was again included as a transfection that longer transcripts, produced in the absence of the U2 control and the probe for 7SK was used to control the 3 box, with or without the addition of the U1 3 box efficiency of the run-on assay. U2 MARK is a 50mer downstream, can be trimmed after complexing with pro- oligonucleotide complementary to the linker in the marked teins in the cytoplasm. However, the pattern of bands is U2 gene plus some flanking sequence (see Materials identical in the absence and presence of the 3 box, and methods) used to detect transcription of the coding suggesting that these ‘precursors’ are generated in the sequence. The probe VEC is an 80mer oligonucleotide same way in both cases. Thus, the slightly 3-extended complementary to the region of the vector downstream transcripts noted above U2M in lane 1 (WT construct) from the U2 3 box which detects readthrough transcription may not represent specific precursors formed by a 3 box- (Figure 7A). Co is an 80mer oligonucleotide complement- dependent mechanism. Alternatively, other sequences ary to a region of the vector upstream from the U2 within the gene may be able to direct a low level of proper promoter and serves as a background control. After 3 end formation in the absence of the 3 box. carrying out the run-on reaction with nuclei from trans- Since the transcripts in this analysis are variable in fected cells, RNA was hybrid-selected using a probe sequence, any differences noted may simply reflect complementary to part of the coding region of the marked changes in RNA stability. However, the drop in detectable U2 gene (see Materials and methods). The selected RNA readthrough products from the U1DS construct is was then hydrolysed and hybridized to nitrocellulose filters consistent with the possibility that the downstream in vivo- (Figure 7B). This procedure ensures that signals are due footprinted region of the U1 gene is capable of terminating to RNA contiguous with the coding region and greatly transcription. reduces the level of background, allowing an accurate quantitation of the low signals over the short U2 MARK The region downstream from the 3 box of the U1 contains a transcription terminator probe. In order to determine unequivocally whether the 3 box RNA from cells transfected with the WT construct is causing termination directly, we have carried out hybridizes to both the U2 MARK and VEC probes, 2873 P.Cuello et al. indicating that transcription is continuing beyond the Termination may be effected directly by proteins binding 3 box in the transfected gene as it does in the endogenous to the sequence within the in vivo-footprinted region, U2 genes. Importantly, no signal is detected over any U2 perhaps in conjunction with a high proportion of T template probe when the PSE is mutated (Figure 7B, P–). residues. Interestingly, the results of our nuclear run-on Neither VAI nor 7SK transcripts have been selected by analysis indicate that the human U2 genes do not have an the probe and remain in the supernatant after the selection efficient termination element within the 250 bp down- procedure, indicating that the selection is specific. Deletion stream of the 3 box. Our results also suggest that of the U2 3 box has little effect on the level of readthrough termination is coupled to recognition of the 3 box. (Δ3 BOX, 89% of the level seen with WT). Introduction of the 3 box of the U1 gene similarly has little effect Why does the U1 gene have a terminator close to (U1, 91.2%). However, the presence of both the U1 the 3 box? 3 box and the immediate downstream region significantly It is not clear why termination of transcription of the U1 reduces the level of readthrough (U1DS, 24.7%). The genes should occur very close to the 3 box while addition of the U1 downstream region alone appears to transcription of the U2 genes continues further down- have a less pronounced effect on the level of readthrough stream. The U1 gene repeat unit is 45 kb long and (DS, 60.5%). The levels of VAI transcripts do not appear repeated 15–30 times on chromosome 1 (Lindgren et al., to vary drastically relative to the coding region transcripts 1985b). This is an extraordinarily large repeat unit con- of the different transfected genes, suggesting that changes sidering that the size of the RNA-coding region of the U1 in the levels of steady-state RNA (Figure 6) are due gene is 200 bp. Unlike the U2 repeat units which are mainly to post-transcriptional effects. arranged tandemly, the U1 genes are loosely clustered and These results lend further support to the hypothesis organized in an irregular and highly polymorphic manner that the in vivo-footprinted element located immediately (Manser and Gesteland, 1982; Lund and Dahlberg, 1984; downstream of the 3 box of the U1 genes is a transcrip- Bernstein et al., 1985). In addition, there are multiple tional terminator. This would satisfactorily explain the genes present in the RNU1 locus. These genes include difference in the run-on profiles of the human U1 and U2 tRNA genes (Van der Drift et al., 1994, 1995) and several genes. The U1 and U2 3 boxes alone appear to have putative neuroblastoma suppressor genes including human little effect on the level of readthrough transcription, Kruppel-related 3 (HKR3) (Maris et al., 1997) and human indicating that these are not efficient termination signals Elk-related kinase (ERK) (Saito et al., 1995). Since not by themselves. However, the putative U1 terminator also all the genes in this region have been identified, or their does not appear to function efficiently in the absence of precise locations determined, it is unclear whether any of the 3 box, suggesting that both elements are necessary these genes lie close to the U1 genes. Perhaps termination for termination of transcription. of transcription of the U1 genes is more critical than for the U2 genes, in order to prevent polymerases reading into downstream genes and causing inhibition of initiation Discussion by transcriptional interference [see Greger et al. (1998) Previous studies on snRNA gene transcription have not for a discussion of this phenomenon]. resolved whether the 3 box functions in the nascent RNA as a processing element or as a DNA element to terminate How might initiation, 3 box function and transcription. The 3 box is necessary for efficient 3 end termination be linked? formation of snRNA gene transcripts (Hernandez, 1985; It is possible that only a fraction of the nascent transcripts Yuo et al., 1985; Ciliberto et al., 1986; Neuman de Vegvar from the human U2 genes are terminated at the 3 box et al., 1986; Ach and Weiner, 1987; reviewed in Hernandez, and processed properly. However, this seems unlikely as 1992), and the function of this element cannot be separated these genes encode very abundant RNAs. In common readily from the process of transcription. Notably, tran- with Yuo et al. (1985) and Neuman de Vegvar et al. scripts containing a copy of the 3 box do not undergo (1986), we favour the hypothesis that the 3 box is an processing when reinjected into Xenopus oocytes RNA processing element present in the nascent RNA and (Ciliberto et al., 1986) and although U1 transcripts that that the 3–16 nucleotide extended precursors detected include the 3 box are detected in HeLa cells, these do in vivo are produced directly by an RNA processing event not seem to be processing substrates (Lobo and Marzluff, mediated by the 3 box. Yuo et al. (1985) have speculated 1987). In addition, recognition of the U1 3 box in vitro further that such a processing reaction may involve appears to be co-transcriptional and only occurs if tran- snRNAs in analogy to 3 end processing of histone scription is initiated from a class II snRNA promoter mRNAs (see below). (Gunderson et al., 1990). Although these findings do not The 3 ends of all other pol II transcripts (mRNAs) rule out that the 3 box is a processing signal, they support are produced by cleavage of nascent RNA. The replication- the notion that this element is a transcription terminator. activated histone mRNAs are produced from precursors The demonstration by Kunkel and Pederson (1985) that by a cleavage event that requires base pairing between transcription terminates very close to the 3 box in the the nascent RNA and the U7 snRNA (reviewed by U1 genes provided the most compelling argument to date Marzluff, 1992). In the case of the remaining mRNAs, that the 3 box is a termination element. However, the cleavage of the nascent RNA and subsequent results presented here show that the 3 box is not sufficient polyadenylation of the resulting 3 end are directed by for efficient termination of transcription. Instead, we the polyadenylation signal in the transcribed RNA implicate a novel element immediately downstream of the (reviewed by Colgan and Manley, 1997). Processing at 3 box in termination of transcription of the U1 gene. the poly(A) site is also linked intimately to both initiation 2874 Termination of transcription of snRNA genes and termination of transcription. Factors required for polyadenylation signals are not recognized efficiently in polyadenylation associate with the C-terminal domain transcripts initiated from the promoters of the U1 or U2 (CTD) of pol II before the poly(A) signal is transcribed, genes (Neuman de Vegvar et al., 1986; Dahlberg and possibly at initiation (Dantonel et al., 1997), and removal Schenborn, 1988; Hernandez and Lucito, 1988; Lobo of the CTD dramatically decreases the efficiency of and Hernandez, 1989). However, recognition of poly(A) polyadenylation in vivo (McCracken et al., 1997). In signals can be uncoupled readily from transcription, addition, termination of transcription of vertebrate mRNA whereas recognition of the 3 box cannot (Lobo and genes requires that splicing of the terminal intron and Marzluff, 1987; Ciliberto et al., 1989; Gunderson et al., polyadenylation take place, even though termination can 1990), as noted above. This suggests that a promoter- occur hundreds of base pairs downstream from the coding dependent activity is also essential for any processing region (Dye and Proudfoot, 1999, and references therein). event directed by the 3 box. Exactly how termination occurs in these genes is not well Termination of transcription of both mRNA genes and understood, but Birse et al. (1998) have shown recently snRNA genes may also occur by a common mechanism that the function of cleavage factors in the polyadenylation involving cleavage of the transcript/release of processing apparatus is required for termination of transcription of factors from the polymerase. Termination of transcription mRNA genes in Saccharomyces cerevisiae. Processing of of the U2 gene would, then, resemble the process in those the RNA at the poly(A) site is likely to influence directly mRNA genes where termination occurs hundreds of base the association of processing factors with the CTD of the pairs downstream from the poly(A) site. In the U1 genes transcribing polymerase. Thus, transcribing a poly(A) instead, factors bound to the DNA just downstream of the signal may convert the polymerase to a more easily processing element terminate the ‘destabilized’ polymerase terminated form as suggested by McCracken et al. (1997) much sooner. and Dye and Proudfoot (1999). In the case of the closely spaced genes encoding the mammalian complement pro- Materials and methods teins, C2 and factor B, however, it has been shown that binding sites for the protein Maz are required in addition Nuclear run-ons and hybrid selection to a functional poly(A) site for efficient termination to The protocol of Ashe et al. (1997) was used with the following changes: ~50 μl of nuclei were collected from two 120 mm dishes of HeLa cells occur (Ashfield et al., 1994). Thus, a combination of for each reaction and 7 μlof [α- P]UTP (3000 Ci/mmol) was added to processing and binding of factors to elements in the DNA each reaction mix. A 1 μg aliquot of each 47–50mer oligonucleotide or can work together to terminate pol II relatively close to 1.6 μg of each 80mer oligonucleotide was denatured and fixed to the the poly(A) site. nitrocellulose filter after slot-blotting. After hybridization, filters were washed three times in 1.0 SSPE/0.1% (w/v) SDS at 37°C (sometimes Efficient recognition of the 3 box in vivo and in vitro with 1 μg/ml RNase A added) and exposed to Kodak AR film at –70°C requires that transcription is initiated from a class II or to Molecular Dynamics Phosphor storage screens for quantitation in snRNA gene promoter, indicating that the pre-initiation a PhosphorImager. complex formed on these genes influences the downstream For run-on analysis of the endogenous U2 genes, the oligos are RNA processing event(s) (Hernandez and Weiner, 1986; complementary to the U2 gene sequence from pTP18 (Pavelitz et al., 1995; DDBJ/EMBL/GenBank accession No. U57614) taking the first Neuman de Vegvar et al., 1986; Gunderson et al., 1990). base pair of the U2 coding sequence as 1: US, –50 to –1; A, 1–47; B, Thus, the initial 3 end processing event of transcripts 48–94; C, 95–141; D, 142–188; E, 189–238; F, 239–288; G, 289–338; from both mRNA and snRNA genes appears to be influ- H, 339–388; I, 399–438; J, 439–489; Z, 789–838. AS is complementary enced directly by events at initiation. The mechanism of to A. The 7SK oligonucloeotide is complementary to sequences 61–110 of the coding region for 7SK (Murphy et al., 1986). co-ordinate promoter/3 box-directed processing of snRNA For run-on analysis of the U1 genes, the oligos A, B and C are gene transcripts may be analogous to the mechanism of complementary to nucleotides 6–55, 101–150 and 171–250, respectively, polymerase-assisted 3 end cleavage and polyadenylation of HSD2 (Manser and Gesteland, 1982) taking the first base of the U1 of mRNA genes. As mentioned above, the structure of coding region as 1. Oligo D is complementary to nucleotides 171–250 the promoters of class II snRNA genes is different from of HU1-1 (Lund and Dahlberg, 1984; DDBJ/EMBL/GenBank accession No. J00318). that of mRNA genes. The essential PSE element is For run-on analysis of transfected genes, the nuclei were collected both necessary and sufficient for initiation and 3 box from two 120 mm dishes of HeLa cells 48 h after transfection with recognition in vivo (Hernandez and Lucito, 1988; Neuman 12.5 μg of the test construct and 1.25 μg of VAI using Lipofectamine de Vegvar and Dahlberg, 1989; Parry et al., 1989). The PSE (Gibco-BRL) according to the manufacturer’s instructions. U2 MARK is recognized by a multisubunit factor known variously as is 5-GAT- ACTACACTTGATCCTCTAGAGCTGGTACCACTAGAGG- ATCTTAGCCA-3. VEC and Co are complementary to bp 61–140 and PTF, SNAPc and PBP (Henry et al., 1998, and references 2792 to 2871 of pGEM 4, respectively, taking the first base pair of therein), which is critical for the formation of a specific the SP6-transcribed region as 1. The VAI oligonucleotide is complement- pre-initiation complex containing TBP and TFIIB (Bernues ary to bp 1–80 of the VAI-coding region. Synthetic RNAs were made et al., 1993; Sadowski et al., 1993). This specialized by T7 or SP6 from sequences cloned into pGEM vectors. Hybrid selection probes were made in the same way with the inclusion of biotin- complex may directly recruit a specialized polymerase 16-UTP (Boehringer Mannheim) as described by Dye and Proudfoot associated with processing factors. Alternatively, the poly- (1999). Hybrid selection was carried out as described by Dye and merase may become associated with specific processing Proudfoot (1999). factors after recruitment, as appears to happen in the interaction of the CTD of pol II with splicing and Constructs The marked U2 gene is described by Murphy (1997) and extends from polyadenylation factors. An obvious scenario is one where the StuI site at –556 of dl-556 (Ares et al., 1985) to the NaeI site at snRNA processing factors rather than mRNA processing 93 downstream of the end of the RNA coding region. This was cloned factors interact with pol II at initiation of transcription of into pGEM4 using the EcoRI and HincII sites in the polylinker. P– was snRNA genes, as already suggested by Neuman de Vegvar made by PCR to replace the sequence between –60 and –49 by et al. (1986). This would account for the finding that TCCCACGGCCGT. The Δ3 BOX construct was made by PCR to 2875 P.Cuello et al. replace the 3 box from 19 to 41 with a C residue to create a unique Bernues,J., Simmen,K.A., Lewis,J.D., Gunderson,S.I., Moncollin,M., MluI site. The U1 sequence from 138 to 187 and the sequence from 192 Egly,J.-M. and Mattaj,I.W. (1993) Common and unique transcription to 250 of pHU1-1 (Lund and Dahlberg, 1984) downstream of the start factor requirements of human U1 and U6 snRNA genes. EMBO J., of transcription were cloned into this site to create U1 and DS, 12, 3575–3585. respectively. For U1DS, MluI–NaeI was replaced with sequences Birse,C.E., Minvielle-Sebastia,L., Lee,B.A., Keller,W. and Proudfoot,N.J. from 138 to 250 of HU1-1. The pGEM 4-based template for the hybrid (1998) Coupling termination of transcription to messenger RNA selection probe for endogenous U2 (Figure 2) was made by PCR from maturation in yeast. Science, 280, 298–301. pTP18 (Pavelitz et al., 1995) using oligos AS and D. For hybrid selection Chanfreau,G., Abou Elela,S., Ares,M. and Guthrie,C. (1997) Alternative of marked U2 transcripts, a fragment containing the sequence from the 3-end processing of U5 snRNA by RNase III. Genes Dev., 11, DdeI site at bp 20 of the coding region to the end of oligonucleotide D 2741–2751. was prepared by PCR of the marked U2 template and cloned into Ciliberto,G., Dathan,N., Frank,R., Philipson,L. and Mattaj,I.W. (1986) pGEM3. Each U2 construct was recloned into the EcoRI site of pGEM4 Formation of the 3 end on U snRNAs requires at least three sequence to place the polylinker region between the T7 promoter and the 3 end elements. EMBO J., 5, 2931–2937. of the U2 sequence. Colgan,D.F. and Manley,J.L. (1997) Mechanism and regulation of mRNA polyadenylation. Genes Dev., 11, 2755–2766. Steady-state RNA analysis Corden,J.L. and Patturajan,M. (1997) A CTD function linking For each 90 cm dish of HeLa cells, a 5 μg aliquot of each U2 construct transcription to splicing. Trends Biochem. Sci., 22, 413–416. was co-transfected with 500 ng of VAI using Lipofectamine (Gibco- Dahlberg,J.E. and Schenborn,E.T. (1988) The human U1 snRNA BRL) as recommended by the manufacturers, and cytoplasmic RNA promoter and enhancer do not direct synthesis of messenger RNA. was collected as described by Whitelaw et al. (1989). RNase protection Nucleic Acids Res., 16, 5827–5840. was carried out as described by Eggemont and Proudfoot (1993). Dantonel,J.C., Murthy,K.G.K., Manley,J.L. and Tora,L. (1997) Riboprobes for analysis of U2 transcripts were made by T7 after Transcription factor TFIID recruits factor CPSF for formation of the digestion with FokI that cuts 42 bp upstream from the start of the U2 3 end of mRNA. Nature, 389, 399–402. coding region. The VAI riboprobe was made by SP6 after digesting the Dye,M.J. and Proudfoot,N.J. (1999) Terminal exon definition occurs co- VAI gene in pGEM4 with BamHI. S1 analysis was carried out as transcriptionally and promotes termination of RNA polymerase II. described by Murphy (1997) using the oligonucleotides described therein. Mol. Cell, 3, 371–378. The products of S1 analysis are 45 and 63 nucleotides for the VAI and Eggermont,J. and Proudfoot,N.J. (1993) Poly(A) signals and U2 RNAs, respectively. transcriptional pause sites combine to prevent interference between RNA polymerase promoters. EMBO J., 12, 2539–2548. In vivo footprinting Elela,S.A. and Ares,J.M. (1998) Depletion of yeast RNase III blocks Cells or purified naked DNA were treated first with DMS (Aldrich) and correct U2 3 end formation and results in polyadenylated but then with piperidine (Aldrich) to prepare in vivo or in vitro DNA functional U2 RNA. EMBO J., 17, 3738–3746. samples, respectively, as described by Greger et al. (1998). Ligation- Greger,I.H., Demarchi,F., Giacca,M. and Proudfoot,N.J. (1998) mediated PCR was carried out on these samples using the following Transcriptional interference perturbs the binding of Sp1 to the HIV-1 nested primers for U1: U1.DS1, 5-CTCGCTTTTTCTCCTATGGC-3; promoter. Nucleic Acids Res., 26, 1294–1300. U1.DS2, 5-CAGGCGACATGTTACTTCCTATTCCGC-3; U1.DS3, 5- Gunderson,S.I., Knuth,M.W. and Burgess,R.R. (1990) The human U1 GGCGACATGTTACTTCCTATTCCGCAGCCCTC-3; and for U2: snRNA promoter correctly initiates transcription in vitro and is U2.US1, 5-GGATTTTTGGAGCAGGGAGA-3; U2.US2, 5-AATA- activated by PSE1. Genes Dev., 4, 2048–2060. GGAGCTTGCTCCGTCCACTCC-3; U2.US3, 5-AGGAGCTTGCTC- Hammarstrom,K., Westin,G., Bark,C., Zabielski,J. and Pettersson,U. CGTCCACTCCACGCATC-3. (1984) Genes and pseudogenes for human U2 RNA. Implications for the mechanism of pseudogene formation. J. Mol. Biol., 179, 157–169. Henry,R.W., Mittal.,V., Ma.,B., Kobayashi,R. and Hernandez,N. (1998) Acknowledgements SNAP19 mediates the assembly of a functional core promoter complex We would like to thank Andre Furger and David Bentley for helpful (SNAPc) shared by RNA polymerases II and III. Genes Dev., 12, suggestions, and Andre Furger and Alexandra Moreira for critical reading 2664–2672. of the manuscript. We would especially like to thank Charlie Birse for Hernandez,N. (1985) Formation of the 3 end of U1 snRNA is directed encouraging us to try using oligonucleotide probes for the run-ons and by a conserved sequence located downstream of the coding region. Ingo Greger for setting up in vivo footprinting in the laboratory. P.C. EMBO J., 4, 1827–1837. was supported by a BBSRC graduate studentship, M.D. was supported Hernandez,N. (1992) Transcription of vertebrate snRNA genes and by a Glaxo graduate studentship, N.J.P. is supported by Wellcome related genes. In McKnight,S.L. and Yamamoto,K.R. (eds), Programme Grant No. 032743, and D.C.B. and S.M. are supported by Transcriptional Regulation. Cold Spring Harbor Laboratory Press, MRC Senior Fellowship No. G117/309. Cold Spring Harbor, NY, Vol. 1, pp. 281–313. Hernandez,N. and Lucito,R. (1988) Elements required for transcription initiation of the human U2 snRNA coincide with elements required References for snRNA 3 end formation. EMBO J., 7, 3125–3134. Hernandez,N. and Weiner,A.M. (1986) Formation of the 3 end of U1 Ach,R.A. and Weiner,A.M. (1987) The highly conserved U small nuclear snRNA requires compatible snRNA promoter elements. Cell, 47, RNA 3-end formation signal is quite tolerant to mutation. Mol. Cell. 249–258. Biol., 7, 2070–2079. Huang,Q., Jacobson,M.R. and Pederson,T. (1997) 3 Processing of Ares,M.,Jr, Mangin,M. and Weiner,A.M. (1985) Orientation-dependent human pre-U2 small nuclear RNA: a base-pairing interaction between transcriptional activator upstream of a human U2 snRNA gene. Mol. the 3 extension of the precursor and an internal region. Mol. Cell. Cell. Biol., 5, 1560–1570. Biol., 17, 7178–7185. Ashe,H.L., Monks,J., Wijgerde,M., Fraser,P. and Proudfoot,N.J. (1997) Kuhn,A. and Grummt,I. (1989) 3-End formation of mouse pre-rRNA Intergenic transcripts and transinduction of the human β-globin locus. involves both transcription termination and a specific processing Genes Dev., 11, 2494–2509. reaction. Genes Dev., 3, 224–231. Ashfield,R., Patel,A.J., Bossone,S.A., Brown,H., Campbell,R.D., Kunkel,G.R. and Pederson,T. (1985) Transcription boundaries of U1 Marcu,K.B. and Proudfoot,N.J. (1994) MAZ-dependent termination small nuclear RNA. Mol. Cell. Biol., 5, 2332–2340. between closely spaced human complement genes. EMBO J., 13, Liao,D. and Weiner,A.M. (1995) Concerted evolution of the tandemly 5656–5667. repeated genes encoding primate U2 small nuclear RNA (the RNU2 Bentley,D.L. and Groudine,M. (1986) A block to elongation is largely locus) does not prevent rapid diversification of the (CT)n·(GA)n responsible for decreased transcription of c-myc in differentiated HL60 microsatellite embedded within the U2 repeat unit. Genomics, 30, cells. Nature, 321, 702–706. 583–593. Bernstein,L.B., Manser,T. and Weiner,A.M. (1985) Human U1 small Lindgren,V., Ares,M., Weiner,A.M. and Franke,U. (1985a) Human genes nuclear RNA genes: extensive conservation of flanking sequences for U2 small nuclear RNA map to a major adenovirus 12 modification suggests cycles of gene amplification and transportation. Mol. Cell. Biol., 5, 2159–2171. site on chromosome 17. Nature, 314, 115–116. 2876 Termination of transcription of snRNA genes Lindgren,V., Bernstein,L.B., Weiner,A.M. and Franke,U. (1985b) Human Van Arsdell,S.W. and Weiner,A.M. (1984) Human genes for U2 small U1 small nuclear RNA pseudogenes do not map to the site of the U1 nuclear RNA are tandemly repeated. Mol. Cell. Biol., 4, 492–499. genes in 1p36 but are clustered in 1q12–q22. Mol. Cell. Biol., 5, Van der Drift,P., Chan,A., van Roy,N., Laureys,G., Westerveld,A., 2172–2180. Speleman,F. and Versteeg,R. (1994) A multimegabase cluster of Lobo,S.M. and Marzluff,W.F. (1987) Synthesis of U1 RNA in isolated snRNA and tRNA genes on chromosome 1p36 harbours an adenovirus/ mouse cell nuclei: initiation and 3 end formation. Mol. Cell. Biol., SV40 hybrid virus integration site. Hum. Mol. Genet., 3, 2131–2136. 7, 4290–4296. Van der Drift,P., Chan,A., Laurets,G., van Roy,N., Sickmann,G., den Dunnen,J., Westerveld,A., Speleman,F. and Versteeg,R. (1995) Lobo,S.M. and Hernandez,N. (1989)A7bp mutation converts a human Balanced translocation in a neuroblastoma patient disrupts a cluster RNA polymerase II snRNA promoter into an RNA polymerase III of small nuclear RNA U1 and tRNA genes in chromosomal band promoter. Cell, 58, 55–67. 1p36. Genes Chromosomes Cancer, 14, 35–42. Lobo,S.M. and Hernandez,N. (1994) Transcription of snRNA genes by Weber,J., Jelinek,W. and Darnell,J.E.,Jr (1977) The definition of a large RNA polymerase II and III. In Conaway,R.C. and Conaway,J.W. (eds), viral transcription unit late in AD2 infection of HeLa cells: mapping Transcription, Mechanisms and Regulation. Raven Press, New York, of nascent RNA molecules labelled in isolated nuclei. Cell, 10, NY, Vol. 3, pp. 127–159. 611–616. Lund,E. and Dahlberg,J.E. (1984) True genes for human U1 small Westin,G., Zabielski,J., Hammarstrom,K., Monstein,H.J., Bark,C. and nuclear RNA: copy number, polymorphism and methylation. J. Biol. Pettersson,U. (1984) Clustered genes for human U2 RNA. Proc. Natl Chem., 259, 2013–2021. Acad. Sci., USA, 81, 3811–3815. Manser,T. and Gesteland,R.F. (1982) Human U1 loci: genes for human Whitelaw,E., Hogben,P., Hanscombe,O. and Proudfoot,N.J. (1989) U1 RNA have dramatically similar genomic environments. Cell, 29, Transcriptional promiscuity of the human α-globin gene. Mol. Cell. 257–264. Biol., 9, 241–251. Maris,J.M., Jensen,J., Sulman,E.P., Beltinger,C.P., Allen,C., Biegel,J.A., Wieben,E.D., Nenninger,J.M. and Pederson,T. (1985) Ribonucleoprotein Brodeur,G.M. and White,P.S. (1997) Human Kruppel-related 3 organisation of eukaryotic RNA. XXXII. U2 small nuclear RNA (HKR3): a candidate for the 1p36 neuroblastoma tumour suppressor precursors and their accurate 3 processing in vitro as ribonucleoprotein gene? Eur. J. Cancer, 33, 1991–1996. particles. J. Mol. Biol., 183, 69–78. Marzluff,W.F. (1992) Histone 3 ends: essential and regulatory functions. Yuo,C., Ares,M. and Weiner,A.M. (1985) Sequences required for 3 end Gene Expression, 2, 93–97. formation of human U2 small nuclear RNA. Cell, 42, 193–202. McCracken,S.N., Fong,N., Yankulov,K., Ballantyne,S., Pan,G.H., Greenblatt,J., Patterson,S.D., Wickens,M. and Bentley,D. (1997) The Received February 10, 1999; revised and accepted March 29, 1999 C-terminal domain of RNA polymerase II couples messenger RNA processing to transcription. Nature, 385, 357–361. Murphy,S. (1997) Differential in vivo activation of the class II and class III snRNA genes by the POU-specific domain of Oct-1. Nucleic Acids Res., 25, 2068–2076. Murphy,S., Tripodi,M. and Melli,M. (1986) A sequence upstream from the coding region is required for transcription of the 7SK RNA genes. Nucleic Acids Res., 14, 9243–9260. Nagai,K. and Mattaj,I.W. (1994) RNA–protein interactions in the splicing snRNPs. In Nagai,K. and Mattaj,I.W. (eds), Frontiers in Molecular Biology. IRL Press at Oxford University Press, Oxford, UK, pp. 150–177. Neugebauer,K.M. and Roth,M.B. (1997) Transcription units as RNA processing units. Genes Dev., 11, 3279–3285. Neuman de Vegvar,H.E., Lund,E. and Dahlberg,J.E. (1986) 3 End formation of U1 snRNA precursors is coupled to transcription from snRNA promoters. Cell, 47, 259–266. Neuman de Vegvar,H.E. and Dahlberg,J.E. (1989) Initiation and termination of human U1 transcription requires the concerted action of multiple flanking sequence elements. Nucleic Acids Res., 17, 9305–9318. Parry,H.D., Tebb,G. and Mattaj,I.W. (1989) The Xenopus U2 gene PSE is a single, compact, element required for transcription initiation and 3 end formation. Nucleic Acids Res., 17, 3633–3644. Pavelitz,T., Rusche,L., Matera,A.G., Scharf,J.M. and Weiner,A.M. (1995) Concerted evolution of the tandem array encoding primate U2 snRNA occurs in situ, without changing the cytological context of the RNU2 locus. EMBO J., 14, 169–177. Sadowski,C.L., Henry,R.W., Lobo,S.M. and Hernandez,N. (1993) Targeting TBP to a non-TATA box cis-regulatory element: a TBP- containing complex activates transcription from snRNA promoters through the PSE. Genes Dev., 7, 1535–1548. Saito,T., Seki,N., Matsuda,Y., Kitahara,M., Murata,M., Kanda,N., Nomura,N., Yamamoto,T. and Hori,T.A. (1995) Identification of the human ERK gene as a putative receptor tyrosine kinase and its chromosomal localization to 1p36.1: a comparative mapping of human, mouse and rat chromosomes. Genomics, 26, 382–384. Seipelt,R.L., Zheng,B., Asuru,A. and Rymond,B.C. (1999) U1 snRNA is cleaved by RNase III and processed through an Sm site-dependent pathway. Nucleic Acids Res., 27, 587–595. Steinmetz,E.J. (1997) Pre-mRNA processing and the CTD of RNA polymerase II: the tail that wags the dog? Cell, 89, 491–494. Strobl,L.J. and Eick,D. (1992) Hold back of RNA polymerase II at the transcription start site mediates down-regulation of c-myc in vivo. EMBO J., 11, 3307–3314. Terns,M.P., Dahlberg,J.E. and Lund,E. (1993) Multiple cis-acting signals for export of pre-U1 snRNA from the nucleus. Genes Dev., 7, 1898–1908.

Journal

The EMBO Journal – Springer Journals

Published: May 17, 1999

Keywords: pol II; RNA processing; snRNA; termination; transcription

There are no references for this article.