The Rpb4 Subunit of Fission Yeast Schizosaccharomyces pombe RNA Polymerase II Is Essential for Cell Viability and Similar in Structure to the Corresponding Subunits of Higher Eukaryotes

Abstract

Both the gene and the cDNA encoding the Rpb4 subunit of RNA polymerase II were cloned from the fission yeast Schizosaccharomyces pombe. The cDNA sequence indicates that Rpb4 consists of 135 amino acid residues with a molecular weight of 15,362. As in the case of the corresponding subunits from higher eukaryotes such as humans and the plant Arabidopsis thaliana, Rpb4 is smaller than RPB4 from the budding yeast Saccharomyces cerevisiae and lacks several segments, which are present in the S. cerevisiae RPB4 subunit, including the highly charged sequence in the central portion. The RPB4 subunit of S. cerevisiae is not essential for normal cell growth but is required for cell viability under stress conditions. In contrast, S. pombe Rpb4 was found to be essential even under normal growth conditions. The fraction of RNA polymerase II containing RPB4 in exponentially growing cells of S. cerevisiae is about 20%, but S. pombe RNA polymerase II contains the stoichiometric amount of Rpb4 even at the exponential growth phase. In contrast to the RPB4 homologues from higher eukaryotes, however, S. pombe Rpb4 formed stable hybrid heterodimers with S. cerevisiae RPB7, suggesting that S. pombe Rpb4 is similar, in its structure and essential role in cell viability, to the corresponding subunits from higher eukaryotes. However, S. pombe Rpb4 is closer in certain molecular functions to S. cerevisiae RPB4 than the eukaryotic RPB4 homologues.

RNA polymerase II in eukaryotes is composed of more than 10 different polypeptides (for example, see reference 29). The genes coding for all 12 putative subunits of RNA polymerase II have been isolated from the budding yeast Saccharomyces cerevisiae (reviewed in references 26 and 27) and humans (10). Sometime ago we reported that the purified RNA polymerase II from the fission yeast Schizosaccharomyces pombe contained at least 10 polypeptides, devoid of the components corresponding to RPB4 and RPB9 of S. cerevisiae (21, 24; for a recent review, see reference 8). Later we cloned the gene and the cDNA for Rpb9 by PCR using the sequence knowledge of subunit 9 from other organisms (23). By Western blot analysis with antibodies against the Rpb9 protein expressed in Escherichia coli, we found that the purified S. pombe RNA polymerase II does indeed contain Rpb9, which had not been detected in the gel pattern because of its comigration with Rpb8 and Rpb11 (23).

Recently, the genes coding for subunit 4 were cloned from humans (10) and the plant Arabidopsis thaliana (15). Human cDNA for RPB4 was cloned by two-hybrid screening of cDNA coding for a protein which interacts with human RPB7 (hRPB7) (10), because S. cerevisiae RPB4 forms a binary complex with RPB7 (6, 11). On the other hand, the gene for the A. thaliana RPB15.9 (AtRPB15.9) subunit, which is a homologue of S. cerevisiae RPB4, was cloned by cross-hybridization using the homologous expressed sequence tag (EST) clone of oilseed rape (Brassica napus) as the probe (15). Both hRPB4 and AtRPB15.9 are smaller than S. cerevisiae RPB4, lacking a segment corresponding to the central portion of S. cerevisiae RPB4. As in the case of S. cerevisiae, subunits 4 and 7 from both human and the plant formed a stable binary complex, but this subassembly seems to be associated with the RNA polymerase II more tightly than that of S. cerevisiae (10, 15). We then reexamined whether the purified RNA polymerase II from S. pombe contains Rpb4 or not. Results herein described indicate that S. pombe contains the gene for Rpb4 and that the Rpb4 protein is essential for cell viability and more similar, in structure and function, to those of higher eukaryotes than that of S. cerevisiae.

MATERIALS AND METHODS

Yeast strains, media, and transformation.

The S. pombe strains used in this study are JY741 (h⁻ ade6-M216 ura4-D18 leu1) and JY746 (h⁺ ade6-M210 ura4-D18 leu1). Media YE, EMM, and ME were prepared as described previously (17). 5-Fluoroorotic acid (Toronto Research Chemicals) was used at 1 mg/ml. S. pombe was transformed by the lithium acetate method (18).

Cloning of the cDNA for rpb4.

Cloning of rpb4 cDNA was performed by a combination of 3′ RACE (rapid amplification of the 3′ end of cDNA) and 5′ RACE (rapid amplification of the 5′ end of cDNA). 3′ RACE was performed by the method described previously (23), using a 3′-Full RACE core set (Takara Shuzo, Kusatsu, Japan) and total mRNA as the template. Total mRNA was isolated from S. pombe JY741 as described previously (22). For amplification of the 3′-proximal region of the rpb4 cDNA, we designed a primer, primer 419, based on the sequence of cosmid c337. Primer 419 corresponds to nucleotide positions 434 to 452 (nucleotide position 1 was set as the first nucleotide of the rpb4 initiation codon) in the 3′-proximal exon (see Table 1 for the primer sequence and Fig. 1A for the location of the primer sequence along the rpb4 gene). For 5′ RACE, the oligo-capping method (22, 25) was employed, using oligo-capped mRNA as the template and a pair of primers, 5′-cap-specific primer, CAP20, and an rpb4-specific 3′ primer, 421R, with the sequence within the last exon (Table 1 and Fig. 1A). Nested PCR was performed using the first PCR products as the template and a combination of the oligo-cap-specific primer and another gene-specific primer, 420R (Table 1 and Fig. 1A). The complete rpb4 cDNA was constructed after combination of the PCR products from 3′ and 5′ RACEs, cloned into pGEM-T vector (Promega), and sequenced.

TABLE 1.

Primers used for PCR

Primer	Sequence	Restriction enzyme sitea	Positionb
419	5′-TTGGGAACACTCTGCTGCGAAAA-3′		434 to 452
420R	5′-CGCGAGACTAGGAATCAAGG-3′		493 to 474
421R	5′-CTCATCCAAGATGCCTTGAAG-3′		535 to 515
422	5′-AAACAACATTCTGAAGAGAGTG-3′		−42 to −21
423R	5′-GACTAATTAAGAAGTGTATGGTG		776 to 744
N420	5′-CATCTATTCcatATGCCGAGGGC-3′	NdeI	−12 to 11
X423R	5′-ATTctcgagATCTTGAAATTTACGCAAAGTGG-3′	XhoI	571 to 540
X426	5′-TTTTCTAGAATAACAGAACTATCTTG-3′	XbaI	−956 to −931
K429R	5′-AGGCgGTAccAACGGTAATTGGCGGACAC-3′	KpnI	1211 to 1183
B424R	5′-ATGAATgGATcCTCGCAAACACTC-3′	BamHI	−2 to −25
X427	5′-GCAATAAcTcGAgCTCAGGCAAGTCTG-3′	XhoI	567 to 593
4-1	5′-AACGAAATTAGTAGAAGAATCCTC-3′		−483 to −460
4-2	5′-TTTGGCAGAGTTTCGAGATGAC-3′		1111 to 1090
CAP20	5′-GAGAGAGACAGGCCTTACTG-3′

^aThe underlined sequences are the restriction enzyme sites indicated in the next column. Nucleotides that do not match those in the rpb4 gene are indicated by lowercase letters. 

^bPositions on the rpb4 gene. Nucleotide 1 is defined as the first nucleotide of the initiation codon. 

FIG. 1.

Structure of the rpb4 gene and the Rpb4 protein. (A) Nucleotide sequences were determined for both the rpb4 gene and its cDNA. The amino acid sequence of Rpb4 was predicted from the cDNA sequence. The Rpb4 open reading frame (large black bars) is interrupted by three introns (small white bars). Nucleotide 1 is defined as the first nucleotide of the initiation codon, while amino acid 1 is defined as the initiation codon. The positions and directions of primers used for PCR are shown by the arrows. For primer sequences, see Table 1. (B) Comparison of the amino acid sequences of RNA polymerase II subunit 4 in various organisms. The amino acid sequence of S. pombe Rpb4 subunit (Sp) is compared with the corresponding subunits from S. cerevisiae (Sc), Homo sapiens (Hs), and A. thaliana (At). The overall identity of the amino acid (aa) sequence of the S. pombe Rpb4 with those of other organisms is shown at the end of each alignment. Amino acids that are identical or similar at least between two species are outlined or shaded, respectively. Gaps introduced to maximize alignment are indicated by dashes.

Disruption of the rpb4 gene.

A 2.2-kb genomic DNA segment including the rpb4 gene from positions −953 to +1201 (+1 is set at the first nucleotide of start codon) was PCR amplified by using primers X426 and K429R (Table 1) and S. pombe genomic DNA as the template. After digestion with XbaI-KpnI, the segment was ligated into pUC18 between XbaI and KpnI sites to construct pUC-rpb4.

For disruption of the rpb4 gene, the entire vector sequence flanked with the 5′- and 3′-terminal sequences of rpb4 (but lacking the Rpb4-coding sequence) was PCR amplified with B424R and X427 as the primers (Table 1) and plasmid pUC-rpb4 as the template, and the PCR product was ligated with a 1.8-kb BamHI-XhoI ura4 fragment to construct pUC-rpb4::ura4, in which the entire coding sequence of rpb4 was replaced by the ura4 gene. A 2.8-kb rpb4::ura4 fragment was PCR amplified with Vent DNA polymerase (New England Biolabs), using primers 4-1 and 4-2 (Table 1) and pUC-rpb4::ura4 as the template; the resulting fragment was used to transform a diploid S. pombe strain constructed by a cross between JY741 and JY746. Ura⁺ transformants were selected on an EMM medium plate containing leucine, and screened on a plate containing 5-fluoroorotic acid for a stable Ura⁺ phenotype. The disruption of one chromosomal copy of the rpb4 gene was confirmed by PCR.

Expression and purification of Rpb4 and other Rpb proteins.

rpb4 cDNA including the complete coding sequence of Rpb4 was amplified by PCR with a pair of primers, i.e., a 5′ primer including an NdeI site sequence (the translational start codon is included in the NdeI sequence) and a 3′ primer including a XhoI site sequence. The PCR product was inserted into pET-21b (Novagen) between NdeI and XhoI sites to construct pET-Rpb4CH, the expression vector of Rpb4 with a hexahistidine (His₆) tag at the C terminus (Rpb4CH). Expression plasmids pET-Rpb4 (coding for Rpb4 without the His₆ tag), pET-Rpb8, pET-Rpb9, and pET-Rpb11 were also derivatives of pET-21b containing rpb4, rpb8, rpb9, and rpb11 cDNA, respectively. These cDNAs were PCR amplified using specific sets of 5′ and 3′ primers, each including the initiation codon (in the NdeI site) and the termination codon, respectively.

The expression plasmids were transformed into E. coli BL21(DE3). Rpb4CH was purified from induced-cell lysates by Ni²⁺-agarose column chromatography.

Anti-Rpb protein antibodies and Western blotting.

Anti-Rpb4 antibodies were raised in rabbits immunized with the purified Rpb4CH. Antibodies against other Rpb proteins were prepared as described previously (7, 23).

For Western blotting, the purified RNA polymerase or the induced E. coli cell lysates for the expression of Rpb proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and proteins were transferred to polyvinylidene difluoride membranes by electroblotting. Protein blots were probed with anti-Rpb4CH, anti-Rpb7CH, anti-Rpb8CH, anti-Rpb9CH, and anti-Rpb11CH antibodies (7, 23), followed by staining with anti-rabbit immunoglobulin G antibodies conjugated with horseradish peroxidase (Cappel). The peroxidase activity was visualized using an enhanced chemiluminescence kit (Amersham), and analyzed with LAS-1000plus lumino-image analyzer (Fuji Film).

Construction of double-expression plasmids of Rpb4 and Rpb7CH.

For simultaneous expression of both subunits 4 (S. pombe Rpb4 [SpRpb4] and S. cerevisiae RPB4 [ScRPB4]) and 7 (S. pombe Rpb7 and S. cerevisiae RPB7) in all possible combinations, four kinds of the double-expression plasmid, pET-Sp4/Sp7CH, pET-Sp4/Sc7CH, pET-Sc4/Sp7CH, and pET-Sc4/Sc7CH, were constructed by using plasmid pET-21b (Novagen) as a vector. First, cDNAs for S. pombe rpb4 and S. cerevisiae RPB4 were PCR amplified using 5′ and 3′ primers, each including the initiation codon (in the NdeI site) and the termination codon, and integrated into pET-21b, while cDNAs for S. pombe rpb7 and S. cerevisiae RPB7 were PCR amplified using 5′ and 3′ primers, each including the initiation codon-NdeI site and the termination codon-XhoI site. All these cDNAs were integrated into pET-21b as to synthesize SpRpb4, ScRPB4, SpRpb7-His₆, and ScRPB7-His₆ proteins. The shorter SphI-BamHI fragment of pET-Rpb4 or pET-RPB4 containing the subunit 4 cDNA and the larger BglII-SphI fragment of pET-Rpb7CH or pET-RPB7CH containing the subunit 7 cDNA sequence fused to the His₆ sequence were ligated in all possible combinations to yield the four kinds of double-expression plasmid. The double-expression plasmids were transformed into E. coli BL21(DE3) for construction of the simultaneous expression system of both subunits 4 and 7 in various combinations (note that both subunit 4 and 7 cDNAs in a plasmid are under independent control of the T7 promoter).

Isolation of complexes of Rpb4 or RPB4 and Rpb7 or RPB7.

The transformed E. coli BL21(DE3) with double-expression plasmids were grown in M9 medium containing 1% Bacto Tryptone, 4% glucose, and 100 μg of ampicillin per ml at 30°C. When the cells reached 80 U, as measured with a Klett-Summerson photometer, isopropyl-β-d-thiogalactopyranoside (IPTG) was added to a final concentration of 0.1 mM. After 3 h of incubation, the cells were harvested by centrifugation and stored at −80°C.

For purification of expressed proteins, the cells were suspended in buffer A (20 mM Tris-HCl [pH 8.0, 4°C], 0.1 M NaCl, 10% glycerol, 10 mM β-mercaptoethanol [β-ME], 0.5 mM phenylmethanesulfonyl fluoride [PMSF]) containing 0.3 mg of lysozyme per ml, incubated on ice for 20 min, sonicated, and centrifuged at 15,000 × g for 20 min at 4°C. Polyethylenimine (pH 7.9) was added to the supernatant to a final concentration of 0.1% (vol/vol) and mixed. The mixture was incubated on ice for 30 min. After centrifugation at 15,000 × g for 20 min at 4°C and then at 150,000 × g for 1 h at 4°C, the resulting supernatant was loaded onto a Ni²⁺-nitriloacetic acid agarose (Qiagen) column equilibrated with buffer A. The column was washed, in a stepwise fashion, with 20 times the column volume of buffer B (20 mM Tris-HCl [pH 8.0, 4°C], 0.5 M NaCl, 10% glycerol, 10 mM β-ME, 0.5 mM PMSF) and 10 times the column volume of buffer C (20 mM Tris-HCl [pH 8.0, 4°C], 0.1 M NaCl, 10% glycerol, 10 mM β-ME, 0.5 mM PMSF, 20 mM imidazole) and then eluted with buffer D (20 mM Tris-HCl [pH 8.0, 4°C], 0.1 M NaCl, 10% glycerol, 10 mM β-ME, 0.5 mM PMSF, 50 mM imidazole).

Gel filtration chromatography.

The complexes of Rpb4 or RPB4 and Rpb7 or RPB7 were loaded onto Superdex 75 PC 3.2/30 columns (Pharmacia) at a flow rate of 40 μl/min in buffer E (20 mM Tris-HCl [pH 8.0, 4°C], 0.3 M NaCl, 5% glycerol, 0.1 mM EDTA, 1 mM dithiothreitol) at 4°C, using the Smart System (Pharmacia). Fractions of 50 μl were collected and analyzed by SDS-PAGE.

Nucleotide sequence accession number.

The DNA sequence of S. pombe rpb4 has been deposited into the DDBJ/GenBank/EMBL database under accession no. AB019575.

RESULTS

Cloning of cDNA for Rpb4 and the rpb4 gene.

The S. pombe Rpb4 subunit has not been identified by SDS-PAGE analysis of the purified RNA polymerase II (21, 23). Likewise, we failed to identify the S. pombe rpb4 gene by PCR with primers which were synthesized by using the highly conserved amino acid sequences among S. cerevisiae RPB4 and both mouse and human ESTs (DNA database accession no. AA139434 and W87848, respectively) (the region corresponding to amino acid residues 99 to 109 of S. pombe Rpb4 [Fig. 1B]). Recently, however, the RPB4 homologue genes were cloned both from humans and a plant (10, 15). We then reinitiated the search for the rpb4 gene in S. pombe. After analysis of the newly publicized sequence of S. pombe genome in PomBase (Sanger Center) using the entire amino acid sequences of S. cerevisiae RPB4 (ScRPB4) and hRPB4 as references, we found several sequence segments which are similar to parts of ScRPB4 and hRPB4 in the cosmid c337 (the sequence information of Rpb4 in the cosmid c337 was kindly provided by Pierre Thuriaux and Olivier Gadal). Based on the comparison between the amino acid sequences of ScRPB4 and hRPB4 proteins and the nucleotide sequence of S. pombe c337 cosmid, we predicted the presence of four exons and three introns within the Rpb4-coding sequence (Fig. 1A).

To confirm our prediction, we tried to determine the rpb4 cDNA sequence using a combination of 3′ RACE and 5′ RACE. After 3′ RACE analysis of total S. pombe mRNA using primer 419 (Fig. 1A), one DNA fragment about 400 bp long was amplified, which included the sequence downstream from the last exon of rpb4 including the 3′ untranslated region (Fig. 1A). On the other hand, the 5′-proximal region of rpb4 cDNA was PCR amplified by 5′ RACE using a 5′-cap primer and one of the rpb4-specific primers 421R (Table 1 and Fig. 1A). The major DNA product of about 550 bp in length, obtained after the second PCR using primer 420R, contained the sequence upstream from the C-terminal proximal exon. The complete cDNA clone was constructed by PCR using a set of primers, 5′-terminal 422 and 3′-terminal 423R (Table 1 and Fig. 1A). A 650-bp-long PCR product was amplified. The sequence of this PCR product was completely identical to the sequence obtained by the combination of 5′ and 3′ RACEs and the sequence predicted from the c337 cosmid sequence.

To isolate the genomic clone of the rpb4 gene, PCR amplification was performed with primers 422 and 423R (Table 1 and Fig. 1A) and with the genomic DNA as the template. The sequence determination of PCR products revealed that the rpb4 gene contains four exons and three introns. The exon-intron junction sequences agree well with the known splice junction sequences in S. pombe (21, 30).

Structure of the Rpb4 protein.

The Rpb4-coding sequence consists of 408 nucleotides, which encodes a polypeptide of 135 amino acid residues with a molecular weight of 15,362 (Fig. 1A). The predicted amino acid sequence of S. pombe Rpb4 was compared with those of the corresponding subunits from three organisms, i.e., human RPB4, A. thaliana RPB15.9, and S. cerevisiae RPB4 (Fig. 1B). S. pombe Rpb4 is similar in size to those of the human and plant RPB4 homologues (142 residues for human and 138 residues for plant), but it was smaller than the S. cerevisiae RPB4 (221 residues) (Fig. 1B and Table 2). The overall sequence identity of S. pombe Rpb4 with the corresponding subunits from humans, A. thaliana, and S. cerevisiae (including the extra sequences) is 36, 31, and 26%, respectively.

TABLE 2.

Subunit composition of the S. pombe RNA polymerase II

Gene product	No. of amino acid residues	Protein mass (kDa)	Gene location (chromosome no.)	No. of amino acid residues (sequence identity [%])
				S. cerevisiaea	H. sapiensb	A. thalianac
Rpb1	1,752	194	2	1,733 (63)	1,970 (53)	1,841 (56)
Rpb2	1,210	138	1	1,224 (68)	1,174 (60)	1,188 (62)
Rpb3	297	34	3	318 (48)	275 (36)	319 (30)
Rpb4	135	15	2	221 (26)	142 (36)	138 (31)
Rpb5	210	24	1	215 (57)	210 (44)	205 (42)
Rpb6	142	16	3	155 (56)	127 (49)	144 (44)
Rpb7	172	19	1	171 (54)	172 (48)	176 (46)
Rpb8	125	14	2	146 (39)	150 (36)	146 (32)
Rpb9	113	13	nd	122 (47)	125 (44)	134 (41)
Rpb10	71	8.3	1	70 (72)	67 (68)	71 (70)
Rpb11	123	14	1	120 (44)	117 (43)	116 (39)
Rpb12	63	7.2	2	70 (39)	58 (46)	51 (44)

^aSubunit composition data from references 26 and 27. 

^bSubunit composition data from reference 10. 

^cSubunit composition data from reference 16. 

The S. cerevisiae RPB4 contains four segments of the extra sequence, two N-terminal proximal segments (residues 7 to 14 and 36 to 43 on the S. cerevisiae sequence) and two segments (residues 72 to 98 and 108 to 138) in the central portion, which are not present in the RPB4 homologue from other organisms (Fig. 1B). The highly charged S. cerevisiae-specific sequence in the central portion is similar, in part, with the E. coli ς⁷⁰ subunit (28). As in the case of both human and plant subunit 4, the S. pombe Rpb4 lacks these S. cerevisiae-specific sequences (Fig. 1B). Nevertheless, hRPB4 is able to complement, albeit at a low efficiency, the altered phenotype of S. cerevisiae rpb4 mutant (10). Together, the results indicate that the S. cerevisiae RPB4-specific sequences are not essential for the subunit 4 function in other organisms and that the S. pombe Rpb4 is more similar, in its structure, to the RPB4 homologues from higher eukaryotes.

Identification of Rpb4 in S. pombe RNA polymerase II.

Previously, we failed to detect Rpb4 in our RNA polymerase II preparations by microsequencing of proteolytic fragments of the subunits separated by SDS-PAGE (21, 24). Since we identified the rpb4 gene sequence in the S. pombe genome, we reexamined whether the rpb4 gene product was expressed in S. pombe and assembled into the RNA polymerase II. For this purpose, the rpb4 cDNA was expressed in E. coli as a fusion with a sequence for His₆ tag, and the putative Rpb4 protein was purified by affinity chromatography. The purified His₆-tagged Rpb4 was used to raise polyclonal antibodies in rabbits. As a test sample, the RNA polymerase II was purified from growing cells of S. pombe by our standard procedure (2).

When the crude enzyme preparation after MonoQ chromatography, which contained more than 20 stained protein bands after SDS-PAGE, was analyzed by Western blotting, we detected an immunostained band with anti-Rpb4 antibodies, which migrated to the same position as Rpb8, Rpb9, and Rpb11 (Fig. 2A). Comigration of Rpb9 (113 amino acid residues) with Rpb8 (125 residues) and Rpb11 (123 residues) interfered with the detection of Rpb9 (23). Here, it became clear that the failure to detect Rpb4 (135 residues) was also caused by the comigration of Rpb4 with these three subunits. To confirm that the anti-Rpb4 antibodies used did not cross-react with other Rpb proteins, all four Rpb proteins were independently expressed in E. coli, the expressed cell lysates were fractionated by SDS-PAGE, and the gel was subjected to Western blot analysis using the anti-Rpb4 antibodies. As shown in Fig. 2B, no cross-reaction was observed between the anti-Rpb4 antibodies and the Rpb8, Rpb9, and Rpb11 proteins.

FIG. 2.

Identification of Rpb4 in purified S. pombe RNA polymerase II. (A) RNA polymerase II was purified by the standard procedure. The crude enzyme preparation after MonoQ chromatography was separated by SDS-PAGE and subjected to Western blotting with anti-Rpb7CH (lane 1), anti-Rpb8CH (lane 2), anti-Rpb9CH (lane 3), anti-Rpb4CH (lane 4), and anti-Rpb11CH (lane 5) antibodies. (B) Cell lysates of E. coli overexpressing the Rpb proteins (lanes 1 to 4, Rpb8, Rpb9, Rpb4, and Rpb11, respectively) were separated by SDS-PAGE, and the gel was subjected to either immunostaining against the anti-Rpb4 antibodies or protein staining with Coomassie brilliant blue (CBB). (C) The MonoQ fraction of RNA polymerase II was further purified by gel filtration chromatography on a Superose 6 column. Aliquots from Superose 6 fractions (lanes 1 to 10) and an aliquot of the Ni²⁺-affinity-purified RNA polymerase II from a S. pombe strain expressing His₆-tagged Rpb3 (lane 11) were separated by SDS-PAGE. The gel was subjected to Western blotting with anti-Rpb4CH (top) and anti-GST-Rpb3 (bottom) antibodies. (D) The purified RNA polymerase II (Pol II) (Superose 6 fraction) was fractionated by SDS-PAGE (lane 4), and the gel was subjected to quantitative immunoblot analysis using anti-Rpb7CH (top) and anti-Rpb4CH antibodies (bottom). For quantitation, various amounts of the purified Rpb4 and Rpb7CH were analyzed in parallel (10 [lane 1], 20 [lane 2], and 50 [lane 3] ng).

Western blot analysis was also performed for an RNA polymerase II preparation at a later step of purification, which was obtained after Superose 6 gel filtration chromatography of the MonoQ fraction. The Rpb4 protein showed the same elution pattern as both the Rpb3 protein (Fig. 2C) and the RNA polymerase activity (data not shown). To confirm these observations, we also tested two different preparations of the RNA polymerase II obtained from different S. pombe strains and by different procedures. One enzyme preparation was purified by glutathione-Sepharose column chromatography from a S. pombe strain carrying the gene coding for a glutathione S-transferase (GST)-Rpb3 fusion in place of the wild-type rpb3 gene (13). The other enzyme preparation was obtained by Ni²⁺-affinity chromatography from a S. pombe strain in which the rpb3 gene was replaced by the gene encoding Rpb3 fused with the His₆ tag (12). We detected the Rpb4 band for both of the RNA polymerase preparations (see Fig. 2C, lane 11, for the RNA polymerase II containing His₆-tagged Rpb3) (the complete set of subunits including Rpb4 was also detected for the RNA polymerase preparation containing the GST-Rpb3 fusion [13]). Thus, we conclude that Rpb4 is indeed associated with the S. pombe RNA polymerase II.

Stoichiometry of the Rpb4 in RNA polymerase II.

The fraction of S. cerevisiae RNA polymerase II containing RPB4 is only about 20% in cells in the exponential growth phase but it gradually increases in the postexponential phase (4, 14). In addition, S. cerevisiae RPB4 is easily dissociated, together with RPB7, from the RNA polymerase II at least under various in vitro situations (5, 6, 20). On the other hand, in humans and plants, subunits 4 and 7 are associated with RNA polymerase II more tightly than in S. cerevisiae even in the exponential growth phase (10, 15). We then tried to determine the stoichiometry of Rpb4 in the RNA polymerase from S. pombe. For this purpose, a quantitative immunoblot analysis was performed with RNA polymerase II, which was purified from exponentially growing cells (5 × 10⁷ cells/ml) by our standard procedure (2). For quantification, we used Rpb7 as a reference, because the stoichiometric amount of Rpb7 associates with the purified S. pombe RNA polymerase II (24). The result, shown in Fig. 2D, indicates that there is as much Rpb4 is as Rpb7.

Disruption of the rpb4 gene in S. pombe.

RPB4 of S. cerevisiae is not essential for cell growth but is required for viability under certain stress conditions (4, 28). To determine whether Rpb4 is essential for cell viability of S. pombe, we constructed a diploid strain carrying one disrupted copy of rpb4 as described in Materials and Methods. The rpb4/rpb4::ura4 cells were sporulated and subjected to tetrad analysis. Of 23 tetrads dissected, 0 viable, 1 viable, and 2 viable spores were observed in 3, 4, and 16 tetrads, respectively, and no more than 2 viable spores were observed (Fig. 3). Moreover, all the 36 viable spores were Ura⁻, indicating that rpb4::ura4 spores are not viable. Microscopic observation revealed that rpb4::ura4 spores germinated but ceased growth after a few divisions. These results demonstrate that the S. pombe rpb4 gene, unlike its S. cerevisiae counterpart, is essential for cell growth.

FIG. 3.

Tetrad analysis of an rpb4 heterozygous diploid. Diploid cells carrying one disrupted copy of rpb4 were sporulated on ME medium at 27°C for 2 days, and tetrads were dissected on YE medium containing adenine and uracil and allowed to grow at 30°C for 3 days. Seven tetrads are shown; the four spores from each tetrad are aligned vertically.

Interaction between Rpb4 and Rpb7.

RPB4 and RPB7 from both S. cerevisiae form a stable binary complex (6). The RPB4 and RPB7 homologues from both humans and A. thaliana were also shown to form heterodimers (10, 15). In order to test whether S. pombe Rpb4 and Rpb7 form a stable heterodimer, we expressed both Rpb4 and Rpb7CH in the same E. coli cells and isolated an Rpb7CH complex(es) by Ni²⁺-affinity chromatography. As shown in Fig. 4A, not only Rpb7CH but also Rpb4 bound to the Ni²⁺-agarose resin, indicating that Rpb4 and Rpb7CH formed a stable complex(es). To confirm the formation of Rpb4-Rpb7 complexes, the Ni²⁺-agarose fractions containing both Rpb4 and Rpb7CH were fractionated by gel filtration chromatography. As shown in Fig. 4B, the Rpb4 and Rpb7CH subunits were coeluted at fractions with an estimated molecular mass of 37 kDa. From the sizes of the two proteins (Rpb4, 16 kDa; Rpb7CH, 22 kDa), we conclude that one molecule each of the two subunits forms the heterodimer. The apparent difference in the staining intensities of the two proteins, Rpb4 and Rpb7, in the equimolar heterodimeric complex is due to the strong binding of the dye, Coomassie brilliant blue, to Rpb4.

FIG. 4.

Formation of complexes of Rpb4 or RPB4 and Rpb7 or RPB7. Two species of the RNA polymerase II subunit were coexpressed in E. coli in the following four combinations: S. pombe Rpb4-S. pombe Rpb7CH, S. cerevisiae RPB4-S. cerevisiae RPB7CH, S. cerevisiae RPB4-S. pombe Rpb7CH, and S. pombe Rpb4-S. cerevisiae RPB7CH, (which are shown as Sp4/Sp7H, Sc4/Sc7H, Sc4/Sp7H, and Sp4/Sc7H over the lanes in panel A). (A) Crude cell extracts were applied to a Ni²⁺-agarose column. Proteins bound were eluted with an elution buffer containing imidazole. Aliquots of the loading fraction, the flowthrough (FT) fraction, and the column-bound fraction were analyzed by SDS-PAGE, and the gel was stained with Coomassie brilliant blue. The migration positions of molecular mass markers are shown on the left. The positions of Rpb4, RPB4, Rpb7CH (Rpb7H), and RPB7CH (RPB7H) from S. pombe (Sp) or S. cerevisiae (Sc) are indicated on the right (note that the migration of S. pombe Rpb4 is faster than Rpb7). (B) The Ni²⁺-agarose elution fractions containing the binary complexes were subjected to gel filtration chromatography on a Superdex 75 PC 3.2/30 column. Aliquots from the fractions were analyzed by SDS-PAGE, and the gel was stained with Coomassie brilliant blue. Fraction numbers are shown at the top, and the peak positions of molecular marker proteins, fractionated on the same column, are indicated at the bottom. The migration positions of Rpb4, RPB4, Rpb7CH, and RPB7CH are indicated on the right.

Human RPB4 homologue does not interact with S. cerevisiae RPB7 as detected by the yeast two-hybrid system but is able to rescue an S. cerevisiae rpb4 disruptant partially (10). We then tested whether chimeric heterodimers are formed between S. pombe and S. cerevisiae. Pairs of two subunits in heterologous combinations, i.e., S. pombe Rpb4 plus S. cerevisiae RPB7CH and S. cerevisiae RPB4 plus S. pombe Rpb7CH were simultaneously expressed in E. coli. Expressed cell lysates were subjected to Ni²⁺-affinity chromatography (Fig. 4A), and the column-bound Rpb7CH or RPB7CH complexes were eluted and fractionated by gel filtration chromatography (Fig. 4B). In both heterologous combinations, the chimeric heterodimers, RPB4-Rpb7CH and Rpb4-RPB7CH, were formed as in the case of homologous combinations, RPB4-RPB7CH and Rpb4-Rpb7CH (Fig. 4B). The elution positions of the heterodimers, except for the S. pombe Rpb4-S. cerevisiae RPB7CH hybrid dimer, are close to those expected based on the assumption that the dimers are composed of one molecule each of the subunits in the dimer. The conformation of the Rpb4-RPB7CH hybrid dimer may be different from those of other dimers. These observations indicate that the extrasequence segments present only in the S. cerevisiae RPB4 are not essential for the dimerization with RPB7 and, moreover, that the S. pombe Rpb4 is closer, in its structure and essential nature in cell viability, to the corresponding subunits of higher eukaryotes, but closer in certain molecular functions, to the S. cerevisiae RPB4 than the RPB4 homologues from higher eukaryotes.

DISCUSSION

Until recently, the subunit 4 (RPB4) of RNA polymerase II was identified only in S. cerevisiae (29). The stoichiometry of RPB4 in the S. cerevisiae RNA polymerase II, however, changes, depending on the growth conditions. In growing cells, the fraction of RNA polymerase II containing RPB4 is about 20% (4, 14), but in the stationary phase, virtually all RNA polymerase II molecules contain Rpb4 (4). In concert with these observations, RPB4 is not essential for the viability of S. cerevisiae (28), and under optimal growth conditions at moderate temperatures, the S. cerevisiae mutant lacking RPB4 can grow, albeit at lower rates than the wild-type counterpart (4). The RPB4 mutant is, however, not viable under various stress conditions such as upon exposure to heat shock (4, 28) or under nutrient starvation at moderate temperatures (4). Overproduction of RPB4 results in almost twofold increase in the growth rate only when the cells were growing slowly (3). In agreement with these in vivo observations, RPB4 is required for the RNA polymerase activity in vitro only at extreme temperatures (19). The RNA polymerase II purified from cells lacking the RPB4 gene is catalytically active in RNA synthesis but is deficient in selective transcription initiation in vitro from certain promoters (6). These observations altogether indicate that in the case of S. cerevisiae, RPB4 is required for functional modulation of the RNA polymerase under certain stress conditions.

The RNA polymerase II purified from an S. cerevisiae mutant lacking the RPB4 gene lacks both RPB4 and RPB7 (14), suggesting that RPB7 interacts with RPB4 to associate with the RNA polymerase. Both RPB4 and RPB7 can be dissociated as a complex from the S. cerevisiae RNA polymerase II when the enzyme is chromatographed on an anion-exchange resin in the presence of 1.2 to 2.0 M urea (6, 20) or when the enzyme is subjected to native gel electrophoresis in the absence of protein denaturants (5). The phenotype of the RPB4 mutant reflects the involvement of the RPB4-RPB7 complex in the modulation of the activity or specificity of RNA polymerase II. Unlike S. cerevisiae RPB4 and RPB7, the corresponding subunits 4 and 7 are more stably associated with animal and plant RNA polymerase II (10, 15). Based on in vitro transcription experiments, RPB4 and RPB7 are known to be dispensable for promoter-independent or nonspecific transcription initiation and RNA chain elongation (6, 20). However, these two subunits are required for promoter-dependent or -specific initiation (6). Thus, RPB4 was once thought to be an accessory protein with no essential role in the RNA polymerase II functions. The association of RPB4 with the other RNA polymerase II subunits may modulate its specificity as to increase the tolerance against various stresses.

Identification of Rpb4 in S. pombe, plants, and animals categorizes the conserved nature of RNA polymerase II subunits in three eukaryotic kingdoms, plants, animals, and fungi. Sequences of the cloned subunits and ESTs suggest that plant, animal, and fission yeast RNA polymerase II enzymes contain 12 subunits that are related to the 12 subunits originally identified in S. cerevisiae (10, 16) (Table 2). The sequence comparison indicated that the S. pombe Rpb4 protein is similar to the corresponding subunits from humans and plants in structure in the following ways. (i) The size of subunit 4 from higher eukaryotes and S. pombe is 60 to 65% of the size of S. cerevisiae RPB4. (ii) The RPB4 homologues from higher eukaryotes and S. pombe lack several segments of the sequence which are present in the S. cerevisiae RPB4, including the highly charged ς⁷⁰-like sequence in the central portion of RPB4 (28).

In concert with the structural difference, the RPB4 homologues from higher eukaryotes and S. pombe are different from the S. cerevisiae RPB4 in function in the following ways. (i) RPB4 is weakly associated with the S. cerevisiae RNA polymerase II, while the RPB4 homologues are tightly associated with the RNA polymerase II from S. pombe (this report) and higher eukaryotes (10, 15). (ii) The stoichiometric amounts of RPB4 homologues are bound with the RNA polymerase II from S. pombe and higher eukaryotes, while the S. cerevisiae RNA polymerase II fraction containing RPB4 is only about 20% of the total RNA polymerase II from the growing cells. (iii) S. pombe Rpb4 is essential for cell growth as analyzed by gene disruption (this report).

The RPB4 homologue from humans does not form stable complexes with the S. cerevisiae RPB7, as evidenced by the yeast two-hybrid assay (10). In agreement with this observation, the human RPB4 homologue is able to substitute only partially for the S. cerevisiae RPB4 as detected by the complementation assay (10). In sharp contrast with the human RPB4 homologue, however, the hybrid dimers are formed in heterologous combinations between S. pombe Rpb4 and S. cerevisiae RPB7 and between S. cerevisiae RPB4 and S. pombe Rpb7, indicating that the S. pombe Rpb4 is closer in function to the S. cerevisiae RPB4. This finding also imply that the extrasequence segments present only in the S. cerevisiae RPB4 (Fig. 1) are not involved in the RPB4-RPB7 dimerization. Instead, an as-yet-unidentified structural element, which is present in both S. pombe Rpb4 and S. cerevisiae RPB4 but absent in the RPB4 homologues from higher eukaryotes, may be involved in the dimerization.

RPB4 is present in the S. cerevisiae cells in excess over the RNA polymerase II during all growth phases, but its affinity to the RNA polymerase II is weak in the exponential growth phase (19). The defective activity of the RPB4-depleted S. cerevisiae mutant cell extract at the nonoptimal temperature can be rescued by the addition of RPB4 subunit from the heated RPB1 mutant cell extract (19). However, this in vitro complementation can be observed only when the RPB4 mutant extract was prepared from the stationary-phase cells. Thus, it appears that the S. cerevisiae RNA polymerase II is modified in the stationary phase so as to recruit the RPB4 subunit. The RNA polymerase II from animals, plants, and the fission yeast might be fixed at the form which is capable of accepting subunit 4. The S. cerevisiae RNA polymerase II, saturated with RPB4, from cells under stress conditions forms high-quality two-dimensional crystals (1, 9). The S. pombe RNA polymerase II, which is tightly associated with Rpb4 even under normal growth conditions, could be a good tool for analysis of three-dimensional structure of the RNA polymerase II.