Genetic, Physical, and Functional Interactions between the Triphosphatase and Guanylyltransferase Components of the Yeast mRNA Capping Apparatus

Abstract

We have characterized an essential Saccharomyces cerevisiae gene, CES5, that when present in high copy, suppresses the temperature-sensitive growth defect caused by the ceg1-25 mutation of the yeast mRNA guanylyltransferase (capping enzyme). CES5 is identical to CET1, which encodes the RNA triphosphatase component of the yeast capping apparatus. Purified recombinant Cet1 catalyzes hydrolysis of the γ phosphate of triphosphate-terminated RNA at a rate of 1 s⁻¹. Cet1 is a monomer in solution; it binds with recombinant Ceg1 in vitro to form a Cet1-Ceg1 heterodimer. The interaction of Cet1 with Ceg1 elicits >10-fold stimulation of the guanylyltransferase activity of Ceg1. This stimulation is the result of increased affinity for the GTP substrate. A truncated protein, Cet1(201-549), has RNA triphosphatase activity, heterodimerizes with and stimulates Ceg1 in vitro, and suffices when expressed in single copy for cell growth in vivo. The more extensively truncated derivative Cet1(246-549) also has RNA triphosphatase activity but fails to stimulate Ceg1 in vitro and is lethal when expressed in single copy in vivo. These data suggest that the Cet1-Ceg1 interaction is essential but do not resolve whether the triphosphatase activity is also necessary. The mammalian capping enzyme Mce1 (a bifunctional triphosphatase-guanylyltransferase) substitutes for Cet1 in vivo. A mutation of the triphosphatase active-site cysteine of Mce1 is lethal. Hence, an RNA triphosphatase activity is essential for eukaryotic cell growth. This work highlights the potential for regulating mRNA cap formation through protein-protein interactions.

The 5′ cap structure of eukaryotic mRNA consists of 7-methylguanosine linked to the end of the transcript via a 5′-5′ triphosphate bridge. Capping occurs by a series of three enzymatic reactions in which the 5′ triphosphate end of nascent pre-mRNA is hydrolyzed to a 5′ diphosphate by RNA triphosphatase, capped with GMP by RNA guanylyltransferase, and then methylated at N7 of guanine by RNA (guanine-7) methyltransferase (21). RNA capping is essential for cell growth. Mutations of the guanylyltransferase or methyltransferase component of the yeast capping apparatus that abrogate catalytic activity are lethal in vivo (4, 10, 15, 30, 31).

A consilience of biochemistry, molecular genetics, and structural biology has illuminated the mechanism of cap guanylylation. Transfer of GMP from GTP to the 5′ diphosphate terminus of RNA occurs in a two-stage reaction involving a covalent enzyme-GMP intermediate (24). The GMP is linked to the enzyme through a phosphoamide (P—N) bond to the ɛ-amino group of a lysine residue within a conserved motif, KxDG (motif I), found in all known cellular and DNA virus-encoded capping enzymes (3, 4, 14, 15, 20, 22, 25). Hakansson et al. (5) have determined the crystal structure of the Chlorella virus capping enzyme in the GTP-bound state and with GMP bound covalently. The protein consists of two domains with a deep cleft between them. The GTP binding site is composed of motif I and five other motifs (III, IIIa, IV, V, and VI) that are conserved in order and spacing among cellular and DNA virus capping enzymes (25). The ɛ-amino group of the active-site Lys in motif I is positioned near the α-phosphate of GTP. The crystal structure reveals a large conformational change in the GTP-bound enzyme, from an open to a closed state, that brings motif VI into contact with the β and γ phosphates of GTP and reorients the phosphates for in-line attack by the active-site Lys. When the crystal is soaked in manganese, guanylyltransferase reaction chemistry occurs in crystallo and the covalent enzyme-GMP intermediate is formed (5). However, only the enzyme in the closed conformation is reactive. This work suggests that conformational changes coordinate substrate entry, reaction chemistry, and product exit during catalysis. Extensive mutational analysis of the Saccharomyces cerevisiae guanylyltransferase confirms that residues which, in the crystal structure, make contact with GTP are essential for capping enzyme function in vivo (22, 31).

What remains largely unexplored is the issue of whether (and how) cap formation might be regulated during eukaryotic gene expression. The guanylyltransferase reaction is likely to be the rate-limiting step in the overall capping pathway (21). Potential regulatory events include (i) those that influence the targeting of the capping enzyme to the transcription apparatus and (ii) those that affect the activity of the capping enzyme. Recent studies suggest that the mammalian and fungal guanylyltransferases are targeted to nascent pre-mRNAs by direct binding to the carboxyl-terminal domain (CTD) of the largest subunit of RNA polymerase II (2, 7, 12, 35). The guanylyltransferase-CTD interaction is contingent on CTD phosphorylation; hence, capping might well be regulated by the dynamics of CTD phosphorylation and dephosphorylation during transcription elongation.

The catalytic activity of the cap-forming enzymes might be modulated by (i) changes in the concentrations of substrates (GTP, S-adenosylmethionine) or inhibitory reaction products (PP_i, S-adenosylhomocysteine), (ii) posttranslational modifications of the enzymes, or (iii) protein-protein interactions that affect enzyme activity. There is a clear precedent for protein-mediated stimulation of cap methylation by the vaccinia virus capping enzyme, whereby the low basal methyltransferase activity of the 95-kDa catalytic subunit is stimulated 50- to 100-fold by its association with the 33-kDa subunit (6, 11).

We have undertaken a genetic analysis of cap formation and cap function in the budding yeast S. cerevisiae, focusing on the CEG1 gene encoding RNA guanylyltransferase. We previously reported the isolation of a collection of temperature-sensitive (ts) guanylyltransferase mutants and the identification of allele-specific multicopy suppressors of the ceg1-ts growth defects (16). We reasoned that capping enzyme suppressor (CES) genes might encode proteins that either interact with Ceg1 or impact on cap-dependent transactions in vivo. Four suppressor genes were identified. CES1 and CES4 encode homologous, functionally redundant proteins that regulate cell morphology and bud formation (1, 16, 17, 34); CES3 is identical to BUD5, which encodes a guanine nucleotide exchange factor involved in bud site selection; and CES2 is identical to ESP1, a gene required for proper nuclear division. The outcome of the initial suppressor screen was remarkable for its suggestion of a genetic nexus between capping and budding and for the absence of straightforward links between the suppressor gene products and mRNA metabolism. Two other CES genes (CES5 and CES6) were isolated in the high-copy suppressor screen but had not been mapped to single open reading frames.

Here we present the identification and characterization of capping enzyme suppressor CES5, which corresponds to open reading frame YPL228W on chromosome XVI. CES5 is an essential gene that encodes a 549-amino-acid polypeptide. While our analysis of CES5 was in progress, Mizumoto and colleagues (27) reported that YPL228W (named CET1 by them) encodes the yeast RNA triphosphatase. We show that purified recombinant Cet1 is a monomeric protein with intrinsic, magnesium-dependent RNA triphosphatase activity. Cet1 binds directly to Ceg1 in solution to form a bifunctional Cet1-Ceg1 heterodimer that we have isolated by glycerol gradient sedimentation. Remarkably, the binding of Cet1 to Ceg1 stimulates the guanylyltransferase activity of Ceg1. These and other results suggest a complex interplay between the triphosphatase and guanylyltransferase components of the capping apparatus.

MATERIALS AND METHODS

Isolation of CES5—a multicopy suppressor of ceg1-25.

Yeast strain YBS2 (MATa leu2 lys2 trp1 ceg1::hisG), bearing the temperature-sensitive ceg1-25 allele on a CEN TRP1 plasmid, was transformed with a yeast genomic DNA library in vector YEp24 (2μ URA3). Approximately 25,000 Ura⁺ transformants were plated on medium lacking uracil at 37°C. The yeast 2μ plasmid was isolated from 10 colonies that grew at 37°C and then transformed into Escherichia coli. Plasmids were prepared from cultures of individual ampicillin-resistant transformants. The DNAs were digested with EcoRI and XhoI to weed out 2μ plasmids containing a genomic insert of the wild-type CEG1 gene. Candidate suppressor clones were retested by transformation into the ceg1-25 strain used for the original selection. One of the plasmids contained CES5, a novel suppressor gene distinct from CES1, CES2, CES3, and CES4, which were described previously (16).

Cloning of CES5.

Limited sequencing of the vector insert junction of the 2μ-CES5 isolate (named pCES5-10.4) indicated that it contained a 10.4-kbp fragment of yeast chromosome XVI from coordinate 116607 to coordinate 127005. A subclone containing the YPL228W open reading frame was constructed by excising a 2.6-kb DNA fragment obtained by digestion of the pCES5-10.4 clone with SphI and HindIII. This restriction fragment, which contained the entire YPL228W coding region plus 886 bp upstream of the start codon and 82 bp downstream of the stop codon, was inserted into YEp24 to yield pCES5-2.6.

Gene disruption.

The chromosomal YPL228W gene was disrupted by insertion of a LEU2 marker as follows. (i) We first constructed pUC18-based plasmid pUC-CES5-3.6, containing a 3.6-kb SphI/SacI fragment extending from 886 bp upstream of the YPL228W start codon to 1,004 bp downstream of the stop codon. (ii) The LEU2 gene was then inserted between the AccI and BsmI sites to yield plasmid pΔces5, in which the YPL228W coding sequence had a deletion from amino acid 200 to amino acid 484. (iii) Linearized pΔces5 was transformed into haploid strain HW-1A (MATa trp1-1 his3-11,15 ura3-1 leu2-3,11 ade2-1 can1-100) that contained plasmid p360-CES5-3.6 (CEN URA3 CES5). (iv) Leu⁺ transformants were selected, and correct insertion of LEU2 into the chromosomal CES5 locus was confirmed by Southern blotting. ces5::LEU2 strain YBS20 could not grow in the presence of 0.75-mg/ml 5-fluoroorotic acid (5-FOA); i.e., cell growth was contingent on maintenance of the CES5 gene on the CEN URA3 plasmid.

Deletion mutants.

A CEN TRP1 plasmid for expression of the full-length YPL228W polypeptide in yeast was constructed in three stages. First, we cloned into pSE-358 (CEN TRP1) a fragment of yeast genomic DNA extending from 886 bp upstream of the translation start codon of CES5 to the start codon; in doing so, we introduced an NdeI site at the start codon and a BamHI site immediately 3′ of the NdeI site. The resulting plasmid was named pCES5-5′. Second, we cloned into the BamHI site of pCES5-5′ a 491-bp fragment comprising the genomic DNA sequence immediately 3′ of the translation stop codon of YPL228W, thereby generating plasmid pCES5-5′3′. Third, we cloned into pCES5-5′3′ a PCR-amplified DNA fragment containing the complete YPL228W coding sequence; NdeI and BamHI sites were introduced at the translation start codon and 3′ of the stop codon, respectively, during PCR. Also, an NdeI site within the open reading frame was destroyed during PCR without affecting the encoded amino acid sequence. The CEN TRP1 CES5 plasmid was named p358-CES5. N-terminal CES5 deletion mutants were constructed by PCR amplification with mutagenic sense strand primers that introduced an NdeI restriction site at the Met-96 codon or an NdeI restriction site and a methionine codon in lieu of the codons for Asn-148, Glu-200, Pro-245, and Leu-300. The PCR products were digested with NdeI and BamHI and then inserted into pCES5-5′3′. The mutated genes were named according to the amino acid coordinates of their polypeptide products, i.e., CES5(96-549), CES5(149-549), CES5(201-549), CES5(246-549), and CES5(301-549). Carboxyl-terminal truncation mutants were constructed by PCR amplification by using antisense primers that introduced translation stop codons in lieu of the codons for Ile-520 and Thr-490 and BamHI sites immediately 3′ of the new stop codons. The PCR products were digested with NdeI and BamHI and then inserted into pCES5-5′3′ to yield plasmids p358-CES5(1-519) and p358-CES5(1-489).

Protein expression and purification.

A 1.7-kbp NdeI/BamHI fragment containing the YPL228W gene was excised from p358-CES5 and inserted into bacterial expression plasmid pET16b (Novagen) to form plasmid pET-CET1. In this context, the yeast polypeptide is fused in frame with a 20-amino-acid N-terminal leader peptide containing 10 tandem histidines and expression of the His-tagged protein is driven by a T7 RNA polymerase promoter. pET-CET1 was transformed into E. coli BL21(DE3). A 100-ml culture of E. coli BL21(DE3)/pET-CET1 was grown at 37°C in Luria-Bertani medium containing 0.1-mg/ml ampicillin until the A₆₀₀ reached 0.5. The culture was adjusted to 0.4 mM isopropyl-β-d-thiogalactopyranoside (IPTG), and incubation was continued at 37°C for 3 h. Cells were harvested by centrifugation, and the pellet was stored at −80°C. All subsequent procedures were performed at 4°C. Thawed bacteria were resuspended in 5 ml of buffer A (50 mM Tris HCl [pH 7.5], 0.15 M NaCl, 10% sucrose). Lysozyme was added to a final concentration of 50 μg/ml, and the sample was sonicated for 30 s. Triton X-100 was added to a 0.1% final concentration. Sonication was repeated, and insoluble material was removed by centrifugation for 45 min at 18,000 rpm in a Sorvall SS34 rotor. The soluble extract was applied to a 1-ml column of Ni-nitrilotriacetic acid-agarose (Qiagen) that had been equilibrated with buffer A containing 0.1% Triton X-100. The column was washed with the same buffer and then eluted stepwise with buffer B (50 mM Tris HCl [pH 8.0], 0.1 M NaCl, 10% glycerol) containing 50, 100, 200, 500, and 1,000 mM imidazole. The polypeptide composition of the column fractions was monitored by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE). The recombinant yeast protein was retained on the column and recovered in the 200 mM imidazole eluate. This fraction was dialyzed against buffer C (50 mM Tris HCl [pH 8.0], 50 mM NaCl, 2 mM dithiothreitol [DTT], 10% glycerol, 0.05% Triton X-100), and the dialysate was applied to a 1-ml column of phosphocellulose that had been equilibrated in buffer C. The column was washed with the same buffer and then eluted stepwise with buffer C containing 0.1, 0.2, 0.5, and 1.0 M NaCl. The recombinant protein was retained on the column and was recovered predominantly in the 0.1 M NaCl fraction. The phosphocellulose preparation was stored at −80°C.

Expression of truncated proteins.

The NdeI/BamHI fragments were excised from p358-CES5(201-549), p358-CES5(246-549), and p358-CES5(301-549) and inserted into pET16b to generate plasmids pET-CET1(201-549), pET-CET1(246-549), and pET-CET1(301-549), respectively. Induced expression of the His-tagged, truncated Cet1 proteins was performed as described above for full-length Cet1. The Cet1(201-549) and Cet1(246-549) polypeptides were purified from soluble bacterial lysates by Ni-agarose and phosphocellulose column chromatography as described above for Cet1. The Cet1(301-549) protein was expressed in bacteria but was recovered exclusively in the insoluble pellet fraction of the cell lysate.

RNA triphosphatase assay.

RNA triphosphatase activity was assayed by liberation of ³²P_i from γ-³²P-labeled triphosphate-terminated poly(A) (23). Standard reaction mixtures (10 μl) containing 50 mM Tris HCl (pH 7.5), 5 mM DTT, 1 mM MgCl₂, 20 pmol (of triphosphate termini) of γ-³²P-labeled poly(A), and enzyme as specified were incubated for 15 min at 30°C. Aliquots of the mixtures were applied to a polyethyleneimine-cellulose thin-layer chromatography plate which was developed with 0.75 M potassium phosphate (pH 4.3). The release of ³²P_i from γ-³²P-labeled poly(A) was quantitated by scanning the thin-layer chromatography plate with a FUJIX BAS2000 Bio-Imaging Analyzer.

Enzyme-GMP complex formation.

Standard reaction mixtures (20 μl) containing 50 mM Tris HCl (pH 8.0), 5 mM DTT, 5 mM MgCl₂, 0.17 μM [α-³²P]GTP, and enzyme were incubated for 10 min at 37°C. The reactions were halted by addition of SDS to a 1% final concentration. The samples were electrophoresed through a 12% polyacrylamide gel containing 0.1% SDS. Enzyme-[³²P]GMP complexes were visualized by autoradiographic exposure of the dried gel and quantitated by scanning the gel with a FUJIX BAS2000 Bio-Imaging Analyzer.

RESULTS

Isolation and characterization of CES5—a multicopy suppressor of ceg1-25.

The high-copy suppressor screen entailed transformation of ceg1-25 with a 2μ plasmid-based wild-type genomic DNA library and selection for Ura⁺ colonies that grew at the nonpermissive temperature (37°C). Plasmid DNA was recovered from individual yeast colonies and transformed into E. coli. Diagnostic restriction enzyme digestion revealed whether wild-type CEG1 had been selected. Candidate suppressors that did not contain the CEG1 gene were retransformed into the ceg1-25 strain. Three genomic clones retested faithfully. These clones derived from three distinct genetic loci, which we named CES1, CES2, and CES5 (CES stands for capping enzyme suppressor). CES1 and CES2 were characterized previously (16, 17). Here, we present a genetic and biochemical analysis of CES5.

Multicopy suppression by CES5 is shown in Fig. 1. Serial 10-fold dilutions of ceg1-25 cells were plated at 25 and 37°C. ceg1-25 cells transformed with the YEp24 vector grew at 25°C but not at 37°C; cells transformed with the wild-type CEG1 gene grew well at both temperatures. The selected 2μ CES5 clone restored growth at the restrictive temperature. The suppressor clone contained a 10.4-kbp insert from yeast chromosome XVI that included one known gene (ALG5) and four open reading frames that had not yet been characterized (Fig. 1). We found that a 2.6-kb subclone containing the YPL228W open reading frame (encoding a 549-amino-acid polypeptide) was as effective as the original clone in suppressing ceg1-25 (Fig. 1). We concluded that YPL228W corresponds to CES5.

FIG. 1.

CES5 is a multicopy suppressor of ceg1-25. ceg1-25 cells were transformed with a 2μ-URA3 plasmid containing the wild-type CEG1 gene, with the 2μ URA3 CES5 plasmid [pCES5(10.4)] that was isolated in the multicopy suppressor screen, with pCES5(2.6 kb) containing the YPL228W reading frame, and with the YEp24 vector plasmid without any insert. Ura⁺ transformants were selected and grown at 25°C in liquid culture in medium lacking uracil. The cultures were adjusted to an A₆₀₀ of 0.1, and aliquots of serial 10-fold dilutions were spotted on agar medium lacking uracil. The plates were photographed after incubation for 3 days at either 25 or 37°C. aa, amino acids.

YPL228W was disrupted by replacement of the coding sequence from amino acid 201 to amino acid 483 with a LEU2 marker gene. The disruption was performed in a haploid strain containing CES5 on a CEN URA3 plasmid. Correct insertion of LEU2 into the chromosomal locus was confirmed by Southern blotting. The ces5::LEU2 strain was unable to grow on medium containing 5-FOA but was able to grow on 5-FOA after being transformed with a CEN TRP1 CES5 plasmid. Thus, CES5 is an essential gene.

A series of truncated versions of CES5 were tested for their function in vivo by using the plasmid shuffle assay. The truncated genes were cloned into CEN TRP1 plasmids; expression was driven by the CES5 promoter. N-terminal deletion mutants CES5(96-549), CES5(149-549), and CES5(201-549) complemented growth of the ces5::LEU2 strain on 5-FOA. However, the more extensive deletions CES5(246-549) and CES5(301-549) were lethal. Carboxyl truncation mutants CES5(1-519) and CES5(1-489) were also lethal (summarized in Fig. 2A). We conclude that the N-terminal 200 amino acids of Ces5 are not essential for its function in vivo.

FIG. 2.

Deletion analysis. (A) CEN TRP1 plasmids encoding full-length and truncated versions of the YPL228W protein were transformed into YBS20. Individual Trp⁺ transformants were selected and patched to plates lacking tryptophan. Cells were streaked on plates containing 0.75-mg/ml 5-FOA. The plates were incubated at 25 and 30°C. Alleles that supported the formation of wild-type size colonies after 3 days on 5-FOA were scored as positive. Lethal truncation alleles (scored negative) were those that formed no colonies after 7 days at either temperature. The full-length and truncated polypeptides are depicted as horizontal bars with N termini at the left and C termini at the right. The region of homology between the YPL228W and YMR180C gene products is in black. (B) The YPL228W amino acid sequence between residues 302 and 532 is aligned with the homologous segment of the yeast YMR180C polypeptide (residues 84 to 310). Identical amino acids are indicated by colons, and conserved residues are denoted by periods. Discontinuities in the alignment are indicated by dashes.

A blastp search of the National Center for Biotechnology Information database with the YPL228W polypeptide revealed sequence similarity (score, 85) to the predicted 320-amino-acid polypeptide encoded by the S. cerevisiae YMR180C open reading frame on chromosome XIII (GenBank accession no. Z49808). The region of sequence similarity spans Ces5 residues 302 to 532 (within the essential C-terminal portion of the protein) and includes 68 identical and 40 similar amino acids (Fig. 2B). The function of YMR180C is unknown.

Allele specificity of multicopy suppression.

The YPL228W coding sequence was cloned into a 2μ URA3 vector to place the gene under the control of a strong constitutive TPI1 promoter. The 2μ TPI-CES5 plasmid suppressed the ceg1-25 ts growth defect (data not shown). The clone was then tested for suppression of nine other ceg1-ts mutations in our collection (16). CES5 suppressed ceg1-6 and, to a lesser extent, ceg1-7 but had no salutary effect on the growth of seven other ceg1 mutants (ceg1-1, ceg1-3, ceg1-5, ceg1-13, ceg1-17, ceg1-27, and ceg1-95) at 37°C (data not shown).

CES5 is identical to CET1 encoding RNA triphosphatase.

Tsukamoto et al. (27) recently identified the YPL228W open reading frame as the gene encoding the RNA triphosphatase component of the yeast mRNA capping apparatus. They named the gene CET1 (capping enzyme triphosphatase 1) and noted, as did we, that disruption of the gene was lethal. Henceforth, we shall adopt their nomenclature and refer to the suppressor gene CES5 as CET1.

Tsukamoto et al. (27) reported that recombinant Cet1 protein expressed in bacteria possessed RNA triphosphatase activity; they also found that soluble Ceg1 interacted with Cet1 immobilized on a membrane (a far-Western blot). We sought to characterize the catalytic activity of the recombinant Cet1 protein and to assess the stoichiometry and functional properties of the Cet1-Ceg1 complex. These issues gain currency in light of the present genetic evidence that overexpression of Cet1 suppresses the growth phenotype caused by mutation of Ceg1.

We expressed Cet1 in E. coli under the control of an inducible T7 RNA polymerase promoter. To facilitate purification, Cet1 was fused to an N-terminal leader peptide containing 10 tandem histidines. The protein was purified from a soluble lysate of induced bacteria by Ni-agarose chromatography. We noted, as did Tsukamoto et al. (27), that recombinant Cet1 migrated anomalously during SDS-PAGE; its apparent size of 80 kDa relative to coelectrophoresed marker polypeptides was greater than the actual size of 62 kDa calculated from the amino acid sequence of the recombinant polypeptide. The Ni-agarose preparation contained Cet1, as well as an array of smaller polypeptides (Fig. 3A, lane Ni). Cet1 was purified away from the smaller contaminants by phosphocellulose chromatography (Fig. 3A). The 0.1 M NaCl phosphocellulose eluate was highly enriched with respect to Cet1.

FIG. 3.

Purification and RNA triphosphatase activity of Cet1. (A) The elution profile of the Cet1 polypeptide during phosphocellulose column chromatography was analyzed by SDS-PAGE. Lanes: Ni, Ni-agarose eluate fraction applied to the phosphocellulose column; F, phosphocellulose flowthrough fraction; W, 50 mM NaCl wash fraction; 0.1, 100 mM NaCl eluate; 0.2, 200 mM NaCl eluate; 0.5, 500 mM NaCl eluate; 1.0, 1.0 M NaCl eluate. A Coomassie blue-stained gel is shown. The values to the right are molecular sizes in kilodaltons. (B) RNA triphosphatase activity. Reaction mixtures (10 μl) containing 50 mM Tris HCl (pH 7.5), 5 mM DTT, 20 pmol (of triphosphate termini) of γ-³²P-labeled poly(A), either 1 mM MgCl₂ (+Mg) or no divalent cation (−Mg), and the indicated amounts of Cet1 (0.1 M NaCl phosphocellulose fraction) were incubated for 15 min at 30°C. P_i release is plotted as a function of input protein. (C) Magnesium titration. Reaction mixtures containing 20 pmol of γ-³²P-labeled poly(A), 0.5 ng of Cet1, and MgCl₂ as specified were incubated for 15 min at 30°C.

The phosphocellulose Cet1 fraction catalyzed the release of ³²P_i from γ-³²P-labeled, triphosphate-terminated poly(A). Activity was proportional to input enzyme, and the reaction proceeded to completion at a saturating enzyme concentration (Fig. 3B). We calculated that recombinant Cet1 released ∼1 fmol of P_i per s per fmol of enzyme. Activity was strictly dependent on inclusion of magnesium in the reaction mixture (Fig. 3B). Optimal activity was seen at 0.5 to 2 mM MgCl₂ (Fig. 3C).

To gauge the native size of the yeast RNA triphosphatase, recombinant Cet1 was centrifuged through a 15 to 30% glycerol gradient. The RNA triphosphatase activity profile coincided with the abundance of the Cet1 polypeptide (Fig. 4A). Cet1 sedimented as a single component of 4.3S relative to marker proteins sedimented in parallel. We conclude that Cet1 is a monomer in solution.

FIG. 4.

Analysis of Cet1-Ceg1 interaction by glycerol gradient sedimentation. Aliquots (0.2 ml) of protein samples were applied to 4.8-ml 15 to 30% glycerol gradients containing 50 mM Tris-HCl (pH 8.0), 0.1 M NaCl, 2 mM DTT, and 0.05% Triton X-100. The gradients were centrifuged at 50,000 rpm for 13 h at 4°C in a Beckman SW50 rotor. Fractions (∼0.21 ml) were collected from the bottom of the tube (fraction 1). Aliquots (25 μl) of alternate fractions were analyzed by SDS-PAGE along with an aliquot of the material that had been applied to the gradient (lane L). The gels were fixed and stained with Coomassie blue dye. The identities of the polypeptides are indicated. (A) Sedimentation of Cet1. A 10-μg sample of the phosphocellulose enzyme fraction was applied to the glycerol gradient. Gradient fractions were assayed for RNA triphosphatase (•). The RNA triphosphatase reaction mixtures contained 20 pmol of γ-³²P-labeled poly(A) and 1 μl of a 1/50 dilution of the indicated gradient fractions. Incubation was for 15 min at 30°C. (B) Sedimentation of Ceg1. A 20-μg sample of the heparin agarose Ceg1 fraction was applied to the gradient. Gradient fractions were assayed for enzyme-GMP complex formation (○). The guanylyltransferase reaction mixtures contained 0.17 μM [α-³²P]GTP and 1 μl of the indicated fractions. The reaction products were analyzed by SDS-PAGE, and the signal intensity of the radiolabeled Ceg1 polypeptide (PSL, photostimulatable luminescence) was measured by scanning the dried gel with a PhosphorImager. (C) Sedimentation of a mixture of Cet1 (10 μg) and Ceg1 (20 μg). The two proteins were mixed in buffer C containing 0.1 M NaCl and then incubated on ice for 30 min before being applied to the glycerol gradient. Gradient fractions were assayed for RNA triphosphatase (•) and enzyme-GMP complex formation (○) as specified for panels A and B. The peaks of the marker proteins catalase, bovine serum albumin (BSA), and cytochrome c (cyt C), which were centrifuged in a parallel gradient, are indicated.

Cet1 and Ceg1 form a heterodimeric capping enzyme complex in vitro.

To test whether Cet1 and Ceg1 interact in solution, recombinant Ceg1 protein was expressed in bacteria and purified by Ni-agarose and heparin agarose column chromatography. Guanylyltransferase activity was monitored during purification by formation of a 52-kDa ³²P-labeled covalent enzyme-GMP complex in the presence of [α-³²P]GTP and magnesium (15). Ceg1 itself sedimented as a monomer when centrifuged in a 15 to 30% glycerol gradient (Fig. 4B). The guanylyltransferase activity profile coincided with the abundance of the 52-kDa Ceg1 polypeptide (Fig. 4B).

Recombinant Cet1 (10 μg) was mixed with recombinant Ceg1 (20 μg) in buffer containing 0.1 M NaCl, and the mixture was analyzed by glycerol gradient sedimentation. Here, Cet1 and the RNA triphosphatase activity sedimented as a single discrete peak of 7.5S, whereas Ceg1 and guanylyltransferase activity sedimented as two discrete peaks, (i) a major 7.5S component cosedimenting with Cet1 and (ii) a less abundant, lighter component corresponding to monomeric Ceg1 (Fig. 4C). The polypeptide composition of the peak fractions of the Ceg1-Cet1 complex (gradient fractions 11 to 13) indicated that the two proteins were present in nearly equimolar amounts. We infer that Ceg1 and Cet1 interact in vitro to form a heterodimer.

We asked whether Cet1 would form a higher-order complex with the RNA guanylyltransferase domain of the mouse capping enzyme, which is structurally homologous to Ceg1 (7). Recombinant mouse guanylyltransferase, Mce1(211-597), was purified to homogeneity, mixed with Cet1 or with a buffer control, and then subjected to sedimentation analysis in parallel with the Cet1-Ceg1 mixtures. The 45-kDa Mce1(211-597) protein alone sedimented as a single monomeric peak (7). Sedimentation of the Mce1(211-597)-plus-Cet1 protein mixture revealed no shift in the distribution of the mouse guanylyltransferase or the yeast triphosphatase to a more rapidly sedimenting form, as gauged by SDS-PAGE analysis of the gradient fractions and by the enzyme activity profiles (not shown).

Characterization of Cet1(201-549) and Cet1(246-549).

His-tagged versions of truncated proteins Cet1(201-549), Cet1(246-549), and Cet1(301-549) were expressed in E. coli. Cet1(201-549) and Cet1(246-549) were purified from soluble bacterial lysates by Ni-agarose and phosphocellulose column chromatography. Cet1(301-549) was insoluble and therefore not amenable to purification. SDS-PAGE analysis of the phosphocellulose fractions of recombinant Cet1(201-549) (Fig. 5A, lane 2) and Cet1(246-549) (Fig. 5A, lane 3) revealed that the proteins were substantially pure and that they migrated more rapidly than full-size Cet1 (Fig. 5A, lane 1). The phosphocellulose preparations of Cet1(201-549) and Cet1(246-549) catalyzed the release of ³²P_i from γ-³²P-labeled, triphosphate-terminated poly(A). The specific activity of Cet1(201-549) was equivalent to that of Cet1, whereas Cet1(246-549) was 50% more active than full-length Cet1 (data not shown). The finding that the CET1(246-549) gene on a CEN plasmid could not support cell growth (Fig. 2), even though the gene product has full RNA triphosphatase activity in vitro, suggests that the catalytic activity of Cet1 may not suffice for Cet1 function in vivo.

FIG. 5.

Stimulation of Ceg1-GMP complex formation by Cet1 and Cet1(201-549). (A) Purification of Cet1(201-549) and Cet1(246-549). Aliquots (3 μg) of the phosphocellulose preparations of full-length Cet1 (lane 1), Cet1(201-549) (lane 2), and Cet1(246-549) (lane 3) were analyzed by SDS-PAGE. A Coomassie blue-stained gel is shown. The positions and sizes (in kilodaltons) of marker proteins are indicated on the left. (B) Effect of Cet1 on Ceg1 guanylyltransferase activity. Ceg1 (300 ng) was mixed with 300 ng of Cet1 (lane 2), Cet1(201-549) (lane 3), or Cet1(246-549) (lane 4) or with 320 ng of the mouse RNA triphosphatase domain Mce1(1-210) (lane 5) in 10 μl of buffer C containing 75 mM NaCl. A control sample contained 300 ng of Ceg1 alone (lane 1). The mixtures were incubated for 30 min on ice. An aliquot (1 μl) of each sample was then assayed for enzyme-GMP complex formation. Reaction mixtures (20 μl) containing 50 mM Tris HCl (pH 8.0), 5 mM MgCl₂, 5 mM DTT, 0.17 μM [α-³²P]GTP, and 1 μl of enzyme were incubated for 10 min at 37°C. The reaction products were analyzed by SDS-PAGE. Ceg1-GMP complex formation was visualized by autoradiography of the dried gel. (C) Cet1 titration. Ceg1 (550 ng) was mixed with 0, 65, 130, 325, 650, or 1,300 ng of Cet1 in 20 μl of buffer C containing 75 mM NaCl. An aliquot (1 μl) of each sample was then assayed for enzyme-GMP complex formation. Reaction mixtures (20 μl) contained 50 mM Tris HCl (pH 8.0), 5 mM MgCl₂, 5 mM DTT, 0.17 μM [α-³²P]GTP, and 28 ng of Ceg1 plus Cet1 as specified. Ceg1-[³²P]GMP complex formation was quantitated by scanning the SDS-PAGE gel. The effect of Cet1 on Ceg1 activity (fold stimulation) was calculated as the ratio of the Ceg1-[³²P]GMP signal intensity in Cet1-containing reaction mixtures to Ceg1-[³²P]GMP signal intensity in the control reaction mixture lacking Cet1. The data shown are averages of two separate titration experiments. (D) GTP titration. Ceg1 (300 ng) was mixed with 300 ng of Cet1 in 10 μl of buffer C containing 75 mM NaCl (+ Cet1). A control sample contained 300 ng of Ceg1 alone (− Cet1). The mixtures were incubated for 30 min on ice. An aliquot (1 μl) of each sample was then assayed for enzyme-GMP complex formation in reaction mixtures (20 μl) containing 50 mM Tris HCl (pH 8.0), 5 mM MgCl₂, 5 mM DTT, and [α-³²P]GTP as specified. Incubation was for 10 min at 37°C. Ceg1-GMP complex formation is plotted as a function of GTP concentration.

Glycerol gradient analysis showed that Cet1(201-549) by itself sedimented as a single 4.1S component that we presume is a monomer (Fig. 6A, fraction 13). The RNA triphosphatase activity also peaked in gradient fraction 13 (Fig. 6A). When Cet1(201-549) was mixed with Ceg1 and the mixture was analyzed by glycerol gradient sedimentation, the two proteins cosedimented as a 6.8S complex (Fig. 6B, fractions 9 to 11), which we presume is a Cet1(201-549)–Ceg1 heterodimer. The RNA triphosphatase and guanylyltransferase activities peaked together in fractions 9 to 11 (Fig. 6B).

FIG. 6.

Cet1(201-549) forms a complex with Ceg1. Aliquots (0.2 ml) of protein samples were applied to 4.8-ml 15 to 30% glycerol gradients containing 50 mM Tris-HCl (pH 8.0), 0.1 M NaCl, 2 mM DTT, and 0.05% Triton X-100. The gradients were centrifuged at 50,000 rpm for 16 h at 4°C in a Beckman SW50 rotor. Fractions (∼0.21 ml) were collected from the bottom of the tube (fraction 1). Aliquots (20 μl) of alternate fractions were analyzed by SDS-PAGE along with an aliquot of the material that had been applied to the gradient (lane L). The gels were fixed and stained with Coomassie blue dye. (A) Sedimentation analysis of Cet1(201-549). A 20-μg sample of the phosphocellulose preparation of Cet1(201-549) was applied to the gradient. Glycerol gradient fractions were assayed for RNA triphosphatase activity in reaction mixtures containing 20 pmol of γ-³²P-labeled poly(A) and 1 μl of a 1/50 dilution of the indicated fractions. Incubation was for 15 min at 30°C. The peaks of the marker proteins catalase, bovine serum albumin (BSA), and cytochrome c (cyt C), which were centrifuged in a parallel gradient, are indicated. (B) Cet1(201-549) (20 μg) and Ceg1 (20 μg) were mixed in buffer C containing 0.1 M NaCl and then incubated on ice for 30 min before being applied to the glycerol gradient. Gradient fractions were assayed for RNA triphosphatase as specified for panel A. Guanylyltransferase reaction mixtures contained 0.17 μM [α-³²P]GTP and 1 μl of the indicated fractions. PSL, photostimulable luminescence.

The more extensively truncated Cet1(246-549) protein did not sediment as a discrete monomer like full-size Cet1 and Cet1(201-549). Rather, most of the Cet1(246-549) protein sedimented as high-molecular-weight aggregates (8S to 13S) that retained RNA triphosphatase activity (data not shown). When a mixture of Cet1(246-549) and Ceg1 was sedimented in a glycerol gradient, most of the Cet1(246-549) remained aggregated; only a minor fraction of the input Ceg1 was shifted to the size expected for a heterodimer (data not shown). We surmise that a deletion spanning residues 201 to 245 (which results in loss of function in vivo despite retention of triphosphatase activity in vitro) affects the interaction of Cet1 with itself and with Ceg1.

Functional consequences of heterodimerization.

In analyzing the glycerol gradient activity profiles, we noted that the total guanylyltransferase activity detected in the gradient containing the mixture of Cet1 and Ceg1 was greater than the activity in the gradient containing Ceg1 alone. This suggested that heterodimerization might stimulate the activity of Ceg1. To further explore this issue, we incubated Cet1 with Ceg1 (at a molar ratio of ∼1:1) for 30 min on ice and then assayed the protein mixtures for enzyme-GMP complex formation. We found that Cet1, which had no guanylyltransferase activity on its own, significantly increased the yield of the Ceg1-GMP complex (Fig. 5B, compare lanes 1 and 2). The extent of stimulation of Ceg1-GMP formation by Cet1 was proportional to the concentration of Cet1 in the protein mixture under conditions in which Ceg1 was in excess (Fig. 5C). Optimal stimulation (∼13-fold) was noted at equimolar concentrations of Ceg1 and Cet1. Cet1(201-549) elicited a similar stimulation of Ceg1-GMP complex formation (Fig. 5B, lane 3), but the more extensively truncated version, Cet1(246-549), did not stimulate guanylyltransferase activity (Fig. 5B, lane 4). These results, together with those of the sedimentation analyses, suggest that stimulation of Ceg1 activity by Cet1 is contingent on heterodimerization. Mixing yeast guanylyltransferase with the 210-amino-acid RNA triphosphatase domain of the mouse RNA-capping enzyme (7) had no effect on Ceg1 activity (Fig. 5B, lane 5). GTP titration experiments showed that Cet1 stimulated Ceg1-GMP complex formation by enhancing the affinity of Ceg1 for the GTP substrate (Fig. 5D). The GTP titration curve was shifted significantly to the left in the presence of Cet1. The fold stimulation of Ceg1-GMP formation by Cet1 was greatest at limiting GTP concentrations but was still significant at 10 μM GTP (Fig. 5D). The amount of Ceg1-GMP complex formed during the 10-min reaction in the presence or absence of Cet1 reflected the final product yield and not an initial rate. We calculated that 20% of the Ceg1 molecules were guanylylated with [³²P]GMP during the in vitro reaction with 10 μM GTP in the presence of Cet1.

In a reciprocal experiment, a fixed amount of Cet1 was incubated with increasing amounts of Ceg1 for 30 min on ice, and aliquots of the mixtures were assayed for RNA triphosphatase. We found that Ceg1 (which had no RNA triphosphatase activity per se) caused modest (two- to threefold) stimulation of the RNA triphosphatase activity of Cet1 (data not shown).

Complementation of a Δcet1 mutation by mouse capping enzyme.

The physical and functional organizations of the triphosphatase and guanylyltransferase components of the capping apparatus have diverged in fungi versus metazoans. The guanylyltransferases of S. cerevisiae (Ceg1; 459 amino acids), Schizosaccharomyces pombe (Pce1; 402 amino acids), and Candida albicans (Cgt1; 449 amino acids) are monofunctional polypeptides that display ∼38% amino acid sequence identity overall (19, 22, 33). They are also functionally homologous, insofar as PCE1 and CGT1 can complement lethal ceg1 mutations in S. cerevisiae (22, 33). The S. cerevisiae RNA triphosphatase is encoded separately by CET1. In metazoans, the capping enzymes are bifunctional polypeptides with triphosphatase and guanylyltransferase activities (7, 12, 26, 28, 31, 32, 35). The 597-amino-acid mouse capping enzyme (Mce1) consists of an N-terminal phosphatase domain (amino acids 1 to 210) and a C-terminal guanylyltransferase domain (amino acids 211 to 597). The primary structure of the mouse guanylyltransferase domain is extensively homologous to that of yeast Ceg1, and the mechanism of catalysis by Mce1 through an enzyme-GMP intermediate is equivalent to that of Ceg1. A cDNA encoding either full-length Mce1 or the guanylyltransferase domain Mce1(211-597) complements the lethality of a ceg1 null mutation in yeast (7, 35).

In contrast, the RNA triphosphatase domain of the mouse capping enzyme bears no structural resemblance to the yeast RNA triphosphatase Cet1 but, instead, has strong similarity to the superfamily of protein phosphatases that act via a covalent phosphocysteine intermediate. The lack of mechanistic and structural similarity between the S. cerevisiae and metazoan triphosphatases prompted us to test whether the mouse capping enzyme could complement growth of the yeast Δcet1 mutant. MCE1 was cloned into a yeast CEN TRP1 expression plasmid under the control of a constitutive yeast TPI1 promoter (7). The MCE1 plasmid was introduced into the cet1::LEU2 strain, and Trp⁺ transformants were plated on medium containing 5-FOA to select against retention of the CEN URA3 CET1 plasmid. Cells bearing the TRP1 MCE1 plasmid grew readily on 5-FOA, whereas cells containing the TRP1 vector plasmid or a plasmid expressing only the mouse guanylyltransferase domain Mce1(211-597) were incapable of growth on 5-FOA (Fig. 7A). A mutated mouse capping enzyme allele, MCE1-C126A, in which the active-site cysteine nucleophile of the triphosphatase (Cys-126) was replaced with alanine, was incapable of supporting yeast growth in the plasmid shuffle assay (Fig. 7A). We concluded that the mammalian capping enzyme is functional as an RNA triphosphatase in vivo in yeast.

FIG. 7.

Complementation of Δcet1 by MCE1. (A) Δcet1 strain YBS20 was transformed with a CEN TRP1 plasmid bearing wild-type CET1, with pYX132(CEN TRP1)-based plasmids containing either wild-type MCE1 or mutant allele MCE1(K294A), MCE1(C126A), or MCE1(211-597), or with the pYX132 vector (7). (B) YBS20 was transformed with pYX132-based plasmid MCE1(1-231) or MCE1(1-210) (two independent isolates of each plasmid were tested), with the pYX132 vector, or with the CET1 control. Individual Trp⁺ transformants were selected and then patched on agar medium lacking tryptophan. Cells from single patches were then streaked on agar medium containing 0.75-mg/ml 5-FOA. The plates were photographed after incubation for 4 days at 30°C.

Two findings emerged from further studies of cet1 complementation by MCE1 derivatives. First, cells transformed with a cDNA encoding only the catalytically active RNA triphosphatase domain Mce1(1-210) were unable to form colonies on 5-FOA (Fig. 7B), signifying that the mouse RNA triphosphatase was functional in yeast only when linked to the guanylyltransferase. A slightly longer version of the triphosphatase domain, Mce1(1-231), was also nonfunctional in yeast (Fig. 7B). Second, the mutant allele MCE1(K294A), in which the active-site nucleophile of the mouse guanylyltransferase (Lys-294) was replaced with alanine, was also incapable of supporting growth of Δcet1 cells (Fig. 7A). Thus, the mouse RNA triphosphatase must be tethered to a catalytically active guanylyltransferase in order to function in yeast. We hypothesize that the mouse guanylyltransferase domain targets the mouse triphosphatase to RNA polymerase II and that the Mce1(K294A) mutant is nonfunctional because it sequesters nascent pre-mRNA ends in a nonproductive complex that is inaccessible to the yeast guanylyltransferase Ceg1.

DISCUSSION

The guanylyltransferase activity of the yeast Ceg1 protein is required for cell growth. We have found that shift of ceg1-ts mutants to the restrictive temperature elicits a rapid decline in the rate of protein synthesis, which correlates with a sharp reduction in the steady-state levels of multiple individual mRNAs (18). ceg1 mutations prevent the accumulation of mRNAs that are newly synthesized at the restrictive temperature. However, uncapped poly(A)⁺ mRNAs do accumulate at the restrictive temperature in ceg1 cells lacking the 5′ exo-RNase Xrn1 (18). These findings indicate that the cap guanylate is critical for mRNA stability in vivo. The cap is likely to play additional roles in yeast mRNA, insofar as eliminating Xrn1 does not bypass the lethality of a ceg1 null mutation (18). We have sought to identify genes that impact on Ceg1-dependent functions by screening for high-copy suppressors of ceg1 mutations. The present characterization of CET1(CES5), which encodes the RNA triphosphatase component of the yeast capping apparatus, as an allele-specific ceg1 suppressor provides new insights into the potential for regulation of RNA capping through protein-protein interactions.

Early work on the purification of yeast RNA guanylyltransferase documented that yeast cell extracts contain several capping enzyme isoforms that differ in chromatographic properties and association, or lack thereof, with an RNA triphosphatase activity (8, 29). The size heterogeneity of the catalytically active guanylyltransferase polypeptide (a mixture of 52- and 47-kDa species) was attributed to partial proteolysis (9, 29). Wang and Shatkin (29) reported that the yeast guanylyltransferase was freed of RNA triphosphatase activity during purification. Itoh et al. (9) characterized a preparation of yeast guanylyltransferase that consisted of two major polypeptides, a 52-kDa guanylyltransferase corresponding to Ceg1 (19) and an 80-kDa RNA triphosphatase, recently identified as Cet1 (27). The distribution of free versus complexed Ceg1 and Cet1 proteins in wild-type yeast cells in vivo has not been established.

Purified recombinant Cet1 catalyzed the release of ∼1 molecule of P_i from triphosphate-terminated poly(A) per enzyme molecule per s. This is comparable to the steady-state turnover number of 0.5 to 0.8 s⁻¹ reported for the RNA triphosphatase component of the vaccinia virus capping enzyme (13). The turnover number of Cet1 is also quite close to the value of 1 to 2 molecules of P_i release per s per enzyme molecule determined for the mouse RNA triphosphatase domain Mce1(1-210) (7). Interaction of Cet1 with Ceg1 results in twofold stimulation of the RNA triphosphatase activity. This modest effect is unlikely to enhance the overall rate of cap formation, because the guanylyltransferase reaction, rather than γ-phosphate cleavage, is rate limiting.

The stimulation of the guanylyltransferase activity of Ceg1 by Cet1 raises the prospect that cap formation in vivo might be regulated by modulating the extent of Cet1-Ceg1 heterodimerization. A plausible scenario to account for suppression of ceg1-25 by 2μ-CET1 is to posit that (i) the mutant Ceg1-25 protein displays diminished affinity for Cet1 at the restrictive temperature and (ii) increasing the level of wild-type Cet1 drives heterodimerization by simple mass action. Our experiments do not distinguish whether growth arrest of ceg1-25 cells at the restrictive temperature is the result of reduced guanylyltransferase activity without changes in the level of Ceg1-25 protein, accelerated degradation of the Ceg1-25 protein, or a combination of both. The narrow allele specificity for suppression of ceg1-25 and ceg1-6 by CET1 is in keeping with the expectation that only a subset of conditional ceg1 mutations will affect the interaction of Ceg1 with Cet1 (as opposed to ts protein folding effects, thermolabile catalytic activity without global folding defects, etc). The CET1-suppressible ceg1-25 allele encodes a polypeptide with three amino acid substitutions: N283Y, D370A, and T378S. The ceg1-6 gene product contains two coding mutations: V289A and E364G. The affected residues fall into two clusters. E364, D370, and T378 are located within and flanking conserved nucleotidyltransferase motif VI, which contacts GTP in the closed conformation of the Chlorella virus capping enzyme (5, 31). N283 and V289 are located between motifs V and VI in a segment of Ceg1 that has no counterpart in the Chlorella virus or mouse guanylyltransferase.

Cet1 interacts with the guanylyltransferase from yeast and not with the mouse guanylyltransferase. Sedimentation analysis revealed no evidence of the formation of a heteromeric complex when purified recombinant Cet1 was mixed with purified Mce1(211-597) under the same experimental conditions that permitted isolation of the Cet1-Ceg1 complex. Also, Cet1 did not stimulate enzyme-guanylate formation by Mce1(211-597) (7a). In turn, the mouse triphosphatase domain Mce1(1-210) had no stimulatory effect on Ceg1 activity.

These and other recent findings may explain why the mouse guanylyltransferase domain Mce1(211-597) can replace yeast guanylyltransferase Ceg1 in vivo, whereas the mouse triphosphatase domain Mce1(1-210) cannot replace the yeast triphosphatase Cet1. To wit: (i) the mouse guanylyltransferase domain need not interact with Cet1 to attain full catalytic activity and is able on its own to locate the polymerase II (pol II) transcription elongation complex by binding to the phosphorylated pol II CTD (7, 35), and (ii) the mouse triphosphatase domain fails to interact with or activate Ceg1 and is unable to bind by itself to the phosphorylated pol II CTD (7). It is clear that the mouse triphosphatase is active in vivo, insofar as full-length MCE1 complements the yeast Δcet1 null mutation. A cDNA encoding the full-length human capping enzyme also complements the Δcet1 null mutation (33a). Our finding that the MCE1(C126A) allele does not complement Δcet1 provides the first genetic evidence that an RNA triphosphatase activity is essential for cell growth.

Does the in vivo requirement for Cet1 reflect its catalytic function in γ-phosphate hydrolysis, its ability to stimulate Ceg1, or both? We infer from the deletion analysis that interaction of Cet1 with Ceg1 is critical for Cet1 function in vivo. The Cet1 protein segment between residues 201 and 245, which is not necessary for RNA triphosphatase activity, may comprise part of the Ceg1-binding surface. To draw conclusions about the requirement for Cet1 catalytic activity, it is necessary to identify Cet1 point mutations that abrogate the triphosphatase function but preserve Cet1-Ceg1 interaction.