INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

mRNA capping occurs by a series of three enzymatic reactions in which the 5' triphosphate terminus of the primary transcript is cleaved to a diphosphate by RNA triphosphatase, capped with GMP by RNA guanylyltransferase, and methylated at the N-7 position of guanine by RNA (guanine 7) methyltransferase (1). In vivo, the capping reactions occur cotranscriptionally, i.e. the substrates for the capping enzymes are nascent RNA chains engaged within RNA polymerase II (pol II)¹ elongation complexes. There must exist a mechanism to target cap formation in vivo to transcripts made by pol II, because the capping enzymes have no inherent specificity for modifying pre-mRNAs in vitro. We and others (2-4) have suggested that targeting is achieved through direct physical interaction of one or more components of the capping apparatus with the phosphorylated carboxyl-terminal domain (CTD) of the largest subunit of pol II. This model is supported by the finding that recombinant yeast guanylyltransferase and methyltransferase proteins bind specifically and independently to the phosphorylated CTD in vitro (2). Are such direct protein-protein interactions conserved in higher eukaryotes? Does the triphosphatase component of the capping apparatus also interact with the CTD?

We know that the physical and functional organizations of the triphosphatase and guanylyltransferase components of the capping apparatus have diverged in fungi versus metazoans. The guanylyltransferases of Saccharomyces cerevisiae (Ceg1; 459 amino acids), Schizosaccharomyces pombe (Pce1; 402 amino acids), and Candida albicans (Cgt1; 449 amino acids) are monofunctional polypeptides that cap diphosphate-terminated RNAs (5-7). Transfer of GMP from GTP to the 5' diphosphate terminus of RNA occurs in a two-stage reaction involving a covalent enzyme-GMP intermediate (8). The GMP is linked to the enzyme through a phosphoamide (P-N) bond to the -amino group of a lysine residue within a conserved motif, KXDG, found in all known cellular and DNA virus-encoded capping enzymes (9). The fungal guanylyltransferases display ~38% amino acid sequence identity overall. They are also functionally homologous, insofar as PCE1 and CGT1 can complement lethal ceg1 mutations in S. cerevisiae (6, 7). The S. cerevisiae RNA triphosphatase is a 549-amino acid polypeptide encoded by the CET1 gene (10). The Ceg1 and Cet1 polypeptides interact in vivo and in vitro.

Metazoan capping enzymes are bifunctional polypeptides with triphosphatase and guanylyltransferase activities (11, 12). Yagi et al. (12) isolated triphosphatase and guanylyltransferase domain fragments of the Artemia salina capping enzyme by partial proteolysis with trypsin. However, it was not clear from this work whether the functional domains overlapped structurally. The first metazoan-capping enzyme gene was isolated recently from Caenorhabditis elegans (13, 14). The 573-amino acid nematode protein consists of a carboxyl-terminal domain homologous to yeast Ceg1 and an amino-terminal domain that has strong similarity to the superfamily of protein phosphatases that act via a covalent phosphocysteine intermediate.

We recently isolated a mouse cDNA encoding the mammalian homologue of the C. elegans capping enzyme (2). The 597-amino acid mouse capping enzyme (Mce1) also consists of an amino-terminal phosphatase domain and a carboxyl-terminal guanylyltransferase domain. Here, we report that Mce1 is functional in vitro and in vivo. An autonomous RNA triphosphatase domain (amino acids 1-210) and an autonomous guanylyltransferase domain (amino acids 211-597) have been purified and characterized. We found that the guanylyltransferase domain per se binds to the pol II CTD and that this interaction requires CTD phosphorylation. The triphosphatase domain of mouse capping enzyme does not bind to the CTD. The phosphorylation-dependent interaction between guanylyltransferase and the CTD is conserved from yeast to mammals and provides a general mechanism for targeting caps to nascent pol II transcripts in vivo.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Yeast Expression Plasmids-- The MCE1 cDNA was cloned into a customized yeast expression vector, pYX1-His, a derivative of pYX132 (CEN TRP1) in which six consecutive histidine codons and a unique NdeI site are inserted between the NcoI and BamHI sites of pYX132 (pYX-132 was purchased from Novagen). The DNA insert of the resulting plasmid pYX1-MCE1 extended from the Mce1 translation start codon to an XhoI site located in the 3' UTR. pYX1-MCE1 encodes the full-length 597-amino acid Mce1 polypeptide fused in-frame with an amino-terminal 12-amino acid leader peptide (MGSHHHHHHSGH). MCE1 expression in this plasmid was under the control of a constitutive yeast TPI promoter. Amino-terminal deletion mutants of MCE1 were constructed by PCR amplification with mutagenic primers that introduced an NdeI restriction site and a methionine codon in lieu of the codons for Ser-210 or Gln-259 or an NdeI site at Met-276. The PCR products were digested with NdeI and BglII and then inserted into pYX1-MCE1 so as to replace a 1.0-kilobase pair NdeI-BglII fragment of the wild type MCE1 gene with restriction fragments encoding deleted Mce1 polypeptides. The mutated genes were named according to the amino acid coordinates of their polypeptide products, i.e. MCE1(211-597), MCE1(260-597), and MCE1(276-597). Alanine-substitution mutations (C126A and K294A) in MCE1 were programmed by synthetic oligonucleotides using the two-stage overlap extension method (15). pYX1-MCE1 served as the template for the first round of amplification. The second-stage PCR products of the C126A and K294A reactions were digested with NdeI and BglII or NdeI and HindIII, respectively, and then ligated into the corresponding sites in pYX1-MCE1 in place of the wild type fragments. The presence of the desired mutations was confirmed in every case by dideoxy sequencing. We sequenced the entire restriction fragment insert in each pXY1 plasmid to exclude the introduction of unwanted mutations. A CEN TRP1 vector, pYX1-CEG1, encoding yeast Ceg1 under the control of a TPI promoter, was constructed by inserting an NcoI-HindIII fragment of plasmid pGYCE-360 (16) into pYX132.

Test of MCE1 Function by Plasmid Shuffle-- Strain YBS2 (MATa ura3 trp1 lys2 leu2 ceg1::hisG pGYCE-360), which is deleted at the chromosomal CEG1 locus, is viable when it maintains an extrachromosomal copy of CEG1 on a CEN URA3 plasmid (pGYCE-360) (16). YBS2 was transformed with pXY1-based plasmids bearing wild type and mutant alleles of MCE1. Trp+ transformants were selected on medium lacking tryptophan. Individual colonies were patched on medium lacking tryptophan. Cells from single patches were then streaked on medium containing 0.75 mg/ml of 5-fluoroorotic acid (5-FOA). The plates were incubated at 30 °C. Mutations scored as lethal were those that did not support colony formation after 7 days. Individual colonies of the viable MCE1 alleles were picked from the 5-FOA plate and patched to plates lacking tryptophan. All 5-FOA survivors were confirmed to be Ura.

Bacterial Expression Plasmids-- NdeI-XhoI fragments of pYX1-MCE1, pYX1-MCE1(211-597), pYX1-MCE1(260-597), and pYX1-MCE1(276-597) were inserted into to the T7-based expression plasmid pET16b (Novagen). The carboxyl-terminal deletion mutant MCE1(1-210) was constructed by PCR amplification from an MCE1 template with an antisense primer that introduced a translation stop at codon 211 and a flanking XhoI restriction site. The PCR product was digested with NdeI and XhoI and inserted into pET16b.

Expression and Purification of Recombinant Mce1 Protein-- A 25-ml culture of E. coli BL21(DE3)/pET-MCE1 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--D-thiogalactopyranoside, and incubation was continued at 17 °C for 17 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, 0.2 M NaCl, 10% sucrose). Lysozyme was added to a final concentration of 50 µg/ml; the suspension was incubated on ice for 10 min and then sonicated for 30 s. Triton X-100 was added to a final concentration of 0.1%, and sonication was repeated to reduce the viscosity of the lysate. 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 0.5-ml column of Ni-NTA-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 step-wise with Buffer B (50 mM Tris-HCl, pH 8.0, 0.1 M NaCl, 10% glycerol) containing 50, 100, 200, 500, and 1000 mM imidazole. The polypeptide compositions of the column fractions were monitored by SDS-polyacrylamide gel electrophoresis (PAGE). The recombinant Mce1 polypeptide was retained on the column and recovered in the 200 and 500 mM imidazole eluates. These two fractions were pooled and dialyzed against Buffer C (50 mM Tris-HCl, pH 8.0, 50 mM NaCl, 2 mM DTT, 10% glycerol, 0.05% Triton X-100).

Expression and Purification of Mce1(1-210) and Mce1(211-597)-- 500-ml cultures of E. coli BL21(DE3) harboring pET-MCE1(1-210) or pET-MCE1(211-597) were grown at 37 °C until the A₆₀₀ reached 0.5. The cultures were adjusted to 0.4 mM isopropyl--D-thiogalactopyranoside and incubated for 3 h at 37 °C. Cells were harvested by centrifugation and stored at 80 °C. Thawed bacteria were resuspended in 25 ml of Buffer A. Soluble lysates were prepared as described in the preceding section and then applied to 4-ml columns of Ni-NTA-agarose, which were eluted step-wise with 50, 100, 200, 500, and 1000 mM imidazole. The Mce1(1-210) and Mce1(211-597) polypeptides were retained on the Ni-NTA-agarose columns and recovered in the 200 and 500 mM imidazole eluates. The preparations were dialyzed against Buffer C. Enzyme fractions were stored at 80 °C and thawed on ice just prior to use. Protein concentrations were determined using the Bio-Rad dye binding assay with bovine serum albumin as a standard.

Glycerol Gradient Sedimentation-- Aliquots (200 µl) of the dialyzed Ni-agarose preparations of Mce1, Mce1(1-210), and Mce1(211-597) were applied to a 4.8-ml 15-30% glycerol gradients containing 50 mM Tris-HCl (pH 8.0), 0.3 M NaCl, 2 mM DTT, 1 mM EDTA, and 0.1% Triton X-100. The gradients were centrifuged at 50,000 rpm for 24 h at 4 °C in a Beckman SW50 rotor. Fractions (0.2 ml) were collected from the bottom of the tube. Marker proteins catalase, bovine serum albumin, soybean trypsin inhibitor, and cytochrome C were sedimented in a parallel gradient.

Enzyme-GMP Complex Formation-- Standard reaction mixtures (20 µl) containing 50 mM Tris-HCl (pH 8.0), 5 mM DTT, 5 mM MgCl₂, [³²P]GTP as specified, and enzyme were incubated for 5 min at 37 °C. The reactions were halted by adding SDS to a final concentration of 1%. 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 was quantitated by scanning the gel with a FUJIX BAS1000 Bio-Imaging Analyzer.

RNA Triphosphatase Assay-- RNA triphosphatase activity was assayed by liberation of ³²P_i from -³²P-labeled triphosphate-terminated poly(A) (17). Standard reaction mixtures (10 µl) containing 50 mM Tris-HCl (pH 7.5), 5 mM DTT, 10 pmol (of triphosphate termini) of [-³²P]poly(A), and enzyme as specified were incubated for 15 min at 37 °C. Reactions was halted by addition of 1 µl of 12 N formic acid. Aliquots of the mixtures were applied to a polyethyleneimine-cellulose TLC plate that was developed with 0.75 M potassium phosphate (pH 4.3). The release of ³²P_i from [-³²P]poly(A) was quantitated by scanning the TLC plate with a FUJIX BAS1000 Bio-Imaging Analyzer.

CTD Affinity Chromatography-- Recombinant glutathione S-transferase (GST)-CTD, consisting of glutathione-S-transferase fused to the first 15 tandem heptad repeats from the mouse pol II CTD (consensus sequence YSPTSPS), was purified and coupled to glutathione agarose as described (2). GST-CTD was phosphorylated in vitro by incubation with HeLa cell extract and ATP (2). The phosphorylation reaction resulted in the addition of approximately 3 phosphates per molecule of GST-CTD. The GST-CTD-PO4 polypeptide was adsorbed to glutathione agarose and the resin was washed with buffer containing 1 M NaCl prior to use in affinity chromatography. The GST-CTD and GST-CTD-PO4 resins contained 3.0 mg of fusion protein per ml of agarose. Affinity chromatography was performed by mixing 25 µl of GST-CTD or GST-CTD-PO4 resins with 180-200-µl samples of Mce1 proteins in binding buffer containing 20 mM HEPES (pH 7.9), 0.1 mM EDTA, 1 mM DTT, 20% glycerol, 0.1 M NaCl, 0.5 µM microcystine, 1 mM -glycerophosphate, 0.1% Nonidet P-40. After mixing for 1 h at 4 °C, the agarose beads were collected by centrifugation, and the supernatant was removed. The resins were washed three or four times with 400-500 µl of binding buffer, and then the bound material was eluted with 100 µl of binding buffer containing 0.9 M NaCl. Aliquots of the input sample and the 0.9 M NaCl eluates were analyzed by SDS-PAGE.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Mammalian Capping Enzyme-- We have isolated a mouse cDNA that encodes Mce1, a putative mRNA capping enzyme (GenBank^TM accession no. AF034568). The size of the Mce1 polypeptide (597-amino acids; 68 kDa) agrees with the size of the 68-kDa enzyme-GMP complex formed by the capping enzyme isolated from mouse cell extracts (18). The amino acid sequence of Mce1 suggests that it is a bifunctional enzyme consisting of an amino-terminal phosphatase domain and a C-terminal guanylyltransferase domain. In this respect, it resembles the C. elegans capping enzyme (13, 14), to which it displays 43% sequence identity. The carboxyl-terminal portion of Mce1 contains the defining sequence motifs of the covalent nucleotidyl transferase superfamily (9), including the KXDGXR sequence that constitutes the active site. The lysine side chain within this motif (Lys-294 in Mce1) reacts with GTP to form a covalent enzyme-GMP intermediate. The amino-terminal portion of Mce1 contains the (I/V)HCXAGXGR(S/T)G signature motif of the dual-specificity protein phosphatase/protein tyrosine phosphatase enzyme family. These proteins catalyze phosphoryl transfer from a protein phosphomonoester substrate to the thiol of a cysteine on the enzyme to form a covalent phosphocysteine intermediate, which is then attacked by water to liberate phosphate (19). The cysteine within the signature motif is the active site of phosphoryl transfer and is thus essential for reaction chemistry. Cys-126 is predicted to be the active site of phosphoryl transfer by Mce1.

Catalytic Activity of Mouse Capping Enzyme-- The biochemical properties of the mammalian capping enzyme were examined after expressing Mce1 and its component domains in bacteria. First, the full-length MCE1 coding sequence was inserted into an inducible T7 RNA polymerase-based pET vector such that a histidine-rich amino-terminal leader (His-tag) was fused to the 597-amino acid Mce1 polypeptide. The pET-Mce1 plasmid was introduced into E. coli BL21(DE3), a strain that contains the T7 RNA polymerase gene under the control of a lacUV5 promoter. A 68-kDa polypeptide corresponding to His-tagged Mce1 was detected by SDS-PAGE in whole-cell and soluble lysates of isopropyl--D-thiogalactopyranoside-induced bacteria (Fig. 1A and data not shown). To assay guanylyltransferase activity of the expressed Mce1 protein, we incubated soluble protein from induced bacteria in the presence of [-³²P]GTP and a divalent cation. This resulted in the formation of a ³²P-labeled nucleotidyl-protein adduct that migrated as a 68-kDa species during SDS-PAGE (Fig. 1A). Three other polypeptides in the range of 45-55 kDa were labeled with GMP to a lesser extent. We detected no label transfer from GTP to polypeptides of these sizes in extracts prepared from bacteria that lacked the MCE1 gene (not shown). Purification of the His-tagged 68-kDa Mce1 guanylyltransferase was achieved by adsorption to Ni-agarose and elution with 200-500 mM imidazole (Fig. 1A). The imidazole eluate fractions were highly enriched with respect to Mce1, as judged by SDS-PAGE (Fig. 1A, left panel). Note that the lower molecular weight polypeptides with guanylyltransferase activity were recovered in the Ni-agarose flow-through and wash fractions (Fig. 1A, right panel), which suggested that these were proteolytic fragments of Mce1 that lacked the amino-terminal His-tag.

View larger version (59K): [in this window] [in a new window]   Fig. 1.   Purification and guanylyltransferase activity of recombinant Mce1 and Mce1(211-597). The polypeptide compositions of recombinant Mce1 (A, left panel) and Mce1(211-597) (B, left panel) during Ni-agarose purification were analyzed by SDS-PAGE: lane 1, soluble bacterial lysate; lane 2, Ni-agarose flow-through; lane 3, Ni-agarose wash; lane 4, 50 mM imidazole eluate; lane 5, 100 mM imidazole eluate; lane 6, 200 mM imidazole eluate; lane 7, 500 mM imidazole eluate; lane 8, 1 M imidazole eluate. The gels were fixed and stained with Coomassie Blue dye. The positions and sizes (in kDa) of coelectrophoresed marker polypeptides are indicated. The guanylyltransferase activity of recombinant Mce1 (A, right panel) and Mce1(211-597) (B, right panel) during Ni-agarose purification was assayed by enzyme-GMP complex formation. Reaction mixtures containing 50 mM Tris-HCl (pH 8.0), 5 mM DTT, 5 mM MgCl2, 0.17 µM [-32P]GTP, and 1 µl of the protein fractions specified above in lanes 1-8 were incubated for 5 min at 37 °C. The reaction products were analyzed by SDS-PAGE. Autoradiographs of the gels are shown. The positions and sizes (in kDa) of prestained marker polypeptides are indicated.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Glycerol gradient sedimentation of Mce1 and Mce1(211-597). A, the Mce1 Ni-agarose preparation was sedimented in a 15-30% glycerol gradient and alternate fractions were assayed for enzyme-GMP complex formation () and RNA triphosphatase activity (). The guanylyltransferase reaction mixtures contained 0.17 µM [-32P]GTP and 1 µl of the indicated fractions. Guanylyltransferase activity was gauged by the signal intensity of the radiolabeled 68-kDa Mce1 polypeptide (PSL, photo stimulatable luminescence), plotted on the y axis at left. The RNA triphosphatase reaction mixtures contained 10 pmol (of triphosphate termini) of [-32P] poly(A) and 0.l µl (1 µl of a 1:10 dilution) of the indicated gradient fractions. Incubation was for 15 min at 37 °C. Activity (in pmol of Pi released) is plotted on the y axis at right. The peaks of marker proteins bovine serum albumin and cytochrome c, centrifuged in a parallel gradient, are indicated by arrows. B, the Mce1(211-597) Ni-agarose preparation (Ni) was sedimented in a 15-30% glycerol gradient, and gradient fractions were analyzed by SDS-PAGE. Polypeptides were visualized by staining with Coomassie Blue dye. The position of Mce1(211-597) is denoted by an asterisk on the right. The position and sizes (in kDa) of marker polypeptides are indicated on the left. Guanylyltransferase reaction mixtures contained 0.17 µM [-32P]GTP and 1 µl of the indicated gradient fractions; incubation was for 5 min at 37 °C. Guanylyltransferase activity was gauged by the signal intensity of the radiolabeled 46-kDa Mce1(211-597) polypeptide (PSL).

Autonomous Guanylyltransferase and Triphosphatase Domains-- To evaluate whether the carboxyl-terminal portion of Mce1 constituted an autonomous functional guanylyltransferase domain, we expressed the protein segment from residues 211-597 as a His-tagged fusion protein. The choice of residue 211 as a domain breakpoint was based on its location 84 amino acids from the putative Mce1 active site Lys-294; the active site lysine of the monofunctional Chlorella virus guanylyltransferase is located 82 amino acids from its amino terminus (20, 21). We found that isopropyl--D-thiogalactopyranoside-induced bacteria accumulated substantial amounts of a soluble 46-kDa polypeptide corresponding to Mce1(211-597), which reacted with [-³²P]GTP to form a 47-kDa radiolabeled enzyme-GMP adduct (Fig. 1B). Mce1(211-597) adsorbed to Ni-agarose and was eluted at 200-500 mM imidazole (Fig. 1B). Mce1(211-597) sedimented as a discrete 2.8 S peak during glycerol gradient centrifugation (Fig. 2B). We surmise that Mce1 the guanylyltransferase component is a monomer in solution.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Purification and RNA triphosphatase activity of Mce1(1-210). A, the polypeptide composition of Mce1(1-210) protein during Ni-agarose purification was analyzed by SDS-PAGE: lane 1, soluble bacterial lysate; lane 2, Ni-agarose flow-through; lane 3, Ni-agarose wash; lane 4, 50 mM imidazole eluate; lane 5, 100 mM imidazole eluate; lane 6, 200 mM imidazole eluate; lane 7, 500 mM imidazole eluate; lane 8, 1 M imidazole eluate. A Coomassie Blue-stained gel is shown. The Mce1(1-210) polypeptide is denoted by an asterisk. B, glycerol gradient sedimentation. The Mce1(1-210) Ni-fraction preparation was sedimented in a 15-30% glycerol gradient. The polypeptide compositions of the indicated fractions were analyzed by SDS-PAGE. A Coomassie Blue-stained gel is shown. Aliquots of the glycerol gradient fractions were assayed for RNA triphosphatase activity. The reaction mixtures contained 10 pmol (of triphosphate termini) of [-32P]poly(A) and 1 µl of a 1:1000 dilution of the indicated gradient fractions. Incubation was for 15 min at 37 °C.

Characterization of the RNA Triphosphatase Domain-- The standard RNA triphosphatase reaction contained 50 mM Tris-HCl (pH 7.5) and 1 µM ³²P-labeled poly(A). The extent of -phosphate hydrolysis during a 15-min incubation at 37 °C was proportional to input Mce1(1-210) (Fig. 4A). In the linear range of enzyme dependence, 1 pmol of ³²P_i was released per fmol of protein. The specific activity of the Ni-agarose preparation was 37 units/µg. (One unit of enzyme releases 1 nmol of ³²P_i from ³²P poly(A) in 15 min.) A kinetic analysis showed that the initial rate of P_i release was proportional to enzyme concentration (Fig. 4B). Mce1(1-210) hydrolyzed 1.2-2 molecules of P_i/enzyme/s at steady state. RNA triphosphatase activity was optimal in 50 mM Tris buffer at pH 7.0-7.5; the extents of -phosphate release at pH 6.0 and pH 9.5 were 70 and 35%, respectively, of the activity seen at pH 7.5 (not shown). Activity was optimal in the absence of a divalent cation and was unaffected by EDTA. Inclusion of divalent cations elicited a concentration-dependent inhibition of RNA triphosphatase activity (Fig. 4C). 75% inhibition was observed at 0.5 mM MgCl₂ or MnCl₂. Mce1(1-210) did not catalyze release of ³²P_i from [-³²P]ATP (not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Characterization of the RNA triphosphatase domain. A, protein titration. Reaction mixtures (10 µl) containing 50 mM Tris-HCl (pH 7.5), 5 mM DTT, 10 pmol (of triphosphate termini) of [-32P]poly(A), and the indicated amounts of Mce1(1-210) (Ni-agarose fraction) were incubated for 15 min at 37 °C. Pi release is plotted as a function of input protein. B, kinetics. Reaction mixtures contained 1 µM [-32P]poly(A) and 0.1, 0.2, or 1 nM Mce1(1-210). The reaction was initiated by the addition of enzyme. Aliquots (10 µl) were withdrawn at the times indicated and quenched with formic acid. Pi release is plotted as function of incubation time. C, inhibition by divalent cations. Reaction mixtures (10 µl) containing 10 pmol of [-32P]poly(A), 5 fmol of Mce1(1-210), and MgCl2 or MnCl2 as indicated were incubated for 15 min at 37 °C. Pi release is plotted as a function of divalent cation concentration.

Characterization of the Guanylyltransferase Domain-- Formation of the covalent enzyme-GMP complex by Mce1(211-597) was dependent on the concentration of GTP from 0.01 to 2 µM (Fig. 5A). The yield of guanylated enzyme plateaued at 5-10 µM GTP; we estimated that at least 30% of the enzyme molecules were labeled with GMP in vitro. (It has been our experience that a significant fraction of any recombinant guanylyltransferase purified from bacteria is already guanylylated and thus cannot be labeled during the in vitro reaction.) Enzyme-GMP complex formation was strictly dependent on a divalent cation cofactor. Either magnesium or manganese sufficed; activity was optimal at 1-2 mM MnCl₂ or 5-10 mM MgCl₂ (Fig. 5B). The rate of enzyme-guanylate formation was slightly greater in the presence of manganese versus magnesium (Fig. 5C). The Ni-agarose preparation of Mce1(211-597) displayed trace levels of RNA triphosphatase activity. The specific activity (0.02 units/µg of protein) was 0.05% that of Mce1(1-210). The residual triphosphatase was resolved from the guanylyltransferase activity during glycerol gradient sedimentation of Mce1(211-597), suggesting that this phosphatase was an E. coli contaminant. We surmise that the guanylyltransferase domain of Mce1 has no triphosphatase activity per se.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Analysis of enzyme-GMP formation by Mce1(211-597). A, GTP dependence. Reaction mixtures (20 µl) containing 50 mM Tris-HCl (pH 8.0), 5 mM MgCl2, 5 mM DTT, 2 pmol of Mce1(211-597) (Ni-agarose fraction), and [-32P]GTP as indicated were incubated for 5 min at 37 °C. The products were analyzed by SDS-PAGE. The amount of 32P-GMP (pmol) transferred to Mce1(211-507) is plotted as a function of GTP concentration. B, divalent cation dependence. Reaction mixtures (20 µl) containing 5 µM [-32P]GTP, 2 pmol of Mce1(211-597), and MgCl2 or MnCl2 as indicated were incubated for 5 min at 37 °C. Enzyme GMP-complex formation is plotted as a function of divalent cation concentration. C, kinetics. Reaction mixtures contained (per 20 µl) 5 µM [-32P]GTP, 2 pmol of Mce1(211-597), and either 5 mM MgCl2 or 5 mM MnCl2. The reaction was initiated by the addition of enzyme. Aliquots (20 µl) were withdrawn at the times indicated. Enzyme GMP-complex formation is plotted as function of incubation time.

Mce1 Activity in Vivo-- To test whether the mouse capping enzyme was functional in vivo, we examined the ability of the wild type full-length MCE1 cDNA to complement a ceg1 null mutation by using the plasmid shuffle technique (6, 16). MCE1 was cloned into a yeast CEN TRP1 expression plasmid under the control of a constitutive yeast TPI promoter. The MCE1 plasmid was introduced into YBS2, a yeast strain in which the chromosomal CEG1 locus is deleted and that is dependent for growth on maintenance of an extrachromosomal copy of CEG1 on a CEN URA3 plasmid. Trp+ transformants were plated on medium containing 5-FOA to select against retention of the wild type CEG1 gene. Cells bearing the TRP1 MCE1 plasmid grew readily on 5-FOA, whereas cells containing the TRP1 vector plasmid were incapable of growth on 5-FOA. A mutated allele, MCE1-K294A, in which the active site nucleophile of the guanylyltransferase Lys-294 was substituted by alanine, was completely incapable of supporting yeast growth in the plasmid shuffle assay (Fig. 6A). We conclude that the mammalian capping enzyme is functional as a guanylyltransferase in vivo.

View larger version (41K): [in this window] [in a new window]   Fig. 6.   Complementation of ceg1 by MCE1. A, the polypeptide encoded by wild type and mutant alleles of MCE1 are depicted as horizontal bars with N termini at the left and C termini at the right. The positions of the active site cysteine (Cys-126; C) for RNA triphosphatase and the active site lysine (Lys-294; K) for the guanylyltransferase are indicated. Gene function was tested by plasmid shuffle. Complementation by plasmids containing MCE1 alleles was gauged relative to a CEG1 control plasmid. Alleles that supports the formation of wild type sized colonies after 3 days on 5-FOA are scored as +++. Alleles that yielded slightly smaller colonies are scored as ++, whereas those that required 7 days to form macroscopic colonies are denoted by +. Lethal mutations () were those that formed no colonies after 7 days. WT, wild type. B, YBS2 derivatives bearing the indicated capping enzyme genes under the control of a TPI promoter were streaked on YPD agar plates. The plates were photographed after incubation for 4 days at 30 °C.

Interaction of Mce1 with the Pol II CTD-- We used affinity chromatography to analyze the binding of mouse capping enzyme and its constituent domains to the pol II CTD. The test samples were incubated in parallel with a glutathione-Sepharose resin containing an immobilized ligand composed of either (i) glutathione S-transferase fused to the first 15 tandem heptad repeats of the mouse pol II CTD (consensus sequence YSPTSPS) or (ii) GST-CTD fusion protein that had been phosphorylated in vitro using HeLa nuclear extract as a source of kinase activity. The phosphorylation reaction resulted in the addition of approximately 3 phosphates per molecule of GST-CTD. After phosphorylation in vitro, the GST-CTD resin was washed extensively with 1 M NaCl to remove residual non-CTD proteins before performing affinity chromatography with mouse capping enzyme. The 68-kDa full-length Mce1 protein (residues 1-597) was prepared by in vitro translation in the presence of [³⁵S]methionine, and the translation product was subjected to affinity chromatography. We found that ³⁵S-labeled Mce1 bound specifically to the phosphorylated CTD (Fig. 7A, lane P) but not to the nonphosphorylated CTD control (lane C). The 46-kDa Mce1(211-597) guanylyltransferase domain by itself also bound specifically to phosphorylated CTD (Fig. 7A). In contrast, the RNA triphosphatase domain Mce1(1-210) did not interact with either phosphorylated or nonphosphorylated CTD (Fig. 7A). We conclude that the triphosphatase domain has no intrinsic affinity for the CTD and that its recruitment to the pol II transcription complex is mediated via its connection in cis to the guanylyltransferase. Guanylyltransferase deletion mutants Mce1(260-597) and Mce1(276-597) were unable to bind the phosphorylated CTD (Fig. 7A). Apparently, deletion of the segment of Mce1 from residues 211-259 disrupts CTD-PO4 recognition as well as guanylyltransferase activity.

View larger version (39K): [in this window] [in a new window]   Fig. 7.   CTD affinity chromatography. A, [35S]methionine-labeled full-length Mce1 polypeptide (1-597) and truncated derivatives Mce1(211-597), Mce1(260-597), Mce1(276-597), and Mce1(1-210) were prepared by translation in vitro in a rabbit reticulocyte lysate. The transcripts encoding these proteins were transcribed in vitro by T7 RNA polymerase from pET-based vectors. CTD-affinity chromatography was performed as described under "Materials and Methods," and aliquots of the input fraction (L) (1.1% of total material loaded) and the high-salt eluates from the nonphosphorylated GST-CTD control resin (C) (10% of eluate) and the phosphorylated GST-CTD resin (P) (10% of eluate) were analyzed by SDS-PAGE. An autoradiograph of the gel is shown. The positions and sizes (in kDa) of prestained marker proteins are indicated on the left. B, partially purified recombinant Mce1(1-210) and Mce1(211-597) were subjected to CTD affinity chromatography. Aliquots of the input fraction (L) (5% of total material loaded) and the high-salt eluates from the nonphosphorylated GST-CTD control resin (C) (10% of eluate) and the GST-CTD-PO4 resin (P) (10% of eluate) were analyzed by electrophoresis through an 8-16% polyacrylamide gradient gel in 0.1% SDS. Polypeptides were visualized by silver staining. The positions of the guanylyltransferase (GTase) and triphosphatase (TPase) polypeptides are denoted by arrows. Aliquots (1 µl) of each fraction (representing 0.5% of the loaded material and 1% of the high salt eluates) were assayed for enzyme-GMP complex formation. The products were analyzed by SDS-PAGE; the labeled covalent intermediate (EpG) was detected by autoradiography of the gel (lower panel).

Cis Negative Effect of an Active Site Mutation of the RNA Triphosphatase on Mce1 Function Yeast-- Yeast cells bearing MCE1-C126A, an allele with an alanine substitution at the active site cysteine of the triphosphatase domain, formed very tiny colonies on 5-FOA plates, even after 6 days of incubation. This contrasted starkly with the behavior of MCE1 cells, which formed colonies of normal size after 2 days on 5-FOA. MCE1-C126A cells displayed a severe growth defect when plated on rich medium at 30 °C (Fig. 6B) and were unable to grow on YPD plates at 37 °C (not shown). The growth suppressive effects of an inactivating point mutation in the triphosphatase were far more severe than a complete deletion of the triphosphatase domain (Fig. 6B). We hypothesize that the C126A mutation interferes with the endogenous yeast triphosphatase, either by sequestering the nascent pre-RNA ends in a nonproductive complex or by blocking access of the yeast triphosphatase to RNA polymerase II.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Functional Domains of Mammalian Capping Enzyme-- Fungi and Chlorella virus encode monofunctional guanylyltransferases, whereas mammals encode a bifunctional triphosphatase-guanylyltransferase enzyme. Here, we demonstrate that the triphosphatase and guanylyltransferase domains of mammalian capping enzyme are distinct and nonoverlapping. The guanylyltransferase domain is a 46-kDa monomer that extends from Mce1 residue 211 to residue 597. The recombinant domain is catalytically active in vitro and is capable of genetically complementing the function of the yeast guanylyltransferase Ceg1 in vivo. A recent study of the C. elegans enzyme Cel1 documented triphosphatase activity of an amino-terminal domain (13), but the guanylyltransferase activity could not be demonstrated. We found that the proximal margin of the active guanylyltransferase domain is located 84 amino acids upstream of the presumptive active site nucleophile Lys-294. This is in keeping with the positions of the N termini of the yeast and Chlorella virus guanylyltransferases 67-82 residues upstream of their respective lysine nucleophiles (6, 7, 16, 20, 22). The fact that the in vivo activity of Mce1 is abrogated by the K294A mutation supports our designation of this residue as the active site. Substitutions of the equivalent residues of Ceg1 and Pce1 are also lethal in vivo (6, 16, 22). The function of the Mce1 guanylyltransferase domain is abolished when residues 211-259 are deleted. Deletion to a similar point of the yeast guanylyltransferase Ceg1 is also lethal (16). In the crystal structure of Chlorella virus guanylyltransferase, the analogous amino-terminal segment is located on the protein surface and consists of two antiparallel -sheets followed by an -helix (21). This region contains no residues that make direct contact with GTP in the enzyme-GTP cocrystal (21). The fact that recombinant Mce1(260-597) expressed in bacteria was predominantly insoluble complicated efforts to distinguish whether the loss of activity upon amino-terminal truncation was caused by defective protein folding or a primary defect in catalysis.

1

Evolution of the Capping Apparatus-- The linear order of amino-terminal RNA triphosphatase and carboxyl-terminal guanylyltransferase domains in Mce1 is superficially similar to the arrangement of the triphosphatase and guanylyltransferase active sites in the vaccinia capping enzyme (24-27). However, the amino acid sequences of the Mce1 and vaccinia amino-terminal segments are unrelated, and the properties of the vaccinia triphosphatase differ from those of the metazoan triphosphatases in two key respects: (i) lack of domain autonomy, and (ii) divalent cation dependence. The vaccinia triphosphatase and guanylyltransferase active sites are distinct, but the two activities are not partitioned into discrete and separable structural modules as they are in Mce1. Instead, moieties on the vaccinia enzyme that are essential for triphosphatase and guanylyltransferase activity overlap within a 545-amino acid polypeptide (23, 26, 27). The triphosphatase activity of the vaccinia capping enzyme depends absolutely on a divalent cation cofactor, whereas the rat liver and brine shrimp triphosphatases, as well as the recombinant C. elegans and mouse triphosphatases, require no divalent cation for activity (in fact, they are inhibited by divalent cations). The sequences of the C. elegans and mouse enzymes indicate that -phosphate cleavage occurs through a phosphoenzyme intermediate. Mutation of the active site cysteine of the C. elegans triphosphatase domain abolished enzyme activity (13). We have been unable to detect a phosphoenzyme intermediate for the vaccinia triphosphatase, which suggests that covalent catalysis does not apply. Mutational analysis of the vaccinia triphosphatase provides additional evidence for a distinct mechanism. We have identified four acidic side chains that are essential for catalysis by vaccinia triphosphatase and are conserved among the poxvirus and African swine fever virus capping enzymes (26, 27). One or more of these acidic residues is likely to bind the essential metal ion(s). The RNA triphosphatase isolated from S. cerevisiae, also depends completely on a divalent cation cofactor (28). The amino acid sequence of yeast Cet1 bears no similarity to the mammalian triphosphatase domains, but it does display local similarities to the vaccinia triphosphatase. We surmise that higher eukaryotes have diverged from vaccinia and yeast with respect to both mechanism and structure of the triphosphatase component of the capping machinery. In contrast, the reaction mechanism and structure of the guanylyltransferase components are conserved in yeasts, metazoans, and DNA viruses (14, 21).

Interaction of the Capping Apparatus with the CTD of RNA Polymerase II-- Cap formation in vivo is targeted to the nascent chains synthesized by pol II. Placing a mammalian pol II transcription unit under the control of a pol III promoter results in a failure to cap the transcript (29). We have elaborated a solution to the problem of how pol II transcripts are specifically singled out for capping whereby the capping enzymes are targeted to pre-mRNA by binding to the phosphorylated CTD of the largest subunit of RNA polymerase II (2). The CTD, which is unique to pol II, consists of a tandem array of a heptapeptide repeat with the consensus sequence Tyr-Ser-Pro-Thr-Ser-Pro-Ser. The mammalian pol II large subunit has 52 tandem repeats, whereas the S. cerevisiae subunit has 27 copies. The pol II largest subunit exists in two forms, a nonphosphorylated IIA form and a phosphorylated IIO form, which are interconvertible and functionally distinct. In vivo, the pol IIO enzyme contains as many as 50 phosphorylated amino acids (primarily phosphoserine) within the CTD (30). During transcription initiation, pol IIA is recruited to the DNA template by the general transcription factors. The pol IIA CTD undergoes extensive phosphorylation and conversion to IIO during the transition from preinitiation complex to stable elongation complex. Several CTD kinase activities have been implicated in CTD hyperphosphorylation, each of which contains a cyclin and cyclin-dependent kinase subunit pair. The cdk7 and cyclin H subunits of the general transcription factor TFIIH catalyze phosphorylation of Ser-5 of the CTD heptapeptide (31). Other CTD kinases include the cdk8/cylin C pair found in the pol II holoenzyme (32, 33), CTDK-I, a heterotrimeric kinase with cdk-like and cyclin-like subunits (34), and P-TEFb, a regulator of polymerase elongation with a cdc2-like subunit (35, 36).