Domain Structure and Mutational Analysis of T4 Polynucleotide Kinase
2001; Elsevier BV; Volume: 276; Issue: 29 Linguagem: Inglês
10.1074/jbc.m103663200
ISSN1083-351X
Autores Tópico(s)RNA and protein synthesis mechanisms
ResumoT4 polynucleotide kinase (Pnk) is the founding member of a family of 5′-kinase/3′-phosphatase enzymes that heal broken termini in RNA or DNA by converting 3′-PO4/5′-OH ends into 3′-OH/5′-PO4 ends, which are then suitable for sealing by RNA or DNA ligases. Here we employed site-directed mutagenesis and biochemical methods to dissect the domain structure of the homotetrameric T4 Pnk protein and to localize essential constituents of the apparently separate active sites for the 5′-kinase and 3′-phosphatase activities. We characterized deletion mutants Pnk(42–301) and Pnk(1–181), which correspond to domains defined by proteolysis with chymotrypsin. Pnk(1–181) is a monomer with no 3′-phosphatase and low residual 5′-kinase activity. Pnk(42–301) is a dimer with no 5′-kinase and low residual 3′-phosphatase activity. Four classes of missense mutational effects were observed. (i) Mutations K15A, S16A, and D35A inactivated the 5′-kinase but did not affect the 3′-phosphatase or the tetrameric quaternary structure of T4 Pnk. 5′-kinase activity was ablated by the conservative mutations K15R, K15Q, and D35N; however, kinase activity was restored by the S16T change. (ii) Mutation D167A inactivated the 3′-phosphatase without affecting the 5′-kinase or tetramerization. (iii) Mutation D85A caused a severe decrement in 5′-kinase activity and only a modest effect on the 3′-phosphatase; the nearby N87A mutation resulted in a significantly reduced 3′-phosphatase activity and slightly reduced 5′-kinase activity. D85A and N87A both affected the quaternary structure, resulting in a mixed population of tetramer and dimer species. (iv) Alanine mutations at 11 other conserved positions had no significant effect on either 5′-kinase or 3′-phosphatase activity. T4 polynucleotide kinase (Pnk) is the founding member of a family of 5′-kinase/3′-phosphatase enzymes that heal broken termini in RNA or DNA by converting 3′-PO4/5′-OH ends into 3′-OH/5′-PO4 ends, which are then suitable for sealing by RNA or DNA ligases. Here we employed site-directed mutagenesis and biochemical methods to dissect the domain structure of the homotetrameric T4 Pnk protein and to localize essential constituents of the apparently separate active sites for the 5′-kinase and 3′-phosphatase activities. We characterized deletion mutants Pnk(42–301) and Pnk(1–181), which correspond to domains defined by proteolysis with chymotrypsin. Pnk(1–181) is a monomer with no 3′-phosphatase and low residual 5′-kinase activity. Pnk(42–301) is a dimer with no 5′-kinase and low residual 3′-phosphatase activity. Four classes of missense mutational effects were observed. (i) Mutations K15A, S16A, and D35A inactivated the 5′-kinase but did not affect the 3′-phosphatase or the tetrameric quaternary structure of T4 Pnk. 5′-kinase activity was ablated by the conservative mutations K15R, K15Q, and D35N; however, kinase activity was restored by the S16T change. (ii) Mutation D167A inactivated the 3′-phosphatase without affecting the 5′-kinase or tetramerization. (iii) Mutation D85A caused a severe decrement in 5′-kinase activity and only a modest effect on the 3′-phosphatase; the nearby N87A mutation resulted in a significantly reduced 3′-phosphatase activity and slightly reduced 5′-kinase activity. D85A and N87A both affected the quaternary structure, resulting in a mixed population of tetramer and dimer species. (iv) Alanine mutations at 11 other conserved positions had no significant effect on either 5′-kinase or 3′-phosphatase activity. polynucleotide kinase polymerase chain reaction isopropyl-β-d-thiogalactopyranoside polyacrylamide gel electrophoresis bovine serum albumin Polynucleotide kinase (Pnk)1 was discovered 35 years ago in extracts of Escherichia coli infected with T-even bacteriophage (1Richardson C.C. Proc. Natl. Acad. Sci. U. S. A. 1965; 54: 158-165Crossref PubMed Scopus (493) Google Scholar, 2Novogrodsky A. Hurwitz J. J. Biol. Chem. 1966; 241: 2923-2932Abstract Full Text PDF PubMed Google Scholar, 3Novogrodsky A. Tal M. Traub A. Hurwitz J. J. Biol. Chem. 1966; 241: 2933-2943Abstract Full Text PDF PubMed Google Scholar). Pnk catalyzes the transfer of the γ phosphate from ATP or other nucleoside triphosphates to the 5′-OH terminus of DNA or RNA to form a 5′-monophosphate polynucleotide and a nucleoside diphosphate; the enzyme also phosphorylates the 5′-OH of nucleoside 3′-monophosphates (1Richardson C.C. Proc. Natl. Acad. Sci. U. S. A. 1965; 54: 158-165Crossref PubMed Scopus (493) Google Scholar, 2Novogrodsky A. Hurwitz J. J. Biol. Chem. 1966; 241: 2923-2932Abstract Full Text PDF PubMed Google Scholar, 3Novogrodsky A. Tal M. Traub A. Hurwitz J. J. Biol. Chem. 1966; 241: 2933-2943Abstract Full Text PDF PubMed Google Scholar). The use of T4 Pnk to label 5′-DNA or RNA ends with 32P was instrumental in the subsequent identification of DNA and RNA ligases and in the development of methods for the analysis of nucleic acid structure, molecular cloning, and nucleic acid sequencing. Early biochemical studies showed that the kinase reaction is reversible (4van de Sande J.H. Kleppe K. Khorana H.G. Biochemistry. 1973; 12: 5050-5055Crossref PubMed Scopus (86) Google Scholar). Although it was initially proposed that T4 Pnk acts via a ping-pong mechanism with a phosphoenzyme intermediate (4van de Sande J.H. Kleppe K. Khorana H.G. Biochemistry. 1973; 12: 5050-5055Crossref PubMed Scopus (86) Google Scholar), kinetic analysis favors a sequential mechanism in which both ATP and the 5′-OH polynucleotide bind the enzyme prior to dissociation of either product (5Lillehaug J.R. Kleppe K. Biochemistry. 1975; 14: 1221-1225Crossref PubMed Scopus (57) Google Scholar). The stereochemical course of the reaction entails inversion of configuration of the transferred phosphate, indicative of an in-line mechanism in which the polynucleotide 5′-OH directly attacks the γ phosphorus of ATP (6Jarvest R.L. Lowe G. Biochem. J. 1981; 199: 273-276Crossref PubMed Scopus (21) Google Scholar). T4 Pnk is a bifunctional enzyme with an intrinsic 3′-phosphomonoesterase activity that removes 3′-PO4 termini from DNA or RNA polynucleotides (7Becker A. Hurwitz J. J. Biol. Chem. 1967; 242: 936-950Abstract Full Text PDF PubMed Google Scholar, 8Cameron V. Uhlenbeck O.C. Biochemistry. 1977; 16: 5120-5126Crossref PubMed Scopus (226) Google Scholar). The kinase and phosphatase active sites are apparently distinct insofar as: (i) the kinase and phosphatase functions have widely divergent pH optima; (ii) the activities are differentially sensitive to the protein-modifying agents diethyl pyrocarbonate, which inactivates the kinase, andN-ethylmaleimide, which inactivates the phosphatase; (iii) the activities are differentially sensitive to digestion of the enzyme with trypsin, which inactivates the kinase, and carboxypeptidase, which inactivates the phosphatase; and (iv) a mutant T4 allele (pseT-1) encodes an enzyme that has normal polynucleotide kinase activity but no 3′-phosphatase activity (8Cameron V. Uhlenbeck O.C. Biochemistry. 1977; 16: 5120-5126Crossref PubMed Scopus (226) Google Scholar, 9Soltis D.A. Uhlenbeck O.C. J. Biol. Chem. 1982; 257: 11340-11345Abstract Full Text PDF PubMed Google Scholar, 10Soltis D.A. Uhlenbeck O.C. J. Biol. Chem. 1982; 257: 11332-11339Abstract Full Text PDF PubMed Google Scholar). Pnk purified from T4-infected bacteria is reported to be either a homotetramer (8Cameron V. Uhlenbeck O.C. Biochemistry. 1977; 16: 5120-5126Crossref PubMed Scopus (226) Google Scholar, 12Panet A. van de Sande J.H. Loewen P.C. Khorana H.G. Raae A.J. Lillehaug J.R. Kleppe K. Biochemistry. 1973; 12: 5045-5050Crossref PubMed Scopus (223) Google Scholar) or a homodimer (10Soltis D.A. Uhlenbeck O.C. J. Biol. Chem. 1982; 257: 11332-11339Abstract Full Text PDF PubMed Google Scholar) of a 33–35-kDa polypeptide. Gene cloning and sequencing revealed that the T4 kinase/phosphatase is a 301-amino acid polypeptide with a predicted size of 34 kDa (11Midgley C.A. Murray N.E. EMBO J. 1985; 4: 2695-2703Crossref PubMed Scopus (49) Google Scholar). The amino acid sequence of T4 Pnk is remarkable for the presence of a nucleotide-binding motif (a Walker A box) near the N terminus and N-terminal sequence similarity to adenylate kinase. There have been no reported efforts to finely map the active sites of this historically important enzyme or to elucidate its structure during the many years since its gene was cloned. As a bifunctional enzyme with a complex quaternary structure and a fascinating biological role in the repair of “broken” RNA molecules (13Amitsur M. Levitz R. Kaufman G. EMBO J. 1987; 6: 2499-2503Crossref PubMed Scopus (205) Google Scholar), T4 kinase is ripe for structure-function investigation. Here we present: (i) a mutational analysis that identifies individual amino acid side chains essential for the 5′-kinase and 3′-phosphatase activities; (ii) low-resolution dissection of structural domains by limited proteolysis and peptide sequencing; (iii) the identification of functional domains based on the proteolytic map; and (iv) the delineation of protein segments that contribute to a tetrameric quaternary structure. Oligodeoxynucleotide primers complementary to the 5′- and 3′-ends of the Pnk gene were used to PCR amplify the open reading frame from T4 genomic DNA (a gift of Dr. Ken Kreuzer, Duke University). The primers were designed to introduce NdeI and BamHI restriction sites at the 5′- and 3′-ends of the gene. The PCR product was digested with NdeI and BamHI and then cloned into the NdeI and BamHI sites of the bacterial expression plasmid pET16b (Novagen) to yield pET-PNK. Dideoxy sequencing of the entire insert of pET-PNK confirmed that no alterations of the genomic T4 DNA sequence were introduced during PCR amplification and cloning of the kinase gene. Missense mutations were introduced into the Pnk gene by using the two-stage PCR-based overlap extension method (14Ho S.N. Hunt H.D. Horton R.M. Pullen J.K. Pease L.R. Gene. 1989; 77: 51-59Crossref PubMed Scopus (6833) Google Scholar). pET-PNKwas used as the template for the first stage PCR reaction.NdeI-BamHI restriction fragments of the mutated second-stage PCR products were inserted into pET16b. The inserts of the resulting plasmids were sequenced to confirm the presence of the desired mutations and the absence of any unwanted coding changes. An N-terminal deletion mutant, Pnk(42–301), was constructed by PCR amplification of the PNK gene with a sense-strand primer that introduced an NdeI restriction site at Met42. A C-terminal deletion mutant, Pnk(1–181), was constructed by PCR amplification with an antisense-strand primer that introduced a stop codon in lieu of the codon for Glu182 and aBamHI site immediately 3′ of the stop codon.NdeI-BamHI restriction fragments containing the truncated genes were inserted into pET16b. The resulting pET-PNKΔN and pET-PNKΔC plasmids were sequenced to exclude the introduction of any unwanted coding changes during amplification and cloning. The wild-type and mutant pET-PNK plasmids were transformed into E. coli BL21(DE3). Single ampicillin-resistant colonies were inoculated into LB medium containing 0.1 mg/ml ampicillin, and 50-ml cultures were grown at 37 °C until the A 600reached 0.5. The cultures were placed on ice for 30 min, then adjusted to 0.3 mm isopropyl-β-d-thiogalactopyranoside (IPTG), and subsequently incubated at 17 °C for 15 h with continuous shaking. Cells were harvested by centrifugation, and the pellets were stored at −80 °C. All subsequent procedures were performed at 4 °C. Cell lysis was achieved by treatment of thawed, resuspended cells with 1 mg/ml lysozyme and 0.1% Triton X-100 in 5 ml of lysis buffer containing 50 mm Tris-HCl (pH 7.5), 1.2m NaCl, 15 mm imidazole, 10% glycerol, 1 mm benzamidine, and 0.2 mm phenylmethylsulfonyl fluoride. The lysates were sonicated to reduce viscosity, and insoluble material was removed by centrifugation at 40,000 × g for 20 min. The supernatants were applied to 1-ml columns of Ni-NTA-agarose (Qiagen). The columns were washed with 5 ml of lysis buffer and then with 2 ml of buffer A (50 mm Tris-HCl (pH 7.5), 200 mm NaCl, 10% glycerol) containing 0.1m imidazole. The columns were step-eluted with 0.3, 0.5, and 1.3 m imidazole in buffer A. The polypeptide compositions of the column fractions were monitored by SDS-PAGE. The peak fractions containing the Pnk polypeptide were pooled and stored at −80 °C. The protein concentrations were determined by using the Bio-Rad dye reagent with bovine serum albumin as the standard. Reaction mixtures (10 µl) containing 70 mm Tris-HCl (pH 7.6), 10 mm MgCl2, 5 mm dithiothreitol, 25 µm[γ-32P]ATP, ∼50 pmol of a synthetic 5′-OH DNA oligonucleotide d(ATTCCGATAGTGACTACA), and Pnk as specified were incubated for 20 min at 37 °C. The reactions were quenched by adding 6 µl of stop solution (95% formamide, 20 mm EDTA, 0.05% bromphenol blue, 0.05% xylene cyanol). The mixtures were analyzed by electrophoresis through a 15-cm 15% polyacrylamide gel containing 7m urea in TBE (90 mm Tris borate, 2.5 mm EDTA). The radiolabeled oligonucleotide products were visualized and quantitated by scanning the gel with a Fujix BAS2500 phosphorimaging device. 3′-Phosphatase activity was measured by the conversion of a 5′-32P-labeled, 3′-PO4 DNA oligonucleotide d(pATTCCGATAGTGACTACAp) to a 3′-OH derivative. The substrate was prepared by enzymatic phosphorylation of the 5′-OH, 3′-PO4 strand d(ATTCCGATAGTGACTACAp) in the presence of [γ-32P]ATP and the 3′-phosphatase-defective deletion mutant Pnk(1–181). The 5′-32P-labeled, 3′-PO4 oligonucleotide was purified by electrophoresis through a 20% native polyacrylamide gel and recovered by electroelution from an excised gel slice into 0.5 ml of 0.2× TBE. 3′-Phosphatase reaction mixtures (10 µl) containing 100 mm imidazole (pH 6.0), 10 mm MgCl2, 10 mm β-mercaptoethanol, 0.1 mg/ml BSA, 1 pmol of 5′-32P-labeled 3′-PO4 oligonucleotide and Pnk as specified were incubated for 20 min at 37 °C. The reactions were quenched by adding 6 µl of stop solution, and the mixtures were analyzed by electrophoresis through a 40-cm 16% polyacrylamide gel containing 7 m urea in TBE. The radiolabeled 3′-OH-containing product, which migrated more slowly than the 3′-PO4-containing substrate oligonucleotide, was quantitated by scanning the gel with a phosphorimaging device. Aliquots (40 µg) of the wild-type and mutant Pnk preparations were mixed with catalase (30 µg), BSA (30 µg), and cytochrome c (30 µg), and the mixtures were applied to 4.8-ml 15–30% glycerol gradients containing 50 mm Tris-HCl (pH 8.0), 0.2 m NaCl, 1 mm EDTA, 2.5 mm dithiothreitol, and 0.1% Triton X-100. The gradients were centrifuged in a SW50 rotor at 50,000 rpm for 14 h at 4 °C. Finer sedimentation analysis of the Pnk deletion mutants was performed by mixing Pnk(42–301) or Pnk(1–181) (40 µg) with BSA (30 µg), ovalbumin (30 µg), and cytochromec (30 µg) and centrifuging the mixtures through 4.8-ml 15–30% glycerol gradients in a SW50 rotor at 50,000 rpm for 24 h at 4 °C. Fractions were collected from the bottoms of the tubes. Aliquots (20 µl) of odd-numbered gradient fractions were analyzed by SDS-PAGE. A goal of the present study was to map the active site of the 5′-OH polynucleotide kinase by identifying individual amino acid functional groups required for the kinase reaction. To do this, we introduced single alanine changes at 17 amino acids in the Pnk polypeptide that are conserved in its closest homologue, a putative polynucleotide kinase/ligase encoded by theAutographa californica nuclear polyhedrosis virus (15Durantel D. Croizier L. Ayres M.D. Croizier G. Possee R.D. Lopez-Ferber M. J. Gen. Virol. 1998; 79: 629-637Crossref PubMed Scopus (37) Google Scholar). The mutated positions are highlighted by asterisks in Fig.1. We focused in particular on the A box motif GXXGXGKS near the N terminus of T4 Pnk, which is present in many NTP-dependent phosphohydrolases and phosphotransferases. The core A box, as well as multiple vicinal residues, are identical in the T4 and baculovirus proteins (Fig. 1). The genes encoding wild-type Pnk and the 17 Pnk-Ala mutants were cloned into a T7 RNA polymerase-based bacterial expression vector so as to fuse the 301-amino acid Pnk to an N-terminal leader peptide containing 10 tandem histidines. The expression plasmids were introduced intoE. coli BL21(DE3), a strain that contains the T7 RNA polymerase gene under the control of a lacUV5 promoter. A prominent ∼35-kDa polypeptide was detectable by SDS-PAGE in whole-cell extracts of IPTG-induced bacteria (not shown). (The calculated size of the His-Pnk polypeptide is 37 kDa.) This polypeptide was not present when bacteria containing the pET vector alone were induced with IPTG. After centrifugal separation of the crude lysate, the Pnk protein was recovered in the soluble supernatant fraction. The His tag facilitated rapid purification of recombinant Pnk by adsorption to an immobilized nickel resin and subsequent elution with buffer containing imidazole. SDS-PAGE analysis of the imidazole eluate fractions showed that the wild-type Pnk and the 17 Pnk-Ala mutants preparation were highly enriched with respect to the His-Pnk polypeptide (Fig. 2 A). The identity of the His-Pnk protein was confirmed by N-terminal sequence analysis (see below). 5′-Kinase activity was measured by label transfer from [γ-32P]ATP to a 5′-OH DNA oligonucleotide. All of the enzyme preparations were assayed in parallel; the kinase reaction mixtures contained 80 ng of input Pnk, an amount sufficient for saturating levels of DNA-labeling by the wild-type enzyme (see Fig.8 B). Conducting the screening assays in this fashion highlighted the most severe mutational effects on catalysis (Fig.2 B). Four mutants were grossly defective in catalysis: K15A, S16A, D35A, and D85A. Two of the essential residues, Lys15and Ser16, are constituents of the A box. One mutation, N87A, elicited a partial loss of function. The other twelve mutant proteins retained 5′-kinase activity: G12A, S13A, G14A, T17A, W18A, R20A, N33A, R34A, T86A, D167A, G168A, and T169A (Fig. 2 B). We surmise that none of these 12 side chains contributes significantly to catalysis of the kinase reaction. The native sizes of the wild-type Pnk and the catalytically defective mutants were investigated by sedimentation through 15–30% glycerol gradients. Marker proteins catalase, BSA, and cytochrome c were included as internal standards. After centrifugation, the polypeptide compositions of the odd-numbered gradient fractions were analyzed by SDS-PAGE (Fig.3). The 37-kDa wild-type Pnk polypeptide sedimented as a discrete peak between catalase (248 kDa) and BSA (66 kDa), consistent with a globular tetrameric quaternary structure for Pnk as reported previously (8Cameron V. Uhlenbeck O.C. Biochemistry. 1977; 16: 5120-5126Crossref PubMed Scopus (226) Google Scholar, 12Panet A. van de Sande J.H. Loewen P.C. Khorana H.G. Raae A.J. Lillehaug J.R. Kleppe K. Biochemistry. 1973; 12: 5045-5050Crossref PubMed Scopus (223) Google Scholar). The catalytically defective K15A, S16A, and D35A proteins sedimented similarly, indicating that the 5′-kinase-inactivating mutations did not grossly affect the quaternary structure of Pnk. We infer that the Lys15, Ser16, and Asp35 side chains are essential constituents of the 5′-kinase active site. The D85A mutation, on the other hand, resulted in a subtle, reproducible alteration in the sedimentation profile, whereby the major component of the D85A protein sedimented less rapidly than wild-type Pnk, at a position closer to the BSA standard, and a minor component of D85A cosedimented with BSA (Fig.4 A). A simple interpretation of the sedimentation profile is that the leading D85A component is an asymmetric tetramer and the lagging component is a dimer. This result implicates Asp85 in subunit-subunit interactions (directly or indirectly) and suggests a more complex interpretation of the deleterious effects of D85A on 5′-kinase activity. Conservative substitutions were introduced at each of the four residues that were defined by alanine scanning as essential for 5′-kinase function in vitro. The seven recombinant proteins, K15R, K15Q, S16T, D35N, D35E, D85N, and D85E, were purified from soluble bacterial extracts by nickel-agarose chromatography (Fig.5 A). Replacement of the A box lysine (Lys15) by either glutamine or arginine did not restore 5′-kinase activity (Fig. 5 B). These data establish a clear requirement for a positively charged residue, and lysine specifically, at this critical position of T4 Pnk. 5′-kinase activity was restored fully when the A box serine (Ser16) was replaced by threonine. Hence, the hydroxyl moiety is critical for the function of this residue in Pnk. Replacement of Asp35 by asparagine was as deleterious to the 5′-kinase activity as the D35A mutation. A low level of activity was restored by the D35E mutation (Fig. 5 B). An acidic side chain at position 35 is evidently essential for catalysis, and there appears to be a steric constraint that precludes full function of the larger glutamate side chain. Strikingly different structure-function relationships were seen at Asp85 where introduction of asparagine restored 5′-kinase activity, whereas a glutamate did not (Fig. 5 B). We surmise that: (i) the acidic carboxylate at position 85 is not critical for 5′-kinase activity; (ii) sufficiency of the amide functional group implies an important hydrogen-bonding interaction; and (iii) there is a steric constraint that hinders the function of glutamate in lieu of aspartate. 3′-Phosphatase activity was measured by conversion of a 5′-32P-labeled oligonucleotide containing a 3′-PO4 into a 5′-32P-labeled 3′-OH-terminated product. The wild-type Pnk and all 17 Pnk-Ala mutants were assayed in parallel; the 3′-phosphatase reaction mixtures contained 2 ng of input Pnk, an amount sufficient for saturating levels of dephosphorylation by the wild-type enzyme (see Fig. 8 C). The D167A mutation elicited the most severe mutational effect on 3′-phosphatase function (Fig. 6). Glycerol gradient analysis showed that the D167A mutant sedimented as a tetramer (Fig. 3). Because the D167A mutation did not affect 5′-kinase function or protein quaternary structure, we construe that Asp167 is a specific constituent of the 3′-phosphatase active site. Fourteen of the alanine mutants retained 3′-phosphatase activity: G12A, S13A, G14A, K15A, S16A, T17A, W18A, R20A, N33A, R34A, D35A, T86A, G168A, and T169A (Fig. 6). It was especially instructive that K15A, S16A, and D35A, which were defective in 5′-kinase activity, were catalytically competent in 3′ dephosphorylation. These data provide strong evidence that the 5′-kinase and 3′-phosphatase active sites are distinct, and they show that the amino acid side chains in and around the A box play no significant role in the 3′-phosphatase reaction. The N87A mutation resulted in a significant reduction in 3′-phosphatase activity (Fig. 6); this mutation also caused a modest reduction in 5′-kinase activity (Fig. 2 B). The nearby D85A mutation, which caused a severe decrement in 5′-kinase activity, had only a modest effect on the 3′-phosphatase (Fig. 6). Glycerol gradient sedimentation of N87A showed two components: a heavier, presumptive tetrameric species sedimenting between catalase and BSA and a lighter species cosedimenting with BSA, which we presume is a dimer (Fig.4 B). The 5′-kinase activity profile of N87A across the glycerol gradient showed that catalytic activity was exclusive to the fraction corresponding to the tetrameric component (not shown). These results suggest several models to account for the effects of the D85A and N87A mutations: (i) the peptide segment85DXN87 may comprise a common component of the 5′-kinase and 3′-phosphatase active sites; (ii) Asp35 and Asn87 are not located at the active sites, and the mutational effects on 5′-kinase and 3′-phosphatase activity are manifestations of a key role for the85DXN87 peptide at the subunit-subunit interface. The N87A sedimentation results suggest that tetramer formation is important for the 5′-kinase activity (see below). The purified His-tagged Pnk was subjected to proteolysis with increasing amounts of chymotrypsin (Fig.7). N-terminal sequencing of the undigested Pnk polypeptide by automated Edman chemistry after transfer from an SDS-gel to a polyvinylidene difluoride membrane confirmed that the N-terminal sequence started from the second residue of the His tag. Apparently, the Pnk suffered removal of the initiating methionine during expression in E. coli. Initial scission of the 37-kDa Pnk by chymotrypsin yielded four major products, C1, C2, C3, and C5 (Fig. 7 A, lane 10). N-terminal sequencing showed that the C1 species consisted of two polypeptides: one arising via chymotryptic cleavage between Tyr30 and Asn31and one generated by cleavage between Met42 and Ala43 (Fig. 7 B). The sites of proteolysis are demarcated by arrows above the Pnk sequence in Fig. 1. The C2 species was also composed of two polypeptides: one cleaved at Tyr50/Lys51 and one at Tyr52/Lys53 (Figs. 1 and 7 B). (The predicted N-terminal products of these cleavages were too small to detect in this gel-based assay.) The C3 species, which retained the His tag, and the C5 species, starting at Asp180, likely arose as the products of a single proteolytic cut at Tyr179/Asp180. High levels of chymotrypsin, in excess of the amount sufficient to cleave all the input Pnk, resulted in the decay of the C1, C2, and C3 species and the appearance of a major lower molecular weight cleavage product C4 (Fig.7 A). A scheme to account for the chymotryptic digestion pattern is shown in Fig. 7 B. The results suggest parallel pathways entailing alternative initial cleavages near the N terminus or else internally at Tyr179. The pathways converge after secondary cleavages of the major products C4 and C5. We surmise that the C4 and C5 fragments are themselves well folded insofar as they were resistant to digestion by concentrations of chymotrypsin sufficient to cleave all the input Pnk. In light of the proteolysis results, we engineered an N-terminal deletion mutant Pnk(42–301) and a C-terminal truncation Pnk(1–181), referred to henceforth as ΔN and ΔC, respectively. The ΔN protein, corresponding to proteolytic species C1, and ΔC protein, corresponding to proteolytic species C3, were produced in bacteria as N-terminal His10-fusions and purified from soluble lysates by nickel-agarose chromatography (Fig.8 A). The ΔN protein was inert with respect to 5′-kinase activity (Fig.8 B); this result was expected given that the A box essential for kinase function was encompassed within the deleted N-terminal peptide (Fig. 1). The ΔC protein retained 5′-kinase function; however, the specific activity was only 5% that of the full-length Pnk (Fig. 8 B). In addition, the dependence of 5′-kinase activity on ΔC concentration was conspicuously sigmoidal. The deletions elicited the converse effects on the 3′-phosphatase activity, i.e. ΔC was inert as a 3′-phosphatase, whereas ΔN retained 3′-phosphatase function, albeit with only 3% the specific activity of full-length Pnk (Fig. 8 C). These findings suggest that an essential constituent of the phosphatase active site is located within the C-terminal domain, Pnk(182–301). Analysis of ΔN and ΔC by glycerol gradient sedimentation under the conditions used for full-sized Pnk and the Pnk-Ala mutants indicated that both deletion variants sedimented as single components at positions between BSA and cytochrome c (not shown); the absence of a component sedimenting with or heavier than BSA suggested that neither ΔC nor ΔN retained the tetrameric quaternary structure characteristic of full-sized Pnk. To better gauge the native size of the truncated proteins, we sedimented them for a longer time in the presence of BSA (66 kDa), ovalbumin (45 kDa), and cytochromec (13 kDa) as internal standards (Fig.9). The ΔC polypeptide, with a calculated mass of 23 kDa, sedimented as a discrete peak between ovalbumin and cytochrome c, consistent with a monomeric native structure for the ΔC protein. In contrast, the ΔN polypeptide, with a calculated mass of 32 kDa, sedimented a discrete peak at a position between BSA and ovalbumin, which indicated that the ΔN protein is a homodimer (Fig. 9). We surmise that the C-terminal domain is essential for attaining any oligomeric structure of Pnk, whereas the N-terminal peptide is required specifically for the transition from dimer to tetramer. T4 Pnk is the founding member of a growing family of 5′-kinase/3′-phosphatase enzymes that heal broken termini in RNA or DNA by converting 3′-PO4/5′-OH ends into 3′-OH/5′-PO4 ends, which are then suitable for sealing by RNA or DNA ligases (13Amitsur M. Levitz R. Kaufman G. EMBO J. 1987; 6: 2499-2503Crossref PubMed Scopus (205) Google Scholar, 16Jilani A. Ramotar D. Slack C. Ong C. Yang X.M. Scherer S.W. Lasko D.D. J. Biol. Chem. 1999; 274: 24176-24186Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar, 17Karimi-Busheri F. Daly G. Robins P. Canas B. Pappin D.J.C. Sgouros J. Miller G.G. Fakhrai H. Davis E.M. Le Beau M.M. Weinfeld M. J. Biol. Chem. 1999; 274: 24187-24194Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar). Here we have employed site-directed mutagenesis and biochemical methods to dissect the domain structure of T4 Pnk and to localize essential constituents of the apparently separate active sites for the 5′-kinase and 3′-phosphatase activities. Our results confirm and extend the early work of Soltis and Uhlenbeck (9Soltis D.A. Uhlenbeck O.C. J. Biol. Chem. 1982; 257: 11340-11345Abstract Full Text PDF PubMed Google Scholar), which implicated the N terminus of T4 Pnk in the 5′-kinase reaction and the C terminus in the 3′-phosphatase function. Our studies were performed using recombinant Pnk protein containing a short N-terminal polyhistidine tag. To determine whether the His tag affects the kinase activity or quaternary structure of T4 Pnk, we produced and purified recombinant Pnk protein containing an N-terminal His-Smt3 domain, then removed the His-Smt3 domain via digestion with Ulp1, a cysteine protease that hydrolyzes the polypeptide backbone exclusively at the junction of His-Smt3 with the downstream polypeptide (28Mossessova E. Lima C.D. Mol. Cell. 2000; 5: 865-876Abstract Full Text Full Text PDF PubMed Scopus (570) Google Scholar). We then purified the untagged Pnk away from the His-Smt3 domain. Pnk lacking the tag sedimented in a glycerol gradient as a discrete tetrameric species and its 5′-kinase specific activity was equivalent to that of the wild-type His-Pnk enzyme used in the studies presented above (data not shown). Therefore, the His tag is unlikely to affect the structure-function relationships illuminated here. We find that the N terminus is required for the 5′-kinase activity of T4 Pnk because it contains the A box element that contributes two essential side chains (Lys15 and Ser16) to the kinase active site. Crystallographic studies of other A box proteins show that the invariant lysine makes direct contact with the β and γ phosphates of the bound nucleotide (18Pai E.F. Krengel U. Petsko G.A. Goody R.S. Kabsch W. Wittinghofer A. EMBO J. 1990; 9: 2351-2359Crossref PubMed Scopus (966) Google Scholar, 19Müller C.W. Schulz G.E. J. Mol. Biol. 1992; 224: 159-177Crossref PubMed Scopus (433) Google Scholar). Our findings that the K15A, K15R, and K15Q mutations inactivate the 5′-kinase are in accord with mutational analyses of several other A box proteins in which the lysine cannot be functionally substituted by arginine (20Sung P. Higgins D. Prakash L. Prakash S. EMBO J. 1988; 7: 3263-3269Crossref PubMed Scopus (221) Google Scholar, 21Byeon L. Shi Z. Tsai M.D. Biochemistry. 1995; 34: 3172-3182Crossref PubMed Scopus (48) Google Scholar, 22Martins A. Gross C.H. Shuman S. J. Virol. 1999; 73: 1302-1308Crossref PubMed Google Scholar). The amino acid adjacent to the A box lysine is typically either threonine or serine (albeit not in adenylate kinase). In T4 Pnk, a threonine is fully active in lieu of Ser16, whereas the S16A mutant is catalytically defective. In p21Ha-rasbound to GMPPNP, the serine hydroxyl in the GKS element interacts with the β phosphate and with magnesium coordinated to the β and γ phosphates (18Pai E.F. Krengel U. Petsko G.A. Goody R.S. Kabsch W. Wittinghofer A. EMBO J. 1990; 9: 2351-2359Crossref PubMed Scopus (966) Google Scholar). Similar contacts are proposed for the GKT motif in the RecA-ADP cocrystal (23Story R.M. Steitz T.A. Nature. 1992; 355: 374-376Crossref PubMed Scopus (560) Google Scholar). The motif I threonine side chains of Rep and PcrA complexed with ADP are also poised near the β phosphate (24Korolev S. Hsieh J. Gauss G.H. Lohman T.M. Waksman G. Cell. 1997; 90: 635-647Abstract Full Text Full Text PDF PubMed Scopus (432) Google Scholar,25Subramanya H.S. Bird L.E. Brannigan J.A. Wigley D.B. Nature. 1996; 384: 379-383Crossref PubMed Scopus (381) Google Scholar). We presume that Ser16 of Pnk engages in similar contacts with its NTP substrate. Asp35 of Pnk is located 20 amino acids downstream of the A box lysine and is critical for 5′-kinase activity. An aspartate is located at the equivalent position in adenylate kinase; this Asp coordinates a water that is in turn coordinated to the essential divalent cation cofactor (26Berry M.B. Phillips G.N. Proteins. 1998; 32: 276-288Crossref PubMed Scopus (98) Google Scholar). The aspartate also forms a salt bridge to an essential arginine. Our mutational findings (Fig. 5) are consistent with similar roles for Asp35 in the 5′-kinase reaction of T4 Pnk, i.e.only a carboxylate can engage in both the hydrogen bond to water and the salt bridge. Although we did not explicitly target the C terminus and the 3′-phosphatase in the present round of alanine-scanning mutagenesis, we did identify Asp167 as an essential, and presumptively catalytic, constituent of the 3′-phosphatase domain. This aspartate is conserved in the baculovirus Pnk homolog (which has not been characterized biochemically) and in the human DNA 5′-kinase/3′-phosphatase and its homologs in Schizosaccharomyces pombe and Caenorhabditis elegans (16Jilani A. Ramotar D. Slack C. Ong C. Yang X.M. Scherer S.W. Lasko D.D. J. Biol. Chem. 1999; 274: 24176-24186Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar, 17Karimi-Busheri F. Daly G. Robins P. Canas B. Pappin D.J.C. Sgouros J. Miller G.G. Fakhrai H. Davis E.M. Le Beau M.M. Weinfeld M. J. Biol. Chem. 1999; 274: 24187-24194Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar). Although it had been suggested based on sequence alignments that this aspartate resides within a putative phosphatase motif FDLDGTL (16Jilani A. Ramotar D. Slack C. Ong C. Yang X.M. Scherer S.W. Lasko D.D. J. Biol. Chem. 1999; 274: 24176-24186Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar,17Karimi-Busheri F. Daly G. Robins P. Canas B. Pappin D.J.C. Sgouros J. Miller G.G. Fakhrai H. Davis E.M. Le Beau M.M. Weinfeld M. J. Biol. Chem. 1999; 274: 24187-24194Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar), the present study provides the first evidence that the aspartate is actually important for the polynucleotide 3′-phosphatase reaction. We also show that the vicinal residues Gly168 and Thr169 are not important for 3′-phosphatase activity of T4 Pnk even though they are conserved in other Pnk-like proteins. The mapping of protease-resistant Pnk fragments, presumably corresponding to folded structural domains of Pnk, led us to characterize recombinant Pnk truncations corresponding to the species generated during proteolysis. We thereby found that the carboxyl domain from 182–301 is essential for both Pnk oligomerization and 3′-phosphatase activity. The residual 5′-kinase activity of the Pnk(1–181) protein displayed a sigmoidal dependence on protein concentration, which would be in keeping with a model whereby the 5′-kinase reaction is performed by an active site composed of residues donated by more than one protomer of Pnk. It would appear that a tetrameric quaternary structure is also important for the 3′-phosphatase, insofar as the deletion of 41 amino acids from the N terminus converted the Pnk(42–301) domain from a tetramer into a homodimer and elicited a significant decrement in the 3′-phosphatase specific activity. Our initial efforts to determine whether the C-terminal domain Pnk(181–301) per se was capable of homodimerization were hampered by the insolubility of the Pnk(181–301) protein produced in bacteria. 2L. K. Wang, unpublished material. In summary, our findings suggest a complex functional organization of T4 Pnk in which the 5′-kinase and 3′-phosphatase sites are distinct, and full catalytic activity depends on a proper tetrameric quaternary structure. We anticipate that further rounds of mutational analysis of T4 Pnk, in tandem with efforts to crystallize the enzyme, will illuminate the phosphoryl transfer and phosphohydrolase reaction mechanisms.
Referência(s)