The 3′–5′ proofreading exonuclease of archaeal family-B DNA polymerase hinders the copying of template strand deaminated bases

Archaeal family B polymerases bind tightly to the deaminated bases uracil and hypoxanthine in single-stranded DNA, stalling replication on encountering these pro-mutagenic deoxynucleosides four steps ahead of the primer–template junction. When uracil is specifically bound, the polymerase–DNA complex exists in the editing rather than the polymerization conformation, despite the duplex region of the primer-template being perfectly base-paired. In this article, the interplay between the 3′–5′ proofreading exonuclease activity and binding of uracil/hypoxanthine is addressed, using the family-B DNA polymerase from Pyrococcus furiosus. When uracil/hypoxanthine is bound four bases ahead of the primer–template junction (+4 position), both the polymerase and the exonuclease are inhibited, profoundly for the polymerase activity. However, if the polymerase approaches closer to the deaminated bases, locating it at +3, +2, +1 or even 0 (paired with the extreme 3′ base in the primer), the exonuclease activity is strongly stimulated. In these situations, the exonuclease activity is actually stronger than that seen with mismatched primer-templates, even though the deaminated base-containing primer-templates are correctly base-paired. The resulting exonucleolytic degradation of the primer serves to move the uracil/hypoxanthine away from the primer–template junction, restoring the stalling position to +4. Thus the 3′–5′ proofreading exonuclease contributes to the inability of the polymerase to replicate beyond deaminated bases.     

Potential roles of 3′-5′exonuclease activity of NM23-H1 in DNA repair and malignant progression

NM23-H1 is a metastasis suppressor protein that exhibits 3′-5′ exonuclease activity in vitro. As 3′-5′ exonucleases are generally required for maintenance of genome integrity, this activity represents a plausible candidate mediator of the metastasis suppressor properties of the NM23-H1 molecule. Consistent with an antimutator function, ablation of the yeast NM23 homolog, YNK1, results in increased mutation rates following exposure to UV irradiation and exposure to the DNA damaging agents etoposide, cisplatin, and MMS. In human cells, a DNA repair function is further suggested by increased NM23-H1 expression and nuclear translocation following DNA damage. Also, forced expression of NM23-H1 in NM23-deficient and metastatic cell lines results in coordinate downregulation of multiple DNA repair genes, possibly reflecting genomic instability associated with the NM23-deficient state. To assess the relevance of the 3′-5′ exonuclease activity of NM23-H1 to its antimutator and metastasis suppressor functions, a panel of mutants harboring defects in the 3′-5′ exonuclease and other enzymatic activities of the molecule (NDPK, histidine kinase) have been expressed by stable transfection in the melanoma cell line, 1205Lu. Pilot in vivo metastasis assays indicate 1205Lu cells are highly responsive to the metastasis suppressor effects of NM23-H1, thus providing a valuable model for measuring the extent to which the nuclease function opposes metastasis and metastatic progression.     

Recognition and processing of double-stranded DNA by ExoX,a distributive 3′–5′ exonuclease

Members of the DnaQ superfamily are major 3′–5′ exonucleases that degrade either only single-stranded DNA (ssDNA) or both ssDNA and double-stranded DNA (dsDNA). However, the mechanism by which dsDNA is recognized and digested remains unclear. Exonuclease X (ExoX) is a distributive DnaQ exonuclease that cleaves both ssDNA and dsDNA substrates. Here, we report the crystal structures of Escherichia coli ExoX in complex with three different dsDNA substrates: 3′ overhanging dsDNA, blunt-ended dsDNA and 3′ recessed mismatch-containing dsDNA. In these structures, ExoX binds to dsDNA via both a conserved substrate strand-interacting site and a previously uncharacterized complementary strand-interacting motif. When ExoX complexes with blunt-ended dsDNA or 5′ overhanging dsDNA, a ‘wedge’ composed of Leu12 and Gln13 penetrates between the first two base pairs to break the 3′ terminal base pair and facilitates precise feeding of the 3′ terminus of the substrate strand into the ExoX cleavage active site. Site-directed mutagenesis showed that the complementary strand-binding site and the wedge of ExoX are dsDNA specific. Together with the results of structural comparisons, our data support a mechanism by which normal and mismatched dsDNA are recognized and digested by E. coli ExoX. The crystal structures also provide insight into the structural framework of the different substrate specificities of the DnaQ family members.     

Effect of guanosine 3′:5′-monophosphate on the hydroosmotic activity of adenosine 3′:5′-monophosphate in toad urinary bladder

Guanosine 3′:5′-monophosphate has a slight hydroosmotic effect on toad urinary bladder. Furthermore, this nucleotide strongly inhibits the responses to 3′:5′-adenosine monophosphate and oxytocin. The response to an increase in medium tonicity is not modified by the guanosine nucleotide. A role for guanosine 3′:5′-monophosphate in the regulation of water permeability in toad urinary bladder is proposed.     

The effect of the 3′ → 5′ exonuclease of T7 DNA polymerase on frameshifts and deletions

Both spontaneous frameshift mutation and deletion mutation were measured in a T7 phage deficient in the 3′ → 5′ exonuclease of T7 DNA polymerase. It was found that the absence of this exonuclease caused a marked increase in the revision of both plus one and minus one mutations. The exonuclease deficiency caused essentially no effect on the frequency of deletion between 10-bp direct repeats even when the segment between the direct repeats contained a 25-bp palindrome.     

A novel single-stranded DNA-specific 3′–5′ exonuclease,Thermus thermophilus exonuclease I,is involved in several DNA repair pathways

Single-stranded DNA (ssDNA)-specific exonucleases (ssExos) are expected to be involved in a variety of DNA repair pathways corresponding to their cleavage polarities; however, the relationship between the cleavage polarity and the respective DNA repair pathways is only partially understood. To understand the cellular function of ssExos in DNA repair better, genes encoding ssExos were disrupted in Thermus thermophilus HB8 that seems to have only a single set of 5′–3′ and 3′–5′ ssExos unlike other model organisms. Disruption of the tthb178 gene, which was expected to encode a 3′–5′ ssExo, resulted in significant increase in the sensitivity to H₂O₂ and frequency of the spontaneous mutation rate, but scarcely affected the sensitivity to ultraviolet (UV) irradiation. In contrast, disruption of the recJ gene, which encodes a 5′–3′ ssExo, showed little effect on the sensitivity to H₂O₂, but caused increased sensitivity to UV irradiation. In vitro characterization revealed that TTHB178 possessed 3′–5′ ssExo activity that degraded ssDNAs containing deaminated and methylated bases, but not those containing oxidized bases or abasic sites. Consequently, we concluded that TTHB178 is a novel 3′–5′ ssExo that functions in various DNA repair systems in cooperation with or independently of RecJ. We named TTHB178 as T. thermophilus exonuclease I.     

Control of SV40 DNA replication by protein phosphorylation: a model for cellular DNA replication?

SV40 DNA replication has been studied extensively as a model for eukaryotic DNA replication. The initiation of SV40 DNA replication depends on certain cellular enzymes and on a multifunctional viral phosphoprotein, T antigen, whose activity is controlled positively and negatively by its phosphorylation state. Several cellular protein kinases and phosphatases that act on T antigen have now been identified. The recent elucidation of the step in initiation that is sensitive to T antigen's phosphorylation state raises the question of whether initiation of cellular DNA replication may utilize a similar regulatory mechanism.     

Isolation of a novel protein,P12—from adult Drosophila melanogaster that inhibits deoxyribonucleoside and protein kinase activities and activates 3′-5′- exonuclease activity

ABSTRACT

We have previously found that Drosophila melanogaster only has one deoxyribonucleoside kinase, Dm-dNK, however, capable to phosphorylate all four natural deoxyribonucleosides. Dm-dNK was originally isolated from an embryonic cell line. We wanted to study the expression of Dm-dNK during development from embryonic cells to adult flies and found declining Dm-dNK activity during development and no activity in adult flies. Surprisingly, the extract from adult flies exhibited a strong inhibitory effect on deoxyribonucloside kinase activity. The dNK-inhibitor was precipitable with ammonium sulfate, and was purified to a high degree by gel-filtration as indicated by LC-MS/MS analysis. Since the inhibitor eluted from G-200 gel-filtration with a size of 10–13 kDa, we named it P12. We tested the purified fraction for specificity towards various enzymes and found that both mammalian and bacterial dNKs were inhibited, whereas there was no effect on hexokinase and pyruvate kinases and acidic phosphatase. However, when tested against cyclin B-dependent kinase, we found a strong inhibitory effect. Both with human Cdk1/CycB and S. pombe Cdc2/B-type cyclin the purified fraction from Superdex 200 that inhibited Dm-dNK, also inhibited the two protein kinases to the same degree. Furthermore, testing P12 in a DNA polymerase based assay we found that the 3′-5′-exonuclease part of the DNA polymerase (Klenow polymerase) was activated.     

Structure–activity relationships in human RNA 3′-phosphate cyclase

RNA 3′-phosphate cyclase (Rtc) enzymes are a widely distributed family that catalyze the synthesis of RNA 2′,3′ cyclic phosphate ends via an ATP-dependent pathway comprising three nucleotidyl transfer steps: reaction of Rtc with ATP to form a covalent Rtc-(histidinyl-N)-AMP intermediate and release PP_i; transfer of AMP from Rtc1 to an RNA 3′-phosphate to form an RNA(3′)pp(5′)A intermediate; and attack by the terminal nucleoside O2′ on the 3′-phosphate to form an RNA 2′,3′ cyclic phosphate product and release AMP. Here we used the crystal structure of Escherichia coli RtcA to guide a mutational analysis of the human RNA cyclase Rtc1. An alanine scan defined seven conserved residues as essential for the Rtc1 RNA cyclization and autoadenylylation reactions. Structure–activity relationships were clarified by conservative substitutions. Our results are consistent with a mechanism of adenylate transfer in which attack of the Rtc1 His320 nucleophile on the ATP α phosphorus is facilitated by proper orientation of the PP_i leaving group via contacts to Arg21, Arg40, and Arg43. We invoke roles for Tyr294 in binding the adenine base and Glu14 in binding the divalent cation cofactor. We find that Rtc1 forms a stable binary complex with a 3′-phosphate terminated RNA, but not with an otherwise identical 3′-OH terminated RNA. Mutation of His320 had little impact on RNA 3′-phosphate binding, signifying that covalent adenylylation of Rtc1 is not a prerequisite for end recognition.     

Evidence that YycJ is a novel 5′–3′ double-stranded DNA exonuclease acting in Bacillus anthracis mismatch repair

The most important system for correcting replication errors that survive the built in editing system of DNA polymerase is the mismatch repair (MMR) system. We have identified a novel mutator strain yycJ in Bacillus anthracis. Mutations in the yycJ gene result in a spontaneous mutator phenotype with a mutational frequency and specificity comparable to that of MMR-deficient strains such as those with mutations in mutL or mutS. YycJ was annotated as a metallo-β-lactamase (MβL) super family member with unknown activity. In this study we carried out a biochemical characterization of YycJ and demonstrated that a recombinant YycJ protein possesses a 5′–3′ exonuclease activity at the 5′ termini and at nicks of double-stranded DNA. This activity requires a divalent metal cofactor Mn²⁺ and is stimulated by 5′-phosphate ends of duplex DNA. The mutagenesis of conserved amino acid residues revealed that in addition to the five MβL family conserved motifs, YycJ appears to have its specific motifs that can be used to distinguish YycJ from other closely related MβL family members. A phylogenetic survey showed that putative YycJ homologs are present in several bacterial phyla as well as in members of the Methanomicrobiales and Thermoplasmales from Archaea. We propose that YycJ represents a new group of MβL fold exonucleases, which is likely to act in the recognition of MMR entry point and subsequent removal of the mismatched base in certain MutH-less bacterial species.     

The 3′ → 5′ exonucleases of both DNA polymerases δ and ε participate in correcting errors of DNA replication in Saccharomyces cerevisiae

DNA polymerases II (ε) and III(δ) are the only nuclear DNA polymerases known to possess an intrinsic 3′ → 5′ exonuclease in Saccharomyces cerevisiae. We have investigated the spontaneous mutator phenotypes of DNA polymerase δ and ε 3′ → 5′ exonuclease-deficient mutants, pol3-01 and pol2-4, respectively. pol3-01 and pol2-4 increased spontaneous mutation rates by factors of the order of 10² and 10¹, respectively, measured as URA3 forward mutation and his7-2 reversion. Surprisingly, a double mutant pol2-4 pol3-01 haploid was inviable. This was probably due to accumulation of unedited errors, since a pol2-4/pol2-4 pol3-01/pol3-01 diploid was viable, with the spontaneous his7-2 reversion rate increased by about 2 × 10³-fold. Analysis of mutation rates of double mutants indicated that the 3′ → 5′ exonucleases of DNA polymerases δ and ε can act competitively and that, like the 3′ → 5′ exonuclease of DNA polymerase δ the 3′ → 5′ exonuclease of DNA polymerase ε acts in series with the PMS1 mismatch correction system. Mutational spectra at a URA3 gene placed in both orientations near to a defined replication origin provided evidence that the 3′ → 5′ exonucleases of DNA polymerases δ and ε act on opposite DNA strands, but were in sufficient to distinguish conclusively between different models of DNA replication.     

3′–5′ exonuclease activity of human apurinic/apyrimidinic endonuclease 1 towards DNAs containing dNMP and their modified analogs at the 3′ end of single strand DNA break

Human DNA apurinic/apyrimidinic (AP-) endonuclease 1 (APE1) is involved in the base excision repair (BER) pathway. The enzyme hydrolyzes DNA from the 5 side of the AP site. In addition to endonuclease activity, APE1 also possesses other slight activities including 3 -5 exonuclease activity. The latter is preferentially exhibited towards mispaired (non-canonical) nucleotides, this being the reason why APE1 is considered as a proofreading enzyme correcting the misincorporations introduced by DNA polymerase beta. We have studied 3 -5 exonuclease activity of APE1 towards dCMP and dTMP residues and modified dCMP analogs with photoreactive groups at the 3 end of the nicked DNA. Photoreactive dNMP residues were incorporated at the 3 end of the lesion using DNA polymerase beta and photoreactive dNTPs. The dependence of exonuclease activity on the "canonicity" of the base pair formed by dNMP flanking the nick at the 3 end, on the nature of the group flanking the nick at the 5 end, and on the reaction conditions has been determined. Optimal reaction conditions for the 3 -5 exonuclease hydrolysis reaction catalyzed by APE1 in vitro have been established, and conditions when photoreactive residues are not removed by APE1 have been chosen. These reaction conditions are suitable for using photoreactive nicked DNAs bearing 3 -photoreactive dNMP residues for photoaffinity labeling of proteins in cellular/nuclear extracts and model APE1-containing systems. We recommend using FAPdCTP for photoaffinity modification in APE1-containing systems because the FAPdCMP residue is less prone to exonuclease degradation, in contrast to FABOdCTP, which is not recommended.     

The mode of action of adenosine 3′:5′-cyclic monophosphate in mammalian islets of Langerhans. Effects of insulin secretagogues on islet-cell protein kinase activity

1. Protein kinase activity was measured in islets of Langerhans that had been incubated in the presence of agents known to affect insulin release. 2. Glucagon, theophylline, caffeine and 3-isobutyl-1-methylxanthine, agents that raise cyclic AMP concentrations in islet cells and stimulate insulin release, increased protein kinase activity. Adrenaline and diazoxide, agents that decrease cyclic AMP concentrations and inhibit insulin secretion, decreased the activity. 3. The increase in protein kinase activity produced by different concentrations of 3-isobutyl-1-methylxanthine was apparently related to the increase in intracellular concentrations of cyclic AMP. 4. The sulphonylureas, tolbutamide and glibenclamide, agents that increase insulin release, also increased the protein kinase activity; however, leucine, arginine and xylitol, which also stimulate insulin release, were without effect on the kinase activity. 5. Increasing the glucose concentration of the incubation medium from 2 to 20mm had no effect on protein kinase activity. Further, the ability of 3-isobutyl-1-methylxanthine to increase the protein kinase activity was not affected by the glucose concentration of the incubation medium. 6. These results suggest that agents which affect insulin secretion by altering cyclic AMP concentrations may exert their effects on hormone release by altering the activity of a cyclic AMP-dependent protein kinase in islet cells.     

Purification and characterization of the 5′ → 3′ exonuclease domain-deleted Thermus filiformis DNA polymerase expressed in Escherichia coli

The gene encoding 5 3 exonuclease domain-deleted Tfi DNA polymerase, named 5 3 Exo^– Tfi fragment, from Thermus filiformis was expressed in Escherichia coli under the control of the tac promoter on a high-copy plasmid, pJR. The expressed enzyme was purified 27-fold with a 19% yield and a specific activity of 2621 U mg^–1 protein. The 5 3 exonuclease domain of Tfi DNA polymerase was removed without significant effect on enzyme activity and stability. PCR conditions for the 5 3 Exo^– Tfi fragment were more tolerant to the buffer composition as compared to the full-length enzyme.     

Studies on the changes in protein fluorescence and enzymic activity of aspartate aminotransferase on binding of pyridoxal 5′-phosphate

1. The alpha and beta subforms of aspartate aminotransferase were purified from pig heart. 2. The alpha subform contained 2mol of pyridoxal 5'-phosphate. The apo-(alpha subform) could be fully reactived by combination with 2mol of cofactor. 3. The protein fluorescence of the apo-(alpha subform) decreased non-linearly with increase in enzyme activity and concentration of bound cofactor. 4. It is concluded that the enzyme activity/mol of bound cofactor is largely independent of the number of cofactors bound to the dimer. 5. The beta subform had approximately half the specific enzyme activity of the alpha subform, and contained an average of one active pyridoxal 5'-phosphate molecule per molecule, which could be removed by glutamate, and another inactive cofactor which could only be removed with NaOH. 6. On recombination with pyridoxal 5'-phosphate the protein fluorescence of the apo-(beta subform) decreased linearly, showing that each dimeric enzyme molecule contained one active and one inactive bound cofactor. 7. The results are not consistent with a flip-flop mechanism for this enzyme.     

Protein kinase activity in membrane preparations from ox brain. Stimulation of intrinsic activity by adenosine 3′:5′-cyclic monophosphate

When Ac(2)-l-Lys-d-Ala-d-Ala and either meso-diaminopimelic acid or Gly-l-Ala are exposed to the exocellular dd-carboxypeptidase-transpeptidase of Streptomyces R61, transpeptidation reactions yielding Ac(2)-l-Lys-d-Ala-(d)-meso- diaminopimelic acid and Ac(2)-l-Lys-d-Ala-Gly-l-Ala occur concomitantly with the hydrolysis of the tripeptide into Ac(2)-l-Lys-d-Ala. The proportion of the enzyme activity which can be channelled in the transpeptidation and the hydrolysis pathways depends upon the pH and the polarity of the environment. Transpeptidation is favoured both by increasing the pH and by decreasing the water content of the reaction mixtures. Kinetics suggest that the reactions proceed through an ordered mechanism in which the acceptor molecule (meso-diaminopimelic acid or Gly-l-Ala) binds first to the enzyme. Both acceptors behave as non-competitive inhibitors of the hydrolysis pathway. Transpeptidation is inhibited by high concentrations of Gly-l-Ala but not by high concentrations of meso-diaminopimelic acid. The occurrence on the enzyme of an additional inhibitory binding site for Gly-l-Ala is suggested.     

Complex of repair DNA polymerase β with autonomous 3′→5′ exonuclease shows increased accuracy of DNA synthesis

The complexes of repair DNA polymerase β with 3′-exonuclease and some other proteins were isolated from the chromatin of hepatocytes of normal rats for the first time. Biopolymers were extracted from the chromatin by the solution of NaCl and Triton X-100. The extract was fractionated by gel-filtration on Sephacryl S-300 columns successively in low and high ionic strength solutions, on hydroxyapatite, and on Sephadex G-100 columns. The complexes have molecular weights of 100 and 300 kDa. They dissociate to DNA polymerase and exonuclease in the course of chromatography on a DNA-cellulose column or after gel-filtration in the presence of 1 M NaCl. The co-purification of the polymerase and exonuclease is reconstituted in 0.1 M NaCl. The fidelity of monomeric and composite DNA polymerase β was measured using phase ?X174 amber 3 as a primer/template. The products of the synthesis were transfected into Escherichia coli spheroplasts, and the frequency of reverse mutations was determined. The complex of DNA polymerase β with 3′-exonuclease was shown to be 30 times more accurate than the monomeric polymerase, which can decrease the probability of repair mutagenesis and carcinogenesis.     

Adenosine 3′:5′-cyclic monophosphate-dependent protein kinase(s) of rat ovarian cells. Gonadotropin regulation of adenosine 3′:5′-cyclic monophosphate-receptor activity

Regulation of cyclic AMP-dependent protein kinase, cyclic AMP-receptor activity and intracellular cyclic AMP concentrations by choriogonadotropin was studied in ovarian cells prepared from 26-day-old rats. A close correlation was observed between phospho-transferase activity and cyclic AMP-receptor activity in 12000g supernatant fractions from rat ovarian homogenate. The apparent activation constant (K(a)) and I(50) (concentration required to produce 50% inhibition) of different cyclic nucleotides for phosphotransferase and cyclic AMP receptor activities respectively were also determined. Cyclic AMP and 8-bromo cyclic AMP were most effective, giving K(a) values of 0.08 and 0.09mum and I(50) of 0.12 and 0.16mum respectively. Other nucleotides were also effective, but required higher concentrations to give a comparable effect. An increased concentration of cyclic AMP produced by choriogonadotropin (1mug/ml) treatment was accompanied by decreased cyclic AMP binding as early as 5min after hormone addition. Choriogonadotropin also stimulated the protein kinase activity ratio (-cyclic AMP/+cyclic AMP) under identical experimental conditions. The phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine potentiated the action of choriogonadotropin on the three parameters measured in a dose- and time-dependent manner. The maximal cyclic AMP-binding capacity, as determined by cyclic AMP-exchange assay, remained unchanged before and after hormone addition. The endogenously bound cyclic AMP was determined from the difference between the maximal binding capacity and the exogenously bound cyclic AMP. With different choriogonadotropin concentrations, a quantitative correlation was established between maximal binding capacity, exogenous binding and endogenous binding activities. Approx. 60% of total binding sites were endogenously occupied in untreated cells, and choriogonadotropin (1mug/ml) treatment fully saturated available binding sites with a parallel 10-fold increase in cellular cyclic AMP. The present results provide evidence for a probable intracellular compartmentalization of cyclic AMP in the ovarian cell, and suggest that in the unstimulated state all cyclic AMP present in the ovarian cell may not be available for protein kinase activation.