| Nucleic Acids Research | Pages |
Pyrophosphorolytic dismutation of oligodeoxy-nucleotides by terminal deoxynucleotidyltransferase
Introduction
Materials And Methods
Materials
Oligonucleotide and dNTP stock solutions
Bovine terminal transferase
Western blot analysis
TdT activity assay
Heat inactivation of TdT
Determination of protein concentration
Radiolabeling and purification of oligonucleotides
Stock metal ion solutions
Pyrophosphorolysis reaction
Urea-PAGE
Approximation of the rates of polymerization and pyrophosphorolysis
Determination of PPi concentrations
Results And Discussion
Demonstration of pyrophosphorolysis
Demonstration of activity as that of TdT
Optimal metal cation concentration
Effects of various concentrations of PPi
Pyrophosphorolysis with p(dC)10 and p(dG)10
Pyrophosphorolysis using ddATP-terminated oligonucleotides
Approximation of the rate of pyrophosphorolysis
Acknowledgement
References
Pyrophosphorolytic dismutation of oligodeoxy-nucleotides by terminal deoxynucleotidyltransferase
Received March 5, 1999; Revised and Accepted June 9, 1999
ABSTRACT Terminal transferase (TdT), when incubated with a purified 32P-5[prime]-end-labeled oligonucleotide of defined length in the presence of Co2+, Mn2+ or Mg2+ and 2-mercaptoethanol in cacodylate or HEPES buffer, pH 7.2, exhibits the ability to remove a 3[prime]-nucleotide from one oligonucleotide and add it to the 3[prime]-end of another. When analyzed by urea-PAGE, this activity is observed as a disproportionation of the starting oligonucleotide into a ladder of shorter and longer oligonucleotides distributed around the starting material. Optimal metal ion concentration is 1-2 mM. All three metal ions support this activity with Co2+ > Mn2+ [cong] Mg2+. Oligonucleotides p(dT) and p(dA) are more efficient substrates than p(dG) and p(dC) because the latter may form secondary structures. The dismutase activity is significant even in the presence of dNTP concentrations comparable to those that exist in the nucleus during the G1 phase of the cell cycle. Using BetaScope image analysis the rate of pyrophosphorolytic dismutase activity was found to be only moderately slower than the polymerization activity. These results may help explain the GC-richness of immunoglobulin gene segment joins (N regions) and the loss of bases that occur during gene rearrangements in pre-B and pre-T cells.
INTRODUCTION
The uniqueness of terminal deoxynucleotidyltransferase (terminal transferase, TdT) as a DNA polymerase has generated much research into the elucidation of its structure and function. Recombinant DNA technology has provided the complete amino acid sequence of the human enzyme which demonstrates great conservation of sequence in bovine and murine species. Evidence generated with regard to the physiological function of TdT suggests a role in creating diversity during immunoglobin (Ig) and T cell receptor gene rearrangements (1-3).
Three catalytic activities characterize most prokaryotic and eukaryotic DNA polymerases: polymerization of dNTPs, pyrophosphate (PPi) exchange and pyrophosphorolysis (4). The equations defining each of these reactions are, respectively:
 |
1 |
 |
2 |
 |
3 |
Polymerization (reaction 1) can be analyzed by measuring the amount of radioactive dNTP incorporated into an acid-insoluble product (DNA). The standard assay examining reaction 2, PPi exchange, quantitates the incorporation of radioactive PPi into a dNTP upon incubation of enzyme, unlabeled DNA and dNTP with radioactive PPi. Pyrophosphorolysis (reaction 3) is simply the reverse of reaction 1 and is usually analyzed by measuring the amount of radioactive PPi incorporated into dNTP upon incubation of the enzyme with radioactive PPi and unlabeled DNA (5,6).
The literature regarding the ability of TdT to catalyze PPi exchange and pyrophosphorolysis appears to be somewhat controversial. Kato et al. (7) and Chang (8) were unable to demonstrate either activity using the standard assay described above. These investigators analyzed product formation by DEAE paper chromatography followed by autoradiography. Srivastava and Modak (6) analyzed product formation by chromatography on PEI-cellulose F254 TLC plates followed by autoradiography and were able to provide evidence that TdT is capable of catalyzing both PPi exchange and pyrophosphorolysis. The quantity of DNA necessary to detect these activities, however, was 10-20 times the amount required to detect polymerization.
Expanding the pyrophosphorolysis reaction:
 |
4 |
 |
5 |
 |
6 |
The unique ability of TdT to polymerize dNTPs in a template-independent manner (7) allows TdT to readily recycle the released dNTP of reaction 4 back into DNA (reaction 5) without regard for complementarity. Perhaps it is this unique ability that has hampered efforts to unequivocally demonstrate pyrophosphorolysis by TdT using the standard assay system, which requires detection of incorporation of radioactive pyrophosphate into the dNTP product. The distributive nature of polymerization exhibited by TdT (7) allows the released dNTP to be incorporated onto the 3[prime]-end of any DNA strand present, not necessarily the strand from which it was removed. The net reaction thus reveals dismutation of a single oligonucleotide species into two products.
Although most DNA polymerases (and RNA polymerases) are capable of catalyzing pyrophosphorolysis and PPi exchange, polymerase [beta] remains a notable exception for which neither PPi exchange nor pyrophosphorolysis has been demonstrated (4). The physiological significance of either reaction, however, is uncertain (9). Several investigators have proposed that PPi exchange may play a functional role in decreasing the number of mismatches occurring during DNA synthesis (10,11). This paper presents a highly sensitive method to demonstrate the pyrophosphorolytic dismutation activity of TdT.
MATERIALS AND METHODS
Materials
All oligonucleotides, 2[prime]-deoxynucleotides and dideoxyATP (ddATP) were purchased from Pharmacia. Calf thymus TdT was purchased from US Biochemical. Yeast inorganic pyrophosphatase was purchased from Sigma. T4 polynucleotide kinase was purchased from New England Biolabs. [[gamma]-32P]ATP and [[alpha]-32P]dGTP were purchased from ICN and [[alpha]-32P]dideoxyATP was purchased from Amersham. Rabbit anti-TdT polyclonal antibody was purchased from Bethesda Research Laboratories. DE-81 filter discs were bought from Whatman Inc. Sterile 12 × 75 mm disposable polystyrene culture tubes were obtained from Fisher Scientific. Pyrophosphate reagent for PPi determinations was purchased from Sigma. All other chemicals were purchased from Sigma or Aldrich.
Oligonucleotide and dNTP stock solutions
Stock solutions of oligonucleotides p(dA)40-60, p(dA)12-18, p(dT)25-30, p(dT)15, p(dA)16, p(dG)10, p(dC)10 and p(dT)16 were prepared by resuspending the lyophilyzed oligonucleotides in 100-300 µl of 10 mM HEPES, pH 7.8. Final concentrations were calculated using an average dNMP molecular weight of 324.3 Da and the A260/mg data provided on the certificate of analysis. An average chain length of 50 nt was used for p(dA)40-60, 15 nt for p(dA)12-18 and 27.5 nt for p(dT)25-30.
A 10-20 mM stock solution of dGTP (specific radioactivity 100-200 c.p.m./pmol) was prepared by adding 500 µCi of [[alpha]-32P]-dGTP (specific radioactivity 3000 Ci/mmol) to 100-200 µl of a 100 mM dGTP solution and brought to a final volume of 1 ml with 10 mM HEPES, pH 7.8.
Bovine terminal transferase
The TdT preparation consisted of the 44 kDa [alpha],[beta] heterodimer (95-98%) and the 45 kDa monomer (2-5%) as judged by Coomassie blue and silver staining of SDS-polyacrylamide gels and western blot analysis.
Western blot analysis
Western blot analysis of SDS gels was carried out according to a method described by Walker and Gaastra (12) using rabbit anti-TdT polyclonal antibodies as the primary antibody and horseradish peroxidase-conjugated goat anti-rabbit IgG as the secondary antibody.
TdT activity assay
Ten microliters of reaction mixture containing TdT was added to 20 µl of assay cocktail. The final concentrations of assay components in 30 µl were: 50 mM HEPES, pH 7.8, 1 mM 2-mercaptoethanol, 1 mM [[alpha]-32P]dGTP (specific radioactivity 50-200 c.p.m./pmol), 10 µM p(dA)12-18 or p(dT)25-30 and 7 mM MgCI2. Incubation was for 2 min at 30°C. The reaction was terminated by application of the entire mixture to a DE-81 paper disc which was washed three times in 5% dibasic sodium phosphate according to Bollum (13). The discs were then washed two or three times in 95% ethanol, dried and counted in a Beckman LS6800 liquid scintillation counter. Formula 989 (Dupont/New England Nuclear) was used as the scintillation fluid. Assays were carried out in sterile 12 × 75 mm disposable polystyrene culture tubes.
Heat inactivation of TdT
An aliquot of TdT was divided into two fractions, one of which was inactivated by incubation at 42°C until residual activity was ~50% that of the unheated control fraction. The heat-treated and control enzyme fractions were then incubated, separately, with purified 5[prime]-end-labeled p(dT)15. At allotted times, 1 µl aliquots were removed for PAGE analysis. Gels were dried and analyzed by BetaScope image analysis (Betagen Corp.).
Determination of protein concentration
All protein determinations were carried out using a modified micro-Lowry procedure described by Peterson (14). Egg white ovalbumin was used as the standard. The precision of protein concentrations determined by this method was checked by spectrophotometric analysis at A275 employing a molar extinction coefficient of 35 000 M-l cm-1 determined for the 44 kDa TdT species (15). Values obtained from the two procedures were within 10% of each other.
Radiolabeling and purification of oligonucleotides
Oligonucleotides labeled on the 5[prime]-end were prepared by using either the polynucleotide kinase exchange reaction or the forward reaction described by Sambrook et al. (16). Labeling of oligonucleotides on the 3[prime]-end was done with [[alpha]-32P]ddATP and TdT. Radiolabeled oligonucleotides were separated from unincorporated nucleotides by urea-PAGE. Radioactive bands were located by autoradiography, excised and extracted into sterile 10 mM Tris-HCI, 1 mM EDTA, pH 8, by gentle agitation overnight at 37°C. They were then desalted on a sterile Sephadex G-25 column eluted with sterile water. Fractions of 1 ml were collected in sterile Eppendorf microcentrifuge tubes. Radioactive peaks were monitored using a Geiger counter. Only fractions at the front of the radioactive peak were pooled and kept for further experiments. Pooled fractions were lyophilized and resuspended in sterile water to a final concentration of 50-150 µM based on absorbance measurements according to Sambrook et al. (16).
Stock metal ion solutions
All metal ion solutions were filtered through a 0.2 µM filter (Nalgene) and autoclaved.
Pyrophosphorolysis reaction
The reaction mixture contained, in a final volume of 25 µl, 100 µM cacodylate buffer, pH 7.2, 1 mM 2-mercaptoethanol, 0.5 µM TdT, 5-10 µM radiolabeled oligonucleotide of defined length, 2 mM divalent metal cation and various concentrations of PPi. The reaction was carried out at 25 or 30°C and quenched by heating the reaction mixture at 110°C for 10 min. One microliter aliquots were removed at allotted times for urea-PAGE analysis.
Urea-PAGE
All urea-PAGE was done according to Sambrook et al. (16). Briefly, the gels were 12% T, 0.4% C polyacrylamide and 7 M urea electrophoresed at a constant 1000 V.
Approximation of the rates of polymerization and pyrophosphorolysis
Four TdT reaction mixtures were prepared containing 100 mM cacodylate buffer, pH 7.2, 1 mM 2-mercaptoethanol, 0.5 µM TdT, 2 mM Co2+ and 10 µM 32P-5[prime]-end-labeled oligonucleotide p(dT)16 (1325 c.p.m./pmol) in a volume of 25 µl. To the first reaction mixture PPi was added to a final concentration of 100 µM and to the second TTP was added to a final concentration of 1 mM. The third reaction mixture was the control having no added PPi or TTP. To ascertain the rate of polymerization in the presence of 100 µM PPi a fourth reaction mixture containing both TTP and PPi was set up. Reaction mixtures were incubated at 30°C and at an allotted time a 1 µl aliquot was withdrawn, placed in loading buffer and heated at 100°C for 5 min. The entire reaction mixture containing TTP was quenched after 45 s. Reaction products were separated by urea-PAGE. Quantitation was by BetaScope image analysis (Betagen Corp.). The rates of polymerization and pyrophosphorolysis were then estimated. Based on the radioactivity present in each band (determined by BetaScope analysis) above or below that which represented the starting material and the known radiospecific activity of the starting oligonucleotide, the quantity of oligonucleotide in each band was determined. The number of TMPs incorporated into each band was then determined by multiplying the quantity of oligonucleotide in each band by the number of nucleotide additions for each band. For example, the first band was multiplied by 1, the second band by 2, the third band by 3, etc. Summing the results for those bands above the starting material permitted calculation of an approximate rate of polymerization expressed as TMP polymerized/min. Similarly, summing the results for those bands below the starting material permitted calculation of an approximate rate of pyrophosphorolysis expressed as oligonucleotide pyrophosphorolyzed/min.
Determination of PPi concentrations
The concentrations of PPi in the potassium phosphate (KPi) TdT storage buffer was measured by the method of O'Brien (17) using the pyrophosphate reagent from Sigma.
RESULTS AND DISCUSSION
Demonstration of pyrophosphorolysis
Reported here is a highly sensitive method for observing and analyzing the pyrophosphorolytic activity of TdT and perhaps other DNA polymerases. Analyzing pyrophosphorolysis by urea-PAGE permits detection and analysis of pyrophosphorolysis using nanomolar concentrations of DNA, providing an increase in sensitivity of one to two orders of magnitude over previously used methods (5,6). This method was used to resolve an apparent controversy in the literature concerning the ability of TdT to catalyze pyrophosphorolysis. Results reported here corroborate those reported by Srivastava and Modak (6); TdT is capable of catalyzing pyrophosphorolysis.
Considering the distributive nature of polymerization exhibited by TdT, the expanded equation for the pyrophosphorolysis reaction given above (equations 4-6) would predict that TdT incubated with PPi and DNA could remove a nucleotide as a dNTP from the 3[prime]-end of a DNA strand and add this nucleotide to another DNA strand. With increasing time of incubation this activity may be observed as a disproportionation of the starting material into bands representing DNA strands having gained one or more nucleotides and DNA strands having lost one or more nucleotides, distributed around the band representing the starting material. If an oligonucleotide of defined length is used as the starting material, then this activity may be visualized by urea-PAGE separation of the reaction products. The result of such an experiment using oligonucleotide p(dT)15 is shown in Figure 1. The absence of bands above or below the starting material in the 0 time lane demonstrates that the bands observed in succeeding lanes are not the result of: (i) oligonucleotide secondary structure; (ii) the presence of contaminating dNTPs; or (iii) contaminating nucleases and polymerases. As can be seen in the indicated lanes, the addition of exogenous PPi leads to a significant rate increase. Lanes without exogenous PPi contained 3.5 µM PPi as a contaminant of the KPi enzyme storage buffer.
Figure 1. Demonstration of pyrophosphorolysis. TdT was incubated with p(dT)15 at 25°C in a 25 µl volume containing: 100 mM cacodylate buffer, pH 7.2, 1 mM 2-mercaptoethanol, 5 µM 5[prime]-end-labeled p(dT)15, 0.5 µM TdT (assuming Mr = 44 kDa), 2 mM Co2+ and PPi. On the left (0-24 h) the PPi concentration was 3.5 µM (endogenous in the KPi enzyme storage buffer) and on the right (0.25-24 h) exogenous PPi was added to a final concentration of 100 µM. Each lane represents a 1 µl aliquot removed at the indicated times and electrophoresed at 1000 V on a 12% polyacrylamide gel containing 7 M urea.
Incubation of 32P-5[prime]-end-labeled p(dA)16 with TdT in the presence of 2-mercaptoethanol and a divalent metal cation in cacodylate buffer, pH 7.2, is shown in Figure 2. Reactions carried out with 2 mM Mn2+ are shown in lanes B-D, lanes E-G were in the presence of 2 mM Mg2+ and lanes H-J were with 2 mM Co2+. Lanes A and K are control lanes lacking either metal cation or TdT, respectively. Lanes C, F and I contain PPi added to a final concentration of 100 µM. Lanes B, E and H contain no exogenously added PPi. The pyrophosphorolytic activity observed in these latter reactions is due to 3.5 µM endogenous PPi contamination carried over from the KPi enzyme storage buffer. Incubation of the TdT preparation with yeast inorganic pyrophosphatase (0.16 µg) for 1 h at 25°C prior to the addition of DNA abolished pyrophosphorolysis (lanes D, G and J). Yeast inorganic pyrophosphatase has no affect on the polymerization reaction.
Figure 2. Metal ion support of pyrophosphorolysis. TdT (0.5 µM) incubated with p(dA)16 at 25°C for 23 h in a reaction volume of 25 µl containing: 100 mM cacodylate buffer, pH 7.2, 2 mM metal cation, 1 mM 2-mercaptoethanol and 10 µM 5[prime]-end-labeled p(dA)16. Lanes A and K are control lanes lacking metal cation or TdT, respectively. Lanes B-D were in the presence of Mn2+; lanes E-G with Mg2+; lanes H-J were in the presence of Co2+. Lanes B, E and H contained 3.5 µM endogenous PPi while lanes C, F and I had PPi added to a final concentration of 100 µM. In lanes D, G and J the TdT preparation was treated for 1 h at 25°C with 0.16 µg yeast inorganic pyrophosphatase prior to incubation with the oligonucleotide.
Figure 2 reveals that the three metals commonly used to assay TdT activity support pyrophosphorolysis with Co2+ > Mn2+ [cong] Mg2+. The absence of bands other than that of the starting material in the two control lanes (A and K) indicate that bands above and below the starting material in all other lanes are not due to: (i) oligonucleotide secondary structure; (ii) the presence of contaminating nuclease and polymerase activities; or (iii) the presence of contaminating free dNTPs. The symmetry of the bands around the starting material allows discrimination between true pyrophosphorolysis and endo- or exonucleolytic hydrolysis. Inhibition of the dismutation activity in the presence of pyrophosphatase and stimulation by the addition of PPi (Figs 1 and 2) unambiguously identify the activity as pyrophosphorolysis. These results also indicate that Mn2+ can support pyrophosphorolysis, in contrast to the observations of Srivastava and Modak (6), who reported that Mn2+ neither supports nor inhibits pyrophosphorolysis. Perhaps at the 1 mM PPi concentration used by Srivastava and Modak (6), Mn2+ is unable to support this activity. Pyrophosphate concentrations used in this report were one to two orders of magnitude lower. Using urea-PAGE to analyze pyrophosphorolysis by TdT also permits analysis of this reaction using submicromolar concentrations of DNA substrate.
The demonstration of significant pyrophosphorolysis by TdT in the presence of low micromolar concentrations of PPi has practical implications. Terminal transferase is often used as a molecular biology reagent for the addition of polynucleotide `tails' to vectors and/or genes to be cloned and expressed. Unless measures are taken to maintain very low (submicromolar) concentrations of PPi before and during polymerization, loss or scrambling of important sequence information may result, particularly if reactions are carried out in phosphate buffers which may have PPi as a contaminant.
Demonstration of activity as that of TdT
TdT is very sensitive to thermal inactivation at temperatures >40°C (13). To ascertain that the activity observed is due to TdT, an aliquot of TdT was incubated at 42°C until the residual was ~50% of the activity before heating. A control aliquot of TdT was maintained at 0°C over the same period of time. Equal quantities of the control and heat-treated enzyme were then incubated with 5[prime]-end-labeled p(dT)15. Reaction products were then separated by urea-PAGE. Quantitation was done by BetaScope image analysis (Betagen Corp.). In the heat-treated sample both the polymerization activity and pyrophosphorolysis activity decreased by ~50% relative to the control, suggesting that both activities are due to the same enzyme, TdT (data not shown). Silver stained and Coomassie stained SDS gels revealed only three polypeptides in the TdT preparation; the 45 kDa monomer and the [alpha] and [beta] polypeptides of the 44 kDa [alpha],[beta] heterodimer. Western blot analysis confirmed the three polypeptides as belonging to TdT. Tripolyphosphate, a known inhibitor of the polymerization reaction (7), at a final concentration of 1 mM was found to inhibit the pyrophosphorolysis reaction (data not shown).
Optimal metal cation concentration
The optimal metal ion concentration is between 1 and 2 mM. Pyrophosphorolysis activity significantly decreases at metal cation concentrations >2 and <0.5 mM (Fig. 3). Zinc is reported to increase the affinity of TdT for DNA and decreases that for the free nucleotide (18). Zinc alone is unable to support pyrophosphorolysis at either 1 or 2 mM (Fig. 3B). Incubation of p(dT)15 with TdT in the presence of 10 µM Zn2+ and 2 mM Mg2+, Mn2+ or Co2+ and 3.5 µM endogenous PPi had no significant affect compared to reaction mixtures lacking Zn2+ (data not shown).
Figure 3. Determination of optimal Mn2+ and Co2+ concentrations. (A) TdT was incubated with p(dT)15 at 25°C in a volume of 25 µl containing: 100 mM cacodylate buffer, pH 7.2, 1 mM 2-mercaptoethanol, 0.5 µM TdT, 3.5 µM PPi, 10 µM 5[prime]-end-labeled p(dT)15 and varying concentrations of Mn2+ as indicated. Time of incubation was 10 h. Aliquots of 1 µl were removed for analysis on a 12% polyacrylamide gel containing 7 M urea. (B) Reactions depicted were carried out as described for (A) except with Co2+ or Zn2+.
Effects of various concentrations of PPi
The rate of pyrophosphorolysis differs little at PPi concentrations between 3.5 and 10 µM, however, the optimal PPi concentration appears to be 100 µM. The rate of pyrophosphorolysis exhibits a sharp decrease at 1 mM PPi. An explanation for this marked decrease is not apparent.
Pyrophosphorolysis with p(dC)10 and p(dG)10
Significant pyrophosphorolysis is observed with homopolymers of thymidine and deoxyadenosine (Figs 1 and 2), but not with homopolymers of deoxycytidine or deoxyguanosine (data not shown). Pyrophosphorolysis of p(dC)10 is supported by both Co2+ and Mn2+, however, only Mn2+ also supported the subsequent polymerization reaction; Mg2+ did not support pyrophosphorolysis of this substrate (data not shown). This is curious in the light of the fact that Co2+ is the superior metal cation for pyrimidine polymerization (7).
Significant pyrophosphorolysis using p(dG)10 as a substrate was not observed with any metal cation (data not shown). Perhaps the virtual lack of pyrophosphorolysis on p(dG)10 is due to secondary structures known to form with homopolymers of this nucleotide. Since the same enzyme preparation and enzyme concentration were used in all experiments, the differences in the efficiency with which different substrates are used is not due to differences in endogenous PPi concentration or differences in enzyme or metal concentrations.
When TdT (0.4 µM) is incubated with 5 µM 5[prime]-end-labeled p(dT)15, 33 µM TTP and 100 µM PPi, pyrophosphorolysis is clearly observed (Fig. 4A). Incubation of TdT (0.4 µM) and 100 µM PPi with 0.16 µg yeast inorganic pyrophosphatase prior to the addition of TTP and 5[prime]-end-labeled p(dT)15 abolishes the pyrophosphorolytic activity, but does not inhibit the polymerization activity (Fig. 4B). The disappearance of the lower bands in Figure 4B with addition of pyrophosphatase indicates that the extra bands in Figure 4A are due to pyrophosphorolysis. Most significant, however, is the clear observation of pyrophosphorolysis in the presence of both polymerization substrates and where TTP is within the measured range of nuclear concentrations (19). This observation may provide a partial explanation for the loss of nucleotides at DNA joins generated by recombination events that take place in pre-T and pre-B cells and the GC-richness of immunoglobulin gene segment joins (N regions) (1,2,20-25). Experiments examining pyrophosphorolysis in the presence of ribonuleotides were not performed.
Figure 4. Pyrophosphorolysis in the presence of polymerization substrates. (A) TdT (0.4 µM) was incubated with 5 µM 5[prime]-end labeled p(dT)15, 33 µM TTP and 100 µM PPi. At the indicated times aliquots were removed and subjected to urea-PAGE on a 12% acrylamide gel. (B) TdT (0.4 µM) was incubated with 0.16 µg yeast inorganic pyrophosphatase and 100 µM PPi for 1 h prior to the addition of 5[prime]-end-labeled p(dT)15 and TTP. At the indicated times aliquots were removed for PAGE analysis.
Experiments carried out with recombinant human TdT show that the full-length (58 kDa) human enzyme is also capable of pyrophosphorolysis (data not shown). This indicates that the activity observed for the 44 kDa [alpha],[beta] heterodimer TdT is not an activity unique to a proteolyzed form of TdT.
Pyrophosphorolysis using ddATP-terminated oligonucleotides
Srivastava and Modak (6) state that TdT could carry out pyrophosphorolysis on oligonucleotides terminating with a ddNTP. To analyze this by urea-PAGE it is necessary to incubate an unlabeled, non-complementary oligonucleotide with oligonucleotide labeled at the 3[prime]-end with [[alpha]-32P]ddNMP. Since a free ddNTP is generated in the first step of the reaction, TdT should be able to incorporate the nucleotide into the unlabeled oligonucleotide, the second step, as shown in Figure 5A. The net result is transfer of label from one oligonucleotide to another effected by TdT. This experiment also demonstrates that added PPi enhances the rate of transfer (Fig. 5B). At this time it is not clear whether the cleaved ddNTP remains bound to TdT or is released free into solution.
Figure 5. Pyrophosphorolysis using a ddAMP-terminated oligonucleotide. TdT (0.5 µM) was incubated at 25°C in a reaction volume of 25 µl containing: 100 mM cacodylate buffer, pH 7.2, 1 mM 2-mercaptoethanol, 2 mM Co2+, 5 µM p(dA)16 labeled at the 3[prime]-end with [[alpha]-32P]ddATP, 5 µM unlabeled p(dA)40-60 and 3.5 µM endogenous PPi or 100 µM added PPi as indicated. (A) A diagram illustrating the reaction in (B). The oligonucleotide labeled A represents p(dA)16-ddAMP with a radioactive phosphate (*P). A[prime] is p(dA)16 after pyrophosphorolysis, B is unlabeled p(dA)40-60 and B[prime] is p(dA)40-60 after addition of [[alpha]-32P]ddATP released from p(dA)16. (B) At various times 1 µl aliquots were removed for analysis on a 12% polyacrylamide gel containing 7 M urea. Lanes L and M are control lanes lacking metal cation or TdT, respectively. Lane A is time 0; lanes B-E are 10, 20, 30 and 60 min, respectively; lanes F-K represent, 2, 4, 6, 8, 10 and 23 h, respectively. The numerals 1 and 2 represent 3[prime]-end-labeled p(dA)16 and p(dA)40-60, respectively.
Approximation of the rate of pyrophosphorolysis
To postulate a possible physiological role for pyrophosphorolysis it is useful to compare the rate of pyrophosphorolysis with that of polymerization. An approximation of the rates of polymerization and pyrophosphorolysis was achieved by BetaScope image analysis (Betagen Corp.) as described in Materials and Methods (1). Analysis of the control reaction, lacking exogenous PPi or TTP, permitted a background rate of 20 nM DNA pyrophosphorolyzed/min to be calculated based on the specific radioactivity of the labeled DNA. A rate of 100 nM DNA pyrophosphorolyzed/min was calculated for the reaction containing 100 µM PPi only. The rate of polymerization in the reaction mixture containing TTP and PPi, a competitive inhibitor of the dNTP substrate (7,26), was calculated to be ~7.4 µM TMP polymerized/min. The rate of polymerization in the reaction containing TTP only was calculated to be ~37.0 µM TMP polymerized/min. Thus the presence of 100 µM PPi effected a 5-fold decrease in the rate of polymerization. It must be stressed that the rate obtained for pyrophosphorolysis by this method is a only conservative approximation since no correction can be made for DNA that has undergone pyrophosphorolysis and then subsequently been converted back into starting material by action of the enzyme. Thus the rate of phosphorolysis is most probably greater than that reported here. Since dNTP concentrations in vivo are maintained at concentrations below the reported Km values and do not reach 1 mM at any time in the cell cycle (19), the in vivo rate of polymerization would be much less than reported here. Because transient local concentrations of PPi and dNTPs around the active site of TdT cannot be accurately determined, any predictions about in vivo rates for these two reactions are tenuous at best. However, the in vitro data presented here suggest that under certain conditions the rate of pyrophosphorolysis may be significant relative to the rate of polymerization; indeed, it may be greater and thus account for the loss of nucleotides observed at some Ig gene segment joins (1,2,20-25).
The genetic relatedness and apparent structural similarities between TdT and Pol [beta] (27,28) would suggest that with appropriate modifications in the nature of the substrate pyrophosphorolysis by Pol [beta] might also be detectable using this method.
The method detailed in this report for observing and analyzing the pyrophosphorolytic activity of TdT should allow for more in-depth investigations into this reaction and, perhaps, the related reaction of PPi exchange (equation 2). The increased accuracy with which the pyrophosphorolytic reaction can be detected may allow a re-examination of this reaction reported for other DNA polymerases and may lead to further insights into the role of PPi in DNA metabolism.
ACKNOWLEDGEMENT
Support for this work came from NIH grant GM25530.
REFERENCES
*To whom correspondence should be addressed at: Department of Biological and Physical Sciences, The Master's College, 21726 Placerita Canyon Road, Santa Clarita, CA 91321, USA. Tel: +1 661 259 3540; Fax: +1 661 253 4080; Email: randerson{at}masters.edu Present address: K. L. Beattie, Senior Staff Member, Health Sciences Research Division, Oak Ridge National Laboratories, PO Box 2008, MS-6123, Oak Ridge, TN 37831, USA
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||