Inactivation of avian myeloblastosis virus DNA polymerase by specific binding of pyridoxal 5'-phosphate to deoxynucleoside triphosphate binding site.

Avian myeloblastosis virus (AMV) DNA polymerase is inactivated by preincubation with pyridoxal 5'-phosphate. This inactivation is relatively specific since various pyridoxal-5'-P analogs cause no inactivation. This effect is reversible but can be made irreversible by reduction with sodium borohydride; the reduced pyridoxal-5'-P adduct exhibits a new absorbance maximum at 325 nm and a fluorescence emission at 392 nm when excited at 325 nm. The evidence presented suggests the formation of a Schiff base between pyridoxal-5'-P and a nucleophilic residue of AMV DNA polymerase. The presence of a deoxynucleoside 5'-triphosphate (dTTP) protected the enzyme from inactivation. Reduction of the pyridoxal-5'-P enzyme complex in the presence or absence of a deoxynucleoside 5'-triphosphate showed that the alpha subunit possesses five reactive amino groups, one of which is essential for catalytic activity; the beta subunit has three reactive amino groups which are not involved in the deoxynucleoside binding site.

Avian myeloblastosis virus (AMV) DNA polymerase is inactivated by preincubation with pyridoxal 5'-phosphate. This inactivation is relatively specific since various pyridoxal-5'-P analogs cause no inactivation.
This effect is reversible but can be made irreversible by reduction with sodium borohydride; the reduced pyridoxal-5'-P adduct exhibits a new absorbance maximum at 325 nm and a fluorescence emission at 392 nm when excited at 325 nm. The evidence presented suggests the formation of a Schiff base between pyridoxal-5'-P and a nucleophilic residue of AMV DNA polymerase.
The presence of a deoxynucleoside 5'triphosphate (dTTP) protected the enzyme from inactivation. Reduction of the pyridoxal-5'-P enzyme complex in the presence or absence of a deoxynucleoside 5'-triphosphate showed that the (Y subunit possesses five reactive amino groups, one of which is essential for catalytic activity; the /I subunit has three reactive amino groups which are not involved in the deoxynucleoside binding site. Pyridoxal 5'-phosphate serves as a prosthetic group for a variety of enzymes including aminotransferases, deaminases, and other enzymes concerned with reactions involving amino acids (1); it is also an essential constituent of phosphorylase (2, 3). In these enzymes it is present as a Schiff base linked to the e-amino acid group of a lysine residue (4, 5). Aside from its normal functions in enzyme catalysis, pyridoxal-5'-P, with its active aldehyde also combines with lysine residues of a number of enzymes even in cases where it is not required for catalytic activity (6-10). In order to gain more information on the nature of the active site of avian myeloblastosis virus DNA polymerase we selected pyridoxal-5'-P as an active site-directed reagent. This reagent is useful because it possesses a phosphate group which can direct it to a triphosphate binding site in the active center of the DNA polymerase and also a reactive aldehyde group which can form an adduct with the e-amino group of lysine (7,10).
In Reactions were incubated at 37" for 30 min and stopped by transfer to 4". To each reaction 0.2 mg of bovine serum albumin, 0.2 mg of yeast RNA, and 10 pmol of pyrophosphate were added followed by cold trichloroacetic acid to final concentration of 10%. After allowing the reaction mixtures to incubate for 10 min at 4", acid-insoluble material was collected on Whatman GF/C filters. Filters were washed with 10% trichloroacetic acid and the radioactivity was counted. AMV DNA polymerase was purified as described by Kacian et al. (14). Isolation and purification of the a subunit was performed according to the procedure described by Papas et al. (15). The purified fraction of AMV DNA polymerase showed two distinct bands on sodium dodecyl sulfate gel electrophoresis (Fig. 11, representing the two subunits p and 01 of M, = 97,000 and 63,000. Protein Measurements -Protein was determined by measuring the relative intensity of fluorescence upon excitation at 375 nm and emission at 490 nm as described by Bohlen et al. (16).

Enzyme
Inactivation-Treatment of DNA polymerase with pyridoxal-5'-P was carried out in the dark. When the enzyme was assayed after reacting with pyridoxal-5'-P the assay mixture contained the same concentration of pyridoxal-5'-P as the preincubation mixture; these conditions prevented dissociation of the complex.
Reducti on with Sodium Borohydride -The enzyme (88 pglml) was incubated with 1 rnM pyridoxal-5'-P at room temperature for 30 min in the dark. The solution was cooled to 0" and brought to pH 4.5 by addition of 1 M acetic acid. Octyl alcohol was added in order to avoid foaming.
A freshly prepared solution of sodium borohydride, dissolved in 0.1 M sodium hydroxide to avoid acid-catalyzed decomposition, was added in lo-fold excess with respect to pyridoxal-5'-P. During the reduction, the pH was maintained at 4.5 by addition of 1 M acetic acid. The reaction was completed within 10 min. The reduced complex was then dialyzed overnight against 10 rnM potassium phosphate buffer, pH 8.0. Absorption spectra of the dialyzed complex right, 5 pg of each of the following markers: P-galactosidase, phosphorylase a, bovine serum albumin, ovalbumin, a-chymotrypsinogen, and 6-lactoglobulin.
were obtained in a Gilford 250 spectrophotometer and fluorescent spectra on an Aminco-Bowman spectrofluorometer.
The quantity of pyridoxal-5'-P bound/m01 of protein was estimated spectrophotomee rically using a value of 8,300 for the extinction coefficient at 325 nm (17). The molecular weight for the 01 subunit was taken as 63,000 and that of crfi at 160,000 (15). Reduced complexes of aldolase and lysoxyme were used as positive controls. Pyridoxal PhosphatelTritium Borohydride Labeling-Of a solution containing 20 pg of AMV DNA polymerase, 150 ~1 were dialyzed in the cold against 85 mM Hepes buffer, pH 8.2, to remove the phosphate anion. The dialyzed solution was split into two aliquots of 75 nl, and 80 ~1 of either water or 32 mM d'l"l'P solution were added and allowed to stand for 10 min at 25" to permit the binding of dTTP to the enzyme. Subsequently, 20 ~1 of a 12.5 mM pyridoxal phosphate solution were added to each sample and the reaction was allowed to proceed for 30 min in the dark at 25". To the solutions cooled in ice were added sequentially 25 ~1 of 1 N acetic acid to bring the pH to 4.5, and 25 ~1 of 70 mM NaB3H4 (specific activity 7 Ci/mmol) in 5 rnM NaOH.
The reductive step was allowed to proceed for 10 min in the cold and the excess of reactants was removed by extensive dialysis against 10 mM sodium phosphate buffer, pH 8. In order to avoid tritium incorporation into foreign products present in dialysis sacks, the dialysis tubing was boiled for 20 min in 0.2 N Na,CO, containing 0.01% NaBH,.
The treated membranes were washed with boiling water, cold water, 0.5 mM HCl, and finally by cold water. The incorporation of tritium label into the polymerase was measured in the trichloroacetic acid-precipitable material, using 70 pg of serum albumin as a carrier. Tritium-labeled enzyme was analyzed electrophoretically in 8% polyacrylamide gels in the presence of sodium lauryl sulfate. The amount of tritium label present in the (Y and 6 subunits was estimated by fractionating the gel in a Gilson Aliquots of the enzyme (44 @g/ml) were preincubated with pyridoxal-5'-P for 5 min at 37" in the dark. Then 5-~1 aliquots were analyzed for enzymatic activity in assays containing the same concentrations of pyridoxal-5'-P. The assays were also performed in the dark.

RESULTS
Inhibition of AMV DNA Polymerase by Pyridoxal-5'-P-DNA polymerase was inactivated by preincubation with pyridoxal-5'-P. To study this effect, the enzyme was preincubated with pyridoxal-5-P and then analyzed in assay mixtures that also contained the reagent. It was necessary to add pyridoxal-5'-P to the assay mixture to avoid dissociation of enzyme . reagent complexes. The inhibition of DNA polymerase activity was dependent on the concentration of pyridoxal-5'-P (Fig. 2); 50% inhibition was obtained at a concentration of 0.2 mM pyridoxal-5-P and complete inhibition at 1 111~. Effect of Analogs of Pyridoxal-5'-P on AMV DNA Polymeruse Activity-Analogs were tested for their effect on the enzyme activity at a concentration of 1 mM (Table Il. In each case, complete inhibition was observed only with pyridoxaM'-P, therefore, both the aldehyde and the phosphate groups were required for complete inhibition. Effect of Primary Amino Groups on Pyridoxal-5'-P Znhibition of AMV DNA Polymerase-Complete inhibition of the enzyme by pyridoxal-5'-P was regularly observed except when the buffer system contained primary amino groups, as in the case of Tris, when only partial inhibition was observed. In the presence of high concentrations of Tris (known to react reversibly with the aldehyde group of pyridoxal 5'-phosphate) the enzyme was inhibited by only 54%. Since the enzyme is capable of reacting with free pyridoxal5'-phosphate obtained from the dissociation of the pyridoxal5'-phosphate.
Tris complex, it appears that the enzyme has a high affinity for pyridoxal 5'- phosphate. This high affinity may result from the phosphate moiety of pyridoxal 5'-phosphate directing the reagent toward the phosphate binding site of the enzyme and facilitating the subsequent reaction of its aldehyde group with the amino group (possible e-lysine) of the enzyme. The essential role of the phosphate for inhibition is demonstrated by the failure of pyridoxal to inhibit the enzymatic activity (Table II). Additionally, cycloserine with a strongly nucleophilic active group almost completely overcomes the inhibition of enzyme activity (Table III).

Absorption
Spectra for Reduced and Unreduced Pyridoxal-5'-P .Enzyme Complex-Spectrophotometric analysis ( Fig.  3A) of solutions containing pyridoxal-5'-P and the enzyme exhibited a characteristic absorption peak at 415 nm, suggesting the formation of a Schiff base between the aldehyde of pyridoxal-5'-P and a primary amino group on the enzyme. After reduction with sodium borohydride, a new absorption peak at 325 nM appeared (Fig. 3B). This spectral change is characteristic of the reduction of a Schiff base to a stable enzyme. pyridoxal-5'-P complex.
Erzyme Complex -Additional information on the pyridoxal-5'-P.enzyme reduced complex is shown in Fig.  4. A fluorescence emission was observed between 390 to 400 nm when excited at 325 nm. This suggests the formation of an Xazolidine-like structure (211, a further derivative of the Schiff base which involves a second nucleophilic residue of the enzyme presumably located in the vicinity of the enzymatic active site. -Preincubation of the enzyme with deoxynucleoside 5'-triphosphate and assaying in the presence of pyridoxal-5'-P resulted in a complete protection against inhibition. No such protective effect was observed when preincubation was carried out in the presence of two template-primers (Table IV). These results are consistent with the interpretation that pyridoxal-5'-P interacts with the deoxynucleoside 5'-triphosphate site and not with that of the template-primer.

Stoichiometry
of Zncorporation of Pyridoxal-5'-P on a and af3AMV DNA Polymerase in Presence and Absence of TTP-As shown earlier, the enzyme. pyridoxal-5'-P complex was made irreversible and stabilized by sodium borohydride. Under these conditions, inactivation was complete. The number of moles of pyridoxal-5'-P incorporated/mol of enzyme was estimated from optical absorption measurements.
Table V gives a quantitative measure of the number of amino groups modified by pyridoxal-5'-P.
The deoxytriphosphate (dTTP) blocks 1 amino acid residue in the case of o( and ap (Table V). Electrophoretic analysis in the presence of sodium dodecyl sulfate showed that pyridoxal phosphate reacts preferentially with the primary amino groups of the a subunit in the a/? form of the AMV DNA polymerase (Fig. 61. That the incorporation of label is not dependent on the proportions of (Y and /3 in the dimer is shown by the fact that the native preparation showed an a:/3 molar ratio of 1:l. From the total tritium label in the a with excess pyridoxal-5'-P at room temperature for 30 min in the dark. This was read against an identical sample without enzyme. A g-fold dilution of each was performed prior to reading. B, absorption spectra of sodium borohydride . reduced pyridoxal-5'-P'enzyme complex. DNA polymerase (88 pg/ml) was incubated with excess pyridoxal-5'-P at room temperature in the dark. After 30 min, the mixture was cooled to 0", reduced, and dialyzed as described under "Experimental Procedures." The reference was a sample treated identically but without pyridoxal-5'-P. and p band in electrophoresis, the relative distribution of radioactivity in the (Y and p subunit was calculated (Table VI). From the relative distribution number and the total number of pyridoxal phosphate groups in the (YP dimer, obtained from spectrophotometric studies (Table V), it was possible to calculate the absolute number of pyridoxal phosphate groups incor-Binding of Pyridoxal 5'-Phosphate to AMV DNA Polymeruse 1429 rr The percentage of inactivation represents the degree of inhibition of the unreduced pyridoxal-5'-P. enzyme complex. Assays were performed as described under "Experimental Procedures." b Purified a subunit (880 pg/ml) and native enzyme ap (880 ng/ ml) were incubated for 10 min at room temperature in the presence or absence of 10 mM TTP. Then 1 mM pyridoxal-5'-P was added and the incubation was continued for 30 min in the dark. The solutions were cooled to O", and the pH adjusted to 4.5 with 1 M acetic acid. Freshly prepared NaBH, was added to a final concentration of 10 rnM and the pH was readjusted to 4.5. The reduction was completed within 10 min and the reaction mixture was dialyzed overnight against 10 mM potassium phosphate buffer, pH 8.0. The quantity of pyridoxal-5'-P bound/m01 of protein was estimated spectrophotometrically using a value of 8,300 for the extinction coefficient at 325 nm. Molecular weights of 01 and alp were 63,000 and 160,000, respectively (15). Approximately 6000 cpm of "H bound to protein were applied in 1% 2-mercaptoethanol and 1% sodium dodecyl sulfate over a 8% acrylamide gel using N,N'-diallyltartadiamide as the crosslinking agent. Electrophoresis was carried out at 7 mA/gel until the bromophenol blue marker reached 8 cm. The fractionation of the gels and measurements of radioactivity were carried out as described in the text. A, :'H-labeled polymerase in the absence of dTTP, B, "Hlabeled polymerase labeled in the presence of 10 rnM dTTP. porated/mol of 01 or p subunit (Table VI). From these data it is apparent that the /3 subunit has three amino groups able to react with pyridoxal phosphate and that dTTP does not interact with any of these groups. The cy subunit possesses five reacting amino groups in the absence of d?TP and four in the presence of dTTP. These results indicate that the interactions between dTl'P and the ~$3 form of the enzyme involves a single essential amino group present in the CY subunit. The formation of the CQ form of the polymerase must involve interactions between the exposed groups of the separate (Y and p subunits. Therefore, it is possible to assume that four of the nine amino groups reactive toward pyridoxal-5'-P present on the isolated CY subunit (Table V) are buried upon surface interactions with the monomeric fl subunit to form the C@ enzyme.

DISCUSSION
Reverse transcriptase is involved in the replicative cycle of RNA tumor viruses (22,23). The enzyme transcribes its high molecular weight RNA genome into DNA, a process known to be essential in the infectious process of these virues. AMV DNA polymerase copies heteropolymeric regions of poly(A1 containing RNAs and under certain conditions it can make a complete cDNA copy (241. The enzyme requires both a template-primer and deoxynucleoside triphosphates and is unable to initiate DNA chain synthesis de nouo in the absence of a primer containing a 3'-OH terminus. The biological importance of this enzyme as well as its usefulness as an analytical tool, prompted us to study both its mechanism of action and chemical structure (13,15,25,26).
Pyridoxal-5'-P, in addition to its catalytic function to certain enzyme systems (1,3), has become a useful active site-directed reagent (27). The inactivation of AMV DNA polymerase is relatively specific for pyridoxal-5'-P since various analogs cannot inactivate the enzyme (Table I); the specificity of inhibition is therefore a result of both the aldehyde and phosphate groups. The interaction of pyridoxal-5'-P with the enzyme has been demonstrated to involve the formation of a Schiff base between the enzyme and amino groups of the protein. Studies with NaBH, reduction of pyridoxal-inactivated enzyme affords supporting evidence for a primary Schiff base formation in the inactivation process. The fluorescence emission band at 390 nm observed when the pyridoxal . enzyme is excited at 325 nm is likewise characteristic of a Schiff base formation.
Preincubation of the enzyme with the substrate (dTTP) prevents the binding of pyridoxal-5'-P and results in the maintenance of catalytic activity; such protection cannot be obtained when the enzyme is preincubated in the presence of  (Table IV). In addition, pyridoxal-5'-P is a competitive inhibitor of deoxynucleoside triphosphate (dlTP) but not of the template-primer (Fig. 5). Thus pyridoxal-5'-P appears to bind at or near the active site of the enzyme. Borders et al. (28) have shown that AMV DNA polymerase is inactivated by butanedione, a reagent which selectively modifies arginyl residues. In their case, the enzymatic activity was protected by template-primer but not by deoxynucleoside triphosphate.. They concluded that arginyl residues are essential for binding of template-primer to the active site. Comparison of their findings with those presented here suggest that the enzyme possesses two independent binding sites: the deoxytriphosphate site which involves a single essential amino group located in the a subunit of the enzyme and a templateprimer binding site involving essential arginine residue(s). Early work by Papas et al. (13) showed that sequential addition of substrates resulted in a synergistic protection of the enzymatic activity against heat inactivation.
This can be explained by a stabilization of the different regions of the active site as a result of interaction between the binding site and their respective ligands. The suggestion of independent binding sites obtained from heat inactivation (13) is now strongly supported by specific chemical modification with butanedione (28) and pyridoxal-5'-P.
Spectrophotometric studies of pyridoxal-5'-P. enzyme formed in the presence or absence of dTTP shows that the aP enzyme form possesses eight reactive groups, one of which is protected by dTTP. Under similar conditions the isolated a subunit possesses nine reactive groups, one of which is protected by dTTP. Reduction of pyridoxal-5'-P.enzyme with 13H1NaBH, allows the incorporation of tritium label into the stable pyridoxal group of the complex. Electrophoretic analysis of labeled aj3 enzyme prepared in the presence and absence of dTl!P substrate allows a calculation of the relative distribution of the pyridoxal groups in the /3 and (Y subunits. The /3 subunit has three reactive groups, none of which is protected by dTTP, while the a subunit has five reactive groups, one of which is protected by dTI'P and is therefore essential for catalytic activity. The single group which is essential for catalytic activity is located in the a subunit and is not buried during the interaction of the subunits. It is interesting to note that the p subunit of the a/3 form does not have a binding site for deoxytriphosphate. It is well documented that a precursor relationship exists between the p and a subunits (15,29,30). The larger p subunit contains a polypeptide with a molecular weight of 31,000 in addition to the (Y sequences (15). The presence of this peptide could induce changes in the tertiary structure of the sequences corresponding to the a region of the p subunit and alter the configuration of the deoxytriphosphate binding site. Alternatively, the absence of this site could be the result of steric hindrance in the region of a and p interactions. Lack of reliable procedures for the isolation of the /3 subunit has limited our efforts to test these possibilities.
We are presently concentrating our efforts on the isolation and sequencing of peptides containing the essential residue for the deoxytriphosphate binding site. These studies will be expanded to other oncornavirus polymerases available in our laboratory including those from fish (311, reptiles (321, and mammals in an effort to determine the degree of conservation of the primary amino acid sequences in the region of the deoxytriphosphate binding sites.