Eur. J. Biochem. 58, 297-302 (1975)

Polynucleotide Kinase from Rat-Liver Nuclei Purification and Properties Hirobumi TERAOKA, Kazutaka MIZUTA, Fumiyasu SATO, Mari SHIMOYACHI, and Kinji TSUKADA Department of Pathological Biochemistry, Medical Research Institute, Tokyo Medical and Dental University (Received April 28/JuIy 9, 1975)

A polynucleotide kinase, which catalyzes the phosphorylation of 5'-hydroxyl ends of deoxyribonucleic acid in the presence of adenosine triphosphate, has been purified 260-fold with a yield of 14 % from 0.15 M NaCl extracts of rat liver nuclei. The purified enzyme has a pH optimum of 5.5. The enzyme is reversibly inhibited by p-chloromercuribenzoate. The So,5value (ligand concentration required for a half-maximal activity) for ATP is 2.5 pM. A bivalent cation is essential for the reaction and ,So,5 values for M$+, Ca2+ and Mn2+ are 3.3 mM, 4 mM and 0.05 mM respectively. Pyrophosphate remarkably inhibits the activity with Zo.5 value (ligand concentration required for a half-maximal inhibition) of 0.2 mM, and sulfate, with Zo.5 of 0.5 mM, whereas phosphate weakly inhibits the activity with Io,5of about 20 mM. An apparent molecular weight of the purified enzyme is estimated to be 8 x 104 by gel filtration on a column of Sephadex G-150, and the Stokes radius of the enzyme molecule is shown to be about 0.36 nm. Sucrose density gradient centrifugation reveals that the enzyme has a sedimentation coefficient of about 4.4 S.

A polynucleotide kinase, which catalyzes the transfer of orthophosphate from ATP to the 5'-hydroxyl groups of a wide variety of nucleic acid compounds, was isolated from TZinfected Escherichia coli by Novogrodsky et al. [1-31 and from T4-infected E. coli by Richardson [4]. Polynucleotide kinase in eukaryotic cells has been found in nuclear extracts from rat liver [3,5]and some properties of this activity in crude extracts of rat liver nuclei have been reported previously by our group [5]. The present paper describes a method for obtaining purified enzyme from nuclear extracts of rat liver, and catalytic and molecular properties.

Centre (Amersham, England). Bovine serum albumin, horse heart cytochrome c, E. coli alkaline phosphatase, pancreatic DNase and snake venom 5'-nucleotidase were purchased from Sigma (St. Louis, U.S.A.). Rabbit muscle lactate dehydrogenase was obtained from Boehringer Mannheim GmbH (Mannheim, Germany). Calf thymus DNA and snake venom phosphodiesterase were from Worthington (New Jersey, U.S.A.). Phosphocellulose (P11) was obtained from Whatman (Maidstone, England), Sephadex G-150 (superfine) from Pharmacia (Uppsala, Sweden) and collodion bags from Sartorius-Membranfilter GmbH (Gottingen, Germany). All other chemicals were of analytical grade.

MATERIALS AND METHODS Preparation of Rat Liver Nuclei [ Y - ~ ~ P I A(sodium TP salt in 50 % aqueous ethanol, 2.0 Ci/mmol) was obtained from the Radiochemical ~~

A66reviurion.s. So,,, ligand concentration requircd for a halfmaximal activity; lo.s,ligand concentration required for a halfmaximal inhibition. Enzymes. Polynucleotide kinase or ATP : S-dephosphopolynucleotide 5'-phosphotransferase (EC 2.7.1.78); lactate dehydrogenase (EC 1.1.1.27); DNase I (EC 3.1.4.5); DNase I1 (EC 3.1.4.6); RNase I (EC 3.1.4.22); 5'-nucleotidase (EC 3.1.3.5); alkaline phosphatase (EC 3.1.3.1); phosphodiesterase (EC 3.1.4.1); polynucleotide ligase (AMP-forming) (EC 6.5.1.I).

Rat liver nuclei were isolated by the method described previously [ 5 ] , similar to that reported by Chauveau et al. [6]. Wistar male rats (150-200 g) maintained on laboratory chow and water ad libitum were killed by decapitation. The livers were quickly removed, rinsed in 0.15 M NaCl, and were homogenized in four volumes of ice-cold 0.25 M sucrose containing 3.3 mM MgC12 in a glass/Teflon homogenizer. The homogenate

Polynucleotide Kinase from Rat-Livcr Nuclei

298

was passed through four sheets of gauze and centrifuged at 1000x g for 10 min at 4 "C. The precipitate was suspended in nine volumes of 2.2 M sucrose containing 3.3 mM MgCl, and centrifuged at 100000 x g for 60 min. The nuclei obtained were washed twice with 0.25 M sucrose containing 3.3 mM MgC1,.

DNase I and I1 activities were determined by measuring acid-soluble 3H from 3H-labelled DNA (2 x lo3 counts x min-' x mg-') in Tris-HC1 buffer, pH 7.5, containing 6 mM MgC1, and in acetate buffer, pH 4.5 respectively. Polynucleotide ligase was assayed as described previously [7].

Preparation o j D N A Containing 5 '-Hydroxyl-Terminated Breaks

Identification of 32P-LabelledD N A Product Formed in Kinase Reaction

5'-Hydroxyl-terminated DNA, a substrate for polynucleotide kinase, was prepared from calf thymus DNA with the use of pancreatic DNase and alkaline phosphatase, as described previously [5]. DNA concentration was expressed as nucleotide equivalent. Assay of Kinuse

The activity of polynucleotide kinase was assayed essentially as described previously [5] with slight modifications. The standard reaction mixture (0.3 ml) contained 25 pmol of succinate buffer, pH 5.5,3 pmol of MgCl,, 5 pmol of 2-mercaptoethanol, 87 nmol of 5'-hydroxyLterminated calf thymus DNA, 2.5 nmol of [y-32P]ATP (specific activity, 4.4 x 10' counts x min-' x pmol-') and the enzyme. The reaction mixture was incubated for 10 min at 37 "C. After the mixture was chilled in ice to stop the reaction, 0.1 ml of 0.5 N NaOH was added and the mixture was boiled for 20 min. To the mixture chilled in ice, 0.1 ml of 0.1 M sodium pyrophosphate was added, followed by 5 ml of 5 % trichloroacetic acid and 1 ml of a suspension of 30 mg celite in 5 % trichloroacetic acid. The precipitate was collected on a pad of celite and washed extensively with trichloroacetic acid, ethanol and ether. The precipitate was dissolved in 0.5 ml of Hyamine, and the radioactivity was counted in a Packard Tri-Carb liquid-scintillation spectrometer. Scintillator solution contained 333 ml of Triton X-100, 2.7 g of 2,5-diphenyloxazole, 0.07 g of 2,2'-pphenylene-bis(5-phenyloxazole)and toluene in a total volume of 1 1. One unit of the enzyme was defined as the amount which catalyzes the incorporation of 1 nmol of Pi into DNA from ATP/min under the standard assay conditions. Specific activity was expressed as units/mg protein. Assay of Interfering Enzymes

Activities of alkaline phosphatase and phosphodiesterase were determined by measuring released p-nitrophenyl anion at 405 nm with the use ofp-nitrophenyl phosphate and bis@-nitrophenyl phosphate) as substrates respectively. b

Identification of 32Pat the 5'-terminus of the polynucleotide chain was carried out as described previously [5].32P-labelledproduct formed in the kinase reaction was treated with each of alkaline phosphatase, phosphodiesterase, 5'-nucleotidase and phosphodiesterase plus 5'-nucleotidase. The 32P released was determined as acid-soluble and Norit-nonabsorbable 32P. Analytical Gel Filtration

Analytical gel chromatography was carried out at 4 "C on a column (0.9 x 52 cm) of Sephadex G-150 equilibrated with 0.01 M potassium phosphate, pH 7.5, containing 0.2 M KCI, 0.5 mM dithiothreitol and 0.1 mM EDTA. Enzyme solution (0.5 ml) was applied on the column and eluted in the same buffer. Fractions of 0.7 ml were collected and 10-pl aliquots were assayed for the enzyme activity. The molecular weight was estimated according to the method of Andrews [8]. The Stokes radius of the enzyme molecule was determined according to the method of Siege1 and Monty [9]. Proteins employed as standards with the following molecular weight and Stokes radii were lactate dehydrogenase (140000, 0.435 nm), bovine serum albumin (68000, 0.35 nm) and cytochrome c (12300, 0.17 cm). Sucrose Density Gradient Centrifugation

The centrifugation was carried out at 4 "C and 38 500 rev./min for 16 h in the Beckman L-5 centrifuge in a 13-ml sucrose linear gradient (5 - 20 %) containing 0.01 M potassium phosphate, pH 7.5, 0.2 M KCI, 0.5 mM dithiothreitol and 0.1 mM EDTA. Lactate dehydrogenase (szO,,, = 7 S), bovine serum albumin ( s , ~ ,=~ 4.4 S ) and cytochrome c (sz0,,, = 1.7 S) were used as standard proteins. Enzyme solution (0.2 ml) was subjected to centrifugation with standard proteins and 0.32-ml fractions were collected from the bottom of the tube. Determination of Protein and D N A

Protein was determined by the method of Lowry et al. [lo] with bovine serum albumin as standard.

--

H. Teraoka, K. Mizuta, F. Sato, M. Shirnoyachi, and K. Tsukada

299

40 were pooled and concentrated to 0.5 ml with a

2

collodion dialysis bag. Step 4 : Sephadex G-150 Column Chromatography. The enzyme solution from step 3 was applied on a column of Sephadex G-150 (1 x 50 cm) equilibrated with buffer A containing 0.2 M KC1. Fractions of 0.7 ml were collected and 10-p1 aliquots were used for the enzyme assay. Fraction numbers 23 - 28 were pooled and concentrated to 0.5 ml with a collodion dialysis bag. The result of the purification is presented in Table 1.

9

General Properties

0.3

0 E

0.1 0

6 hl c

0.05

z

0 Fraction number

Fig. 1. Pho.vphocrllulose column chromatograph!' of kinose. Detailed Enzyme ) activity; conditions arc described in the text. (M (0-0) absorbance at 280 nm; ( x -x) salt concentration

Protein concentrations on column chromatography were determined by measuring absorbance at 280 nm. DNA was determined by the procedure of Burton [11I.

RESULTS Purfication of Kinase

All operations described below were carried out at 0-4 "C. Step I : Preparation of Nuclear Extracts. 7 g (wet weight) of nuclei were suspended in 140 ml of 0.15 M NaCl containing 5 mM 2-mercaptoethanol, and stirred gently with a magnetic stirrer for 20 min. The suspension was centrifuged at 10000 x g for 15 min to obtain nuclear extracts. Step 2: p H 5.0 Treatment. The nuclear extracts (136 ml) were brought to pH 5.0 by the dropwise addition of 0.2 N acetic acid and stirred for 5 min. After the solution had been centrifuged at 10000 x g for 10 min, the resulting supernatant was adjusted to pH 7.5 by the addition of 2 N ammonium hydroxide. Step 3 :Phosphocellulose Column Chromatography. The enzyme solution (136 ml) was applied on a column of phosphocellulose (2 x 6 cm) previously equilibrated with buffer A (0.1 M potassium phosphate, pH 7.5, 0.5 mM dithiothreitol and 0.1 mM EDTA) containing 0.15 M KCl, washed with 15 ml of the same buffer and eluted with a linear gradient between 20ml of buffer A containing 0.15 M KC1 and 20 ml buffer A containing 0.4 M KCl. Fractions of 2.7 ml were collected and 10yl aliquots were used for the assay of the enzyme activity. The enzyme activity was eluted at about 0.22 M KCl (Fig. 1). Fraction numbers 38-

The purified polynucleotide kinase (steps 3 and 4) was stable for at least a week at 0-4 "C without significant loss of its activity. The enzyme with every step, however, lost more than 80% of its activity when treated with ammonium sulfate, as in the case of ammonium sulfate fractionation, even for a short time, and the activity was not overcome by depriving of salts. In the enzyme preparation of steps 3 and 4, activities of DNase I, DNase 11, phosphodiesterase, alkaline phosphatase and polynucleotide ligase were not detectable. Therefore, the purified enzyme was considered to be preferable to nuclear extract for the preparation of 5'-32P-labellednicked DNA as a substrate for polynucleotide ligase [7; 12,131. As shown in Table 2, the incorporation of 32Pi from [ Y - ~ ~ P I A into T P DNA by the purified enzyme was completely dependent on both DNA and MgC12, and the incorporation was abolished by the addition of pancreatic DNase. Preliminary experiments indicate that the 5'-hydroxyl RNA can act as a phosphate acceptor and the incorporation of 32Pifrom [ Y - ~ ~ P ] ATP into RNA was sensitive to RNase but insensitive to DNase. The purified enzyme was inhibited byp-mercuribenzoate and this inhibition was reversed completely by the addition of 2-mercaptoethanol (Table 2), indicating that the enzyme contains sulfhydrylgroup(s) essential to the activity [5]. A 32P-labelled DNA product was prepared and subjected to alkaline phosphatase, phosphodiesterase, 5'-nucleotidase and phosphodiesterase plus 5'-nucleotidase. When the DNA product was subjected to alkaline phosphatase, more than 90% of 32P was released as Pi. Neither phosphodiesterase nor 5'nucleotidase released 32P as Pi from the [32P]DNA, but combined addition of both enzymes resulted in the release of more than 90% of 32P as Pi [5]. These results indicate that the product of the kinase reaction was a 5'-phosphate-terminated DNA. p H Optimum of the Enzyme

The effect of pH on the enzyme activity was studied by using the purified enzyme in Tris-acetate buffer,

300

Polynucleotide Kinase from Rat-Liver Nuclei

Table 1. Pur[ficarion of polynucleotide kinase from rat liver nuclei The purification and assay procedure of kinase activity are described in the text Step

1. 2. 3. 4.

Nuclear extract pH 5.0 treatment Phosphocellulose Sephadex G-150

Protein

Activity

Specific activity

Yield

Purification

mg

units

units/mg

%

-fold

92.4 51.6 1.10 0.05

15.0

0.16 0.33 3.08 41.8

100 113 23 14

1 2 19 26 1

17.0 3.39 2.09

Table 2. Requirementsfor kinase reaction The reaction mixtures were the same as described in Materials and Methods except for the presence of 0.6 pg of step 3 enzyme with or without indicated materials Components

32P incorporated

pmol/lO min Complete Minus DNA Minus MgCI2 Plus DNase (5 pg) Plus RNase ( 5 pg) Minus 2-rnercaptoethanol Minus 2-mercaptoethanol plus p-chloromercuribenzoate (0.033 mM) Minus 2-mercaptoethanol plus p-chloromercuribenzoate (0.033 mM) and 2-mercaptoethanol(i6 mM)

18.9 0.1 0.1 0.1 18.4 18.0

"

4

5

6 PH

0.6 18.6

pH 4.0- 8.0. Polynucleotide kinase showed a pH optimum of 5.5 (Fig.2), as previously determined in the enzyme of nuclear extract [ 5 ] . The enzyme activity decreased to about one-fifth at pH 7.0 and to about one-seventh at pH 4.5.

7

8

Fig. 2. p l i dependence of kinase reaction. Experimental conditions were the same as described in Materials and Methods except that the reaction mixture contained 0.06 pg of step 4 enzyme and 25 pmol of Tris/acetate buffer with indicated pH value, instead of succinate buffer. Reaction velocity was expressed as pmol of 32Pincorporated into DNA in 10 min

/-

*Ot

Kine tic Proper ties

The kinase reaction was dependent on both ATP and 5'-hydroxyl DNA. As shown in Fig. 3, the relation between reaction velocity and ATP concentrations was hyperbolic and the So.5value of 2.2 pM was obtained from the double-reciprocal plots. The So,5 value for DNA was 35.5 pM and the value was significantly altered with different preparations of 5'hydroxyl DNA. The kinase reaction exhibited an absolute requirement for bivalent cations (see Table 2). Ca2+ as well as Mg2' and Mn2+ was effective for the reaction, and Zn2+ or Coz+ was less effective. So.5 values for Mg2+ and CaZ+ were 3.3 mM and 4 mM respectively. In the case of Mn2+ the relation between reaction velocity and thc metal concentrations was observed to be slightly sigmoidal and the So.5 value was estimated to be 0.05 mM. Pyrophosphate, phosphate and sulfate significantly inhibited the kinase reaction. Pyrophosphate inhibited

J

0 0.5 l / [ A T P ] (IM-')

-0.5

0

0

1.0

I

I

I

I

I

I

2

4

6 [ATPI

8

10

12

(IN

Fig. 3. .Ef;ecl u f A TP concmtrafiuris 011 kinasr acfivity. Experimental conditions were as described in Materials and Methods except that the reaction mixture contained 0.06 pg of step 4 enzyme and ATP in the indicated concentrations. Reaction velocity ( I : ) was expressed as pmol of 32Pincorporated in 10 min

the enzyme with an 10.5value of 0.2mM (Fig.4A). Phosphate was less inhibitory than pyrophosphate with an 10.5value of about 20 mM (Fig. 4B). The enzyme activity was also affected by sulfate (Zo.5 = 0.5 mM) (Fig. 4C).

30 1

H. Teraoka, K . Mizuta, F. Sato, M. Shimoyachi, and K. Tsukada

[Pyrophosphate] (mM)

[Phosphate] (rnM)

[Sulfate] (rnM)

Fig. 4. €/ficr oJnnion concentrutions on kiriase activity. Experimental conditions were as described in Materials and Methods except that the reaction mixture contained 0.6 pg of step 3 enzyme and the indicated concentration or anion. Reaction vclocity was expressed as pmol of 3zP incorporated in 10 min. (A) Pyrophosphatc: (B) phosphate; (C) sulfate

Molecular Properties When the step 4 enzyme was applied on a column of Sephadex (3-150, the activity was eluted as a single symmetrical peak ahead of the elution volume of bovine serum albumin. An apparent molecular weight of about 8 x lo4 was obtained assuming that polynucleotide kinase is a globular protein. The Stokes radius of the enzyme was estimated to be about 0.36 nm from the same gel filtration data. The sedimentation coefficient was obtained from sucrose density gradient centrifugation. Polynucleotide kinase (step 4) was sedimented as a single symmetrical peak at the same position as bovine serum albumin ( s ~ =~ 4.4 , ~S), employed as standard. Although the behavior of kinase relative to that of bovine serum albumin was different in gel filtration and sedimentation experiments, it seems likely that the disagreement is due to asymmetrical features of kinase molecule.

DISCUSSION Polynucleotide kinase has been purified from T2-infected [2] and T4-infected Escherichia coli [4]. No activity has been found in uninfected cells. The results presented above indicate a method to purify polynucleotide kinase from rat-liver nuclei and some of its catalytic and molecular properties. Some fundamental properties in catalysis, such as pH optimum, requirements of ATP and Mg2+ and inhibition by y-chloromercuribenzoate are the same as those obtained previously in the crude enzyme preparation [ 5 ] . The crude kinase in nuclear extract of rat liver has been used to prepare 32P-labelled nicked DNA at pH 5.5 as a substrate of polynucleotide ligase [7,12,13]. Since the enzyme preparations of step 3 or 4 do not contain any detectable activities of some of enzymes interfering with the kinase assay, the purified enzyme seems to be preferable to the nuclear extract for the preparation of 32P-labelled nicked DNA.

The kinase activity was undetectable in lo5 x g supernatant solution from rat liver at various pH values. Therefore, the enzyme seems to be located almost exclusively in nuclei. The enzyme activity was distributed not only in a 0.15 M NaCl extract of nuclei but also in the chromatin fraction ; the NaC1-extract enzyme is 3 - 5-fold higher in total and specific activities than the chromatin enzyme. In regenerating rat liver, nuclear DNA synthesis begins to increase about 12- 16 h after the operation and reaches maximum at 19-24 h [14]. During the regeneration polynucleotide ligase, which catalyzes fixation of single-strand breaks in a duplex DNA strand, increases several-fold with increasing DNA synthesis [7,13]. On the other hand, the polynucleotide kinase activity both in the 0.15 M NaCl extract and chromatin fraction remains constant, even at 22 h after partial hepatectomy. Although the function of the polynucleotide kinase remains unknown in mammalian cells at present, this enzyme is presumed to play a significant role to repair damaged DNA in co-operation with polynucleotide ligase. The authors are indebted to Miss Reiko Ishiwatari for the preparation of this manuscript. This investigation was supported in part by the Scientific Research Fund for Cancer from the Ministry of Education, Japan.

REFERENCES 1. Novogrodsky, A. & Hurwitz, J. (1965) Fed. Proc. 24, 602. 2. Novogrodsky, A. & Hurwitz, J. (1966) J . Biol. Chem. 241, 2923 - 2932. 3. Novogrodsky, A., Tal, M., Traub, A. & Hurwitz, J. (1966) J. Biol. Chem. 241,2933 - 2943. 4. Richardson, C. C. (1965) Proc. Nut1 Acad. Sci. U.S.A. 54,

158-165.

5. Ichimura, M. & Tsukada, K . (1971) J . Biochern. (Tokyo) 69, 823 - 828. 6. Chauveau, J., Moule, Y .& Rouiller, C. (1956) Exp. Cell Res. 11,317-321. 7. Tsukada, K. & Ichimura, M. (1971) Biochern. Biophys. Res. Commun. 42,1156- 1161.

302

H. Teraoka, K. Mizuta, F. Sato, M. Shimoyachi, and K. Tsukada: Polynucleotide Kinase from Rat-Liver Nuclei

8. Andrews, P. (1965) Biochem. J . 96, 595-606. 9. Siegel, L. M. & Monty, K. J. (1966) Biochim. Biophys. Acta, 112,346- 362. 10. Lowry, 0. H., Rosebrough, N. J . , Farr, A. L. & Randall, R. J. (1951) J . Biol. Chem. 193, 265-275. 11. Burton, K . (1955) Biochem. J . 61,473-483.

12. Tsukada, K., Hokari, S., Hayasaki, N. & Ito, N. (1972) Cancer R ~ s32, . 886-888. 13. Tsukada, K . (1974) Biochem. Biophys. Res. Commun. 57, 758- 762. 14. Tsukada, K., Moriydma, T., Lynch, W. E. & Lieberman, I. (1968) Nuture (Lond.) 220, 162- 164.

H. Teraoka and K. Tsukada, Department of Pathological Biochemistry, Medical Research Institute, Tokyo Medical and Dental University, Yushima, Bunkyo-ku, Tokyo, Japan K. Mizuta, Research Laboratories, Otsuka Pharmaceutical Co., Tokushima, Japan

F. Sato and M. Shimoyachi, Department of Biochemical Pathology, Drug Research Institute, Toyama University, Gofuku, Toyama, Japan