Poly(ADP-Ribose) Synthetase, a Main Acceptor of Poly(ADP-Ribose) in Isolated Nuclei*

Poly(ADP-ribose) synthetase was identified as the main acceptor of this polymer produced in isolated nuclei of rat liver. When the nuclei were incubated with [32P]NAD at a limited concentration (2.4 PM) and for a brief period (10 s), a protein with M, = 110,000 was predominantly poly(ADP-ribosyl)ated, as judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The modification of this protein increased upon longer incubations or at higher NAD concentrations, and induced a marked increase in the apparent molec- ular weight. A comparison with poly(ADP-ribose) synthetase (Mr = 110,000) of rat liver under various con- ditions suggested that the increase in the molecular weight of the acceptor resembled that of the synthetase undergoing multiple auto-poly(ADP-ribosy1)ation. This interpretation was further supported by the following observations:

Poly(ADP-ribose) is a macromolecule synthesized from NAD on various protein acceptors in nuclei (1)(2)(3)(4). Its biological function has not yet been fully elucidated, but several lines of evidence have suggested its participation in DNA replication (5), DNA repair (6), cell differentiation (7, €9, or neoplastic transformation (9). Whether or not these possible functions are correlated to modification of specific acceptor(s) has not been extensively studied. So far, histones (mainly H1 and H2B) have been reported as the main acceptors in vitro (9)(10)(11)(12)(13)(14) and in vivo (15). It should be noted, however, that this view holds true only for the material extractable with dilute acid more than a half of poly(ADP-ribose) synthesized in nuclei and other systems remains uncharacterized in the residue. We recently reported that a protein with M , = 110,000 was mainly modified in lymphocytes before as well as after * This work was supported in part by grants-in-aid for Scientific and Cancer Research from the Ministry of Education, Science and Culture, Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. treatment with a DNA-damaging reagent, N-methyl-N'-nitro-N-nitrosoguanidine, and that this protein behaved closely to the synthetase of this polymer (16). In the present study, we extended the analysis to rat liver nuclei, and obtained evidence that the main acceptor in the nuclear system was not histone but poly(ADP-ribose) synthetase.

Materials-[AMP-"PINAD
(34 Ci/mmol) and [Ac~~-'~C]NAD (266 Ci/mol) were obtained from New England Nuclear and the Radiochemical Centre, Amersham, respectively. Molecular weight calibration kits, radioactive and nonradioactive, were the products of the Radioactive Centre, Amersham, and Pharmacia, respectively. Calf thymus DNA (type I) was purchased from Sigma, proteinase K from Merck, and 3-aminobenzamide from Tokyo Kasei. Nuclei and poly(ADP-ribose) synthetase were prepared from rat liver as previously described (13,17). Poly(ADP-ribose) glycohydrolase was partially purified from calf thymus by a modification of the method described by Miwa et al. (18); the specific activity was about 3000 units/mg.
Poly(ADP-ribosy1)ation of Nuclei or Synthetase-Nuclei (2 X 10') or poly(ADP-ribose) synthetase (60 ng) was incubated with ["PINAD at 15 "C in a solution (100 pl) containing 100 m~ Tris/HCI (pH 8.0), 10 mM MgClz, 1.25 mM dithiothreitol, and, for the case of synthetase, 20 p g / d of DNA. The concentration of NAD and the incubation period were specified in each experiment. Incubation was terminated by the addition of 20 p1 of 100% (w/v) CClsCOOH. An aliquot (10 p1) was removed from the mixture and assayed for 20% CCLCOOHinsoluble radioactivity.
Analysis of Poly(ADP-ribosy1)ated Products with Gel Electrophoresis-Another aliquot (10 pl) from the above incubation mixture was centrifuged at 17,000 X g for 10 min, and the precipitate, rinsed with ethanol, was suspended in a solubilizing buffer containing 1% sodium dodecyl sulfate (16). After standing at 25 "C for 3 h, a portion was electrophoresed in a 0.1% sodium dodecyl sulfate/l2.5% polyacrylamide slab gel as previously described (16). After electrophoresis, the gel was stained, dried, and autoradiographed (16), and, when specified, the autoradiogram was analyzed by scanning with a densitometer (Jookoo, Tokyo). Molecular weights were estimated by comparison with reference proteins, radioactive or nonradioactive, obtained from commercial sources and electrophoresed side by side.

RESULTS AND DISCUSSION
Incubation of isolated nuclei with 2. 110,000 is referred to as the "IlOK acceptor" in this paper. All these bands disappeared almost completely when the incubation was carried out in the presence of 3 mM 3-aminobenzamide, an inhibitor of poly(ADP-ribose) synthetase (20). The protein nature of aLl these acceptors was indicated by the disappearance of the bands upon incubation (37 "C, 10 min) of nuclei with proteinase K (5 pg/106 nuclei; pH 8.0). Upon longer incubations up to 10 min or at higher NAD concentrations up to 500 p~, the incorporation of ADP-ribose into 20% CC1,COOH-insoluble material increased almost linearly (data not shown). There were parallel increases in incorporation into the several protein acceptors (Fig. 1). Under these con-' Migration patterns of histones in sodium dodecyl sulfate gels have been found to be anomalous with respect to their molecular weights (19).  (Fig. lC), indicating that the diversity of the electrophoretic mobility was bestowed by poly(ADP-ribosy1)ation to varying extents of a single type acceptor. The product located at the top of the gel appeared to resist the glycohydrolase digestion, probably due to formation of aggregates too large to be solubilized. The same treatment with poly(ADP-ribose) glycohydrolase did not reduce the amounts nor the electrophoretic mobilities of other minor bands, an indication that the ADP-ribose residues attached to these acceptors existed mostly in monomeric or oligomeric forms. The molecular weight, llO,OOO, of the main acceptor was the same as that of poly(ADP-ribose) synthetase of rat liver (17). This coincidence, together with the fact that the synthetase catalyzes automodification (17,(21)(22)(23)(24), suggested the possibility that the llOK acceptor might be the enzyme. We investigated this possibility by comparing the electrophoretic behaviors of the products synthesized under various conditions. The autoradiograms of automodified poly(ADP-ribose) synthetase showed a molecular weight change that closely resembled that of the llOK acceptor described above (Fig. 2).
The view that the llOK acceptor was probably poly(ADPribose) synthetase itself was further supported by the following experiment. When a KC1 extract was made from nuclei incubated with [=PINAD and chromatographed on a hydroxyapatite column, the main peak of ADP-ribosyl protein co-eluted with poly(ADP-ribose) synthetase activity (Fig. 3). The peak fractions, as were isolated, gave a distinct band at   (Fig. 3, insets). These resulta were in accord with the view that the synthetase activity was associated with the llOK acceptor.
So far, the major acceptors of poly(ADP-ribose) in nuclei have been reported as histones with no reference to the fraction unextractable with dilute acid (9)(10)(11)(12)(13)(14). This was due, partly, to the apparent "insolubility" of poly(ADP-ribose) in the residual fraction upon electrophoresis, and, partly, to its marked heterogeneity in various chromatographic systems. In the light of our present results and our previous observation that as many as 15 molecules of the polymer composed of up to about 80 ADP-ribose units were attached to a single poly(ADP-ribose) synthetase molecule (24), these properties appear to be ascribable to auto-poly(ADP-ribosy1)ation to varying extents of poly(ADP-ribose) synthetase.
Jump et al. (25) reported that the major acceptors in isolated HeLa cell nuclei were a 125,000-dalton protein and histone H3. They assigned the former protein to the synthetase on the basis of molecular weight and co-purification through some steps. We recently demonstrated that the main acceptor in permeabilized lymphocytes was a protein, probably the synthetase, of M, = 110,000 prior to as well as subsequent to DNA damage (16). The present work showed that automodification of the synthetase took place also in isolated liver nuclei, and that it represented a main portion of poly(ADP-ribosy1)ation in this system.
Recently, Benjamin and Gill (26) distinguished two classes of poly(ADP-ribosy1)ation in cultured cells subjected to DNA breakage; one was dependent upon the introduction of strand breaks into DNA and the product was a polymer, while the other was independent of DNA breakage and the product was a monomer or short oligomers attached to a variety of proteins. Our present results suggesting the presence of two kinds of ADP-ribosylation in isolated nuclei, one on poly(ADPribose) synthetase and polymeric, and the other on histones or other proteins and mostly monomeric or oligomeric, are consistent with their observation, except that the polymer was not free but bound to the synthetase in our experiment.