INFORMATION TO USERS
This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality of the material submitted.
The following explanation o f techniques is provided to help clarify markings or notations which may appear on this reproduction.
1.The sign or “ target” for pages apparently lacking from the document photographed is “Missing Page(s)” . I f it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.
2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image of the page can be found in the adjacent frame. I f copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of “sectioning” the material has been followed. It is customary to begin filming at the upper left hand comer of a large sheet and to continue from left to right in equal sections with small overlaps. I f necessary, sectioning is continued again—beginning below the first row and continuing on until complete.
4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.
Uni International 300 N. Zeeb Road Ann Arbor, Ml 48106 8318416
Pantaleone, David Peter
STUDIES ON THE PURIFICATION, CHARACTERIZATION AND MECHANISM OF POLY(ADP-RIBOSE) POLYMERASE FROM CALF THYMUS
The Ohio Slate University Ph.D. 1983
University Microfilms International300 N. Zeeb Road, Ann Arbor, M I 48106 PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .
1. Glossy photographs or pages______
2. Colored illustrations, paper or print______
3. Photographs with dark background______
4. Illustrations are poor copy______
5. Pages with black marks, not original copy____
6. Print shows through as there is text on both sides of page______
7. Indistinct, broken or small print on several pages
8. Print exceeds margin requirements ______
9. Tightly bound copy with print lost in spine______
10. Computer printout pages with indistinct print______
11. Page(s)______lacking when material received, and not available from school or author.
12. Page(s)______seem to be missing in numbering only as text follows.
13. Two pages num bered______. Text follows.
14. Curling and wrinkled pages______
15. Other______
University Microfilms International STUDIES ON THE PURIFICATION, CHARACTERIZATION
AND MECHANISM OF POLY(ADP-RIBOSE)
POLYMERASE FROM CALF THYMUS
DISSERTATION
Presented in Partial Fulfillment of the Requirements
for the Degree Doctor of Philosophy in the
Graduate School of The Ohio State University
By
David P. Pantaleone, B.A. * * * *
The Ohio State University
1983
Reading Committee: Approved By
Robert M. Mayer Reynaldo C. Pless Ming-Daw Tsai Adviser (/ Department of Chemistry DEDICATION
To my wife, Jennifer
ii ACKNOWLEDGMENTS
I would like to express my gratitude to my adviser,
Dr. Robert M. Mayer, for his guidance throughout this research project and the preparation of this dissertation.
I also wish to thank the colleagues in my laboratory for their support throughout my endeavors, as well as my family for their encouragement.
I would like to thank Mr. Peter Pantaleone and Mr.
James Pantaleone for obtaining the fresh calf thymus; also
Mr. Herman Guiney for making, with Mr. Peter Pantaleone, the sieving device which aided in the preparation of the
DNA-agarose affinity column.
Finally, I wish to acknowledge the financial support given me by the Chemistry Department.
iii VITA
June 17, 1955 ...... Born - Joliet, Illinois
1977 ...... B.A., Lewis University Romeoville, Illinois
1977-1982 ...... Teaching Assistant Chemistry Department The Ohio State University Columbus, Ohio
19 82-1983 ...... Research Associate Chemistry Department The Ohio State University Columbus, Ohio
FIELD OF STUDY
Major Field: Biochemistry TABLE OF CONTENTS
Page
DEDICATION ...... ii
ACKNOWLEDGMENTS ...... iii
VITA ...... iv
LIST OF TABLES ...... ix
LIST OF FIGURES ...... X
Chapter
I. INTRODUCTION ...... 1
A. Enzyme Reaction ...... 2 1. Reactants a. Donor Substrate b. Acceptor Substrates 2. Product a. Structure of Poly(ADP-ribose) b. Poly(ADP-ribose) Degradation 3. Mechanism of Catalysis 4. Purification and Properties of Poly(ADP-ribose) Polymerase a. Poly(ADP-ribose) Synthesis In Vitro as Influenced by Histones
B. Physiological Role of Poly(ADP-ribose) ... 37 1. DNA Repair 2. Differentiation and Poly(ADP-ribose) 3. DNA Synthesis
C. Purpose of this Investigation ...... 48
v CONTENTS (continued)
Page
II. MATERIALS AND METHODS ...... 50
A. Materials ...... 50 1. Tissue 2. Enzymes and Proteins 3. Nucleotides and Related Compounds 4. Chromatography Materials a. Preparation and Assay of DNA- agarose 5. Filtration Membranes 6. Chemicals 7. Synthesis of ADP-ribonolactone
B. Methods ...... 65 1. Assay Procedure a. Millipore Filter Assay b. Paper Chromatography Assay 2. Radioactive Analysis a. Aqueous Samples b. Polyacrylamide Gel Slices 3. Chromatographic and Electrophoretic Methods a. Paper Chromatography b. Column Chromatography i. General Equilibration Techniques ii. General Monitoring c. Paper Electrophoresis d. Thin Layer Electrophoresis (TLE) e. SDS-Polyacrylamide Gel Electro phoresis i. Electrophoresis According to Weber and Osborn (0.1% SDS) ii. Electrophoresis According to Fairbanks et al. (1.0% SDS) f. Acid-Urea Polyacrylamide Gel Electro phoresis g. Isoelectric Focusing
VI CONTENTS (continued)
Page
4. Preparation of Poly(ADP-ribosyl)ated Polymerase a. SDS-PAGE of Poly(ADP-ribosyl)ated Material b. Average Chain Length Analysis 5. Protein Assay 6. General Determinations
III. RESULTS ...... 76
A. Purification of Poly(ADP-ribose) Polymerase from Calf Thymus ...... 77 1. Crude Extract 2. Ammonium Sulfate Fractionation 3. DNA-agarose Column Chromatography 4. Hydroxylapatite Column Chromatography 5. Blue Sepharose CL-6B Column Chromato graphy 6. Separation on the Basis of Molecular Weight a. Concentration on Hydroxylapatite b. Sephadex G-150 Column Chromatography 7. Summary
B. Homogeneity ...... 96 1. Polyacrylamide Gel Electrophoresis in the Presence of SDS 2. Isoelectric Focusing 3. Qualitative Product Evaluation
C. Properties of Poly(ADP-ribose) Polymerase ...... 114 1. Requirements for Maximal Activity 2. Storage of the Purified Enzyme 3. Stability of the Purified Enzyme 4. Molecular Weight Determination
vii CONTENTS (continued)
Page
a. SDS—PAGE According to Weber and Osborn b. SDS-PAGE According to Fairbanks et a l . Product Analysis
D. Mechanistic Studies 139 1. Isotope Exchange Reaction 2 . Binding Sites a. Substrate Site i. Kinetic Studies with Nico tinamide ii. Substrate Derivatives b. Acceptor Site i. DNA and Whole Histone in Equal Proportions ii. Whole Histone and Histone HI versus DNA Concentration iii. Effect of NAD+ Concentration and Histone HI on ADP-ribose Incorporation c. Automodification Site i. Effect of Acceptor Histone HI on the Automodification Reac tion. Average Chain Length A n a l y s i s .
IV. DISCUSSION 177
V. BIBLIOGRAPHY 193
viii LIST OF TABLES
Table Page
1. Tissues and Cells from which Poly(ADP-ribose) has been Purified...... 28
2. Properties of Poly(ADP-ribose) Polymerase Purified from Calf Thymus by Ito et a l . (68)...... 30
3. Electrophoretic Analysis of Nucleotides Along with ADP-ribonolactone...... 64
4. Purification of Poly(ADP-ribose) Polymerase 95
5. Requirements of Purified Enzyme for Poly (ADP-ribose) Synthesis...... 115
6. Stability of Poly(ADP-ribose) Polymerase Frozen in Liquid Nitrogen versus Bulk Frozen as a Function of Time at -70°C...... 122
7. Effect of Substrate Analogs on Poly (ADP-ribose) Synthesis...... 151
8. Effect of Kistone Hi on the Automodification Reaction...... 176
9. Properties of Poly(ADP-ribose) Polymerase from Calf Thymus Purified as Described in This Study ...... 179
ix LIST OF FIGURES
Figure Page
1. Structure of the Substrate, NAD+ ...... 5
2. A. Linear Structure of Poly(ADP-ribose) B. Repeating Unit of Poly (ADP-ribose) ...... 15
3. A. Branch Structure of Poly(ADP-ribose) B. Structure of the Branch Point of Poly(ADP-ribose) ...... 19
4. Enzymes that Degrade Poly(ADP-ribose) ...... 21
5. Proposed Involvement of Poly(ADP-ribose) Polymerase in DNA Repair ...... 43
6. A. Proposed Involvement of Poly(ADP-ribose) Polymerase in Cell Differentiation B. Poly(ADP-ribosyl)ation of Nuclear Proteins ... 46
7. Reaction Scheme for the Synthesis of ADP- ribonolactone ...... 57
8. First Bio-Gel P-2 Column ...... 59
9. Second Bio-Gel P-2 Column ...... 62
10. DNA-agarose Column Chromatography ...... 81
11. Hydroxylapatite Column Chromatography ...... 84
12. Blue Sepharose CL-6B Column Chromatography ...... 87
13. Hydroxylapatite Concentrating Column ...... 90
14. Sephadex G-150 Column Chromatography ...... 93
15. Purity of Poly(ADP-ribose) Polymerase as Shown by SDS-Polyacrylamide Gel Electrophoresis ...... 99
x FIGURES (CONTINUED)
Figure Page
16. Silver Staining of SDS-Polyacrylamide Gels in Figure 1 5 ...... 101
17. Isoelectric Focusing of Poly(ADP-ribose) Polymerase...... 104
18. Paper Chromatographic Analysis of the Polymerase Reaction Using Crude Extract...... 106
19. Paper Chromatographic Analysis of the Polymerase Reaction Using the (NH^^SO^ Fraction...... 108
20. Paper Chromatographic Analysis of the Polymerase Reaction Using the DNA-agarose Fraction...... Ill
21. Paper Chromatographic Analysis of the Polymerase Reaction Using Purified Poly(ADP-ribose) Polymerase...... 113
22. Paper Chromatographic Analysis of the Polymerase Reaction Using Purified Enzyme Minus DNA ...... 118
23. Paper Chromatographic Analysis of the Polymerase Reaction Using Purified Enzyme Minus Whole Histone...... 120
24. Protection of Enzyme Activity with DNA and Whole Histone...... 125
25. Heat Inactivation...... 128
26. Log Molecular Weight versus Rf ...... 131
27. Ferguson Plot. Log R^ versus Gel Percentage...... 133
28. Molecular Weight versus KR ...... 136
29. Log Molecular Weight versus Rf Using SDS-Polyacrylamide Gel Electrophoresis...... 138
30. Paper Chromatographic Analysis of Phospho diesterase Digestion Reaction Products...... 141
31. Paper Chromatographic Analysis of the Isotope Exchange Reaction...... 144
32. Inhibition by Nicotinamide...... 149
xi FIGURES (CONTINUED)
Figure Page
33. A. Protein Scan of Whole Histone B. Protein Scan of Histone HI ...... 155
34. ADP-ribose Incorporation with Differing Conditions of DNA and Whole Histone...... 158
35. ADP-ribose Incorporation with Whole Histone and Histone HI and Varying DNA Concentrations.... 161
36. Effect of NAD+ Concentration and Histone HI on ADP-ribose Incorporation...... 164
37. SDS-Polyacrylamide Gel Electrophoresis of Poly (ADP-ribosyl) ated Material...... 169
38. Effect of Histone HI on the Automodification Reaction...... 172
xii I. INTRODUCTION
Poly(adenosine diphosphate ribose)[poly(ADP-ribose)] synthetase or polymerase (E.C.2.4.99.-) activity was first reported nearly 20 years ago by Chambon, Weill, and Mandel
(1). Since that time much progress has been made in the investigation of this complex enzyme system. This is evi dent by the fact that several comprehensive reviews (2-10), proceedings of the last four international meetings (11-
14), as well as other brief accounts (15-17) have been pub lished. Such research areas as enzyme purification, poly-
(ADP-ribose) structure, and the possible physiological functions of this polymer, to name a few, have been exten sively investigated. Although no single biological func tion has been established for poly(ADP-ribose), it has been implicated to be involved in regulating such nuclear processes as DNA synthesis (18-22), DNA transcription (23,
24), DNA repair (25-31), and cellular differentiation and development (32-37) as well as others. These highly impor tant processes are central to nuclear metabolism, which
appears to be regulated by the enzyme poly(ADP-ribose)
polymerase.
1 2
A. Enzyme Reaction
The reaction catalyzed by poly(ADP-ribose) polymerase,
an enzyme which is tightly associated with chromatin and is
found in the nucleus of all eukaryotes, can be represented
by the following general equation:
nADPR-N+ + X -v (ADPR) - X + nN + nH+ (1) n
in which ADPR-N+ stands for adenosine diphosphoribosyl nico
tinamide [nicotinamide adenine dinucleotide(NAD+)] and X
stands for a macromolecular acceptor. These acceptors are
the nuclear proteins, histones, which are tightly associ
ated with DNA in the nucleus and have been shown to play an
important role in chromatin structure (38-40) . In addition,
other nuclear proteins such as high mobility group proteins
(HMG) (41), A24 protein (42) , and a Ca+2,Mg+2-dependent
endonuclease (43) serve as acceptors for poly(ADP-ribose)
polymerase. It has been proposed that such post-transla
tional modification of these nuclear proteins may be in
volved in the regulation of important nuclear processes.
1. Reactants
a. Donor Substrate
The structure of NAD+, the most abundant respiratory
coenzyme (5), which is the natural donor substrate for 3 poly(ADP-ribose) polymerase, is shown in Figure 1. Analocs of NAD+ have been tested in vitro for their donor substrate ability, resulting in limited success. For example, the
NAD+ analog in which the adenosine moiety has been replaced by tubercidin or 3^-deoxy NAD+ yielded average chain lengths of poly(ADP-ribose) of 1.7 and 1.4, respectively (44).
NAD+ usually functions in an oxidation-reduction capa city in conjunction with such enzymes as dehydrogenases or reductases. In these reactions, hydride transfer to the four position of the nicotinamide moiety takes place, resulting in a reduced form of the coenzyme, NADH. However,
NAD+ can also function in another way, taking into account the nature of its two high energy bonds, the pyrophosphate bond and the glycosidic linkage at the quaternary nitrogen of the pyridine ring. The reactions utilizing the high energy bonds can be viewed as transfer reactions, whereby the chemical potential of these bonds is the driving force of the reaction.
In the case of the pyrophosphate bond, the enzyme polynucleotide ligase from Escherichia coli catalyzes the transfer of AMP to the enzyme with the release of nicotina mide mononucleotide (NMN). This AMP-enzyme is an inter mediate form involved in ligation of 5 '-phosphoryl and 3'- hydroxyl termini of "nicked" double-stranded DNA (45). Figure 1 Structure of the Substrate for Poly(ADP-ribose) Polymerase, Nicotinamide Adenine Dinucleotide
(NAD+ )
4 NAD (D P N ) nh2
o0 o0 n H2C-0-P-0-P-0-CH2 ° ° c? 0 OH OH ADP~Ribosyl Nicotinamide
Figure 1 In the case of the N-glycosidic bond, the ADP-ribose residue is transferred to an acceptor with the release of nicotinamide and a proton. There are several enzymes that catalyze reactions of this type, as illustrated by the fol lowing equations:
ADPR-N + H20 -v ADPR + N + H+ (2)
ADPR-N + Base ADPR-Base + N (3)
ADPR-N + X -* ADPR-X + N + H+ (4)
n ADPR-N + X (ADPR) -X + nN + n H+ (5) n
Reaction 2 is catalyzed by the enzyme NAD+ glycohydrolase
(NADase) where serves as the acceptor to release free
ADP-ribose plus nicotinamide and a proton. Reaction 3 illustrates a transglycosidation reaction which is cata lyzed by certain types of NADases. In this reaction ana logs of NAD+ are formed through an exchange reaction with a variety of pyridine bases. Reaction 4 is catalyzed by the enzyme ADP-ribosyl transferase where X represents a macro- molecular acceptor and where only one ADP-ribose residue is attached to that acceptor. Both diphtheria and Pseudomonas aeroginosa toxins have been shown to mono ADP-ribosylate elongation factor 2, which results in the inhibition of protein synthesis (4). Other ADP-ribosyl transferases have also been studied (5). Reaction 5, as previously shown, is catalyzed by the enzyme poly(ADP-ribose) polymerase where X is a nuclear protein acceptor to which a homopolymer of
ADP-ribose is attached. This enzymatic reaction has been
found to occur in the nuclei of all types of eukaryotes (9)
including plants (46) and slime mold (47).
These reactions described above have the common
feature of N-glycosidic bond cleavage. The free energy of hydrolysis of this bond, reported to be -8.2 kcal/mol at pH 7 and 25°C, must provide sufficient energy to drive these reactions toward product formation (48).
b. Acceptor Substrates
As previously stated, the major protein acceptors for poly(ADP-ribose) polymerase are the nuclear proteins his- tones, however others have been reported (2-14). These highly conserved, basic proteins, along with DNA, form the
fundamental unit of eukaryotic chromatin called the nucleo- some (38-40) . The nucleosome core is composed of an octa- mer of histones complexed with approximately 200 base pairs of DNA. This octamer is composed of two copies of each of the slightly lysine-rich histones H2A and H2B and two copies each of the arginine-rich histones H3 and H4. His tone Hi is associated with the linker DNA, a O to 80 base pair segment of DNA which joins core nucleosomes. Since no enzyme activity has been found to be associated with his tones, their function besides structural involvement has not been clearly understood. Therefore post synthetic modifications such as poly(ADP-ribosyl)ation are of great interest since they may relate to non-structural functions of these nuclear proteins. They may also explain the bio logical role of poly(ADP-ribose) polymerase.
While many studies have been conducted to investigate the involvement of the polymerase in histone modification, only the major findings will be cited. Hayaishi and co workers (49) were first to observe that incubation of radio- labeled NAD+ with a nuclear preparation from rat liver resulted in an association of newly synthesized poly(ADP- ribose) with histones Hi, H2A, H2B and H3. Since then, investigators using other systems have reported similar results. Using He La cell nuclei, Giri et al. (41) found histones Hi and H2B to be the major acceptors of poly(ADP- ribose) while H2A and H3 were modified to a lesser extent.
In addition, minor modification of the nonhistone nuclear proteins (HMG, M1-M4) was also observed.
Other findings of nonhistone proteins serving as acceptors for poly(ADP-ribose) polymerase have been re- 2 2 ported. Koide and co-workers (43) found a Ca+ ,Mg+ - dependent endonuclease to be poly(ADP-ribosyl)ated and
Muller and Zahn (23) described the same type of modifica tion of RNA polymerase. 9
A unique nuclear protein, called A24, a complex of histone H2A and ubiquitin (a nonhistone protein) joined by an isopeptide linkage, resulting from an e-amino group of a lysine residue of A24 being joined to the carboxyl ter minus of ubiquitin, has also been shown by Okayama and
Hayaishi (42) to be ADP-ribosylated. Since the proposed function of A24 has been reported to be a repressor of ribosomal gene activity (50/51), this modification by ADP- ribosylation may be very important in regulating such func tion .
A unique observation has been made by Kidwell et al.
(52) in terms of the involvement of poly(ADP-ribosyl)ation with regard to higher orders of chromatin structure. Using
He La cell nuclei, they have observed a complex consisting of two molecules of histone Hi linked by a bridge of 15
ADP-ribose units, with one detectable covalent linkage of polymer to protein. Even though only one such linkage be tween protein and polymer has been detected, the noncova- lently bound molecule of Hi must be very tightly associated as it will not exchange with free Hi and does not dissoci ate in SDS. Thus speculation arises that poly(ADP-ribose) may be involved in chromatin condensation by crosslinking
Hi molecules (53).
The acceptor proteins described above all have the common feature of being covalently modified with poly(ADP- ribose) . The nature of this linkage has been established 10
on the basis of its properties and chemical characteristics,
and will be described briefly.
Early studies by Hayaishi and co-workers (54) indi
cated that the linkage between ADP-ribose and histone was
acid stable but unstable in neutral N ^ O H and mild alkali.
These results suggest an ester bond. Later studies (55)
revealed that the cleavage of ADP-ribosyl histone H2B by
alkali as well as by NI^OH involved a fast and a slow
phase. These results were also confirmed by Adamietz and
co-workers (56) using He La histone Hi. Despite the heter
ogeneity in the stability of the ADP-ribosyl histone link
age, Hayaishi and co-workers (57) identified the linkage
to histone H2B as an ester bond via glutamic acid residue
number two. They have also identified three modification
sites of rat liver histone Hi as being ester linkages (53).
Dixon et ad. (58) and Koide and co-workers (59,60) have
also reported that glutamic acid residues were ADP-ribo-
sylated in histones from trout testis and rat liver, res
pectively. On the other hand, Smith and Stocken (61) re ported that poly(ADP-ribose) was bound to rat liver histone
HI through a phosphoserine residue. This was concluded
from an experiment using [32P] NAD+ to poly(ADP-ribosyl)ate histone Hi in vivo followed by protease digestion and mild
acid hydrolysis, which yielded phosphoserine and ADP-ribose.
In addition to the acceptor proteins described above, which were reported for systems employing nuclei or 11
partially purified enzyme, another acceptor has been ob
served in experiments that utilized the purified poly
merase. In these studies it was found that the enzyme it
self can function as an acceptor. In other words, purified
poly(ADP-ribose) polymerase catalyzes an automodification
reaction in the presence of DNA and in the absence of added
acceptor.
This automodification reaction was first suggested by
Yoshihara et al. (62) for purified polymerase from bovine
thymus. Since then purified enzyme from rat liver (63),
calf thymus (64,65), and He La cells (66) has been shown
to catalyze an automodification reaction.
Along these lines, Jump et a_l. (67) have investigated
the level of chromatin structure required for the ADP-ribo-
sylation of chromatin material. They found that in small
oligonucleosomes (2-4 nucleosomes), the predominant ADP-
ribosylation reaction was the automodification of the poly merase, whereas in large oligonucleosomes (10-18 nucleo
somes) , automodification was reduced approximately 25% and histone ADP-ribosylation was up from 8% to 30%. These
results suggest that where folding tends to promote inter
action between the enzyme and the various histones (large oligonucleosome), the specific structural geometry is such
to provide a direct interaction between the polymerase and
its target proteins. 12
More recent studies by Ferro and Olivers (65) have shown the importance of buffer ionic strength on the auto modification reaction. Also, Hayaishi and co-workers (63) have shown that the automodified enzyme was less active than the unmodified polymerase; the Km value for NAD+ gradually increased and Vmax decreased as the modification proceeded. In addition, using pulse-chase experiments, it has been suggested that the poly(ADP-ribosyl) enzyme was not an intermediate in the poly(ADP-ribosyl)ation of other proteins (63) . Therefore if the automodified enzyme is not an intermediate in the modification of other acceptor pro teins, then it is possible for this type of modification to be a means of regulating the activity of the polymerase it self (68). This may be comparable with that of rabbit muscle cAMP-dependent protein kinase I, in which the enzyme, autophosphorylates its own catalytic subunit, and the phos phate residue is not transferred to other protein sub strates, including histone Hi (69).
The structure of the polymer which is attached to the acceptors described above will be discussed next.
2. Product
a. Structure of Poly(ADP-ribose)
Poly(ADP-ribose) is a unique homopolymer derived from the coenzyme NAD+ whose synthesis is catalyzed by 13
the enzyme poly(ADP-ribose) polymerase. The physical and
chemical properties of this polymer have been extensively
reviewed by Hilz and Stone (4). The ADP-ribose units are
linked by a ribose-ribose glycosidic bond and the polymer
is nearly always found linked to an acceptor protein
(Figure 2A).
The correct structure of the bond linking the repeat
unit was elucidated after the polymer was subjected to
snake venom phosphodiesterase digestion (see Figure 4).
This yielded AMP, a nucleotide composed of one molecule of
adenine and two molecules each of ribose and phosphate,
and one molecule of ribose 5-phosphate attached to the pro
tein acceptor (70-72). After treatment of this nucleotide with E. coli phosphatase, followed by methylation and sub
sequent acid hydrolysis, a structure shown in Figure 2B was proposed. This figure illustrates that the bonding be
tween repeat units involves a linkage between ribosyl resi
dues and that the bond of adjacent ADP-ribose units is be
tween the 1" and 2' carbons. The configuration of this
linkage was later determined to be an a(l" -> 2') linkage by Miwa et al. (73) using n m r . This a configuration was confirmed using 1H NMR by Inagaki et al. (74) and
Ferro and Oppenheimer (75).
Relatively few compounds are known to have ribose-
ribose bonds (7) and of these compounds, poly(ADP-ribose) Figure 2.
A. Linear Structure of Poly(ADP-ribose)
B. Repeating Unit of Poly(ADP-ribose)
14 t•— 1 unit— t 1 ADP- ribose
A.
Adenine - P - 0
OH OH OH
B. 0— P— | I 0 J n Figure 2 16 was the first shown to have an a anomeric configuration.
More recently the prosthetic group of citrate lyase from
Klebsiella aerogenes was also found to have an a (1" -* 2')
ribose-ribose bond (76).
The chain length of poly(ADP-ribose) has been reported
to vary from 1 to 50 monomers, depending on the experimental
conditions (5). Average chain length determination is based on the snake venom phosphodiesterase digestion products of the polymer, 5'-AMP which is derived from terminal units and isoADP-ribose molecules which are derived from inter nal positions. After quantitation by either paper or column chromatography, the average chain length can be estimated by taking the ratio of total ADP-ribose units to the number of terminal units. Since a branch point (see below) has been observed, the average chain length esti mates must be modified to take this into account (63).
Initial structural analyses indicated that poly(ADP- ribose) was a linear structure as shown in Figure 2A. Re cently, however, with the development of a method for the separation of different sized ADP-ribose polymers by poly acrylamide gel electrophoresis (77) and analysis of the average chain length, a discrepancy was realized. This was the fact that above molecular weights corresponding to 30 or more ADPR units, the average chain length of the polymer remained constant at approximately 22. This suggested that poly(ADP-ribose) may not always be linear but may have a 17 branched structure (Figure 3A). This idea was confirmed in studies which showed that in addition to the expected pro ducts from snake venom phosphodiesterase digestion, a new compound was found. It has been characterized as 2 ' [ 1 " ~ ribosyl-2 " ( 1 " ^-ribosyl) ] adenosine- 5 " , 5 " ,5 " "-tris (phos phate) by Miwa and co-workers (78). Its structure is shown in Figure 3B.
Although the function of the branching in the struc ture of poly(ADP-ribose) is not known, it is possible that the conformation of poly(ADP-ribose) may be stabilized by the presence of this branched structure. Recent studies on the conformation of long chain poly(ADP-ribose) by Minaga and Kun (79) have indicated a significant secondary struc ture. From their studies employing spectral analysis and circular dichroism analysis with highly purified poly(ADP- ribose) , they suggest a transition from single stranded to random coil occurs with increasing temperature and there fore for the first time identify poly(ADP-ribose) as a polymer with nucleic acid-like properties.
b. Poly(ADP-ribose) Degradation
Poly(ADP-ribose) has been shown to be degraded by as many as three different types of enzymes (Figure 4). These are poly(ADP-ribose) glycohydrolase, phosphodiesterase, and
ADP-ribosyl histone hydrolase. Figure 3.
A. Branch Structure of Poly(ADP-ribose)
B. Structure of the Branch Point of Poly(ADP-ribose)
18 19
Ade Ade I I Rib-Rib Rib i i i p p — p
— Rib Rib — Rib Rib - Rib Rib -
A.
III
H0-P-0CH2
Rl" R2
B.
Figure 3 Figure 4. Enzymes that Degrade Poly(ADP-ribose)
20 ADP-ribose glycohydrolase phosphodiesterase ADP-ribose histone hydrolase
Figure 4 22 Poly(ADP-ribose) glycohydrolase is an exoglycohydro-
lase which splits the ribose-ribose linkage to produce ADP-
ribose. This enzyme was first discovered by Miwa and Sugi-
mura (80) in calf thymus nuclei and was later shown to be
the major degradative enzyme for poly(ADP-ribose) by Miwa
et al. (81) . In addition, it has recently been shown to
hydrolyze the ribose-ribose bond involved in forming the branch portion of poly(ADP-ribose) (82).
As previously mentioned, snake venom phosphodiesterase
is widely used for the structural analysis of poly(ADP-
ribose) . Other phosphodiesterases, one from rat liver nuclei (83) and one from rat liver mitochondria (84) have been shown to hydrolyze poly(ADP-ribose) internally. The phosphodiesterase from rat liver nuclei has been shown to hydrolyze poly(ADP-ribose) proceeding from the AMP terminus
(85) and yields isoADP-ribose as the major product.
The enzyme ADP-ribosyl histone hydrolase has been partially purified from rat liver nuclei by Okayama and co-workers (86). This enzyme removes the ADP-ribose from modified histones. The most effective substrate was mono-
(ADP-ribosyl) histone H2B followed by mono(ADP-ribosyl) histone HI with modified nonhistone protein being least effective (15).
It has been suggested by Ueda et al. (15) that since both biosynthesis and the principal degradation of 23 poly(ADP-ribose) takes place at ribose-ribose or ribose- protein linkages and nowhere else, frequent modification
and degradation may be possible in vivo.
3. Mechanism of Catalysis
Reactions in which the high energy N-glycosidic bond of NAD+ are cleaved can be thought of as a type of group transfer reaction. Either the ADP-ribose moiety is trans
ferred to H 2O, a protein, or a protein-bound ADP-ribose residue in reactions catalyzed by the enzyme NAD+ glyco- hydrolase, ADP-ribosyl transferase, or poly(ADP-ribose) polymerase, respectively. The cleavage of the N-glycoside can proceed according to several pathways and the stereo chemical outcome can be either retention or inversion of configuration. The generalized pathways will be outlined below, followed by a discussion of how these pathways relate to the possible reaction mechanism of poly(ADP- ribose) polymerase.
As proposed by Koshland (87,88), enzymes catalyzing group transfer reactions of the type
BX + Y -> BY + X where BX is the donor substrate and Y is the acceptor sub strate can proceed by two different reaction mechanisms. 24
The direct displacement mechanism is characterized by a single SN2 nucleophilic substitution in which a group on the acceptor substrate (Y) attacks the donor substrate (BX) from the backside. The stereochemical outcome of such a reaction results in an overall inversion of configuration.
The double displacement mechanism is characterized by two SN2 nucleophilic substitutions in which the enzyme directly participates through the formation of a covalent enzyme intermediate. This occurs by a nucleophilic group on the enzyme attacking the donor substrate (BX) from the backside to form a covalent enzyme intermediate with the release of X. This is followed by the acceptor substrate
(Y) attacking the intermediate from the backside with the release of free enzyme and the formation of the product
(BY). This can be shown by the following equations:
E + BX -*■ E-B + X
E-B + Y + BY + E
The stereochemical outcome of this type of reaction, being two, SN2 inversions results with overall retention of con figuration. Enzymes which employ such covalent enzyme intermediates have been listed by Bell and Koshland (89) and more recently by Spector (90).
Another type of mechanism that has been proposed in volves the formation of an oxocarbonium ion in the 25 transition state (91-9 3). As shown in the scheme for the hydrolysis of glycosides © H H H © <3 0-R '0-R H OH OH
h 2° H H OH
OH oxocarbonium ion the first step involves the protonation of the departing alcohol, ROH (94). Next a glycosyl C-l carbonium ion is generated after the loss of ROH which is stabilized by the lone pair of electrons on the ring oxygen. This oxocar bonium ion is probably the major form contributing to the resonance hybrid. The last step results in an attack by
H20 to yield the hemiacetal form of the product. Since 2 the oxocarbonium ion is planar (sp hybridized), the attacking H20 molecule can enter from either side and therefore the stereochemical outcome may be retention or inversion of configuration.
This type of mechanism involving the formation of a 2 planar sp hybridized intermediate at C-l has been probed 26
using substrate analogs which are also planar at C-l. Such
compounds as 1,5-gluconolactone and the 6-lactone from
tetra-N-acetylchitotetraose have been employed with the
enzymes glycogen phosphorylase (95) and lysozyme (96),
respectively. Both have been shown to be very powerful
inhibitors of their respective reactions and, as noted by
Wolfenden (97,98), may be classified as transition state
analogs. These types of analogs have been successfully
used in studying enzyme mechanisms as well as in the field
of drug design (99).
Since poly(ADP-ribose) polymerase catalyzes three
types of reactions, initiation, elongation and branching,
using a number of acceptor proteins as well as itself being
an acceptor and requires DNA as an activator, the reaction mechanism is probably very complex. At present not much is
known in terms of the mechanism, although attempts have
been made to simplify this complicated system (65). On
the other hand, since the reaction involves the N-glyco-
sidic bond cleavage of 3-NAD+ and the formation of an
a (1" ->■ 2') ribose-ribose bond, this inversion of configu
ration is consistent with either a single displacement (or
an odd number of displacements) or a carbonium ion mecha nism. Further investigation must be conducted in order
to establish which is the case as well as sort out the
details of this complicated reaction. 27
4. Purification and Properties of Poly(ADP-ribose)
Polymerase
Before any progress could be made in terms of eluci dating the biological function of the polymerase or ana lyzing its reaction mechanism, it was essential to employ purified enzyme. During the past seven years, a number of purification procedures have been published from a variety of tissues and cultured cell systems, as shown in Table I.
Since the enzyme is tightly associated with chromatin, solubilization and separation from the bulk of DNA has been a major concern. Most purification procedures achieve solubilization by using high salt concentrations. Separa tion from DNA has been accomplished using ultracentrifu gation (102), protamine sulfate (65,68), or hydroxylapatite chromatography (100,107). The enzyme is like other nuclear enzymes (8), since it is highly unstable. In order to pre serve enzyme activity, such stabilizing agents as thiol reagents, glycerol, and high salt concentrations, as well as protease inhibitors such as sodium bisulfite or phenyl- methylsulfonyl fluoride, have been used (68,106). Tso- panakis et al. (103,104) have also employed subzero tem perature chromatography with an aqueous organic solvent, ethylene glycol, to stabilize the polymerase during the purification. 28
Table I
Tissues and Cells from which Poly(ADP-ribose)
Polymerase has been Purified
Source Reference
Bovine thymus (adult) Yoshihara et al. (100)
Ferro and Olivera (65)
Bovine thymus (calf) Niedergang et al. (101,102)
Ito et al. (68)
Pig thymus Tsopanakis et al. (103,104)
Lamb thymus Petzold et al. (105)
Human tonsils Carter and Berger (106)
Rat liver Okayama et al. (107)
Bull testes Farina et al. (108)
Mouse testes Agemori et al. (109)
Ehrlich ascites cells Holtlund et al. (110)
Kristensen and Holtlund (111)
He La cells Jump and Smulson (66)
Leukemia cells Ellison (112) 29
The purification procedures listed in Table I gener ally employ separation techniques based on affinity, ion exchange, and gel filtration chromatography. Specific examples of effective chromatography media include DNA- agarose (109,111), hydroxylapatite (105-107), and Sephadex
G-150 (65,68) column chromatography.
Purified poly(ADP-ribose) polymerase has been charac terized by some general properties, as shown in Table 2.
In addition, the purified enzyme is inhibited by nicotin amide. These properties are very similar for all the puri fied poly(ADP-ribose) polymerases with the following major exceptions.
The molecular weight for pig thymus (103,104) and rat liver (107) polymerase has been reported to be 63,000 and
50,000 daltons, respectively. More recently, Holtland et al. (113) reported pig thymus enzyme to have a molecular weight of 112,000 daltons and the molecular weight of rat liver enzyme was revised to 110,000 daltons by Kawaishi et al . (114). A new method for determining the molecular weight of poly(ADP-ribose) polymerase in nuclear homogenates from different tissues using a protein blotting technique has confirmed the similarities in molecular weights re ported above (115).
Another major exception to the purified calf thymus poly(ADP-ribose) polymerase (68) was the isoelectric point Table 2
Properties of Poly(ADP-ribose) Polymerase
Purified from Calf Thymus by Ito et al. (6 8)
Property Value
Molecular Weight 120,000
Isoelectric point (pi) 9.8 pH optimum 8.0 - 8.5
Km for NAD+ 55 pM . -1 -1 V 1430-2200 nmol mm mg max for calf thymus polymerase purified by Niedergang et al.
(101/102). It was reported to have a pi of 6.5, consi derably lower than 9.8 as reported by Ito et ad. (68).
This has been suggested to be the result of residual DNA in the enzyme preparation (101,102). In agreement with this suggestion is the fact that adding DNA to purified polymerase leads to a drastic lowering of the apparent pi of the enzyme (116).
Purified poly(ADP-ribose) polymerase exhibits an absolute requirement for DNA. This requirement is speci fic and could not be replaced by other polyanions such as heparin, poly(vinylsulfate), or by other polynucleotides such as RNA (9,68). DNA which separates from bovine thy mus polymerase on hydroxylapatite column chromatography during the purification has been shown to be 20 times as effective in activating the enzyme as the bulk of calf thymus DNA (117) and has been termed "active DNA." This
"active DNA" has also been found to be associated with the calf thymus polymerase (101) and has been estimated to have an average size of about 100-200 base pairs using electrophoresis on polyacrylamide-agarose gels. It has been suggested that its unique activating ability is due to a DNA sequence(s), concentrated on the DNA, and that the enzyme has a high affinity for these sites (117). Fragmented DNA stimulates poly(ADP-ribose) synthesis more than native DNA. Benjamin and Gill (118) have shown, using a crude cell-free extract of calf thymus, that with plasmid DNA in a covalently closed form, little poly(ADP- ribose) synthesis is observed. However, using restriction enzymes to induce strand breakage, poly(ADP-ribose) syn thesis is stimulated in proportion to the frequency of strand breaks. These results support the idea of poly(ADP- ribose) being somehow involved in DNA repair (see physio logical role). It would be of interest to determine whether or not purified poly(ADP-ribose) polymerase would yield similar results.
a. Poly(ADP-ribose) Synthesis in Vitro as
Influenced by Histones
As previously discussed, histones serve as acceptor proteins for poly(ADP-ribose) polymerase. This was first established using crude enzyme systems. With the recent availability of purified enzyme, studies have also been conducted to see what effect histones might have on the purified polymerase. Results of these studies have been somewhat conflicting, probably owing to variations in the conditions of the enzyme reaction. Such conditions as enzyme concentration, histone concentration, and the pre sence or absence of NaCl and Mg+2 in the buffer have been confirmed as being extremely important factors in the ADP- ribosylation of histones in vitro with purified poly(ADP- ribose) polymerase (114,119-122). The major findings regarding the modification of histones by purified poly-
(ADP-ribose) polymerase will be presented below.
In the initial report of the purification of poly(ADP- ribose) polymerase from rat liver by Okayama et ad. (107), evidence was given for histones being allosteric activators rather than acceptors of the polymerase. Similar results were obtained for purified polymerase from bovine thymus and He La cells by Yoshihara et ad. (100) and Jump and
Smulson (66), respectively.
In more recent studies using bovine thymus polymerase,
Tanaka et _al. (119) have observed two types of enzyme reac tions, one which is Mg+2_dependent and the other histone- dependent. In the absence of Mg+2, histone HI is ADP- ribosylated and when analyzed by acid-urea polyacrylamide gel electrophoresis, 4 0% of the product corresponds to HI and the remaining 60% is randomly distributed on the gel between the origin and the position of HI. Also, both the • o Mg -dependent and histone-dependent reactions have been shown to proceed using activated DNA-cellulose as an im mobilized solid support (123).
Other studies by Ferro and Olivera (65) using bovine thymus polymerase have shown histone HI to be an acceptor 34
+ 2 in the presence or absence of Mg . These investigators have also examined the poly(ADP-ribosyl)ation of the en zyme and histone HI in the presence and absence of high +2 salt and Mg in the buffer and found that the rate and extent of the poly(ADP-ribosyl)ation reaction changed.
They believe this to result from the enzyme becoming more negatively charged by poly(ADP-ribosyl)ation, which would repel the negatively charged DNA required for enzyme acti vation and therefore a shielding of the charges would exist at higher ionic strength. This "repulsion point" would be changed in the presence of histone HI and therefore poly-
(ADP-ribosyl)ation of HI would be dependent on how many HI molecules could be modified before this "repulsion point" is reached. In addition, steric requirements must also be considered when interpreting these results. Although the authors conclude that this is probably a much more simpli fied picture than what exists in vivo, they feel that it is a prerequisite for rationalizing the role of the enzyme in more complex systems (65).
Subsequent work using He La cell polymerase (67) has shown that in vitro almost 30% of the total protein-bound
ADP-ribose was associated with histones when activity on polynucleosomes of sufficient 'size were analyzed. These studies have suggested a strict steric requirement for effective histone modification. Similar studies with calf thymus polymerase containing endogenous DNA (10% w/w) by Okazaki et al. (120,121) have shown that all histone fractions become ADP-ribosylated.
After this endogenous DNA ("active DNA") was removed via hydroxylapatite column chromatography; however, only his tone HI was modified. This raises some very important questions as to the nature of this "active DNA" and its interaction with the polymerase.
In contrast to the findings of Okayama et al. (107) for purified rat liver poly(ADP-ribose) polymerase, more recent studies (114,124) have indicated that histone HI could serve as an acceptor for the polymerase reaction.
In addition, evidence was provided which indicates that the polymerase catalyzes both the initiation and elonga tion of the polymer. Chain elongation was demonstrated using a chemically synthesized (ADP-ribose)•histone adduct as described by Kun et al. (125), which was radiolabeled in the adenine moiety. Upon incubation of this labeled adduct with the polymerase and unlabeled NAD+ followed by product analysis with snake venom phosphodiesterase, the production of labeled isoADP-ribose was observed. This clearly indicates that elongation of the synthetic adduct occurred on the ADP-ribose residue chemically preattached to histone and the elongation proceeded via a terminal addition mechanism, i.e. the ADP-ribose units being trans ferred to free adenosine termini. Kawaichi et al. (114)/ using purified polymerase from
rat liver, examined initiation of polymer formation and
showed histone HI to be an acceptor. A reaction was car
ried out by incubating purified polymerase with labeled
NAD+ in the presence of higher concentrations of histone
HI than had been previously used (50-100 yg/ml). Upon
extraction of the product with 5% HCIO^ and analysis on
SDS-polyacrylamide gels, the label comigrated with protein
(HI). This indicates that Hi is an acceptor for purified
rat liver polymerase. In looking at initiation, poly(ADP-
ribosyl) histone HI was prepared from [ribose (NMN)-1‘♦C,
adenine-3H] NAD+ purified, rat liver enzyme. After re
leasing the polymer from HI with mild alkali, product ana
lysis using snake venom phosphodiesterase digestion af
forded 3 compounds: [llfC] ribose-5-phosphate, [ 3H] AMP, and
[1 ,3H] isoADP-ribose. Since ribose-5-phosphate was
detected, which was derived from the end of the polymer
attached to the histone, initiation of polymer formation must have occurred. By examining the ratio of [lttC] ribose-
5-phosphate to [3H] AMP (0.78), the authors conclude that
the majority of the polymer was synthesized in direct link
age to histone HI. They also speculate the remaining 22%
may represent polymers elongated from preexisting ADP-
ribose or perhaps a branched structure as previously de
scribed (78,122). 37
The in vitro systems developed employing purified enzyme (114/119-122) appear to reproduce the modification reactions using intact nuclei. Evidence has now accumu lated to suggest that a single enzyme can catalyze three distinct reactions involved in poly(ADP-ribose) synthesis:
(1) initiation, the transfer of ADP-ribose to an acceptor protein (114,119-121); (2) chain elongation, addition of
ADP-ribose units to preexisting mono- or oligo(ADP-ribose)
(114,119-121,124); and (3) branching of the linear poly-
(ADP-ribose) chain (122). Further studies with purified poly(ADP-ribose) polymerase may provide insight in terms of understanding the role this enzyme plays in modifying chromatin structure.
B . Physiological Role of Poly(ADP-ribose)
As previously mentioned, nearly 20 years have passed since the discovery of poly(ADP-ribose) polymerase and yet its biological function still remains unclear. Although implicated in a number of nuclear functions such as chroma tin structure, DNA repair and synthesis, and cellular dif ferentiation, it has been very difficult if not impossible to correlate enzyme activity from isolated nuclei with a particular stage of the cell cycle. This problem may be somewhat overcome with the development of permeabilized-cell systems (20) which may provide a more realistic model than 38 isolated nuclei (6). Because so many nuclear functions appear to be affected by poly(ADP-ribose) polymerase or its reaction product, poly(ADP-ribose), only the major ones, DNA repair, cellular differentiation, and DNA syn thesis will be presented below. Information regarding other processes affected can be found in the following reviews (2-10).
1. DNA Repair
Many investigations have been conducted in an effort to establish some correlation between poly(ADP-ribose) polymerase activity and the DNA repair process. Since poly(ADP-ribose) synthesis has been shown to increase in response to DNA strand breaks (25-27,30,31), studies em ploying UV irradiation and DNA-damaging agents have been conducted. In conjunction with these studies, specific inhibitors of poly(ADP-ribose) polymerase have also aided the understanding of the polymerase's involvement in DNA repair. Since most of the investigations of the biologi cal role of poly(ADP-ribose) polymerase are centered around DNA repair, some of the major studies are presented below.
Studies by Berger et ail. (26) using permeabilized normal human lymphocytes indicated a close correlation between the stimulation of poly(ADP-ribose) activity and 39 the repair mode of DNA synthesis. In response to such
DNA-damaging agents as bleomycin, N-methyl-N'-nitro-N- nitrosoguanidine (MNNG) or N-acetoxy-acetyl aminofluorene and UV light, an abrupt increase in unscheduled DNA syn thesis along with a 3 to 4 fold stimulation in poly(ADP- ribose) synthesis was seen. By density labeling the DNA, it was found that the unscheduled synthesis was a result of the repair mode of DNA synthesis. Also, using the pro tein- synthesis inhibitor cycloheximide prior to the treat ment with the various drugs or immediately after exposure to UV indicated that the increase in poly(ADP-ribose) syn thesis is not dependent upon new protein synthesis. This suggests that the polymerase is inactive until required.
Durkacz et al. (27) have demonstrated an even closer correlation between poly(ADP-ribose) synthesis and DNA excision repair. By treating mouse L1210 leukemic lympho blast cells with dimethyl sulfate (known to cause DNA strand breaks), cellular NAD+ and DNA levels are lowered with an increase in poly(ADP-ribose) polymerase activity.
With time, NAD+ levels and poly(ADP-ribose) synthesis return to control levels and the DNA damage is repaired.
However, when polymerase inhibitors such as nicotinamide or 3-aminobenzamide are included along with dimethyl sul fate, NAD+ content doesn't drop and the damaged DNA is not + repaired. In addition, when cellular NAD levels were 40 lowered by nutritional deprivation, and L1210 cells were then treated with dimethyl sulfate, the DNA became frag mented and strand joining was inhibited. These results show that the biosynthesis of poly(ADP-ribose) is required for efficient excision repair and survival following damage with monofunctional agents.
Recent studies by Berger et al. (29) have employed long term lymphocyte cell lines from normal donors and from patients with xeroderma pigmentosum (XP). These cells are known to be defective in their ability to repair UV- induced DNA damage (126); however, they will repair some types of DNA damage such as that caused by MNNG (127).
When normal donor lymphocyte cells were treated with MNNG or UV irradiated, poly(ADP-ribose) synthesis was shown to increase. On the other hand, XP cell lines showed an in crease in poly(ADP-ribose) synthesis when treated with
MNNG, but they were defective in their poly(ADP-ribose) synthesis in response to UV irradiation. In order to de termine whether or not UV irradiation destroyed the XP cells' ability to synthesize poly(ADP-ribose) in response to other types of DNA damage after UV irradiation, XP cells were then permeabilized and treated with DNase I. These results showed a normal increase in their ability to syn thesize poly(ADP-ribose), thus supporting the fact that XP 41 cells truly fail to demonstrate an increase in poly(ADP- ribose synthesis in response to UV irradiation.
In summary, there appears to be a good correlation between poly(ADP-ribose) polymerase activity and the DNA repair process. This has been depicted by Berger et al.
(28) as shown in Figure 5, however the mechanism by which poly(ADP-ribose) participates in this process is not fully understood.
2. Differentiation and Polv(ADP-ribose)
The fact that cellular NAD+ levels may be involved with cellular differentiation was recognized by Caplan and co-workers (128) who have been responsible for much of the work in this area. In studies that followed, poly(ADP- ribose) was also implicated in cellular differentiation.
Using mesodermal cells of embryonic chick limbs which have the capacity to differentiate into either muscle or carti lage, Caplan and Rosenberg (32) were the first to correlate the net rate of poly(ADP-ribose) synthesis with the dif ferentiation of chondrogenic cells from mesodermal cells.
This expression of the chondrogenic phenotype as well as poly(ADP-ribose) synthesis could be inhibited by treatment of the cells with nicotinamide or bromodeoxyuridine. By treating cells "with 3-acetylpyridine, a stimulation of poly(ADP-ribose) synthesis was observed along with the Figure 5: Proposed Involvement of Poly(ADP-ribose)
Polymerase in DNA Repair by Berger et al. (28)
42 DNA ______STRAND ALTERATION RESTITUTION CHROMATIN ^ EXCISION DAMAGE BREAKS REPAIR CHROMATIN STRUCTURE STRUCTURE I
Poly(ADP- + ribose) NAD ------Poly(ADP- glycohydro1ase Polymerase ribose) ► ADP-ribose
Figure 5 potentiation of chondrogenic expression. It is believed that these studies indicate that changes in the intra cellular NAD+ levels are 11 sensed" by the chromatin-associ ated poly(ADP-ribose) synthesizing machinery and result in differential rates of synthesis of this polymer. This is in turn correlated with the process of differentiation (see
Figure 6A). More recent studies by Caplan and co-workers
(33) demonstrated that poly(ADP-ribose) levels change, whether analyzed in situ or in cell culture, in chick limb mesenchymal cells undergoing developmental changes. During early developmental periods, this change is characterized by a three-fold decrease in poly(ADP-ribose) per unit DNA; however during the events involved in muscle formation, the amount of poly(ADP-ribose) was relatively unchanged.
These observations were the first to indicate that the absolute levels of poly(ADP-ribose) do not play a major role in later expressional events (muscle formation) but may be involved with the early commitment events in the cellular development of chick limb mesenchymal cells.
Other investigators have employed systems such as
Xenopus laevis embryos (37), erythroleukemic mouse spleen cells (36), human blood cells (15), and SV40 transformed cells (35). In agreement with the work of Caplan, these studies have suggested the levels of poly(ADP-ribose) pre sent to be important in the process of cellular differen tiation. Figure 6.
A. Proposed Involvement of Poly(ADP-ribose)
Polymerase in Cell Differentiation by Caplan (34)
B. Poly(ADP-ribosyl)ation of Nuclear Proteins by
Sucrimura et al. (7)
45 46
MUSCLE CARTILAGE
A.
Nucleosome H2B. A 24
Poly (ADP-Rib) HMG
B.
Figure 6 47
3. DNA Synthesis
It was first postulated by Burzio and Koide (18) in
1970 that poly(ADP-ribose) may be involved in DNA synthesis.
This was the result of the observation that when isolated rat liver nuclei were treated with NAD+ , the incorporation of [3H]-labeled deoxyribonucleotides into acid-insoluble material markedly decreased. This decrease was suggested to be due to poly(ADP-ribose) formation. Since that time many other investigators have reported the involvement of poly(ADP-ribose) with DNA synthesis for different types of cells and cell systems, however the results have been some what conflicting (see reviews 6,8,9).
Using permeabilized mouse L cells, Berger et al. (20,
21) have observed that under conditions which decrease DNA synthesis, for example acute glucose deficiency, infection with vaccinia virus, or treatment with cytosine arabinoside, an associated increase in intrinsic poly(ADP-ribose) syn thesis is seen. These studies demonstrate for mouse L cells that suppression of DNA synthesis by a number of dif ferent physiological mechanisms is always associated with an increase in the intrinsic activity of poly(ADP-ribose) synthesis.
When normal and chronic lymphocytic leukemia (CLL) lymphocytes were examined by Berger et al. (22) and Burzio et al. (19) using a permeabilized cell system and isolated 48 nuclei, respectively, similar results were obtained. That is, when compared to normal lymphocytes, CLL cells had higher initial activities of poly(ADP-ribose) synthesis.
These high levels of poly(ADP-ribose) polymerase activity of CLL cells can be added to a number of other disorders of enzyme activities that have been reported for this disease
(22) .
These studies cited exemplify the studies dealing with the relationship between DNA synthesis and poly(ADP-ribose).
The individual studies cited in this section on the physi ological role of poly(ADP-ribose) illustrate the major cur rent thoughts in the areas of DNA repair, cellular differen tiation and DNA synthesis in regard to poly(ADP-ribose).
More details on the physiological role of this polymer can be obtained in the following reviews (2-10). As shown in
Figure 6B, poly(ADP-ribose) polymerase modifies a number of nuclear proteins. It is possible that the physiological function of poly(ADP-ribose) polymerase may be to regulate the structure of chromatin and thus control a number of nuclear functions as those described above.
C. Purpose of Investigation
The enzyme poly(ADP-ribose) polymerase appeared to be similar to another polymerizing enzyme, dextransucrase, already being studied in our laboratory, and it would be of interest to compare these two enzymes in terms of their mechanistic properties. However, before such studies could be performed, it was necessary to obtain pure enzyme, which
is required for mechanistic studies of enzymes. Therefore part of the focus of this work has been the development of
a reproducible enzyme purification procedure followed by
characterization of the purified material. This characteri
zation, based on general properties such as product analy
sis, molecular weight, isoelectric point, etc. was very
important to ensure that the purified enzyme was similar to
that reported in the literature.
Subsequent studies have been focused on various mecha
nistic aspects of the polymerase reaction. This has in
cluded experiments involving the substrate, acceptor, and
automodification binding sites. The nature of the inter
action between the acceptor site and the automodification
site has also been addressed. II. MATERIALS AND METHODS
A. Materials
1. Tissue
Calf thymuses were obtained from Swissland Packing Co.
(Ashkum, IL). Immediately after slaughter, thymuses were trimmed of fat, placed in zip-loc plastic freezer bags
(about 250 g per bag) and quick frozen by placing on dry ice. The tissues were stored at -70°C.
2. Enzymes and Proteins
Commercial enzyme preparations were purchased from these sources: deoxyribonuclease I, DNase (Type I from bo vine pancreas), phosphodiesterase I, SVPDE (Type II from
Crotalus adamanteus venom) and (Type VII from Crotalus atrox v e n o m ) , 8-galactosidase (Grade VIII from Escherichia coli), a-chymotrypsin A (Type II from bovine pancreas) from Sigma
Chemical Company (St. Louis, MO); trypsin (from hog pancreas) from ICN Pharmaceuticals, Inc. (Cleveland, OH); and RNA polymerase (from Escherichia coli) from Boehringer Mannheim
Biochemicals (Indianapolis, IN).
Commercial protein preparations were purchased from these sources: myosin, ovalbumin, bovine serum albumin
(BSA), crosslinked BSA, calf thymus histones: whole histones
50 51
(type IIA) and HI (type Ills) from Sigma; trypsin inhibitor
(from soybean) from Boehringer Mannheim.
3. Nucleotides and Related Compounds
Nucleotides and related compounds were obtained from
the following sources: deoxyribonucleic acid, DNA (Type I:
sodium salt from calf thymus, "highly polymerized"), 3-
nicotinamide adenine dinucleotide, NAD+ (Grade V from yeast),
3-nicotinamide adenine dinucleotide, reduced form, NADH
(Grade III: disodium salt from yeast), nicotinamide, adeno
sine 5"-monophosphate, AMP (Type II: monosodium salt from yeast), adenosine 5"-diphosphate, ADP (Grade I: disodium
salt from equine muscle), and adenosine 5"-diphosphoribose,
ADPR (monosodium salt) from Sigma Chemical Company.
Radioactive pyridine nucleotides were obtained as fol
lows: [adenine-2,8- 3H] nicotinamide adenine dinucleotide
(26.1 Ci/mmol) from New England Nuclear (Boston, MA); [ade- n ine-U-14C] nicotinamide adenine dinucleotide (265 mCi/mmol)
and [carbonyl-14C] nicotinamide (53 mCi/mmol) from Amersham
Corporation (Arlington Heights, IL).
4. Chromatography Materials
Chromatography materials were obtained from the fol
lowing sources: agarose from Miles Laboratories, Inc.
(Elkhart, IN); hydroxylapatite (Bio-Gel HTP), Bio-Gel P-2 52
(100-200 mesh) and the cation exchange resin (AG 50W-X8,
50-100 mesh, hydrogen form) from Bio Rad Laboratories
(Richmond, CA); Blue Sepharose CL-6B and Sephadex G-150
from Pharmacia Fine Chemicals (Piscataway, NJ). Whatman
1MM and 3MM chromatography paper from Whatman, Inc. (Clif
ton, NJ); plastic-backed thin layer plates, Silica Gel
60 F 254 with fluorescent indicator from E. Merck (Darm
stadt, Germany).
a. Preparation and Assay of DNA-aqarose
DNA-agarose was prepared according to Schaller et al.
(129) and Bendich and Boulton (130) as follows: Calf thymus
DNA (1.5 g) was dissolved in 0.02 M NaOH (100 ml). This
solution was heated to 100°C in a boiling water bath for
15 minutes to separate the polynucleotide strands and then mixed with an equal volume of an 8% agarose solution in
water. This dark yellow mixture was stirred with a magnetic
stirrer with continued heating at 100°C for approximately 10
minutes, at which time small amounts were removed and cooled
in an ice cold glass dish. The solidified gel was cut into 2 pieces about 1 cm using a razor blade and passed through a
60-mesh brass sieve. This was accomplished using a homemade
device consisting of a caulking gun over the end of which the
sieve was fitted, forcing the gel through under pressure.
Two passes through the sieve were sufficient to produce very
uniform particles. The DNA-agarose particles were then 53
suspended in 0.01 M Tris-HCl (pH 7.5), 1 mM EDTA, 0.1 M
NaCl, packed in a column and washed extensively at room
temperature until no DNA could be detected (measurement at
260 nm) in the wash.
In order to determine the DNA content of the gel, the
DNase digestion procedure of Umansky et al_. (131) was fol
lowed. The extensively washed DNA containing gel was packed
into a small column (2 ml) and one column volume of 0.01 M
Tris-HCl (pH 7.5), 0.01 M MgCl 2 buffer containing 50 yg/ml
of DNase I was percolated through followed by closing the
column outlet. After incubation for one hour at 24°C, the
column was washed with 0.01 M Tris-HCl (pH 7.5) and the
nucleotide content of the eluate was determined by measure
ment of absorbance at 260 nm. An absorbance coefficient of
25 at 26 0 nm was used for 1 mg/ml solution treated with
DNase (129). The final preparation contained 1.4 mg DNA/ml
settled bed volume and was stable for over one year.
5. Filtration Membranes
Filtration membranes (25 mm in diameter, pore size
0.45 ym) were obtained from Millipore Corporation (Bedford,
MA). Spectra/Por I dialysis membrane tubing was obtained
from Spectrum Medical Industries, Inc. (Los Angeles, CA).
6 . Chemicals
Reaqent grade chemicals were obtained from the fol
lowing sources: isobutyric acid and magnesium chloride from Aldrich Chemical Co. (Milwaukee, WI): pyronin Y from
Allied Chemical Corp. (Morristown, NJ); bromophenol blue
(BPB) from Beckman Instruments, Inc. (Palo Alto, CA); Bio-
Lyte 3/10 from Bio Rad; hydroxylamine hydrochloride from
Eastman Kodak Co. (Rochester, NY); 30% hydrogen peroxide, and sodium thiosulfate from Fisher Scientific Co. (Fair
Lawn, NJ); disodium dihydrogen ethylene diaminetetraacetate dihydrate (EDTA) and sodium meta periodate from G. Frederick
Smith Chemical Co. (Columbus, OH); ammonium persulfate, iodine, lithium carbonate, ferric chloride, phosphorus pen- toxide, and sodium phosphate from Mallinckrodt, Inc. (Paris,
KY); ammonium sulfate, glycerol, trichloroacetic acid (TCA) and urea from Matheson, Coleman, and Bell (Cincinnati, OH);
Insta-Gel from Packard Instrument Co. (Downers Grove, IL); acrylamide, N,N'-methylene-bis-acrylamide, bis-p-nitrophenyl phosphate, Coomasie Brilliant Blue G, D-ribonic acid-y- lactone, dithiothreitol (DTT), mercaptoethanol, l,4-bis-[2-
(5-phenyloxazolyl)]-benzene (POPOP), 2,5-diphenyloxazole
(PPO), sodium dodecyl sulfate (SDS), p-tosyl-L-arginine methyl ester (TAME), N,N,N",N'-tetramethylethylene diamine
(TEMED), and tris (hydroxymethyl) aminomethane, Trizma Base
(Tris) from Sigma Chemical Co. (St. Louis, MO). All other chemicals used in this study were of reagent or spectro scopic grade and were purchased from commercial sources. 55 7. Synthesis of ADP-ribonolactone
ADP-ribonolactone was synthesized via alkaline iodine
oxidation of ADP-ribose according to the method of Schuber
and Pascal (132) as shown in Figure 7. ADP-ribose (15 ymol)
was dissolved in 3.4 ml of a solution containing 0.01 M I2,
0.5 M Nal, 0.05 M Li 2CC>2 and stirred in the dark at 4°C.
Aliquots (20 pi) were removed, acidified, and the residual
iodine titrated with a standardized sodium thiosulfate solu
tion. The reaction was judged complete after approximately
12 hours with the reduction of one equivalent of iodine per
equivalent of ADP-ribose. The reaction was terminated by
adjusting the pH to 4 with 5 N H 2SO^ and then extracting the
excess iodine with benzene (4 x 10 ml). After lyophiliza-
tion, most of the iodide salts were extracted with acetone
at -20°C. The residue was redissolved in 1.0 ml H20 and
applied to a-Bio-Gel.P-2 column (1.5 x 59 cm) which was
developed in the descending mode (0.17 ml/min) using H20 as
the solvent (see Figure 8). Fractions 49 - 78 were pooled
(9.1 pmol by A^q) and when analyzed for reducing sugar
gave negative results. This is consistent with there being
no unreacted ADP-ribose.
Before applying the oxidized ADP-ribose to a Dowex AG 50W-
X 8 column, the Dowex was bleached as described by Dunaway-
Mariano and Cleland (133) as follows: the hydrogen form of
Dowex AG 50W-X8 was converted to the sodium form by exten
sively washing with 1 N NaOH. Then a solution of 1% Br2 in Figure 7: Reaction Scheme for the Synthesis of
ADP-Ribonolactone
Shown is the synthetic scheme according to Schuber and Pascal (132) for the preparation of ADP-ribonolactone.
56 Synthesis of ADP-Rlbonolactone
NHa NH2 NH2
N,S r (P (p H2C-O P 0-p-0-CH2CH2 H2C0f-0-p*0*CH2 a DOWEX O 0 I 2 OH HCr 00 /SnndJ OH H+ form N,/ 'O OH OH OH OH OH OH
Figure 7 Figure 8 : First Bio-Gel P-2 Column
After quenching the iodine oxidation described in
Figure 7 with acid, the excess iodine was extracted with benzene and then the aqueous fraction was lyo-
philized. After lyophilization, the residue was
extracted with acetone at -20°C to remove most of the
iodine salts and then was redissolved in 1.0 ml of water and applied to a Bio-Gel P-2 column (1.5 x 59
cm) equilibrated with water. Elution was with water
and fractions were analyzed for nucleotide by the
absorbance at 260 (•-----• ) or for salts by conducti
vity (o o) . Fractions 49 -78 were pooled accounting
for 9.1 ymol by A 26q.
58 8.0
7.0
6.0
5.0
4.0
3.0 30
2.0 20
0 20 40 60 80 100 120 FRACTION No.
Figure 8 60
1 N NaOH was added and stirred with the resin until the dark yellow color bleaches to light yellow (about 15 minutes).
After decanting the Br ^ solution and washing with water until neutral, the hydrogen form was regenerated by washing extensively with 1 N HC1 followed by water until neutral.
The resin was packed in a column and equilibrated with 10 bed volumes of starting buffer before applying any sample.
The oxidized ADP-ribose was applied to a Dowex AG
50W-X8 column (hydrogen form) (1.1 x 15 cm) equilibrated with water. Fractions 4-17 were pooled and immediately lyophilized. The residue was dissolved in 1.1 ml water
(5.2 ymol by A^gg) and applied to a second Bio-Gel P-2 column as before (see Figure 9). This step separates any
AMP which may be formed from ADP-ribonic acid during pas sage over the Dowex column. Fractions 34 - 50 were pooled, lyophilized and placed in a vacuum dessicator over P 2°5 ^or three days. The white solid (2.1 ymol by A 260' overall) was dissolved in 1.0 ml of water and consisted of a mixture of ADP-ribonolactone (44%) as measured by the lactone assay described in Methods, and ADP-ribonic acid
(56%) taken as the difference between the total ^ g Q and the amount of lactone measured. The sample was stored at
-70°C.
The following criteria were used to confirm the pre sence of ADP-ribonolactone: 1) A reducing sugar assay using the Nelson arsenomolybdate method (134) gave negative results. Figure 9 : Second Bio-Gel P-2 Column
The lyophilized peak from the Dowex column was dissolved
in 1.1 ml water (5.2 ymol by and applied to a Bio-
Gel P-2 column (1.5 x 59 cm) equilibrated with water.
Elution was with water and fractions were analyzed for
nucleotide by the absorbance at 260 nm. Fractions 34 -
50 were pooled and after lyophilization and storage in
a vacuum dessicator for three days afforded 2.1 ymol by
A 2 6 0 ’
61 62
A260 7.0
60
50
40
30
20
0 20 40 60 80 FRACTION No.
Figure 9 This indicates that all the ADP-ribose has been oxidized at the anomeric carbon. 2) The product gives a positive reac tion with hydroxylamine-ferric chloride (135), indicating the presence of a lactone. Of the 2.1 ymol by A 2gQ re covered, 0.9 4 ymol (44%) are lactone positive. 3) The ultraviolet spectrum was characteristic of an unmodified adenosine derivative (A 257 - 258 nm, e = 15,400 M -1 max cm , pH 7.0); therefore the C-8 of the adenine ring is not modified. 4) Using paper electrophoresis in 0.05 M sodium acetate, pH 5.0 at 2000 V for 2.5 hours, two products were detected by ultraviolet light from the oxidation reaction of ADP-ribose. ADP-ribonic acid was identified after base treatment of the mixture as the faster moving compound.
Also, thin layer electrophoresis in 0.05 M sodium citrate, pH 3.5 at 500 V for two hours was employed to analyze the mixture. The mobilities of standards and sample on paper and thin layer electrophoresis are listed in Table 3.
This mixture was then used with purified poly(ADP- ribose) polymerase to assess its inhibitory characteristics
(discussed under Binding sites). TABLE 3
Electrophoretic Analysis of Nucleotides
Along with ADP-Ribonolactone
cm from the origin Compound toward the anode
Paper TLE
AMP 10.5 0.35
ADP 18.2 1.45
ADP-Ribose 15.9 1.15
ADP-Ribonolactone 16.5 1.30
ADP-Ribonic acid 19.1 1.85 65
B. Methods
1. Assay Procedure
a. Millipore Filter Assay
Poly(ADP-ribose) polymerase was routinely assayed using the commonly employed Millipore filtration method analogous to that described by Shizuta et a^. (136). Each reaction contained 100 mM Tris-HCl (pH 8.0), 10 mM MgCl2,
1 mM dithiothreitol (DTT), 10 yg calf thymus DNA, 10 yg calf thymus whole histone, 0.2 mM [adenine-2,8- 3H] (10,000 dpm/ nmol) or [adenine-U-1^C] NAD+ (5,500 dpm/nmol) and varying amounts of enzyme in a total volume of 0.1 ml. After incu bating for 2 minutes at 25°C, 2 ml of ice-cold 10% tri chloroacetic acid (TCA) was added to precipitate the pro tein. The precipitate was collected on a Millipore filter, washed with 5% TCA (2 x 5 ml), 95% ethanol (1 x 2 ml) and then dried in an oven for 10 minutes at 65°C.
Each dried filter was placed in a vial containing 10 ml of scintillation fluid (15.2 g PPO and 380 mg POPOP per gal lon of spectral grade toluene) and the radioactivity quan titated using a Packard Tri-Carb 460 CD liquid scintillation spectrometer. A dpm standard curve was prepared by counting standards on Millipore filters in the presence of increasing amounts of acetone as a quencher. 66 One unit of enzyme activity is defined as being equiva lent to one nmol ADP-ribose incorporated per minute into acid-insoluble material and calculated as follows:
ADPM = Total DPM - Background DPM
- (Control DPM - Background DPM)
Activity (units) =
ADPM x specific Radioactivity (DPM/nmol) x time (min)
b. Paper Chromatography Assay
The product formed by poly(ADP-ribose) polymerase was qualitatively analyzed using descending paper chromatography.
This assay consisted of the same reaction mixture as de scribed for the Millipore filter assay in a total volume of
0.05 ml. After appropriate times, the reaction mixture was spotted on Whatman 1MM or Whatman 3MM filter paper and developed in solvent system I, 95% ethanol: 1 M ammonium acetate, pH 7.0 (5/2 v/v). After the solvent reached the bottom of the paper, the chromatograms were air-dried, cut
into 1 cm strips and analyzed for radioactivity. Each strip was placed in a scintillation vial and quantitated analogously to the Millipore filters as previously described.
A dpm standard curve was similarly prepared for Whatman 1MM and Whatman 3MM as described for Millipore filters. Non- radioactive, UV absorbing standards were visualized under
UV light (254 n m ) .
2. Radioactive Analyses
a. Aqueous Samples
Aqueous samples were counted as a gel using Insta-Gel as the scintillation cocktail. Three ml of 1^0 were added to the sample followed by 5 ml of Insta-Gel. The mixture was immediately vortexed to give an opaque solution which gelled upon cooling in the scintillation counter.
b. Polyacrylamide Gel Slices
Polyacrylamide gel slices were prepared for scintilla tion counting as follows: Each slice was incubated with
0.5 ml of 30% H 2O 2 at 93°C until a clear solution resulted
(approximately 1 hour). After cooling and transferring to a scintillation vial with 2.7 ml of H 2O, 5 ml of Insta-Gel were added. After vigorous vortexing and then cooling, the
samples were counted.
3. Chromatographic and Electrophoretic Methods
a. Paper Chromatography
Descending paper chromatography was carried out using
Whatman 3MM filter paper with the following solvent systems solvent system I, 95% ethanol: 1 M ammonium acetate, pH 7.0
(5/2 v/v); solvent system II, isobutyric acid: concentrated
NH 4OH: H 2O (66/1/33 v/v); and solvent system ill, n-butanol glacial acetic acid: H 2O (50/25/25 v/v). Radioactive ana lysis was carried out as previously described.
b. Column Chromatography
i. General Equilibrium Monitoring
All columns throughout the enzyme purification were determined to be equilibrated when the pH, conductivity, and absorbance at 280 nm of the eluate was equal to that of the equilibration buffer. Depending on the chromatographic media, it took up to 10 bed volumes for equilibration.
ii. General Monitoring
Columns were monitored for protein by the absorbance at
280 nm and where too dilute by the method of Bradford (137).
Enzyme activity using the Millipore filter assay was rou tinely used throughout the enzyme purification as previously described. Nucleotides were followed by the absorbance at
260 n m .
c. Paper Electrophoresis
High voltage paper electrophoresis was carried out using a Shandon model L-24 water-cooled flat bed apparatus along with a Shandon 5000 V/200 mA power supply. Samples 69 were spotted on Whatman 3MM paper near the center and electrophoresed at 2000 V for 2.5 hours. Compounds were visualized with a UV light. Solvent system was the fol lowing: 0.05 M sodium acetate (pH 5.0).
d. Thin Layer Electrophoresis (TLE)
Thin layer electrophoresis was carried out using the same apparatus described for paper electrophoresis. Plas tic-backed silica thin layer plates with fluorescent indi cator were used (6.5 x 20 cm) employing Whatman 3MM as filter paper wicks. After applying the sample 5 cm from the cathode end and allowing to dry, each end of the plate was immersed in electrolyte solution for about 30 seconds, being careful not to wet the sample. Electrolyte was ap plied to the sample by placing an electrolyte-soaked piece of Whatman 3MM filter paper over it and applying even pres sure. After blotting excess buffer with filter paper, electrophoresis was carried out at 500 V for 2 hours. Com pounds were visualized with a UV light. The following buf fer system was used: 0.05 M sodium citrate (pH 6.5).
e. SDS-Polvacrvlamide Gel Electrophoresis
i. Electrophoresis According to Weber and
Osborn (0.1 SDS)
Gel electrophoresis was carried out in the presence of 0.1% SDS using the method of Weber and Osborn (13 8) with 70 the following modifications: Protein samples were precipi tated with an equal volume of ice cold 24% TCA at 0°C for 20 minutes. After centrifuging and removing the supernatant fluid, the precipitated protein was washed with acetone
(2 x 1.0 ml) to remove the TCA, centrifuged again and the acetone supernatant fluid removed using an aspirator. The precipitate was dissolved in 8 yl of a 2% SDS w/v, 2% mercap- toethanol, 8 M urea solution plus 7 yl electrode buffer [0.1
M sodium phosphate (pH 7.2) 0.1% SDS] and 1 y 1 of 0.05% bro- mophenol blue. Samples were dissociated by heating at 50°C for 5 minutes and were applied to the gel under tray buffer using a 25 y 1 syringe.
Electrophoresis was carried out at a constant current of
8 mA/gel for 0.5 x 6.0 cm gels or 2.5 mA/gel for 0.3 x 6.0 cm gels. Migration of the tracking dye was followed and the run halted when it was 1 cm from the bottom of the gel; usually
3.5 to 4 hours.
After removing the gel from the tube with a thin wire, it was sliced at the dye front and either stained for protein according to Weber and Osborn (138) or frozen in dry ice, sliced and counted for radioactivity.
ii. Electrophoresis According to Fairbanks
(1.0% SDS)
Polyacrylamide gel electrophoresis in 1.0% SDS was carried out according to Fairbanks, Steck and Wallach (139) with the following modifications: Protein samples were 71 precipitated by the method previously stated for 0.1% SDS gels and the precipitate was dissolved in reduction solu tion where 5% glycerol was used in place of 10% sucrose.
After dissociating at 50°C for 10 minutes, the sample was applied under electrode buffer using a 10 yl syringe.
Electrophoresis was carried out at a constant current of 4.6 mA/gel for 0.5 x 7.5 cm gels. Running time was 1.75 to 2 hours as judged by the tracking dye. After removing the gels from the tubes as before and slicing at the dye front, each gel was placed in 200 ml of 25% isopropyl alco- hol-10% acetic acid and shaken overnight on a rotary shaker.
Staining was by the Coomassie blue-perchloric acid method of Reisner et al. (140) and destaining was with 7.5% acetic acid. In addition, after staining with Coomassie blue, gels could be stained using the method of Wray et al. (141).
This sensitive silver staining procedure was modified by using 40% methanol instead of 50%.
f. Acid-Urea Polyacrylamide Gel Electrophoresis
Acid-urea polyacrylamide gel electrophoresis was per formed according to Panyim and Chalkley (142) with the fol lowing modifications: Polyacrylamide gels of 7.5% (0.5 x
7.5 cm) and 6.25 M urea were used. Samples of 20 yl were applied to each gel containing 2 M urea instead of 15% sucrose. Gels were not pre-electrophoresed and were run at 2.5 mA/gel for 1.5 hour using methyl green as the 72 tracking dye. After slicing at the dye front, gels were stained according to Bonner ejt al. (143) in 0.1% Coomassie blue G, 40% ethanol, 5% acetic acid for about 8 hours and passively destained in 20% ethanol, 5% acetic acid.
g. Isoelectric Focusing
Isoelectric focusing in polyacrylamide gels was car ried out similarly to Righetti and Drysdale (144). The gels, 0.3 x 10 cm contained the final concentrations of the following: 5% acrylamide, 0.133% N,N'-methylene bis- acrylamide, 2% (w/v) Biolyte 3/10, 5% glycerol, 0.062%
N,N,N^,N'-Tetramethylethylenediamine (TEMED) and 0.125% ammonium persulfate. The electrode solutions contained
0.01 M H 3PO 4 as the anode and 0.02 M NaOH as the cathode, with the cathode uppermost.
Salt-free protein samples were mixed with an equal volume of 4% (w/v) Biolyte 3/10 that was adjusted to pH 10 with 1.0 M NaOH. Before underlaying the samples, the gels were pre-electrophoresed for 15 minutes at 1 mA/gel.
Focusing was carried out at 4°C at a constant current of
1 mA/gel for 12 hours.
Gels stained for protein were first treated with 200 ml of 10% TCA at 37°C to remove ampholytes before staining with Coomassie blue-perchloric acid (140). The pH gradient was measured by sectioning the gel (0.5 cm/slice), incu bating the slices for 45 minutes in 0.5 ml of 0.1 M NaCl 73
(pH 7.0) at 37°C and then reading on an Orion model 701 pH meter at room temperature.
4 . Preparation of Poly(ADP-ribosyl)ated Polymerase
Poly(ADP-ribose) polymerase was poly(ADP-ribosyl)- ated similar to the method of Kawaichi, Ueda, and Hayaishi
(63). The reaction was carried out in a polyethylene cen trifuge tube containing 100 mM Tris-HCl (pH 8.0), 10 mM
MgCl2, 1 mM DTT, 100 yM [adenine-U-1 ] NAD+ (10 dpm/nmol),
50 yg/ml calf thymus DNA and enzyme in a total volume of
1 ml. The reaction was carried out for 30 seconds at 25°C and terminated by adding 1 ml of ice cold 25% TCA. After removing an aliquot for a Millipore filter assay, the remaining solution was centrifuged at 15,000 x g for 20 minutes. The supernatant fluid was removed and the pellet was washed with ice cold 10% TCA (2 x 1.0 ml) and then with
1 ml of diethyl ether. Variations in the above procedure were carried out depending on the type of analysis of the poly(ADP-ribosyl)ated material and will be described in the Results section.
a. SDS-PAGE of Poly(ADP-ribosyl)ated Material
The precipitated poly(ADP-ribosyl)ated material was dissolved in 20 yl of 1.5 mM sodium phosphate (pH 7.0)-3%
SDS and incubated at 25°C for 2.5 hours. After adding
20 yl of 2% SDS, 2% mercaptoethanol, 8 M urea and 1.0 yl of 0.05% bromophenol blue, the sample was electrophoresed according to Weber and Osborn (138) on 7.5% polyacrylamide gels as previously described. Determination of radioacti vity in polyacrylamide gels has also been previously described.
b. Average Chain Length Analysis
The average chain length of poly(ADP-ribose) was determined essentially by the method of Yamada and Sugimura
(145) with modifications for branching by Kawaichi, Ueda and
Hayaishi (63). The precipitated poly(ADP-ribosyl)ated material was treated with 100 yl of 1 N NH 4OH for 30 m i nutes at 25°C to release poly(ADP-ribose) from acceptors.
After lyophilization, 40 yl of phosphodiesterase (0.06 units) in 50 mM Tris-HCl (pH 8.0), 10 mM M gC^ and 2.5 mM
AMP were added and the reaction was incubated at 37°C for
6 hours. Trypsin was sometimes included in this digestion
if the material contained a large amount of protein.
The products were analyzed by paper chromatography using solvent system II and the radioactivity determined as previously described. Since the phosphodiesterase used contained 5 '-nucleotidase activity, the average chain
length equation was modified as follows: 75
Average Chain Length =
dpm (AMP) + dpm (Adenosine) + dpm (isoADP-ribose)
[dpm (AMP) + dpm (Adenosine)]
+ dpm (phosphoribosyl-isoADP-ribose)
- dpm (phosphoribosyl - isoADP-ribose)
5- Protein Assay
Protein was assayed by either the method of Bradford
(137) or the method of Bensadoun and Weinstein (146) using
bovine serum albumin as a standard.
6 . General Determinations
The concentration of adenine containing nucleotides
(e.g. AMP, ADP, ADPR etc.) and NAD was determined by the
ultraviolet absorption at 259 nm (pH 7.0) using extinction
coefficients of 15,400 M -1 cm-1 and 18,000 M -1 cm"^, res
pectively (14 7).
Total phosphate was determined using the modified
Chen method (148) with adenosine monophosphate as a standard.
Lactone was determined by the hydroxylamine assay
described by Couling and Goodey (135) using ribonolactone as
a standard.
Reducing sugar was determined by the arsenomolybdate method of Nelson (134) using glucose as a standard. III. RESULTS
The goal of the research was to explore issues related to the mechanism of poly(ADP-ribose) polymerase, and the sites of interaction with substates and effectors. How ever, before such studies could be carried out, a purifi cation scheme had to be developed. Although purification procedures had been published (68,100-103), these were irreproducible and yielded enzyme of low purity. There fore initial efforts were focused on the development of a reproducible purification- procedure. A modification of the schemes of Ito et al. (68) and Yoshihara et al. (100) will be discussed here. In addition, experiments designed to characterize the purified enzyme in terms of such gene ral properties as molecular weight, isoelectric point, kinetic constants, and others will be presented, which are very important for comparison of purified enzyme by this procedure to that reported in the literature from other procedures. In a subsequent section, experiments will be described which probe the NAD+ binding site, the acceptor binding site, and the automodification or polymer site.
Also, experiments designed to look at the effect of the acceptor histone HI on the automodification reaction will be presented. 77
A. Purification of Poly(ADP-ribose) Polymerase from Calf
Thymus
Calf thymus was chosen as the source of enzyme because
of its availability. The general steps in the purification
procedure involve ammonium sulfate precipitation and affi
nity, ion exchange and gel filtration chromatography. The
developed procedure takes a week to 10 days and yields en
zyme in sufficient quantity and quality to employ in fur
ther experiments. Each step in the purification will be
discussed in detail below. All steps were carried out at
4°C and are summarized in Table 4.
1. Crude Extract
Calf thymus, stored at -70°C, was broken into pieces while still frozen using a hammer. Following passage
through a meat grinder, it was mixed with 2 1 of 50 mM
Tris-HCl, 0.3 M NaCl, 10 mM mercaptoethanol, 50 mM sodium
bisulfite and 10% glycerol adjusted to pH 8.0 at 4°C (200 g
tissue/1 buffer). The mixture was blended in a Waring
blender for 30 seconds on the highest setting, and centri
fuged in a Sorval RC-2B centrifuge with a GS-3 rotor at
11,000 x g for 20 minutes. The supernatant fluid was fil
tered through a glass wool plug and designated the crude
extract. The specific activity of the crude extract was
about one unit/mg protein. 78
2. Ammonium Sulfate Fractionation
Ammonium sulfate precipitation was used to concentrate the polymerase as well as remove the bulk of non-enzyme protein. This was accomplished as follows; Solid
(NH^)2S°4 was added to the gently stirring crude extract over an hour time period to 40% saturation. After conti nuously stirring overnight/ the solution was centrifuged
at 12,000 x g for 20 minutes in a GS-3 rotor. The super
natant fluid was carefully decanted, filtered through a glass wool plug and then brought to 85% saturation with
respect to (NH^)2S° 4* After stirring overnight and cen
trifuging as before, the supernatant fluid was decanted and the red pellet resuspended in approximately 250 ml of 50 mM
Tris-HCl, 0.2 M NaCl, 10 mM mercaptoethanol, 10% glycerol
and 0.02% NaN3; the solution was adjusted to pH 8.0 at 4°C
(Buffer A). Dialysis against Buffer A served to remove
the (NH^) 2S04 an<^ t^ie resulting solution was designated the
ammonium sulfate fraction. The recovery was 68% and the
specific activity was 2.9 units/mg.
3. DNA-agarose Column Chromatography
Since the polymerase requires DNA for activity, the
use of a DNA-agarose affinity column provides an effective
purification step (68,109,111). The ammonium sulfate 79
fraction was applied to a DNA-agarose column (5.0 x 12 cm)
(preparation previously described), equilibrated with Buf
fer A. The column was washed with Buffer A modified to a total NaCl concentration of 0.47 M, until the absorbance 4 at 280 nm was below 0.10. The polymerase was then eluted with Buffer A containing a total NaCl concentration of
1.0 M (Buffer B) and designated the DNA-agarose fraction.
The column could be regenerated by washing with approxi mately 3 column volumes of 50 mM Tris-HCl (pH 8.0) con
taining 10% glycerol and 1.2 M NaCl. At this point it was
ready for equilibration.
As shown in Figure 10, a large amount of contaminating
protein was removed in the first wash with very little en
zyme. The recovery of enzyme activity was usually greater
than 100%, possibly owing to the removal of competing en
zymes or of an inhibitor of the polymerase. The specific
activity was 173 units/mg.
4. Hydroxylapatite Column Chromatography
The DNA-agarose fraction had a significant absorbance
at 260 nm and did not require added DNA for activity. It was concluded that endogenous DNA was present. In order to
remove this endogenous DNA from the polymerase, a hydroxyl
apatite column in high salt was used (100,101). The DNA-
agarose fraction was applied directly to a hydroxylapatite Figure 10. DNA-agarose Column Chromatography
The ammonium sulfate fraction of 303 ml (10,744 u) was applied to a DNA-agarose column (5.0 x 12 cm) equilibrated with Buffer A. After washing with approximately 1.5 1 of
Buffer A containing a final NaCl concentration of 0.47 M, the polymerase was eluted with Buffer B. Fractions 115 to
150 were pooled. Protein was monitored by absorbance at
280 nm (----- ) and enzyme activity by the Millipore fil tration assay ( ------) .
80 DNA-AGAROSE 0.47 1.0 U/mL M.NoCI (....) 120 25 20 100
80
60 1.00
0.75 40 0.50 20 0.25
0.00 _L 0 0 20 40 60 80 100 FRACTION
Figure 10 column (2.5 x 5.0 cm) equilibrated with Buffer B. After washing with approximately 1.5 column volumes of Buffer B containing 1 M KC1 followed by 1.5 column volumes of Buf fer B containing 2 M KCl, the enzyme was eluted with a 400 ml linear gradient of potassium phosphate (0 to 50 mM) in
Buffer B containing 2 M KCl. This was followed with 100 ml of Buffer B containing 2 M KCl and 0.50 M potassium phos phate. The enzyme was pooled and extensively dialyzed against 50 mM Tris-HCl, 0.05 M NaCl, 10 mM mercaptoethanol,
10% glycerol and 0.02% N a N 3 adjusted to pH 8.0 and was de signated the hydroxylapatite fraction. Regeneration of this column was accomplished by washing with approximately
3 column volumes of 1.5 M potassium phosphate (pH 7.2).
As shown in Figure 11, the polymerase eluted at ap proximately 20 mM phosphate along with the major protein peak. The second peak of A2gg eluting with the final wash has been shown to be DNA (100). The recovery of enzyme activity was 54% and the specific activity was 188 units/mg.
The polymerase at this stage of purification required added
DNA for activity. In view of this dependency, it would appear that the endogenous DNA had been removed.
5. Blue Sepharose CL-6B Column Chromatography
The dye Cibacron blue covalently linked to Sepharose has been employed quite successfully in the purification Figure 11. Hydroxylapatite Column Chromatography
The DNA-agarose pool of 475 ml (13,316 u) was applied to a hydroxylapatite column (2.5 x 6.0 cm) equilibrated with
Buffer B. After washing the column with 45 ml of Buffer B containing 1 M KCl followed by 45 ml of Buffer B containing
2 M KCl, the polymerase was eluted with a 400 ml linear gradient of potassium phosphate (0 to 50 mM) in Buffer B containing 2 M KCl. Fractions 190 to 237 were pooled and dialyzed as previously described. Protein was monitored by absorbance at 280 nm (------) and enzyme activity by the Millipore filtration assay (----- ) .
83 HYDROXYLAPATITE U/ml 280 2.0 M KCl 0.5 M ®1.0 (—-) 30
I.0 0
0.80 20
0.60 - 15
0.40 mM 0.20 - 5 50
0.00 20 40 140 I60 I80 200 220 240 260 FRACTION No.
Figure 11 85 of many proteins including kinases, dehydrogenases, and most enzymes requiring adenylyl-containing substances (e.g.
NAD+) (149). It is believed that the anthraquinone and the
terminal phenylsulfonate rings of the dye can be oriented
as the adenine and the nicotinamide of NAD+ , respectively,
and thus serve as an affinity support for NAD+ requiring
enzymes (150). Since NAD+ is the substrate for poly(ADP-
ribose) polymerase, it appeared that the use of Blue Sepha
rose would be an appropriate step to employ for purifica
tion purposes (110).
The hydroxylapatite fraction was applied to a Blue
Sepharose CL- 6B column (1.5 x 17 cm) equilibrated with Buf
fer A. After washing with Buffer A until the absorbance
at 280 nm was constant, the polymerase was eluted with Buf
fer A containing a final NaCl concentration of 0.4 M. A
final wash was with Buffer B. The appropriate enzyme frac
tions were pooled and designated the Blue Sepharose frac
tions I, II, III and IV. The column was regenerated by washing with 10 column volumes each of 0.1 M Tris-HCl (pH
8.5) containing 0.5 M NaCl and 0.1 M sodium acetate (pH
4.5) containing 0.5 M NaCl as suggested by Pharmacia.
As shown in Figure 12, the polymerase elutes with the
major protein peak at 0.4 M NaCl. The total recovery of
enzyme activity was 89% and the specific activity of Pool
II (used in the next step) was 49 3 units/mg. Figure 12. Blue Sepharose CL-6B Column Chromatography
The hydroxylapatite fraction of 270 ml (7,136 u) was applied to a Blue Sepharose CL- 6B column (1.5 x 17 cm) equilibrated with Buffer A. After washing the column with 70 ml of Buf fer A, the polymerase was eluted with Buffer A containing a final NaCl concentration of 0.4 M. A final wash was with
Buffer B. The enzyme activity was pooled as designated in the figure into four fractions I, II, III and IV. Protein was monitored by absorbance at 280 nm (------) and enzyme activity by the Millipore filtration assay (------).
86 BLUE SEPHAROSE CL-6B
280 0.2 0.4 1.0 M.NaCI U/mL
1.00 50
0.80 40
0.60 30
0.40 20
0.20
0.00 0 100 200 300 400 500 VOLUME(ml)
Figure 12 88
6 . Separation on the Basis of Molecular Weight
a. Concentration on Hydroxylapatite
From analysis on SDS-polyacrylamide gels# the major contaminants were of lower molecular weight than the poly merase and therefore a separation step employing gel fil tration would be appropriate. However, before this could be carried out, it was necessary to concentrate the enzyme.
This was accomplished by using a short hydroxylapatite column (2.0 x 1.0 cm) equilibrated with Buffer A. The enzyme was either diluted with Buffer A containing no NaCl or dialyzed against the same buffer until the conductivity was below that of Buffer A. After applying the enzyme to the column and washing it with 15 ml of Buffer A, the poly merase was eluted with Buffer A containing 0.5 M potassium phosphate.
A typical elution profile from this hydroxylapatite column is illustrated in Figure 13. As shown, no active enzyme is eluted until the phosphate concentration is in creased to 0.5 M at which point a peak was observed fol lowed by a tail, which is very often seen on hydroxy1- apatite. Nevertheless, the enzyme activity that eluted with the protein was concentrated about 5-fold and the recovery usually between 50 and 60%. Figure 13. Hydroxylapatite Concentrating Column
The enzyme to be concentrated, 31 ml, 2,534 u, was applied
to a hydroxylapatite column (2.0 x 1.0 cm) equilibrated with Buffer A. After washing with 15 ml of Buffer A, the
polymerase was eluted with Buffer A containing 0.5 M potas
sium phosphate. Fractions 29 to 32 (6.3 ml) and 33 to 45
(20.0 ml) were pooled. Protein was measured by the absor
bance at 280 nm (•—• ) and enzyme activity by the Milli-
pore filter assay (O O).
89 U/mL 0 .5 M ®
320
280 280
2 4 0
.00 200
0.80 160
0.60 120
0.40 80
0.20 4 0
0.00 20 30 40 50 FRACTION No.
Figure 13 91
Other methods for concentrating the polymerase were studied, such as ultrafiltration using an Amicon stirred pressure cell with a variety of membranes, as well as con centration in a dialysis bag against solid sucrose. The pressure cell method resulted in large losses of enzyme activity, probably due to binding of the enzyme to the ultrafiltration membrane and the solid sucrose methods were too slow. More recently, a method has been described by Ferro and Olivera (65) employing dialysis against satu rated ammonium sulfate.
b. Sephadex G-150 Column Chromatography
The concentrated Blue Sepharose Pool II was applied
to a Sephadex G-150 column (two columns: 1.5 x 119 cm and
1.6 x 65 cm, connected in series) equilibrated with 50 mM
Tris-HCl, 10 mM mereaptoethanol, 0.5 M KC1, 10% glycerol
and 0.02% NaN^; the solution was adjusted to pH 8.0 at 4°C.
The enzyme was eluted with the same buffer. The polymerase,
eluting at a volume of 150 to 200 ml, was pooled and dia-
lyzed against Buffer A. Depending on the protein concen
tration, it was sometimes necessary to concentrate the en
zyme on a hydroxylapatite column before storing at -70°C.
As shown in Figure 14, the polymerase eluted with the
major protein peak and the total recovery of enzyme was 83%.
The specific activity of the leading edge of the activity Figure 14. Dual Sephadex G-150 Column Chromatography
The concentrated Blue Sepharose fraction (Pool II.) of 10 ml
(2,466 u) was applied to a Sephadex G-150 column (two
columns in series: 1.5 x 119 cm and 1.6 x 6 5 cm) equili brated with 50 mM Tris-HCl, 10 mM mercaptoethanol, 0.5 M
KC1, 10% glycerol and 0.02% NaN^; the solution was adjusted
to pH 8.0 at 4°C. Elution was with the equilibration buf
fer. The active enzyme fractions were pooled as designated
in the figure. Protein was monitored by the method of
Bradford (137) (------) and enzyme activity by the Milli- pore filtration assay ( ----- ).
92 SEPHADEX G 150 595
0.12
0.10
0.08 120
0.06 90
0.04 -6 0
0.02 30
0.00 0 50100 150 200 250 VOLUME( m l)
Figure 14 94
peak, Pool II was 3,287 units/mg and that of Pool I was
1,413 units/mg. The highest specific activity for poly(ADP-
ribose) polymerase from calf thymus, which is entirely de
pendent on added DNA, reported to date, has been 1,250 units/mg (68).
7. Summary
The steps of the purification scheme discussed above are summarized in Table 4. The combined overall yield is
13%, and is comparable to what has been reported (64,68); however the specific activity of both Sephadex G-150 pools
is higher than what has been reported for calf thymus poly-
(ADP-ribose) polymerase free of DNA (68). Since the speci
fic activity is a measure of purity, the higher value being more pure, this is an indication that the enzyme is
relatively pure. However, because of the complexity of the system, the specific activity depends on the conditions of assay. Since the assay conditions used are not identi cal, caution must be taken in assessing purity on this basis. On the other hand, from these and other results presented below, the enzyme obtained by this purification procedure appeared to be sufficiently pure to be used for further studies. The homogeneity of this preparation will be discussed below. Table 4
Purification of Poly(ADP-ribose) Polymerase
Step Total Protein Total Units Specific Activity mg nmol/min units/mg
Crude Extract 14,300 15,761 1.1
Ammonium Sulfate 3,700 10,744 2.9
DNA-Agarose 77 13,316 173
Hydroxylapatite 38 7,136 188
Blue Sepharose CL-6B 5.0 2,466 493
Dual Sephadex G-150 I 1.1 1,554 1,413
II 0.15 493 3,287
vo Ul 96
B. Homogeneity
As previously stated, it was very important to have a reproducible enzyme purification procedure which would yield enzyme of high purity in good yield. The high degree of purity was necessary before any mechanistic studies could be carried out. The homogeneity of poly(ADP-ribose) polymerase was investigated using SDS-polyacrylamide gel electrophore sis (SDS-PAGE), isoelectric focusing (IEF), and a paper chromatographic assay method. Using SDS-PAGE allows one to detect impurities with different molecular weights and IEF will detect protein contaminants with different isoelectric points (pi). The latter is indicative of DNA contamination since DNA is known to shift the pi from 9.4 to 4.0 (116).
The paper chromatographic assay was employed in order to detect the presence of contaminating enzymes which would degrade NAD+ as well as possibly any enzymes which may de grade the product of the polymerase reaction. The results of analyses employing these methods will be presented below.
1. Polyacrylamide Gel Electrophoresis in the Presence
of SDS
Samples at various stages of the purification were examined using SDS-PAGE according to Fairbanks et aJL. (139) as described in Methods and the results are shown in Figure
15. As seen in Gel 2 (DNA-agarose fraction), the polymerase 97 band designated by the arrow becomes visible and with each subsequent purification step, contaminating proteins are removed. This band is assigned as the polymerase by virtue of the fact that approximately equivalent units were loaded onto each gel (except crude) and that the intensity of this band is relatively constant throughout the purification.
In addition, as will be discussed later, when the polymer ase is incubated with [adenine-2,8- 3H] NAD+ and the mixture then applied to an SDS gel, radioactivity and protein co- migrate. This is consistent with a known reaction cata
lyzed by the enzyme, i.e. the automodification reaction.
Gel 6 (Sephadex G-150 Pool II) shows the major band being enzyme with some faint bands of higher molecular weight.
An estimate of the homogeneity was made from the densito- metric tracing using the cut and weigh method. The esti mate is 95% homogeneity.
Since only small amounts of protein were applied to gels 4, 5 and 6 , staining these gels with silver which has been reported to be 100 times more sensitive than Coomassie blue (151) could provide a more rigorous test of purity.
As shown in Figure 16, no proteins other than those de
tected by Coomassie blue are detected by the silver
staining procedure in the Sephadex G-150 Pool II enzyme,
confirming its high degree of purity. Fraction 15. Purity of Poly(ADP-ribose) Polymerase as
Shown by SDS-Polyacrylamide Gel Electro
phoresis
Polyacrylamide gel electrophoresis was carried out in the presence of 1.0% SDS according to Fairbanks et ad. (139) as described in Methods. Samples at various stages, in the purification procedure were examined as follows: 1) crude extract, 51.1 yg; 2) DNA-agarose fraction, 25.4 yg; 3) hydroxylapatite fraction, 35.1 yg; 4) Blue Sepharose Pool
II, 11.0 yg; 5) Sephadex G-150 Pool I, 2.7 yg; and 6) Sepha dex G-150 Pool II, 2.4 yg. The amounts of protein utilized correspond to approximately the same number of units of en zyme (between 4 and 8 units), with the exception of the crude extract. Protein samples were precipitated using an equal volume of ice cold 24% TCA. Gels were stained according to Reisner (140) after removing the SDS by shaking each gel in 200 ml of 25% isopropanol-10% acetic acid overnight.
The position of the arrow corresponds to the poly(ADP- ribose) polymerase band.
98 99
Figure 15 Figure 16. Silver Staining of SDS-Polyacrylamide Gels in
Figure 15
Gels 4, 5 and 6 , described in Figure 15, were subsequently stained with silver using the method of Wray et al. (141).
The position of the arrow corresponds to the poly(ADP- ribose) polymerase band.
100 101
Figure 16 102
2. Isoelectric Focusing
Isoelectric focusing in polyacrylamide gels was car ried out according to Righetti and Drysdale (144) as de tailed in Methods. As shown in Figure 17, purified poly-
(ADP-ribose) polymerase exhibits a sharp peak corresponding to a pi of 9.4. This value is in very good agreement with values that have been published (68,116). The fact that there is only one peak is a good indication that the poly merase is very pure. In addition, since the pi is 9.4 and not less, the presence of DNA is probably negligible (116), especially since the enzyme is inactive without added DNA.
3. Qualitative Product Evaluation
In preliminary studies it was observed that when crude extract was used in an assay for the polymerase an^ then spotted on paper and chromatographed in solvent system I
(see Methods), NAD+ was cleaved to produce a fast moving product. At this point it was realized that the use of a paper chromatography assay would be helpful in qualita tively detecting the polymerase at different stages of the purification. In addition, such an analysis would be able to detect contaminating enzymes that degrade NAD+ as well as enzymes that would degrade the reaction product, poly-
(ADP-ribose). Figure 17. Isoelectric Focusing of Purified Poly(ADP-
ribose) Polymerase Using Polyacrylamide Gel
Rods
Isoelectric focusing was carried out according to the pro cedure of Righetti and Drysdale (144) as stated in Methods.
Purified poly(ADP-ribose) polymerase (5.6 yg) was mixed with an equal volume of 4% Biolyte 3/10 that was adjusted to pH 10 with 1.0 N NaOH. The sample was applied under the cathode buffer (0.02 M NaOH) to a pre-electrophoresed 5% gel and focused at a constant current of 1 mA/gel for 12 hours at 4°C. The anode buffer was 0.01 M H^PO^. The gel containing the sample was stained by the Coomassie blue - perchloric acid method of Reisner (140) after removing the ampholytes by soaking the gel in 10% TCA overnight. The gel was then scanned at 620 nm (------). The pH gradient
(•— — •) was measured in another gel without sample as described in Methods.
103 pH •) 10
8
6
4
2
0 2 4 6 8
DISTANCE (cm) 104
Figure 17 105
Figures 18, 19, 20 and 21 show the results of the paper chromatography assay using the crude extract,
(NH^)2^0^ fraction, DNA-agarose fraction, and purified polymerase, respectively. As shown in Figures 18 and 19,
NAD+ is cleaved to a faster moving product as well as being incorporated into origin material (polymer). How ever, after the DNA-agarose column (Figure 20) and for the purified polymerase (Figure 21), no fast moving pro duct is formed and only a shift in radioactivity from the substrate NAD to the material at the origin is seen.
This indicates that enzymes that degrade NAD+ under these conditions have been removed by chromatography on DNA- agarose and that the final enzyme preparation is equally devoid of these enzymes. Furthermore, if any polymer de grading enzymes were present, such as poly(ADP-ribose) glycohydrolase or a phosphodiesterase, free ADP-ribose or isoADP-ribose would be detected, respectively. The results in Figures 20 and 21 indicate that this is not the case.
It was therefore concluded on the basis of these results and the results of SDS-PAGE and IEF, that the polymerase is sufficiently pure to employ in further studies. Figure 18. Paper Chromatographic Analysis of the Polymerase
Reaction Using Crude Extract as the Enzyme Source
Crude extract (0.15 u) was incubated with 0.1 M Tris-HCl (pH
8.0), 10 mM MgCl2 , 1 mM DTT, 100 yg/ml calf thymus DNA, 100
yg/ml whole histone, and 0.2 mM [adenine-2,8- 3H] NAD+ (10,000 dpm/nmol) in a final volume of 50 yl at 25°C for 30 minutes.
The reaction was then spotted on Whatman 3 MM filter paper and chromatographed using solvent system I. Analysis of the radioactivity was like that described in Methods. A control with boiled enzyme was also chromatographed and is designated
(•---- • ) and the experimental is designated (□---- □ ). Stan- + dards 1 through 7 are ADP, NAD , AMP, ADP-ribose, adenosine, adenine, and inosine, respectively and were detected by using an ultraviolet light.
106 20 DPM (%)
12 16 20 24 28 DISTANCE OF MIGRATION (cm)
Figure 18 107 Figure 19. Paper Chromatographic Analysis of the Poly-
merase Reaction Using the (NH4)2S04 Fraction
as the Enzyme Source
The ammonium sulfate fraction (0.70 u) was incubated as described in the legend to Figure 18 and chromatographed in solvent system I. A control with boiled enzyme (• • ) was chromatographed along with the experimental (□ □ ).
Standards 1 through 7 are ADP, NAD+ , AMP, ADP-ribose, ade nosine, adenine, and inosine, respectively and were de tected using an ultraviolet light.
108 25
(%)
DISTANCE OF MIGRATION (cm)
Figure 19 Figure 20. Paper Chromatographic Analysis of the Poly
merase Reaction Using the DNA-agarose Fraction
as the Enzyme Source
The DNA-agarose fraction (0.56 u) was incubated as de scribed in the legend to Figure 18 and chromatographed in solvent system I. A control with boiled enzyme (• • ) was chromatographed along with the experimental (CD a) .
Standards 1 through 7 are ADP, NAD+ , AMP, ADP-ribose, ade nosine, adenine, and inosine, respectively and were de tected using an ultraviolet light.
110 DPM (%)
8 12 16 DISTANCE OF MIGRATION (cm)
Figure 20 Figure 21. Paper Chromatographic Analysis of the Poly
merase Reaction Using Purified Poly(ADP-
ribose) Polymerase
Purified poly(ADP-ribose) polymerase (1.0 u) was incubated as described in the legend to Figure 18 and chromatographed in solvent system I. A control reaction with boiled enzyme
(•--- •) was chromatographed along with the experimental
(□--- □). Standards 1 through 7 are ADP, NAD+ , AMP, ADP- ribose, adenosine, adenine, and inosine, respectively and were detected using an ultraviolet light.
112 7 0
6 0
50
4 0
DPM (%)
2 0 .
B 12 16 20 24 28 36 DISTANCE OF MIGRATION (cm)
Figure 21 114
C . Properties of Poly(ADP-ribose) Polymerase
Before any mechanistic studies could be carried out, it was necessary to characterize the polymerase to verify that its properties are similar to those that have been published. Such properties as the requirements for full activity, the stability, molecular weight, and the analysis of products were studied. The Sephadex G-150 pools are referred to as purified enzyme and are used in the following studies unless designated otherwise.
1. Requirements for Maximal Activity
Poly(ADP-ribose) polymerase from calf thymus (68), rat liver (107) and Ehrlich ascites tumor cells (116) as well as from other sources has been reported to have an absolute +2 requirement for DNA. Also histone, DTT and Mg have been shown to be required for full activity. These requirements were investigated using the standard Millipore filter assay as described in Methods with the indicated omissions shown in Table 5. These results indicate that the purified poly merase has an absolute requirement for DNA, analogous to what has been reported. In addition, the enzyme requires +2 histone, DTT, and Mg for maximal activity.
The requirement for DNA and whole histone was also evaluated using the paper chromatography assay. Reactions 115
Table 5
Requirements of Purified Enzyme for
Poly(ADP-ribose) Synthesis
pmol %
Complete 1022 100 Boiled Enzyme 0 0 - DNA 0 0 - DNA, -Histone 20 2 - Histone 430 42 - DTT 427 42 - MgC l 2 622 61
Experimental details: Poly(ADP-ribose) polymerase was assayed by the Millipore filter assay as described in Methods. The complete reaction mixture contained 100 mM Tris-HCl (pH 8.0), 10 mM MgCl 2 , 1 mM DTT, 0.2 mM (adenine- 2,8-3H] NAD+ (10,000 dpm/nmol), 10 pg each of calf thymus DNA and calf thymus whole histone and purified polymerase in a total volume of 0.1 ml. After incubation for 2 minutes at 25°C, the protein was precipitated by adding 2 ml of ice cold 10% TCA, collected on a Millipore filter and quanti tated via liquid scintillation counting as described in Methods. Reactions with the indicated omissions were conducted similarly. 116 were carried out as described in Methods with the indicated
omissions.
Figures 22 and 23 show the reaction with purified poly merase in the absence of DNA and in the absence of whole
histone, respectively. The lack of radioactivity at the
origin in the absence of DNA supports the idea that the
enzyme is DNA dependent. Without whole histone, the polymer
forms, but not to the same extent as in its presence (Figure
21). Using the radioactivity at the origin as a measure of
polymer formation, the percentage of activity in the absence
of DNA and whole histone can be calculated. For purified
polymerase in the presence of 100 yg/ml DNA and 100 yg/ml
whole histone (standard amounts), 66.3% of the total radio
activity applied to the chromatogram was at the origin. In
the absence of DNA, only 0.77% of the total radioactivity
applied remained at the origin, while without whole histone
24.2% of the total radioactivity applied remained at the
origin. Using 66.3% as a measure of the total enzyme acti
vity in a complete assay reaction, the omission of DNA re
sulted in a 99% reduction of activity, while the omission of
whole histone resulted in a 64% reduction. This was quite
similar to what has been observed using the Millipore fil
ter assay. Figure 22. Paper Chromatographic Analysis of the Polymerase
Reaction Using Purified Enzyme in the Absence of
Added DNA
Purified enzyme (1.0 u) was incubated with 0.1 M Tris-HCl
(pH 8.0), 10 mM MgC l2 , 1 mM DTT, 100 pg/ml whole histone, and 0.2 mM [adenine-2,8- 3H] NAD+ (10,000 dpm/nmol) in a final volume of 50 yl at 25°C for 30 minutes. The reaction was then spotted on Whatman 3 MM filter paper and chromato graphed using solvent system I. Analysis of the radioacti vity was like that described in Methods. A control with boiled enzyme was also chromatographed and is designated
(•---- •) and the experimental is designated (o □) .
Standards 1 through 7 are ADP, NAD+ , AMP, ADP-ribose, ade nosine, adenine, and inosine, respectively and were de tected using an ultraviolet light.
117 25
20 DPM (%)
0 4 8 12 16 20 24 28 32 36 DISTANCE OF MIGRATION (cm)
Figure 22 118 Figure 23. Paper Chromatographic Analysis of the Polymerase
Reaction Using Purified Enzyme in the Absence of
Added Whole Histone
Purified enzyme (1.0 u) was incubated as described in the legend to Figure 22 except that whole histone was omitted from the incubation mixture and 10 0 yg/ml calf thymus DNA was included. Chromatography was in solvent system I and a boiled enzyme control (•—— •) along with the experimental
(□---- □ ) was run. Standards 1 through 7 are ADP, NAD+ , AMP,
ADP-ribose, adenosine, adenine, and inosine, respectively, and were detected using an ultraviolet light.
119 30
DPM 2 0 (%)
3 620 24 28 3620 DISTANCE OF MIGRATION (cm) 120
Figure 23 121
2. Storage of the Purified Enzyme
Once the polymerase had been purified, it was necessary
to find an appropriate means of storage and to evaluate its
stability. A very convenient and effective method of stor
age which has been used for other enzymes involves a quick
freezing process using liquid nitrogen. The enzyme is per mitted to drip into liquid nitrogen through a thin gauge
syringe needle. Upon entering the nitrogen, the enzyme
freezes rapidly into small pellets which are stored at -70°C.
Samples are easily removed with a spatula (152).
Purified poly(ADP-ribose) polymerase was either frozen with liquid nitrogen as described above or frozen in bulk
(1.0 ml aliquot) at -70°C. Both were stored at -70°C. Ali
quots of the liquid nitrogen frozen enzyme were thawed
along with the 1.0 ml bulk frozen aliquot and were assayed
(10 yl) for enzyme activity using the Millipore filter as
say. After the bulk frozen pool had thawed and an aliquot
removed for assay, it was immediately refrozen at -70°C.
The results shown in Table 6 indicate that quick
freezing with liquid nitrogen does not preserve the enzyme
activity and in addition, the enzyme is inactivated just by
storing at -70°C. Also, repeated freezing and thawing of
the bulk frozen pool results in a significant loss in en
zyme activity. This is quite similar to what has been re ported by Ito et al. (68). The stability problem can be 122
Table 6
Stability of Poly(ADP-ribose) Polymerase
Frozen in Liquid Nitrogen versus Bulk
Frozen as a Function of Time at -70°C
Days at -70°C Liquid ^ Frozen Bulk Frozen
% activity % activity u/ml remaining u/ml remaining
0 36. 3 100 36. 3 100 1 27.4 76 31.6 87 2 19. 6 54 28. 3 78 3 17.6 44 23.5 65 4 11. 7 32 18.0 50
Experimental details: Purified poly(ADP-ribose) polymerase was frozen in liquid nitrogen as described in the text or bulk frozen (1.0 ml aliquot) at -70°C. After storing at -70°C for the indicated times, an aliquot of the liquid nitrogen frozen enzyme was thawed, as was the entire sample of the bulk frozen enzyme. Those were then assayed in a reaction mixture containing 100 mM Tris-HCl (pH 8.0), 10 mM MgC l 2 / 1 niM DTT, 1.74 mM [adenine-2,8- 3H] NAD+ and 10 yg each of calf thymus DNA and calf thymus whole histone in a total volume of 0.1 ml. The reaction was stopped with 2 ml of 10% ice cold TCA after incubating at 25°C for 2 minutes. Activity was determined using the Millipore filter assay as described in Methods. The bulk frozen enzyme pool was immediately refrozeii after removing an aliquot for assaying. 123 overcome by storing the enzyme at higher protein concen tration (data not shown) and by storing the enzyme in small volumes in order to circumvent the need to thaw the entire pool each time enzyme is needed.
The length of time of storage at -70°C was also an im portant factor. After one year at -70°C, Blue Sepharose pools I, III, and IV were assayed for enzyme activity and were found to have 76, 62 and 79% activity remaining, res pectively. These results indicate that storage at -70°C for periods of time up to one year only show a slight loss in activity. It is probable that shorter storage periods would exhibit smaller losses in activity.
3* Stability of the Purified Enzyme
The stability of purified poly(ADP-ribose) polymerase was investigated at 25°C in the presence and absence of DNA and whole histone using the standard Millipore filter assay.
Purified enzyme (0.31 u) was pre-incubated at 25°C in the presence or absence of DNA and whole histone. The activity of the enzyme mixture was assayed as a function of time, and the results are shown in Figure 24, These results show that in the presence of DNA and whole histone, there was no loss in activity for up to 15 minutes; however without DNA and whole histone, the polymerase loses 27% of its catalytic activity in 15 minutes. Therefore, it can be concluded that Figure 24. Protection with DNA and Whole Histone
Purified poly(ADP-ribose) polymerase (0.31 u) was preincu bated at 25°C with 10 yl 1 M Tris-HCl (pH 8.0), 5 yl 0.2 M
MgCl2 f 5 yl 20 mM DTT, 40 yl H 20 with or without 10 yl
1 yg/yl DNA and 10 yl 1 yg/yl whole histone. At the indi cated times, 10 yl 1.74 mM [adenine-2,8- 3H] NAD+ (9,770 dpm/nmol) were added to the reactions containing DNA and histone (O O) and then assayed using the Millipore fil ter assay previously described in Methods. To those reac tions in which DNA and whole histone had been omitted, 10 yl
1 yg/yl DNA, 10 yl 1 yg/yl whole histone (•----• ) and 10 yl
1.74 mM [adenine-2,8- 3H] NAD+ (9,770 dpm/nmol) were added and assayed as above. The control activity taken as 100% is that for no preincubation.
124 01______i______i______i______i______i__ 0 20 40 60 80 100120 140 160
TIME (MIN)
Figure 24 to cn 126
DNA and whole histone protect the polymerase from inacti vation at 25°C for short periods of time. This observation was similar to that of Niedergang et al. (101) , whose DNA- independent polymerase was more stable than their DNA-depen- dent polymerase. This was due to the fact that the DNA- independent enzyme preparation contained endogenous DNA.
This protective effect of DNA on enzyme stability has also been observed by Petzold et ad. (105) for poly(ADP-ribose) polymerase from lamb thymus.
The heat inactivation of purified poly(ADP-ribose) polymerase (0.31 u) was investigated by heating at 93°C.
Samples were removed as a function of time, and enzyme activity was determined. The results shown in Figure 25 indicate that after 3 minutes, there is no catalytic acti vity remaining, < 12.5 seconds. In view of these obser vations, it was possible to use boiled enzyme controls in the standard assay by heating for at least 3 minutes at
9 3 °C .
4. Molecular Weight Determination
It was important to determine the molecular weight for purified poly(ADP-ribose) polymerase for two reasons: (1) to compare it to the published value, and (2) to have a reliable value to use in experiments in which an accurate number of moles of enzyme was important. Two methods using Figure 25. Heat Inactivation
Purified poly(ADP-ribose) polymerase (0.31 u) was preincu bated with 10 yl 1 M Tris-HCl (pH 8.0), 5 yl 0.2 M MgCl2,
5 yl 20 mM DTT, and 40 yl H 20 at 93°C. At the indicated times, the mixtures were rapidly cooled on. ice and 10 yl
DNA (1 yg/yl) and 10 yl whole histone (1 yg/yl) were added.
After warming to room temperature, 10 yl 1.74 mM [adenine-
2, 8- 3H] NAD+ (9,770 dpm/nmol) were added and the mixtures assayed using the Millipore filter assay previously de scribed in Methods. The control activity taken as 100% is that for no preincubation.
127 iue 25 Figure % CONTROL ACTIVITY 100 20 0 4 0 6 0 8 0 1.0 IE (MIN) TIME 2.0 3.0 128 129 polyacrylamide gel electrophoresis in the presence of the detergent SDS (SDS-PAGE) were employed.
a. SDS-PAGE according to Weber and Osborn (138)
Polyacrylamide gel electrophoresis in the presence of
0.1% SDS was carried out according to the procedure of Weber and Osborn (138) using differing gel concentrations (4-7%) and the data analyzed by the method of Frank and Rodbard
(153). A calibration curve was established, using a series of proteins of known molecular weight (138,154). Protein standards chosen were a-chymotrypsinogen (25,700), oval bumin (46,000), and the monomer (68,000) and dimer (136,000) of bovine serum albumin. For each gel percentage, plots of log molecular weight versus were constructed and are shown in Figure 26. It can be seen that the molecular weight of the polymerase was quite constant over the four gel percentages employed, being 120,000, 130,000, 126,000 and 124,000 for 4, 5, 6 and 7% gels, respectively. The average of these determinations was 125,000. The data from the same gels can be used to plot according to the procedure of Ferguson (155). Figure 27 shows these plots, in which log Rf versus gel percentage for the protein standards and the polymerase is used. From the slope of these lines (re tardation coefficient, KR) and the molecular weight of the standards, the molecular weight of the polymerase can be Figure 26. Log Molecular Weight versus R^
SDS-PAGE according to Weber and Osborn (138) was carried out as described in Methods using gels of 4, 5, 6 and 7% polyacrylamide (0.3 x 6.0 cm). The protein standards em ployed were: (1) a-chymotrypsinogen (25,700); (2) oval bumin (46,000); and (3) bovine serum albumin (monomer,
68,000; dimer, 136,000) applied at a level of 3 vig. Gel percentages are designated as follows: 4% (•), 5% (□),
6% (O) and 7% (★). Purified poly (ADP-ribose) polymerase was applied (2.5 yg) and is designated (O). After staining and destaining as described in Methods, gels were scanned at 620 nm and the Rf's determined from the densi- tometric tracings. All lines were generated via linear regression analysis of the data obtained.
130 131 cn LOG LOG MOLECULAR WEIGHT
Figure 26 Figure 27. Ferguson Plot. Log Rf versus Gel Percentage
The data were taken from the experiment described in Figure
26. Protein standards are designated as follows: a- chymotrypsinogen ( □ ), ovalbumin (O ), BSA monomer ( ★ ) and BSA dimer ( • ). Poly(ADP-ribose) polymerase is desig nated by (O ). All lines were generated via linear re gression analysis of the data obtained.
132 GEL %
Figure 27 133 134 calculated from the relationship shown in Figure 28. In this analysis, the polymerase was found to have a molecu
lar weight of 118,000. Both of the determined values are
close to that reported (120,000) for calf thymus poly(ADP- ribose) polymerase (68,102).
b. SDS-PAGE according to Fairbanks et al. (139)
Polyacrylamide gel electrophoresis was carried out in the presence of 1.0% SDS according to the procedure de
scribed by Fairbanks et all. (139) using 6% gels as pre viously described in Methods. As shown in Figure 29, puri
fied poly(ADP-ribose) polymerase exhibits a molecular weight of 125,000, which is in good agreement with that obtained using the method of Weber and Osborn (138) .
Therefore a molecular weight of 125,000 will be used throughout in all calculations.
5. Product Analysis
In order to determine whether the product of the enzyme reaction was a polymer composed of ADP-ribose units, the average chain length was investigated.
Subjecting poly(ADP-ribose) to snake venom phospho diesterase digestion and subsequent paper chromatographic analysis is the typical method for determining the average chain length (see Methods). As shown in Figure 30, the Figure 28. Molecular Weight versus KR
The values for KR were taken from Figure 27. Protein standards are designated as follows: a-chymotrypsinogen
( □ ) , ovalbumin (O ), BSA monomer ( ★ ) and BSA dimer ( • ).
Poly(ADP-ribose) polymerase is designated by (O). The line was generated via linear regression analysis of the data obtained.
135 136
20
o o
X
0 2 4 6 8 10 12 14 MOLECULAR WEIGHT X I0‘ 4
Figure 28 Figure 29. Log Molecular Weight versus using SDS-PAGE
according to Fairbanks et.al. (139)
SDS-PAGE was carried out according to Fairbanks et al.
(139) using 6% polyacrylamide gels (0.5 x 7.5 cm) as de scribed in Methods. Protein standards (138,154,156) used
(3-5 yg) were as follows: (1) Myosin (200,000); (2) RNA polymerase &' subunit (165,000); (3) RNA polymerase 6 sub unit (155,000); (4) g-galactosidase (130,000); (5) Bovine serum albumin monomer (68,000); (6) RNA polymerase a sub unit (39,000); (7) a-chymotrypsinogen (25,700); and (8) soybean trypsin inhibitor (21,500). Purified poly(ADP- ribose) polymerase (2.5 yg) was used and is designated as
(O ). The line was generated via linear regression ana lysis of the data obtained.
137 MOLECULAR iue 29 Figure 4 5 0.1 0.2 0.3 R .4 0 j 0.5 6 0.6 7 0 0.8 u> H 00 139 paper chromatographic profile of a phosphodiesterase di gestion reaction revealed the production of phosphoribosyl- isoADP-ribose (arrow), isoADP-ribose (migrating with ADP- ribose), AMP and adenosine. From these data, the average chain length was calculated to be 3.89 with whole histone and 2.83 without whole histone (data not shown) under the conditions described in the legend to Figure 30. This establishes the reaction product as a polymer and strongly suggests that this polymer is composed of repeating units of ADP-ribose.
D. Mechanistic Studies
The preceeding investigations demonstrate that the properties of the enzyme preparation obtained by the de scribed purification procedure were comparable to those reported for other preparations. The homogeneity was suf ficiently high to permit its use in studies pertaining to its mechanism.
1. Isotope Exchange Reaction
Many group transfer enzymes catalyze an isotope ex change reaction. Frequently these reactions have important mechanistic implications. The question of whether poly-
(ADP-ribose) polymerase can catalyze such an exchange be tween nicotinamide and NAD+ was examined. Figure 30. Paper Chromatographic Analysis of Phosphodi
esterase Digestion Reaction Products
Purified poly(ADP-ribose) polymerase (3.86 yg) was incu bated at 27°C for 30 minutes with 100 mM Tris-HCl (pH 8.0),
10 mM M g C l 2 , 1 mM DTT, 50 yg/ml calf thymus DNA, 50 yg/ml whole histone and 0.21 mM [adenine-U-1^C] NAD+ (5.88 x 105 dpm/nmol) in a total volume of 0.5 ml. After terminating the reaction with 1.0 ml of ice cold 25% TCA, the protein precipitate was collected by centrifugation and washed as described in Methods. The precipitate was treated with
100 yl of I N NH^OH for 30 minutes at 25°C, lyophilized and then dissolved in 40 yl of 50 mM Tris-HCl (pH 8.0),
10 mM M g C l 2 and 2.5 mM AMP containing 0.06 units of phos phodiesterase and 0.7 yg trypsin. After incubating for 6 hours at 37°C the mixture was spotted on Whatman 3 MM filter paper and chromatographed in solvent system II. The arrow indicates the position where phosphoribosyl isoADP- ribose would migrate. Standards 1, 2 and 3 are ADP-ribose,
AMP and adenosine, respectively and were detected using an ultraviolet light.
140 DPM X 10“ CD
1/10
0 10 20 30 DISTANCE OF MIGRATION (cm)
Figure 30 141 142
The experimental reaction consisted of purified poly-
(ADP-ribose) polymerase (0.38 u) in 100 mM Tris-HCl (pH
8.0) , 10 mM MgC^/ 1 mM DTT, 100 pg/ml DNA and various con centrations of NAD+ (0.75 pM, 7.5 pM and 75 pM) in a total volume of 25 pi. These concentrations of NAD+ represent levels above and below the Km for NAD+ (54 pM). The reac tion was started by adding enzyme and was maintained for one minute at 25°C. Then a solution (25 pi) containing
1.2 mM [carbonyl-1^C] nicotinamide (53 pCi/pmol) in 0.2 M
Tris-HCl (pH 8.0), 0.02 M MgCl2, 100 pg/ml DNA and 2 mM DTT was added and the reaction was allowed to proceed for an additional minute. This concentration of nicotinamide is much greater than the for nicotinamide (13 pM). At this point, 50 pi of 1.74 mM NAD+ was added (preheated to 50°C) to trap any labeled NAD+ which may have formed and the reaction mixture was heated at 50°C for two minutes. A control reaction was run in which boiled enzyme was used.
Aliquots (50 pi) were removed, subjected to descending paper chromatography in solvent system III, and analyzed for radioactivity as described in Methods.
As shown in Figure 31, the experimental with 0.75 pM
NAD+ is identical to the boiled enzyme control. In addi tion, with 7.5 pM NAD+ and 75 pM NAD+ , the data (not shown) were identical to that in Figure 31. Figure 31. Paper Chromatographic Analysis of the Isotope
Exchange Reaction
Purified poly(ADP-ribose) polymerase (0.38 u) was incubated in 100 mM Tris-HCl (pH 8.0), 10 mM MgC^, 1 mM DTT, 100 yg/ml DNA and 0.75 yM NAD+ in a total volume of 25 yl for one minute at 25°C. Then 1.2 mM [carbonyl-1^C] nicotina mide (53 yCi/ymol) in 0.2 M Tris-HCl (pH '8.0), 0.02 M
MgCl2, 100 yg/ml DNA, and 2 mM DTT was added (25 yl) and further incubated for another minute at 25°C. The reac tion was terminated by adding 50 yl of 1.74 mM NAD+ (pre heated to 50°C) followed by heating at 50°C for 2 minutes.
A 50 yl aliquot was spotted on paper, chromatographed in solvent system III, and analyzed for radioactivity as de scribed in Methods. A control with boiled enzyme (------) was chromatographed in addition to the experimental
(----- ). Non-radioactive standards were detected using an ultraviolet light.
143 144
DPM (%) NAD NIC
0 .80
0.60
0.40
0.20
0.00 20 40 60 80 100 120 140 RNIC
Figure 31 145
Many variations of this experiment were carried out in order to rigorously establish whether or not the poly merase could catalyze an isotope exchange reaction between
NAD+ and nicotinamide- Such variations included higher
NAD+ concentrations, longer incubation times, and the presence of acceptor whole histone. These are briefly de scribed below.
Purified poly(ADP-ribose) polymerase (1.0 u) was incu bated in 100 mM Tris-HCl (pH 8.0) , 10 mM M g C ^ / 1 mM DTT,
50 yg/ml DNA, 0.2 or 0.5 mM NAD+ and 1.32 mM [carbonyl-14C] nicotinamide (53 yCi/ymol) in a total volume of 0.15 ml.
The reaction was started with the addition of enzyme and then 35 yl was immediately removed and spotted on paper as a control. The mixture was incubated at 25°C for various times (20 and 60 minutes) and then a 50 yl aliquot was removed and spotted for paper chromatography as described above. The results were the same as in Figure 31, and showed that no [11+C] NAD+ was found. In addition, the same experiment was carried out in the presence of 50 yg/ml whole histone at both NAD+ concentrations and using a boiled enzyme control and no enzyme control at 0.2 mM NAD+ .
The results (not shown) were identical to those in Figure
31. It is evident from the experiments described that poly(ADP-ribose) polymerase does not catalyze an isotope exchange reaction between NAD+ and nicotinamide. Therefore 146 it would appear that the equilibrium of the polymerase reaction must lie far to the right in favor of the cleavage of NAD+ .
2. Binding Sites
In order for poly(ADP-ribose) polymerase to catalyze the reactions it does, a number of different molecules must interact with the enzyme. This includes the sub
strate, activator, and acceptor which are NAD+ , DNA, and histone, respectively. The regions on the enzyme which interact with these molecules can be thought of as binding domains or binding sites. Therefore an examination of the different binding sites of poly(ADP-ribose) polymerase may provide some evidence as to how the enzyme catalyzes certain reactions. Studies involving the substrate, acceptor, and automodification binding sites are presented below.
a. Substrate Site
-L The substrate binding site is the location where NAD interacts with the enzyme. These interactions may be more significant for certain portions of the NAD+ molecule than for others. For example, nicotinamide has been shown to be a very potent competitive inhibitor of NAD+ for puri fied poly(ADP-ribose) polymerase from calf thymus (68,101) 147 and human lymphoid tissue (106). With this in mind, the following experiments employing nicotinamide as well as other substrate derivatives were carried out to probe the substrate binding site.
i. Kinetic Studies with Nicotinamide
The following kinetic experiment was carried out in order to determine the effect of .nicotinamide on the puri fied polymerase. Purified poly(ADP-ribose) polymerase
(0.5 u) was incubated with 100 mM Tris-HCl (pH 8.0), 10 mM
MgC^, 1 mM DTT, 100 pg/ml DNA, 100 pg/ml whole histone, 1 + varying concentrations of [adenine-2,S-^H] NAD (0.04 to
0.40 mM; 0.09 pCi), and nicotinamide (0, 25 or 50 pM) in a total volume of 0.1 ml. The reaction was started with the addition of enzyme and allowed to proceed for 2 minutes at
25°C before being terminated with 2 ml of ice cold 10% TCA.
The protein precipitate was collected on a Millipore filter and analyzed as described in Methods.
As shown in Figure 32, nicotinamide is a competitive inhibitor of NAD+ , which is similar to that observed by others (68,101,106). From this data the apparent Km for
NAD+ is 54 pM and the V is 1180 nmol min ^ mg ■*". The max inhibitor constant, K^, for nicotinamide can be calculated from the secondary plot (data not shown) to be 13 pM. The. values of K and V are in close agreement with those m max Figure 32. Inhibition by Nicotinamide
Purified poly(ADP-ribose) polymerase (0.5 u) was incubated in a final volume of 0.1 ml containing 100 mM Tris-HCl (pH
8.0), 10 mM MgCl 2 , 1 mM DTT, 100 pg/ml DNA, 100 pg/ml whole histone and varying concentrations of [adenine-2,8- 3H] NAD+
(0.04 to 0.40 mM; 0.09 pCi) and nicotinamide: 0 pM (♦— — ♦);
25 pM (□----□); 50 pM (■---- ■) . The reaction was started by adding enzyme, and after 2 minutes, terminated with ice cold 10% TCA. The analysis of enzyme activity was completed by the Millipore filter assay as described in Methods. The units of V are nmol min-1. Data was plotted according to
Lineweaver and Burk (157) using linear regression analysis on the data obtained.
148 149
30
20
10
0 10 20
Figure 32 150 published for calf thymus and adult bovine thymus poly(ADP- ribose) polymerase (68,100). The for nicotinamide is in close agreement with that of the DNA-independent polymerase
(101) .
ii. Substrate Derivatives
The affect of other substrate analogs was assessed in order to provide additional information about other binding
interactions. Reaction mixtures contained 0.31 u purified polymerase in 100 mM Tris-HCl (pH 8.0), 10 mM MgCl2, 1 mM
DTT, 100 yg/ml DNA, 100 yg/ml whole histone, 1.74 mM
[adenine-2,8- 3H] NAD+ (9,770 dpm/nmol) and various analogs in a final volume of 0.1 ml. After incubating at 25°C for
2 minutes, the reaction was terminated with ice cold 10%
TCA and the protein precipitate collected on a Millipore filter and analyzed as described in Methods.
As shown in Table 7, nicotinamide and NADH inhibit the polymerase strongly at approximately 0.1 mM and each to nearly the same extent. In contrast, AMP and ADP are only slightly inhibitory at concentrations 40 times greater.
From this it can be concluded that the binding interactions between the enzyme and the nicotinamide portion of the sub strate NAD+ are much stronger than the binding interactions between the enzyme and the adenine-containing portion of
NAD+ . In addition, ADP-ribose and ADP-ribonolactone at 151
Table 7
Effect of Substrate Analogs on
Poly(ADP-ribose) Synthesis
Addition Concentration ADP-ribose Incorporated
mM pmol %
None 243 100 AMP 3.95 203 84 ADP 4. 35 177 73 Nicotinamide 0 .12 89 37 NADH 0.10 85 35 ADP-ribose 0.18 202 83 ADP-ribonolactone 0.19* 192 79
Synthesized as previously described, consisting of a mixture of ADP-ribonolactone and ADP-ribonic acid. The 0.19 mM is based only on the ADP-ribonolactone. 152 roughly equal concentrations inhibit to approximately the same extent. Although the type of inhibition by ADP-ribose
and ADP-ribonolactone cannot be established from this data, it indicates the lactone is not much better as an inhibitor than ADP-ribose.
As previously mentioned in the Introduction, lactone derivatives of substrates have provided information about mechanisms involving carbonium ion intermediates (9 5,96)
and many have been designated as transition state analogs.
Because the lactone is sp^ hybridized and therefore planar
at C-l, it would mimic a carbonium ion if one was important
in the reaction pathway. This has been shown for the enzyme
lysozyme, in which the 6-derived lactone from tetra-N-acetyl-
chitotetraose has been shown to be 110 times more potent an
inhibitor than the unmodified tetrasaccharide (96) . Since
Wolfenden states that suitable transition state analogs
should be powerful enzyme inhibitors (97), it can be con
cluded that ADP-ribonolactone is neither a potent inhibitor
nor a transition state analog. Furthermore, this suggests
that a carbonium ion is not important in the reaction path way catalyzed by the enzyme poly(ADP-ribose) polymerase. 153
b . Acceptor Site
The roles which the acceptor proteins, histones, play in the poly(ADP-ribose) polymerase reaction were examined.
Before looking at this, however, the commercially available acceptors were screened for protein composition using acid- urea polyacrylamide gel electrophoresis as described in
Methods.
As shown in Figure 33A, whole histone is composed of at least five protein bands. On the other hand, histone HI, as shown in Figure 33B, is composed of one major protein band.
i. DNA and Whole Histone in Equal Proportions
As mentioned above, the enzyme poly(ADP-ribose) poly merase can catalyze the addition of poly(ADP-ribose) to histones. It has also been shown that this poly(ADP-ribo- syl)ation reaction is dependent on DNA. Since histones are known to interact quite strongly with DNA, it was of in terest to examine the effect of histone alone or with DNA on the poly(ADP-ribosyl)ation reaction.
The following experiment was designed to examine the dependence of the reaction on DNA and whole histone at various concentrations, but in equal proportions. Purified poly(ADP-ribose) polymerase (0.1 yg) was incubated in a final volume of 0.1 ml containing 100 mM Tris-HCl (pH 8.0), Figure 33.
A. Protein Scan of Whole Histone
Acid-urea polyacrylamide gel electrophoresis was carried out according to Panyim and Chalkley (142) using 12% polyacryla mide gels (0.5 x 7.5 cm) as described in Methods. Whole histone (10 pg) was loaded and gels were stained and de stained according to Bonner et al. (143). The gel was scanned at 620 nm using an ISCO Model 1310 Gel Scanner with a Model UA-5 Absorbance Monitor with an absorbance range setting of 2 .0 .
B. Protein Scan of Histone HI
Conditions were the same as in A, only 10 pg histone HI were used.
154 155
A.
B.
Figure 33 156
10 mM M g C ^ , 1 mM DTT, 100 mM NaCl, 0.174 mM [adenine-2,8-
3H] NAD+ (10,701 dpm/nmol) and varying amounts of DNA and whole histone in equal proportions (0 to 250 yg/ml of each).
The reaction was started by adding enzyme and terminated
after 2 minutes at 25°C with ice cold 10% TCA. The reaction mixture was analyzed using the Millipore filter assay as described in Methods. In addition, the above reaction was
carried out in the absence of whole histone but with DNA added (0 to 250 yg/ml) and in the absence of DNA but with whole histone added (0 to 250 yg/ml).
As shown in Figure 34, in the absence of DNA there is no reaction. This data also indicate that in the presence of DNA and whole histone at the level of 75 yg/ml each, there is maximum ADP-ribose incorporation. At higher levels of DNA and histone, the incorporation decreases. In addi tion, there is a 3.7-fold stimulation in the amount of ADP- ribose incorporated by the addition of whole histone at 100
yg/ml. Similar observations have been reported by others
(68,100,107).
ii. Whole Histone and Histone HI versus DNA
Concentration
As shown in Figure 33, the protein composition of whole histone is remarkably different from that of histone
HI, whole histone being a mixture of a number of different Figure 34. ADP-ribose Incorporation with Differing Con
ditions of DNA and Whole Histone
Poly(ADP-ribose) polymerase (0.1 yg) was incubated in a final volume of 0.1 ml containing 100 mM Tris-HCl (pH 8.0),
10 mM MgC^, 1 mM DTT, 100 mM NaCl, 0.174 mM [adenine-2,8-
3H] NAD (10,701 dpm/nmol) and varying amounts of DNA and whole histone in equal proportions (0 to 250 yg/ml)
(♦----♦). The reaction was started with the addition of enzyme and terminated after 2 minutes with ice cold 10% TCA.
The determination of ADP-ribose incorporation was by the
Millipore filter assay as described in Methods. Similar reactions were also carried out in the absence of DNA with added whole histone (0 to 250 yg/ml) (•— — •) and in the absence of whole histone with added DNA (0 to 250 yg/ml)
(■---- ■ ). ADP-ribose incorporation is expressed per minute of reaction.
157 iue 34 Figure ADP-Ribose Incorporated (nmol) 0.05 0.35 0.45 0.25 itn adb DA (/xg/ml) DNAHistone and/br 100 200 0 0 3 158 159
histone fractions. Because of the greater complexity,
the interactions of whole histone with DNA might be dif
ferent than the interaction of histone HI with DNA.
Therefore the effect of these different protein acceptors
on the poly(ADP-ribosyl)ation reaction was examined in
relation to DNA concentration in the following experiment.
The reaction conditions were the same as described in the
legend to Figure 34 with the following exceptions: His
tone, either whole or HI, was held constant at 100 yg/ml
and the DNA concentration was varied (0 to 400 yg/ml).
As shown in Figure 35, there appears to be a major
difference between whole histone and HI. Maximum incor
poration was observed in the presence of whole histone at
a concentration of 100 yg/ml and 75 yg/ml DNA. In addi
tion, at high DNA concentrations, an inhibitory effect is
seen similar to that shown in Figure 34 for high concen
trations of DNA and whole histone in equal proportions.
On the other hand, the maximum ADP-ribose incorporation with HI is much lower than that seen in the presence of whole histone. It is not until the ratio of DNA to HI is
3 to 1 that the incorporation of ADP-ribose in the pre
sence of each of the histones becomes equal. At 100 yg/ml whole histone and 10 0 yg/ml DNA, ADP-ribose incorporation
is 3.5 times that observed with 100 yg/ml HI and the same
amount of DNA. It can be concluded from this experiment Figure 35. ADP-ribose Incorporation with Whole Histone and
Histone HI and Varying DNA Concentrations
Poly (ADP-ribose) polymerase (.0.1 yg) was incubated in a final volume of 0.1 ml containing 100 mM Tris-HCl (pH 8.0),
10 mM MgC^f 1 mM DTT, 100 mM NaCl, 0.174 mM [adenine-2,8-
3H] NAD+ (10,701 dpm/nmol) . Whole histone (■— — ■) or histone HI (•— — •) at 100 yg/ml and varying amounts of
DNA. The reaction was started with the addition of enzyme and terminated after 2 minutes with ice cold 10% TCA. Ana lysis of ADP-ribose incorporated was using the Millipore filter assay described in Methods. ADP-ribose incorpora tion is expressed per minute of reaction.
160 0 .6 0 . o E c 0.50
■ O < u S 0.40 o Q. k. O o 0.30. c
Q) W O 0.20 .O ir ■ CL O 0.10 < 0.00 2 0 0 300 DNA (/tQ/ml)
Figure 35 162 that whole histone, being a mixture of different histones, results in more ADP-ribose incorporated than histone HI.
This may be possible for a number of reasons. With whole histone, the other histones present in addition to histone
HI may be better acceptors than histone HI alone; or the
possibility exists that whole histone may stimulate the
automodification reaction more than histone HI. Both of
these would result in more ADP-ribose incorporation with whole histone. The nature of the histone-DNA interaction may also be important. It is not known whether DNA binds to the polymerase in order to modify histone or if the histone-DNA complex is necessary for reaction. From recent evidence using histone HI (65) , it has been shown that HI molecules not bound to DNA were poly(ADP-ribosyl)ated in
preference to DNA-bound HI. In any case, the interaction
of whole histone with DNA could result in the formation of
a nucleosome-like structure, whereas with HI alone, the
interaction with DNA would probably be much different.
This could possibly explain the difference in ADP-ribose
incorporation between whole histone and histone HI.
iii. Effect of NAD+ Concentration and Histone
HI on ADP-ribose Incorporation
The following experiment was designed to examine the
effect of NAD+ and histone HI on ADP-ribose incorporation. 163
This was carried out in order to establish an appropriate
NAD+ concentration and range of histone HI concentrations to be employed in a study regarding the automodification site. Poly (ADP-ribose). polymerase (0.16 ug) was incubated in a final volume of 0.1 ml containing 100 mM Tris-HCl (pH
8.0) , 10 mM M g C ^ , 1 mM DTT, 50 yg/ml DNA, varying concen trations of Iadenine-2,8- 3H] NAD+ (211,000 dpm) and varying amounts of histone HI. The reaction was initiated by the addition of enzyme and terminated after one minute at 25°C with ice cold 10% TCA. The reaction mixture was analyzed using the Millipore filter assay as described in Methods.
As shown in Figure 36, ADP-ribose is incorporated into acid insoluble material even in the absence of added HI.
As will be discussed later, this is due to the automodifi cation reaction catalyzed by the polymerase. As in the case of whole histone, Figure 35, HI also inhibits the reac tion at higher concentrations. However, under the condi tions employed in this study, i.e. lower ionic strength and lower DNA concentration, the concentration of HI required for inhibition is much lower. The ionic strength has re cently been shown to affect ADP-ribose incorporation (6 5) and the DNA-HI ratio may also be important for determining the HI concentration range in which inhibition is observed. Figure 36. Effect of NAD+ Concentration and Histone HI on
ADP-ribose Incorporation
Poly(ADP-ribose) polymerase (0.16 yg) was incubated in a final volume of 0.1 ml containing 100 mM Tris-HCl (pH 8.0),
10 mM M g C ^ , 1 mM DTT, 50 y g/ml DNA, various concentrations of [adenine-2,8-3H] NAD+ , 2.436 yM (■----■ ), 24.36 yM
(O O) or 243.6 yM (•— —•) (211,000 dpm) and varying amounts of histone Hi. The reaction was initiated by ad ding enzyme and terminated after one minute at 25°C with ice cold 10% TCA. The reaction mixture was analyzed using the Millipore filter assay as described in Methods. ADP- ribose incorporated is expressed per minute of reaction.
164 iue 36 Figure ADP-RIBOSE INCORPORATED (pmol) 200 240 160 120 40 80 0 20 I (/xg/mL) HI 40 60 80
100 D-IOE NOPRTD (pmol) INCORPORATED ADP-RIBOSE H Ul CJ> 166
c. Automodification Site
Poly (ADP-ribose). polymerase has been shown to catalyze an automodification reaction in the absence of added accep tors (62-66). The following experiment was conducted to see if this was also the case for the polymerase purified by the procedure described in this report. Purified poly-
(ADP-ribose) polymerase (1.6 yg) was incubated in a solu tion containing 50 mM Tris-HCl (pH 8.0), 10 mM M g C ^ , 1 mM
DTT, 50 yg/ml DNA and 24.36 yM {adenine-2,8- 3H] NAD+ (10 yCi) in a final volume of 0.5 ml for 30 seconds at 25°C.
The reaction was initiated with enzyme and terminated with
0.3 ml of ice cold 10% TCA. After centrifuging and washing the poly(ADP-ribosyl)ated material as described in Methods, the precipitate was dissolved in 25 yl of 1.5 mM sodium phosphate (pH 7.2) in the presence of 3% SDS and incubated for 1.5 hours at 25°C. Then 1 yl of 0.05% bromophenol blue and 25 yl of 2% SDS, 2% mercaptoethanol and 8 M urea were added and the solution was applied to a 7.5% polyacrylamide gel (0.5 x 7.5 cm). Electrophoresis was carried out ac cording to Weber and Osborn (138) as described in Methods.
A gel with an equal amount of enzyme (1.6 yg) which had not undergone reaction was also run, stained with Coomassie blue, and scanned. The gel with the reacted enzyme was sliced, and counted for radioactivity. 167
As shown in Figure 37, the radioactivity and the pro tein comigrate. This has been seen by others (62-66) when the extent of automodification is not sufficiently great to alter the molecular weight. This is principally with short reaction times and low substrate concentrations.
Therefore, from these results it would appear that poly-
(ADP-ribose) polymerase as purified by the.described pro cedure catalyzes an automodification reaction in the ab sence of added acceptor protein.
i . Effect of the Acceptor Histone HI on the
Automodification Reaction. Average Chain
Length Analysis
In order to determine how the automodification reac tion is affected by adding acceptor proteins, the following experiment was performed. Varying amounts of HI (0, 5, 10,
30 and 70 yg/ml) were added to a reaction mixture described previously for the automodification reaction. After SDS-
PAGE, each gel was sliced and the radioactivity determined.
Since SDS-PAGE will separate poly(ADP-ribosyl)ated polymer ase from poly(ADP-ribosyl)ated HI, the amount of ADP-ribose incorporated into each can be quantified.
As shown in Figure 38, increasing the HI concentration causes an inhibition in the extent of automodification.
This has also been shown by Ferro and Olivera (65). Figure 37. SDS-Polyacrylamide Gel Electrophoresis of Poly-
(ADP-ribosyl)ated Material
Poly(ADP-ribose) polymerase (1.6 yg) was incubated in a solution containing 50 mM Tris-HCl (pH 8.0), 10 m M M g C l 2 ,
1 mM D T T , 50 yg/ml DNA and 24.36 yM Iadenine-2,8-3H] NAD+
(10 yCi) in a final volume of 0.5 ml for 30 seconds at
2 5 °C. The reaction was started by adding enzyme and ter minated with 0.3 ml of ice cold 25% TCA. The mixture was centrifuged at 15,000 x g for 20 minutes and the super natant fluid was removed. The precipitate was washed as described in Methods and then resuspended in 25 yl of 1.5 mM sodium phosphate (pH 7.2) containing 3% SDS and incu bated for 1.5 hours at 25°C. Bromophenol blue was added
(1 yl of 0.05%) plus 25 yl of 2% SDS, 2% mercaptoethanol, in 8 M urea and this was applied to a 7.5% polyacrylamide gel (0.5 x 7.5 cm) containing 0.1% SDS and electrophoresed according to Weber and Osborn (138) as described in Methods.
This gel was sliced and counted for radioactivity (•----• ) and another gel (1.6 yg enzyme) was stained, destained, and scanned for protein at 620 nm ( ) as described in
M e t h o d s . 168 169
DPM ,-3 620
4 .
3
0.2 2
0.1
0 . 0.0
0 10 20 30 SLICE No.
Figure 37 170
However, HI does not act as an acceptor under these condi tions, even though it has been shown (6 5,114) that HI is an acceptor of poly(ADP-ribose) polymerase under certain con ditions. Tanaka et al. (119) have reported that the poly merase reaction from purified bovine thymus is dependent on + 2 the histone only in the absence of Mg . In addition, the ionic strength of the buffer has been shown to play an im portant role in HI modification (65).
Since it can be seen in Figure 38 that the ADP-ribose incorporated into enzyme at 0 pg/ral HI (5.65 pmol) is greater than that incorporated with the addition of 30 pg/ml HI (3.58 pmol)., it was of interest to investigate the average chain length of the polymers associated with en zyme under these different conditions. Therefore the fol lowing experiment was designed: Three reactions were car ried out, two without HI and one with 30 pg/ml HI, and they will be referred to as I, II and III, respectively.
The reaction mixture consisted of purified poly(ADP-ribose) polymerase (1.6 pg) in a solution containing 50 mM Tris-HCl
(pH 8.0), 10 mM MgCl2, 1 mM DTT, 50 pg/ml DNA, 24.36 mM
[adenine-2, 8- 3H] NAD+ (10 pCi) and with or without HI, as indicated, in a final volume of 0.5 ml. The reaction was initiated with enzyme and incubated for 30 seconds at 25°C, followed by protein precipitation with 0.3 ml of ice cold
25% TCA. After centrifuging and washing the precipitate Figure 38. Effect of Histone HI on the Automodification
Reaction
Poly (ADP-ribose) polymerase (.1.6 yg) was incubated in a
solution containing 50 mM Tris-HCl (pH 8.0), 10 mM MgCl2,
1 mM DTT, 50 pg/ml DNA, 24.36 pM [adenine-2,8- 3H] NAD+ (10
pCi) and the indicated amounts of histone HI in a final
volume of 0.5 ml. The reaction was initiated with enzyme
and after 30 seconds at 259C was terminated with 0.3 ml of
ice cold 2 5% TCA. After the solution was centrifuged at
15,000 x g for 20 minutes, the supernatant fluid was re
moved and the precipitate was washed with ice cold 10% TCA
(2 x 1.0 ml) followed by 1.0 ml of diethyl ether. The washed precipitate was dissolved in 25 pi of 1.5 mM sodium
phosphate (pH 7.2) in 3% SDS and incubated for 1.5 hours at
25°C. After adding 25 pi of a solution containing 2% SDS,
2% mercaptoethanol, 8 M urea and 1 pi of 0.05% bromophenol
blue, the solution was applied to a 7.5% polyacrylamide gel
(0.5 x 7.5 cm). Electrophoresis was carried out according
to Weber and Osborn (138) as described in Methods. Gels
were frozen, sliced, and the radioactivity associated with
the polymerase (•—— •) and with HI (★--- ★ ) was determined
as in Methods. ADP-ribose incorporated is expressed per 30
seconds of reaction. 171 172 ADP-RIBOSE INCORPORATED ADP-RIBOSE INCORPORATED (pmol) CJl x \ |— 3
Figure 38 173 as described in Methods and resuspending in buffer as be fore, each mixture was applied to a 7.5% polyacrylamide gel
(0.5 x 7.5 cm) and electrophoresed according to Weber and
Osborn (13 8). The gel for reaction I was sliced and coun ted for radioactivity. From this profile of radioactivity versus slice number (not shown but similar to Figure 37), the radioactivity associated with the polymerase could be determined.
Gels for reactions II and III were sliced and the slices corresponding to the enzyme band for each gel were treated as follows to remove the polymer from the gel:
First, the slices were soaked in 25% isopropyl alcohol (IPA)
(3 x 1.0 ml), 30 minutes for each soak to remove most of the
SDS (139). After removing the final IPA wash, the slices were macerated with a glass rod and stirred for 3 hours in 1.0 ml of 1 N NH^OH to release the polymer from the protein and from the gel. Two subsequent NH^OH extractions were per formed, one for 6 hours and another overnight; the slices were filtered using a glass fritted funnel and washed with approximately 0.5 ml of 1 N NH^OH. Aliquots were taken
from the combined IPA washes, and the NH^OH washes and the radioactivity was determined by scintillation counting.
The macerated slices were digested with as ^escr;i-ked in Methods and then counted for radioactivity. 174
Of the total radioactivity in the IPA washes, NH^OH washes and remaining in the slices, 70% of the radioacti vity was in the NH^OH washes. This material was subjected to chain length analysis. In addition, for the gel of reaction I, 44.5% of the radioactivity applied to the gel was associated with the polymerase. Since reactions I and
II were identical and assuming the same percentage radio activity associated with the polymerase in II, then the radioactivity recovered in the IPA washes, NH^OH washes and that remaining in the slices for II accounts for 10 3% of the total radioactivity associated with the polymerase.
The average chain length analysis was performed on the
NH^OH washes of II and III as follows: The combined NH^OH washes were lyophilized and the residue dissolved in 75 yl of phosphodiesterase {0.04 units) in 50 mM Tris-HCl (pH
8.0), 10 mM MgCl2, and 2.5 mM AMP. The mixture was incu bated overnight at 37°C and then applied to filter paper for descending chromatography in solvent system II. Radio activity and average chain length calculations were deter mined as described in Methods.
As shown in Table 8, in the absence of HI the average chain length of enzyme associated polymer is 9.21, and with 30 yg/ml HI it is 2.94. Therefore a decrease in the average chain length by 6 8% is seen when 30 yg/ml HI is added. From Figure 38, 5.65 pmol and 3.58 pmol of 175
ADP-ribose are incorporated into the polymerase in the ab sence and presence of 30 yg/ml HI, respectively. In the absence of HI the total number of chains is 0.61 pmol and in the presence of 30 yg/ml HI the total number of chains is 1.2 pmol. From this it can be concluded that there are twice as many chains in the presence of HI than in the ab sence of HI. In addition, in the absence of HI there are
0.0 48 chains/mol of enzyme and with HI there are 0.09 5 chains/mol of enzyme. In other words, without HI, 1 in 20 molecules of polymerase are modified, but with Hi, 1 in 10 molecules of enzyme are modified. The role HI plays in affecting the automodification reaction is far from clear and although the extent of automodification is slight, this reaction could still be important to the function of poly-
(ADP-ribose) polymerase in vivo. 176
Table 8
Effect of Histone HI on the
Automodification Reaction
Total ADP-ribose Chain HI Chain Incorporated2 Number3 Chains/mol (yg/ml) Length1 (pmol)______(pmol) enzyme1*
0 9.21 5.65 0.61 0.048
30 2.94 3.58 1.22 0.095
determined as described in Methods.
2From Figure 38
3Total ADP-ribose incorporated/average chain length.
4Picomoles of enzyme were calculated from the enzyme used in
the reaction (1.5 yg, 12.8 pmol) using a molecular weight
of 125,000. IV. DISCUSSION
Poly(ADP-ribose) polymerase, a nuclear enzyme tightly
associated with chromatin, catalyzes a unique post-trans
lational modification of certain nuclear proteins. This modification involves the covalent attachment of a homo polymer of ADP-ribose units to these proteins. A substan tial body of evidence supports the idea that the modifica
tion is responsible for the regulation of many important
nuclear functions. A better understanding of this complex
enzyme in terms of its properties, function, and mechanism may provide some answers to the fundamental questions in volved in the regulation of nuclear metabolism.
In order for studies involving the mechanism of action of the poly(ADP-ribose) polymerase reaction to be carried out, it was important to obtain highly purified enzyme.
The reproducible purification procedure of poly(ADP-ribose) polymerase from calf thymus, which has been described in detail, results in yields of 13% and a homogeneity of 95%.
This is comparable to what has been reported for other preparations of poly(ADP-ribose) polymerase from calf and bovine thymus (68,100). A comparison of the properties of this enzyme preparation with those reported for other
177 178 preparations indicates that it is nearly identical to them.
As shown in Table 9, these properties are very similar to that of purified calf thymus polymerase previously reported
(see Table 2). The specific activity of the enzyme pre paration described in this dissertation is higher than that reported for different preparations from the same source.
This is probably due to either different assay conditions
[e.g. different DNA, (100)] or the retention of more acti vity after purification. Other major similarities include the dependence on added DNA for full activity, formation of a product with a branched structure, competitive inhi bition by nicotinamide, and the ability to catalyze the automodification reaction. In view of these similarities, it would appear that the use of this enzyme would be quite appropriate for conducting mechanistic studies where the use of purified enzyme is essential.
The catalytic mechanism of poly(ADP-ribose) polymerase appears to be quite complex. It must be recognized that the enzyme catalyzes three types of reactions. The first involves the initiation step of polymer formation. This reaction can be characterized by the formation of a cova lent bond between a protein acceptor and the C - l " carbon of the ADP-ribose moiety derived from NAD+ . The second involves the elongation of the polymer, and the third is the formation of branches in the polymer. The second and Table 9
Properties of Poly(ADP-ribose) Polymerase
from Calf Thymus Purified
as Described in This Study
Property Value
Molecular Weight 125,000
Isoelectric Point (pi) 9.4
Km for NAD+ 54 yM
V 1180 nmol min”"*" mg’ max 3 180 third reactions both involve the formation of a ribose- ribose bond between ADP-ribose residues already linked to
an acceptor protein and a new ADP-ribosyl group derived
from NAD+ . However, all three reactions have the common feature of the cleavage of the N-glycosidic bond of NAD+ with the release of nicotinamide and the transfer of the
ADP-ribose moiety to an acceptor.
Another facet of the complexity of the reaction is the number of acceptor substrates and effectors involved.
A number of different nuclear proteins can serve as accep tors, such as histones, A24 protein, the polymerase itself,
and others. Also, the linkage of ADP-ribose to these
acceptors may be different. In addition, the polymer
length as well as the degree of branching can vary. Be
cause of these variations and their effect on the initia tion, elongation and branching steps, the investigation of the mechanism of action of poly(ADP-ribose) polymerase in vitro has been difficult. However, attempts have been made to simplify the in vitro systems (65) by using carefully defined reaction conditions such as the use of one histone
as an acceptor.
The overall stereochemical course of the reaction provides some insight into the catalytic mechanism. The
N-glycosidic bond in NAD+ which is cleaved has a B-confi- guration. In terms of the initiation reaction, the 1 8 1 stereochemistry of the ADP-ribose-acceptor bond is not known; however, for elongation and branching steps, the linkage of the respective ribose-ribose bonds have a- configurations (73-75,82). The overall inversion of con figuration could result from the formation of an inter mediate carbonium ion, or by a single SN2 displacement (or an odd number of displacements).
Compounds which mimic a transition state in biochemi cal reactions, called transition state analogs (97,98), can be employed to aid in distinguishing between these two pathways. For example, results from experiments that uti lized transition state analogs containing a lactone func- 2 tional group at the anomeric carbon resulting in sp hybri dization (planar), have been used as evidence to support a carbonium ion mechanism for the enzymes glycogen phosphory- lase (95) and lysozyme (96) . These analogs have been shown to be powerful inhibitors of their respective enzymes. In light of these findings, a potential transition state ana- 2 log, ADP-ribonolactone (Figure 7), which is also sp hybri dized (planar) at the anomeric carbon, was examined. The lack of inhibition as compared to approximately the same concentration of ADP-ribose suggests that a carbonium ion is not involved in the catalytic mechanism and that a direct displacement may be involved. In contrast, ADP- ribonolactone has been shown to have an affinity 9 times 182
greater for the active site of calf spleen NAD+ glycohydro-
lase relative to ADP-ribose, which supports the postulated
intermediary ADP-ribosyl oxocarbonium ion (132) in that
reaction.
In addition to the interaction of ADP-ribonolactone
with poly(ADP-ribose) polymerase, other substrate deriva
tives have been examined to assess other binding inter
actions. It was shown that nicotinamide and NADH are very
potent inhibitors of the polymerase at concentrations of
approximately 0.1 mM and that nicotinamide is competitive with respect to NAD+. NADH has also been shown to be a
competitive inhibitor and the reported for NADH (55 yM)
(68) is equal to the for the substrate, NAD+. This
suggests that the positive charge on the pyridinium nitro
gen of the nicotinamide ring is not important for binding.
Other inhibitors which bear similar structural features to
nicotinamide, such as substituted benzamides and thymidine
analogs, have been tabulated by Ueda and associates (158).
These studies are consistent with the idea that the nico
tinamide ring of the substrate NAD+ interacts quite
strongly with the enzyme.
On the other hand, substrate derivatives such as AMP,
ADP and ADP-ribose inhibit the polymerase, but net nearly
to the same extent as does the same concentration of nico
tinamide or NADH (see Table 7). This has been observed by 183 others (101,113,159) and suggests that the adenine ring of the substrate does not interact strongly with the polymer ase.
Other NAD+ requiring enzymes whose crystal structures are known have been examined quite thoroughly in regard to their NAD+ binding site (160). From studies with NAD+ dependent dehydrogenases, a number of common features for the NAD+ binding domain have been shown to exist. Such major features are that the coenzyme binding domain con tains parallel stranded s-sheets of nearly identical topo logy, NAD+ is bound in an extended configuration, the binding pocket for the adenine ring is not highly specific, and the nicotinamide ring is bound in a cavity which is hydrophobic on one side and hydrophilic on the other, depending on which side interacts with the substrate (161).
More specifically, in regard to the NAD+ binding domain of liver alcohol dehydrogenase, the ADP-ribose part of the coenzyme is quite rigidly fixed in the active site and acts as an anchor for the nicotinamide moiety, which can change its orientation within the freedom of motion pro vided by the glycosyl linkage (162).
Although the NAD+ binding domain is characteristic of enzymes using NAD+ in redox reactions, the binding to
Blue Sepharose can be used to predict the presence of such a binding domain in other enzymes (150). Since 184 poly(ADP-ribose) polymerase binds to Blue Sepharose, it is possible that the NAD+ binding site possesses similar features to the well-studied NAD+ binding domains for de hydrogenases. On the other hand/ in contrast to liver alcohol dehydrogenase/ it appears that the nicotinamide moiety of poly(ADP-ribose) polymerase interacts much more strongly with the enzyme than does the ADP-ribose part of the substrate, NAD+ .
Since poly(ADP-ribose) polymerase catalyzes the trans fer of an ADP-ribose moiety to acceptors, it might be expected that the nicotinamide moiety would serve to anchor
NAD+ to the enzyme, whereas the ADP-ribose moiety may interact less specifically so as to facilitate its transfer to an acceptor. The data obtained so far suggests that this may be so, however before any definite conclusions can be made regarding the NAD+ binding domain of poly(ADP- ribose) polymerase, more investigations are needed.
In addition to interacting with the substrate, NAD+, poly(ADP-ribose) polymerase must interact with a number of other molecules in order for catalysis to occur. For example, the enzyme requires DNA for activity, and thus can be thought of as an activator. Also, proteins such as histones are modified by the enzyme with the covalent at tachment of a homopolymer of ADP-ribose units. The enzyme itself undergoes a similar modification in a reaction 185 designated automodification. Therefore/ along with the
NAD+ binding site, the polymerase possesses other binding sites. These include the following: the activator binding site where DNA interacts with the polymerase, which is dis tinct from the NAD+ binding site; the acceptor binding site where histones interact with the enzyme, which is distinct from the substrate and activator binding sites; and the autopolymerization site where the polymerase itself accepts poly(ADP-ribose), which is also distinct from the other binding sites. Recently, studies have been carried out by
Ferro and Olivera (65) and Nishikimi et al. (163) with purified poly(ADP-ribose) polymerase which have supported the distinct nature of the activator and acceptor binding sites and the activator and autopolymerization binding sites, respectively.
In addition to histones, poly(ADP-ribose) polymerase can utilize other nuclear proteins as acceptors (2- 10) and more recently it has been shown that low molecular weight nucleotides can serve in this capacity (164,165). With regard to histones, using chromatin from rat liver, both core histones and histone HI were found to be poly(ADP- ribosyl)ated. These modifications are near the polar regions of the histones, which strongly interact with DNA
(15,53,57), and may serve to alter such interactions.
Early investigators who utilized purified poly(ADP-ribose) 186 polymerase regarded 'histones as allosteric effectors rather
.than acceptors (107). However, changing conditions, e.g. greater histone and DNA concentrations, led to the conclu sion that histones were acceptors. Yoshihara and co-workers
(119) have reported a histone-dependent reaction only in + 2 the absence of Mg and have evaluated the effects of various histones on this reaction. Other investigators have not detected a difference in histone being an acceptor with or without Mg+2 (65).
Results have been presented in this study which show that ADP-ribose incorporation is quite different in the presence and absence of added histones. With added whole histone, which is a mixture of histones, there is a stimu lation of ADP-ribose incorporation seen at a concentration of 75 jjg/ml with an equal amount of DNA. At higher levels, however, the amount of ADP-ribose incorporated appears to be inhibited. In the absence of added whole histone, no stimulation is observed. Under these latter conditions the ADP-ribose incorporation is most likely on the enzyme itself, i.e. automodification. The extent of this incor poration is less than observed with added acceptor. This suggests that under these conditions, whole histone is a better acceptor than the polymerase itself.
It has also been shown by Yoshihara and co-workers
(119) that histone HI causes a stimulation of ADP-ribose + 2 incorporation with increasing DNA concentration when Mg was omitted from the reaction. However, in the present study, a stimulation of ADP-ribose incorporation was ob served with increasing DNA concentration with whole his- + 2 tone, but not with histone HI in the presence of Mg
Also, higher ADP-ribose incorporation was observed with whole histones than with histone HI. It is not known whether the increase in ADP-ribose incorporation is due to whole histone serving as a better acceptor or whether whole histone stimulates the enzyme itself into becoming a better acceptorc This difference is probably due to differences in protein composition of the two histone fractions or other reaction conditions and therefore the results ob served here cannot be compared to that of Yoshihara and co-workers (119).
In regard to the recent finding that low molecular weight nucleotides such as diadenosine, 5 % 5 ' ' '-P1, P4- tetraphosphate (Ap4A) can serve as acceptors, new questions have been raised concerning the acceptor site. This reac tion is analogous to the elongation or branching steps catalyzed by the polymerase because a ribose-ribose bond has been shown to form between Ap4A and the ADP-ribose moiety derived from NAD+ (164). Since the modification of
Ap4A requires histone HI, it has been suggested that his tone HI is a cofactor in this type of reaction (165). 188
However, since it has been stated that Ap4A binds to his tone HI (164), poly(ADP-ribose) polymerase may recognize
Ap4A as being an ADP-ribose residue which is attached to the histone.
Because of the variety of acceptors for poly(ADP- ribose) polymerase, the acceptor binding site is probably of broad specificity. Initially it may be thought that the acceptor may bind to DNA in order to provide the proper interaction between the enzyme and acceptor for reaction.
However, using [125I] histone HI, Ferro and Olivera (65) have shown that HI molecules not bound to DNA were poly-
(ADP-ribosyl)ated in preference to DNA-bound HI.
A number of studies have been carried out with puri fied poly(ADP-ribose) polymerase in efforts to examine the automodification reaction, and the effect of acceptors such as histone HI on it (63-65,121). It has been shown in this study that the automodification reaction is inhi bited by increasing levels of histone HI and that histone
HI is not an effective acceptor. A possible reason that histone HI was not a good acceptor was the fact that the amount of DNA (50 pg/ml) in the reaction may have been enough to complex all the HI. Ferro and Olivera (65) have suggested that histone HI bound to DNA was poly(ADP-ribo- syl)ated only slowly and since the reactions employed in the present study were only 30 seconds, it might explain 189 this anomaly. The inhibition of poly(ADP-ribose) ribosyla- tion of enzyme caused by HI was also observed by Ferro and
Olivera (65), however under the conditions they used, his tone HI was an acceptor.
The nature of this inhibition was further examined in this study by determining if there was any effect on the average chain length of the polymer associated with the enzyme. It was shown that the average chain length de creased by 68% when 30 pg/ml histone Hi was added as com pared to when it was omitted. On the other hand, it was shown that the number of chains doubled. The extent of modification was limited under the conditions employed, which yielded 0.048 and 0.095 chains/enzyme molecule in the absence and presence of 30 pg/ml, respectively. This increase in the presence of histone HI may be the result of new sites on the enzyme being exposed. Histone HI may cause a conformational change which is responsible for the increased availability of the sites. Another possibility may be the fact that termination occurs prematurely before the polymer reaches its usual length, thus decreasing the average chain length. This may be followed by initiation at a new site.
No direct evidence has been given as to whether the automodification site (polymer site) is a distinct site from the catalytic site, but preliminary studies by 190
Hayaishi and co-workers (63,122) "have suggested that they are distinct. They have also suggested that the automodi fication reaction is at least in part intermolecular. This would seem likely since polymer growth occurs at the AMP end of a chain which may be able to serve as an acceptor for another enzyme molecule.
As previously mentioned, poly(ADP-ribose) polymerase requires double stranded DNA for activity. Studies have shown that strand breaks in the DNA stimulate enzyme acti vity (25-27,30,31) and that DNA cannot be replaced by other polyanions with purified enzyme (9,68). DNA which is associated with the enzyme and isolated during the puri fication, designated "active DNA," has been shown to have a high affinity for the enzyme, which probably results in its unusual ability to highly activate the enzyme (166).
The "active DNA" associated with the polymerase from calf thymus has been estimated to have an average site of about
100-200 base pairs (101). More recently, the DNA binding domain has been separated from the automodification domain using partial papain digestion followed by DNA-cellulose chromatography (163). It has been proposed that the DNA functions in two capacities. One is to position the en zyme on specific binding sites such as single- or double stranded breaks on the DNA, and the other is to induce an active conformation of the enzyme (166) . This may be due 191 to the fact that poly(ADP-ribose) polymerase is a basic protein and the acceptors are also basic proteins. The requirement of DNA may be to aid in the neutralization of charges between the two proteins.
With the data presented in this study as well as that by others [see reviews (2-10)], a model for the binding sites of the polymerase can be formulated. These binding sites are the substrate, acceptor, activator and automodi fication binding sites where NAD+ , histone, DNA and polymer interact, respectively. Within the NAD+ binding site is the catalytic site which contains groups on the enzyme that participate in N-glycosidic bond cleavage, and in the transfer of the ADP-ribose moiety to the acceptor or auto modification site. Presumably, at both the acceptor and polymer sites the two main types of reactions can occur, namely initiation (ADP-ribose bond formation to protein) and elongation (glycosidic bond formation). The substrate and acceptor sites are probably in close proximity to each other in order for efficient transfer of ADP-ribose resi dues to occur. Since the polymerase has been shown to be an acceptor, a separate site for polymer attachment pro bably exists. It is not known whether this automodifica tion involves an intramolecular or intermolecular reaction; preliminary evidence has suggested it may be intermolecular.
The purpose of this automodification may be an 192
auto-regulation phenomenon whereby the polymerase controls its own activity and under certain conditions modifies it
self before (or instead of) modifying acceptors.
Poly(ADP-ribosyl)ation is a type of post-translational modification of histones similar to acetylation, methyla- tion and phosphorylation. These latter modifications/ how ever, would appear to result in minor structural modifica tions as compared to poly(ADP-ribosyl)ation. This struc tural modification of chromatin may be the key step in the
regulation of certain nuclear functions. This regulation has been implicated as being like other interconvertible
enzyme cascades (167).
Data has been presented here regarding the character
istics of the enzyme responsible for this modification, poly(ADP-ribose) polymerase. A better understanding of this enzyme in vitro will aid in the understanding of
such important nuclear functions as DNA repair, DNA syn thesis, cellular differentiation and others to which this
enzyme has been linked. V. BIBLIOGRAPHY
1 . Chambon, P., Weill, J. D. and Mandel, P. (1963) Biochem. Biophys. Res. Commun. 11, 39.
2. Sugimura, T. (1973) Prog. Nucleic Acid Res. Mol. Biol. 13, 127.
3. Sugimura, T., Miwa, M., Kanai, Y., Oda, K., Segawa, K., Tanaka, M. and Sakura, H. (1976) Prog. Cancer Res. Ther. 1, 231.
4. Hilz, H. and Stone, P. (1976) Rev. Physiol. Biochem. Pharmacol. 16_, 1.
5. Hayaishi, O. and Ueda, K. (19 77) Annu. Rev. Biochem. 46_, 95.
6 . Purnell, M. R., Stone, P. R. and Whish, W. J. D. (1980) Biochem. S o c . Trans. £, 215.
7. Sugimura, T., Miwa, M., Saito, H., Kanai, Y., Ikejima, M . , Terada, M . , Yamada, M. and Utakoji, T. (1980) Adv. Enzyme Regul. lj^, 195.
8 . Mandel, P., Okazaki, H. and Niedergang, C. (19 82) Prog. Nucleic Acid Res. Mol. Biol. 2_7, 1.
9. Pekala, P. H. and Anderson, B. M. (1982) in "The Pyridine Nucleotide Coenzymes," (J. Everse, B. Anderson and K. You, eds.) Academic Press, New York, pp. 325-377.
10. Hayaishi, O. and Ueda, K., eds. (1982) "ADP-ribosyla- tion Reactions," Academic Press, New York.
11. Harris, M . , ed. (1973) "Poly(ADP-ribose), International Symposium," Fogerty Int. Cent. Proc. No. 26, Washington, D.C.
12. "Proceedings of the Seminar on Poly(ADP-ribose) and ADP-ribosylations of Proteins," (1975) J. Biochem. (Tokyo) 77, lp.
193 194
13. Hilz, H., ed. (1976) "Proceedings of the 4th Inter national Symposium on Poly(ADP-ribose) and ADP- ribosylation of Proteins," Walter de Gruyter, Berlin.
14. Smulson, M. and Sugimura, T . , eds. (1980) "Proceedings of the 5th International Conference on Novel ADP-ribosylations of Regulatory Enzymes and Proteins," Elsevier/North Holland, Amsterdam.
15. Ueda, K., Hayaishi, O., Kawaichi, M., Ogata, N., Ikai, K . , Oka, J. and Okayama, H. (1979) in "Modulation of Protein Function," (D. Atkinson and C. Fox, eds.) Academic Press, New York, p. 47.
16. Hilz, H., Adamietz, P., Bredehorst, R. and Wielckens, K. (1979) Adv. Enzyme Regul. 11_, 195.
17. Smulson, M. (1979) Trends Biochem. Sci. 4_, 225.
18. Burzio, L. and Koide, S. S. (1970) Biochem. Biophys. Res. Commun. 4J), 1013.
19. Burzio, L., Reich, L. and Koide, S. S. (1975) P r o c . Soc. Exptl. Biol. Med. 149, 933.
20. Berger, N. A., Weber, G. and Kaichi, A. S. (1978) Biochim. Biophys. Acta. 519, 87.
2 1 . Berger, N. A., Weber, G., Kaichi, A. S. and Petzold, S. J. (1978) Biochim. Biophys. Acta. 5 1 9 , 105.
22 . Berger, N. A., Adams, J. W., Sikorski, G. W., Petzold, S. J. and Shearer, W. T. (1978) J. Clin. Invest. 62, 111.
23. Muller, W. E. G. and Zahn, R. K. (1976) Mol. Cell. Biochem. 1_2, 14 7.
24. Mullins, D. W., G i r i , C. P. and Smulson, M. E. (1977) Biochemistry 16_, 506.
25. Smulson, M. E., Schein, P., Mullins, D. W. and Sudhakar, S. (1977) Cancer Res. 37_, 3006.
26. Berger, N. A., Sikorski, G. W., Petzold, S. J. and Kurohara, K. K. (1979) J. Clin. Invest. 63, 1164. 195
27. Durkacz, B. W. , Omidiji, 0., Gray, D. A. and Shall, S. (1980) Nature 283, 593.
28. Berger, N. A., Sikorski, G. W. and Petzold, S. J. (1980) in "Novel ADP-ribosylations of Regulatory Enzymes and Proteins," (M. E. Smulson and T. Sugimura, eds.) Elsevier/North Holland, Amsterdam, pp. 185-195.
29. Berger, N. A., Sikorski, G. W., Petzold, S. J. and Kurohara, K. K. (1980) Biochemistry 19^ 289.
30. Benjamin, R. C. and Gill, D. M. (1980) J. Biol. Chem. 255, 10493.
31. Jacobson, E. L., Antol, K. M., Juarez-Salinas, H. and Jacobson, M. K. (1983) J. Biol. Chem. 258, 103.
32. Caplan, A. I. and Rosenberg, M. J. (1975) P r o c . Natl. Acad. Sci. U.S.A. 12_, 1852.
33. Caplan, A. I., Niedergang, C., Okazaki, H. and Mandel, P. (1979) Dev. Biol. 72, 102.
34. Caplan, A. I. (1980) in "Novel ADP-ribosylations of Regulatory Enzymes and Proteins," (M. E. Smulson and T. Sugimura, eds.) Elsevier/North Holland, Amsterdam, pp. 315-324.
35. Miwa, M., Oda, K., Segawa, K. Tanaka, M., Irie, S., Yamaguchi, N., Kuchino, T., Shirski, K., Shimojo, H., Sakura, H., Matsushima, T. and Sugimura, T. (1977) Arch. Biochem. Biophys. 181, 313.
36. Rastl, E. and Swetly, P. (1978) J. Biol. Chem. 253, 4333.
37. Farzaneh, F. and Pearson, C. (1979) Dev. Biol. 72^, 254.
38. Kornberg, R. D. (1977) Annu. Rev. Biochem. 4£, 931.
39. McGhee, J. and Felsenfeld, G. (1980) Annu. Rev. Biochem. 49^, 1115.
40. Igo-Kemenes, T . , H o r z , W. and Zachen, H. G. (1982) Annu. Rev. Biochem. 51_, 89.
41. Giri, C. P., West, M. H. P., Ramirez, M. L. and Smulson, M. (1978) Biochemistry 17, 3501. 196
42. Okayama, H. and Hayaishi, 0. (1978) Biochem. Biophys. Res. Commun. 84^, 755.
43. Yoshihara, K., Tanigawa, Y., Burzio, L. O. and Koide, S. S. (1975) Proc. Natl. Acad. S c i . U.S.A. 7 2 , 289.
44. Suhadolnik, R. J. , Baur, R., Lichtenwalner, D. M . , Uematsu, T., Roberts, J. H. Sudhakar, S. and Smulson, M. (1977) J. Biol. Chem. 252, 4134.
45. Lehman, I. R. (1974) in "The Enzymes," 3rd Ed., Vol. 10 (P. D. Boyer, ed.) Academic Press, New York, pp. 237-259.
46. Payne, J. F. and B a l , A. K. (1976) Exp. Cell. Res. 9 9 , 428.
47. Brightwell, M. and Shall, S. (19 71) Biochem. J. 125, 67p.
48. Zatman, L. J., Kaplan, N. 0. and Colowick, S. P. (1953) J. Biol. Chem. 200, 197.
49. Nishizuka, Y., Ueda, K., Honjo, T. and Hayaishi, O. (1968) J. Biol. Chem. 243, 3765.
50. Ballal, N. R. and Busch, H. (1973) Cancer Res. 3 3 , 2737.
51. Ballal, N. R., Kang, Y. J., Olsen, M. 0. and Busch, H. (1975) J. Bioi. Chem. 250, 5921.
52. Stone, P., Lorimer, W. S. and Kidwell, W. R. (1977) Eur. J. Biochem. 81_, 9.
53. Ogata, N., Ueda, K., Kagamiyama, H. and Hayaishi, O. (1980) J. Biol. Chem. 255, 7616.
54. Nishizuka, Y., Ueda, K., Yoshihara, K., Yoshihara, H., Takeda, M. and Hayaishi, O. (1969) Cold Spring Harbor Symp. Quant. Biol. 3A_, 7 81.
55. Ueda, K . , Narumiya, S. and Hayaishi, O. (1972) Hoppe- Seyler's Z. Physiol. Chem. 353, 846.
56. Adamietz, P., Bredehorst, R. and Hilz, H. (1978) Eur. J. Biochem. 91, 317. 197
57. Ogata, N . , Ueda, K. and Hayaishi, O. (1980) J. Biol. Chem. 255, 7610.
58. Dixon, G. H., Wong, N. and Poirier, G. G. (1976) Fed. P r o c . 35^, 1623.
59. Riquelme, P. T., Burzio, L. 0. and Koide, S. S. (1979) J. Biol. Chem. 254, 3018.
60. Burzio, L. 0., Riquelme, P. T. and Koide, S. S. (1979) J. Biol. Chem. 254, 3029.
61. Smith, J. A. and Stocken, L. A. (1975) Biochem. J. 14 7 , 523.
62. Yoshihara, K., Hashida, T., Yoshihara, H., Tanaka, Y. and Ohgushi, H. (1977) Biochem. Biophys. Res. Commun. 7_8, 1281.
63. Kawaichi, M . , Ueda, K. and Hayaishi, 0. (1981) J. Biol. Chem. 256, 9483.
64. Caplan, A. I., Niedergang, C., Okazaki, H. and Mandel, P. (1979) Arch. Biochem. Biophys. 19 8 , 60.
65. Ferro, A. M. and Olivera, B. M. (1982) J. Biol. Chem. 257 , 7808.
66 . Jump, D. B. and Smulson, M. (1980) Biochemistry 1 9 , 1024.
67. Jump, D. B., Butt, T. R. and Smulson, M. (1980) Bio chemistry 19_, 1031.
68. Ito, S., Shizuta, Y. and Hayaishi, O. (1979) J. Biol. Chem. 254. 3647.
69. Chiu, Y. S. and T a o , M. (1978) J. Biol. Chem. 2 5 3 , 7145 .
70. Doly, J. and Petek, F. (1966) C. R. Hebd. Seances Acad. Sci. Ser. D. 263, 1341.
71. Chambon, P., Weill, J. D., Doly, J., Strosser, M. T. and Mandel, P. (196 6) Biochem. Biophys. Res. Commun. 25_, 6 38.
72. Doly, J., Meihac, M., Chambon, P. and Mandel, P. (1972) Hoppe-Seyler's Z. Physiol. Chem. 3 5 3 , 843. 198
73. Miwa, M., Saito, H., Sakura, K., Saikawa, N., Watanabe, F., Matsushima, T. and Sugimura, T. (1977) Nucleic Acids Res. £, 3997.
74. Inagaki, F., Miyazawa, T., Miwa, M., Saito, H. and Sugimura, T. (1978) Biochem. Biophys. Res. Commun. 85, 415.
75. Ferro, A. M. and Oppenheimer, N. J. (1978) P r o c . Natl. Acad. Sci. U.S.A. 75^, 809.
76. Oppenheimer, N. J., Singh, M., Sweeley, C. C., Sung, S. J. and Srere, P. A. (1979) J. Biol. Chem. 254, 1000.
77. Tanaka, M., Hayashi, K., Sakura, H., Miwa, M., Matsushima, T. and Sugimura, T. (1978) Nucleic Acids Res. 5_, 3183.
78. Miwa, M., Saikawa, N., Yamaizumi, Z., Nishimura, S. and Sugimura, T. (1979) Proc. Natl. Acad. Sci. U.S.A. 76., 595.
79. Minaga, T. and Kun, E. (1983) J. Biol. Chem. 2 5 8 , 725.
80. Miwa, M. and Sugimura, T. (1971) J. Biol. Chem. 246, 6362 .
81. Miwa, M., Nakatsugawa, K., Hara, K., Matsushima, T. and Sugimura, T. (1975) Arch. Biochem. Biophys. 1 6 7 , 54.
82. Miwa, M., Ishihara, M., Takishima, S., Takasuka, N., Maeda, M., Yamarzumi, Z., Sugimura, T., Yokoyama, S. and Miyazawa, T. (1981) J. Biol. Chem. 256, 2916 .
83. F u t a i , M. and Mizuno, D. (1967) J. Biol. Chem. 242, 5301.
84. Ferro, A. M. and Kun, E. (1976) Fed. Proc. 3_5, 1541.
85. Futai, M . , Mizuno, D. and Sugimura, T. (1968) J. Biol. Chem. 243, 6325.
86. Okayama, H . , Honda, M. and Hayaishi, O. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 2254. 199
87. Koshland, D. E., Jr. (1953) Biol. Revs. Cambridge Phil. Soc. 28^ 416.
88. Koshland, D. E., Jr. (1959) in "The Enzymes," 2nd Ed., Vol. 1 (P. D. Boyer, H. Lardy and K. Myrback, eds.) Academic Press, New York, pp. 305-346.
89. Bell, R. M. and Koshland, D. E., Jr. (1971) Science 172, 1253.
90. Spector, L. B. (1982) "Covalent Catalysis by Enzymes," Springer-Verlag, New York.
91. Philips, D. C. (1967) Proc. Natl. Acad. Sci. U.S.A. 57, 484.
92. Dahlquist, F. W., Rand-Meir, T. and Raftery, M. A. (1969) Biochemistry 8^, 4214.
93. Bull, H. G., Ferraz, J. P., Cordes, E. H., Ribbi, A. and Apitz-Castro, R. (1978) J. Biol. Chem. 253, 5186 .
94. Mayer, F. C. and Larner, J. (1958) J. Am. Chem. Soc. 81, 188.
95. Graves, D. J. and Wang, J. H. (1972) in "The Enzymes," 3rd Ed., Vol. 7 (P. D. Boyer, ed.) Academic Press, New York, pp. 435-482.
96. Secemski, I. I., Lehrer, S. S. and Lienhard, G. E. (1972) J. Biol. Chem. 247, 4740.
97. Wolfenden, R. (1972) A c c t s . Chem. Res. 5_, 10.
98. Wolfenden, R. (1976) Annu. Rev. Biophys. Bioeng. 5^, 271.
99. Coward, J. K. (1978) in "Transition States in Bio chemical Processes," (R. D. Gandour and R. L. Schowen, eds.) Plenum Press, New York, pp. 579- 591.
100. Yoshihara, K. , Hashida, T., Tanaka, Y., Ohgushi, H., Yoshihara, H. and Kamiya, T. (1978) J. Biol. Chem. 253, 6459.
101. Niedergang, C., Okazaki, H. and Mandel, P. (19 79) Eur. J. Biochem. 102, 43. 200
102. Mandel, P., Okazaki, H. and Niedergang, C. (1977) FEBS Lett. 84^, 331.
103 . Tsopanakis, C., McLaren, E., Tsopanakis, A. and Shall, S. (1977) Biochem. Soc. Trans. 5,. 932.
104 . Tsopanakis, C. Leeson, E., Tsopanakis, A. and Shall, S. (1978) Eur. J. Biochem. 90, 337.
105. Petzold, S. J., Booth, B. A., Leimbach, G. A. and Berger, N. A. (1981) Biochemistry 2_0, 7075.
106 . Carter, S. G. and Berger, N. A. (19 82) Biochemistry 21, 5475.
107 . Okayama, H., Edson, C. M., Fukushima, M., Ueda, K. and Hayaishi, O. (1977) J. Biol. Chem. 2 5 2 , 7000.
108 . Farina, B., Rosario, M., Mennella, F. and Leone, E. (1979) M a c r o m o l . Fun c t . Cell. Proc. Sov.-Ital. Symp. 28 3.
109. Agemori, M., Kagamiyama, H., Nishikimi, M. and Shizuta, Y. (1982) Arch. Biochem. Biophys. 2 1 5 , 621.
110 . Holtlund, J., Kristensen, T. and Sletten, K. (1980) Biochem. J. 185, 779.
111. Kristensen, T. and Holtlund, J. (1976) Eur. J. Bio chem. 10_, 441.
112. Ellison, R. P. (1978) J. Cell. Biol. 79^, 131a.
113 . Holtlund, J., Kristensen, T., Ostvold, A.-C. and Laland, S. G. (1981) Eur. J. Biochem. 119, 23.
114. Kawaichi, M . , Ueda, K. and Hayaishi, O. (1980) J. Biol. Chem. 255, 816.
115. Jongstra-Bilen, J., Ittel, M. E., Jongstra, J. and Mandel, P. (1981) Biochem. Biophys. Res. Commun. 103, 383.
116 . Kristensen, T. and Holtlund, J. (1978) Eur. J. Bio chem. 88 , 4 95. 201
117. Hashida, T., Ohgushi, H. and Yoshihara, K. (1979) Biochem. Biophys. Res. Commun. 8_8 , 305.
118. Benjamin, R. C. and Gill, D. M. (1980) J. Biol. Chem. 255, 10502.
119. Tanaka, Y., Hashida, T., Yoshihara, H. and Yoshihara, K. (1979) J. Biol. Chem. 254, 12433.
120. Okazaki, H., Niedergang, C., Couppez, M. , Martinage, A., Sautiere, P. and Mandel, P. (1980) FEBS Lett. 110, 227.
121. Okazaki, H., Niedergang, C. and Mandel, P. (1980) Biochemie 6_2, 147.
122. Ueda, K., Kawaichi, M., Oka, J. and Hayaishi, O. (1980) in "Novel ADP-ribosylations of Regulatory Enzymes and Proteins," (M. E. Smulson and T. Sugimura, eds.) Elsevier/North Holland, Amster dam, pp. 47-57.
123. Yoshihara, K., Hashida, T., Tanaka, Y., Matsunami, N., Yamaguchi, A. and Kami y a , T. (1981) J. Biol. Chem. 256, 34 71.
124 . Ueda, K., Kawaichi, M., Okayama, H. and Hayaishi, O. (1979) J. Biol. Chem. 254, 679.
125. Kun, E . , Chang, A. C. Y., S h a r m a , M. L., Ferro, A. M. and Nitecki, D. (1976) Proc. Natl. Acad. Sci. U.S.A. 7_3, 3131.
126 . Setlow, R. B., Regan, J. D., German, J. and Carrier, W. L. (1969) Proc. Natl. Acad. Sci. U.S.A. 6 4 , 1035.
127 . Cleaver, J. E. (1973) Cancer Res. 3_3 , 362.
128. Caplan, A. I. and Stoolmiller, A. C. (1973) Proc. Natl. Acad. Sci. U.S A. 70^, 1713.
129. Schaller, H., Nusslein, C Bonhoeffer, F. J., Kurz, C. and Nietzschmann, I. (1972) Eur. J. Biochem. 26, 474.
130 . Bendich, A. J. and Bolton E. T. (196 8) Methods Enzymol. 12B, 6 35. 202
131. Umansky, S. R., Kovalev, Y. I. and Tokarskaya, V. I. (1975) Biochim. Biophys. Acta. 3 8 3 , 242.
132. Schuber, F. and Pascal, M. (1977) FEBS Lett. 7_3, 92.
133. Dunaway-Mariano, D. and Cleland, W. W. (1980) Bio chemistry 19_, 1496.
134. Nelson, N. (1944) J. Biol. Chem. 1 5 3 , 375.
135 . Couling, T. E. and Goodey, R. (1970) Biochem. J. 1 1 9 , 303.
136 . Shizuta, Y., Ito, S., Nakata, K. and Hayaishi, 0. (1980) Methods Enzymol. 66_, 159.
137 . Bradford, M. M. (1976) Anal. Biochem. 17^, 248.
138. Weber, K. and Osborn, M. (1969) J. Biol. Chem. 2 4 4 , 4406 .
139. Fairbanks, G., Steck, T. L. and Wallach, D. F. H. (1971) Biochemistry 10^, 2606 .
140. Reisner, A. H., Nemes, P. and Bucholtz, C. (1975) Anal. Biochem. 64^ 509 .
141. Wray, W., Boulikas, T., Wray, V. P. and Hancock, R. (1981) Anal. Biochem. 118, 197.
142. Panyim, S. and Chalkley, R. (1969) Arch. Biochem. Biophys. 130, 337.
143. Bonner, W. M. , West, M. H. P. and Stedman, J. D. (1980) Eur. J. Biochem. 109, 17.
144 . Righetti, P. and Drysdale, J. W. (1971) Biochim. Biophys. Acta. 236, 17.
145. Yamada, M. and Sugimura, T. (1971) Biochemistry 1 2 , 3303.
146 . Bensadoun, A. and Weinstein, D. (1976) Anal. Biochem. 7£, 241.
147 . Pabst Laboratories (1956) Circular No. OR-10, p. 21.
148. Ames, B. and Dublin, D. T. (1960) J. Biol. Chem. 2 3 5 , 769. 203
149. "Affinity Chromatography, Principles and Methods," (1979) Pharmacia Fine Chemicals, Uppsala, Sweden, p. 74.
150. Stellwagen, E. (1977) Acct. Chem. Res. ]j0, 92.
151. Oakley, B. R., Kirsch, D. R. and Morris, N. R. (1980) Anal. Biochem. 105, 361.
152. Parnaik, Veena K. (1979) Ph.D. Thesis, The Ohio State University.
153. Frank, R. N. and Rodbard, D. (1975) Arch. Biochem. Biophys. 171, 1.
154 . Dunker, A. K. and Rueckert, R. R. (1969) J. Biol. Chem. 244, 5074.
155. Ferguson, K. A. (1964) Metabolism 1J3 , 985.
156 . Boehringer Mannheim, Combithek®, Cat. No. 161365.
157. Lineweaver, H. and Burk, D. L. (19 34) J. Am. Chem. S o c . 5 6 , 6 5 8.
158 . Ueda, K . , Kawaichi, M. and Hayaishi, O. (1982) in "ADP-ribosylation Reactions," (O. Hayaishi and K. Ueda, eds.) Academic Press, New York, pp. 117- 155.
159 . Tanaka, Y . , M atsunami, N. and Yoshihara, K. (1981) Biochem. Biophys. Res. Commun. 99_, 837.
160. Sund, H . , ed. (1977) "Pyridine Nucleotide-Dependent Dehydrogenases," Walter de Gruyter, Berlin.
161. Schulz, G. E. and Schirmer, R. H. (1979) in "Principles of Protein Structure," Springer- Verlag, New York, pp. 220-224.
162. Dutler, H. (1977) in "Pyridine Nucleotide-Dependent Dehydrogenases," (H. Sund, ed.) Walter de Gruyter, Berlin, pp. 339-350.
163. Nishikimi, M., Ogasawara, K., Kameshita, I., Taniguchi, T. and Shizuta, Y. (1982) J. Biol. Chem. 257, 6102. 204
164. Yoshihara, K. and Tanaka, Y. (1981) J. Biol. Chem. 256, 6756.
165. Tanaka, Y., Matsunami, N., Itaya, A. and Yoshihara, K. (1981) J. Biochem. (Tokyo) 90^, 1131.
166. Ohg u s h i , H., Yoshihara, K. and Kamiya, T. (1980) J. Biol. Chem. 255, 6205.
167. Chock, P. B., Rhee, S. G . , Stadtman, E. R. (1980) Annu. Rev. Biochem. 49, 813.