Yasuhiro Ohashi

Inhibition of DNA polymerase α, DNA polymerase β, terminal deoxynucleotidyl transferase, and DNA ligase II by poly(ADP-ribosyl)ation reaction in vitro☆

Incubation of DNA polymerase α1, DNA polymerase β, terminal deoxynucleotidyl transferase, or DNA ligase II in a reconstituted poly(ADP-ribosyl)ating enzyme system markedly suppressed the activity of these enzymes. Components required for poly(ADP-ribose) synthesis including poly(ADP-ribose) polymerase, NAD+, DNA, and Mg2+ were all essential for the observed suppression. Purified poly(ADP-ribose) itself, however, was slightly inhibitory to all of these enzymes. Furtheremore, the suppressed activities of DNA polymerase α, DNA polymerase β, and terminal deoxynucleotidyl transferase were largely restored (3 to 4-fold stimulation was observed) by a mild alkaline treatment, a procedure known to hydrolyze alkaline-labile ester linkage between poly(ADP-ribose) and an acceptor protein. All of these results strongly suggest that the four nuclear enzymes were inhibited as a result of poly(ADP-ribosyl)ation of either the enzyme molecule itself or some regulatory proteins of these enzymes.

ADP-ribosylation by type C1 and D botulinum neurotoxins: Stimulation by guanine nucleotides and inhibition by guanidino-containing compounds

We recently reported that type D botulinum neurotoxin ADP-ribosylates a specific protein of Mr 21,000 in membrane fractions of various tissues (Ohashi, Y. and Narumiya, S. (1987) J. Biol. Chem. in press). We examined similar enzyme activities in other types (types A, B, C1 and E) of botulinum neurotoxins. Of these, only type C1 toxin showed the activity similar to type D toxin and ADP-ribosylated the same Mr 21,000 protein in membranes of mouse brain. No enzyme activities were detected in type A, B and E toxins under the present experimental conditions. GTP stimulated ADP-ribosylation by the two toxins in a concentration dependent manner from 10 nM to 100 microM. The maximum stimulation was about 6 fold. GDP was 10 times less potent than GTP and achieved similar maximum at 1 mM, while GMP, ADP and ATP had little effect. Several guanidino-containing compounds dose-dependently inhibited the activities of both toxins. The IC50 values were 8.5, 14.5 and 45 mM for agmatine, L-arginine methyl ester and guanidine, respectively.

Mutagenicity of nifurtimox in Escherichia coli

The effects of nifurtimox, a nitrofuran derivative, on killing and mutation induction in 3 Escherichia coli strains having different DNA-repair systems for UV lesion were studied and compared with the effects of furylfuramide. Nifurtimox induces mutations at a high frequency in both Hs3OR (uvrA-) and H/r30R (radiation-resistant), although no significant killing effect is detected with Hs30R. No significantly induced mutation frequency could be detected with NG30 (recA-), which is very sensitive to killing by nifurtimox. The characteristics of lesions of DNA induced by nifurtimox and the mechanism of mutation induction in Hs30R are discussed.

Poly(ADP-ribosyl)ation of DNA polymerase βinvitro☆

DNA polymerase beta purified from bovine thymus is markedly inhibited when incubated in a reconstituted poly(ADP-ribosyl)ating reaction system. Analyses of the reaction product synthesized in this system by SDS-polyacrylamide gel electrophoresis and subsequent fluorography of the gel indicated that ADP-ribose is covalently attached to DNA polymerase beta molecule (Mr = 44,000).

A method for determining oligo- and poly(ADP-ribosy)ated enzymes and proteins in vitro

A new method to determine oligo- and poly(ADP-ribosyl)ated enzymes and proteins in vitro has been developed. This method is based on the facts that in Mg2+-depleted condition automodification of poly(ADP-ribose)polymerase is minimized and exogenously added acceptor protein is oligo(ADP-ribosyl)ated predominantly, and in Mg2+-fortified conditions the exogenous acceptor can be poly(ADP-ribosyl)ated. When 13 proteins, including several enzymes, were subjected to this system, dimeric bovine seminal RNase and micrococcal nuclease were found to be oligo(ADP-ribosyl)ated under Mg2+-depleted conditions but their activity was unchanged. Under Mg2+-fortified conditions however, the RNase was deactivated concomitantly with its extensive poly(ADP-ribosyl)ation. When dimeric bovine seminal RNase was monomerized in advance by treatment with dithiothreitol and urea, the enzyme lost ADP-ribose-accepting ability in spite of a significant residual enzyme activity. As used here successfully, the Mg2+-depleted and Mg2+-fortified ADP-ribosylation and subsequent chromatographic analysis of various proteins and enzymes might be an useful method for proving their oligo- and poly(ADP-ribosyl)ation.

An ovulation inducing agent containing clomiphene citrate causes DNA-strand breaks without SOS responses in Escherichia coli

Effects of Clomid, an ovulation-inducing drug containing clomiphene citrate, on Escherichia coli were investigated. Radiation-sensitive mutants, uvrA and recA, were more sensitive to Clomid than the parental wild-type strain. DNA synthesis in these two strains was more depressed by Clomid than that in the wild-type strain. Clomid caused DNA-strand breaks, but few SOS responses such as mutation, induction of prophage and expression of the umuC+ gene were induced.

DNA breakage of human leukemia cells by clomiphene, an ovulation-inducing agent

When HPB-ALL, human lymphoblastic leukemia cells were treated with clomiphene citrate, an ovulation-inducing agent, poly(ADP-ribose) synthesizing activity of the cells increased up to 4 fold. This stimulatory effect was almost comparable to that of bleomycin, a typical DNA strand breaking agent. Since the agents causing DNA breakage stimulate poly(ADP-ribose) synthesis in cells [Berger, N. A., Sikorski, G. W., Petzold, S. J., and Kurohara, K. K. J. Clin. Invest. 63, 1164-1171 (1979)], clomiphene citrate is suggested to damage DNA of the cells. In fact, an increase of single strand breakage of the DNA was detected by using alkaline sucrose density gradient centrifugation when the HPB-ALL cells were treated with increasing concentration of clomiphene. The inhibition of the cell growth by clomiphene citrate (IC50 = 5 micrograms/ml) appeared to be ascribable to its potent DNA-damaging effect. Although bleomycin activated purified poly(ADP-ribose) synthetase by cleaving covalently closed circular plasmid pBR322 DNA in vitro, clomiphene citrate per se did not.

Induction of SOS responses in Escherichia coli by Panfuran-S, 3-di(hydroxymethyl)amino-6-(5-nitro-2-furylethenyl)-1,2,4-triazine

The inducibility of SOS responses by Panfuran-S, which has been used as an antimicrobial medicine in Japan, was studied in Escherichia coli cells having different DNA-repair capacities for UV lesions. Panfuran-S induced mutations at high frequencies in uvrA and the wild-type strains, and significant killing effects of Panfuran-S were detected in DNA-repair-deficient strains, uvrA and recA. The effective prophage induction was detected in two kinds of lambda-lysogenized cells treated with Panfuran-S. The expression of the umuC+ gene was apparently induced in uvrA and the wild-type strains, but not induced in lexA and recA strains. In particular, high inducibility of the gene expression was detected in uvrA strain as compared with the wild-type strain. From these results, we conclude that Panfuran-S is a DNA-damaging agent and may induce the error-prone SOS responses.

Poly(ADP-Ribos)ylation of Nuclear Enzymes

Poly(ADP-ribose) polymerase catalyzes a sequential transfer of an ADP-ribose portion of NAD+ to various chromatin proteins [1] and to the polymerase itself (automodification [2]), forming a polymer of ADP-ribose, which is covalently bound to protein at one end [3]. Recent studies elucidated that two chromatin enzymes, Ca2+, Mg2+-dependent endonuclease [4, 5] and DNA topoisomerase [6, 7], were markedly inhibited as a result of poly(ADP-ribos)ylation of the enzyme proteins. RNA polymerase I [8] and DNA ligase II [9] also are suggested to be poly(ADP-ribos)ylated, although the latter enzyme seems to be activated after poly(ADP-ribos)ylation in vivo. Furthermore, bull seminal RNase [10, 29] and micrococcal nuclease [29] also have been shown to be the acceptors of ADP-ribose in the enzyme reaction in vitro. These results suggest a possibility that poly(ADP-ribose) polymerase randomly modifies many kinds of chromatin enzymes rather than it selecting a few kinds of specific enzymes as its targets. Thus, in order to study whether the modification reaction is specific only for the enzymes described above, we examined six nuclear enzymes, which are involved in metabolism or function of chromatin. After several unsuccessful trials using standard and modified conditions of the reconstituted ADP-ribosylating system, we found that all of these enzymes except DNA ligase I were markedly inhibited when the enzymes were incubated in an ADP-ribosylating reaction mixture containing a limited concentration of buffer (5 mM).