The EJ-ras gene was placed under the transcriptional control of the steroid-inducible mouse mammary tumor virus promoter/enhancer and introduced into Rat-1 fibroblasts, yielding the 14C cell line. When these cells were exposed to dexamethasone in vitro, EJ-ras mRNA was induced 15- to 20-fold, the cells grew in agar, and, after injection of cells into syngenic Fischer 344 rats, they produced lethal fibrosarcomas. Inhibitors of poly(ADP ribose) polymerase, which prevent the activation of the purified enzyme by a synthetic octadeoxyribonucleotide duplex, inhibited both in vivo tumorigenicity and in vitro growth in soft agar. The enzyme inhibitor 1,2-benzopyrone, which was studied in detail, and other polymerase inhibitors had no effect on EJ-ras mRNA or p21 protein expression. Poly(ADP ribose) polymerase [NAD+:poly(adenosine diphosphate D-ribose) ADP-D-ribosyltransferase, EC 2.4.2.30] was inhibited by the drug in both untreated and dexamethasone-treated cells both in vitro and in vivo to the same extent, but biological consequences of enzyme inhibition were manifest only when the cells were in the transformed tumorigenic state.

No abstract

No abstract

669f i g~r a t i o n ,~ Indeed, we have proposed that many retention reactions at asymmetric silicon proceed by mechanism SNi-Si, and we wish to report below stereochemical results which validate the above prediction and thereby provide strong additional evidence for the importance of the SNi-Si m e~h a n i s m . ~ We have found that a variety of functional groups on silicon can be reduced to give a silicon-hydrogen bond with i-BunAIH. Reductions have been carried out using optically active R3Si*X, in which R,Si* is a-naphthylphenylmethylsilyl.2a Table Z gives the stereochemical results for reduction of seven optically active silanes with i-BuzAIH in hexane solvent. In this solvent i-BunAlH is trimeric and association involves hydrogen bridge^.^ R3Si*X + i-BuzAIH + R,Si*H + ButAIX 3 H nmr singlets at 6 0.99, 1.04, and 1.70 (broad), and signals at 2.8 (m, 1 H, C-l(H)), 3.5 (m, 1 H, C-5(H)), and 5.3-5.8 (m, 3 H, vinyl protons), in a 4 : 1 ratio. An alternative formulation for the bicyclic product, 2,2,7trimethylbicyclo[3.2.0]hepta-3,6-diene, is excluded since a C(6)-H resonance well below the other olefinic absorptions is absent. sec-'; it is thus about 100 times slower than the conversion 1 + 2,' but at least 101l-fold faster than the degenerate and unobserved bicyclo[3.3.0]octa-2,6-diene rearrangement.3,4 That a cis,trans,cis tropone might be more accessible, relative to its bicyclic precursor, than a cis,trans,cis tropilidene seems resonable, considering the larger bond angle and lower bending force constant for the C-CO-C unit.Whatever the final resolution of this mechanistic controversy may be, the propensity shown by bicyclo-[4.2.0]-and bicyclo[3.2.0]dienes 3 and 5 for this type of rearrangement, and the absence of rearrangement in bicyclo[3.3.0]octa-2,6-diene at 450" for 85 min, does not deserve to be discounted as a pertinent fact through postulating that heteroatom functionality is essential.

The catalytic activity of highly purified poly(ADP-ribose) polymerase was determined at constant NAD+ concentration and varying concentrations of sDNA or synthetic octadeoxyribonucleotides of differing composition. The coenzymic activities of deoxyribonucleotides were compared in two ways: (i) graphic presentation of the activation of poly(ADP-ribose) polymerase in the presence of a large concentration range of deoxyribonucleotides and (ii) by calculating kn values for the deoxyribonucleotides.As determined by method i, auto-mono-ADP-ribosylation of the enzyme protein at 25 nM NAD+ was maximally activated at 1: 1 octamer/enzyme molar ratios by the octadeoxyribonucleotide derived from the regulatory region of SV40 DNA (duplex C). At a 0.41 sDNA/enzyme ratio, sDNA was the most active coenzyme for mono-ADP-ribosylation.At 200 PM NAD+, resulting in polymer synthesis and with histones as secondary polymer acceptors, duplex C was the most active coenzyme, and the octamer containing the steroid hormone receptor binding consensus sequence of DNA was a close second, whereas sDNA exhibited an anomalous biphasic kinetics. sDNA was effective on mono-ADP-ribosylation at a concentration 150-200 -times lower than on polymer formation. When comparison of deoxyribonucleotides was based on method ii (kn values), by far the most efficiently binding coenzyme for both mono and polymer synthesis was sDNA, followed by duplex C, with (dA-dT)s exhibiting the weakest binding. The synthetic molecule 6-amino-1,2-benzopyrone (6-aminocoumarin) competitively inhibited the coenzymic function of synthetic octadeoxyribonucleotides at constant concentration of NAD+, identifying a new inhibitory site of poly(ADP-ribose) polymerase. Octadeoxyribonucleotide;Poly(ADP-ribose) polymerase; 6-Amino-1,Zbenzopyrone site