Gene, 149 (1994) 173-178 0 1994 Elsevier Science B.V. All rights reserved. 0378-l 119/94/$07.00 173 GENE 08097 Inhibition of prokaryotic and eukaryotic transcription by the 2: 1 2,9-dimethyl- 1 , 1zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 0-phenanthroline-cuprous complex, a ligand specific for open complexes* (Transcriptional inhibitors; HIV; RNA polymerase; redox-inactive (Neocup),Cu+; David M. Perrin, Lori Pearson, Abhijit ~azumder** open complexes) and David S. Sigman zyxwvutsrqponmlkjihgfedcbaZYXW Department of Biological Chemistry, School of M edicine, Department of Chemistry and Biochemistry, M olecular Biology Institute, University of California at Los Angeles, Los Angeles, CA 90024- 1570, USA Received by J.W. Larrick: 15 February 1994; Accepted: 11 April 1994; Received at publishers: 16 May 1994 SUMMARY The redox-stable, tetrahedral cuprous chelate of neocuproine (2,9-dimethyl-l,lO-phenanthroline) binds to the singlestranded DNA formed in open complexes and is an effective inhibitor of eukaryotic and prokaryotic transcription. Despite the many kinetic and structural differences between prokaryotic and eukaryotic transcription systems, they are all similarly inhibited by neocuproine copper, suggesting that all open complexes may share a homologous structure. INTRODUCTION Faithful transcription of the template strand is not possible without the separation of double-stranded DNA because the sequence information is masked by the complementary noncoding DNA strand (But and McClure, Correspondence to: Dr. D.S. Sigman, Molecular Biology Institute, University of California at Los Angeles, Los Angeles, CA 90024-1570, USA. Tel. (l-310) 825-890~ Fax (l-310) 206-7286; e-mail: sigman~ewald.mbi.ucla.edu *Presented at the Palo Alto Institute of Molecular Medicine Symposium on “Pharmaceutical Design: Anti-Sense, Triple Helix, Nucleic Acid-Binding Drugs”, January 31-February 1, 1994; Hyatt Rickeys, Palo Alto, CA, USA. **Present address: Laboratory of Molecular Pharmacology, National Cancer Institute, Bethesda, MD 20892, USA. Tel. (l-301) 496-5944. Abbreviations: bp, base pair(s); DTT, dithiothreitol; HIV, human immunodeficiency virus; Inr, initiator element with homology to and including the terminal deoxynucleotyltransferase initiator; I,, inhibitor concentration of 50% activity; kb, kilobasefs) or 1000 bp: LTR, long terminal repeat; (Neocup),Cu iI neocuproine redox-inactive complex (see Fib. 1); nt, nucleotide(s); NTP, nucleoside t~phosphate(s); (OP),Cu +, redox-active 2:l l,lO-ph-cuprous complex; PAGE; polySDS, acrylamide-gel electrophoresis; l,lO-ph, l,lO-phenanthroline; sodium dodecyl sulfate; ss, single strand(ed); TPA, tetradeconoyl phorbol acetate; rsp, transcription start point(s). SSDI 0378-l 119(94)00308-F 1985; But, 1987). All RNA polymerases must form steady-state intermediates during catalysis in which the template strand is transiently single-stranded (ss) and accessible to incoming NTP. The first intermediate, known as the open complex, is characterized by enzymestabilized ss DNA. Its formation is typically the rate limiting step in transcription initiation (McClure et al., 1978; McClure, 1980). Designing molecules which bind specifically to the open complex is an appealing strategy for generating potent transcription inhibitors because enzyme-bound, ss DNA at the tsp has a different reactivity than other regions of DNA in the genome. For exampie, it will react with base specific chemical modification reagents (Kirkegaard et al., 1983; &alla, 1985) like dimethyi sulfate which modifies ss cytosines (Kirkegaard et al., 1983) and K.permanganate which modifies ss thymidines (Sasse-Dwight and &alla, 1989). Ongoing studies with the chemical nuclease activity of l,lO-ph-copper have also shown that the template strand of three E. coli lac promoters reacts with the tetrahedral cuprous chelate of l,lO-ph ((OP),Cu+) and its derivatives (Spassky and Sigman, 1985; Spassky, 1986; Sigman, 1990). The template strands of the wild-type, P, and 174 lacUV5 promoters at precisely were strongly Unlike by (OP),Cu+ those sites which had demonstrated with base-specific E. coli promoters (Spassky cleaved to be ss chemical modification reagents. Other have exhibited comparable reactivity et al., 1988; Frantz the bimolecular reaction and O’Halloran, mechanism 1990). of the base modification reagent, the site specific scission by (OP),Cu+ likely reflects the creation of a binding site for this tetrahedral coordination complex complex. Since the isosteric, redox-inactive 2,9-dimethyl- 1JO-ph, (Neocup),Cu+, tetrahedral geometry in the open as the chelate shares the corresponding of same redox- tor with the IQcUVS transcription unit. It was found to be an effective inhibitor of RNA synthesis with the unique property of binding directly to the ss DNA within the open complex (~azumder et al., 1993; 1994). All of our studies with the lacUV5 promoter have indicated that the redox-inactive (Neocup),Cu+ blocks the oxidative scission of template chelates RESULTS AND DISCUSSION (a) Inhibition of a transcription unit with a demonstrated open complex There are two methods cuprous active l,lO-ph-cuprous chelate (Sigman et al., 1993), (Neocup),Cu’ was investigated as a transcription inhibi- cuprous units share a common structural motif with binding affinity for tetrahedral I,lO-ph-cuprous complexes. strand DNA by isosteric of l,lO-ph, as indicated redox-active in Fig. 1. for demonstrating binding of tetrahedral cuprous chelates of l,lO-ph to the open complex of eukaryotic transcription units; the first and most direct is the demonstration previously in bacterial of scission systems. For retardation analyses demonstrated graphic fraction containing TFIIE/F ing of RNA polymerase promoter-T~IID-TFII tionally competent gel, hypersensitive as demonstrated II to adenovirus complexes. example, gel- that the chromatostimulates the bindmajor late When these transcrip- complexes are footprinted within cleavage sites are observed that the are strikingly similar to those observed with the open complex of the lac promoters (Kuwabara and Sigman, 1987; Buratowski et al., 1991). The (OP)$LI+ scission sites In this paper, we report that (Neocup)~Cu ’ also inhibits eukaryotic transcription units. These results strongly indicate that the open complex formed with the adenovirus major late promoter is similar in structure to that suggest observed that open complexes of different transcription in the prokaryotic system (Buratowski et al., zyxwvutsrq HP> Template strand Catalytic&dative zyxwvutsrqponmlkjihgfedcbaZYXWVU nicked s&.&n Ope n c om ple xDN A Stable association and inhibiti~ Fig. 1. Redox-active (OP),Cu+ and redox-inactive (Neocup)@+ compete for the transcription bubble formed at kinetically competent transcriptjon start sites. The preferential scission of template strand of the ss DNAs of open complexes suggests that they create favorable binding sites for the redox-active tetrahedral complex, (OP),Cu+. These results suggested that its redox inactive isostere, (Neocup),Cu+, should bind reversibly to the open-complex and inhibit transcription. (Neocup),Cu + is the only reversible inhibitor of transcription which binds at this site. 1991). (OP),Cu+ hyperreactivity may be a shared property of many different transcription units. However, footprinting data on eukaryotic systems are difficult to obtain without purified components, since only 1% of the promoter DNA is present in a kinetically competent open complex when conventionally prepared HeLa extract is used to initiate transcription (Kadonaga, 1990; Kamakaka et al., 1991). Therefore, the second approach to monitoring the binding of a tetrahedral cu20 0 300 150 too 50 25 IO prous chelate is to assay for transcriptional inhibition by HIV LTR (Neocup),Cu+. This inhibition was initially examined 0 Transcript using a synthetic eukaryotic promoter derived from the adenovirus E4 promoter (Wang et al., 1992). This syn[(Neocup)@7 (micromolar) thetic promoter contains a TATA sequence, multiple binding sites for the chimeric virus-yeast transcriptional Fig. 2. Inhibition of transcription of HIV-LTR by (Neocup),Cu+. Gel inset: Kinetic assays. Lane 1, full-length transcript in the absence activator, GAL4-VP16, and a sequence of six thymidines of inhibitor; lanes 2-7, Decreasing concentrations of (Neocup),Cu” which extended from position - 3 to + 3 relative to the as indicated. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPO M ethods; Neocuproine was purchased as tsp. Base-specific modification reagents have been used 2,9-dimethyl-l,lO-ph (G.F. Smith). The inhibitor is initially prepared successfully to detect open complex formation with this fresh as the cuprous chelate at a stock concentration of 10mM CuSO$I40 mM neocuproine/69 mM mercaptopropionic acid (Aldrich), promoter (Wang et al., 1992). These reagents may be in water. It is then serially diluted into 10 x stocks to give the specified superior to (OP),Cu+ because they are less reactive to concentrations, and discarded after use. The HIV-LTR promoter region B-DNA and therefore contribute less background when from - 179 to + 80 upstream from the cat gene is contained on a 3-kb only 1% of the DNA is bound. In the presence of ATP, plasmid conferring ampicillin resistance (gift of Dr. Richard Gaynor, Southwestern Medical School) which has been overexpressed in zyxwvutsrqponm E. co& HeLa cell extract, and the GALCVPlG fusion protein, purified by the Qiagen system, and linearized with NcoI such that fullthe six thymidines at the tsp are reactive to permanganate length transcript is 620 nt in length. HeLa cell extracts were purchased if more than five binding sites for the fusion protein are from Promega. Transcription and inhibition analysis is carried out as present upstream from the zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA tsp . Even though (OP),Cu+ follows: 9 pl transcription buffer (20 mM Hepes pH 7.9/1OOmM KCl’O.2 mM EDTAiO.5 mM DTT/ZO% glycerol) 7.4 ul H,0/3.6 ul scission is not observed, (Neocup),Cu+ inhibits by 80% MgCl,/l ul containing 700 ng of DNA template/6 ul(8 units) HeLa cell at 16 uM (Neocup)~~u+ under the conditions where pernuclear extract, are preincubated for 10 min at 25°C. 3 ul of 10 x inhibimanganate demonstrates open complex formation. (b) In vitro inhibition of HIV LTR tor stock solution to give the final inhibitor concentrations is added and samples are incubated for another 5 min at 25°C. Transcription was initiated first by the addition of 8 ul of a cocktail containing 5 ul Hepes transcription buffer above 1.8 ul H,0/1.2 ul 50 mM MgCl,, followed by 1 ul of the NTP complement that is 600 uM each of ATP, CTP and UTP, and 40 uM GTP and 1 u1 of [w~‘PJGTP obtained from Amersham with a specific activity of 3000 Ci/mmol and 10 uCi/ul. The reactions are continued for 60 min at 3o”C, quenched with 350 ul stop buffer (7 M urea/350 mM NaCl/lO mM EDTA/lOO mM Tris pH 7.4/l% SDS). Samples are phenol-chloroform extracted and ethanol precipitated by standard methods. The precipitates were resuspended in formamide dye, and analyzed directly by a 7 M urea-8% PAGE. The autoradiogram is then analyzed by densitometry. Diagram: Analysis of kinetic data. The bands in the gel were analyzed by densitometry and percent inhibition was calculated from (R,-RJR, . where R is the densitometry reading. The values are plotted. Transcription of integrated DNA copies of the HIV is regulated by the long terminal repeat (LTR) which contains multiple cis-acting sequences that serve as upstream regulatory sites for both general and T-cell-specific transcriptional activators (Gaynor et al., 1988). The HIV LTR promoter contains a TATA box, binding sites for SP-1, two tandem recognition sequences for NFKB and a binding site for nuclear factors of activated T cells, NEAT (Li et al., 1991). The basal activity of this transcription unit can be assayed using HeLa cell extracts in the absence of the virally encoded positive regulatory protein, Tat. RNA synthesis from this transcription unit is inhibited by (Neocup)~Cu’ (Fig. 2). The assay was carried out by observed, (Neocup)~Cu+ most likely inhibits transcrippreincubating the DNA template for 10 min in HeLa cell tion initiation rather than elongation. extract followed by the addition of a full complement of The inhibition of HIV and lacUV5 transcription by share an important similarity. Both nucleotide triphosphates. Under these conditions, the I,, (Neocup),Cu+ depend of is observed at a concentration of 50 pM (Fig. 2). very sharply on the concentration Comparable levels of inhibition are observed when the (Neocup)~Cu+. Estimations of the Hill coefficient from inhibitor and nucleotide triphosphates are added simultathe data reported in Fig. 2 provide a value of 1.7. The Hill coefficient for the inhibition of the ZacUV-5 promoter neously, suggesting that (Neocup),Cu+ is a reversible inhibitor. Since full-length transcripts are predominantly zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA by (Neocup),Cu+ is approx. 1.8 (Mazumder et al., 1993). 176 This higher order dependence on the concentration of inhibitor is consistent with the scission results that have scription in a reaction The time of incubation been observed where formation of both open complexes and eliminate the kinetics of assembly as a variable in the assay. Both tran- and the scription units were inhibited by (Neocup),Cu’: the I,, for the TATA-dependent promoter is 30 uM for the InI RNA in the footprinting polymerase open of the lucUV5 complex E. coli with (OP),Cu+ three strong cleavage sites are observed (Spassky Sigman, 1985; Thederahn et al., 1990). Although multiplicity of sites could be attributable tion of a diffusible oxidative to the genera- species, an alternate explana- tion, consistent with the inhibition data, is that there is more than one binding site for the tetrahedral chelate in the open complex. the 4-phenyl-, The fact that the cuprous 5-phenyl- l,lO-phenanthroline sites on the template chelates of 3,4.7,8-tetramethyl- generate l-3 additional cleavage strand implies that the open com- plex may have additional (Mazumder and binding loci for these chelates et al., 1994). mixture using the HeLa extracts. was sufficiently long to allow the dependent, 70 uM (Fig. 3). These experiments demonstrate that despite the difference in the kinetics of assembly of the two promoters, the resulting open complexes are still susceptible to inhibition by the tetrahedral neocuproine-cuprous chelate. The results further show that the TATA sequence created itself does not compose for (Neocup),Cu’ within the binding site the open complex. (d) Cytotoxicity of the cuprous complexes of neocuproine and derivatives The biological activity of the neocuproine-cuprous late has previously containing an Spl site In contrast to the HIV and adenovirus-derived antimetabolites potentially useful in the chemotherapy of infectious disease and cancers. Toxicity of the free ligand otic promoters, eukary- many cellular genes do not contain TATA had been been demonstrated screened in order che- (c) In vitro inhibition of a TATA-less ‘initiator’ promoter for mycoplasms to identify (Antic et al., sequences (Smale and Baltimore, 1989). There are two subsets of these TATA-less genes. One contains G+C- 1977), trypanosomes (Shapiro et al., 1982) fungi (Blank, 1951) and both Gramand Gram’ bacteria (Butler rich sequences at the tsp and tend to encode ‘housekeeping genes’. The second is not G + C-rich and is likely to et al., 1969). In these studies, the unusual encode genes which are tightly regulated during develop- ment. Terminal deoxynucleotide transferase whose expression is restricted to precursor B and T cells has a TATA-less promoter which lacks a G + C-rich promoter (Smale and Baltimore, element 1989). It contains of the cuprous complex was not explicitly recognized probably because neocuproine scavenged the trace copper from the growth media. In two perceptive studies, Rabinovitz and colleagues demonstrated that the cytotoxic effect of zyxwvutsrq a 17-bp initiator ‘r (1~) with the sequence: -6 +11 6 ‘. 5’-GCCCTCATTCTGGAGAC-3’ (A represents effectiveness -m- TATA promoter -*- Inr promoter the tsp= + 1) This sequence can direct accurate the promoter strength is strongly basal transcription but enhanced by insertion of a TATA box or other components of heterologous promoters (O’Shea-Greenfield and Smale, 1992; ZenzieGregory et al., 1992). A variety of promoters have been constructed on the pSP72 vector to permit the comparison of the basal transcription directed by TATA box or Inr sequences in the presence of the SV-40 21-bp repeats (plasmid V and VI, respectively, of Smale et al., 1990). A significant difference between these two promoters is that the transcription unit containing the Znr sequence assembles more rapidly than the TATA containing plasmid. To investigate the similarity of the open complexes formed with these two promoter constructs, we assayed the inhibition by (Neocup),Cu+. To carry out the transcription assay, a complete complement of NTP was added to initiate tran- I * I I I. I * “1 1 200 250 50 100 150 0 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLK [(Neocup),Cu’] Fig. 3. Inhibition of transcription (micromolar) of Inr and TATA-containing promot- ers by (Neocup),Cu+. Concentrations of (Neocup),Cu * of 0, 2, 10, 50, 250 pM were used. The Inr and TATA promoters each contain upstream SP-1 sites and are present on a pSP72 derived vector (gift of Dr. Steven Smale, UCLA; see Smale et al., 1990) which has been overexpressed in E. coli and purified by the Qiagen system. The supercoiled plasmid is transcribed in vitro as noted in the methods for Fig. I. The internally labeled transcripts are analyzed directly by a 7 M urea-8% PAGE. The autoradiogram is then analyzed by densitometry and plotted as shown above. 177 neocuproine was copper dependent (Mohindru et al., 1983a). They demonstrated that the cellular uptake of neocuproine and cytotoxicity at a concentration of 100 nM required the addition of cupric ion which would be reduced to cuprous ion by the growth media. In control experiments, both uptake and cytoxicity were blocked by 4,7-diphenyl-neocuproine-disulfonic acid, a cuprous-specific ligand with no transcriptional inhibitory activity in vitro (Mohindru et al., 1983a,b; Mazumder et al., 1993). Subsequent studies in our laboratory indicated that the cuprous chelate of neocuproine was especially cytotoxic to the Ames tester strain TAlOl when it was used as a control for evaluating the mutagenicity of its redox-active isostere (OP),Cu+ (Feig et al., 1988). Our demonstration that (Neocup),Cu+ inhibits transcription in vitro suggested a possible mechanism for this cytotoxicity. To test this hypothesis, we investigated the inhibition of transcription of primary response genes induced in Swiss 3T3 cells by the phorbol ester TPA in the presence of the translation inhibitor cycloheximide (Fletcher et al., 1992; Kujubu et al., 1993). This system was chosen for two reasons: (i) the TPA-induced early response gene messages are synthesized in the absence of new protein synthesis and are not essential for cellular viability; and (ii) the kinetics of TPA-induced mRNA synthesis in the absence of new protein synthesis are much more rapid than the kinetics of cell death. This system therefore provided the opportunity of determining if transcription inhibition or cell death was responsible for the reduced level of observed transcription. After a 2-h induction, mRNA was isolated and the level of expression monitored by a Northern analysis. Although inhibition of transcription was apparent with cells that had been incubated with (Neocup),Cu+ at concentrations as low as 100 nM, the cells also apparently lost viability as measured by a replating assay. Despite the design of the experiment, the reduction in the amount of observable transcript could be due either to a decrease in the number of viable cells or to the inhibition of transcription. The fact that toxicity is observed at much lower concentrations than required for in vitro transcription is puzzling. It is possible that transcription inhibition is not responsible for cell death and that (Neocup),Cu+ triggers an apoptotic response (Williams et al., 1992; Zhong et al., 1993). Tetrahedral chelates of cuprous ion and l,lO-ph and its derivatives bind specifically to transcriptionally competent open complexes. Two lines of experiments support this conclusion. The first is that hyperreactive scission sites are observed when either prokaryotic or eukaryotic transcription complexes are footprinted with l,lO-phcopper. The preferential cutting of the open complex can be attributed to an RNA polymerase-induced binding site for the redox-active coordination chelate which then will cut the DNA (and possibly RNA polymerase) with high efficiency. The second line of evidence is the general inhibition of transcription by the redox-inert, exchange-stable cuprous chelate of neocuproine, (Neocup),Cu+. This simple coordination complex, which is isosteric with the redox-active (OP),Cu+, inhibits all transcription units that have been examined. In addition to blocking bacterial transcription units, it inhibits the HIV promoter which requires a range of activator promoters that use chimeric activators and synthetic promoters both with and without functional TATA sequences. These results suggest that the specificity of the binding site is entirely determined by the local structure of the open complex just upstream from the site of transcription initiation. These results are consistent with our early observation that lac promoter variants all show a common cleavage pattern when footprinted with (OP),Cu+ (Spassky and Sigman, 1985). The binding site generated at open complex exhibits interesting specificity (Mazumder et al., 1993; 1994). Octahedral phenanthroline complexes prepared with iron or nickel ion are not nearly as inhibitory. Moreover, redox-inactive isosteres of copper complexes which do not exhibit hyperreactive sites are poor inhibitors. For example, the cuprous complex of 4,7-diphenyl-l,lO-ph does not cleave the open complex of the lac promoter. As a result, its redox-inactive isostere, the cuprous complex of (4,7-diphenyl-Neocup),Cu+, is not nearly as effective an inhibitor. Since the determinants which govern the binding affinity for both the redox-active and redox-inactive coordination complexes appear to be similar, inhibition is a convenient method for screening the specificity of the binding site created in transcriptionally active open complexes. The cytotoxicity of the copper chelates have been previously reported. However, the present studies do not unambiguously show that the cytotoxicity is due to transcription inhibition. Cells are killed at much lower concentrations of (Neocup),Cu+ than are required to inhibit transcription in vitro. Further studies will focus on the mechanism of cytotoxicity, as well as defining the structural determinants of the RNA polymerases, the associated transcription factors and the promoter sequences which contribute to the binding specificity of these tetrahedral chelates to the open complex. zyxwvutsrqponmlkjihgfedcba ACKNOWLEDGEMENTS This research was supported by USPHS GM21199 and GM 39558. We thank Michael Carey, Harvey Herschman, Stephen Smale and the members of our research group for helpful discussions. 17x promoter REFERENCES open complcxc~. Proc. Natl. Acad. ( 199.3) Sci USA 90 8140 8144. ~4~~ulnder. A.. Perrin. D.M.. McMillen, D. and Siyman, U.S. Antic, B.M.. Van der Goat. H.. Nauta, W.T., Balt, S.. DC Bolster, W.G.. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Interactions of Tr~lllscripti~~l~ Inhibit(~rs with the I:‘. co/i FNA polyStouthamer. AH.. Verheul. H. and Vis, R.D.: The influence ofcopper mcrasc-irrcUV5 promoter open-complex. Biochemistry 33 ( 1994) ions on the growth-inhibitory effect of 2.2.bipyridyl and related 2262m 2268. compounds on mycoplasmas. Eur. J. Med. Chcm I2 (1977) McClure, 573-57s. Blank, F.: In vitro fungistatic action of phenanthrolines genic fungi. Nature 168 { 1951) 516 517. But. H.: Initiation of prokaryotic transcriptiol~ approaches. In: Eckstein. against Kin&c patho- and structural F. and Lilley. D.M.J. (Eds.), Nucleic Acids and Molecular Biology. Vol. I. Springer, Berlin. 1911’7,pp. 1X6-195. But, H. and McClure, R.: Kinetics of open complex formation betvveen Esrhuric,/rirr ccl/i RNA polymerase and Evidence for a sequential mechanism Biochemistry 24 (1985)2714-2723. Buratowski. S.. Sopta. M.. Greenblatt. mcrase II-associated change in the transcription proteins /lu,UV5 involving J. and Sharp. promoter. three steps. P.A.: RNA poly- for a DNA are required initiation Sci. USA 88 (1991) 750997513. Butler. H.M., Hurse, A., Thursky. the complex. conformation Proc. Natl. Acad. activity of E. and Shulman. A.: Bactericidal l.lO-phenanthroline-copper Biophys. Res. Commun. 155 (1988) 3388343. Fletcher, B.S., Kujubu. D.A.. Pcrrin. D.M. and Structure of the mitogen inducible ion. Biochem. Herschman, H.R.: TS IO zcne. J. Biol. Chem. 267 ( 1993 ) 4338 -4344. Frantz. B. and O’Halloran, T.V.: DNA distortion accompanies tran- scriptional activation by the metal-responsive gene-regulatory protein MerR. Biochemistry 29 (1990) 4747 4751. Gaynor. R.B.. Kuwabara, M.D.. Wu, F.K.. Garcia, J.A.. Harrich. D.. Briskin. for the RNA M.. Wall. R. and S&man. D.S.: Repeated B motifs in the HIV-I LTR enhancer are not equivalent and exhibit non cooperative factor binding. Proc. Natl. Acad. Sci. USA X5 (1488)94069410. Gralla, J.D.: Rapid ‘footprinting’ on supercoiled DNA. Proc. Natl. polymerase illiti~ition Mohindru. A.. Fisher, 2.9-Dunethyl-l,IO-phcnanthroline dependent cytotoxin 32 (lY83a) 3627 D.E.: A steady reaction. Proc. Natl. state assay J. Biol. Chem. 753 J.M. and Rabinovitz. (neocuproine): a potent. with anti-tumor activity. Biochem. M.: copper- Pharmacol. 3632. Mohindru. A.. Fisher, J.M. and Rabinovitz, M.: Bathocuproine sulphomite: a tissue ctllture-c[~tnpatible indicator of copper-in~~iiated toxicity. Nature 303 (l983b) O’Shea-Greenheld. Sasse-Dwight, S. and Gralla. and direction of RNA J. Biol. Chem. 267 (1992) 1391-1402. J.: KMnO, and inechanism A., Nathan Rescigno. chelators S.T.: Roles of TATA and initiator the start site location II transcription. DNA melting 8074 80X1. Shapiro. 64-65. A. and Smale. in eetermining polymerase H.C.. Hutner, as a probe for iric promoter in vivo. J. Biol. Chem. S.H.. Garofalo, 264 (1989) J.. McLaughlin, S.D.. D. and Bacchi, C.J.: In vivo and in vitro activity by diverse against Trypc~rtosonzo hrucei hruc,ei. J. Protoz.ool. 29 (1982) XS 90. Sigman. Sigman. D.S.: Chemical nuclcases. Biochemistry 29 ( 1990) 9007-9105. U.S.. Mazumder. A. and Perrin. D.M.: Chemical nucleases. them. Rev. 93 ( 1993) 22Y5-- 23 16. Smale. ST. and Baltimore. element. D.: The “initiator” as a transcription control Cell 57 (19X9) 103- 113. Smale, S.T., Schmidt. M.C.. Berk. A.J. Transcriptional activation by Spl as directed and Baltimore, D.: through TATA or initi- ator: specilic re~luirement for mammalian transcriptio~l Proc. Nntl. Acad. Sci. USA 87 ( 1990) 450994513. Spassky, Acad. Sci. USA 82 (1985) 307830X1. Kadonaga. J.T.: Assembly and disassembly steps in RNA chain mitiation. ( 197X) 8Y41 X948. clcments action of selected phenanthroline chelates and related compounds. Aust. J. Exp. BIoI. Med. Sci. 47 ( 1969)541 552. Feig, A.L.. Thederahn. T. and Sigman, D.S.: Mutagenicity of the nuclease W.R.: Rate-limiting Acad. Sci. USA 77 ( 19X0) .5634-563X. McClure. W.R.. Cech, C.L. and Johnston, A.: Following the progression factor of the transcription IID. bubble. J. Mol. Biol. I88 (1986) 99 103. of the zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Drosophila RNA Spassky. A. and Sigman. U.S.: Nuclease activity of I. IO-phellanthrolinetranscription. J. Biol. Chem. 265 polymerase II complex during (1990) 2624 2631. Kamakaka. R.T.. Tyree. C.M. and Kadonaga. cient RNA polymerase J.T.: Accurate and effi- II tr~~tlscription with a soluble nuclear frac- tion derived from Drosopiiifrc embryos. Proc. Nat]. Acad. Sci. USA 88 (1991) 1024- 1028. Kirkegaard, K.. Spassky, A., But, H. and Wang, J.: Mapping of singlestranded regions in duplex DNA at the sequence level: single-strand- specific cytosine methylation in RNA polymer~~sc-prolnoter plexes. Proc. NatI. Acad. Sci IJSA 80 ( 1983) 2544 2548. Kujubu. D.A.. Reddy. ST.. Fletcher, B.S. and Herschman. Expression of the protein product of prostaglandin synthetase com- copper operon. ion. Conformational analysis and footprinting Biochemistry 24 ( 1985) X050--8056. Spassky. A., Rimsky. the conformation promoter a structural strength: S.. But. H. and Busby, S.: Correlation between of Est~hcrich cdi IO hexamer sequences and use of orthophenanthroline index. EMBO 2/TIS cuprous complex as J. 7 (1988) 1871~ 1879. Thederahn, T., Spassky. A.. Kuwahara, M.D. and Sigman, Chemical nuclease activity of 5-phenyl-1.10-phenanthrohne-copper ion detects intermcdiate~ H.R.: of the Itrc in transcription initiation polymerasc. Biochem. Biophys. Res. Commun. Wang, W., Carey. M. and Gralla, J.D.: Polymerase D.S.: by E. co/i RN.4 168 (1YYO) 756 762. II promoter activa- 10 gene in mitogen-stimulated Swiss 3T3 cells. J. Biol. Chem. 268 ( 1993) 5425 5430. Kuwabara. M.D. and Sigman, D.S.: Footprinting DNA-protein complcxes in situ f~~llowin~ gel retardation assays using tion: closed complex formation and ATP-driven start site opening. Science 255 (1992) 450 ~453. Williams. G.T.. Smith. C.A.. McCarthy, N.J. and Grimes, E.A.: Apoptosis: final control point in cell biology. Trends Cell Biol. 3 1, IO-pl~e~~a~lthroline-copper ion: ~s~~7~~j~~~jucoii RNA polymerase(fit promoter complexes. Biochemistry 26 ( 1987) 7234 7238. Li. C.. Lai, C.. Sigman, D.S. and Gaynor. R.B.: Cloning of a cellular (1992) 263 267. Zenzie-Gregory, B., O’Shea-Greenfield. A. and Smale, S.T.: Similar mechanisms for transcription initiation mediated through a TATA factor, interleukin binding factor, that bind to NFAT-like motifs in the human immunodehciency virus long terminal repeat. Proc. Nat]. Acad. Sci. USA 88 (1991) 7739-7743. Marumder, A., Perrin, D.M.. Watson, K.H. and Sigman. D.S.: A tran- box or an initiator element. J. Biol. Chem. 267 (1992) 3823 2X30. Zhong, L.-T., Sarahan. T.. Kane. D.J., Charles. A.C.. Mah. S.P.. Edwards, R.H. and Bredesen, D.E.: Bcl-2 inhibits death of central neural cells induced by multiple agents. Proc. Natl. Acad. Sci. USA YO ( 1Y93) 453.1~-4537. scription inhibitor specific for unwound DNA in RNA polymerase-