- Email: [email protected]
A Novel Series of DNA Triple Helix-Binding Ligands
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
224, 717–720 (1996)
1089
A Novel Series of DNA Triple Helix-Binding Ligands1 Keith R. Fox,*,2 David E. Thurston,† Terence C. Jenkins,‡ Athanasia Varvaresou,§ Andrew Tsotinis,§ and Theodora Siatra-Papastaikoudi§ *Department of Physiology & Pharmacology, University of Southampton, Bassett Crescent East, Southampton, SO16 7PX, United Kingdom; †School of Pharmacy, University of Portsmouth, Park Building, King Henry 1st Street, Portsmouth, PO1 2DZ, United Kingdom; ‡CRC Biomolecular Structure Unit, Institute of Cancer Research, Cotswold Road, Sutton, Surrey, SM2 5NG, United Kingdom; and §Department of Pharmacy, Division of Pharmaceutical Chemistry, University of Athens, Panepistimiopoli-Zografou, GR-157 51 Athens, Greece Received June 7, 1996 We have examined the effect of a series of substituted imidazothioxanthones on the stability of an intermolecular DNA triple helix by DNase I footprinting. We find that several of these compounds promote the formation of a complex between T5C5 and the target site A6G6rC6T6 , suggesting that they bind specifically to triplex DNA. The only inactive derivative lacked a protonatable function in the side chain, suggesting that this is an essential feature for triplex stabilization. These compounds, which are amongst the first triplex-binding ligands which possess an uncharged chromophore, are selective for the TrAT rather than the C/rGC triplet. q 1996 Academic Press, Inc.
Synthetic oligonucleotides can bind to homopurine sequences in duplex DNA, generating intermolecular triple helices which are stabilized by the formation of specific hydrogen bonds between bases in the third strand and functional groups located in the DNA major groove [13]. Two types of triple helix have been characterized, which vary according to the orientation of the third strand with respect to the duplex purine strand. Pyrimidine-rich third strands bind in a parallel orientation and are characterized by C/rGC and TrAT triplets [4,5], while purinerich oligonucleotides bind antiparallel generating GrGC, ArAT and TrAT triplets [6-8]. Although the formation of intermolecular DNA triple helices offers the potential for designing compounds with exquisite sequence recognition properties, the binding of the third strand oligonucleotide may not be strong. One means of improving the interaction is to design compounds which bind to triplex, but not duplex, DNA. In recent years several such compounds have been described including the benzopyridoindole derivatives BePI and BgPI [9-12], coralyne [13], naphthoquinoline derivatives [14-16], and 2,6-disubstituted anthraquinones [17]. Each of these compounds is presumed to act by selectively intercalating into the DNA triple helix. Binding to duplex DNA is disfavoured as a result of their geometry and large aromatic surface area. All of the compounds described to date contain a positive charge on the intercalating ring, which may limit the interaction with C/rGC triplets. In this paper we show that a series of functionalised imidazothioxanthones can also stabilize intermolecular DNA triple helices. MATERIALS AND METHODS Chemicals. The structures of compounds are shown in Figure 1. These were prepared as previously described [18,19]. Oligonucleotides were purchased from Oswel oligonucleotide service, stored at 0207C and used without further purification.
1 2
In memory of Professor Aspasia Papadaki-Valiraki. Corresponding author. 717 0006-291X/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC
Vol. 224, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Chemical structures of the imidazothioxanthone derivatives.
DNA fragments. Plasmid pAG1, containing the sequence A6G6rC6T6 cloned into the SmaI site of pUC18, was prepared as previously described [16,20]. The plasmid was cut with HindIII, labelled at the 3*-end with a-[32P]dATP using reverse transcriptase and cut again with EcoR1. This procedure labels the purine-containing strand of the target site. The radiolabelled DNA fragment of interest was separated from the remainder of the plasmid on an 8% polyacrylamide gel and dissolved in 10mM Tris-HCl pH 7.5 containing 0.1mM EDTA. Footprinting. Samples were prepared by mixing 1.5ml of radiolabelled DNA with 1.5ml oligonucleotide (dissolved in 50mM sodium acetate, pH 5.5 containing 10mM MgCl2) and 1.5ml of triplex binding ligand (dissolved in 50mM sodium acetate, pH 5.5 containing 5mM MgCl2). These were left to equilibrate for at least 30 minutes before adding 2ml DNase I (dissolved in 20mM NaCl, 2mM MgCl2 , 2mM MnCl2 at a concentration of about 0.01 units/ml). The reaction was stopped after 1 minute by adding 3ml of formamide containing 10mM EDTA and 0.1% (w/v) bromophenol blue. The products of digestion were resolved on 10% polyacrylamide gels containing 8M urea. 40cm gels were run at 1500V for about 2 hours, then fixed in 10% (v/v) acetic acid, transferred to Whatman 3MM paper and dried under vacuum at 807C. Gels were exposed to autoradiography at 0707C using an intensifying screen.
RESULTS
Figure 2a shows DNase I digestion patterns for the DNA fragment containing the triplex target site A6G6rC6T6 in the presence and absence of the third strand oligonucleotide T5C5 , together with each of the putative triplex binding ligands. These experiments were performed at pH 5.5 so as to stabilize the formation of the C/rGC triplets. It can be seen that, as previously described [16] the oligonucleotide alone does not generate a footprint under these conditions. This inability to form a stable triplex is attributed to its short length, generating only 10 triplets, and because the complex contains five contiguous C/rGC triplets. In the presence of a naphthoquinoline triplex-binding ligand [14-16] (lane labelled 10mM T5C5 / 10mM naphtho) a clear footprint is generated, which extends over the entire target site. A similar protection is afforded by 10mM concentrations of each of the compounds 1, 2, 4, 5, 6 and 7. These footprints each cover the entire target site, extending in the 5* (upper) direction by about four bases and terminating at the lower (3*) edge of the target site. The protection is most clearly seen by the absence of the doublet at the upper edge of the target site and the single band at the centre of the target. None of these compounds affects the DNase I digestion pattern in the absence of the oligonucleotide. Compound 3, which possesses a simple ester instead of a cationic side chain, is the only one of these ligands which does not appear to facilitate triplex formation under these solution conditions. Figure 1b shows the effect of varying concentrations of compound 2 on the complex formed with 10mM oligonucleotide. A clear footprint is produced with 10mM ligand with attenuated 718
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC
Vol. 224, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. DNase I digestion patterns showing the binding of T5C5 to a DNA fragment containing the target site A6G6 .C6T6 in the presence and absence of the triplex binding ligands. a) con indicates the control digestion of DNA in the absence of oligonucleotide or added ligand. The fourth lane (10mM T5C5 / 10mM naphtho) shows the footprint produced by T5C5 in the presence of the naphthoquinoline derivative described in [14–16]. Lanes 1–7 show the digestion in the presence of 10mM T5C5 with 10mM of each of the compounds in turn. The track labelled ‘‘GA’’ is a Maxam-Gilbert sequencing lane specific for purines. b) con indicates the control digestion of DNA in the absence of oligonucleotide or added ligand. The last five lanes show the digestion in the presence of 10mM T5C5 with varying concentrations of compound 2. The ligand concentration (mM) is shown at the top of each gel lane. In both panels the position of the insert of is indicated by the square brackets.
DNase cleavage still evident at concentrations of 3mM and 1mM; the cleavage pattern has returned to that in the control by 0.3mM. This concentration dependence is similar to that observed with other triplex binding ligands [15,16]. In similar series of experiments we have attempted to determine whether these compounds discriminate between TrAT and C/rGC triplets, by comparing its ability to stabilize the interaction of T6C2 or T2C6 at the same target site. The former should generate a complex containing 61TrAT and 21C/rGC triplets, while the latter contains 21TrAT and 61C/rGC triplets. Neither oligonucleotide produces a footprint in the absence of a stabilizing ligand. We find that compound 2 only promotes complex formation with T6C2 (not shown), suggesting that it is selective for the TrAT triplet. 719
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC
Vol. 224, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DISCUSSION
The results presented above demonstrate that these imidazothioxanthones can promote intermolecular triple helix formation, and suggest that a protonated or positively charged side chain is an essential part of their structure. The cationic side chain is also required for the interaction of these compounds with duplex DNA as well as for their cytotoxicity [19]. We assume that these compounds bind by intercalation, positioning the charged side group in one or other of the DNA grooves, where it can make contacts with the phosphodiester backbone(s). Although these compounds bind to duplex DNA [19], they do not affect DNase I cleavage at concentrations which stabilize triplex formation (10mM). This suggests that their large aromatic surface area may be able to form better stacking interactions with base triplets than with duplex DNA. The shape of the molecule may also be important for triplex-binding; the fused aromatic rings are not in a linear array (as too for BePI and coralyne) producing better overlap with the triplets than might be obtained with a simple linear polycyclic aromatic compound. We had hoped that, since these compounds possess an uncharged chromophore (unlike coralyne or BePI and BgPI) they might be better agents for stabilizing C/rGC triplets. BePI does not bind well to C/rGC triplets, presumably as a result of repulsion between the positive charges on the ligand and the triplet. However we find that complexes containing predominantly TrAT triplets are stabilized by these ligand, while those containing mainly C/rGC are not. This suggests that factors other than charge repulsion account for the selective action on TrAT triplets. ACKNOWLEDGMENTS This work was supported by grants from the Cancer Research Campaign and the Medical Research Council (KRF).
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Thuong, N. T., and He´le`ne, C. (1993) Angewandte Chemie 32, 666–690. Chubb, J. M., and Hogan, M. E. (1992) Trends in Biotechnol. 10, 132–136. Moffat, A. S. (1991) Science 252, 1374–1375. Moser, H. E., and Dervan, P. B. (1987) Science 238, 645–650. Le Doan, T., Perrouault, L., Praseuth, D., Habhoub, N., Decout, J. L., Thuong, N. T., Lhomme, J., and He´le`ne, C. (1987) Nucleic Acids Res. 15, 7749–7760. Beal, P. A., and Dervan, P. B. (1991) Science 251, 1360–1363. Radhakrishnan, I., de los Santos, C., and Patel, D. J. (1993) J. Mol. Biol. 234, 188–197. Radhakrishnan, I., and Patel, D. J. (1993) Structure 1, 135–152. Mergny, J. L., Duval-Valentin, G., Nguyen, C. H., Perrouault, L., Faucon, B., Rouge´e, M., Montenay-Garestier, T., Bisagni, E., and He´le`ne, C. (1992) Science 256, 1681–1684. Pilch, D. S., Waring, M. J., Sun, J.-S., Rouge´e, M., Nguyen, C.-H., Bisagni, E., Garestier, T., and He´le`ne, C. (1993) J. Mol. Biol. 232, 926–946. Pilch, D. S., Martin, M.-T., Nguyen, C. H., Sun, J.-S. Bisagni, E., Garestier, T., and He´le`ne, C. (1993) J. Am. Chem. Soc. 115, 9942–9951. Escude, C., Nguyen, C. H., Mergny, J.-L., Sun, J.-S., Bisagni, E., Garestier, T., and He´le`ne, C. (1995) J. Am. Chem. Soc. 117, 10212–10219. Lee, J. S., Latimer, L. J. P., and Hampel, K. J. (1993) Biochemistry 32, 5591–5597. Wilson, W. D., Tanious, F. A., Mizan, S., Yao, S., Kiselyov, A. S., Zon, G., and Strekowski, L. (1993) Biochemistry 32, 10614–10621. Chandler, S. P., Strekowski, L., Wilson, W. D., and Fox, K. R. (1995) Biochemistry 34, 7234–7242. Cassidy, S. A., Strekowski, L., Wilson, W. D., and Fox, K. R. (1994) Biochemistry 33, 15338–15347. Fox, K. R., Polucci, P., Jenkins, T. C., and Neidle, S. (1995) Proc. Natl. Acad. Sci. U.S.A. 92, 7887–7891. Varvaresou, A., Tsotinis, A., Papadaki-Valiraki, A., and Siatra-Papastaikoudi, Th. (1996) Bioorg. & Med. Chem. Lett. 6, 861–864. Varvaresou, A., Tsotinis, A., Siatra-Papastaikoudi, Th., Papadaki-Valiraki, A., Thurston, D. E. Jenkins, T. C., and Kelland, L. R. (1996) Bioorg. & Med. Chem. Lett. 6, 856–870. Fox, K. R. (1994) Nucleic Acids Res. 22, 2016–2021.
720
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC
224, 717–720 (1996)
1089
A Novel Series of DNA Triple Helix-Binding Ligands1 Keith R. Fox,*,2 David E. Thurston,† Terence C. Jenkins,‡ Athanasia Varvaresou,§ Andrew Tsotinis,§ and Theodora Siatra-Papastaikoudi§ *Department of Physiology & Pharmacology, University of Southampton, Bassett Crescent East, Southampton, SO16 7PX, United Kingdom; †School of Pharmacy, University of Portsmouth, Park Building, King Henry 1st Street, Portsmouth, PO1 2DZ, United Kingdom; ‡CRC Biomolecular Structure Unit, Institute of Cancer Research, Cotswold Road, Sutton, Surrey, SM2 5NG, United Kingdom; and §Department of Pharmacy, Division of Pharmaceutical Chemistry, University of Athens, Panepistimiopoli-Zografou, GR-157 51 Athens, Greece Received June 7, 1996 We have examined the effect of a series of substituted imidazothioxanthones on the stability of an intermolecular DNA triple helix by DNase I footprinting. We find that several of these compounds promote the formation of a complex between T5C5 and the target site A6G6rC6T6 , suggesting that they bind specifically to triplex DNA. The only inactive derivative lacked a protonatable function in the side chain, suggesting that this is an essential feature for triplex stabilization. These compounds, which are amongst the first triplex-binding ligands which possess an uncharged chromophore, are selective for the TrAT rather than the C/rGC triplet. q 1996 Academic Press, Inc.
Synthetic oligonucleotides can bind to homopurine sequences in duplex DNA, generating intermolecular triple helices which are stabilized by the formation of specific hydrogen bonds between bases in the third strand and functional groups located in the DNA major groove [13]. Two types of triple helix have been characterized, which vary according to the orientation of the third strand with respect to the duplex purine strand. Pyrimidine-rich third strands bind in a parallel orientation and are characterized by C/rGC and TrAT triplets [4,5], while purinerich oligonucleotides bind antiparallel generating GrGC, ArAT and TrAT triplets [6-8]. Although the formation of intermolecular DNA triple helices offers the potential for designing compounds with exquisite sequence recognition properties, the binding of the third strand oligonucleotide may not be strong. One means of improving the interaction is to design compounds which bind to triplex, but not duplex, DNA. In recent years several such compounds have been described including the benzopyridoindole derivatives BePI and BgPI [9-12], coralyne [13], naphthoquinoline derivatives [14-16], and 2,6-disubstituted anthraquinones [17]. Each of these compounds is presumed to act by selectively intercalating into the DNA triple helix. Binding to duplex DNA is disfavoured as a result of their geometry and large aromatic surface area. All of the compounds described to date contain a positive charge on the intercalating ring, which may limit the interaction with C/rGC triplets. In this paper we show that a series of functionalised imidazothioxanthones can also stabilize intermolecular DNA triple helices. MATERIALS AND METHODS Chemicals. The structures of compounds are shown in Figure 1. These were prepared as previously described [18,19]. Oligonucleotides were purchased from Oswel oligonucleotide service, stored at 0207C and used without further purification.
1 2
In memory of Professor Aspasia Papadaki-Valiraki. Corresponding author. 717 0006-291X/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC
Vol. 224, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Chemical structures of the imidazothioxanthone derivatives.
DNA fragments. Plasmid pAG1, containing the sequence A6G6rC6T6 cloned into the SmaI site of pUC18, was prepared as previously described [16,20]. The plasmid was cut with HindIII, labelled at the 3*-end with a-[32P]dATP using reverse transcriptase and cut again with EcoR1. This procedure labels the purine-containing strand of the target site. The radiolabelled DNA fragment of interest was separated from the remainder of the plasmid on an 8% polyacrylamide gel and dissolved in 10mM Tris-HCl pH 7.5 containing 0.1mM EDTA. Footprinting. Samples were prepared by mixing 1.5ml of radiolabelled DNA with 1.5ml oligonucleotide (dissolved in 50mM sodium acetate, pH 5.5 containing 10mM MgCl2) and 1.5ml of triplex binding ligand (dissolved in 50mM sodium acetate, pH 5.5 containing 5mM MgCl2). These were left to equilibrate for at least 30 minutes before adding 2ml DNase I (dissolved in 20mM NaCl, 2mM MgCl2 , 2mM MnCl2 at a concentration of about 0.01 units/ml). The reaction was stopped after 1 minute by adding 3ml of formamide containing 10mM EDTA and 0.1% (w/v) bromophenol blue. The products of digestion were resolved on 10% polyacrylamide gels containing 8M urea. 40cm gels were run at 1500V for about 2 hours, then fixed in 10% (v/v) acetic acid, transferred to Whatman 3MM paper and dried under vacuum at 807C. Gels were exposed to autoradiography at 0707C using an intensifying screen.
RESULTS
Figure 2a shows DNase I digestion patterns for the DNA fragment containing the triplex target site A6G6rC6T6 in the presence and absence of the third strand oligonucleotide T5C5 , together with each of the putative triplex binding ligands. These experiments were performed at pH 5.5 so as to stabilize the formation of the C/rGC triplets. It can be seen that, as previously described [16] the oligonucleotide alone does not generate a footprint under these conditions. This inability to form a stable triplex is attributed to its short length, generating only 10 triplets, and because the complex contains five contiguous C/rGC triplets. In the presence of a naphthoquinoline triplex-binding ligand [14-16] (lane labelled 10mM T5C5 / 10mM naphtho) a clear footprint is generated, which extends over the entire target site. A similar protection is afforded by 10mM concentrations of each of the compounds 1, 2, 4, 5, 6 and 7. These footprints each cover the entire target site, extending in the 5* (upper) direction by about four bases and terminating at the lower (3*) edge of the target site. The protection is most clearly seen by the absence of the doublet at the upper edge of the target site and the single band at the centre of the target. None of these compounds affects the DNase I digestion pattern in the absence of the oligonucleotide. Compound 3, which possesses a simple ester instead of a cationic side chain, is the only one of these ligands which does not appear to facilitate triplex formation under these solution conditions. Figure 1b shows the effect of varying concentrations of compound 2 on the complex formed with 10mM oligonucleotide. A clear footprint is produced with 10mM ligand with attenuated 718
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC
Vol. 224, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. DNase I digestion patterns showing the binding of T5C5 to a DNA fragment containing the target site A6G6 .C6T6 in the presence and absence of the triplex binding ligands. a) con indicates the control digestion of DNA in the absence of oligonucleotide or added ligand. The fourth lane (10mM T5C5 / 10mM naphtho) shows the footprint produced by T5C5 in the presence of the naphthoquinoline derivative described in [14–16]. Lanes 1–7 show the digestion in the presence of 10mM T5C5 with 10mM of each of the compounds in turn. The track labelled ‘‘GA’’ is a Maxam-Gilbert sequencing lane specific for purines. b) con indicates the control digestion of DNA in the absence of oligonucleotide or added ligand. The last five lanes show the digestion in the presence of 10mM T5C5 with varying concentrations of compound 2. The ligand concentration (mM) is shown at the top of each gel lane. In both panels the position of the insert of is indicated by the square brackets.
DNase cleavage still evident at concentrations of 3mM and 1mM; the cleavage pattern has returned to that in the control by 0.3mM. This concentration dependence is similar to that observed with other triplex binding ligands [15,16]. In similar series of experiments we have attempted to determine whether these compounds discriminate between TrAT and C/rGC triplets, by comparing its ability to stabilize the interaction of T6C2 or T2C6 at the same target site. The former should generate a complex containing 61TrAT and 21C/rGC triplets, while the latter contains 21TrAT and 61C/rGC triplets. Neither oligonucleotide produces a footprint in the absence of a stabilizing ligand. We find that compound 2 only promotes complex formation with T6C2 (not shown), suggesting that it is selective for the TrAT triplet. 719
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC
Vol. 224, No. 3, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DISCUSSION
The results presented above demonstrate that these imidazothioxanthones can promote intermolecular triple helix formation, and suggest that a protonated or positively charged side chain is an essential part of their structure. The cationic side chain is also required for the interaction of these compounds with duplex DNA as well as for their cytotoxicity [19]. We assume that these compounds bind by intercalation, positioning the charged side group in one or other of the DNA grooves, where it can make contacts with the phosphodiester backbone(s). Although these compounds bind to duplex DNA [19], they do not affect DNase I cleavage at concentrations which stabilize triplex formation (10mM). This suggests that their large aromatic surface area may be able to form better stacking interactions with base triplets than with duplex DNA. The shape of the molecule may also be important for triplex-binding; the fused aromatic rings are not in a linear array (as too for BePI and coralyne) producing better overlap with the triplets than might be obtained with a simple linear polycyclic aromatic compound. We had hoped that, since these compounds possess an uncharged chromophore (unlike coralyne or BePI and BgPI) they might be better agents for stabilizing C/rGC triplets. BePI does not bind well to C/rGC triplets, presumably as a result of repulsion between the positive charges on the ligand and the triplet. However we find that complexes containing predominantly TrAT triplets are stabilized by these ligand, while those containing mainly C/rGC are not. This suggests that factors other than charge repulsion account for the selective action on TrAT triplets. ACKNOWLEDGMENTS This work was supported by grants from the Cancer Research Campaign and the Medical Research Council (KRF).
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Thuong, N. T., and He´le`ne, C. (1993) Angewandte Chemie 32, 666–690. Chubb, J. M., and Hogan, M. E. (1992) Trends in Biotechnol. 10, 132–136. Moffat, A. S. (1991) Science 252, 1374–1375. Moser, H. E., and Dervan, P. B. (1987) Science 238, 645–650. Le Doan, T., Perrouault, L., Praseuth, D., Habhoub, N., Decout, J. L., Thuong, N. T., Lhomme, J., and He´le`ne, C. (1987) Nucleic Acids Res. 15, 7749–7760. Beal, P. A., and Dervan, P. B. (1991) Science 251, 1360–1363. Radhakrishnan, I., de los Santos, C., and Patel, D. J. (1993) J. Mol. Biol. 234, 188–197. Radhakrishnan, I., and Patel, D. J. (1993) Structure 1, 135–152. Mergny, J. L., Duval-Valentin, G., Nguyen, C. H., Perrouault, L., Faucon, B., Rouge´e, M., Montenay-Garestier, T., Bisagni, E., and He´le`ne, C. (1992) Science 256, 1681–1684. Pilch, D. S., Waring, M. J., Sun, J.-S., Rouge´e, M., Nguyen, C.-H., Bisagni, E., Garestier, T., and He´le`ne, C. (1993) J. Mol. Biol. 232, 926–946. Pilch, D. S., Martin, M.-T., Nguyen, C. H., Sun, J.-S. Bisagni, E., Garestier, T., and He´le`ne, C. (1993) J. Am. Chem. Soc. 115, 9942–9951. Escude, C., Nguyen, C. H., Mergny, J.-L., Sun, J.-S., Bisagni, E., Garestier, T., and He´le`ne, C. (1995) J. Am. Chem. Soc. 117, 10212–10219. Lee, J. S., Latimer, L. J. P., and Hampel, K. J. (1993) Biochemistry 32, 5591–5597. Wilson, W. D., Tanious, F. A., Mizan, S., Yao, S., Kiselyov, A. S., Zon, G., and Strekowski, L. (1993) Biochemistry 32, 10614–10621. Chandler, S. P., Strekowski, L., Wilson, W. D., and Fox, K. R. (1995) Biochemistry 34, 7234–7242. Cassidy, S. A., Strekowski, L., Wilson, W. D., and Fox, K. R. (1994) Biochemistry 33, 15338–15347. Fox, K. R., Polucci, P., Jenkins, T. C., and Neidle, S. (1995) Proc. Natl. Acad. Sci. U.S.A. 92, 7887–7891. Varvaresou, A., Tsotinis, A., Papadaki-Valiraki, A., and Siatra-Papastaikoudi, Th. (1996) Bioorg. & Med. Chem. Lett. 6, 861–864. Varvaresou, A., Tsotinis, A., Siatra-Papastaikoudi, Th., Papadaki-Valiraki, A., Thurston, D. E. Jenkins, T. C., and Kelland, L. R. (1996) Bioorg. & Med. Chem. Lett. 6, 856–870. Fox, K. R. (1994) Nucleic Acids Res. 22, 2016–2021.
720
AID
BBRC 5066
/
6904$$$461
07-10-96 10:43:03
bbrca
AP: BBRC