Bioconjugafe Chem. 1994, 5, 497-500
497
Target-PromotedAlkylation of DNA Tianhu Li,t Qinping Zeng, and Steven E. Rokita* Department of Chemistry, State University of New York at Stony Brook, Stony Brook, New York 11794. Received June 30, 1994@
Inducible and selective alkylation of DNA was accomplished under neutral conditions by use of a silyl-protected phenol that served as a precursor for a highly reactive quinone methide. As expected, addition of fluoride triggered reaction of a model compound, 3-(tert-butyldimethylsiloxy)-4-[(pnitrophenoxy)methyl]benzamide,and its oligodeoxynucleotide conjugate. Surprisingly, the silyl phenol was also specifically yet more slowly activated by the environment of duplex DNA in the absence of fluoride. This alternative process was associated with the hybridization of probe and target strands, and single-stranded DNA was unable to induce a similar activation. Therefore, DNA appears to effect its own alkylation by promoting the formation of a n electrophilic and nondiffusible intermediate.
DNA alkylation is central to the action of many anticancer therapeutics despite its inherent lack of specificity. Most efforts to enhance the selectivity have relied on the technique of affinity modification (1-3) in which a n electrophile is attached to a DNA-binding ligand. Much greater specificity would be possible through the use of mechanism-based inactivation. In this case, selective modification could be controlled by the binding and catalytic properties of a target as widely applied in enzymology (4). Extension of this to the field of nucleic acids is currently restricted by the limited reactions catalyzed by this class of molecules. For DNA, only the hydrolysis of benzo[alpyrenediol epoxide has been investigated repeatedly (5-8), but other reports now indicate that DNA can also chemically activate the antitumor antibiotics CC-1065 (9, 10) and neocarzinostatin chromophore (11).For RNA, splicing reactions (12-14) and most recently esterase activity (15, 16) have received considerable attention. This letter reports the discovery of a silyl phenol ether that is converted by the environment of duplex DNA to a reactive electrophile available for alkylation of a target sequence. Quinone methide formation is often employed in the design of enzyme inactivators (17-20) and is also associated with the biological activity of DNA-targeted drugs such as mitomycin (21-24) and anthracyclins (25-27). Our laboratory has recently studied model systems that form related intermediates under control of local pH (28, 29), reducing agents (301, and near-UV irradiation (30). A silyl-protected phenol provided a complementary source of quinone methide (Scheme 1). Stability of such a precursor is greatest a t physiological pH (311, and quinone methide generation is easily triggered by addition of fluoride (32). Affinity modification based on this chemistry exhibited the expected properties: (a) only a target sequence of DNA was alkylated; (b) modification was initiated by fluoride; and ( c ) the alkylating species was produced transiently (33). Surprisingly, target modification was also evident when fluoride was replaced with chloride, bromide, phosphate, or perchlorate (33). This result has since been investigated below and shown t o depend uniquely on the duplex formed by hybridization of probe and target strands. + Current address: Department of Chemistry, Scripps Research Institute. Abstract published in Advance ACS Abstracts, October 1, 1994. @
Scheme 1 ,TBDMS 0
on
0
0 lb, 2b
la, 2s
'+
TBDMS = Si
I
X
=
O
O
N
O
2
To distinguish between the intrinsic and inducible reactivity of the silyl phenol ether, its stability and modification were examined first in the absence of DNA, and then attached to a single-stranded oligodeoxynucleotide, and finally held adjacent to duplex DNA. In each system, the fluoride-dependent activity served as a positive control for generating the quinone methide (Scheme 1). In the presence of 4 mM KF, desilylation of the low molecular weight model l a occurred with a halflife of 4 min as observed by lH NMR (Figure 1). The resulting phenol lb was moderately stable under these conditions and could even be isolated after silica gel flash chromatography albeit in low yield (unoptimized, 9%).l Elimination of nitrophenol (tl/z > 60 h) and addition of water to form Id proceeded slowly under conditions equivalent to those of Figure 1.2Thus, as expected, the silyl phenol moiety was chemically competent for fluorideinduced reaction in the absence of DNA. 'Synthesis and characterization of la, lb, and Id are included in the supplementary material. Preparation of 2a was reported previously (33).
1043-1802/94/2905-0497$04.50/00 1994 American Chemical Society
Li et al.
498 Bioconjugate Chem., Vol. 5, No. 6,1994 (B) alkylation of
(A) stability of probe strand
target strand
2a + KF 2a + 3
U U
KF
+
11
incubate 1-30 min
incubate 1-30 min
assay (rx with 3) 10 20 30
1
min
1
10 20
30
x-link
iane
2
1
3
4
5
6
7
8
Figure 2. Autoradiograms of denaturing polyacrylamide (20%) gels used to monitor nonproductive consumption of singlestranded 2a vs interstrand cross-linking of 2a 3 in the presence of fluoride. For lanes 1-4,2a (9 nM)was preincubated with 200 mM KF in 1 mM MES pH 7 for the indicated times before addition of 5’-[32Pl-3(9 nM). These final solutions were quenched after 30 min. For lanes 5-8, oligodeoxynucleotides 2a and 5’-[32P]-3(9 nM each) were hybridized for one min in 1 mM MES pH 7 before addition of 200 mM KF. Samples were then incubated under ambient temperature and quenched (33) a t the indicated times.
+
0
2
6
4
8
10
12
14
16
18
20
Time (minutes)
Figure 1. Fluoride dependent reaction of la. Conversion of la to l b was monitored by integrating the NMR signals of the benzylic protons of these compounds (5.21 and 5.16 ppm, respectively). Remaining starting material was determined as the fraction of la/(la lb) and this was fit to a first order process using nonlinear regression (-). Reaction was initiated by adding 4 mM KF to a solution of la (0.5 mM), 70 mM morpholinoethansulfonate(MES) and 2.5 M CD3CN in D20 pD 6.5 at 20 “C.
+
The fluoride-initiated activity of the oligonucleotide conjugate 2a alone and in complex with a complementary target, 3,5’-d(AGTGCCACCTGACGTCTAAG), was also consistent with the activation process outlined in Scheme 1. In the absence of a target strand, the electrophilic intermediate 2c was consumed via solvent, and possibly intramolecular, reaction. This process was illustrated by treating 2a with fluoride for 1-30 min and then observing the mixture’s diminished ability to form interstrand cross-links upon subsequent addition of 3 (Figure 2A). Both the decomposition of single strand 2a and the crosslinking of duplex 2a 3 (Figure 2B) were induced by fluoride with similar efficiency. This suggests the common transformation of 2a to 2c controlled both reactions independent of the surrounding DNA structure. In contrast, neither the model silyl phenol (la)nor its single-stranded derivative (2a)exhibited any intrinsic or ion-inducible reactivity in the absence of fluoride. Compound la did not undergo modification when another salt such as LiC104 (4-100 mM) replaced KF under the conditions described in Figure 1. NMR analysis of such incubations revealed that la persisted beyond 10 days (20 “C) without change, and no trace of desilylation or substitution was detected. A related model lacking the carboxamide group was also inert in the presence of 100 mM LiC104 or NaCl(34). Similarly, these salts did not promote any of the possible solvent and intramolecular reactions that consumed the single-stranded reagent 2a after its exposure to KF. Incubation of 2a with NaCl in place of KF did not diminish its later ability to alkylate 3 (Figure 3A). Consequently, the silyl-protected phenol of 2a expressed no spontaneous reactivity as a single strand.
+
Reaction of the model lb proceeded more slowly than the related reaction of 2b, oligodeoxynucleotidecross-linking, in part due to solvent effects (Q. Zeng, unpublished observations). The solubility of 1 was limited under aqueous conditions, and therefore, a mixed organidaqueous system was used during its investigation. The oligodeoxynucleotide derivative 2 was examined under aqueous conditions.
(B) alkylation of target strand
(A) stability of probe strand
2a
t
NaCl
2a
U
+3+
NaCl
11
incubate 1-240 min
incubate 1-240 min
8
assay (rx with 3) min
1
10 30 60 120240
1 10 30 60 120 240
x-link 5pp1-3
lane
1
2
3
4
5
6
7
8
9
10
11
12
Figure 3. Autoradiograms of denaturing polyacrylamide (20%) gels used to monitor nonproductive consumption of singlestranded 2a vs interstrand cross-linking of 2a 3 in the presence of NaCl. For lanes 1-6,2a (9 nM) was preincubated with 200 mM NaCl in 1 mM MES pH 7 for the indicated times before addition of 5’-[32Pl-3(9 nM). These final solutions were quenched after 240 min. For lanes 7- 12, oligodeoxynucleotides 2a and 5’-[32P]-3(9 nM each) were hybridized for one min in 1 mM MES pH 7 before addition of 200 mM NaC1. Samples were then incubated under ambient temperature and quenched (33) at the indicated times.
+
Duplex formation played a n obligatory role in the activation of the phenol derivative when fluoride was excluded. Hybridization of 2a and its complement 3 effected strand cross-linking under conditions that did not promote any reaction of single-stranded 2a (Figure 3A vs 3B). Probe-target modification was observed in the presence of NaCl (Figure 3B) albeit at a rate lower than that induced by KF at equivalent ionic strength. Ultimately, both conditions produced the same yield of crosslinking (ca. 30%) (33). These results are not consistent with simple affinity modification since la and 2a exhibited no intrinsic reactivity that could be directed to a DNA sequence. Furthermore, a mechanism involving direct displacement could not explain both the similar reactivity of 2a and 2a 3 in the presence of fluoride and the dissimilar reactivity of 2a and 2a 3 in the absence of fluoride. Instead, the structure established by 2a 3 most likely converted the attached silyl phenol ether into a n electrophilic intermediate for target alkylation. This transformation could mimic the fluoride-dependent mechanism
+
+
+
Bioconjugafe Chem., Vol. 5, No. 6, 1994 499
Letters Scheme 2
no deconposhion
t
*
YTBDMS
0
J
target alkylation
by effecting loss of the silyl group and formation of the quinone methide (Scheme 1). Such a n analogous pathway would also account for the similarity in cross-linking yields induced by fluoride and strand hybridization. Selective modification of DNA is traditionally based on preferential binding and orientation of reactive functional groups (35). However, the intrinsic activity of such groups often causes in vivo modification of nontargeted components of a cell or organism. The inducible alkylation demonstrated here should now minimize these undesirable events since the reactive species is generated only after target recognition (Scheme 2). ACKNOWLEDGMENT
This research was generously supported by the Center for Biotechnology, State University of New York a t Stony Brook, in conjunction with the New York State Science and Technology Foundation. We thank the Mass Spectrometry Facility of UC Riverside for mass spectral analysis. Supplementary Material Available: Preparation and characterization of la, lb, and Id (4 pages). Ordering information is given on any current masthead page. LITERATURE CITED (1) Dervan, P. B. (1991) Characterization of Protein-DNA Complexes by Affinity Cleaving. Methods Enzymol. 208,497513. (2) Uhlmann, E., and Peyman, A. (1990) Antisense Oligonucleotides: A New Therapeutic Principle. Chem. Rev. 90, 543584. (3) Sigman, D. S., Bruice, T. W., Mazumder, A., and Sutton, C. L. (1993) Targeted Chemical Nucleases. Acc. Chem. Res. 26, 98-104. (4) Silver", R. B. (1988) Mechanism-Based Enzyme Znactivation: Chemistry and Enzymology, Vols. I and 11, CRC Press, Boca Raton.
(5) Kootsra, A., Haas, B. L., and Slaga, T. J. (1980) Reaction of Benzo[a]-pyrene Diol-Epoxides with DNA and Nucleosomes in Aqueous Solution. Biochem. Biophys. Res. Commun. 94, 1432-1438. (6) Geacintov, N. E., Yoshida, H., Ibanez, V., and Harvey, R. G. (1982) Noncovalent Binding of 7/3,8a-Dihydroxy-Sa,lO/3epoxytetrahydrobenzo[alpyrene to Deoxyribonucleic Acid and Its Catalytic Effect on the Hydrolysis of the Diol Epoxide to Tetrol. Biochemistry 21, 1864-1869. ( 7 ) Michaud, D. P., Gupta, S. C., Whalen, D. L., Sayer, J . M., and Jerina, D. M. (1983) Effects of pH and Salt Concentration on the hydrolysis of a Benzo[alpyrene 7,8-Diol-9,10-Epoxide Catalyzed by DNA and Polyadenylic Acid. Chem.-Biol. Znteract. 44, 41-52. ( 8 ) Huang, C.-R., Milliman, A., Price, H. L., Urnao, S., Fetzer, S. M., and LeBreton, P. R. (1993) Photoemission Probes of Catalysis of Benzo[a]pyrene Epoxide Reactions in Complexes with Linear, Double-Stranded and Closed-Circular, SingleStranded DNA. J . Am. Chem. Soc. 115, 7794-7805. (9) Lin, C. H., Beale, J. M., and Hurley, L. H. (1991) Structure of the (+)-CC-1065-DNA Adduct: Critical Role of Ordered Water Molecules and Implications for Involvement of Phosphate Catalysis in the Covalent Reaction. Biochemistry 30, 3597-3602. (10) Boger, D. L., Munk, S. A., Zarrinmayeh, H. (1991) (+)-CC1065 DNA Alkylation: Key Studies Demonstrating a Noncovalent Binding Selectivity Contribution to the Alkylation 113, 3980-3983. Selectivity. J . Am. Chem. SOC. (11) Kappen, L. S., Goldberg, I. H. (1993) DNA Conformationinduced Activation of an Enediyne for Site-Specific Cleavage. Science 261, 1319-1321. (12) Cech, T. R. (1990) Self-splicing of Group I Introns. Annu. Rev. Biochem. 59, 259-271. (13) Pace, N. R., and Smith, D. (1990) Ribonuclease P: Function and Variation. J . Biol. Chem. 265, 3587-3590. (14) Symons, R. H. (1992) Small Catalytic RNAs. Annu. Rev. Biochem. 61,641-647. (15) Noller, H. F., Hoffarth, V., and Zimniak, L. (1992) Unusual Resistance of Peptidyl Transferease to Protein Extraction Procedures. Science 256, 1416-1419. (16) Piccirilli, J. A., McConnell, T. S., Zaug, A. J., Noller, H. F., and Cech, T. R. (1992) Aminoacyl Esterase Activity of the Tetrahymena Ribozyme. Science 256, 1420-1424. (17) BBchet, J.-J., Dupaix, A., Yon, J., Wakselman, M., Robert, J . C., and Vilkas, M. (1973) Inactivation of a-Chymotrypsin by a Bifunctional Reagent, 3,4-Dihydro-3,4-dibromo-6-bromomethyl Coumarin. Eur. J . Biochem. 35, 527-59. (18) Harper, J. W., and Powers, J. C. (1985) Reaction of Serine Proteases with Substituted 3-Alkoxy-4-chloroisocoumarines and 3-Alkoxy-7-amino-4-chloroisocoumarines: New Reactive Mechanism-Based Inhibitors. Biochemistry 24, 7200-7213. (19) Lee, C.-H., and Skibo, E. B. (1987) Active-Site-Directed Reductive Alkylation of Xanthine Oxidase by Imidazo[4,5-g]quinazoline-4,9-diones Functionalized with a Leaving Group. Biochemistry 26, 7355-7362. (20) Myers, J . K., and Widlanski, T. S. (1993) Mechanism-Based Inactivation of Prostatic Acid Phosphatase. Science 262, 1451-1453. (21) Tomasz, M., and Lipman, R. (1981) Reductive Metabolism and Alkylating Activity of Mitomycin C Induced by Rat Liver Microsomes. Biochemistry 20, 5056-5061. (22) Egbertson, M., and Danishefsky, S. J . (1987) Modeling of the Electrophilic Activation of Mitomycins: Chemical Evidence for the Intermediacy of a Mitosene Semiquinone as the Active Electrophile. J. Am. Chem. Soc. 109, 2204-2205. (23) Tomasz, M., Chawla, A. K., and Lipman, R. (1988) Mechanism of Monofunctional and Bifunctional Alkylation of DNA by Mitomycin C. Biochemistry 27, 3182-3187. (24) Han, I., Russell, D. J.,and Kohn, H. (1992) Studies on the Mechanism of Mitomycin C (1) Electrophilic Transformations: Structure-Reactivity Relationships. J . Org. Chem. 57, 1799-1807. (25) Fisher, J., Ramakrishnan, K., and Becvar, J . E. (1983) Direct Enzyme-Catalyzed Reduction of Anthracyclines by Reduced Nicotinamide Adenine Dinucleotide. Biochemistry 22, 1347-1355.
500 Bioconjugate Chem., Vol. 5, No. 6, 1994 (26) Gaudiano, G., Frigerio, M., Bravo, P., and Koch, T. H. (1990) Intramolecular Trapping of the Quinone Methide from Reductive Cleavage of Daunomycin with Oxygen and Nitrogen Nucleophiles. J . Am. Chem. SOC.112, 6704-6709. (27) Schweitzer, B. A,, and Koch, T. H. (1993) Synthesis and Redox Chemistry of 5-Deoxydaunomycin. A Long-Lived Hydroquinone Tautomer. J . A m . Chem. SOC.115, 5440-5445. (28) Woolridge, E. M., Rokita, S. E. (1991) 6-(Difluoromethyl)tryptophan as a Probe for Substrate Activation during the Catalysis of Tryptophanase. Biochemistry 30, 1852-1857. (29) Woolridge, E. M., and Rokita, S. E. (1991) The Use of 6-(Difluoromethyl)indole to Study the Activation of Indole by Tryptophan Synthase.Arch. Biochem. Biophys. 286,473-480. (30) Chatterjee, M., and Rokita, S. E. (1994) The Role of a Quinone Methide in the Sequence Specific Alkylation of DNA. J. A m . Chem. SOC.116, 1690-1697.
Li et al. (31) Shirai, N., Moriya, K., and Kawazoe, Y. (1986) pH Dependence of Hydrolytic Removal of Silyl Group from Trialkylsilyl Ethers. Tetrahedron 42, 2211-2214. (32) Marino, J. P., and Dax, S. L. (1984) A n Efficient Desilylation Method for the Generation of o-Quinone Methides: Application to the Synthesis of (+I- and (-)-Hexahydrocannabinol. J. Org. Chem. 49, 3671-3672. (33) Li, T., and Rokita, S. E. (1991) Selective Modification of DNA Controlled by a n Ionic Signal. J . Am. Chem. SOC.113, 7771-7773. (34) Li, T. (1992) Sequence Selective Modification of DNA with a Silyl Phenol Ether-Oligodeoxynucleotide Conjugate, Ph.D. Dissertation, The State University of New York at Stony Brook. (35) Warpehoski, M. A., and Hurley, L. H. (1988) Sequence Selectivity of DNA Covalent Modification. Chem. Res. Toxic01 1, 315-333.