Dna interactions of two clinical camptothecin drugs stabilize their active lactone forms

Full text of DOI:10.1021/ja973433j

J. Am. Chem. Soc. 1998, 120, 2979-2980
2979
DNA Interactions of Two Clinical Camptothecin
Drugs Stabilize Their Active Lactone Forms
Danzhou Yang,†,‡ J. Thompson Strode,†,‡
H. Peter Spielmann,‡,§ Andrew H.-J. Wang,| and
Thomas G. Burke*,†,‡
DiVision of Medicinal Chemistry and Pharmaceutics
College of Pharmacy
Department of Biochemistry, College of Medicine
Department of Chemistry, College of Arts and Sciences
Markey Cancer Center, UniVersity of Kentucky
Lexington, Kentucky 40506
Figure 1. Kinetic evaluation of the rate of lactone ring opening for
topotecan (left panel) and CPT-11 (right panel) in the presence and
absence of duplex DNA oligonucleotides ((dG-dC)15 or (dA-dT)15). Data
for drug in the absence of DNA (1), drug in the presence of(dG-dC)15
(b) or (dA-dT)15 (O) are shown. All experiments were conducted in PBS
(pH 7.40 ( 0.05) at room temperature. Drug and DNA concentrations
of 10 µM and ∼30 mM base, respectively, were employed. Each profile
represents the average of at least three independent kinetic runs with the
same sampling schedules. The standard deviation of each point was
typically 5% or less.
Department of Cell & Structural Biology
UniVersity of Illinois at Urbana-Champaign
Urbana, Illinois 61801
ReceiVed October 2, 1997
Camptosar (CPT-11) and Hycamtin (topotecan, TPT), shown
in Figure 1, are two clinically useful anticancer drugs of the
camptothecin family which function by inhibiting human DNA
topoisomerase I (TopoI).1 Successful inhibition of TopoI by
camptothecins is known from structure-activity studies to require
an intact lactone ring (ring E) functionality.2-4 Unfortunately,
this lactone moiety is subject to hydrolysis under physiological
conditions (i.e., at pH 7 and above) with each camptothecin agent
existing in equilibria with its corresponding ring-opened car-
boxylate form. The position of the equilibria is pH-dependent,
with the carboxylate form predominating at physiological pH after
1 h of incubation.5 Previous equilibrium dialysis studies which
evaluated the interactions of extensively incubated (16 h) and
hydrolyzed camptothecin with sonicated calf thymus DNA and
plasmid DNA (2 mM base concentrations or less) provided little
to no evidence of DNA binding.6 In this paper, we investigate
the effect of synthetic duplex DNA oligonucleotides on CPT-11
and TPT stability as a function of incubation time and DNA
concentration using a combination of HPLC, UV, and NMR
methods. Our studies demonstrate that the positively charged and
water-soluble TPT and CPT-11 congeners, as well as uncharged
camptothecin, are capable of interacting directly with double-
stranded DNA (dsDNA). Moreover, our results indicate that the
dsDNA interactions of the camptothecin drugs of interest result
in a marked stabilization of their active lactone forms.
in human TopoI). Camptothecin agents are thought to interact
with this covalent intermediate at the nick site, thereby preventing
the religation and ultimately leading to DNA fragmentation and
cell death.1 As a result of intense study following their discovery
as TopoI inhibitors,1-4,6,9 the camptothecins are now known to
act through the formation of stable ternary complexes between
drug, TopoI, and DNA.1 However, strong evidence that these
agents are capable of interacting directly with DNA in the absence
of TopoI has not been identified until the present study.
The impact of two different 30mer dsDNA oligonucleotides,
(dA-dT)15 and (dG-dC)15, on drug stability in phosphate-buffered
(9) (a) Hertzberg, R. P.; Caranfa, M. J.; Holden, K. G.; Jakas, D. R.;
Gallagher, G.; Mattern, M. R.; Mong, S.-M.; Bartus, J. O.; Johnson, R. K.;
Kingsbury, W. D. J. Med. Chem. 1989, 32, 715-720. (b) Jaxel, C.; Kohn, K.
W.; Wani, M. C.; Wall, M. E.; Pommier, Y. Cancer Res. 1989, 49, 1465-
1469. (c) Pommier, Y.; Kohlhagen, G.; Kohn, K. W.; Leteurtre, F.; Wani, M.
C.; Wall, M. E. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 8861-8865. (d) Crow,
R. T.; Crothers, D. M. J. Med. Chem. 1992, 35, 4160-4164.
(10) Phosphate-buffered saline (PBS) contained 8 mM Na2HPO4, 1 mM
KH2PO4, 137 mM NaCl, and 3 mM KCl (pH 7.4 unless specified otherwise).
The oligonucleotides d(CGTACG) were synthesized on an automated DNA
synthesizer at the Genetic Facility of UIUC. (dG-dC)15 and (dA-dT)15 were
purchased from IDT (Coralville, IA). CPT-11 was generously provided by
Yakult (Tokyo), and TPT was obtained from the NCI. The drug lactone stock
solutions were prepared in aqueous solution at 2 mM, pH 5. Solutions of the
various drug-DNA complexes for NMR studies were prepared by mixing
the appropriate amounts of drug stock solution and DNA stock solution in
PBS, followed by pH adjustment to the desired value. Drug-DNA complexes
solutions were vacuum-dried in a SpeedVac at room temperature and then
dissolved in 0.5 mL 99.8% D2O for 1D 1H NMR spectra or in 0.5 mL
90%H2O/10%D2O for 1D H2O spectra. The final DNA duplex concentrations
ranged between 0.7 and 1 mM for all 1D NMR spectra. 1D H2O NMR spectra
were collected using the 1-1 pulse sequence (Sklenar, V.; Brooks, B. R.;
Zon, G.; Bax, A. FEBS Lett. 1996, 216, 249-252). The NMR spectra were
recorded on Varian VXR500 (University of Illinois, Urbana, IL) and Varian
Inova 500 (University of Kentucky, Lexington, KY) 500 MHz spectrometers.
The chemical shifts (in ppm) were referenced to the HDO peak which was
calibrated to a sodium 2,2-dimethyl-2-silapentane-5-sulfonate (DSS) external
standard at different temperatures. The NMR data were processed with the
program FELIX v.1.1 (Hare Research, Woodinville, WA) or FELIX 95.0
(MSI) on Silicon Graphics workstations. The UV spectrum of a drug-DNA
complex was obtained by subtracting the spectrum of free DNA from the
spectrum of the complex. The hydrolysis kinetics of both drugs were
determined by the quantitative reversed-phase high-performance liquid
chromatographic (HPLC) methods as decribed previously (Mi, Z.; Malak, H.;
Burke, T. G. Biochemistry 1995, 34, 13722-13728. Warner, D. L.; Burke,
T. G. J. Chromatogr. B 1997, 691, 161-171). DNA stock solutions were
prepared by dissolving the oligos in PBS at a concentration of ∼30 mM base
concentration, with adjustment of pH to 7.4; 5 µL of 1 mM drug stock solutions
(either lactone form or carboxylate form) were then added to 0.5 mL of pH
7.4 PBS or DNA stock solutions and assayed by HPLC to determine lactone
stability or relactonization parameters, respectively.
In the absence of drug, TopoI mediates the relaxation of
supercoiled DNA. TopoI first binds DNA and then nicks it on
one strand, rotates the helix by one turn and finally rejoins the
nicked strand.1,7,8 The nicking of DNA by TopoI creates a
covalent intermediate in which the 3′-phosphate at the nick site
is attached to the phenolic hydroxyl group of a tyrosine (Tyr723
† College of Pharmacy, University of Kentucky.
‡ Markey Cancer Center, University of Kentucky.
§ Colleges of Medicine and Arts and Sciences, University of Kentucky.
| University of Illinois at Urbana-Champaign.
(1) (a) Liu, L. F.; Duann, P.; Lin, C.-T.; D’Arpa, P.; Wu, J. Ann. N.Y.
Acad. Sci. 1996, 803, 44-49. (b) Takimoto, C. H.; Arbuck, S. G. Cancer
Chemother. Biother. 1996, 463-484. (c) Slichenmeyer, W. J.; Rowinsky, E.
K.; Donehower, R. C.; Kaufmann, S. H. J. Natl Cancer Inst. 1993, 85, 271-
291.
(2) Wani, M. C.; Nicholas, A. W.; Manikumar, G.; Wall, M. E. J. Med.
Chem. 1987, 30, 1774-1779.
(3) Wani, M. C.; Nicholas, A. W.; Wall, M. E. J. Med. Chem. 1987, 30,
2317-2319.
(4) Jaxel, C.; Kohn, K. W.; Wani, M. C.; Wall, M. E.; Pommier, Y. Cancer
Res. 1989, 49, 5077-5082.
(5) Fassberg, J.; Stella, V. J. J. Pharm. Sci. 1992, 81, 676-684.
(6) Hertzberg, R. P.; Caranfa, M. J.; Hecht, S. M. Biochemistry 1989, 28,
4629-4638.
(7) Pommier, Y. Cancer Chemother. Pharmacol. 1993, 32, 103-108.
(8) Henningfeld, K. A.; Arslan, T.; Hecht, S. M. J. A. Chem. Soc. 1996,
118, 11701-11714.
S0002-7863(97)03433-1 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/12/1998

2980 J. Am. Chem. Soc., Vol. 120, No. 12, 1998
Communications to the Editor
Figure 2. UV absorption spectra of topotecan lactone (left panel) and
CPT-11 lactone (right panel) in the absence and presence of duplex DNA
oligonucleotides ((dG-dC)15 or (dA-dT)15) in PBS (pH 5.00 ( 0.05 at
room temperature). Drug, (dG-dC)15, and (dA-dT)15 concentrations of
10 µM, 16 mM base, and 24 mM base, respectively, were employed.
saline10 (PBS) solution at pH 7.4 and room temperature is depicted
in Figure 1. A reversed-phase HPLC assay was employed to
assess the rate of hydrolysis and quantify the proportion of the
lactone and carboxylate forms at equilibrium.10 In the absence
of DNA, 10 µM samples of CPT-11 and TPT incubated for 3 h
hydrolyzed with pseudo-first-order kinetics and displayed percent
lactone at equilibrium values of 24% and 12%, respectively. The
extent of lactone ring opening for both agents was markedly
reduced, however, in the presence of dsDNA (∼30 mM base
concentration). Lactone levels for CPT-11 remained at 95%
following 3 h of incubation in the presence of (dA-dT)15 or (dG-
dC)15, and at levels of 78% and 95%, respectively, for TPT. Even
for time points out to 5 days, lactone levels remained high (50-
60%) in the presence of DNA for both agents. Experiments
conducted with 10 µM camptothecin demonstrated that the lactone
form of this agent was also efficiently stabilized by the presence
of ∼30 mM base (dG-dC)15 with lactone levels remaining above
80% after 3 h.
Addition of pure carboxylate forms of CPT-11 and TPT to ∼30
mM base dsDNA solutions in PBS followed by incubation at room
temperature was found to result in relactonization and the
promotion of high levels of active lactone (∼30% after 30 h).
HPLC analyses have also been completed on solutions containing
single-strand DNA (ssDNA) in the form of ∼33 mM base (dT)30
(see Figure S1, Supporting Information). Samples containing
ssDNA showed significantly diminished effects on stabilizing both
TPT and CPT-11, with dsDNA solutions some 10- to 100-fold
more dilute than the (dT)30 solutions affording approximately
equal lactone stabilization. Figure 2 depicts changes in the UV
spectra of the lactone forms of CPT-11 and TPT in PBS (pH 5)
following addition of the (dA-dT)15 or (dG-dC)15 30mers.10 Note
that a substantial hypochromic and red spectral shift is observed
for both drugs in in the presence of DNA, suggestive of a possible
intercalation mode of binding. Consistent with the HPLC data,
the UV data also provides evidence that TPT displays a sequence
preference of (dG-dC)15 over (dA-dT)15, while no signs of any
sequence preference were observed for CPT-11.11
Figure 3. NMR spectra depicting the titration of (A) CPT-11 to
d(CGTACG) at pH 7.0, 2 °C and (B) TPT lactone to d(CGTACG) at pH
5 at 2 °C. The ratios shown in the figure are for drug/DNA.
the drugs to DNA at increasing ratios resulted in upfield shifts
of the DNA resonances, especially for the imino resonances G2
and G6. The drug resonances also shifted upfield in a correlated
fashion. For TPT, the aromatic resonances of the H7, H12, and
H11 protons shifted from 8.62, 7.87, and 7.43 ppm to 8.55, 7.73,
and 7.24 ppm, respectively. Such upfield shifts of both DNA
and drug resonances are characteristic of an intercalative mode
of drug binding.12,13 The unequivocal assignments of the
resonances associated with the lactone/carboxylate forms (Figure
3) also allow us to confirm the stabilizing effect which DNA
elicits on the lactone form of each agent.14 The 20-methyl
resonances of CPT-11 lactone occur at 0.95 ppm, which changes
to 1.06 ppm (20M) upon the formation of opened-ring species.
The NMR time course studies of CPT-11 and TPT concerning
the differential rates of ring opening in the presence and absence
of d(CGTACG) were consistent with determinations made using
the HPLC method.
In summary, we have shown that the active lactone forms of
CPT-11 and TPT are stabilized through interactions with dsDNA.
The presence of dsDNA, in fact, was found to promote the
conversion of inactive carboxylate to active lactone. Our results
thus provide the first evidence that duplex DNA devoid of TopoI
may play a functional role in the biological activities of the
camptothecins through the promotion of active lactone levels
within the cell nucleus. These results suggest that the agents,
upon reaching chromosomal DNA, may interact directly with
DNA prior to the action by TopoI (although the site of drug
binding to DNA is not necessarily at the site of topoisomerase I
action). The DNA-associated drugs are likely to be in their active
lactone forms and ready for the subsequent drug-DNA-enzyme
complex formation.
Further insight into the molecular interactions between CPT-
11 and TPT with dsDNA oligonucleotide d(CGTACG) was
obtained from NMR studies. Figure 3 shows that addition of
Acknowledgment. This work was supported by the NIH CA 63653
and American Cancer Society DHP-138 and RPG-95-059-03-CDD to
T.G.B., American Cancer Society RPG-94-014-04 to A.H.-J.W., and
American Cancer Society (IN-163) and the Kentucky-NSF EPSCoR
program (EHR-91-08764 and DE-FGO2-91ER75657) to H.P.S. We
gratefully acknowledge the support of the Center of Computational
Sciences, University of Kentucky.
(11) A stronger hypochromic effect shift in the UV spectra of TPT in the
presence of (dG-dC)15 vs (dA-dT)15 is consistent with a stronger interaction
with the former, albeit differences in transition-state dipole interactions could
also account for the observed effect.
(12) Feigon, J.; Denny, W. A.; Leupin, W.; Kearns, D. R. J. Med. Chem.
1984, 27, 450-465.
(13) Wang, A. H.-J. Curr. Opin. Struct. Biol. 1992, 2, 361-368
(14) 2D NOESY experiments have also been completed for both CPT-11
and TPT complexed to d(CGTACG) at different drug: DNA ratios of 1:1,
2:1, and 3:1.
Supporting Information Available: Figure S1 (2 pages, print/PDF).
See any current masthead page for ordering information and Web access
instructions.
JA973433J