Anal. Chem. 1996,67, 1010-1013
Platinum Anticancer Drug Binding to DNA Detected by Thickness-Shear-ModeAcoustic Wave Sensor Hongbo Su, Patrick Williams, and Michael Thompson* Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario M5S 1A 1, Canada
Nucleic acid has been attached to the electrodes of thickness-shear-modeacoustic wave devices to produce a biosensor for platinum-based drugs. The decreases in series resonant frequency for interactions of DNA with both cis- and transplatin are indicative of two distinct kinetic processes. The results of a kinetic analysis are interpreted in terms of nucleic acid binding of the hydrolysis products of the two drugs. Concentration-dependent decreases of series resonant frequency show that the limit of detection for the drugs is approximately lo-‘ M. Motional resistance changes for nucleic acid-drug interactions also convey information regarding the chemistry of the macromolecules at the interface. Nucleic acid-binding molecules have found extensive use as chemotherapeutic agents. A number of these species owe their usefulness to their ability to covalently link together or to crosslink bases of the same strand, of different strands, or of different helices. cis-Diaminedichloroplatinum(II)(cisplatin) is an example of one of these agents,’ in that its antitumor activity has been employed for the treatment of ovarian, bladder, and testicular cancer.2 The interaction of DNA with cisplatin has been extensively studied over the past 20 years by a variety of techniques, including NMR spectroscopy and X-ray diffra~tion.~ Only the GG and AG adducts arise from cisplatin analogs from the six different kinds of major reaction products between cisplatin and DNA These adducts are coordinated through the N7 position and the drug in a “head-to-head orientation. The trans isomer (transplatin), which differs only in its ligand coordination geometry, is less mutagenic and cytotoxic and is ineffective as an antitumor agent. The reasons for this are still not fully understood. The binding of cis- and transplatin to DNA is kinetically rather then thermodynamically controlled. Four steps are involved in the DNA-platin binding p r o ~ e s s .The ~ initial step, involving complexation, is hydrolysis of the first chloride ion (reaction I), which is also the rate-limiting reaction. Second, the complex coordinates primarily to the N7 positions of guanine and adenine bases in the major groove to form adducts (reaction ID. The third step is associated with the further hydrolysis of monofunctional adducts (reaction III). Finally, these species react with a second nucleophile, forming primarily intrastrand adducts (reaction IV) 51!
(1) Lippard, S. J. In Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy; Howell, S. B., Ed.; Plenum Press: New York, 1991; pp 1-6. (2) Sherman, S. E.: Lippard, S. J. Chem. Rev. 1987,87, 1153. (3) Bancroft, D. P.; Lepre, C. A,; Lippard, S. J. J. Am. Chem. SOC.1990,112, 6860.
1010 Analytical Chemistry, Vol. 67, No. 5,March 1, 1995
Although the structure of the adducts has received signiticant attention, little work has been performed on the direct detection of the drugs in a biosensor format, Le., where a real-time signal is obtained for adduct formation on a transducer surface. Research in our laboratory has centered on the role of interfacial physical chemistry in determining the response of the thicknessshear-mode (TSM) acoustic wave sensor in the liquid pha~e.68~ Particular attention has been paid to the potential application of acoustic network analysis.8~9In recent articles, we have demonstrated the direct real-time detection of RNA homopolymerloand plasmid DNA hybridization” at the TSM sensor-liquid interface. Additionally, the kinetics of nucleic acid hybridization at a solidliquid interface have been examined by acoustic network analysis.12 The present work is concerned with the direct monitoring of DNA-platin drug interactions through measurement of timedependent changes in series resonant frequency. Furthermore, we describe alteration in motional resistance obtained from acoustic network analysis. EXPERIMENTAL SECTION
Reagents. Double-strand calf thymus DNA, cisplatin, and transplatin were obtained from Sigma (St. Louis, MO) and used without further purification. The Millipore water with specific resistance 18.2 MSZ cm was employed during all the experiments. Apparatus. AT-cut SMHz piezoelectric crystals with gold electrodes were supplied by International Crystal Manufacturing Co., Oklahoma City, OK. Palladium was sputtered onto the gold electrodes using a Perkin-Elmer Ultek 2400.8SA radio frequency sputtering system. The instrument used to characterize TSM devices in liquid was an HP 4195A network/spectrum analyzer (Hewlett-Packed, Palo Alto, CA). The values of the equivalent circuit elements for the device are calculated internally by the analyzer from measured data. The TSM sensor was housed in a cell, the construction of which is described elsewhere.11 One side of the crystal was immersed in solution, while the other face was kept under nitrogen. The two halves of the cell encasing the device are separated by O-rings and Teflon. Reedijk, J. In Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy; Howell, S. B., Ed.: Plenum Press: New York, 1991: pp 1322. Eashnan, A; Barry, M. A. Biochemistry 1987,26, 3303. RajakoViC, Lj.V.; CaviE-Vlasak, B. A; Ghaemmaghami, V.: Kallury, IC M. R; Kipling, A L.; Thompson, M. Anal. Chem. 1991,63, 615. Yang, M.; Thompson, M. Anal. Chem. 1993,65, 1158. Kipling, A L.; Thompson, M. Anal. Chem. 1990,62,1514. Yang, M.;Chung, F. L.; Thompson, M. Anal. Chem. 1993,65,3713. Su, H.; Yang, M.; Kallury, IC M. R; Thompson, M. Analyst 1993,118, 309. Su, H.; Kallury, K. M. R; Thompson, M.; Roach, A Anal. Chem. 1994,66, 769. Su, H.;Thompson, M. Biosens. Bioelectron, in press. 0003-2700/95/0367-1010$9.00/0 0 1995 American Chemical Society
200 a
0
A
-200 -400
. -
$1000
N
c . . .... . .. . .. . . .
8
% 100
8.96
8.98
9.00
9.02
-1000' 400 I
1
0
1
2
3
4
5
6
6
7
9.04
Frequency (MHz)
CO
L Rm
Lm
Cm
Figure 1. Typical plots of the magnitude (14)and phase angle (e) of impedance and equivalent electrical circuit of a 9-MHz TSM sensor upon water loading. (a) Measured phase angle of the impedance. (b) Simulated phase angle from the equivalent circuit. (c) Measured magnitude of the impedance. (d) Simulated impedance from the equivalent circuit.
Immobilization of DNA on TSM Sensor Surfaces. A 40pL sample of a double-strand calf thymus DNA solution (1 mg mL-l) was placed on the palladium electrodes of TSM devices. The DNA binding capacity onto the sensor surfaces can be increased by oven heating and UV cross-linking. The sensor was then rinsed copiously with water, followed by incubation in 5 mL of water overnight. Previous work has show that nucleic acid is strongly bound to the palladium oxide surface of the electrodes.lOJ1 DNA-Cis- or Transplatin Interaction and Acoustic Network Analysis. In single concentration studies, 1mM cisplatin and 6 mM transplatin solutions were injected into the measurement cell which housed the DNA-sensor combination. The devices were pretreated in water prior to drug injection. Control experiments with a non-DNAcoated device were performed according to the same procedure. Experiments involving variable concentrations of drug were performed according to following procedure. First, the one face of a TSM sensor was exposed to Millipore water while the other side was kept under nitrogen gas. The steady series resonant frequencyf, was obtained after 5-20 min. The water was then drained off, and a 1pM solution of drug was used in situ to wash the TSM sensor three times and then injected into the same side. This final addition of drug to the cell constituted the zero time reference point for the acoustic network measurement. The steady-state series resonant frequency was measured again after the sample was incubated for certain periods of time. Similar procedures were employed with higher concentrations of cisplatin and transplatin solutions. RESULTS AND DISCUSSION
Before discussing the results obtained for the drug-DNA studies, we briefly review the liquid-phase operation of TSM
-800
- 1200 - 1600 -2000
1 i x
I'
-1
i
B 0
1
2
3
4
5
Time (1000 secs) Figure 2. Series resonant frequency shift vs time for exposure of sensors to (a) cisplatin and (b) transplatin. Plot B is for sensor with surface-coated calf thymus DNA, and plot A is for a bare PdO electrode.
sensors. (More detailed treatments of piezoelectric studies can be found e1~ewhere.l~ ) A piezoelectric material placed in an electric field will experience strains which are proportional to the strength of the field and will change sign if the direction of the field is reversed. The final result of equations of motion analysis is an expression for the current density flowing through the crystal in terms of the voltage across the device. The equivalent circuit of a quartz crystal is shown in Figure 1. The four circuit elements are the motional resistance, R,, motional inductance, L,,,,motional capacitance, ,C, and parallel capacitance, C, for one mode of vibration of the crystal. As shown in Figure 1,the series resonant frequency, f,, is the lower frequency and the parallel resonant frequency, fp, is the higher frequency at which impedance is a pure resistance. Figure 1 also depicts typical plots of the magnitude and phase angle of impedance of a SMHz TSM sensor in the solution phase. The four equivalent circuit parameters can also be extracted from the impedance measurements by standard circuit analysis. Typical real-time plots of change in f, during cis- and transplatin-DNA interaction are shown in Figure 2, together with the results of a control experiment involving a bare PdO electrode surface. The latter exhibits a small decrease of about 50 Hz over 1h, which we attribute to the adsorption of drug onto the electrode surface. The total change in f, for the coated sensor is in the range of several hundred hertz for each drug, conliming the DNA-platin interaction at the sensor-liquid interface. The responses with time for both molecules are reflective of two (13) Bottom, V. E. Introduction to Quartz Ctystal Unit Design; Van Nostrand Reinhold: New York, 1982.
Analytical Chemistry, Vol. 67, No. 5, March 1, 1995
1011
Table 1. Comparison of Kinetics for the DNA-Platln Reactions In Solution and at the Solid-Solution Interface
reaction I I1 111
Iv
rate parameters solid- solution interfaceb solution phasea cisplatin transplatin cisplatin transplatin s-l) tl/z (min) k s-l) tl/z (min) k s-l) t l / z (min) k s-l) t 1 / 2 (min)
k
1.2 0.10
20 222
2.2 0.14
0.102 1.96 0.092
11 174
114 6 126
