cis-Dichloro(α,ω-diamino carboxylate ethyl ester)palladium(II) as Palladium(II) versus Platinum(II) Model Anticancer Drugs: Synthesis, Solution Equilibria of Their Aqua, Hydroxo, and/or Chloro Species, and in Vitro/in Vivo DNA-Binding Properties

Article abstract of DOI:10.1021/ic961226z

cis-Dichloro(d,l-2,3-diaminopropionate ethyl ester)palladium(II), cis-[Pd(Etdap)Cl₂] (I), and cis-dichloro(d,l-2,4-diaminobutyrate ethyl ester)palladium(II), cis-[Pd(Etdab)Cl₂] (II), were synthesized and characterized by elemental an

Full text of DOI:10.1021/ic961226z

1806
Inorg. Chem. 1997, 36, 1806-1812
cis-Dichloro(r,ω-diamino carboxylate ethyl ester)palladium(II) as Palladium(II) Wersus
Platinum(II) Model Anticancer Drugs: Synthesis, Solution Equilibria of Their Aqua,
Hydroxo, and/or Chloro Species, and in Vitro/in ViWo DNA-Binding Properties
M. L. Gonza´lez,1a J. M. Tercero,1a A. Matilla,*,1a J. Niclo´s-Gutie´rrez,1a M. T. Ferna´ndez,1a
M. C. Lo´pez,1b C. Alonso,1c and S. Gonza´lez1a
Departamento de Qu´ımica Inorga´nica, Facultad de Farmacia, Universidad de Granada,
E-18071 Granada, Spain, Instituto de Parasitolog´ıa y Biomedicina “Lo´pez-Neyra” (CSIC),
Granada, Spain, and Centro de Biolog´ıa Molecular “Severo Ochoa” (CSIC), Facultad de Ciencias,
Universidad Auto´noma de Madrid, E-28049 Madrid, Spain
ReceiVed October 10, 1996X
cis-Dichloro(d,l-2,3-diaminopropionate ethyl ester)palladium(II), cis-[Pd(Etdap)Cl2] (I), and cis-dichloro(d,l-2,4-
diaminobutyrate ethyl ester)palladium(II), cis-[Pd(Etdab)Cl2] (II), were synthesized and characterized by elemental
analysis, IR spectroscopy, and TG-DTA thermal analysis. The equilibrium model and log âpqr constants of
hydrolytic species of I and II were also investigated by potentiometric methods (I ) 0.15 M (NaClO4) and 37
°C). Aqueous solutions of cis-[Pd(Etdaa)(H2O)2](ClO4)2 (Etdaa ) Etdap (III) or Etdab (IV)) were prepared by
stoichiometric reaction of I or II with AgClO4. Thus, log âpqr of the corresponding cis-aquahydroxo (pqr )
1,0,-1) and di(µ-hydroxo) (pqr ) 2,0,-2) species were obtained from E(H+) data of alkalimetric titrations of
solutions of III or IV. Such constants were then used as fixed values to obtain log âpqr of chloro-containing
species (cis-dichloro (1,2,0), cis-chloroaqua (1,1,0), and cis-chlorohydroxo (1,1,-1)) from E(H+) and E(Cl-)
data pairs simultaneously obtained by titration of solutions of III or IV with NaOH and NaCl. All log âpqr were
fitted by the SUPERQUAD program. Appropriate log âpqr sets give good simulations of experimental titration
curves and several species distribution diagrams. In addition the interaction of I and/or II with DNA in Vitro and
in ViVo was studied by various methods. UV spectral data and melting and renaturation curves indicate that both
compounds destabilize the DNA helicoidal structure. Compound II disrupts in ViVo the structure and transcription
process of polytene chromosomes of Drosophila hydei.
Introduction
impels the cis-coordination to the more labile ligand (two
chloro5-11,13-19 or nitrato7 or one dicarboxylato.7 In such cases,
the chelating ligand prevents cis f trans isomerization around
Pd(II) in spite of its higher lability (approximately up to 105-
fold) relative to its analogous Pt(II). This essential kinetic
difference promotes the use of Pd(II) compounds as models for
the Pt(II) ones in mechanistic and kinetic4 or potentiometric5,9
and/or spectrophotometric studies.12,20
Research into cis-dichlorodiammineplatinum(II) (cis-DDP)
and related compounds has determined several structural features
that favor anticancer activity.2,3 Most active Pt(II) drugs are
neutral at plasma pH and chloride ion concentration (104 mM)
but give positively charged species upon hydrolytic reactions
at physiological pH and low [Cl-] ) 4 mM within cells.
Consequently, many active Pt(II) drugs have two cis leaving
groups (Cl-, NO3-, dicarboxylato) as well as two inert groups
(NH3, chelating diammines). Primary and secondary amines
(with N-H groups which form hydrogen bonds) are also
preferred.3
The remarkable analogy between Pt(II) and Pd(II) coordina-
tion stereochemistry has promoted studies of palladium(II)
compounds as anticancer drugs.4 Recent advances in this sense
have been focused on Pd(II) complexes with one N,N-
diamino5-12 or N,S-aminothioether13-19 chelating ligand, which
(7) Gill, D. S. In Platinum Coordination Complexes in Cancer Chemo-
therapy; Hacker, M. P., Douple, E. B., Krakhoff, I. K., Eds.; Martinus
Nijhoff: Boston, 1984; pp 267-278.
(8) Anderegg, G. Inorg. Chim. Acta 1986, 111, 25.
(9) Hohmann, H.; van Eldik, R. Inorg. Chim. Acta 1990, 174, 87.
(10) Navarro-Ranninger, C.; Zamora, F.; Pe´rez, J. M.; Lo´pez-Solera, I.;
Mart´ınez-Carrera, S.; Masaguer, J. R.; Alonso, C. J. Inorg. Biochem.
1992, 46, 267.
(11) Navarro-Ranninger, C.; Pe´rez, J. M.; Zamora, F.; Gonza´lez, V. M.;
Masaguer, J. R.; Alonso, C. J. Inorg. Biochem. 1993, 52, 37.
(12) Lomozik, L; Odani, A.; Yamauchi, O. Inorg. Chim. Acta 1994, 219,
107.
* Author to whom correspondence should be addressed. E-mail ad-
dress: amatilla@goliat.ugr.es.
(13) Warren, R. C.; McConnell, J. F.; Stepherson, N. C. Acta Crystallogr.
1970, 26B, 1402.
(14) Biogli, F.; Leporati, E.; Pellinghelli, M. A. Acta Crystallogr. 1979,
35B, 1465.
(15) Al-Salem, N. A.; El-Ezaby, M. S.; Marafie; H. M.; Abu-Soud, H. M.
Polyhedron 1986, 5, 633.
(16) Marafie, H. M.; Shuaib, N.; El-Ezaby, M. S. Polyhedron 1987, 6, 1391.
(17) Marafie, H. M.; Shuaib, N. M.; Ghodsiam, R.; El-Ezaby, M. S. J.
Inorg. Biochem. 1990, 38, 27.
X Abstract published in AdVance ACS Abstracts, March 15, 1997.
(1) (a) Universidad de Granada. (b) Instituto de Parasitologia y Biome-
dicina “Lo´pez-Neyra (CSIC). (c) Universidad de Auto´noma de Madrid.
(2) Brown, S. J.; Chow, Ch. S.; Lippard, S. J. In Encyclopedia of Inorganic
Chemistry; King, R. B., Ed.; Wiley: New York-Chichester, 1994;
pp 3305-3315, and references cited therein.
(3) (a) Lippard, S. J. Pure Appl. Chem. 1987, 59, 731. (b) Reedijk, J.;
Fichtinger-Schepman, A. M. J.; van Oosteron, A. T.; van de Putte, P.
Struc. Bonding 1987, 67, 53.
(18) Taki Khan, B.; Bhatt, J.; Najmuddin, K.; Shamsuddin, S. J. Inorg.
Biochem. 1991, 44, 55.
(4) Rau, T.; van Eldik, R. In Metal Ions In Biological Systems; Sigel, H.,
Sigel, A., Eds.; Marcel Dekker: New York-Basel, 1996; Vol. 31, pp
339-378, and references cited therein.
(5) Lim, M. C.; Mart´ın, R. B. J. Inorg. Nucl. Chem. 1976, 38, 1911.
(6) Graham, R. D.; Williams, D. R. J. Inorg. Nucl. Chem. 1979, 41, 1245.
(19) Mansuri-Torshizi, H.; Srivastava, T. S.; Parekh, H. K.; Chitnis, M. P.
J. Inorg. Biochem. 1992, 45, 135.
(20) Mart´ın, R. B. In Platinum, Gold and other Metal Chemotherapeutic
Agents; Lippard, S. J., Ed.; ACS Symposium Series No. 209; American
Chemical Socirty: Washington, D.C., 1983; Chapter 11.
S0020-1669(96)01226-8 CCC: $14.00 © 1997 American Chemical Society

Pd(II) Versus Pt(II) Model Anticancer Drugs
Inorganic Chemistry, Vol. 36, No. 9, 1997 1807
Inagaki et al.21 have prepared cis-dichloroplatinum(II) com-
plexes with R,ω-diaminocarboxylic acids (namely, 2,3-diami-
nopropionic (Hdap), 2,4-diaminobutyric (Hdab), and 2,3-
diaminosuccinic (H2dasa) acids) as well as their ethyl ester
derivatives (Etdap, Etdab, Et2dasa) as N,N-chelating inert
groups. The reactivity of these latter esterified complexes with
calf thymus DNA (CT-DNA) and the antitumor properties
against leukemia L1210 were explained on the basis of their
noncharged nature and their expected ability to permeate into
cells. In contrast, the negatively charged cis-dichloro(diami-
nocarboxylato)platinum(II) complexes exhibit lower reactivity
with CT-DNA and the absence of antitumor activity.
biochemical and biological properties on the basis of the nature
of their species in solution.
Experimental Section
Synthesis of 2,3-Diaminopropionate Ethyl Ester Dihydrochloride
(Etdap‚2HCl) and 2,4-Diaminobutyrate Ethyl Ester Dihydrochlo-
ride (Etdab‚2HCl). A mixture of d,l-2,3-diaminopropionic acid
hydrochloride (5 g; Sigma) or of d,l-2,4-diaminobutyric acid dihydro-
chloride (5 g; Sigma) in absolute ethanol (500 mL) was stirred and
refluxed for 24 h under a flow of dried HCl until a clear solution was
obtained. The hot solution was filtered. After cooling with ice and
NaCl (-15 °C), the desired diamino ester dihydrochloride precipitates
as a white polycrystalline product, which was filtered, washed with
absolute ethanol and acetone, and air-dried: yield, >95%. Anal. Calcd
for Etdap‚2HCl (C5H14Cl2N2O2): C, 29.29; H, 6.88; N, 14.00.
Found: C, 29.37; H, 7.25; N, 13.67. Mp(dec) ) 160 °C. Calcd for
Etdab‚2HCl (C6H16Cl2N2O2): C, 32.82; H, 7.34; N, 12.76. Found: C,
33.06; H, 8.02; N, 12.81. Mp (dec) ) 182 °C.
Synthesis of cis-[Pd(Etdap)Cl2] (I) and cis-[Pd(Etdab)2Cl] (II).
Yellow-orange polycrystalline samples of compounds I and II were
obtained by equimolar reactions of K2PdCl4 in water with Etdap‚2HCl
and Etdab‚2HCl, respectively, followed by neutralization with NaHCO3.
A similar procedure was reported for cis-dichloro(diaminosuccinate
diethyl ester)palladium(II) in ref 21: yield. >80%; Calcd for I,
C5H12Cl2N2O2Pd: C, 19.41; H, 3.91; N, 9.05; Cl, 22.91. Found: C,
19.51; H, 3.91; N, 9.18; Cl, 23.90. Mp (dec) 230 °C. Calcd for II,
C6H14Cl2N2O2Pd: C, 22.28; H, 4.36; N, 8.66; Cl, 21.92. Found: C,
22.30; H, 4.32; N, 8.25; Cl, 20.39. Mp (dec) ) 245 ° C. Both
complexes can also be prepared by reacting a solution of PdCl2 (4%)
in hot 0.1 M HCl with an equimolar amount of each diaminoacidate
ethyl ester dihydrochloride. The pH of the resulting solutions is
adjusted to 6-7 with aqueous NaHCO3. The desired products
precipitate during 24 h at room temperature as yellow powders which
can be recrystallized from dilute NaCl aqueous solutions.
We have recently reported the synthesis of cis-[Pd(H2dasa)-
Cl2] and cis-[Pd(Et2dasa)Cl2] as well as the molecular structure
of the latter and that of [Pd(2,2′-bipy)(dasa)]‚3H2O (bipy ) 2,2′-
bipyridine).22 Both cis-dichloropalladium(II) complexes have
been shown to induce conformational changes in the covalent
closed circular form of pUC8 plasmid. Both compounds
showed significant cytotoxicity against MDA-MB468 and HL-
60 human cancer cell lines.22 In these and closely related cis-
dichloro(chelating diammine)palladium(II) compounds it can be
assumed that the chelating ligand remains effectively bounded
to Pd(II) in a wide pH range,4,7,10,12,23,24 so that is called a
“spectator” ligand.4 However, at physiological pH 7.4 and ionic
strength I ) 0.15 M such cis-dichloro complexes would give a
variety of aqua and/or hydroxo and/or chloro species (including
certain polynuclear ones), depending on the low (inside of cells)
or rich (plasma or blood) chloride ion concentration and on the
drug concentration itself. We think that equilibrium studies of
such labile systems would be instructive for a better understand-
ing of the modeling role of Pd(II) complexes for the inert Pt(II)
ones. In this sense, Hohmann and van Eldik9 have reported
interesting potentiometric and spectrophotometric results for
water by chloride substitution reactions on cis-diaqua(ethylene-
diamine)palladium(II), which include the use of a selective
electrode to measure the free-chloride ion concentration at
equilibrium. In our laboratories we investigated selected cis-
dichloropalladium(II) complexes as potential anticancer drugs25
as well as model systems for their platinum(II) analogues. For
these purposes we have attempted to develop a more general
experimental methodology in order to fit the complexation
model and accurate formation constants for the aqua and/or
hydroxo and/or chloro species in solutions of cis-dichloropal-
ladium(II) drugs. This general procedure supposes the use of
the SUPERQUAD program26 to fit, as a first step, formation
constants log âpqr of the hydroxo species from E(H+) poten-
tiometric data of alkalimetric titrations of cis-diaqua complex
solutions. In a second step the above formation constants log
âpqr (as fixed values) and the referred program are used with
series of E(H+) and E(Cl-) data pairs, simultaneously obtained
from several titrations of the corresponding diaqua complex
solutions, to fit log âpqr for the chloro-containing species. In
this paper we report the synthesis and some DNA binding
properties of cis-dichloro(d,l-2,3-diaminopropionate ethyl ester)-
palladium(II) (I) and cis-dichloro(d,l-2,4-diaminobutyrate ethyl
ester)palladium(II) (II). We also apply our potentiometric
procedure to compounds I and II in order to explain their
Physicochemical Characterization of the Products. Solid samples
of diaminoacidate ethyl ester dihydrochlorides, and the corresponding
cis-dichloropalladium(II) complexes (I and II) were characterized by
infrared spectroscopy and TG-DTA thermal analysis as described in
ref 22.
Potentiometric Titrations, Complexation Model, and Complex
Formation Constant Refinement. All required solutions were
prepared with CO2-free doubly distilled water. The ionic strength of
solutions of the complexes (see below) as well as titrant reagents (0.1
M NaOH or NaCl) were adjusted to I ) 0.15 M (NaClO4). The
temperature of the sample in the double wall reaction cell was
maintained at 37.00 ( 0.05 °C by circulating temperature-controlled
water. A stream of N2 presaturated with 0.15 M NaClO4 flowed over
the tested solutions. Aqueous solutions of cis-diaquapalladium(II)
complexes with Etdap or Etdab were prepared as perchlorate salts by
the reaction of carefully measured amounts of I or II, respectively,
dissolved in the smallest possible volume of water with 2 equiv of a
fresh and standardized AgClO4 aqueous solution (working with topaz
stained glass material in a darkroom and removing AgCl by filtration).
Clear solutions of cis-[Pd(Etdap)(H2O)2](ClO4)2 (III) and cis-[Pd-
(Etdab)(H2O)2](ClO4)2 (IV) were diluted with appropriate amounts of
a stock solution of NaClO4 and water to obtain virtually chloride-free
solutions (estimated [Cl-] , 10-5 M) of III or IV salts with I ) 0.15
M (NaClO4). These solutions alone, or diluted with 0.15 M background
electrolyte, form the initial solutions for the two steps of the poten-
tiometric study. As a first step, aliquots (50 mL) of three initial
solutions of III (3.079 × 10-3, 2.207 × 10-3, and 1.333 × 10-3 M)
and of IV (3.277 × 10-3, 2.344 × 10-3, and 1.390 × 10-3 M) were
titrated with standardized 0.1 M NaOH in a Metrohm Dosimat 665
Titroprocessor equipped with a digital (pH/mV)-meter Crison 2002 with
a glass electrode (Ingold 10-401-3664) and a (Ag/AgCl) reference
electrode (Ingold 373-90-WTE-ISE-S7) with an intermediate electrolyte
(21) Inagaki, K.; Kidani, Y.; Suzuki, K.; Tashiro, T. Chem. Pharm. Bull.
1980, 28, 2286.
(22) Matilla, A.; Tercero, J. M.; Dung, N.-H.; Viossat, B.; Pe´rez, J. M.;
Alonso, C.; Mart´ın-Ramos, J. D.; Niclo´s-Gutierrez, J. J. Inorg.
Biochem. 1994, 55, 235.
(23) Wilson, E.; Mart´ın, R. B. Inorg. Chem. 1970, 9, 528.
(24) Pitner, T. P.; Wilson, E. W.; Mart´ın, R. B. Inorg. Chem. 1972, 11,
738.
(25) Gonza´lez Moles, M. L. Ph.D. Dissertation, University of Granada,
Granada, Spain, 1990.
(26) Gans, P.; Sabatini, A.; Vacca, A. J. Chem. Soc., Dalton Trans. 1985,
1195.

1808 Inorganic Chemistry, Vol. 36, No. 9, 1997
Gonza´lez et al.
chamber. The intermediate electrolyte (0.15 M NaClO4) was replaced
daily. The standard electrode potential E°, the concentration of the
NaOH titrant, and Kw ) (4.57 ( 1) × 10-14 were checked before and
after each experiment by the titration of a known amount (50 mL) of
∼10-2 M HClO4 in 0.15 M NaClO4. This calibration procedure of the
potentiometric system is similar to that described by Leporati27 and
was proved successful in a system involving Cu(II)-chelated hydroxo-
complex formation.28 The equilibrium model and the corresponding
hydroxo-complex formation constants (log âpqr) can be defined by the
general equilibrium 1 (charges omitted for simplicity)
pPdL(H2O)2 + qCl ( rH a (PdL)pClq(H2O)2H(r
(1)
where L represents Etdap or Etdab, q ) 0 for chloride-free species,
pqr ) 1,0,0 corresponds to cis-[PdL(H2O)2] ) PdL(H2O)2 and, for
example, pqr ) 1,0,-2 indicates PdL(OH)2 ) cis-[PdL(OH)2]. Obvi-
ously equilibrium 1 can also be used for reactions involving chloro (q
) 1) and dichloro species (q ) 2). The formation constants of the
hydroxo complexes reported herein were obtained from 158 or 139
E(H+) data of three solutions of diaqua-complex salts (III or IV) treated
with the method of rigorous least-squares by means of the SUPER-
QUAD program.26 In these calculations we use σV ) 0.002 mL and
σE ) 0.1 mV. These potentiometric data correspond to the range 3.5
e pH e 7. This upper pH limit was decided upon because in the
alkaline range it becomes evident that there is an overlap of the ester
hydrolysis reaction with the formation of dihydroxo species type PdL-
(OH)2, which prevents the calculation of log â1,0,-2. Fortunately, these
uncharged species PdL(OH)2 have no significant abundance at physi-
ological pH values. The second step was to use the formation constants
of aquahydroxo and di(µ-hydroxo) species as fixed values to fit the
chloro-complex formation constants (log âpqr, q ) 1 or 2). In this step,
we have simultaneously used two potentiometric systems which measure
E(H+) (as described above) and E(Cl-) potential data pairs. For this
purpose the titroprocessor was equipped with a second digital (pH/
mV) meter with a (Ag2S/AgCl) chloride ion selective electrode (Ingold
15-213-3000) and the above-mentioned reference electrode with an
intermediate electrolyte chamber (salt bridge 0.15 M NaClO4, replaced
daily). When necessary, we replaced the alkali titrant container with
another one to add aliquots of 0.1 M NaCl in 0.15 M NaClO4 (second
titrant reagent). Using these two potentiometric systems simultaneously,
we carried out a variety of experimental procedures, changing the order
of titrant (NaOH or NaCl) additions. The best results correspond to
the addition of NaOH to reach the first buffer pH range of the titration
(pH ∼ 4.5) before making successive small additions of the NaCl
solution. Each of these additions can produce changes in both E(Cl-)
and E(H+) potentials if the H2O or OH- ligand of the aqua and/or
hydroxo species undergo substitution reactions by Cl- ions to give
chloro-containing species. E(H+) and E(Cl-) data pairs were measured
after each of these additions. From 1.745 × 10-3 and 3.490 × 10-3
M solutions of compound III and from 1.750 × 10-3 and 3.500 ×
10-3 M solutions of compound IV we treated 127 and 70 E(H+) and
E(Cl-) pairs of data, respectively, by the SUPERQUAD program to
fit the formation constants of the cis-chloroaqua, cis-chlorohydroxo,
and cis-dichloro species (pqr ) 1,1,0; 1,1,-1; and 1,2,0, respectively)
formed from cis-[PdL(H2O)2] (L ) Etdap or Etdab). These calculations
were performed with σV ) 0.002 mL, σE ) 0.1 mV for H+, and 0.2
mV for Cl-. The entire set of fitted log âpqr values for Etdap or Etdab
Pd(II) complexes were used to simulate the alkalimetric titrations of I
or III and of II or IV, respectively, as well as to obtain a variety of
diagrams which illustrate the species abundance in the studied systems
as a function of pH, pCl ) -log[Cl-] and total molar complex
concentration.
Figure 1. Experimental potentiometric data points for 3.079 mM cis-
[Pd(Etdap)(H2O)2]2+ (lower), 3.489 mM cis-[Pd(Etdap)Cl2] (upper),
3.277 mM cis-[Pd(Etdab)(H2O)2]2+ (lower), and 3.506 mM cis-[Pd-
(Etdab)Cl2] (upper) solutions and the corresponding simulated titrations
(unbroken lines) with NaOH (0.0992, 0.0994, 0.0988, and 0.0983 M,
respectively). V0 ) 50 mL.
latter having 80% of GC pairs). Drug:DNA complexes were formed
by incubation at 37 °C of compounds cis-DDP, I, or II and DNA in a
buffered solution (pH ) 7.2) 0.1 × SSC (SSC ) 0.15 M NaCl and
0.015 M sodium citrate). The periods of complex formation were 15
min and 48 h (so that the equilibria with cis-DDP were also reached).
The UV spectra were recorded on a Beckman Acta III spectrophotom-
eter attached to a temperature programmer. Melting profiles were
recorded at 258 nm using a rate of temperature increase of 1 °C/min
from 25 to 90 °C. In the renaturation assays a phosphate buffer (pH
) 7.2) was used. Salmon sperm DNA was denaturated by heating it
for 20 min at 100 °C and then allowing it to renature at 10 °C below
the melting temperature (Tm) of native DNA. The activity in ViVo of
the drugs was studied by analysis of the transcription process using as
a model system the polytene chromosomes of Drosophila hydei. For
this purpose larvae from the third instar were microinjected with 0.5
µL of a 10-6 M solution of compound II. Four hours after the
microinjection the chromosomes were isolated in physiological Ringer
solution following standard methods, pre-fixed with ethanol:acetic acid
(3:1), fixed with 50% acetic acid, flattened between cover glass, and
frozen in liquid nitrogen. After the cover glass was removed, the
chromosomes on the slide were introduced into absolute ethanol for
10 min before being stained with lactoacetic orcein. Transcription was
followed by the detection of chromosomal nascent RNA visualized by
the formation of DNA:RNA hybrids in situ according to Alcover et
al.29 The hybrids were detected by an anti-DNA:RNA antibody
followed by a fluorescent antibody. The chromosomes were photo-
graphed in a Zeiss fluorescence microscope using Tri-X Pan Kodak
film. The amount of fluorescence was determined by densitometry of
10 chromosomes in control as well as in treated chromosomes as a
measurement of transcription.
DNA Binding Studies. A comparative study of the interaction of
compounds cis-DDP, cis-[Pd(Etdap)Cl2] (I), and cis-[Pd(Etdab)Cl2] (II)
with native DNA samples was carried out in Vitro by differential
spectrophotometry methods. In addition to measuring the UV differ-
ence spectral data of drug:DNA complexes, melting profiles and
renaturation curves of DNA samples treated with the metal complexes
at several molar ratios r (complex/nucleotide of DNA) were recorded.
We used DNA from salmon sperm or Streptomyces sp. sources (this
Results and Discussion
Chemical Characterization. The IR spectra (cm-1) of
Etdap‚2HCl and Etdab‚2HCl show bands of the modes ν(CdO)
(1742 and 1735, respectively) of -COOEt and δd (∼1600) and
+
δs (1475-1500 cm-1) of -NH3 groups. The IR spectra of
complexes I and II show a double νas band (3240 and 3170, I;
3270 and 3200, II) and a νs band (3090, I; 3120, II) of -NH2
(27) Leporati, E. J. Chem. Soc., Dalton Trans. 1985, 1605; 1988, 421.
(28) Tercero, J. M.; Matilla, A.; Niclo´s, J.; Castin˜eiras, A. J. Coord. Chem.
1995, 34, 139.
(29) Alcover, A.; Izquierdo, M.; Stollar, B. D.; Kitagawa, Y.; Miranda,
M.; Alonso, C. Chromosoma 1982, 87, 262.

Pd(II) Versus Pt(II) Model Anticancer Drugs
Inorganic Chemistry, Vol. 36, No. 9, 1997 1809
Table 1. Formation Constants (log âpqr)a of the Hydroxo and/or Chloro Species in Aqueous Solutions of cis-Dichloro(diaminoacidate ethyl
ester)palladium(II), cis-[Pd(Etdaa)Cl2], for I ) 0.15 M (NaClO4) and t ) 37 °C
Etdaa ) Etdap
Etdaa ) Etdab
Etdaa ) Et2dasad
log âd
complexb (pqr)
log â
stat. Dc
log â
stat. Dc
aquahydroxo (10-1)
di(µ-hydroxo) (20-2)
-5.99(2)
-7.14(4)
Z ) 158
σ ) 1.09
ø2 ) 8.68
Z ) 127
σ ) 1.88
ø2 ) 9.75
-5.06(1)
-7.35(2)
Z ) 139
σ ) 3.50
ø2 ) 9.89
Z ) 70
-5.25
-6.55
dichloro (120)
chloroaqua (110)
chlorohydroxo (11-1)
5.78(1)
3.38(1)
-2.97(4)
5.98(2)
3.32(1)
-2.37(2)
5.86
3.65
-2.68
σ ) 2.32
ø2 ) 10.40
a Values in parentheses are standard deviations. b Univalent ligands bounded to one (p ) 1) or two (p ) 2) Pd(Etdaa) moieties. c Statistical data
for accuracy: Z ) total number of experimental data points used in the refinement; ø2 ) observed statistical parameter based on weighted residuals
of emf readings (values < 12.6 represent a confidence level of >95%);27 σ ) ratio of the mean square root of the weighted residuals to the
estimated error in the actual working conditions (σV and σE). d log âpqr values of the hydrolytic species of cis-dichloro(diaminosuccinate diethyl
ester)palladium(II), cis-[Pd(Et2dasa)Cl2].31
groups, two weak bands of ν(Pd-N) (560 and 520, I; 550 and
510, II) and a typical one of ν(Pd-Cl) near 300 cm-1. These
bands are absent in the ligand dihydrochloride spectra. Other
expected absorptions such as the ν(CdO) stretching (1735) and
δ(NH2) bands (1555, I; 1565, II) were also observed. These
assignments are consistent with those reported for Etdap-Pt(II)
and Etdab-Pt(II) analogues in ref 21. The TG-DTA diagrams
reveal that compounds I and II are stable in an air atmosphere
below 230 and 245 °C, respectively. At higher temperatures
the pyrolysis of the organic ligand probably produces a crude
residue of PdCl2 (near 600 °C).
) 1,2,0; 1,1,0; and 1,1,-1) will be formed in the presence of
Cl- (Table 1). Other di- or polynuclear species can be formed
in more concentrated complex solutions.
The fitness of both sets of formation constants reported herein
are shown in the remarkable agreement of the simulated and
experimental titrations (Figure 1). These constants can be
compared with those reported for related diaminopalladium(II)
complexes.8,9,20,31,32 Our log âpqr data are very similar to those31
of the closely related cis-[Pd(Et2dasa)Cl2] and cis-[Pd(Et2dasa)-
(H2O)2]2+ (Table 1). However, the dimerizing ability of the
aquahydroxo complexes by the general equilibrium
Hydrolytic Species of cis-[Pd(Etdaa)Cl2] (I and II) and
cis-[Pd(Etdaa)(H2O)2](ClO4)2 (III and IV). Alkalimetric
titrations of I and II and the corresponding diaqua complexes
III and IV (perchlorate salts) show a defined inflection to a )
1 (a ) equiv of base/mol Pd(II), Figure 1) followed by another
one to a ) 2 (extremely smooth, not shown). These behaviors
are consistent with the titration of two groups with different
acidities, in separate steps, to give hydroxo species with OH/
Pd(II) ratios 1:1 and 2:1, respectively. In Cl--free solutions of
III or IV (diaqua complexes, pqr ) 1,0,0) the assumption that
only mononuclear hydroxo complexes (pqr ) 1,0,-1) are
formed for a e 1 does not seem reasonable because of the
remarkable difference in their two acidities. The formation of
dinuclear di-(µ-hydroxo) complexes have been suggested in the
hydrolysis of related complexes.5,8,9,30-32 However, the forma-
tion of only such dinuclear di-(µ-hydroxo) complexes [(Etdaa)-
Pd(OH)2 Pd(Etdaa)]2+ (pqr ) 2,0,-2) also seems unreasonable
because the “apparent” pK1 value (the pH value after the addition
of a half of alkali equivalent) does not remain as a “constant”
but decreases as the complex concentration increases. Thus,
the first hydrolysis of III and IV will involve at least the
overlapped formation of mononuclear hydroxo and di-(µ-
hydroxo) complexes (and perhaps other ones).30-32 These
features preclude all attempts to estimate a constant pKa for
Pd(II)-H2O groups. In the second step (1 e a e 2, pH > 7)
[Pd(Etdaa)(OH)2] will be formed, but a slow and persistent
lowering of pH (by ester hydrolysis) prevents the calculation
2[Pd(Etdaa)(H2O)(OH)]+ a
[(Etdaa)Pd(µ-OH)2Pd(Etdaa)]2+ + 2H2O
are different depending on the spectator chelating ligand.4
Indeed the dimerization constants log KD ) log â2,0,-2 - 2 log
â1,0,-1 are in the order of log KD ) 4.84, 3.95, and 2.77 for daa
) Etdap, Et2dasa, and Etdab, respectively. Thus, such dimer-
ization ability seems mainly related to the chelate ring size
formed by the spectator ligand (Pd(Etdab) moiety with a larger
six-membered ring should better hinder the dimerization than
the Pd(Etdap) and Pd(Et2dasa) ones with shorter five-membered
rings).
The hydrolysis of III and IV depends on pH, [Cl-], and the
diaqua complex concentration. A variety of illustrative diagrams
are given in Figures 2-4. In acid to neutral solutions, the
hydrolysis of cis-[Pd(Etdaa)(H2O)2]2+ (100, Etdaa ) Etdap or
Etdab) produces cis-aquahydroxo and di(µ-hydroxo) species
(10-1 and 20-2). The mole fraction of the dimeric species
depends on the diaqua complex concentration because dilution
favors dissociation (Figure 2). In addition, as expected for log
KD values, the hydrolysis of cis-[Pd(Etdap)2(H2O)]2+ produces
a higher proportion of di-(µ-hydroxo) complex than that of cis-
[Pd(Etdab)(H2O)2]2+
.
As an illustrative case, Figure 3 shows the distribution
diagram of species for 1 mM cis-[Pd(Etdab)(H2O)2](ClO4)2 at
the physiological pH 7.4 as a function of pCl ) -log [Cl-]. In
such conditions, at [Cl-] ) 1 M (pCl ) 0) the abundance of
the neutral dichloro complex becomes 90%, while the remaining
10% corresponds to the also neutral chlorohydroxo one. A
decrease in [Cl-] produces an increasing hydrolysis which gives
the chlorohydroxo complex (1,1,-1) as the predominant species
as well as positively charged aquahydroxo and di(µ-hydroxo)
ones (1,0,-1 and 2,0,-2). Note that in a wide range of [Cl-]
the chloroaqua complex cation (1,1,0) does not have significance
at physiological conditions. At [Cl-] ) 104 mM (plasma) all
of the species formed are neutral chloro ones (1,2,0 and 1,1,-
1). In clear contrast, at [Cl-] ) 4 mM (inside of cells) the
neutral chlorohydroxo species and two positively charged ones
(1,0,-1 and 2,0,-2) are formed.
of log â1,0,-2
.
Previous suggestions were taken into account in the fitting
of the hydrolytic model and formation constants (log âpqr) by
the SUPERQUAD program (Table 1). In the experimental
conditions of the present work, only mononuclear (pqr ) 1,0,-
1) and dinuclear species (pqr ) 2,0,-2) seem involved in the
first hydrolysis of III and IV. These two hydroxo complexes
as well as dichloro, chloroaqua, and chlorohydroxo ones (pqr
(30) Anderegg, G. Inorg. Chim. Acta 1986, 113, 101.
(31) Gonza´lez, S.; Tercero, J. M.; Matilla, A.; Pe´rez, J. M.; Gonza´lez, V.
M.; Alonso, C.; Niclo´s-Gutie´rrez, J. J. Inorg. Biochem. 1996, 61, 26.
(32) Tercero-Moreno, J. M.; Matilla-Herna´ndez, A.; Gonza´lez-Garc´ıa, S.;
Niclo´s-Gutie´rrez, J. Inorg. Chim. Acta 1996, 235, 23.

1810 Inorganic Chemistry, Vol. 36, No. 9, 1997
Gonza´lez et al.
Figure 2. Distribution diagrams for hydrolytic species (indicated as pqr codes) of the diaqua complexes as a function of pH and complex concentration.
Solutions of cis-[Pd(Etdap)(H2O)2]2+ (III) at 1 mM (a) or 1 µM (c) and solutions of cis-[Pd(Etdab)(H2O)2]2+ (IV) at 1 mM (b) or 1 µM (d).
DNA Binding Studies. The data shown in Table 2 indicate
the spectral behavior of the DNA bound to the complexes tested
in this study. The data (λmax, H, and ∆A270/∆A295) obtained
for the DNA-cis-DDP complex (used as a reference system)
are in agreement with other data reported in the literature.21,33,34
The UV spectral pattern of the DNA samples treated with
compounds I and II was, in general, similar to that obtained
for the DNA-cis-DDP complex. Bathochromic shifts ac-
companied by hyperchromic effects are generally observed.
However, our data reveal that the interaction between cis-DDP
and DNA increases with the incubation time, while the
interactions of the Pd(II) complexes (I and II) with DNA are
time independent (the data obtained at 15 min or 48 h of
incubation for Pd(II) complexes are the same). These features
are related to the kinetic differences existing between Pt(II) and
Pd(II) complexes. Thus, the latter can react up to 105-fold faster
Figure 3. Distribution of species (indicated by pqr codes) in solutions
of 1 mM cis-[Pd(Etdab)(H2O)2](ClO4)2 at pH ) 7.4 as a function of
chloride ion concentration (pCl ) -log[Cl-]). Dotted lines correspond
to plasma (104 mM) and into cell 4 mM Cl- ion concentrations.
than the Pt(II) ones. Furthermore, it was interesting to note
that the effects induced by binding of compounds I and II are
stronger than those induced by cis-DDP binding. Indeed,
concentrations of compounds I or II (i.e., r ) 0.0025) at even
100-fold lower than cis-DDP (i.e., r ) 0.25) give similar or
higher values of λmax, H, and ∆A270/∆A295. Particularly interest-
ing is the ∆A270/∆A295 ) 1.8 value presented by compound II
(at r ) 0.0025), quite similar to the ∆A270/∆A295 ) 2 value
presented by the antitumoral agent cis-DDP at r ) 0.25. These
features suggest that the secondary structure of DNA may be
disrupted in the presence of compounds I and II to an extent
similar to or greater than that produced by cis-DDP.
In view of the reactivity with DNA, it is interesting to estimate
the total percentage of neutral and charged species in physi-
ological conditions (Figure 4). Noteworthy, both diaqua
complexes III and IV will produce >96% of neutral species in
plasma (1 mM to 1 µM complex concentrations). In contrast,
they produce abundant cationic species (∼24-72% for [Pd-
(Etdap)(H2O)2]2+ and ∼34-45% for [Pd(Etdab)(H2O)2+]2) at
the same conditions. Such a pathway can be significant for
analogous inert platinum(II) compounds (where µ-hydoxo
oligomers seem resposiblie of toxicity) but not for kinetically
labile palladium(II) ones which can quickly displace a variety
of complex formation equilibria. Our results are very suitable
for the diffusion of compounds I and II across biological
membranes as well as for interaction with DNA into cells.
The data obtained from the melting curves of DNA samples
treated with compound I, II, or cis-DDP (reference) are
(33) Inagaki, K.; Kidani, Y. Inorg. Chim. Acta 1980, 46, 35.
(34) Lo´pez, M. C.; Ruiz, L. M.; Cracianescu, D.; Doadrio, A.; Osuna, A.;
Alonso, C. Chem. Biol. Interact. 1986, 18, 99.

Pd(II) Versus Pt(II) Model Anticancer Drugs
Inorganic Chemistry, Vol. 36, No. 9, 1997 1811
Figure 4. Total percentage of neutral and charged species in diaqua complex solutions at pH ) 7.4 and chloride concentrations as in plasma or
within cells: 1 mM (a) and 1 µM (b) cis-[Pd(Etdap)(H2O)2]2+ 1mM (a) and 1µM (b) and 1 mM (c) and 1 µM (d) cis-[Pd(Etdab)(H2O)2]2+
.
Table 2. Ultraviolet Spectral Dataa of Native DNA and DNA
Samples after the Addition of Pd(II) Complexes I and II and
cis-DDPb
Table 3. Melting Temperature (Tm, °C), Hyperchromicity (H, %),
and Width of Transition 2σ (°C) of Native DNA and DNA
Complex Systems
∆A270
/
Tm
(°C)a (%) (°C)
Hb
2σc
a
systems
native DNA
DNA-compound II 0.0005
r
λmaxa (nm) Ha (%) ∆A295
systems
native DNA
DNA-compound II
r
258
259
260.5
261
261.3
262
68.9 33.5
68.9 28.4
65.7 24.4
47.8 11.0
40.6 10.1
48.0 16.3
8.8
9.1
9.9
9.2
5.4
6.0
9.9
1.4
4.0
11.5
13.4
22.1
7.5
1.06
1.56
1.6
1.65
1.8
1.4
1.3
0.0005
0.0010
0.0015
0.0020
0.0020
0.0010
0.0015
0.0020
0.0025
0.0025
0.25(15 min)
0.25(24 h)
DNA-compound I
DNA-cis-DDP
DNA-compound I
DNA-cis-DDP
258
0.2500 (t ) 15 min) 65.7 35.4
260
263
(-0.8)
4.5
0.2500 (t ) 48 h)
64.9 28.2 10.1
84.7 27 5.1
53.4 11.9 16.0
2.0
DNA (Streptomyces sp.)
DNA-compound II
0.0020
a λmax ) wavelength at which maximal ultraviolet absorption is
produced; H (%) ) hyperchromic (or hypochromic, if negative) effect,
as percentage of the native DNA; ∆A270/∆A295 ) increase in the
absorption ratio at 270 versus 295 nm. b Compounds I and II (Pd(II)
complexes) show time independent effects on DNA, whereas those of
cis-DDP are time dependent.
a Tm ) temperature value at half denaturation. b H ) hyperchromicity
by heating. c 2σ ) temperature range corresponding to 15 and 85%
denaturation.
the effects of cis-DDP that increase with the incubation time.
The data for Tm and H obtained for compounds I and II even
at r ) 0.001 are quite similar to those of cis-DDP (r ) 0.25,
incubation time 48 h). Thus, lower concentrations of Pd(II)
drugs can produce DNA destabilization similar to that of cis-
DDP. Both compounds at r ) 0.0025 produce an almost
complete DNA denaturation. With GC 80% rich DNA (Strep-
tomyces sp) the behavior seems different. The width transition
values (2σ) of this native DNA increase from 5.1 to 16.0 (r )
0.002) for its DNA-compound II complex, while this parameter
decreases from 8.8 for native DNA (salmon sperm) to 5.4 for
its DNA-compound II complex (r ) 0.002). These results
summarized in Table 3. The melting temperature (Tm) and
hyperchromicity (H) values of DNA-compound II, DNA-
compound I, and DNA-cis-DDP systems are lower than the
Tm ) 68.9 °C and H ) 33.5% values of the native DNA and
decrease with the increase of the drug concentrations (r). This
behavior reveals the metal-complex interaction with the DNA
bases leading to the destabilization of its double helix with
respect to heat denaturation. The destabilization is greater as
the concentration of metal-complexes increases. The effects
produced by the Pd(II) complexes on the melting profiles of
DNA are independent of the incubation time, in contrast with

1812 Inorganic Chemistry, Vol. 36, No. 9, 1997
Gonza´lez et al.
Figure 5. (Upper) DNA renaturation (%) of DNA treated with different
concentrations (r) of cis-[Pd(Etdab)Cl2] (II): (a) Addition of complex
to heat denatured DNA; (b) addition of complex before DNA
denaturation process. (Lower, c) Hyperchromicity (%) of DNA treated
with different concentrations of cis-[Pd(Etdab)Cl2] (II) (T ) 25 °C).
Figure 7. Abundance (%) of hydrolytic species and total of neutral
and charged species for cis-[Pd(Etdap)Cl2] and cis-[Pd(Etdab)Cl2] in
experimental conditions of in Vitro DNA binding studies (pH ) 7.2,
[Cl-] ) 0.15 M, [complex] ) 1 µM). Species indicated by pqr codes.
in ViVo incorporaction of 3H-uridine. It was observed that these
regions did not incorporate the precursor, while as expected the
puffed regions in the control chromosomes did (data not shown).
The immunograms of the nascent RNA transcripts as detected
by DNA:RNA fluorescence indicated that the control chromo-
somes accumulated 4-fold more RNA than the treated ones after
4 h of incubation with compound II as an indication that the
drug partially inhibited RNA transcription.
It is instructive to correlate the biochemical results with the
hydrolytic behavior of Pd(II)-complexes I and II. Figure 7
shows the percentage of species formed in the hydrolysis of
the Pd complexes under the experimental conditions of the DNA
binding studies in Vitro (pH ) 7.2, [Cl-] ) 0.15 M and [Pd(II)
complex] e 10-6 M). On the basis of Inagaki’s suggestions,21
we think that the charged species of drugs I and II would be
the more active ones in the interaction mechanism of these drugs
with DNA. However, such cationic species arise from the
hydrolysis of both Pd(II) complexes (drugs I and II) in very
low proportions (2.28 and 1.41% for I and II, respectively).
This fact does not suppose a low interaction with DNA because
the hydrolytic equilibria of Pd(II) complexes are labile, and
consequently the equilibrium displacement could provide new
amounts of the charged species to interact with DNA, as the
biochemical experiments reveal. In contrast, the hydrolytic
equilibria of analogous Pt(II) complexes are not labile. The
kinetic differences of the Pt(II) complexes with respect to the
Pd(II) ones and not their thermodynamic resemblances are more
probably responsible for the hydrolysis of Pt(II) complexes
making a high proportion of charged species available to
produce an important interaction with DNA.21
Figure 6. Photographic images of polytene chromosomes extracted
from salivary glands of Drosophila hydei larvae and stained with
lactoacetic orcein: (a) chromosome of a larva microinjected with 0.5
µL of a solution of 1 µM cis-[Pd(Etdab)Cl2] (II); (b) chromosome
control (untreated).
indicate that the interaction of drug II with DNA depends upon
the GC content, with the GC-rich DNA being destabilized more
effectively.
The results of the DNA renaturation assays are given in Figure
5. The addition of compound II to a heat denatured DNA (line
a) does not modify the DNA renaturation percentage, this being
practically complete even at a complex concentration corre-
sponding to r ) 0.0025. However, the addition of the complex
before the DNA denaturation process produces an inhibition of
the DNA renaturation (line b). This effect increases with the
value of r. Since the inhibition percentage of the DNA
renaturation (line b) is similar to the hyperchromicity percentage
caused by the drug on native DNA (at the same drug concentra-
tion) (line c), it is likely that the interaction of drug II with
DNA stimulates its denaturation, while it partially blocks its
renaturation.
The effects of compound II upon chromatin structure and
transcription were revealed by analysis of the polytene chro-
mosomes of Drosophila hydei (Figure 6). It was observed that
the chromatin of the treated chromosomes appeared more
compact than that of the untreated ones. In particular the
actively transcribing regions (puffs) present in the untreated
chromosmes (Figure 6b) collapsed after in ViVo administration
of compound II (Figure 6a). That in the treated chromosomes
these regions were transcriptionally inactive was confirmed by
The available biochemical and/or biological data indicate that
compound II has a DNA binding capacity higher than that of
compound I, although the first produces a total percentage of
charged species lower than the last one when both compounds
are hydrolyzed (Figure 7). Perhaps, in addition to charged
species, different neutral species (dichloro or chlorohydroxo
ones) of the studied drugs may influence their reactivity with
DNA by different ways. For example, we can assume that the
chlorohydroxo species (1,1,-1) could form hydrogen bonds.
This feature will favor the activity of drug II with respect to
drug I, although the total neutral species for both drugs is
similar.
IC961226Z