A combination of access to preassociation sites and local accumulation tendency in the direct vicinity of G-N7 controls the rate of platination of single-stranded DNA

Article abstract of DOI:10.1039/b418966c

Adduct formation between cationic reagents and targets on DNA are facilitated by the ability of DNA to attract cations to its surface. The electrostatic interactions likely provide the basis for the documented preference exhibited by cisplatin and related compounds for nuclear DNA over other cellular constituents. As an extension of a previous communication, we here present an investigation illustrating how the rate of adduct formation with the naturally occuring base guanine (G-N7) can be modulated by i) bulk solvent conditions, ii) local nature and size of the surrounding DNA and, iii) increasing DNA concentration. A series of single-stranded DNA oligomers of the type d(T_nGT_m); n = 0, 2, 4, 6, 8, 10, 12, 14, 16 and m = 16 - n or n = m = 4, 6, 8, 12, 16, 24 were allowed to react with the active metabolite of a potential orally active platinum(IV) drug, cis-[PtCl(NH ₃)(c-C₆H₁₁NH₂)(OH₂)] ⁺ in the presence of three different bulk cations; Na⁺, Mg²⁺, and Mn²⁺. For all positions along the oligomers, a change from monovalent bulk cations to divalent ones results in a decrease in reactivity, with Mn²⁺ as the more potent inhibitor as exemplified by the rate constants determined for interaction with d(T₈GT ₈): 10³ × K_obs/s^-1 = 6.5 ± 0.1 (Na⁺), 1.8 ± 0.1 (Mg²⁺), 1.0 ± 0.1 (Mn²⁺) at pH 4.2 and 25°C. Further, the adduct formation rate was found to vary with the exact location of the binding site in the presence of both Na⁺ and Mg²⁺, giving rise to reactivity maxima at the middle position. Increasing the size of the DNA-fragments was found to increase the reactivity only up to a total length of ca. 20 bases. The influence from addition of further bases to the reacting DNA was found to be salt dependent. At [Na⁺] = 0.5 mM a retardation in reactivity was observed whereas [Na⁺] ≥ 4.5 mM give rise to length independent kinetics. Finally, for the first time we have here been able to evaluate the influence from an increasing concentration of non-reactive DNA bases on the adduct formation process. The latter data were successfully fitted to an inhibition model suggesting that non-productive association of the platinum complex with sites distant from G-N7 competes with productive ones in the vicinity of the G-N7 target. Taken together, the kinetics support a reaction mechanism in which access to suitable association sites in the direct vicinity of the target site controls the rate of platination. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b418966c

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
A combination of access to preassociation sites and local
accumulation tendency in the direct vicinity of G-N7 controls
the rate of platination of single-stranded DNA†
a
a,b
Ase Sykfont Snygg, Malgorzata Brindell, Grazyna Stochelb and Sofi K. C. Elmroth*a
a Inorganic Chemistry, Chemical Center, Lund University, P. O. Box 124, SE-221 00, Lund,
Sweden. E-mail: Sofi Elmroth@inorg.lu.se; Fax: 46 46 222 4439; Tel: 46 46 222 8106
b Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060, Krakow, Poland
˚
Received 17th December 2004, Accepted 15th February 2005
First published as an Advance Article on the web 1st March 2005
Adduct formation between cationic reagents and targets on DNA are facilitated by the ability of DNA to attract
cations to its surface. The electrostatic interactions likely provide the basis for the documented preference exhibited
by cisplatin and related compounds for nuclear DNA over other cellular constituents. As an extension of a previous
communication, we here present an investigation illustrating how the rate of adduct formation with the naturally
occuring base guanine (G-N7) can be modulated by i) bulk solvent conditions, ii) local nature and size of the
surrounding DNA and, iii) increasing DNA concentration. A series of single-stranded DNA oligomers of the type
d(TnGTm); n = 0, 2, 4, 6, 8, 10, 12, 14, 16 and m = 16 − n or n = m = 4, 6, 8, 12, 16, 24 were allowed to react with the
active metabolite of a potential orally active platinum(IV) drug, cis-[PtCl(NH3)(c-C6H11NH2)(OH2)]+ in the presence
of three different bulk cations; Na+, Mg2+, and Mn2+. For all positions along the oligomers, a change from
monovalent bulk cations to divalent ones results in a decrease in reactivity, with Mn2+ as the more potent inhibitor as
exemplified by the rate constants determined for interaction with d(T8GT8): 103 × kobs/s−1 = 6.5 0.1 (Na+), 1.8
0.1 (Mg2+), 1.0 0.1 (Mn2+) at pH 4.2 and 25 ◦C. Further, the adduct formation rate was found to vary with the exact
location of the binding site in the presence of both Na+ and Mg2+, giving rise to reactivity maxima at the middle
position. Increasing the size of the DNA-fragments was found to increase the reactivity only up to a total length of
ca. 20 bases. The influence from addition of further bases to the reacting DNA was found to be salt dependent. At
[Na+] = 0.5 mM a retardation in reactivity was observed whereas [Na+] ≥ 4.5 mM give rise to length independent
kinetics. Finally, for the first time we have here been able to evaluate the influence from an increasing concentration of
non-reactive DNA bases on the adduct formation process. The latter data were successfully fitted to an inhibition
model suggesting that non-productive association of the platinum complex with sites distant from G-N7 competes
with productive ones in the vicinity of the G-N7 target. Taken together, the kinetics support a reaction mechanism in
which access to suitable association sites in the direct vicinity of the target site controls the rate of platination.
intracellular constituents.8,9,13 As for cisplatin, nuclear DNA is
Introduction
believed to be the most important intracellular target.14,15 The
Among transition metal complexes, the platinum containing
detailed molecular and cellular response is somewhat altered
ones have proven to be particularly efficient for the treatment
however, as a result of the presence of the cyclohexyl moiety in
of various types of cancers. Cisplatin, the first discovered
the intact coordination sphere of the platinum centre.14,16–19
anticancer active compound in this family,1,2 is still used
The polyanionic nature of DNA creates a local environment
routinely in the clinic, with well documented efficacy against
around it in solution that strongly influences the conditions for
testicular, small-cell lung, bladder, ovarian, and head and neck
approaching charged species. We have previously shown that
cancers. Nuclear DNA is one of the more important intracellular
the formation of coordinate covalent bonds between cationic
targets for the drug, and the resulting consequences for the
metal complexes and DNA models are facilitated, whereas
repair machinery and mechanisms for induction of apoptosis
reactions involving anionic metal complexes are inhibited.20–23
are currently under intense investigation.3
One of the major clinical limitations for the use of cisplatin
The increased reactivity observed for cations can be explained
by introduction of a preassociation step preceding the rate-
is intimately related to the development of resistance towards
determining adduct formation reaction. Our previous studies
the drug.4 The search for replacement drugs has so far resulted
of DNA model systems have indicated that the extent of
in two structurally similar compounds approved worldwide,
preassociation influences the reaction kinetics already in small
carboplatin and oxaliplatin, and a handful of compounds being
single-stranded systems.21,24,25 For example, the rate of adduct
evaluated in clinical trials.3a,5 A promising candidate of the latter
type is the orally administered Pt(IV) compound cis,trans,cis-
formation with the charged phosphodiester backbone was found
to be approximately linearly dependent on the size of the
[PtCl2(OAc)2(c-NH2C6H11)] (Satraplatin, JM 216), currently
studied oligomers in the range 2–16 bases.20 The variation in
undergoing Phase III clinical trials.6,7 Its mode of action
reactivity was found to correlate well with the expected cation
includes reduction in vivo and formation of the neutral Pt(II)
accumulation tendency in oligoelectrolyte systems dominated
compound cis-[PtCl2(NH3)(c-NH2C6H11)] (JM 118).6,8–11 Inside
by end effects.26–31 The purpose of the present study has been
the cell, the low chloride concentration favours formation of the
to further verify the influence of electrostatic properties on the
corresponding positively charged aqua complexes,12 a reaction
adduct formation process, with a focus on the naturally preferred
step that then allows for further facile interaction with many
platinum target; the guanine N7-atom (G-N7). Our preliminary
investigation32 showed that the rate of adduct fomation with
the non-charged G-N7 follows the same trends as reported by
us for interaction with the phosphodiester backbone, i.e. with
† Electronic supplementary information (ESI) available: Derivation of
the rate law. See http://www.rsc.org/suppdata/dt/b4/b418966c/
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
D a l t o n T r a n s . , 2 0 0 5 , 1 2 2 1 – 1 2 2 7
1 2 2 1
©

View Article Online
a rate maximum for interaction with the middle position. We
here further investigate the kinetic response to i) a change of
monovalent bulk cations for divalent ones, ii) an increase of the
total length of the DNA up to 49 bases, and iii) an increased
concentration of DNA during constant concentration of reactive
metal complex. Taken together, the kinetic observations all
support a reaction mechanism in which access to suitable
association sites in the direct vicinity of the target site controls
the rate of platination.
The synthesis of the complex cis-[PtCl2(NH3)(c-C6H11NH2)],
1a, has previously been described.35 Solutions of the monochloro
complex cis-[PtCl(NH3)(c-C6H11NH2)X]+/0, X = DMF or NO3
,
−
1b, were obtainedby addition of 0.98mole equivalents of AgNO3
(Baker Analysed Reagent) to the dichloro compound dissolved
in DMF (LAB-Scan). The mixture was vortexed in the dark
for about 20 h prior to removal of the precipitated AgCl by
centrifugation. Stock solutions of the monochloro compound
in DMF were stored in the dark at 8 ◦C. In aqueous solution, 1b
rapidly and quantitatively converts to the monoaqua derivative
cis-[PtCl(NH3)(c-C6H11NH2)(OH2)]+, 2, compare Scheme 1.
Experimental
HPLC measurements
Materials and solutions
The HPLC analysis was carried out on a LaChrom (Merck
Hitachi) chromatograph system with a D-7000 interface and a
D-7400 UV/Vis detector set at 260 nm and at constant temper-
ature. Separation of platinated from unreacted oligonucleotides
was obtained by using reversed phase Protein & Peptide columns
equipped with guards. Oligomers containing 9–17 bases were
separated using a C18 Vydac column (250 × 4.6 mm id, 10 lm
particle diameter) and those containing 25–49 bases by use of a
Pro-C4 YMC column (250 × 4.6 mm id, 5 lm particle diameter).
Different gradient systems were applied for these columns: i)
C18 column; 0.10 M ammonium acetate buffer (Merck) pH 6.0
was used as the mobile phase with an acetonitrile (LAB-Scan,
HPLC grade) gradient from 10% to 14% for 18 min at a flow of
Aqueous buffer solutions were prepared from KHC8H4O4
(ACROS) according to a standard procedure,33 and the pH
was adjusted by addition of HClO4 (Merck, p.a.). The ph-
thalate buffer was diluted 100 times for the kinetics measure-
ments, resulting in a constant potassium ion concentration of
0.50 mM. The variation of cation concentration was achieved
by further addition of small volumes of concentrated, aqueous
stock solutions of NaClO4 (Merck, p.a.), Mg(ClO4)2 (ACROS)
or Mn(NO3)2 (Labkemi). The pH was recorded by a standard
pH electrode (Methrom 744). All stock solutions were prepared
from HPLC quality water (Maxima, USF ELGA) and were kept
at room temperature.
The single-stranded DNA, d(TnGTn) where n = 4, 6, 8 and
12 and d(TnGT16−n) where n = 0, 2, 4, 6, 10, 12, 14 and 16
were purchased from Scandinavian Gene Synthesis AB, and the
oligonucleotides d(TnGTn) where n = 16 and 24 were bought
from Sigma-Genosys Ltd, see Scheme 1 for an overview of
studied DNAs. The oligonucleotides were received in aqueous
solutions and were kept frozen at −80 ◦C. Concentrations of
the oligomers were determined by absorption measurements at
260 nm using calculated extinction coefficients.34 Spectra were
recorded by use of a Milton Roy 3000 Diode-array, at 25 ◦C and
ambient pressure.
◦
1 ml min−1, t = 25 C, ii) Pro-C4 column; solutions of 0.10 M
triethylammonium acetate (TEAA, Calbiochem), pH 6.0 with
an acetonitrile gradien◦t from 12% to 14% for 12 min at a flow
of 1 ml min−1, t = 30 C. The chromatograms were evaluated
by use of an on-line HPLC System Manager Software working
under Microsoft Windows NT Workstation version 4.0.
Kinetics measurements
The adduct formation reactions were studied under pseudo-first-
order conditions with 2 in excess. If nothing else is stated, con-
stant concentrations of platinum complex and oligonucleotide
were employed; [Pt(II)] = 1.60 × 10−4 M and [DNA] = 4.0 ×
10−6 M, thus allowing for a direct comparison of the obtained
rate constants. The study of the DNA-concentration dependence
was performed in the interval [oligonucleotide] = (1.0–32) ×
◦
10−6 M. All reactions were studied at 25 C and pH 4.2 0.1.
This pH is well below the pKa value of 2 which allows the reactive
aqua-complex to dominate over the less reactive corresponding,
deprotonated hydoxo complex in solution.12,36 The platination
reactions were initiated by addition of a small volume of
1b in DMF to a thermostated and buffered solution of the
DNA. Aliquots were withdrawn at different time intervals and
directly quenched by dilution in buffer. The samples were then
immediately frozen, and if needed, stored in liquid nitrogen at
−196 ◦C. HPLC analysis was performed directly after thawing.
The time-dependent decrease of the integrated peak areas of the
non-reacted DNA was used to follow the kinetics. The observed
pseudo-first-order rate constant, kobs, was determined by a fit of
a single exponential function to the experimental data points.
Results
Kinetics evaluation
The kinetics for the platination reaction was evaluated by
use of the time-dependent decline of the peak eluting with
a retention time corresponding to unreacted DNA. Typical
examples of HPLC chromatograms are shown in Fig. 1, here
illustrated by the reaction of 2 with d(T6GT6) and d(T24GT24),
respectively. The inserted graphs show fits of the normalised
integrated peak areas of the unreacted oligonucleotide to a single
exponential function, confirming the assumption of pseudo-
first-order kinetics.
Scheme 1 Schematic illustration of studied DNAs and platinum(II)
reagent.
1 2 2 2
D a l t o n T r a n s . , 2 0 0 5 , 1 2 2 1 – 1 2 2 7

View Article Online
Fig. 1 Typical HPLC chromatograms illustrating the conversion of unreacted oligonucleotides to platinated ones for (A) reaction of d(T6GT6) with
2; [d(T6GT6)] = 4.0 × 10−6 M and (B) reaction of d(T24GT24) with 2; [d(T24GT24)] = 1.0 × 10−6 M. The inserted graphs illustrate the fit of a single
exponential function to the normalised, integrated peak areas vs. time for the unplatinated oligonucleotide (A: tr = 12.5 min and B: tr = 9.5 min).
Similar bulk reaction conditions were employed for both experiments; [2] = 1.60 × 10−4 M, [Na+] = 0.50 mM, [K+] = 0.50 mM.
Reaction profile in the presence of Na+, Mg2+ and Mn2+
A preliminary report of the kinetic influence from the DNA
environment as a function of position of G-N7 in d(TnGT16−n),
where n = 0, 2, 4, 6, 8, 10, 12, 14 and 16, at [Na+] + [K+] =
1.0 mM has been reported earlier by us.32 Further investigation
of the influence from the location of the target size in the presence
of divalent cations is presented here. The rate of platination of
the oligonucleotides d(TnGT16−n), where n = 0, 4, 8, 12 and 16,
was studied at the following cation concentrations; [Mg2+] =
0.5 mM, [Mn2+] = 0.5 mM, [Mg2+] = 34.5 mM and [Na+] =
34.5 mM. A summary of the observed pseudo-first-order rate
constants is given in Table 1 and selected data is shown in Fig. 2.
Fig. 2a illustrates the reactivity of G-N7 as a function
of position in d(T8GT8) at a constant cation concentration
of 1.0 mM. As can be seen here, platination is favoured at
the central region of the oligomer, both in the presence of
monovalent and divalent cations. At any given position within
the oligonucleotide, the same reactivity trend is observed with
k
obs(Na+) > kobs(Mg2+) > kobs(Mn2+). A comparison of the relative
reactivity along the oligonulceotide, see Fig. 2b, shows that the
reactivity profiles obtained in the presence of Na+ and Mg2+
are rather similar. A well pronounced reactivity maximum is
Table 1 Summary of observed pseudo-first-order rate constants for
platination of d(TnGT16−n) as a function of type and concentration of
cation in the reaction mediuma,b
Fig. 2 (A) Reactivity profiles for platination of G-N7 as a function of
its position along the 17-mer oligonucleotide d(TnGT16−n) at [cation] =
1.0 mM. [K+] = 0.5 mM and [Mn+] = 0.5 mM, where Mn+ = Na+ (closed
circles), Mg2+ (open squares) and Mn2+ (closed squares), pH 4.2 0.1,
t = 25 ◦C, [2] = 1.60 × 10−4 M and [d(TnGT16−n)] = 4.0 × 10−6 M. (B)
Relative reactivity compared to that obtained at the 5ꢀ-end as a function
of the location of G-N7 along the oligomer with Mn+ = Na+ (black
bars), Mg2+ (hatched bars), Mn2+ (white bars).
103 × kobs/s−1 for [cation]/mM
DNA-fragment
d(GT16)
Cationc
1.0
35
Na+
1.88 0.47
0.53 0.15
0.38 0.07
3.08 0.25
4.59 0.07
1.47 0.15
0.83 0.09
6.01 0.28
6.46 0.04
1.81 0.08
1.00 0.07
5.98 0.02
5.15 0.22
1.54 0.13
0.96 0.06
4.19 0.06
2.48 0.13
1.20 0.13
0.81 0.06
0.30 0.05
0.06 0.04
—
Mg2+
Mn2+
Na+
d(T2GT14)
d(T4GT12)
—
obtained at the middle of the oligomer in the presence of both
salts, and similar relative reactivity changes are observed in
particular at the 5ꢀ-side of the middle. In contrast, the presence
of Mn2+ seems to reduce the influence from local position
significantly, particularly at the 3ꢀ-side of the middle where
uniform reactivity is observed. In addition, the relative change
in reactivity on going from the 5ꢀ-end towards the middle is less
pronounced compared with the results obtained in the presence
of Na+ and Mg2+.
An increase of the cation concentration to 35 mM significantly
reduces the reactivity both in the presence of mono- and divalent
cations,‡ see Fig. 3. A pronounced reactivity profile is obtained
in the presence of the higher Na+-concentration, whereas the
Na+
0.97 0.04
0.12 0.01
—
Mg2+
Mn2+
Na+
d(T6GT10)
d(T8GT8)
—
Na+
1.18 0.18
0.15 0.01
—
Mg2+
Mn2+
Na+
d(T10GT6)
d(T12GT4)
—
Na+
1.08 0.14
0.13 0.01
—
Mg2+
Mn2+
Na+
d(T14GT2)
d(T16G)
—
Na+
0.56 0.12
0.09 0.01
—
Mg2+
Mn2+
a 25 ◦C, pH 4.2 0.1, buffered with KHC8H4O4. b Errors correspond to
‡ The reaction in the presence of Mn2+ was too slow to follow due to
complications arising from extensive hydrolysis of the Pt-compound.
one standard error. c [K+] = 5.0 × 10−4 M in all experiments.
D a l t o n T r a n s . , 2 0 0 5 , 1 2 2 1 – 1 2 2 7
1 2 2 3

View Article Online
Fig. 4 The observed pseudo-first-order rate constants for platination
of the central position of each oligomer as a function of increasing
oligonucleotide length (2n + 1) and increasing salt concentration. The
oligonucleotides d(TnGTn), n = 4, 6, 8, 12, 16 and 24, were used as targets
under the following conditions: [K+] = 0.5 mM and [Na+] = 0.5 mM
(closed circles), 4.5 mM (open squares), 14.5 mM (closed squares). [2] =
1.60 × 10−4 M and [d(TnGTn)] = 4.0 × 10−6 M.
Fig. 3 Comparison of reaction profiles in the presence of (A) monova-
lent salt and (B) divalent salt by plots of the observed pseudo-first-order
rate constants for platination of G-N7 as a function of position along
the 17-mer oligonucleotide d(TnGT16−n), [2] = 1.60 × 10−4 M and
[d(TnGT16−n)] = 4.0 × 10−6 M. (A) [K+] = 0.5 mM and [Na+] = 0.5 mM
(closed squares), 34.5 mM (open squares). (B) [K+] = 0.5 mM and
[Mg2+] = 0.5 mM (closed circles), [Mn2+] = 0.5 mM (closed diamonds),
[Mg2+] = 34.5 mM (open circles).
DNA concentration dependence
The rate of platination was studied as a function of increasing
concentration of the oligonucleotides, d(T8GT8) and d(T24GT24).
The concentration of 2 was kept constant and in excess (5–
160-fold) with respect to the oligonucleotide concentration
during these experiments, thus formally allowing for evaluation
according to pseudo-first-order conditions in all experiments
except the 5-fold excess. The obtained observed rate constants
are listed in Table 3. Fig. 5 shows a fit of the experimentally
obtained rate constants to the reaction model outlined in
Scheme 2. A similar decline in reactivity is observed as a function
of increasing DNA-concentration for both oligonucleotides, for
example with a reduction from kobs = 7.6 s−1 and 6.2 s−1 at 80-
fold excess to kobs = 2.0 s−1 and 1.0 s−1 at 10-fold excess of 2
reacting with d(T8GT8) and d(T24GT24), respectively.
variation in reactivity as a function of position along the
oligonucleotide in the presence of higher [Mg2+] is significantly
reduced, all in agreement with trends observed for adduct
formation processes with the phosphodiester backbone.25
In summary, the tendency of 2 to react at different sites within
the 17-mer at the various reaction conditions studied here can
be expressed in the following order: 0.5 mM Na+ > 0.5 mM
Mg2+ > 35 mM Na+ ≈ 0.5 mM Mn2+ > 35 mM Mg2+.
Oligonucleotide length dependence
The rate of adduct formation at G-N7 was studied as a function
of oligonucleotide length at [Na+] + [K+] = 1.0, 5.0 and
15 mM. The observed rate constants from the reaction of 2
with d(TnGTn), where n = 4, 6, 8, 12, 16 and 24, are presented
in Table 2 and illustrated in Fig. 4. For all salt concentrations
studied, a linear increase of the reactivity can be observed as
a function of increasing oligomer length up to ca. 17 bases. At
the lowest salt concentration studied, 1.0 mM, further increase
of the oligomer size results in a significant reduction of the
platination rate. At the higher cation concentrations employed,
5.0 and 15.0 mM this decline become less distinct and is better
characterized as a region with constant platination rate.
Scheme 2 Reaction mechanism including the non-productive preasso-
ciation step.
Table 2 Observed pseudo-first-order rate constants for platination of
d(TnGTn), n = 4, 6, 8, 12, 16 and 24a,b,c
Table 3 Observed pseudo-first-order rate constants for platination of
d(T8GT8) and d(T24GT24) at different oligonucleotide concentrationsa,b,c
103 × kobs/s−1 for [cation]/mM
103 × kobs/s−1
d(T8GT8)
DNA-fragment
1.0d
5.0d
15.0d
106 × [DNA]/M
Excess [Pt(II)]
d(T24GT24)
d(T4GT4)
d(T6GT6)
d(T8GT8)
d(T12GT12)
d(T16GT16)
d(T24GT24)
2.93 0.03
4.52 0.32
6.46 0.04
4.34 0.13
3.74 0.46
3.62 0.51
1.84 0.01
2.90 0.28
3.57 0.27
3.28 0.21
3.37 0.33
2.60 0.12
0.69 0.03
1.33 0.15
1.81 0.12
1.58 0.10
1.71 0.13
1.69 0.09
32
16
8.0
4.0
2.0
1.0
5
10
20
40
80
0.62 0.10
1.95 0.42
4.03 0.05
6.46 0.04
7.56 0.23
—
—
1.00 0.04
1.95 0.04
3.62 0.51
6.20 0.74
8.35 0.49
160
a [Pt(II)] = 1.6 × 10−4 M, [DNA] = 4.0 × 10−6 M. b 25 ◦C, pH 4.2
0.1, buffered with KHC8H4O4. c Errors correspond to one standard
deviation. d [cation] = [Na+ + K+] with [K+] = 5.0 × 10−4 M.
a [Na+ + K+] = 1.0 mM, [Pt(II)] = 1.6 × 10−4 M. b 25 ◦C, pH 4.1 0.1,
buffered with KHC8H4O4. c Errors correspond to one standard error.
1 2 2 4
D a l t o n T r a n s . , 2 0 0 5 , 1 2 2 1 – 1 2 2 7

View Article Online
salt concentration, [cation] = 1.0 mM, the apparent reactivity
in the presence of Na+ exceeds that obtained in the presence of
Mg2+ and Mn2+ with up to ca. one order of magnitude.
The variation in reactivity observed as a function of choice
and concentration of bulk metal ions, Mn+, is here accounted for
by use of the mechanisms and corresponding rate laws outlined
in eqns. (1)–(4).
Pt+ + DNA − −GN7· · ·Mn+DNA − −GN7· · ·Pt+ + Mn+
(1)
k
2
DNA − −GN7· · ·Pt+ ꢀ DNA − −GN7− − Pt+
v = k2 Kass[Pt+][DNA–GN7 · · · Mn+]/[Mn+]
(2)
(3)
kobs = k2Kass[Pt+]/[Mn+]
(4)
k2,app = k2Kass/[Mn+]
In the first step, rapid preassociation of the Pt(II) complex is as-
sumed to take place in parallel with release of bulk cations from
the DNA surface, eqn. (1). In the second reaction step, surface
bound Pt(II)-complexes located in the vicinity of G-N7 undergo
the adduct formation reaction in a formally monomolecular
reaction step to form the final adduct, i.e. formation of the
coordinate covalent bond, see eqn. (2) and bottom path in
Scheme 2. The rate law thus implies a rate dependence on three
factors: i) the bulk salt concentration, ii) the magnitude of Kass
and iii) the rate-constant for coordinate covalent bond formation
k2, the latter being independent on bulk salt conditions. All our
experimental findings are in agreement with the predictions that
can be made using this rate law. First of all, for a constant
choice of bulk metal ion, the rate of platination is always
decreasing with increasing bulk cation concentration, see for
example Fig. 3A. Second, the rate of platination should exhibit
a dependence on the magnitude of the binding constant Kass. By
a change from monovalent to divalent bulk cations a reduction
of Kass can be expected, and thus the observed platination rate,
due to the combined contributions from a) decreased attraction
between the approaching Pt-complex and the DNA due to the
more efficient screening of the phosphodiester backbone and b)
the increased free energy needed for release of divalent cation
from the DNA surface. The experimental findings nicely support
prediction as illustrated by a comparison of the data presented
in Fig. 3A and 3B.
A comparison of the reactivity data obtained in the presence
of Mg2+ and Mn2+ reveals that not only concentration and
net charge but also actual choice of metal ion influences the
rate by which platinum adducts are formed in a rather subtle
way. Despite their similar charge, Mg2+ and Mn2+ have different
characteristics with respect to their mode of interaction with
ribonucleic acids. Magnesium(II) preferentially interacts with
the phosphodiester backbone, whereas manganese(II) has been
shown to interact electrostatically with the G-N7 moiety, i.e. the
target of the Pt(II)-complex. The data summarized in Figs. 2A
and 3B show clearly that Mn2+ prevents platination more
efficiently than does Mg2+. In addition, a comparison of the
relative reactivity of each position of the oligomer, see Fig. 3B,
reveals a close resemblance for reactions in the presence of Na+
and Mg2+ with a similar reactivity maximum in the middle. In
contrast, uniform reactivity is observed on the 3ꢀ-side of the
oligomer in the presence of Mn2+. Taken together these data
seem to suggest that Mn2+ might suppress the platination rate
by competitive inhibition for the association sites in the vicinity
of G-N7, possibly in combination with a change of the DNA
structure in such a way that the ability of the Pt(II) complex
to discriminate between different association sites along the
oligomer is significantly reduced.42–45
Fig. 5 Observed pseudo-first-order rate constants for platination of
(A) d(T8GT8) and (B) d(T24GT24) as a function of phosphodiester
group concentration, [DNA · · · Pi], together with a fit of the data to the
inhibition model outlined in eqns. (5)–(7). The following experimental
conditions were used: [K+] = 0.50 mM, [Na+] = 0.50 mM, [Pt(II)] =
1.60 × 10−4 M, (A) [d(T8GT8)] = 2.0–32 × 10−6 M and (B) [d(T24GT24)] =
1.0–16 × 10−6 M.
Discussion
Reaction profile in the presence of Na+, Mg2+ and Mn2+
In solution, the overall negative charge of ribonucleic acids
is partly reduced by cations from the bulk. The ability of
the polymer to attract cations to its close vicinity, i.e. its
condensation layer, depends on factors such as size and structure
of the polymer and type of bulk cations in the surrounding
electrolyte.37 For example, extended rod-like structures such as
double-stranded DNA carry more cations in the condensation
layer compared with small, non-linear structures such as hairpin
DNAs. Further, divalent cations are more tightly bound to
the surface region than monovalent ones.38,39 The tendency for
cation accumulation at a given position, and the corresponding
local binding constant, varies with its actual location within the
oligomer and its overall structure. Theoretical calculations sug-
gest for example that cation accumulation on DNA-fragments
shorter than ca. 50 bases are highly non-uniform and is better
described as dominated by oligoelectrolyte end-effects with an
approximately linear increase of accumulation tendency towards
the middle of the oligomer.26–31
Previous studies by us and others indicate that preassociation
on the DNA surface is an important reaction step in the reaction
mechanism during covalent modification of DNA with cationic
metal complexes in a broad reactivity range.20–23,40,41 In addition,
we have also been able to show how the local positioning of a
given interaction site, in a constant sequence context, affects the
rate by which adducts are being formed.25,32 More specifically,
the rate of adduct formation was studied as a function of target
position within a given size DNA fragment, i.e. systems of the
type d(TnXT16−n) with X as a phosphorthioate group or guanine,
p(S) or G-N7 respectively. In the present study we have extended
these studies to include also the influence of divalent bulk cations
on the kinetics of adduct formation with G-N7, i.e. in the
presence of Mg2+ and Mn2+, respectively. As shown in Fig. 2a,
distinct reaction profiles are obtained for adduct formation with
G-N7 in the presence of all types of bulk cations. At low total
Oligonucleotide length dependence
Previous studies by us have shown that the rate of adduct
formation with cationic complexes increases with increasing
D a l t o n T r a n s . , 2 0 0 5 , 1 2 2 1 – 1 2 2 7
1 2 2 5

View Article Online
oligomer size in the interval ca. 4–16 bases.20,21 For the first
time we here present reactivity data including a systematic study
of oligomers up to a total length of 49 bases, see Fig. 4. For
all salt concentrations studied, a common tendency of increased
reactivity in the interval 9–17 bases is observed. The effect of
further increase of the oligomer length depends on the bulk salt
concentrations, however. At low salt concentration, [monovalent
salt] = 1.0 mM, further addition of bases to the oligomer
results in a successive decline in apparent reactivity that is most
pronounced in the region 20–30 bases. In contrast, for the two
higher salt concentrations used, 5.0 and 15.0 mM respectively, a
constant but salt dependent reactivity is obtained in the interval
17–49 bases.
A plausible explanation for the different reaction profiles
can be found by assuming the monovalent Pt(II) complex to
serve as a non-specific counterion for the negatively charged
phosphodiester group, and thus being trapped in a position
where product formation is prevented. Under the experimental
conditions chosen for this study, the maximum concentration
of phosphodiester groups is defined by the experiment with
the oligomers containing 49 bases; ca. 2 × 10−4 M. For
the two higher salt concentrations employed, 5.0 mM and
15.0 mM, charge neutralization of the phosphodiester groups
can thus be achieved without significant change of the bulk salt
concentrations. In addition, these salt concentrations are more
than one order of magnitude above that of the Pt(II) complex
thus effectively suppressing participation of Pt(II) in its role as
a non-specific charge neutralization reagent. The amount of
Pt(II) complex used for neutralization of the charge created by
the phosphodiester backbone is thus likely to be minimized,
despite the preference found for DNA association with Pt(II)
over Na+. The data obtained at 1.0 mM salt concentration
indicates however that the effective concentration of the Pt(II)
complex is reduced when the ratio [salt] : [phosphate] is below
ca. 10 : 1, a value that agrees well with what can be expected
using the apparent inhibition constant that can be calculated
from available experimental data, vide infra.
corresponding equilibrium constant Kass,i, see eqn. (5).
K
ass,j
Pt+ + DNA − −Pi· · ·Mn+ ꢀ DNA − −Pi· · ·Pt++Mn+
(5)
The resulting general expression for [Pt+], under presently used
experimental conditions with [Pt+] >> [G-N7], is given in
eqn. (6).
[Pt+] = CPt[Mn+]/([Mn+] + Kass,i [DNA–Pi · · · Mn+])
The final expression for the observed rate constant is eqn. (7),
(6)
kobs = k2,appCPt[Mn+]/([Mn+] + Kass,i[DNA–Pi · · · Mn+])
(7)
the latter as a result of insertion of the expression for [Pt+] from
eqn. (6) into eqn. (4). As can be seen in Fig. 5, the experi-
mentally observed decline in kobs as a function of increasing
phosphodiester concentration could successfully be fitted to the
expression in eqn. (7), allowing for determination of an average
binding constant to individual phosphate groups, Kass,i, = 18
7 and 16
3 for association with d(T8GT8) and d(T24GT24),
respectively. These binding constants support the idea that weak
electrostatic interactions may contribute to facilitate formation
of specific hydrogen bonds, e.g. between ammine ligands of
cisplatin and JM 118 (cis-[PtCl2(NH3)(c-NH2C6H11)]) and the
adjacent 5ꢀ-phosphate in the DNA duplex,17,48,49 that are crucial
in determining the rate and/or efficacy by which nucleic acids
are platinated in vivo.
Conclusions
The present study provides quantitative evidence for the pres-
ence of a preassociation step between the monovalent, active
metabolite of the compound cis,trans,cis-[PtCl2(OAc)2NH3(c-
C6H11NH2)] and DNA. The conditional association constants
determined in the presence of monovalent cations, Ki ≈ 10–
20, gives a rationale for the well documented preference of for
example cisplatin for the targeting of nucleic acids in vivo. In
addition, the inhibitory effect caused by divalent metal ions,
most pronounced by Mn2+, further indicates that steric factors,
i.e. access to preassociation sites in the direct vicinity of the final
target site, effectively controls the rate of platination.
DNA concentration dependence
As discussed above, several independent observations now
indicate that preassociation may play a dual role in the reaction
mechanism of anticancer active Pt(II) complexes during their
interaction with DNA.20,21,24,25 As a global phenomenon, the
electrostatically driven preassociation facilitates compartmen-
talization of cations in the living cell by increasing the local
concentration of cations in the DNA vicinity. Once located
here, target accessibility will influence the adduct formation rate.
Under extreme conditions however, association may effectively
prevent adduct formation due to unfavourable geometrical
constraints. In the present study a more subtle balance has
been observed, however, with divalent cations as efficient kinetic
inhibitors for the adduct formation process.
References
1 B. Rosenberg, L. Van Camp and T. Krigas, Nature, 1965, 205, 698.
2 B. Rosenberg, L. Van Camp, J. E. Trosko and V. H. Mansour, Nature,
1969, 222, 385.
3 (a) For recent reviews see: T. Boulikas and M. Vougiouka, Oncology
Reports, 2003, 10, 1663; (b) J.-C. Bertrand, DNA Repair in Cancer
Therapy, Humana Press, 2004, p. 51; (c) K. R. Barnes and S. J.
Lippard, Met. Ions Biol. Syst., 2004, 42, 143.
4 (a) For recent reviews see: R. P. Wernyj and P. J. Morin, Drug
Resistance Updates, 2004, 7, 227; (b) G. Giaccone, Drugs, 2000, 59,
9; (c) M. A. Fuertes, C. Alonso and J. M. Perez, Chem. Rev., 2003,
103, 645.
To further test the hypothesis concerning the non-productive
preassociation pathway, a series of experiments were made in
which the concentration of Pt(II) was kept constant during
increase of the DNA concentration. Two DNA oligomers
were used in these experiments, d(T8GT8) and d(T24GT24). The
decrease in reactivity observed as a function of increasing DNA-
concentration was interpreted as a result of a reaction mech-
anism where non-productive preassociation occurs in parallel
with the adduct formation step, see Scheme 2, top and bottom
reaction paths, respectively. The alternative of introducing a
reversible step involving a coordinatively bound phosphate
was discarded due to the slow rate by which such complexes
can be estimated to be formed, i.e. t1/2 ≥ 10 h under present
experimental conditions.46,47 The decrease in reaction rate was
accounted for by introduction of an equilibrium between
reactive metal complex in the bulk solution, Pt+ and non-
reactive phosphate-associated complex, DNA–Pi · · · Pt, with
5 (a) For recent reviews see: L. R. Kelland, Am. J. Cancer, 2002, 1, 247;
(b) J. Lokich, Cancer Invest., 2001, 19, 756.
6 L. R. Kelland, B. A. Murrer, G. Abel, C. M. Giandomenico, P. Mistry
and K. R. Harrap, Cancer Res., 1992, 52, 822.
7 L. R. Kelland, S. Y. Sharp, C. F. O’Niell, F. I. Raynaud, P. J. Beale
and I. R. Judson, J. Inorg. Biochem., 1999, 77, 111.
8 F. I. Raynaud, D. E. Odell and L. R. Kelland, Br. J. Cancer, 1996,
74, 380.
9 F. I. Raynaud, F. E. Boxall, P. Goddard, C. F. Barnard, B. A. Murrer
and L. R. Kelland, Anticancer Res., 1996, 16, 1857.
10 G. K. Poon, F. I. Raynaud, P. Mistry, D. E. Odell, L. R. Kelland,
K. R. Harrap, C. F. J. Bernard and B. A. Murrer, J. Chromatogr. A.,
1995, 712, 61.
11 L. R. Kelland, Expert Opin. Invest. Drugs, 2000, 9, 1373.
12 S. J. Barton, J. Kevin, A. Habtemariam, R. E. Sue, E. Rodney and
P. J. Sadler, Inorg. Chim. Acta, 1998, 273, 8.
13 S. J. Barton, K. J. Barnham, U. Frey, A. Habtemariam, R. E. Sue
and P. J. Sadler, Aust. J. Chem., 1999, 52, 173.
14 J. F. Hartwig and S. J. Lippard, J. Am. Chem. Soc., 1992, 114, 5646.
1 2 2 6
D a l t o n T r a n s . , 2 0 0 5 , 1 2 2 1 – 1 2 2 7

View Article Online
˚
15 K. J. Mellish, C. F. Barnard, B. A. Murrer and L. R. Kelland,
Int. J. Cancer, 1995, 62, 717.
16 K. J. Yarema, J. M. Wilson, S. J. Lippard and J. M. Essigmann, J. Mol.
Biol., 1994, 236, 1034.
17 A. P. Silverman, B. Wu, S. M. Cohen and S. J. Lippard, J. Biol. Chem.,
2002, 277, 49743.
18 C. F. O’Niell, B. Koberle, J. R. W. Masters and L. R. Kelland,
Br. J. Cancer, 1999, 81, 1294.
32 A. Sykfont, A. Ericson and S. K. C. Elmroth, Chem. Commun., 2001,
1190.
33 CRC Handbook of Chemistry and Physics, ed. D. R. Lide, CRC Press,
Boca Raton, FL, 1994.
34 http://paris.chem.yale.edu/extinct.html.
35 C. M. Giandomenico, M. J. Abrams, B. A. Murrer, J. F. Vollano,
M. I. Rheinheimer, S. B. Wyer, G. E. Bossard and J. D. Higgins,
Inorg. Chem., 1995, 34, 1015.
19 S. M. Cohen and S. J. Lippard, Prog. Nucl. Acid Res., 2001, 67, 93.
20 S. K. C. Elmroth and S. J. Lippard, J. Am. Chem. Soc., 1994, 116,
3633.
21 S. K. C. Elmroth and S. J. Lippard, Inorg. Chem., 1995, 34, 5234.
22 A. Ericson, J. L. McCary, R. S. Coleman and S. K. C. Elmroth, J. Am.
Chem. Soc., 1998, 120, 12680.
23 A. Ericson, Y. Iljina, J. L. McCary, R. S. Coleman and S. K. C.
Elmroth, Inorg. Chim. Acta, 2000, 297, 56.
24 J. Kjellstro¨m and S. K. C. Elmroth, Chem. Commun., 1997, 1701.
25 J. Kjellstro¨m and S. K. C. Elmroth, Inorg. Chem., 1999, 38, 6193.
26 M. O. Fenley, G. S. Manning and W. K. Olson, Biopolymers, 1990,
30, 1191.
36 S. J. Berners-Price, T. A. Frenkiel, J. D. Randford and P. J. Sadler,
J. Chem. Soc., Chem. Commun., 1992, 10, 789.
37 G. S. Manning, Q. Rev. Biophys., 1978, 11, 179.
38 J. A. Cowan, Chem. Rev., 1998, 98, 1067.
39 A. Pullman and B. Pullman, Q. Rev. Biophys., 1981, 14, 289.
40 Y. Wang, N. Farrell and J. D. Burgess, J. Am. Chem. Soc., 2001, 123,
5576.
41 H. T. Chifotides, J. M. Koomen, M. Kang, S. E. Tichy, K. R. Dunbar
and D. H. Russell, Inorg. Chem., 2004, 43, 6177.
42 J. Reuben and E. J. Gabbay, Biochemistry, 1975, 14, 1230.
43 A. Yamada, K. Akasaka and H. Hatano, Biopolymers, 1976, 15,
1315.
27 M. C. Olmsted, C. F. Anderson and M. T. J. Record, Proc. Natl.
Acad. Sci. USA, 1989, 86, 7766.
28 W. Zhang, J. P. Bond, C. F. Anderson, T. M. Lohman and M. T. J.
Record, Proc. Natl. Acad. Sci. USA, 1996, 93, 2511.
29 W. Zhang, H. Ni, M. W. Capp, C. F. Anderson, T. M. Lohman and
M. T. J. Record, Biophys. J., 1999, 76, 1008.
44 J. Granot, J. Feigon and D. R. Kearns, Biopolymers, 1982, 21, 181.
45 J. Granot and D. R. Kearns, Biopolymers, 1982, 21, 203.
46 R. N. Bose, R. E. Viola and R. D. Cornelius, J. Am. Chem. Soc.,
1984, 106, 3336.
47 M. S. Davies, S. J. Berners-Price and T. W. Hambley, Inorg. Chem.,
2000, 39, 5603.
30 M. C. Olmsted, J. P. Bond, C. F. Anderson and M. T. J. Record,
Biophys. J., 1995, 68, 634.
31 V. M. Stein, J. P. Bond, M. W. Capp, C. F. Anderson and M. T. J.
Record, Biophys. J., 1995, 68, 1063.
48 P. M. Takahara, A. C. Rosenzweig, C. A. Frederick and S. J. Lippard,
Nature, 1995, 377, 649.
49 P. M. Takahara, C. A. Frederick and S. J. Lippard, J. Am. Chem.
Soc., 1996, 118, 12309.
D a l t o n T r a n s . , 2 0 0 5 , 1 2 2 1 – 1 2 2 7
1 2 2 7