Oxidation of 5′-dGMP, 5′-dGDP, and 5′-dGTP by a platinum(IV) complex

Article abstract of DOI:10.1007/s00775-015-1312-0

We previously reported that a Pt(IV) complex, [Pt^IV(dach)Cl₄] [trans-d,l-1,2-diaminocyclohexanetetrachloroplatinum(IV)] binds to the N₇ of 5′-dGMP (deoxyguanosine-5′-monophosphate) at a relatively fast rate and oxidizes it to 8-oxo-5′-dGMP. Here, we further studied the kinetics of the oxidation of 5′-dGMP by the Pt^IV complex. The electron transfer rate constants between 5′-dGMP and Pt^IV in [H₈-5′-dGMP-Pt^IV] and [D₈-5′-dGMP-Pt^IV] were similar, giving a small value of the kinetic isotope effect (KIE: 1.2 ± 0.2). This small KIE indicates that the deprotonation of H₈ in [H₈-5′-dGMP-Pt^IV] is not involved in the rate-determining step in the electron transfer between guanine (G) and Pt^IV. We also studied the reaction of 5′-dGDP (deoxyguanosine-5′-diphosphate) and 5′-dGTP (deoxyguanosine-5′-triphosphate) with the Pt^IV complex. Our results showed that [Pt^IV(dach)Cl₄] oxidized 5′-dGDP and 5′-dGTP to 8-oxo-5′-dGDP and 8-oxo-5′-dGTP, respectively, by the same mechanism and kinetics as for 5′-dGMP. The Pt^IV complex binds to N₇ followed by a two-electron inner sphere electron transfer from G to Pt^IV. The reaction was catalyzed by Pt^II and occurred faster at higher pH. The electron transfer was initiated by either an intramolecular nucleophilic attack by any of the phosphate groups or an intermolecular nucleophilic attack by free OH^- in the solution. The rates of reactions for the three nucleotides followed the order: 5′-dGMP > 5′-dGDP > 5′-dGTP, indicating that the bulkier the phosphate groups are, the slower the reaction is, due to the larger steric hindrance and rotational barrier of the phosphate groups.

Full text of DOI:10.1007/s00775-015-1312-0

J Biol Inorg Chem (2015) 20:1327–1341
DOI 10.1007/s00775-015-1312-0
ORIGINAL PAPER
Oxidation of 5′‑dGMP, 5′‑dGDP, and 5′‑dGTP by a platinum(IV)
complex
1
1
1
Ioannis Kipouros · Sebastian Matias Fica‑Contreras · Gregory Joon Kee Bowe ·
Sunhee Choi
1
Received: 9 September 2015 / Accepted: 2 November 2015 / Published online: 21 November 2015
©
SBIC 2015
Abstract We previously reported that a Pt(IV) com-
Keywords Oxidation · 5′-dGMP · 5′-dGDP · 5′-dGTP ·
Platinum(IV) complex
IV
plex, [Pt (dach)Cl ] [trans-d,l-1,2-diaminocyclohexan
4
etetrachloroplatinum(IV)] binds to the N7 of 5′-dGMP
(
deoxyguanosine-5′-monophosphate) at a relatively fast rate
and oxidizes it to 8-oxo-5′-dGMP. Here, we further studied
the kinetics of the oxidation of 5′-dGMP by the PtIV com-
plex. The electron transfer rate constants between 5′-dGMP
Introduction
Oxidation of DNA plays a major role in carcinogenesis
[1, 2], aging [3, 4], and neurodegenerative diseases such
as Parkinson’s and Alzheimer’s disease [5, 6]. Among the
four nucleobases, guanine (G) has the lowest reduction
IV
IV
IV
and Pt in [H -5′-dGMP–Pt ] and [D -5′-dGMP–Pt ]
8
8
were similar, giving a small value of the kinetic isotope
effect (KIE: 1.2 ± 0.2). This small KIE indicates that the
IV
deprotonation of H in [H -5′-dGMP–Pt ] is not involved
potential (E = 1.3, 1.4, 1.6 and 1.7 V vs. NHE for dGuo,
8
8
0
in the rate-determining step in the electron transfer between
dAdo, dCyt and dThy, respectively) and, therefore, is most
easily oxidized [7]. Many transition metal complexes
including those of Cr, Mn, Fe, Co, Ni, Cu, Ir, Os, Re, and
Rh mediate oxidation of guanine via many different mech-
anisms and all require oxidizing radiation, photosensitiz-
ers, or oxidants [8, 9].
IV
guanine (G) and Pt . We also studied the reaction of
5
′-dGDP (deoxyguanosine-5′-diphosphate) and 5′-dGTP
IV
(
deoxyguanosine-5′-triphosphate) with the Pt complex.
IV
Our results showed that [Pt (dach)Cl ] oxidized 5′-dGDP
4
and 5′-dGTP to 8-oxo-5′-dGDP and 8-oxo-5′-dGTP,
III
IV
respectively, by the same mechanism and kinetics as for
The Ru and Pt complexes can also oxidize purine
derivatives, but unlike the oxidation by other transition
metal complexes cited above, this process does not involve
oxidizing radiation, photosensitizers, or oxidants. Both
IV
5
′-dGMP. The Pt complex binds to N followed by a two-
7
IV
electron inner sphere electron transfer from G to Pt . The
reaction was catalyzed by Pt and occurred faster at higher
II
III
IV
pH. The electron transfer was initiated by either an intramo-
lecular nucleophilic attack by any of the phosphate groups
Ru and Pt complexes are substitution inert but bind
slowly to the N of G (N ) followed by a two-electron
7
7
−
or an intermolecular nucleophilic attack by free OH in the
inner sphere redox reaction. Clarke’s group reported that
III
III
solution. The rates of reactions for the three nucleotides
followed the order: 5′-dGMP > 5′-dGDP > 5′-dGTP, indi-
cating that the bulkier the phosphate groups are, the slower
the reaction is, due to the larger steric hindrance and rota-
tional barrier of the phosphate groups.
the H of G (H ) bound to Ru (H G–Ru ) can be depro-
8
8
8
−
tonated by OH in basic solution (intermolecular nucleo-
phile), and a two-electron oxidation product of 8-oxo-G is
produced in the presence of O molecules [10–12]. Later,
2
IV
IV
our group reported that the H bound to Pt {H G–Pt :
8
8
IV
IV
Pt = [Pt (dach)Cl ]} can be deprotonated by 5′-OH or
4
5
′-phosphate (intramolecular nucleophile) forming a two-
*
Sunhee Choi
electron oxidation product of either cyclic-5′-O-C8-G or
choi@middlebury.edu
8
-oxo-G without involving O molecules [13, 14]. We also
2
IV
1
discovered that the redox reaction between Pt and G can
occur in basic solution or phosphate buffer, where OH or
Department of Chemistry and Biochemistry, Middlebury
College, Middlebury, VT 05753, USA
−
1
3

1
328
J Biol Inorg Chem (2015) 20:1327–1341
H
N
O
N
Cl
(
a)
Cl-
H-OH
-
-
PtIV
OH2
NH
O
N
O
N
II
O
Pt
..
N
+
O
N
N
NH - H
NH
NH2
-
H O
2
O-
NH2
H
O
O
P
O P O
-O
N
O-
N
- H+
O
NH2
-
O
O
O P O
O
OH
-
O
OH
OH
-
H O
2
- H+
Cl
Cl
PtIV
N
(
b)
PtIV
H-OH
N
O
O
- Cl-
NH - PtII
O
N
O
N
OH2
O
.
.
N
NH
O
N
O
NH
- H+
O
NH
NH2
H
O
O
O
-O
P
O
-O
P
O
-O
P
O-
O P O
N
N
N
NH2
H
N
N
NH2
H
N
NH2
O
O
O
O
-
O
OH
OH
OH
OH
IV
Scheme 1 Proposed mechanism on basis of Choi et al. for the oxidation of 5′-dGMP by [Pt (dach)Cl ] [14]. Deprotonation of H in a is
4
8
involved in the rate-determining step, but not in b [19]
phosphate in solution (intermolecular nucleophile) attacks
N to deprotonate at high pH to induce large negative charge
1
II
C of G (C ) to produce 8-oxo-G and Pt [15]. Through this
on C . Therefore, the main driving force for the redox reac-
8
8
8
mechanism, every G in the oligonucleotides can be oxi-
dized to 8-oxo-G in phosphate buffer.
tion should be the polarization of H by electron-deficient
Pt and Ru .
8
IV
IV
We further discovered that Ru complexes like the PtIV
III
One of our major questions in this investigation was at
complex, can go through the two-electron redox mecha-
what point the deprotonation of H occurs. Does it occur
8
III
nism without involving O [16, 17]. The [Ru (NH ) L]
right after the nucleophile attacks C (Scheme 1a) or after
2
3 5
8
IV
(
L = dGuo, dIno, Xao) complex disproportionates in basic
the two-electron transfer between G and Pt is complete
II
IV
solution to the corresponding Ru and Ru complexes,
followed by a two-electron redox reaction between Ru
(Scheme 1b)? Recent density function theory (DFT) calcu-
IV
lations on the oxidation of 5′-dGMP suggested that H is
8
and L that does not involve O . When L was dGuo, the
two-electron oxidation product was cyclic-5′-O-C8-dGuo
deprotonated after the redox reaction (Scheme 1b) [19]. To
shed light on which mechanism is more plausible, we deter-
mined the kinetic isotope effect (KIE = k /k ) of the elec-
2
[
16], but when L was dIno or Xao, the oxidation product
H
D
was 8-oxo-L [17].
tron transfer rate. In Scheme 1a, the rate-determining step
What is the main driving force for the redox reaction
between G and Pt (and Ru /Ru )?
is deprotonation of H , which should give a primary KIE of
8
IV
III
IV
7 [20]. On the other hand, in Scheme 1b, the H is depro-
8
IV
Coordination of a metal ion at N inductively withdraws
tonated after a two-electron transfer from G to Pt , which
7
electron density from G, placing a partial positive charge at
C and promotes the exchange of H [18]. Clarke’s group
should give a secondary KIE of a value much smaller than
IV
7. Our results showed that the KIE of fac-[Pt (dach)Cl (5′-
8
8
3
III
showed that the autoxidation rate of [Ru (NH ) L] (L = Ino,
dGMP)] was 1.2 ± 0.2, indicating that the deprotonation
3
5
IV
1-MeIno, 1-MeGuo, Guo, and dGuo) at high pH to the corre-
of H -5′-dGMP–Pt is not involved in the rate-determining
8
sponding 8-oxo complex depends on the polarity of H [12].
step. Therefore, the plausible mechanism is not the paths in
Scheme 1a but rather in Scheme 1b, which is in agreement
with the DFT calculations [19].
Another objective of this study was to investigate the
mechanism and kinetics of the oxidation of other nucleo-
tides such as 5′-deoxyguanosine diphosphate (5′-dGDP)
and 5′-deoxyguanosine triphosphate (5′-dGTP), and com-
pare them with the conclusions made for the reaction
with 5′-dGMP [13–15, 21]. This new objective is biologi-
cally important as 5′-dGDP and 5′-dGTP comprise a major
8
The change in polarity of C is reflected in the chemical shift
8
III
1
of H of [Ru (NH ) L] in the H NMR spectra at −31.24,
8
3 5
−
−
−
−
30.37, and −20.22 ppm for dGuo , Guo , and Ino ,
respectively. The rate follows the order dGuo < Guo < Ino
indicating that the more polar the H is, the faster the
8
redox reaction becomes. The oxidation of the 1-MeIno (or
1-MeGuo) complex proceeds approximately 3 times faster
than that of the dGuo (or Guo) complex. The H of 1-MeGuo
8
is highly susceptible to polarization because there is no H at
1
3

J Biol Inorg Chem (2015) 20:1327–1341
1329
fraction of the intracellular nucleotide pool, and their modifi-
cation by transitional metal complexes can lead to mutations
through their incorporation in the DNA chain during replica-
tion. These indirect modifications of the DNA molecule could
cause mutagenesis in the same fashion as the direct modifica-
collected. The solid was isolated by vacuum filtration and
identified by H NMR and ESI-LC/MS. The H NMR peak
1
1
at 8.3 ppm corresponding to the H G of 5′-dGMP had dis-
8
appeared and the ESI-LC/MS showed an m/z peak of 348.5
(m/z for 5′-dGMP = 347.5), indicating that the reaction was
completed and the 5′-dGMP was completely deuterated.
tion of the normal guanine bases of DNA [22]. Our results
IV
showed that [Pt (dach)Cl ] oxidized 5′-dGDP and 5′-dGTP
4
to 8-oxo-5′-dGDP and 8-oxo-5′-dGTP, respectively, by the
Preparation of reactions
same mechanism and kinetics as in the case of 5′-dGMP
IV
IV
[
13–15, 21]. The Pt complex binds to N followed by a two-
Stock solutions of [Pt (dach)Cl ] (10 mM) were prepared by
7
4
IV
IV
electron inner sphere electron transfer from G to Pt . The
dissolving appropriate amounts of [Pt (dach)Cl ] in water
4
II
16
18
reaction was catalyzed by Pt and occurred faster at higher
(in the form of H O, H O, or D O) followed by stirring and
2 2 2
pH. The electron transfer was initiated by either an intramo-
lecular nucleophilic attack by any of the phosphate groups
or an intermolecular nucleophilic attack by free OH in the
gentle heating (maximum 50 °C) for at least 6 h. An appro-
priate amount of NaCl was also added to a final concentra-
tion of 100 mM to decrease the hydrolysis of [Pt (dach)
−
IV
solution. The rates of reactions for the three nucleotides fol-
lowed the order: 5′-dGMP > 5′-dGDP > 5′-dGTP, indicating
that the bulkier the phosphate groups are, the slower the reac-
tion is, due to the larger steric hindrance and rotational barrier
of the phosphate groups.
Cl ] through the replacement of its chloride ligands by water.
4
II
When needed [Pt (dach)Cl ] was also dissolved in the solu-
2
tion in low concentrations (~0.02 mM). A fresh solution of
IV
IV
[Pt (dach)Cl ] was prepared for each reaction as [Pt (dach)
4
Cl ] hydrolyzes when in solution for extended periods of
4
time. Buffers were not used to avoid complications arising
from buffer coordination to platinum. Appropriate amounts
of 5′-dGMP, 5′-dGDP or 5′-dGTP were weighted in 1.5-mL
Materials and methods
Eppendorf tubes, in which the desired volume of the dis-
IV
Chemicals
solved [Pt (dach)Cl ] solution was transferred. The solu-
4
tion was vortexed for 5 min and the pH was then raised to
7.0 or 8.6 using 1.0 M NaOH (or NaOD). Each solution
was then incubated at 37 °C in a Teckam Dri-Block with a
DB-1 Heater, while covered in aluminum foil to avoid any
form of photodegradation. All the samples for NMR analysis
IV
[
Pt (dach)Cl ] was obtained from the National Cancer
4
Research Institute, Drug Synthesis and Chemistry Branch,
Development Therapeutics Program, Division of Cancer
Treatment. The deoxy-nucleotides, 5′-dGMP (>99 %) and
5
5
′-dGDP (>90 %) were obtained from Sigma-Aldrich, and
′-dGTP (>98 %) was obtained from GE Healthcare Life Sci-
were prepared in D O (>98 %). Most reaction solutions were
filtered through a syringe-end filter disk (Gelman Acrodisk
0.45 µm pore size) before ESI-LC/MS and NMR analysis.
2
ence, and were all used as sodium salts. NaCl, NaOH, HCl,
ammonium acetate, and HPLC-grade acetonitrile were pur-
1
8
chased from Fisher Scientific. Radiolabeled H O (97 %)
2
1
31
was purchased from Berry & Associates, Synthetic Medicinal
Chemistry. All other chemicals were obtained from Sigma-
Aldrich. All chemicals were used as purchased without any
further purification. All water used was purified by a Hydro
Picotech 2 purification system. The pH of reactions and sol-
vents was adjusted with 1.0 M NaOH and 0.5 M HCl, using an
Orion Research Expandable Ion Analyzer EA 940 equipped
with Corning Semi-Micro Combo electrode. For NMR experi-
H and P NMR spectroscopy
NMR spectra were recorded on a Bruker 400 Ultrashield
spectrometer in D O at room temperature. The H NMR
1
2
spectra of all reactions were taken at 400.1319 MHz over
50–256 scans. No water-suppression method was used,
since the peaks of interest had chemical shifts that were
not close to that for water. For the P NMR spectra, 2000
scans were taken at 16233.77 Hz. All free inductions
decays (FID) were Fourier transformed and analyzed using
Bruker TOPSPIN 1.3 software. Spectra were edited using
the XWIN-PLOT software.
3
1
ments D O (99.9 %) was used as the solvent, and the pH of
2
the solution samples was adjusted with 1.0 M NaOD (99.9 %),
both chemicals were obtained from Sigma-Aldrich.
Synthesis of D ‑5′‑dGMP
8
Electron spray ionization–liquid chromatography/mass
spectrometry (ESI‑LC/MS)
D -5′-dGMP was prepared by adapting a previously pub-
8
lished procedure [23]. Briefly, 0.0203 g 5′-dGMP was dis-
solved in 5 mL of D O and heated at 80 °C for 24 h. The
Liquid chromatography coupled to a mass spectrometer
was used to identify the intermediates and final oxidation
2
solution was dried under N gas, and a white solid was
2
1
3

1
330
J Biol Inorg Chem (2015) 20:1327–1341
products of the reactions. A 1100 Series LC, coupled to an
LC-MSD Trap XCT Plus system from Agilent Technolo-
gies, was used to perform sample analysis. The liquid chro-
matography component was equipped with a photo diode
array detector and an Eclipse X-D8-C8 column. Isocratic
elution was performed with 0.5 % aqueous formic acid
solution at a flow rate of 0.5 mL/min. The mass spectrom-
eter was set to scan in the range of 50–2200 m/z, detect-
ing positive ions with electrospray ionization mass spec-
trometry (ESI–MS). The nitrogen nebulizer pressure was
set to 50 psi, the dry gas (helium) flow to 9 L/min, and the
dry temperature to 365 °C. The ion trap was set to collect
targeted molecules and acquire their fragmentation mass
spectra.
k
s
PtIV–Cl
+
G
+
PtII
PtIV–G
PtIV–G
PtII
+
PtII
G (-2e)
(1)
(2)
ke
+
IV
13
Scheme 2 Kinetic model for the oxidation of G by [Pt (dach)Cl ]
4
As indicated by the decrease in A360, the disappearance of
the reactant, [Pt (dach)Cl ], is slow in the beginning, but
increases in rate as the reaction progresses because the
reaction is auto-catalyzed by the product, [Pt (dach)Cl ].
The A390 curve arises from the intermediate, fac-[Pt (dach)
(5′-dGMP)Cl ] . The kinetic curves of A and A390 vs.
time were fitted with the DynaFit Software [24] utilizing
the kinetic model in Scheme 2 [21]. The substitution rate
constant (k ) and the electron transfer rate constant (k ) giv-
IV
4
II
2
IV
+
3
360
s
e
UV/Vis spectroscopy
ing the best fit to both A360 and A390 curves were identical.
The k and k values giving the best fit to the curves
s
e
UV/Vis spectroscopy was used to monitor the reaction rates
of A360 and A390 (Fig. 1) were basically the same for H8-
IV
of [Pt (dach)Cl ] with 5′-dGMP, 5′-dGDP and 5′-dGTP.
5′-dGMP and D -5′-dGMP. The k of H -5′-dGMP and
4
8
s
8
UV/Vis spectra were obtained in 10-mm-path length micro
quartz cells on a Varian Cary 4000 Spectrophotometer with
Cary Win UV kinetic assay software. All kinetic experi-
ments were replicated at least in triplicate. The baseline
absorbance was measured using water prior to experimen-
tal measurements. Analysis of the obtained kinetic data was
performed using the Kaleidagraph software.
D -5′-dGMP at pH 7.0 and 40 °C were 5.02 ± 0.88
8
−2 −1
and 4.76 ± 0.07 M s , respectively. The k of H -
e
8
5′-dGMP and D -5′-dGMP at pH 7.0 and 40 °C were
8
−4
−1
−4 −1
3.13 (± 0.08) × 10
s
and 3.06 (± 0.03) × 10 s ,
respectively. These rate constants are comparable with
the previously found k and k values of 5′-dGMP at pH
s
e
−2 −1
8.3 and 40 °C, which were 8.1 ± 0.2 M s and 3.6
± 0.5) × 10 s , respectively [21]. The kinetic iso-
tope effect (KIE) of k which is the ratio of the k of H -
−4
−1
(
e
e
8
Results and discussion
5′-dGMP to the k of D -5′-dGMP was calculated to be
e 8
1
.2 ± 0.2. This small KIE value suggests that the depro-
IV
IV
Kinetic isotope effect of the reaction of 5′‑dGMP
tonation of H bound to Pt , [H -5′-dGMP–Pt ], is not
8
8
IV
with [Pt (dach)Cl ]
involved in the rate-determining step in the electron transfer
between G and Pt . This result is consistent with the DFT
4
IV
Figure 1 compares the kinetic curves of the reaction of H -
calculation which predicted that H deprotonates after two
electrons from 5′-dGMP are transferred to Pt , the rate-
8
8
IV
IV
5
′-dGMP and D -5′dGMP with [Pt (dach)Cl ] at pH 7.0
8
4
and 40 °C. Both reactions show the same kinetics trends.
determining step in the electron transfer (Scheme 1b) [19].
Fig. 1 Kinetic curves at a 360
1.6 (a)
(b)
.8
0
and b 390 nm of the reactions
IV
of [Pt (dach)Cl ] with H -
H (Exp)
H (Exp)
4
8
8
8
5
1
′-dGMP and D -5′-dGMP in
00 mM NaCl at 40 °C with a
8
H (Fit)
H (Fit)
8
8
starting pH 7.0 (t = 0). Fits to
1.2
D (Exp)
8
0.6
D (Exp)
both A3 and A curves give
8
60
390
D (Fit)
identical k and k values
s
e
8
D (Fit)
8
0
.8
.4
0
.4
.2
0
0
0
50
100
150
200
250
300
0
50
100 150 200 250 300
Time (min)
Time (min)
1
3

J Biol Inorg Chem (2015) 20:1327–1341
1331
IV
It is interesting to note that the KIE of k of fac-[Pt (dach)
α-phosphate in the ring structure, and the other one the
β-phosphate. At 24 h the peak at 6.33 ppm appears as a
triplet with no overlapping, suggesting that one cyclic
intermediate is more stable than the other and disappears
e
+
(
9-EtG)Cl ] (9-EtG = 9-ethylguanine) at high pH (pH 12)
3
is reported to be 7.2 ± 0.2 [16]. At high pH, the electron
−
transfer is initiated by OH in the solvent (inter molecular
nucleophile), while at pH 7, the electron transfer is initi-
faster. The peaks at 6.10 and 6.15 ppm arise from H of
1′
II
ated by 5′-phosphate (intramolecular nucleophile). These
the final products of 8-oxo-dGDP and Pt -bound-5′-dGDP,
respectively.
−
results indicate that OH is a more powerful nucleophile
than phosphate. However, the KIE of autoxidation of
Figure 3 shows an overlay of the time course of 31P-
NMR spectra of the reaction of 5′-dGDP (20 mM) with
III
−
Ru (NH ) (Ino)] assisted by O and OH is reported to be
3 5 2
[
IV
a secondary KIE of 1.6 [11]. It seems that different systems
have different KIEs, indicating that deprotonation occurs at
different stages, depending on the nucleophiles involved.
[Pt (dach)Cl ] (10 mM) in D O with 100 mM NaCl
4
2
and a starting pH of 8.6. At the beginning of the reaction
(t = 0 h) there are two doublets at −5.95 and −10.20 ppm
3
1
corresponding to the resonance of P nuclei of the β- and
α-phosphate groups of free 5′-dGDP, respectively [25]. The
IV
Reaction of 5′‑dGDP with [Pt (dach)Cl ]
4
3
1
P peak of the β-phosphate of free 5′-dGDP at −5.95 ppm
IV
The time course of the reaction of [Pt Cl (dach)]/5′-
shifts upfield over the course of the reaction to −9.0 ppm
4
1
dGDP (10/20 mM) was monitored by H-NMR spec-
after 48 h. This upfield chemical shift is due to the protona-
troscopy, and the results were in parallel with those for
tion of the phosphate group (pK = 6.8), as the pH of the
a
1
5
′-dGMP [14]. Figure 2 shows the H NMR spectra in the
reaction mixture decreases over time due to the production
of two protons in the two-electron redox reaction [13, 14].
It is known that both inorganic and nucleotide phosphates
shift upfield when phosphate is protonated [26, 27]. After
12 h of reaction two new doublets are formed at −11.3 and
−19.0 ppm and gradually disappear. These are assigned
to the (5′-P O-C8)-dGDP and (5′-P O-C8)-dGDP cyclic
area where the chemical shifts of H and H appear. When
8
1′
G is free, H resonates around 8.2 ppm (not shown). When
8
IV
II
G is bound to the electron-deficient Pt and Pt at the
N position, the chemical shifts of H change to 9.03 and
7
8
8
1
.32 ppm, respectively. The 9.03 ppm peak appears after
2 h and disappears after around 48 h, indicating that Pt -
IV
β
α
bound G is an intermediate. The 8.32 ppm peak appears
after 24 h and grows in intensity, indicating that the Pt -
intermediates, respectively. The upfield chemical shifts of
II
5 and 9 ppm of (5′-P O-C8)-dGDP and (5′-P O-C8)-dGDP,
β
α
bound G is a product.
respectively, compared to the free 5′-dGDP, are relatively
The H of free G resonates at 6.17 ppm. The two over-
close to that of 10 ppm upfield reported for the cyclic
1
′
3
1
lapping peaks around 6.33 ppm correspond to the sec-
ond intermediates and were assigned to the H1′ of the
cyclic (5′-PO-C8)-dGDP intermediate, one involving the
intermediate in 5′-dGMP. The P peaks of free 5′-dGMP
and (5′-PO-C8)-dGMP appear around 5.0 and −5.5 ppm,
respectively [14].
6.17
(a)
(b)
8.32
6.10
6.15
48 h
4
8 h
4 h
9.03
2
4 h
2 h
2
6.33
1
12 h
0
h
0
h
6.4
6.35
6.3
6.25
6.2
ppm
6.15
6.1
6.05
6
9.2
9
8.8
8.6
8.4
8.2
ppm
1
IV
Fig. 2 Time-dependent H NMR spectra of the reaction of 10 mM [Pt Cl (dach)] with 20 mM 5′-dGDP in 100 mM NaCl at 37 °C with a start-
4
ing pH 8.6 (t = 0)
1
3

1
332
J Biol Inorg Chem (2015) 20:1327–1341
+
Cl
α
H
2
N
N
Cl
Cl
Pt
H
2
β
9.0
O
N
-
10.20
N
NH
NH
-
H
O
P
O
P
N
2
4
3
2
8 h
6 h
4 h
6
6
6
5
2
10
HO
O
O
O
β
α
OH OH
OH
8
41.5
α
α
1.5 10
1 10
β
8
42.5
8
43.5
840.5
39.5
β
-
11.3
- 19.0
8
β
1
2 h
α
-
5.95
844.5
845.5
5
10
0
h
-5
-10
-15
-20
ppm
0
8
Fig. 3 Time-dependent 31P NMR spectra of the reaction of 10 mM
30
835
840
845
m/z
850
855
860
IV
[
Pt Cl (dach)] with 20 mM 5′-dGDP in 100 mM NaCl at 37 °C with
4
a starting pH 8.6 (t = 0)
IV
Fig. 4 Mass spectrum of fac-[Pt (dach)(5′-dGDP)Cl ] along with its
3
theoretical isotope patterns
7
Two more doublets, with low intensities, appear after
7 10
(1)
1
2 h of reaction with chemical shifts 0.5 ppm upfield from
16
8
-oxo-dGDP( O)
1
8
the peaks of α and β-phosphate of free 5′-dGDP. These
8-oxo-dGDP( O)
3
1
443.6
peaks probably correspond to the P nuclei of the neigh-
boring phosphate groups in the two cyclic intermediates,
which do not participate in the phosphodiester bond. This
assignment is supported by the fact that these low-intensity
peaks show up in the same timeframe with the other two
intermediate peaks, and have significantly smaller changes
in chemical shifts, as expected for nuclei that do not par-
ticipate directly in chemical changes in the molecule. It is
4
45.6
7
3.5 10
IV
also possible that these peaks correspond to the Pt –dGDP
0
intermediate.
438
441
444
m/z
447
450
To further verify the identity and presence of the PtIV–
dGDP intermediate, the above reaction mixture was ana-
lyzed by ESI-LC/MS. Figure 4 shows the mass spectrum of
Fig. 5 Mass spectra of 8-oxo-5′-dGDP produced from the reaction
mixture of 5′-dGDP (20 mM) with [Pt (dach)Cl ] (10 mM) in (a)
H O and (b) H O in 100 mM NaCl after 48 h at 37 °C with a start-
IV
IV
4
Pt –dGDP obtained after 16 h of reaction. The major peak
1
2
6
18
2
IV
8
41.5 (m/z) is close to the expected mass of Pt –dGDP
ing pH 8.6 (t = 0)
(
840.0 amu). The multiple surrounding peaks appear due to
the several platinum isotopes.
To confirm the 31P NMR results indicating that both
phosphate groups (α and β) participate in the intramolecu-
18
18
[8-oxo-dGDP( O)]. The mass of 8-oxo-5′-dGDP( O)
(445.6 m/z) is 2 amu higher than the mass of 8-oxo-5′-
1
6
lar nucleophilic attack to produce the cyclic phosphodiester
intermediate, the reactions of [Pt (dach)Cl ] with 5′-dGDP
were carried out in both normal H O and H O and their
products were analyzed by ESI-LC/MS. In particular, the
fragmentation mass spectra of 8-oxo-dGDP from both
reactions after 48 h were obtained and analyzed. Figure 5
shows the overlay of the mass spectra of 8-oxo-5′-dGDP
obtained from the reaction in H O (8-oxo-5′-dGDP( O))
and 8-oxo-5′-dGDP obtained from the reaction in H O
dGDP( O) indicating that the former compound con-
tains an O atom in its structure. The absence of a peak
at 443.6 (m/z) in the reaction in H O suggests that there
was not any significant contamination by H O. Further-
more, the mass of 5′-dGDP from a control experiment of
incubating 5′-dGDP in H O without [Pt (dach)Cl ] for
48 h did not increase 2 amu. This result confirmed that the
2 amu increase of 8-oxo-5′-dGDP( O) was not due to the
IV
18
4
16
2
18
2
18
2
16
2
18
2
IV
4
16
2
16
18
18
2
oxygen exchange in the phosphates of 5′-dGDP, but due
1
3

J Biol Inorg Chem (2015) 20:1327–1341
1333
7
2.5 10
16
[3]
18
8
-oxo-dGDP( O)
8
-oxo-dGDP( O)
O
N
H
7
7
7
6
[
3]
N
2
10
NH
NH2
O
8
N
[
1]
2
1
.5
2
1
0
7
7
7
(
a)
[3]
2
47.5
247.5
10
(
c)
(b)
1
.5 10
.5
10
O
HO Pβ
O
2
1
.5
2
1
0
6
6
6
107
106
1
5
O
P
α O
[
1]
167.5
2
49.5 [3']
O
10
1
67.5
OH
OH
.5
10
0
2
46
248
m/z
250
252
OH
[2]
1
5
10
10
6
5
1
69.5 [1']
1
5
10
10
0
[4] [5]
1
66
167
168
m/z
169
170
171
5
4
3
10
10
10
6
6
6
3
27.6
4
10
6
6
6
[4]
[
4]
2.5
1
0
6
6
179.5
[
2]
3
.2
10
327.6
[2]
[5]
2
10
345.6
345.6
2.4
10
179.5
2
1
10
10
6
6
[
1]
1.5
1
0 6
3
29.6 [4']
1
.6
8
1
0
6
5
[
5]
1
5
10
10
6
5
10
181.5 [2']
0
347.6 [5']
3
25.
5
327.3
m/z
329
330.8
0
1
78
179
180
m/z
181
182
183
0
3
45
346
m/z
347
348
349
0
1
50
200
250
m/z 300
350
400
450
IV
Fig. 6 Fragmentation mass spectra of 8-oxo-5′-dGDP obtained from the reaction mixture of 5′-dGDP (20 mM) with [Pt (dach)Cl ] (10 mM) in
4
16
18
a H O and b H O in 100 mM NaCl after 48 h at 37 °C with a starting pH 8.6 (t = 0). Inset structure of 5′-dGDP along with the bond break-
2
2
ing positions
18
2
IV
16
18
to the involvement of H O in the reaction of [Pt (dach)
8-oxo-5′-dGDP( O) and 8-oxo-5′-dGDP( OP ) are gen-
β
Cl ] and 5′-dGDP. Similarly, the absence of any peaks cor-
erated, while fragments [4′] and [5′] with m/z = 329.6
4
1
6
responding to a mass higher than 8-oxo-5′-dGDP( O) by
or 6 amu further indicates that oxygen in water did not
and 347.6, are generated, respectively, from 8-oxo-5′-
18
18
4
dGDP( OP ) or 8-oxo-5′-dGDP( OC ). These results
8
α
18
18
exchange in the phosphates of 5′-dGDP.
clearly indicate that O atom in 8-oxo-5′-dGDP( O) can be
Figure 6 displays the overlay of fragmentation mass
located at P , P , or C , indicating that both an intermolecu-
β
α
8
16
18
spectra of 8-oxo-5′-dGDP( O) and 8-oxo-5′-dGDP( O)
from m/z 150 to 450. The fragmentation mass spectrum
lar and an intramolecular nucleophilic attack to the Pt(IV)–
G intermediate is possible. The possible structures of all the
fragments along with their m/z are summarized in Table 1.
16
of 8-oxo-5′-dGDP(O ) shows five major peaks arising
1
31
from fragments [1], [2], [3], [4], and [5]. All of these peaks
Together the H and P NMR, and ESI–MS/MS stud-
ies led us to propose the mechanism in Scheme 3. The
18
appear in 8-oxo-5′-dGDP( O) as well, accompanied by
the corresponding weaker peaks at 2 amu higher arising
from fragments [1′], [2′], [3′], [4′], and [5′]. The structure
of 8-oxo-5′-dGDP, the possible bond breakages, and the
five fragments are shown in the inset in Fig. 6. When the
bonds break at (a) and (c) positions, the fragments [1] and
IV
[Pt (dach)Cl ] complex binds to the N and a two-electron
4
7
IV
transfer from G to Pt occurs. The two-electron transfer
step is initiated by an intramolecular nucleophilic attack by
either the α or β phosphate groups followed by cyclization
and hydrolysis, or by an intermolecular nucleophilic attack
−
[
2] are produced. The fragment [1] with m/z = 167.5 due to
by OH .
16
guanine can be generated from both 8-oxo-5′-dGDP( O)
18
18
IV
and 8-oxo-5′-dGDP( OP ) or 8-oxo-5′-dGDP( OP ), but
Reaction of 5′‑dGTP with [Pt (dach)Cl ]
β
α
4
[
5
1′] with m/z = 169.5 can be generated only from 8-oxo-
18
IV
′-dGDP( OC ). The fragment [2] with m/z = 179.5 due
The time course of the reaction of [Pt Cl (dach)]/5′-dGTP
8
4
1
to ribose-phosphate can be generated from both 8-oxo-
5
(10/20 mM) was monitored by H-NMR spectroscopy, and
the results were in parallel with those for 5′-dGMP [14]
and 5′-dGDP described above. Figure 7 shows the H NMR
spectra in the area where the chemical shifts of H and H
16
18
′-dGDP( O) and 8-oxo-5′-dGDP( OP ) or, 8-oxo-5′-
β
18
1
dGDP( OC ), but [2′] with m/z = 181.5 from 8-oxo-5′-
8
18
dGDP( OP ). When the bond breaks at (b), the fragment [3]
α
8
1′
IV
with m/z = 247.5 due to guanosine is generated from both
of guanosine (G) occurs. The H peak due to Pt -bound
8
1
6
18
II
8
5
-oxo-5′-dGDP( O) and 8-oxo-5′-dGDP( OP ) or 8-oxo-
dGTP appears at 9.00 ppm, and the Pt -bound dGTP peak
β
18
′-dGDP( OP ), but [3′] with m/z = 249.5 from 8-oxo-5′-
at 8.35 ppm. The H peaks due to the cyclic (5′-PO-C8)-
dGTP intermediate, 8-oxo-dGTP, and Pt -bound dGTP
α
1′
18
II
dGDP( OC ). When the bond breaks at (c), the fragments
8
[4] and [5] with m/z = 327.6 and 345.6, respectively, from
appear at 6.29, 6.07, and 6.13 ppm, respectively.
1
3

1
334
J Biol Inorg Chem (2015) 20:1327–1341
Table 1 Possible structures of fragments of 8-oxo-5′-dGDP corresponding to the peaks in Fig. 6
H2O16
H2O18
[1]
[1]
[1']
m/z: 169.5
m/z: 167.5
m/z: 167.5
[2]
[2]
[2']
m/z: 181.5
m/z: 179.5
m/z: 179.5
[
3]
4]
[3]
[3']
m/z: 249.5
m/z: 247.5
m/z: 327.6
m/z: 247.5
[
[4]
[4']
[4']
m/z: 329.6
m/z: 327.6
m/z: 329.6
[
5]
6]
[5]
[6]
[5']
[5']
m/z: 345.6
m/z: 363.6
m/z: 345.6
m/z: 347.6
m/z: 347.6
[
[6']
[6']
m/z: 363.6
m/z: 365.6
m/z: 365.6
[
7]
[7']
[7']
m/z: 407.6
m/z: 409.6
m/z: 409.6
[7]
m/z: 407.6
[
7']
m/z: 409.6
[8]
[8']
m/z: 425.6
m/z: 427.6
m/z: 425.6
[8]
[8']
[8']
m/z: 427.6
m/z: 427.6
Figure 8 shows an overlay of the 31P-NMR spectra of
−5.5 ppm correspond to α and γ phosphate groups of free
5′-dGTP, respectively, and the triplet at 21.1 ppm to the
P nuclei of the β phosphate. This assignment is in agree-
IV
the reaction of 5′-dGTP (20 mM) with [Pt (dach)Cl ]
4
3
1
(
10 mM) in D O with 100 mM NaCl and a starting pH of
2
8
.6, without the presence of buffer, at different times. At
ment with previously published studies [25]. The chemi-
the beginning of the reaction, the doublets at −10.6 and
cal shift corresponding to the γ-phosphate of free 5′-dGTP
1
3

J Biol Inorg Chem (2015) 20:1327–1341
1335
O
N
PtIV
N
N
NH
NH2
H
O
O
HO Pβ O Pα O
O
O
O
OH
-
OH attack
α attack
β attack
PtIV
O
N
PtIV
N
O
PtIV
N
O
N
OH
O
N
NH
NH2
NH
NH2
NH
H
H
H
N
N
O
P
N
NH2
O
O
O
Pβ
O
O
P
N
HO P O Pα O
HO
O
O
HO
P
O
O
O
O
O
O
O
O
O
O
OH
OH
OH
-
H
PtIV
O
PtIV
O
N
PtIV
O
N
O
P
O
O
N
N
NH
NH2
NH
N
OH
O
NH
HO
O
Pα
O
P
β
O
N
O
O
N
O
H
N
NH2
OH H
O
O
N
NH2
O
P
O-
HO
P
O
P
O
O
O
O
O
O
OH
- PtII
OH
OH
- PtII
PtII
- H
-
- H
O
O
N
O
N
NH
NH2
O
P
O
O
N
N
O
OH
NH
NH2
N
N
NH
NH2
HO
O
P
α
O
Pβ
N
H2O
O
O
O
N
O
OH
O
O
N
HO
P
O
P
O
O
P
O
O
O
O
O
O
OH
H2O
OH
OH
- H
-
H
O
-
H
O
N
O
N
H
H
N
H
N
NH
NH2
N
NH
NH2
NH
NH2
O
O
O
N
O
O
N
N
N
O
O
O
O
HO
P
O
P
O
O
HO
P
O
P
α
O
HO Pβ O
P
O
O
O
O
O
O
OH
OH
O
OH
OH
OH
IV
IV
II
II
Scheme 3 Proposed mechanism for the oxidation of 5′-dGDP by [Pt (dach)Cl ] (represented as Pt ). [Pt (dach)Cl ] is represented as Pt . The
[
4
2
II
II
Pt (dach)Cl ] complex reacts further with excess 5′-dGDP and produces [Pt (dach)(5′-dGDP) ]
2
2
shifts upfield over the course of the reaction to −8.1 ppm
after 36 h. This upfield chemical shift is due to the proto-
nation of the phosphate group (pKa ~ 6.7), as the pH of
the reaction mixture decreases over time [26, 27]. After
0.5 ppm upfield from the corresponding peaks of α, β, and
γ-phosphate of free 5′-dGTP. These peaks probably corre-
3
1
spond to the P nuclei of the neighboring phosphate groups
in the two cyclic intermediates, which do not participate in
the phosphodiester bond, or due to the peaks arising from
1
2 h of reaction, two new doublets are formed at −19.7
IV
and −11.35 ppm, and a triplet at −23.7 ppm, all of which
the binding of [Pt (dach)Cl ] to the N of 5′-dGTP.
4
7
IV
gradually disappear over time. They were assigned to the
To further verify the identity and presence of the Pt –
dGTP intermediate, the above reaction mixture was ana-
lyzed by ESI-LC/MS. Figure 9 shows the mass spectrum of
3
1
P nuclei of the participating in the phosphodiester bond in
the cyclic intermediates (5′-P O-C8)-dGTP, (5′-P O-C8)-
γ
α
IV
dGTP, and (5′-P O-C8)-dGTP, respectively. After 12 h of
Pt –dGTP obtained after 16 h of reaction. The major peak
β
IV
reaction, low-intensity peaks appear with chemical shifts of
921.5 (m/z) is close to the expected mass of Pt –dGTP
1
3

1
336
J Biol Inorg Chem (2015) 20:1327–1341
8
.35 ppm
4
8 h
6
.13 6.07
1
1
48 h
2
4 h
2
4 h
8
.90 ppm
6
.29
6
h
6
h
0
0
6
.4
6.35
6.3
6.25
6.2
ppm
6.15
6.1
6.05
6
9
.2
9
8.8
8.6
8.4
8.2
ppm
1
IV
Fig. 7 Time-dependent H NMR spectra of the reaction of 10 mM [Pt Cl (dach)] with 20 mM 5′-dGTP in 100 mM NaCl at 37 °C with a start-
4
ing pH 8.6 (t = 0)
(b)
-
11.35
α
10.9
(
a)
β
-
γ
-
22.0
-
11.1 -11.2 -11.3 -11.4 -11.5 -11.6
-
8.1
-
11.35
-23.7
-
19.7
36 h
(c)
-19.7
β
γ
24 h
β
-
19
-19.2 -19.4 -19.6 -19.8
-20
γ
γ
12 h
(d)
-23.7
-
10.6
α
-21.1
-5.5
0
h
-
5
-10
-15
-20
-25
ppm
-22.5
-23
-23.5
-24
-24.5
-25
Fig. 8 Time-dependent 31P NMR spectra of the reaction of 10 mM [Pt Cl (dach)] with 20 mM 5′-dGTP in 100 mM NaCl at 37 °C with a start-
IV
4
ri na gn g pe Ha f8t e. 6r 2( t4=h 0). a −5 to −25 ppm range; b −11.1 to −11.6 ppm range after 24 h; c −19 to −20 ppm range after 24 h; d −22 to −23.5 ppm
1
3

J Biol Inorg Chem (2015) 20:1327–1341
1337
+
106
Cl
5
H
N
2
Cl
Cl
Pt
N
-oxo-dGTP(16O)
H
2
8
O
N
N
NH
NH
6
H
523.5
2
10
O
P
O
P
O
P
N
2
-oxo-dGTP (18O)
8
HO
O
O
O
O
9
21.5
OH OH OH
.5 106
2
525.5
OH
6
6
5
1
.5 10
922.5
9
23.5
9
20.5
0
5
1
5
10
10
20
522
524
m/z
526
528
9
19.5
9
24.4
925.4
Fig. 10 Mass spectra of 8-oxo-5′-dGTP produced from the reac-
tion mixture of 5′-dGTP (20 mM) with [Pt (dach)Cl ] (10 mM) in a
H O and b H O in 100 mM NaCl after 48 h at 37 °C with a starting
IV
4
1
2
6
18
2
pH 8.6 (t = 0)
0
9
10
915
920
925
930
935
940
m/z
Figure 11 displays the overlay of fragmentation mass
spectra of 8-oxo-5′-dGTP( O) and 8-oxo-5′-dGTP( O)
1
6
18
IV
Fig. 9 Mass spectrum of fac-[Pt (dach)(5′-dGTP)Cl ] along with its
from m/z 150 to 450. The fragmentation mass spectrum of
3
theoretical isotope pattern
16
8
-oxo-5′-dGTP( O) shows eight major peaks arising from
fragments [1], [2], [3], [4], [5], [6], [7], and [8]. All these
1
8
peaks appear in 8-oxo-dGTP( O) as well, accompanied
by the corresponding weaker peaks at 2 amu higher arising
from fragments [1′], [2′], [3′], [4′], [5′], [6′], [7′], and [8′].
The structure of 8-oxo-5′-dGTP, possible bond breakages,
and eight fragments are shown in the inset in Fig. 11. The
fragments [1] to [5] were generated exactly the same way
as for the 5′-dGDP. The fragments [7] and [8] indicate that
(
920.0 amu). The multiple surrounding peaks appear due to
the several platinum isotopes.
To confirm the 31P NMR results indicating that all three
phosphate groups (α, β, and γ) participate in the intramo-
lecular nucleophilic attack to produce the cyclic phospho-
diester intermediate, the reactions of [Pt (dach)Cl ] with
5
and the results were analyzed by ESI-LC/MS. In particular,
the fragmentation mass spectra of 8-oxo-dGTP from both
reactions after 48 h were obtained and analyzed.
IV
4
16
2
18
2
18
,
′-dGTP were carried out in both normal H O and H O
an O atom should reside at Pγ and the fragment [7′] and
1
8
[8′] should contain O at Pβ, Pα, or C8.
1
31
Together, the H and P NMR, and ESI–MS/MS studies
IV
led us to propose the mechanism in Scheme 4. The [Pt (dach)
Figure 10 shows the mass spectra of 8-oxo-5′-dGTP
Cl4] complex binds to the N7, followed by a two-electron
transfer from G to Pt . The two-electron transfer step is initi-
1
6
16
IV
obtained from the reaction in H O [8-oxo-5′-dGTP( O)]
2
1
8
and 8-oxo-5′-dGTP obtained from the reaction in H O
8-oxo-5′-dGTP( O)]. The mass of 8-oxo-5′-dGTP( O)
525.5 m/z) is higher than the mass of 8-oxo-5′-dGTP( O)
ated by an intramolecular nucleophilic attack by either the α,
β, or γ phosphate group followed by cyclization and hydroly-
2
1
8
18
[
(
1
6
−
sis, or by an intermolecular nucleophilic attack by OH .
by 2 amu indicating that the former compound contains an
1
8
O atom in its structure. The absence of a peak at 523.5
Kinetics of the reaction of 5′‑dGMP, 5′‑dGDP
and 5′‑dGTP with [Pt (dach)Cl4]
1
8
IV
(m/z) in the reaction in H O suggests that there was not
2
1
6
any significant contamination by H O. Furthermore, the
2
mass of 5′-dGTP from a control experiment of incubat-
Figure 12 shows the changes in the electronic absorption at
360 and 390 nm over time, for the reaction of [Pt (dach)
1
8
IV
IV
ing 5′-dGTP in H O without [Pt (dach)Cl ] for 48 h did
2
4
not increase 2 amu. This result confirmed that the 2 amu
Cl4] with 5′-dGMP, 5′-dGDP, and 5′-dGTP at 35 °C and at
a starting pH of 8.6. The changes in the electronic absorp-
tion at 360 nm (A360) indicate the depletion of the unbound
1
8
increase of 8-oxo-5′-dGTP( O) was not due to the oxy-
gen exchange in the phosphates of 5′-dGTP, but due to the
18
2
IV
IV
involvement of H O in the reaction of [Pt (dach)Cl ] and
[Pt (dach)Cl4], while those at 390 nm (A390) the formation
4
IV
5
′-dGTP. Similarly, the absence of any peaks correspond-
and depletion of the Pt –G intermediate. The change in
1
6
ing to a mass higher than 8-oxo-dGTP( O) by 4 or 6 amu
further indicates that oxygen in water did not exchange in
the phosphates of 5′-dGTP.
A360 of 5′-dGMP is significantly larger than that of 5′-dGDP,
which in turn is higher than that of 5′-dGTP. This trend
IV
suggests that [Pt (dach)Cl4] reacts faster with 5′-dGMP,
1
3

1
338
J Biol Inorg Chem (2015) 20:1327–1341
6
6
6
6
5
Fig. 11 Fragmentation mass
spectra of 8-oxo-5′-dGTP
obtained from the reaction
2
.5 10
2
.5 10
6
16
[3]
[
3]
425.6
425.6
427.6
[8]
8
-oxo-dGTP( O)
[8]
O
N
1.25 10
6
H
N
[8']
mixture of 5′-dGTP (20 mM)
18
NH
NH2
O
8
2 10
8
-oxo-dGTP( O)
N
424
0
425
426 m/z
427
428
429
IV
with [Pt (dach)Cl ] (10 mM) in
[1]
4
(a)
16
18
a H O and b H O in 100 mM
(d)
(c)
(b)
1.2 10
6
2
2
O
HO Pγ
O
Pβ
OH
O
Pα
OH
4
07.6
NaCl after 48 h at 37 °C with a
starting pH 8.6 (t = 0)
O
O
O
O
[7]
5
409.6
OH
6
10
407.6
1.5 10
OH
2]
[7']
[
006
407
408 m/z
409
410
411
[4] [5] [6]
4
[
7] [8]
6
4
2
10
10
10
5
[
6]
363.6
363.6
1
5
10
10
5
5
365.6
[7]
[
6']
3
060
362
m/z 364
366
368
[
6]
[2]
[4]
[
5]
[1]
0
1
50
225
m/z 300
375
450
O
PtIV
N
N
NH
NH2
H
O
HO Pγ
O
O
O
N
O
Pβ O P α O
O
-
α attack
O
O
OH attack
OH
β attack
O
γ attack
PtIV
N
O
N
PtIV
N
PtIV
N
O
N
HO
PtIV
N
O
NH
NH2
NH
NH2
H
H
NH
NH2
NH
NH2
O
P
O
P
O
N
N
H
H
O
O
Pβ
O
O
N
N
O
HO
O
O
P
α
O
O
O
P
O
O
P
O
O
O
P
N
N
O
HO
P
O
O
P
O
O
O
P
O
O
O
HO
P
γ
O
O
O
O
HO
P
O
O
O
O
O
OH
O
O
O
O
OH
OH
OH
-
H
PtIV
N
O
PtIV
N
O
N
O
O
PtIV
N
O
N
O
PtIV
N
O
O
P
O
P
O
Pα O
O
P
HO
P
O
O
Pβ
O
H
NH
NH2
O
P
O
NH
NH2
HO
O
O
NH
NH2
γ
NH
HO
O
O
OH
O
O
O
N
H
O
N
N
O
O
O
N
O
H
O
P
N
N
NH2
O
O
HO
P
O
P
O
P
O
O
O
O
O
P
O
O
O
O
O
OH
OH
OH
OH
- PtII - H
-
PtII - H
PtII
- H
- PtII
-
O
H2O
O
N
O
N
O
P
O
H O
O
O
O
2
O
O
N
N
N
N
N
O
O
P
Pβ
O
O
N
N
NH
NH2
HO
NH
NH2
O
P
NH
NH2
HO
P
O
P
O P O
α
O
P
γ
O
HO
N
O
NH
O
O
O
N
O
O
P
O
P
O
P
O
OH
O
O
N
HO
O
O
O
O
O
NH2
O
O
P
O
O
O
O
O
OH
OH
O
OH
H2O
OH
-
H
-
H
- H
O
- H
O
O
H
N
H
O
O
N
N
H
N
H
NH
NH2
NH
NH2
N
N
O
O
NH
NH
NH2
O
O
O
P
O
P
O
N
N
O
O
O
N
N
N
NH2
O
P
O
O
P
O
O
P
O
N
O
HO Pγ
O
P
O
O
P
O
HO
O
O
Pα
OH
O
HO
P
O
Pβ
O
P
O
O
O
O
O
HO
O
O
O
O
O
O
O
O
OH
O
OH
OH
OH
OH
OH
IV
IV
II
II
Scheme 4 Proposed mechanism for the oxidation of 5′-dGTP by [Pt (dach)Cl ] (represented as Pt ). [Pt (dach)Cl ] is represented as Pt . The
[
4
2
II
II
Pt (dach)Cl ] complex reacts further with excess 5′-dGTP and produces [Pt (dach)(5′-dGTP) ]
2
2
IV
followed by 5′-dGDP, and the slowest with 5′-dGTP. The
change in A390 follows the exact same trend, with the maxi-
mum peak for 5′-dGMP (3.8 h) appearing earlier than that of
that the formation of an intermediate, Pt –G, and its deple-
tion by electron transfer initiated by intra- or intermolecu-
lar nucleophilic attack, occurs faster for 5′-dGMP, followed
by 5′-dGDP, and last by 5′-dGTP. In our previous kinetic
5′-dGDP (9.3 h), and of 5′-dGTP (14 h). This trend indicates
1
3

J Biol Inorg Chem (2015) 20:1327–1341
1339
Fig. 12 Changes in the elec-
tronic absorption at a 360 nm
and b 390 nm over time, for
1.3
0.8
0.6
0.4
0.2
(
a)
(b)
5'-dGTP
5
'-dGTP
IV
the reaction of [Pt (dach)
Cl ] (5 mM) with 5′-dGMP
4
1.2
(
50 mM), 5′-dGDP (50 mM)
and 5′-dGTP (50 mM) in
00 mM NaCl at 35 °C with
5'-dGDP
5
'-dGDP
1
[
II
Pt (dach)Cl ] (0.02 mM) at a
2
5
'-dGMP
starting pH of 8.6 (t = 0)
1.1
1
5'-dGMP
0
5
10
Time (h)
15
20
0
5
10
Time (h)
15
20
Fig. 13 Change in the elec-
tronic absorption at a 360 nm
1.2
0.75
(
a)
(b)
Without PtII
With PtII
and b 390 nm over time, for
IV
the reaction of [Pt (dach)Cl ]
4
(
5 mM) with 5′-dGTP (50 mM)
II
with and without [Pt (dach)Cl ]
2
(
0.02 mM) in NaCl (100 mM)
at 45 °C and a starting pH of 8.6
t = 0)
0.5
(
With PtII
Without PtII
1
0.25
0
2
4
6
8
0
2
4
6
8
Time (h)
Time (h)
Fig. 14 Change in the elec-
tronic absorption at a 360 nm
1.1
1
0.6
(
a)
(b)
and b 390 nm over time, for
IV
the reaction of [Pt (dach)Cl ]
4
0.5
(
5 mM) with 5′-dGTP (50 mM)
II
with and without [Pt (dach)Cl ]
2
(
0.02 mM) in NaCl (100 mM)
at 45 °C and a starting pH of 8.6
t = 0) and 6.0 (t = 0
pH 6.0
pH 8.6
0.4
pH 6.0
pH 8.6
(
0
.9
.8
0
.3
.2
0
0
0
2
4
6
8
0
3
6
9
12
Time (h)
Time (h)
study on the PtIV reaction with 3′-dGMP and 5′-dGMP, we
found that 3′-dGMP reacts faster than 5′-dGMP due to its
low steric hindrance and low rotational barrier of 5′-group
electron transfer reactions of 5′- and 3′-dGMP indicated
that the steric hindrance and the ability of the 5′-group to
form H-bonds are the main factors influencing substitution,
while the rotational barrier determines the electron transfer
[21]. The activation parameters for the substitution and
1
3

1
340
J Biol Inorg Chem (2015) 20:1327–1341
IV
step. Increasing the length of the phosphate chains should
increase steric hindrance and rotational barrier, and decrease
the overall reaction rate.
inner sphere electron transfer from G to Pt . The redox
II
reaction was catalyzed by Pt and occurred faster
at higher pH. The electron transfer was initiated by
either an intramolecular nucleophilic attack by any
of the phosphate groups or an intermolecular nucleo-
In previous studies, we also showed that the presence
II
[
[
Pt (dach)Cl ] can catalyze the substitution reaction of
2
IV
−
Pt (dach)Cl ] to 5′-dGMP, as shown in Scheme 2 [21].
philic attack by free OH in the solution. The rates of
4
II
Figure 13 shows the effect of the [Pt Cl (dach)] on the
reactions for the three nucleotides follow the order:
5′-dGMP > 5′-dGDP > 5′-dGTP, indicating that the bulk-
ier the phosphate group, the slower the reactions, due to
the larger steric hindrance and rotational barrier of the
phosphate groups.
2
changes in A and A390 for the reaction of 5′-dGTP with
3
60
IV
[
Pt (dach)Cl ] at 45 °C and at a starting pH of 8.6. The
4
II
presence of [Pt (dach)Cl ] shortened the induction time
Fig. 13a) and produced the [Pt –G] intermediate at an
2
IV
(
earlier time than when the reaction was run without the cat-
Acknowledgments This work was partially supported by the
National Science Foundation (CHE-08480720). The mass spectrom-
eter used in this study was purchased with funds from the NSF (CHE-
0520708). IK, SMFC, and GJKB are grateful to the Easton, Gleason,
and DeWitt families, respectively, for their summer research funds
at Middlebury College. We thank Mr. Bruce O’Rourke (MS) of the
Department of Chemistry Mass Spectrometry Facility at the Univer-
sity of Vermont for confirming our ESI–MS/MS spectral results.
alyst (Fig. 13b). Similar results were obtained for 5′-dGDP
II
(
not shown). Therefore, we concluded that [Pt (dach)
Cl ] acts as a catalyst for the substitution reaction for both
2
5
′-dGDP and 5′-dGTP.
Figure 14 shows the change of A360 and A390 for the reac-
IV
tion of 5′-dGDP with [Pt (dach)Cl ] at 45 °C and at two dif-
4
ferent starting pH values of 6.0 and 8.6. The changes of both
wavelengths over time show that a decrease in pH from 8.6
to 6.0 significantly slows down the reaction. A decrease in
pH leads to the protonation of the β-phosphate of 5′-dGDP,
References
decreasing its ability to form H-bonds with the amine groups
1. Schumacker PT (2015) Cancer Cell 27:156–157
IV
2
.
Klaunig JE, Kamendulis LM, Hocevar BA (2009) Toxicol Pathol
8:96–109
of [Pt (dach)Cl ] and thus decreasing the rate of substitu-
4
3
tion reaction, as well as decreasing the nucleophilicity of
the β-phosphate group and thus slowing the electron trans-
fer step. Therefore, lowering the pH decreases the overall
reaction rate. However, another important aspect is the role
3
4
.
.
Balaban RS, Nemoto S, Finkel T (2005) Cell 120:483–495
Mangialasche F, Polidori MC, Monastero R, Ercolani S, Camarda
C, Cecchetti R, Mecocci P (2009) Ageing Res Rev 8:285–305
Abdel Moneim AE (2015) Curr Alzheimer Res 12: 335-349
5
.
−
6. Aslan M, Ozben T (2004) Curr Alzheimer Res 1:111–119
of the intermolecular attack by free OH groups. A decrease
7
8
.
.
Steenken S, Jovanovic SV (1997) J Am Chem Soc 119:617–618
MullerJG, KayserLA, PaikoffSJ, DuarteV,TangN, PerezRJ, Rotica
SE, Burrows CJ (1999) Coord Chem Rev 185–186:761–774
. Burrows CJ, Muller JG (1998) Chem Rev 98:1109–1151
of the pH from 8.6 to 6.0 lowers the concentration of free
−
OH in the solution, decreasing the rate of intermolecular
IV
9
nucleophilic attack to C [G–Pt ]. Overall, the results from
8
1
1
0. Clarke MJ, Morrissey PE (1984) Inorg Chim Acta 80:L69–70
1. Gariepy KC, Curtin MA, Clarke MJ (1989) J Am Chem Soc
our kinetic study provide a qualitative framework to inves-
tigate in detail the factors that influence the reaction of
1
11:4947–4952
IV
[
Pt (dach)Cl ] with 5′-dGDP and 5′-dGTP. The fact that the
12. Rodriguez-Bailey VM, LaChance-Galang KJ, Doan PE, Clarke
MJ (1997) Inorg Chem 36:1873–1883
13. Choi S, Cooley RB, Voutchkova A, Leung CH, Vastag L,
Knowles DE (2005) J Am Chem Soc 127:1773–1781
4
mechanisms for the reactions of 5′-dGDP and 5′-dGTP with
IV
[
Pt (dach)Cl ] contain a greater number of intramolecular
4
nucleophiles and cyclic phosphodiester intermediates, as
well as a more significant role for the intermolecular nucleo-
philes than that of 5′-dGMP, will greatly increase the com-
plexity of the quantitative kinetic studies.
1
1
4. Choi S, Cooley RB, Hakemian AS, Larrabee YC, Bunt RC,
Maupaus SD, Muller JG, Burrows CJ (2004) J Am Chem Soc
1
26:591–598
5. Choi S, Personick ML, Bogart JA, Ryu DW, Redman RM, Lar-
yea-Walker E (2011) Dalton Trans 40:2888–2897. doi:10.1039/
c0dt00822b
1
1
6. Choi S, Ryu DW, DellaRocca JG, Wolf MW, Bogart JA (2011)
Inorg Chem 50:6567–6574. doi:10.1021/ic2003518
7. Wolf MW, Choi S (2012) J Biol Inorg Chem 17:1283–1291.
doi:10.1007/s00775-012-0942-8
Conclusion
IV
The deprotonation of H -5′-dGMP bound to Pt (H -G–
8
8
18. Noszal B, Scheller-Krattiger V, Martin RB (1982) J Am Chem
Soc 104:1078–1081
1
IV
Pt ) is not involved in the rate-determining step in the
IV
IV
9. Ariafard A, Tabatabaie ES, Aghmasheh S, Najaflo S, Yates B
2012) Inorg Chem 51:8002–8013. doi:10.1021/ic300038m
electron transfer between G and Pt . The [Pt (dach)
(
Cl ] complex oxidized 5′-dGDP and 5′-dGTP to 8-oxo-
4
2
0. Chang R The chemical educator, vol. 2, No. 3: Springer: New
York, 1997; pp 1–3
5
′-dGDP and 8-oxo-5′-dGTP, respectively, by the same
mechanism and kinetics as in the case of 5′-dGMP. The
21. Choi S, Vastag L, Leung CH, Beard AM, Knowles DE, Larrabee
JA (2006) Inorg Chem 45:10108–10114
IV
Pt complex binds to N followed by a two-electron
7
1
3

J Biol Inorg Chem (2015) 20:1327–1341
1341
2
2. Nakabeppu Y, Tsuchimoto D, Furuichi M, Sakumi K (2004) Free
Radic Res 38:423–428
26. Bose RN, Fonkeng BS, Moghaddas S, Stroup D (1998) Nucleic
Acids Res 26:1588–1596
2
3. Schweizer MP, Chan SI, Helmkamp GK, Ts’O POP (1964) J Am
Chem Soc 86:696–700
27. Robitaille PML, Robitaille PA, Brown GG, Brown GG Jr (1991)
J Magn Reson 92:73–84
2
2
4. Kuzmic P (1996) Anal Biochem 237:260–273
5. Reily MD, Hambley TW, Marzilli LG (1988) J Am Chem Soc
1
10:2999–3006
1
3