Relative rates of reaction of Pt(en)Cl (NH2R)+ with guanosine monophosphate as a function of amino group substituent: Toward efficient labeling of DNA for TEM imaging

Article abstract of DOI:10.1016/j.jorganchem.2012.11.003

In an attempt to understand the factors that govern the rates of reaction of the complexes [Pt(en)Cl(NH₂R)]⁺NO₃ ^- (en = ethylene diamine) with guanosine monophosphate (dGMP) a series of amine complexes, where RC

Full text of DOI:10.1016/j.jorganchem.2012.11.003

Journal of Organometallic Chemistry 734 (2013) 53e60
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Relative rates of reaction of Pt(en)Cl (NH2R)þ with guanosine monophosphate as
a function of amino group substituent: Toward efficient labeling of DNA for TEM
imaging
,
b
c
Rakesh Kumar a, Edward Rosenberg a , Miriam Inbar Feske , Antonio G. DiPasquale
*
a Department of Chemistry, University of Montana, Missoula, MT 59812, USA
b Latham and Watkins, Menlo Park, CA 94025, USA
c Department of Chemistry, University of California, Berkeley, CA 94720, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
In an attempt to understand the factors that govern the rates of reaction of the complexes [Pt(en)
Cl(NH2R)]þNO3ꢀ (en ¼ ethylene diamine) with guanosine monophosphate (dGMP) a series of amine
complexes, where R]C8H9NO2 (benzo[d][1,3]dioxol-5-ylmethanamine) (1), C8H11N (phenethylamine)
(2), C7H9N (benzylamine) (3), C6H7N (aniline) (4), C6H6IN (p-iodo-aniline) (5) C3H9NO (2-methoxy-
ethylamine) (6) and C6H13N (cyclohexylamine) (7), were synthesized and their reactions with deoxy-
guanosine monophosphate (dGMP) were followed by 1H NMR. Compound 1 was initially chosen because
it showed significant water solubility. Compound 1 reacted quantitatively but slowly with dGMP and
a subsequent Transmission Electron Microscopy (TEM) study of the binding 1 to a GATC DNA repeat gave
a TEM micrograph that showed selective labeling of DNA at guanine, using a technique that allowed the
laying down of a straight single strand of DNA on a carbon platform. The TEM suggested a possible side
reaction with adenine and so a study of the reaction of 1 with adenine was performed and showed slow
and what appeared to be non-specific binding to deoxyadenosine monophosphate (dAMP). The reactions
of compounds 2e7 with dGMP were then studied by 1H NMR and it was found that 2 reacted much faster
than 1 with dGMP while the remaining complexes reacted more slowly. No reaction of 2 with dAMP was
observed in the same time frame. The ultimate goal of the project was to bind a third row transition
metal cluster to guanine and given the effective binding of 1 to DNA the synthesis of the complex
[Os3(CO)11PPh2(CH2)2NH2(en)PtCl]NO3 (9) is also reported that contains Pt as a linker to label guanine.
The synthesis was performed by reacting Os3(CO)10(CH3CN)2 with Ph2PCH2CH2NH2 which gave an h2
chelate complex Os3(CO)10PPh2(CH2)2NH2 (8). Complex 8 was reacted with [Pt(en)Cl(DMF)]NO3 in a CO
atmosphere to give 9. 1H and 195Pt NMR indicate formation of an adduct with dGMP but too slowly to be
of use in labeling DNA. The solid-state structure of 8 is also reported.
Received 28 September 2012
Received in revised form
30 October 2012
Accepted 1 November 2012
Keywords:
Platinum complexes
DNA
TEM
dGMP
Osmium clusters
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
TEM. DNA has only light element (C,H,N,O,P), that are inherently
transparent to TEM. Heavy metal (Z > 70) staining is well known to
Beer and Moudrianakis (1962) first suggested that it may be
possible to sequence DNA with the aid of the electron microscope
[1]. Transmission Electron Microscopy (TEM) has the potential to
bring the base pair reading length up to 105 base-pairs per minute if
the bases could be specifically labeled with heavy atoms. Long
reading length enables detection of long repeating patterns in the
genetic code and facilitates high-speed sequencing. Hence, the cost
of DNA sequencing can be reduced significantly with the use of
increase image contrast in TEM, and heavy metal salts have long
been used in TEM to render nucleic acids visible [1e4]. The relative
number of electrons that are scattered to a detector is approxi-
mately proportional to Z1.5. The inherent drawback in using metal
salts’ staining is the lack of specificity to a nucleotide base owing to
non-specific interactions. Therefore, covalent adduct formation
between a DNA base and a metal complex is necessary for deter-
mination of the position of a specific base in a DNA molecule, Thus,
functionalized heavy metal complexes capable of base-specific
labeling of DNA can provide a tool for DNA sequencing. Further-
more, the labels need to have high reactivity to attach themselves
specifically to one of the four bases, GATC, along with very high
* Corresponding author.
E-mail address: edward.rosenberg@mso.umt.edu (E. Rosenberg).
0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2012.11.003

54
R. Kumar et al. / Journal of Organometallic Chemistry 734 (2013) 53e60
specificity. The key to getting sequence information from the heavy
atom labeling is the ability to stretch a single strand of DNA onto
a platform where multiple images of the strands can be added
together to give the exact sites of heavy atom labeling. Recently,
such a technique has been developed, by Halcyon Molecular and
provided the opportunity to develop heavy atom labels for DNA [5].
We have been studying the interaction of benzoheterocycle
complexes of triosmium carbonyl clusters for the last ten years [6e
8]. These studies revealed some of the structural requirements for
selective binding of modified triosmium clusters with guanosine
monophosphate (dGMP). Selectivity for guanine over the other
bases was also observed [7,8]. However, subsequent attempts to
obtain a straight and stretched single stranded DNA containing
a covalently bound triosmium cluster failed. In order to demon-
strate proof of concept for the technique of visualizing a guanine
bound metal complex on DNA with TEM we studied the reaction of
showed a single peak at ꢀ2625 ppm relative to chloroplatinic acid
(Fig. 1b) [10]. To this solution of 1 was added an equimolar (based
on the piperonylamine) amount of dGMP and a shift to ꢀ2511 ppm
was observed (Fig. 1c). This gives evidence for the formation of
a single adduct of 1 with dGMP.
2.2. The reaction of 1 with dGMP followed by 1H NMR
Equimolar solutions of 1 and dGMP were combined in D2O and
the reaction followed by 1H NMR over the course of 24 h using
solvent suppression for the HDO peak (Fig. 2). Resonances assign-
able to residual DMF were observed at 8.3 ppm (CH) and 2.3 and
2.4 ppm (2 Me). Thirty minutes after addition of dGMP a new
resonance appeared at 8.5 ppm downfield of the H(8) of dGMP at
7.9 ppm which we assign to the formation of the adduct 1-dGMP
(Equation (2)).
the complexes [Pt(en)Cl(NH2R)]þNO3ꢀ with dGMP and DNA. It was
previously been shown that complexes of this type were selective
for binding to guanine when R was a luminescent molecule [9]. We
report here the successful visualization of the Pt atoms in the
complex [Pt(en)Cl(NH2R)]þNO3ꢀ (R]C8H9NO2 (benzo[d][1,3]dioxol-
5-ylmethanamine, also known as piperonylamine) (1)) bound to
DNA by TEM as well as a qualitative kinetic study of a series of
related complexes reacting with dGMP and the synthesis and dGMP
binding of a Pt-triosmium conjugate.
With increasing time the resonance at 7.9 ppm continues to
decrease while the resonance at 8.5 ppm. After 19 h the resonance
at 7.9 ppm has completely disappeared and only the peak assign-
able to 1-dGMP is observed accompanied by the DMF resonance at
8.3 ppm in the downfield region of the spectrum. Other changes in
the resonances above 7.0 ppm also observed.
2.3. Transmission electron microscopy of 1 incubated with ss-DNA
GATC repeat
2. Results
Based on these results 1 was incubated with a 25 mmol solution
of a single-stranded GATC repeat (60 bases in length) at 40 ꢁC for
6 h in aqueous phosphate buffer using a six-fold molar excess of
label (compound 1) relative to the number of guanines. After
dialysis to remove excess label a micro drop of the solution was
subjected to the stretching and straightening technique developed
by Halcyon Molecular [5]. Multiple strands were deposited on glass
2.1. Synthesis of 1 and its reaction with dGMP
Complex 1 was synthesized by the reaction of benzo[d][1,3]
dioxol-5-ylmethanamine with [(en)PtCl2] via halide abstraction in
dimethylformamide (Equation (1)).
H2
N
H2
N
H2
+
AgNO3
DMF
+
Cl
Cl
N
Cl
O
O
O
Cl
+
H2N
-
Pt
NO3
Pt
Pt
(1)
H
N
O
O
N
N
N
H2
H2
-
H2
H2
NO3
N
Me
1
Me
The complex was characterized by 1H, 195Pt NMR and mass
spectrometry. The 195Pt NMR of the DMF intermediate complex
showed two overlapping resonances in DMF/THF-d8 which we
attribute to a mixture of the DMF and THF-d8 complexes (Fig.1a). To
this solution was added 0.8 equivalents of piperonylamine, the
solution evaporated to dryness and dissolved in D2O. The 195Pt NMR
slide, sprayed with carbon and imaged on the 300 kV VG Micro-
scope at Oak Ridge National Lab [11]. Fig. 3 shows a TEM image of
a segment of the labeled GATC repeat. There are 15 G bases in the
segment using the calculated value 0.77 nm/base and 11 of them
are labeled (green dots). There are several labels that are out of
sequence (red dots) and these could be labeled adenines.

R. Kumar et al. / Journal of Organometallic Chemistry 734 (2013) 53e60
55
repulsive interactions can be produced between the NH2 in position
6 of adenine and a platinum-bonded amine ligand [9].
2.5. Kinetic studies of amine variants, compounds 2e7, with dGMP
The TEM experiment constitutes a proof of concept in that it
illustrates that it is possible to visualize a single platinum atom with
base selectivity, bound to stretched single-stranded DNA. However,
for this labeling scheme to be effective the reaction of 1 with dGMP
must be much faster. To this end, a study of the rates of reaction of
a series of complexes related to 1, Pt(en)Cl(NH2R)]þNO3 where R]
C8H11N (phenethylamine) (2), C7H9N (benzylamine) (3), C6H7N
(aniline) (4), C6H6IN (p-iodo-aniline) (5) C3H9NO (2-methoxy-eth-
ylamine) (6) and C6H13N (cyclohexylamine) (7) were synthesized
and their rates of reaction with dGMP were followed by 1H NMR
(Chart 1). The amines were chosen for their differences in steric
bulk and basicity of the donor atom. The aromatic amines in this
series were chosen based on the prior work of Heetebrij et al. who
found that aromatic amines showed a higher affinity for guanine
relative to the other bases and attributed this to p-stacking effects
[9]. The aliphatic amines used were chosen for comparison to see
how important the stacking effect was for the rate of reaction as
well as the selectivity. In the case of 5 a second heavy atom was
included to increase electron scattering to improve visualization of
the label. The experiments were conducted with equimolar
amounts of the amine complexes and dGMP in D2O with water
suppression. The reaction of 2 with dGMP is shown in Fig. 5 and it
can be seen that it reacts much more quickly 1, with the reaction
being almost complete after just 2 h. Fig. 6 compares the rate of
conversion of 1 with 2 and 4. Complex 3 showed the same rate of
conversion as 1, both being benzyl amines. Fig. 7 compares the rate
of conversion of 4 with 5 and shows that inclusion of the iodine
atom does not have huge influence on the initial rate but does
decrease the overall % conversion after 24 h. Complexes 6 and 7
reacted only sluggishly with dGMP and showed <50% after 24 h. It
is clear from these results that the longer tether in the phene-
thylamine complex results in the fastest rate of conversion and
represents the best candidate for single metal atom labeling of
guanine in single-stranded DNA.
2.6. Synthesis and structure [Os3(CO)10PPh2(CH2)2NH2] (8)
Fig. 1. a) 195Pt NMR of [Pt(en)Cl(DMF)]NO3 in THF-d4/DMF. b) 195Pt NMR of [Pt(en)
Cl(piperonyl)]NO3 in D2O. c) 195Pt NMR of [Pt(en)Cl(piperonyl)]NO3 þ dGMP in D2O.
The studies with complexes 1e5 showed that a two-carbon
tether gave the best rate of conversion. However, visualizing
a single heavy metal atom label would require the very powerful
TEM located at ORNL [11]. In the hope of combining the effectiveness
of the two-carbon tether with a metal cluster label, which would
allow in-house imaging with more conventional TEM instruments,
we synthesized the compound [Os3(CO)11PPh2(CH2)2NH2(en)PtCl]
NO3 (9) using the bidentate ligand 2-diphenylphosphino-ethyl-
amine (Scheme 1). Os3(CO)11(CH3CN) [12] was reacted with 2-
diphenylphosphino-ethylamine at ambient temperature in methy-
lene chloride. The spectroscopic data indicated that the reaction had
2.4. 1H NMR studies of 1 with dAMP
In order to test this hypothesis the reaction of deoxyadenosine
monophosphate (dAMP) with 1 was studied. The 1H NMR of
equimolar amounts of dAMP and 1 in D2O was followed over a 24 h
period using water suppression. Although evidence of reaction is
observed in the form of new resonances appearing in the downfield
(8 ppm) and upfield (2e7 ppm) regions of the spectrum the H(8) of
the adenine is still present after 24 h (Fig. 4). Thus the reaction 1
with dAMP is much slower than with dGMP and appears to be less
specific, given the greater number new resonances that appeared
over the 24 h period. Nonetheless, this experiment does lend
credence to the idea that the outlier labels (red dots) are 1-dAMP
although it is not clear if this product has the same structure as
1-dGMP (Equation (2)). Monofunctional Pt complexes give simpler
Pt-G monoadducts.
resulted in the displacement of
a carbonyl ligand to give
[Os3(CO)10PPh2(CH2)2NH2] (8) (Scheme 1). However, we could not
determine if the ligand was coordinated in a bridging or chelating
mode and so a solid-state structure was undertaken. The solid-state
structure of 8 is shown in Fig. 7 with selected bond lengths and
angles in the figure caption. The crystal and collection data are given
in Table 1. The 2-diphenylphosphino-ethylamine ligand is chelated
to one Os atom. There is considerable precedent in the literature for
the chelate structure being preferred for bidentate ligands with an
ethylene bridge in the chemistry of triosmium clusters. However,
the equilibrium between the chelate and two metal atom bridging
structure is highly dependent on the nature of the of the two donor
N7 of both adenine and guanine can be platinated, however N7
of guanine shows a greater kinetic preference. This tendency results
from the stronger basicity of that nitrogen and from possible
simultaneous hydrogen-bond interactions between NHþ3 protons
and the O(6) of guanine. In contrast, in the case of adenine, only

56
R. Kumar et al. / Journal of Organometallic Chemistry 734 (2013) 53e60
Fig. 2. 1H NMR of 1 þ equimolar dGMP versus time.
atoms, i.e. (PP), (PS), (PN) or (PO) [13e17]. The structure of 8 consists
of an Os triangle with the chelating diphenylphosphino-2-
ethylamine ligand on Os(3). The Os(2)eOs(3) bond length is
2.7. Conversion of (8) to [Os3(CO)11PPh2(CH2)2NH2(en)PtCl]NO3 (9)
and attempted reaction with dGMP
ꢀ
slightly shorter than (2.86(2) A) than the other two OseOs bonds
The opening of the chelate ring at the Os(3)eN(1) bond could be
accomplished by treating 8 with CO gas at 50 ꢁC in THF-d8. The
reaction is followed by 31P NMR and after 1 h the 31P NMR reso-
nance at 33.70 ppm assigned to 8 is replaced by a new resonance at
35.55 ppm. To the THF solution was added an equimolar amount of
ꢀ
(2.91(2) A.) The Os(3)eP(1) and Os(3)e(N1) bond lengths (2.32(2)
ꢀ
and 2.25(2) A) respectively are fairly typical of these bonds in tri-
osmium clusters [14e18].
Fig. 3. TEM image of 1 incubated with a GATC repeat using a 300 kV VG microscope at
Oak Ridge National lab.
Fig. 4. 1H NMR studies of 1 þ equimolar dAMP versus time.

R. Kumar et al. / Journal of Organometallic Chemistry 734 (2013) 53e60
57
Chart 1.
[Pt(en)Cl(DMF)]NO3. The Pt-cluster adduct was isolated by
preparative reverse phase TLC and characterized by IR and NMR
spectroscopies as [Os3(CO)11PPh2(CH2)2NH2(en)PtCl]NO3 (9).
A solution of 9 was then combined with an equimolar amount of
dGMP in 90:10 D2O:CD3OD. After 4 h there was some evidence of
formation of an adduct as noted by the appearance of a new
resonance at 8.6 ppm downfield of the H(8) resonance of guanine
as noted for the other dGMP adducts reported here. However with
increasing time a precipitation was noted and the overall intensity
of the NMR resonances increased. Thus it appears that if dGMP-9 its
solubility in this solvent mixture was too limited to make further
studies possible.
dGMP suggesting that there are several ways that the dAMP is
interacting with the label (Figs. 3 and 4).
Although 1 was reasonably selective for guanine its reaction
with dGMP was rather slow. Following the rates of the reactions of
complexes 2e7 revealed some of the factors that control the rate of
reaction of these Pt complexes with DNA. Complex 2, the phene-
thylamine analogue showed fastest reactivity as observed by NMR
spectroscopy. It was consumed in 2 h compared to 1, which took
12 h at RT. Complex 3 reacted at the same rate as 1 being a ben-
zylamine, illustrating that the dioxo ring has little influence on the
rate of reaction with dGMP, but is preferred for its better water
solubility. Complex 4 reacted more slowly than 1 or 2 and taken
together the results shown in Fig. 5 suggest that tether length to the
aromatic ring is the dominant factor in affecting the rate of reaction
with dGMP. The differences in basicity may also play a role but
based on the results for the remaining complexes this is apparently
of secondary importance. Complex 5 reacts more slowly than 4 and
this may be due to the electron withdrawing effect of the iodo
group (Fig. 6). This is unfortunate in that the iodo group would offer
enhanced imaging possibilities. Complexes 6 and 7 reacted very
slowly and reaction with dGMP was <50% converted after 24 h. We
tentatively assign this to steric considerations where the greater
motional degrees of freedom of 6 and the relative bulkiness of the
cyclohexyl group in 7 compared to phenyl interferes with the
displacement of chloride by the dGMP.
3. Discussion
The TEM results reported here for the labeling of GATC DNA with
1 using the stretching technique and the distance grid developed by
Halcyon Molecular represents a proof of concept for visualizing
a single platinum atom selectively bound to guanine [5]. The
selectivity for guanine is based on the clear results of our studies in
solution on the binding of 1 with dGMP. This selectivity is also
based on our previous studies with dGMP and triosmium cluster
labels [8]. Outlier spots were observed in the TEM and were
attributed side reactions with adenine that could be due to either
covalent bonding as with guanine or to non-specific electrostatic
interactions. Following the reaction of dAMP with 1 resulted in
much more complex changes in the 1H NMR of the dAMP relative to
The synthesis of complexes 8 and 9 opens the way for the labeling
of DNA bases with clusters attached to a platinum complex.
However, water solubility of the 9-dGMP conjugate or 9 itself
presents a significant obstacle. Previous studies have shown
Fig. 5. Kinetic plots of the reaction complexes 1, 2 and 4 with dGMP run at ambient
temperatures in D2O and based on integration of the adduct peak to the H(8) peak of
dGMP relative to the HDO peak.
Fig. 6. Kinetic plots of the reaction complexes 4 and 5 with dGMP run at ambient
temperatures in D2O and based on integration of the adduct peak to the H(8) peak of
dGMP relative to the HDO peak.

58
R. Kumar et al. / Journal of Organometallic Chemistry 734 (2013) 53e60
application of high-powered TEM for visualization of single metal
atoms attached to bio-macromolecules. However, taken together
with the other unpublished work done in collaboration with
Halcyon Molecular it does not appear that fast throughput labeling
using heavy atom labels and the DNA stretching technique will be
a viable pathway for fast throughput genetic sequencing [20]. Other
techniques appear to be much further down this path [21].
5. Experimental
5.1. Materials
All amines, TMAO (trimethylamine N-oxide), Pt(en)Cl2, dAMP
and dGMP were purchased from SigmaeAldrich Chemicals and
used as received. Osmium carbonyl and diphenylphophino-2-
ethylamine were purchased from Strem Chemicals and used as
received. Deuterated solvents were purchased from Cambridge
Isotopes Inc and used as received. Acetonitrile and methylene
chloride were distilled from calcium hydride directly before use.
Other solvents were reagent grade and were purchased from J. T.
Baker and used as received. Syntheses of the Os-cluster compounds
carried out under a nitrogen atmosphere, but purification processes
were carried out in air using preparative thin layer chromatography
Fig. 7. Solid-state structure of [Os3(CO)10PPh2(CH2)2NH2] (8). Selected bond distances:
N(1)eOs(3), 2.249(3); N(1)eH(1A), 0.9200; N(1)eH(1B), 0.9200; P(1)eOs(3),
ꢀ
2.3206(9); Os(1)eOs(2), 2.9028(2); Os(1)eOs(3), 2.9151(2); Os(2)eOs(3), 2.8626(2) A.
ꢀ
Average CO bond distance: 1.147(5) A. Average OseCeO bond angle: 176.1(3) deg.
(20 ꢂ 20 cm plates coated with 500
mm silica from Dynamic
Absorbents or Uniplate reverse phase, hydrocarbon impregnated
however that the introduction of water-soluble phosphines onto the
cluster label causes distortion of the stretched single-stranded DNA
making accurate calculations of the ladder grid impossible [19].
silica gel 1000 mm).
5.2. Methods
4. Conclusions
1H, 195Pt and 31P NMR spectra were obtained on a Varian NMR
system 500 MHz spectrometer at 500, 109.1 and 202.6 MHz respec-
tively. Infrared spectra were obtained on a Thermo-Nicolet 633 FT-IR
spectrometer. Prior to the NMR experiments all Pt complexes (except
9) werepurifiedona Biotage(Isolera primeFlash purificationsystem)
The relative rates reported here for the reactions of 1e7 with
dGMP could prove useful for future applications of the reactions of
Pt complexes with DNA. The visualization of 1 attached to
a stretched single stranded DNA is proof of concept for the
Scheme 1.

R. Kumar et al. / Journal of Organometallic Chemistry 734 (2013) 53e60
59
[Pt(en)Cl(dmf)]þNO3ꢀ were removed by filtration through
membrane filters (0.2 m). The clear filtrate was lyophilized to give
the final product (1) in approximately 45% yield. 1H NMR (D2O,
500 MHz, 298 K): 2.2 (s, 3NH2, en), 2.6 (s, en), 3.6 (s, CH2), 5.8 (s,
CH2 ppr), 6.7 (s, CH, Ar), 6.8 (s, CH, Ar), 6.9 (s, CH, Ar). 195Pt NMR
Table 1
Crystal data and structure refinement for compound 8.
m
Empirical formula
C24 H16 N O10 Os3 P
d
Formula weight
Temperature
Wavelength
Crystal system
Space group
Unit cell
1079.95
100(2) K
ꢀ
0.71073 A
(D2O, 500 MHz, 298 K)
d
ꢀ2624.8. MS (ES): m/z 443.
Monoclinic
P2(1)/n
5.3.3. 1-dGMP
a ¼ 8.9448(3) A
a
b
g
¼ 90ꢁ
ꢀ
dimensions
b ¼ 14.7147(4) A
¼ 101.5820(10)ꢁ
¼ 90ꢁ
ꢀ
In an NMR tube 1 (4.5 mg, 0.089 mmol, 1 equiv.) and dGMP
(3.0 mg, 0.089 mmol, 1 equiv.) were mixed in 0.5 mL D2O. The
mixture was kept at 40 ꢁC for overnight. A new resonance
at ꢀ2511 ppm appeared in 195Pt NMR and the ꢀ2625 ppm peak
disappeared completely.
ꢀ
c ¼ 20.9890(7) A
3
ꢀ
Volume
Z
Density
(calculated)
Absorption
coefficient
F(000)
2706.32(15) A
4
2.651 Mg/m3
In a separate reaction the reaction was followed by 1H NMR
14.165 mmꢀ1
a new peak appeared at
completely.
d 8.2 and 7.9 H(8) peak disappeared
1960
0.07 ꢂ 0.05 ꢂ 0.04
mm3
Crystal size
5.3.4. [Pt(II)phenethylamine(en)Cl](NO3) (2)
Crystal color/
habit
Theta range for
data collection
Index ranges
Yellow plate
Phenethylamine (14.7 mg, 0.122 mmol, 0.8 equiv.) was dissolved
in xylene (12 mL) and added to the [Pt(en)Cl(DMF)]NO3
1.70e25.39ꢁ
(0.122 mmol) solution. 1H NMR (D2O, 500 MHz, 298 K):
d 2.2 (s,
ꢀ10 ꢃ h ꢃ 10, ꢀ17
ꢃ k ꢃ 17, ꢀ24 ꢃ l ꢃ 25
50804
3NH2), 2.4 (s, en), 2.8 (s, CH2), 2.9 (s, CH2), 7.2 (m, CHphy), 7.4 (m,
CHphy). MS (ES): m/z 412.
Reflections
collected
Independent
reflections
4970 [R(int) ¼ 0.0318]
5.3.5. [Pt(II)benzylamine(en)Cl](NO3) (3)
Benzylamine (13.07 mg, 0.122 mmol, 0.8 equiv.) was dissolved in
xylene (12 mL) and added to the [Pt(en)Cl(DMF)]NO3 (0.122 mmol)
Completeness
to theta ¼
100.0%
solution. 1H NMR (D2O, 500 MHz, 298 K):
d 2.0 (s, 3NH2), 2.8 (t, en),
25.00ꢁ
Absorption
correction
Semi-empirical from
equivalents
3.8 (s, CH2), 7.3 (m, CH, Ar). MS (ES): m/z 399.
Max. and min.
transmission
Refinement method
0.6011 and 0.4372
5.3.6. [Pt(II)aniline(en)Cl](NO3) (4)
Full-matrix least-squares
on F2
4970/0/352
Aniline (11.36 mg, 0.122 mmol, 0.8 equiv.) was dissolved in
xylene (12 mL) and added to the [Pt(en)Cl(DMF)]NO3
Data/restraints/
parameters
(0.122 mmol) solution. 1H NMR (D2O, 500 MHz, 298 K):
d 2.0 (s,
3NH2), 2.8 (t, en), 6.6 (m, CH, Ar), 6.8 (m, CH, Ar), 7.2 (m, CH, Ar).
MS (ES): m/z 385.
Goodness-of-fit
1.128
on F2
Final R indices [I >
2sigma(I)]
R indices (all data)
Largest diff. peak
and hole
R1 ¼ 0.0150, wR2 ¼ 0.0299
R1 ¼ 0.0187, wR2 ¼ 0.0315
5.3.7. [Pt(II)(4-iodoaniline)(en)Cl](NO3) (5)
4-iodoaniline (27.72 mg, 0.122 mmol, 0.8 equiv.) was dissolved
in xylene (12 mL) and added to the [Pt(en)Cl(DMF)]NO3
ꢀꢀ3
0.844 and ꢀ0.798 e A
(0.122 mmol) solution. 1H NMR (D2O, 500 MHz, 298 K):
d 2.0 (s,
3NH2), 2.8 (t, en), 6.4 (m, CH, Ar), 7.4 (m, CH, Ar). MS (ES): m/z 511.
usingChromobondFlashFM15/2C18,9cmandChromobondFlashFM
25/5C-18,10cmcolumns.Massspectrafortheaminecomplexeswere
obtained on a Shimadzu LCMS 2020. The mass spectrum for
compound 8 was obtained on a Waters e2695 separation module
combined with a Micromass technologies LCT Premier XE.
5.3.8. [Pt(II)(2-methoxyethanamine)(en)Cl](NO3) (6)
2-methoxyethanamine (9.16 mg, 0.122 mmol, 0.8 equiv.) was
dissolved in xylene (12 mL) and added to the [Pt(en)Cl(DMF)]NO3
(0.122mmol)solution.1H NMR(D2O,500MHz,298K):
d2.0(s, 3NH2),
2.7 (t, CH2), 2.8 (s, en), 3.3 (s, CH3), 3.8 (t, CH2). MS (ES): m/z 377.
5.3. Syntheses
5.3.9. Pt(II)(cyclohexylamine)(en)Cl(NO3) (7)
Cyclohexylamine (12.34 mg, 0.122 mmol, 0.8 equiv.) was dis-
solved in xylene (12 mL) and added to the [Pt(en)Cl(DMF)]NO3
5.3.1. [Pt(en)Cl(DMF)]NO3
[Pt(en)Cl2] (50 mg, 0.153 mmol) was suspended in DMF (5 mL)
and AgNO3 (24.8 mg, 0.146 mmol, 0.95 equiv.) in DMF (1 mL) was
added. The solution was stirred for 16 h at room temperature in the
(0.122 mmol) solution. 1H NMR (D2O, 500 MHz, 298 K):
d 1.1 (m,
CH2), 1.5 (m, CH2), 1.7 (m, CH2), 2.0 (s, 3NH2), 2.7 (m, CH), 2.8 (s, en),
3.3 (s, CH3), 3.8 (t, CH2). MS (ES): m/z 391.
dark and filtered through membrane filters (0.2
mm) to remove
AgCl.
5.3.10. [Os3(CO)10PPh2(CH2)2NH2 (8)
[Os3(CO)11(NCCH3)] [12] (370 mg, 0.402 mmol) was dissolved in
10 mL CH2Cl2 and PPh2(CH2)2NH2 (0.092 mg, 0.402 mmol) was
added. The solution was stirred at room temperature for overnight.
Analytical thin layer chromatography (30% CH2Cl2 and 70% hexane
as eluent) was used to confirm complete conversion. The solution
was rotary evaporated under vacuum to dryness. Two bands were
observed on preparative TLC. Faster moving band gave a cyclized
5.3.2. [Pt(II)(piperonylamine)(en)Cl]þNO3ꢀ (1)
Piperonylamine (18.5 mg, 0.122 mmol, 0.8 equiv relative to
[Pt(en)Cl2]) was dissolved in xylene (12 mL) and added to the
[Pt(en)Cl(DMF)]þNO3ꢀ solution. The mixture was stirred overnight
at 40 ꢁC. Solvents were removed en vacuo and the remaining
product was redissolved in a minimum amount of Milli-Q water
and stored overnight at 4 ꢁC. Insoluble yellow particles of residual
aminoephosphine complex 8. Yield for 8: 45%. IR n(CO) in KBr:

60
R. Kumar et al. / Journal of Organometallic Chemistry 734 (2013) 53e60
2079 m, 2028 s, 2004 m, 1974 w cmꢀ1. 1H NMR (500 MHz in CH2Cl2
at 25 ꢁC) 7.53 (m, 3H, H(3,4,5)), 7.42 (m, 2H, H(2,6)), 2.43 (s (NH2)),
1.56 (m, 2H, CH2b-NH2), 1.09 (m, 2H, CH2a-NH2). 13C NMR (CH2Cl2):
187.35 (s, 4CO), 187.45 (s, 4CO), 183.95 (s, 2CO). 31P NMR (CH2Cl2):
33.70 (s, 1P, PPh2(CH2)2NH2). Mass spectrum: calcd 1090 amu;
observed 1089 amu and 1112 (parent ion plus Naþ).
consistent with the proposed structure. All non-hydrogen atoms
were refined anisotropically by full-matrix least-squares (SHELXL-
97). All hydrogen atoms were placed using a riding model. Their
positions were constrained relative to their parent atom using the
appropriate HFIX command in SHELXL-97.
Acknowledgements
5.3.11. Chelate ring opening of 8 with CO complexation with [Pt(en)
Cl(DMF)]NO3
We gratefully acknowledge Halcyon Molecular and the Univer-
sity of Montana for support of this research and Christopher Own
for help with the TEM imaging.
40 mg (0.036 mmol) of compound 8 dissolved in THF-d8 (10 mL)
and CO gas was bubbled through the solution at 50 ꢁC for 1 h. The
reaction was monitored by 31P NMR spectroscopy the resonance at
33.70 ppm assignable to 8 shifted to 35.55 ppmwhichweassigntothe
ring-opened compound [Os3(CO)11PPh2(CH2)2NH2]. A solution of
[Pt(en)Cl(DMF)]NO3 (0.03mmol)in1.5mLDMFwasthenaddedtothe
reaction flask and was stirred at room temperature for overnight. The
Pt-cluster adduct was then precipitated with diethyl ether to remove
DMF and purified by reverse phase preparative TLC (50% MeOH, 50%
CH2Cl2). One faint band was observed and isolated by extraction with
methanol and rotary evaporation. 35 mg (0.023 mmol) of a greenish-
brownsolidwasobtainedwhosespectroscopicdata isconsistentwith
Appendix A. Supplementary materials
Crystallographic data for 8 have been deposited with the Cam-
bridge Crystallographic Data Center CCDC No. 903534. Copies of the
information may be obtained free of charge from the Director,
CCDC, 12 Union Road, Cambridge, CB2 IEZ UK (fax: 44-1223-336-
033); email: deposit@ccdc.ac.uk or http://wwwccdc.cam.ac.uk.
[Os3(CO)11PPh2(CH2)2NH2(en)PtCl]NO3 (9).
n(CO) in KBr: 1966 m,
References
2003 s, 2047 m, 2081 w cmꢀ1. 1H NMR (500 MHz in CD3OD at 25 ꢁC)
7.8e8.2 (broadm,12H, Phen), 5.9 (s, 2H, NH2), 3.6(s, 2H, CH2eP), 3.1(t,
4H, en), 1.6 (s, 2H, CH2eN). 31P NMR (CD3OD) 35.55.
[1] M. Beer, E.N. Moudrianakis, Proc. Natl. Acad. Sci. U S A 48 (1962) 409.
[2] H.P. Erickson, M. Beer, Biochemistry 6 (1967) 2694. N. Y.
[3] J.P. Langmore, N.R. Cozzarelli, A.V. Crewe, Proc. Electron Microsc. Soc. Am. 30
(1972) 184.
[4] J.J. Frizell, J. Theor. Biol. 59 (1976) 241.
[5] W. Andregg, M. Andregg, PCT Int. Appl. (2009). WO 2009046445 A1
20090409.
[6] E. Rosenberg, F. Spada, K. Sugden, B. Martin, L. Milone, R. Gobetto, A. Viale,
J. Fiedler, J. Organomet. Chem. 668 (2003) 51.
[7] E. Rosenberg, F. Spada, K. Sugden, B. Martin, R. Gobetto, L. Milone, A. Viale,
J. Organomet. Chem. 689 (2004) 4729.
[8] E. Rosenberg, R. Kumar, Dalton Trans. 41 (2012) 714.
[9] R.J. Heetebrij, E.G. Talman, A.v. Velzen, R.P.M.v. Gijlswijk, S.S. Snoeijers,
M. Schalk, J. Wiegant, F.v.d. Rijke, R.A.K. Kerkhoven, H.J. Tanke, J. Reedijk,
H.J. Houthoff, ChemBioChem 4 (2003) 573.
[10] J.R.L. Priqueler, I.S. Butler, F.D. Rochon, Appl. Spectrosc. Rev. 41 (2006) 185e
226.
[11] Materials Science and technology Group Oak Ridge National Laboratory
P.O. Box 2008, Building 4515 1 Bethel Valley Road Oak Ridge, TN 37831e
6071.
[12] B.F.G. Johnson, J. Lewis, J. Inorg. Synth. 13 (1972) 93.
[13] D.A. Horvat, E. Nordlander, M.G. Richmond, Organometallics 31 (2012) 6608.
[14] Antony J. Deeming, Kenneth I. Hardcastle, Shariff E. Kabir, J. Chem. Soc. Dalton
Trans. Inorg. Chem. (1988) 827.
5.3.12. Reaction between dGMP and [Os3(CO)11PPh2(CH2)2NH2(en)
PtCl]NO3 (9)
Equimolar molar amounts of dGMP (0.023 mmol) and
[Os3(CO)11PPh2(CH2)2NH2(en)PtCl]NO3 (0.023 mmol) were added
in CD3OD and D2O (90:10). Reaction was followed by 1H NMR. Some
conversion to dGMP-9 was observed after 4 h as noted by the
decrease in H(8) resonance of dGMP and the appearance of a new
resonance at 8.6 ppm but due to precipitation of either the dGMP-9
or 9, conversion rate could not be followed precisely.
5.4. Solid-state structure determination of 8
A yellow plate 0.07 ꢂ 0.05 ꢂ 0.04 mm in size was mounted on
a Cryoloop with Paratone oil. Data were collected in a nitrogen gas
stream at 100(2) K using phi and omega scans. Crystal-to-detector
distance was 60 mm and exposure time was 20 s per frame using
a scan width of 1.0ꢁ. Data collection was 100.0% complete to 25.00ꢁ
[15] X. Zhang, S. Kandala, L. Yang, W.H. Watson, X. Wang, D.A. Hrovat, W.T. Borden,
M.G. Richmond, Organometallics 30 (2011) 1253.
in
q. A total of 50804 reflections were collected covering the
[16] R. Persson, M. Monari, R. Gobetto, A. Russo Andrea, S. Aime, M.J. Calhorda,
E. Nordlander, Organometallics 20 (20) (2001) 4150e4160.
[17] R. Persson, M.J. Stchedroff, B. Uebersezig, R. Gobetto, J.W. Steed, P.D. Prince,
M. Monari, E. Nordlander, Organometallics 29 (2010) 2223.
[18] E. Rosenberg, S.E. Kabir, L. Milone, R. Gobetto, D. Osella, K.I. Hardcastle,
S. Hajela, E. Wolf, K. Moizeau, D. Espisita, Organometallics 16 (1997)
2665.
[19] S. Kemmish, E. Rosenberg, unpublished results.
[20] E. Rosenberg, R. Kumar, W. Andregg, unpublished results.
[21] Oxford Nanopore Technologies.www.nanoporetech.com.
indices, ꢀ10 ꢃ h ꢃ 10, ꢀ17 ꢃ k ꢃ 17, ꢀ24 ꢃ l ꢃ 25. 4970 reflections
were found to be symmetry independent, with an Rint of 0.0318.
Indexing and unit cell refinement indicated a primitive, monoclinic
lattice. The space group was found to be P2(1)/n (No. 14). The data
were integrated using the Bruker SAINT software program and
scaled using the SADABS software program. Solution by direct
methods (SIR-97) produced a complete heavy-atom phasing model