Multinuclear 1D and 2D NMR investigations on the interaction between the pyrimidic nucleotides 5'-CMP, 5'-dCMP, and 5'-UMP and diethyltin dichloride in aqueous medium

Article abstract of DOI:10.1002/(SICI)1099-0682(200003)2000:3<513::AID-EJIC513>3.0.CO;2-M

The interactions between diethyltin dichloride and 5'-CMP, 5'-dCMP, and 5'-UMP in aqueous solution were investigated by multinuclear 1D and 2D NMR techniques including ¹¹⁹Sn, ¹⁵N and ³¹P nuclei. These studies were combined with electrospray mass spectrometry, infrared spectroscopy, solid state ¹³C, ³¹P and ¹¹⁷Sn CP-MAS NMR, and elemental analysis. As demonstrated by ³¹P-¹H HOESY spectroscopy, the diethyltin moiety interacts with the phosphate group of the pyrimidic mononucleotides in the pH range 0.5-3.5. Compound 8 (X = Cl), the solid isolated in this pH range from 5'- CMP, contains two tin atoms bridged by one oxygen and one chlorine atom, each tin atom being linked to the phosphate group of a nucleotide moiety. For 5'- UMP the solid isolated (12) has a dimeric structure with two different tin atoms; it can be formed by dimerization of compound 11 with the elimination of two water molecules. As demonstrated by ¹H-¹¹⁹Sn HMQC correlation NMR data and the ²J(¹¹⁹Sn-O-¹¹⁷Sn) coupling constant of 156 Hz, the diethyltin moiety reacts with the O(2') and O(3') oxygen atoms of the nucleotides to form a dimeric diethyldioxastannolane at pH > 9.0. Between pH 5.0 and 9.0, no evidence for any interaction between the diethyltin moiety and the nucleotides was found.

Full text of DOI:10.1002/(SICI)1099-0682(200003)2000:3<513::AID-EJIC513>3.0.CO;2-M

FULL PAPER
Multinuclear 1D and 2D NMR Investigations on the Interaction between the
Pyrimidic Nucleotides 5’-CMP, 5’-dCMP, and 5’-UMP and Diethyltin
Dichloride in Aqueous Medium
Laurent Ghys,[a,b] Monique Biesemans,[b,c] Marcel Gielen,[c] Achilleas Garoufis,[d]
[b]
Nick Hadjiliadis,[e] Rudolph Willem,*[b,c] Jose C. Martins
´
Keywords: Tin / Nucleotides / NMR spectrometry / Mass spectroscopy
The interactions between diethyltin dichloride and 5’-CMP,
each tin atom being linked to the phosphate group of a nuc-
5’-dCMP, and 5’-UMP in aqueous solution were investigated leotide moiety. For 5’-UMP the solid isolated (12) has a di-
by multinuclear 1D and 2D NMR techniques including 119Sn,
15N and 31P nuclei. These studies were combined with elec-
trospray mass spectrometry, infrared spectroscopy, solid state
13C, 31P and 117Sn CP-MAS NMR, and elemental analysis. As
demonstrated by 31P-1H HOESY spectroscopy, the diethyltin
moiety interacts with the phosphate group of the pyrimidic
mononucleotides in the pH range 0.5–3.5. Compound 8 (X =
Cl), the solid isolated in this pH range from 5’-CMP, contains
two tin atoms bridged by one oxygen and one chlorine atom,
meric structure with two different tin atoms; it can be formed
by dimerization of compound 11 with the elimination of two
water molecules. As demonstrated by 1H-119Sn HMQC cor-
relation NMR data and the 2J(119Sn–O–117Sn) coupling con-
stant of 156 Hz, the diethyltin moiety reacts with the O(2’)
and O(3’) oxygen atoms of the nucleotides to form a dimeric
diethyldioxastannolane at pH Ͼ 9.0. Between pH 5.0 and 9.0,
no evidence for any interaction between the diethyltin moi-
ety and the nucleotides was found.
Introduction
phenanthroline) with 5’-dGMP in aqueous medium using
[trans-en2Os(η-H2)](CF3SO3)2, a versatile H NMR probe.
1
Many organometallic compounds exhibit interesting
antitumoural activity against several human cancer cell
lines.[1] The well-known complex cisplatin, Pt(NH3)2Cl2,
clinically used in cancer chemotherapy, proved to interact
mainly with the N7 atoms of two adjacent guanines in the
same DNA strand.[2–5] For several years, we have been in-
terested in the synthesis and characterisation of organotin
complexes which, in some cases, display higher antitumour
activities in vitro than cisplatin.[6] In contrast to platinum
compounds, very little is known about the origin of the
antitumoural activity of organotin compounds although, as
for cisplatin, DNA was proposed to be the target.[7] Some
work has already been performed on the interactions be-
tween organotins and DNA or its precursors. Barbieri et al.
have investigated the solid resulting from the reaction of
ethanolic organotins [SnR2Cl2(EtOH)2 and SnR3Cl(EtOH)]
with aqueous DNA or mononucleotide solutions. However,
only 119Sn Mössbauer spectroscopy was used.[8–10] The con-
densate was assumed to be a diorganotin species interacting
with two vicinal phosphate groups of DNA. Li et al. have
investigated the interaction of (C2H5)2SnCl2(phen) (phen ϭ
They also studied solid mixtures of the same diethyltin
dichloride complex with 5’-AMP, 5’-CMP, and 5’-GMP dis-
1
solved in DMSO, using H and 31P NMR and UV spectro-
scopy.[11a] More recently, Hadjiliadis et al.[11b] studied the
interaction of the purine nucleotides 5’-IMP and 5’-GMP
1
with Et2SnCl2 using various techniques, including H, 13C
and 31P 1D NMR and Mössbauer spectroscopy. Finally, a
very recent paper described the investigation of the interac-
tion between dimethyltin dichloride and 5’-GMP, 5’-ATP
1
and 5’-d(CGCGCG)2 by potentiometric titrations and H
and 31P NMR spectroscopy.[11c] All these authors con-
cluded that the organotin moiety reacts with the phosphate
group of the nucleotides. However, these interactions have
not been investigated further using spectroscopic techniques
that provide more detailed structural data, such as 119Sn
NMR spectroscopy. Potentially, this technique is an ideal
tool of investigation, comparable to 195Pt NMR spectro-
scopy for platinum compounds.[12]
This study investigates the interaction between a dior-
ganotin dichloride, Et2SnCl2, and pyrimidic nucleic acid
precursors within a wide pH range in aqueous medium. The
three mononucleotides investigated, cytidine-5’-mono-
phosphate (5’-CMP), 2’-deoxycytidine-5’-monophosphate
(5’-dCMP) and uridine-5’-monophosphate (5’-UMP), were
selected on the basis of their increased solubility when they
are complexed to tin (Figure 1).
First, we monitored the 31P chemical shift of the monon-
ucleotides in aqueous solution as a function of pH in the
presence or absence of diethyltin dichloride and the 119Sn
chemical shift of a mixture of the latter with the mononu-
[a]
´
Universite Libre de Bruxelles, Service de Chimie Organique,
Av. F. D. Roosevelt, 50, B-1050 Brussels, Belgium
Vrije Universiteit Brussel, High Resolution NMR Centre,
Pleinlaan, 2, B-1050 Brussels, Belgium
Vrije Universiteit Brussel, Laboratory of General and Organic
Chemistry,
[b]
[c]
Pleinlaan, 2, B-1050 Brussels, Belgium
Technological Institute of Athens, Department of Chemistry,
Physics and Materials Technology,
[d]
[e]
12210 Egaleo, Greece
University of Ioannina, Department of Chemistry,
45110 Ioannina, Greece
Eur. J. Inorg. Chem. 2000, 513Ϫ522
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ1948/00/0303Ϫ0513 $ 17.50ϩ.50/0
513

R. Willem et al.
FULL PAPER
Table 1. Electrospray mass spectrometry data of a MeOH/H2O (90/
10) solution of Et2SnCl2 at pH 2.2
m/z
Assignment
653
635
441
277
245
5 ϩ H3Oϩ
5 ϩ Hϩ
2 ϩ Hϩ
[Et2SnCl(CH3OH)2]ϩ
[Et2SnCl(CH3OH)]ϩ
Figure 1. Structures and numbering scheme of the 5’-CMP, 5’-
dCMP and 5’-UMP mononucleotides
Table 2. Experimental elemental analysis data for the solid com-
pound isolated from an aqueous solution of Et2SnCl2 at pH 2.8 as
well as calculated data for the proposed structure
%C
%H
%Cl
%Sn
cleotides. The differences in 31P chemical shift titration pro-
files are used as indicators for an interaction between di-
ethyltin dichloride and the mononucleotides. Subsequently, [ClEt2SnOSnEt2Cl] (2)
Et2SnCl2 isolated at pH 2.8
21.8
21.8
4.3
4.6
14.0
16.1
52.2
53.9
the structural features of the interaction are characterised
1
by 1D and 2D NMR correlation techniques involving H,
13C, 15N, 31P and 119Sn nuclei, completed by 35Cl NMR
The pH of a 0.5  aqueous solution of Et2SnCl2 is ca.
2.2. A precipitate is generated when the pH of the supernat-
ant is raised above 2.2 with aqueous NaOH. Below pH 2.2,
a clear solution is observed. An electrospray mass spectrum
of a mixture obtained by adding methanol to an aqueous
solution of Et2SnCl2 (pH 2.2) with a final MeOH/H2O ratio
of 90/10 reveals two species. On the basis of data from Dak-
ternieks et al.[22a] these species can be assigned to com-
pounds 5 and 2 (Table 1). The fragment detected at m/z 441
in the electrospray mass spectrum is the protonated form of
intermediate 2 (Figure 2).
data where appropriate. Proton detected 1H-119Sn
HMQC[13–15]
and
phosphorus
detected
31P-1H
HOESY[16,17] have been reported to be very efficient for the
characterisation of organotin complexes in solution.[18]
Whenever precipitation occurs, the solid is isolated and
characterised by solid state 13C, 31P, and 117Sn CP-MAS
NMR and IR spectroscopy as well as elemental analysis.
None of the solids isolated provided crystals suitable for X-
ray diffraction analysis.
The other fragment ions correspond to [Et2SnCl]ϩ solv-
ated by one or two methanol molecules or to hydrated 5
(Table 1). These results clearly indicate that in solution at
pH 2.2 only one chloride anion dissociates from diethyltin
dichloride.
Results and Discussion
Aqueous Chemistry of Diethyltin Dichloride
The precipitate obtained at pH 2.8 was isolated. Its ele-
mental analysis data (Table 2) indicate that it basically
corresponds to compound 2, probably contaminated by 3,
considering the slight chlorine deficit.
Figure 2 shows the different intermediates (in their
monomeric form, for convenience) generated upon hydro-
lysis of Et2SnCl2 as mentioned in the literature.[19–21] The
NMR data collected during this preliminary investigation
are suitable references for the interpretation of data ob-
tained in the presence of nucleotides.
The 117Sn NMR spectrum of this sample in CDCl3 at
89.150 MHz displays eight resonances between δ ϭ –87 and
δ ϭ –182. They can be divided into two pairs of equally
intense resonances B and C and one set of four equally
intense resonances A. Set B (δ ϭ –91.7 and –139.9) displays
one pair of satellites with a coupling of 67 Hz. It is assigned
to intermediate 2 which is described by Otera[22b] as a dimer
for R ϭ Me, Et and Bu with similar 119Sn parameters for
R ϭ Et:[22b] δ(119Sn1) ϭ –140.6; δ(119Sn2) ϭ –91.8;
2J(119Sn–O–117/119Sn) ϭ 73 Hz. Set C (δ ϭ –157.0 and –
176.2) displays at least one pair of 2J(117Sn–O–117/119Sn)
satellites with a coupling of 212 Hz, others being unclear
due to a low signal-to-noise ratio. The 117Sn resonances of
Figure 2. Different species formed under hydrolysis of Et2SnCl2 in
aqueous solution
514
Eur. J. Inorg. Chem. 2000, 513Ϫ522

Interaction between the Pyrimidic Nucleotides 5’-CMP, 5’-dCMP, and 5’-UMP and Diethyltin Dichloride
FULL PAPER
Figure 3. (a) 31P NMR titration curve of 5’-CMP and a mixture of 5’-CMP and Et2SnCl2; full circles represent free 5’-CMP; crosses
correspond to 5’-CMP/Et2SnCl2; (b) 119Sn NMR titration curve for 5’-CMP/Et2SnCl2; multiple points related by straight lines for a given
pH value account for the presence of multiple resonances
set A (δ ϭ –87.9, –130.9, –151.8, and –181.8) display a pair Heteronuclear Overhauser Spectroscopy) technique[16–18]
of signals with a coupling constant of 98 Hz and another which correlates nuclei through dipolar couplings was pre-
pair with a coupling constant of 175 Hz. Although the spe- viously used by Barnham et al.[26] to characterise the Pt–
cies generating A and C cannot be identified, these NMR NH 5’-phosphate intermolecular interaction in the com-
data point to a mixture of intermediates including 3 and 4, plex [Pt(en)(5’-GMP–N7)2] 9H2O. This technique provided
possibly aggregated to various degrees during hydrolysis of a 2D spectrum of 5’-CMP/Et2SnCl2 at pH 2.2 with four
Et2SnCl2 depicted in Figure 2.
31P-1H NOE cross-peaks, two from the CH2 and CH3 pro-
tons of diethyltin dichloride and two from the H5’ and H6
protons of 5’-CMP. Similar results were obtained for the
mixtures of the two other nucleotides with Et2SnCl2, i.e. 5’-
dCMP/Et2SnCl2 and 5’-UMP/Et2SnCl2 at pH 2.2. With the
ratio (γ1H/γ31P) Ͼ 2.38 in mind, the distances between the
ethyl protons and the phosphorus atom of the three nucle-
Interactions of Diethyltin Dichloride with the Nucleotides in
Acidic Medium
Above pH 2.2, precipitation occurs from the solutions
of the three 2:1 mixtures of 5’-CMP/Et2SnCl2, 5’-dCMP/
Et2SnCl2, and 5’-UMP/Et2SnCl2. Consequently, samples
for NMR experiments in solution are obtained from the
supernatant of this precipitate. Because in acidic medium
the three nucleotides behaved similarly in the presence of
Et2SnCl2, only the mixture 5’-CMP/Et2SnCl2 is discussed
in detail.
˚
otides cannot exceed 4 A, proving that the tin moiety inter-
acts with the phosphate group of the nucleotide. 31P-1H-
NOE’s are always positive, irrespective of the correlation
time.[27] Hence, weak or missing NOE’s cannot be assigned
to a zero cross point of the NOE effect. The observed cross
peak between the H6 proton and the 31P nucleus of 5’-
CMP, 5’-dCMP, and 5’-UMP arises from the anti con-
formation of the base relative to the sugar unit which is also
known for free 5’-GMP.[28]
In the pH range 0.5–3.5, the addition of 0.5 equivalent
of diethyltin dichloride to 5’-CMP, 5’-dCMP and 5’-UMP
causes the 31P resonance of the phosphate group of the nuc-
leotides to shift to lower frequencies, with the maximum
change in chemical shift in the pH range 2.0–2.5. The titra-
tion curve of the 31P chemical shift is displayed in Figure 3a
for 5’-CMP and 5’-CMP/Et2SnCl2.
1
As assessed by the 119Sn chemical shifts and the J(13C–
119Sn) values,[29,30] the tin atom is six-coordinated in all
three mixtures, i.e. 5’-CMP/Et2SnCl2, 5’-dCMP/Et2SnCl2,
and 5’-UMP/Et2SnCl2 at pH 2.2 (Table 3). Empirical cor-
relations between C–Sn–C angles and coupling constants
involving the 119Sn nucleus[31] (Table 3) provide C–Sn–C
angle values typically ranging from 150° to 170°, which are
in agreement with distorted trans-octahedral, six-coordinate
geometries at the tin atom.
This 31P chemical shift modification can be traced back
to an interaction of diethyltin dichloride with the phosphate
group of the three nucleotides. H-119Sn, H-31P, 31P-119Sn,
and 13C-31P HMQC experiments[23,24] performed on 5’-
CMP/Et2SnCl2 at pH 2.2 failed to show any correlation
peaks between the resonances of the organotin moiety and
those of 5’-CMP. However, this does not formally exclude
1
1
As shown by 13C, 31P, and 117Sn CP-MAS NMR, the
the interaction between diethyltin dichloride and the phos- solid isolated from the mixture 5’-CMP/Et2SnCl2 (2:1) at
phate group of 5’-CMP. Indeed, the coupling can be very pH 2.2 contains both Et2Sn and 5’-CMP moieties. The 31P
small or even nonexistent and undetectable depending on CP-MAS NMR spectrum exhibits 5’–CMP two resonances
the bond strength.[25] The 31P-1H HOESY (31P detected at δ ϭ –11 and δ ϭ –12 which are low frequency shifted
Eur. J. Inorg. Chem. 2000, 513Ϫ522
515

R. Willem et al.
FULL PAPER
Table 3. 119Sn chemical shifts (ppm) and coupling constants (Hz) determined at pH 2.2 for the three nucleotides in the presence of
Et2SnCl2 in the molar ratio 2:1
5’-CMP/Et2SnCl2
5’-dCMP/Et2SnCl2
5’-UMP/Et2SnCl2
δ(119Sn) (D2O)
δ(117Sn) (solid)
1J(13C–119Sn)
–287
–274
850
–278
*
–298
–262 and –281
834
93
876
92
163°
154°
162°
2J(1H–119/117Sn)
C–Sn–C angle[a]
C–Sn–C angle[b]
C–Sn–C angle[c]
96
172°
152°
160°
166°
150°
158°
[a]
[b]
Calculated from the empirical relationships of Howard et al.:[31a] Θ (deg) ϭ 2.28 [2J(1H–119Sn)] – 46.4. –
Lockhart et al.:[31b]
Θ
(deg) ϭ 0.088 [1J(13C–119Sn)] ϩ 76.754. – Holecek et al.:[31c] Θ (deg) ϭ 0.1 [1J(13C–119Sn)] ϩ 74.675; * irreproducible, contaminated
by Et2SnO and other unidentified hydrolysis products.
[c]
with respect to the single one in solution at ca. δ ϭ 1. The six-coordinate tin atoms in solution. These CP-MAS NMR
assignment of the resonances in the 13C CP-MAS spectrum data, combined with the solution NMR data, suggest that
is made by comparison with the solution 13C chemical shifts the tin environments in the solid and solution states are sim-
of 5’-CMP/Et2SnCl2 at pH 2.2. In solution, 1H-13C HMBC ilar.
and HMQC[23,24,32] spectra allowed the assignment of the
The vibrational data (Table 5) account exclusively for the
1
13C and H resonances that is further supported by literat- existence of a tin–phosphate interaction, the ring vibration
ure data.[33]
frequencies not being affected by the addition of Et2SnCl2.
As suggested above for the variation of the pyrimidic 13C
chemical shifts, the shift of the ν(C(2)ϭO) frequency [∆ν ϭ
ν(5’-CMP/Et2SnCl2) – ν(5’-CMPNa2) ϭ 23 cm–1] can be
explained by different H-bonding networks.[34] The Raman
spectrum displays two characteristic bands at 434 and
568 cm–1 that were tentatively assigned to the ν˜(Sn–O–P)
and ν˜(Sn–O–Sn) vibration frequencies, respectively.[35–37]
On the basis of all previous data, four tin–phosphate
complexes (Figure 4) are conceivable for the solid-state
structure obtained from mixing 5’-CMP and Et2SnCl2. Un-
expectedly, elemental analysis data from samples of the
complex depend on the acid (HCl or HBr) used for the
adjustment of the pH of the supernatant from which this
solid precipitates; nevertheless, these data are self-consistent
and reproducible (Table 6). With HCl, the chlorine detected
originates not only from Et2SnCl2 but also, and even pre-
dominantly, from the excess HCl. However, chlorine is also
detected when HBr is used in excess. In the latter case, the
solid analysed contains two complexes, one only with brom-
ine (ca. 80%) and one only with chlorine (ca. 20%). Struc-
tures 6, 7 (X ϭ Cl, OH), as well as structure 9 (X ϭ OH)
(Figure 4) can easily be rejected on the basis of the chlorine
elemental analysis. Structure 8 (X ϭ Cl) is favoured over
structure 9 (X ϭ Cl) because the calculated carbon and tin
Table 4. Comparison of the solid and solution state 13C chemical
shifts at pH 2.2 for the mixtures 5’-CMP/Et2SnCl2 (2:1) and 5’-
UMP/Et2SnCl2 (2:1)
5’-CMP/Et2SnCl2
5’-UMP/Et2SnCl2
solid
163.9
154.0
141.7
98.9
solution
solid
163.1
152.2
139.4
102.3
100.2
93.1
solution
C4
C2
C6
C5
159.2
148.5
144.1
95.3
166.6
152.2
142.0
103.0
C1’
C4’
C2’
C3’
87.2
85.2
83.4
89.8
83.4
74.2
69.2
89.3
83.5
74.3
70.1
84.4
81.5
75.2
76.2
75.3
72.9
72.1
68.0
19.9
18.4
17.8
9.2
71.8
71.3
67.6
19.7
19.0
17.2
9.7
C5’
CH2
63.8
25.6
65.1
27.0
CH3
9.2
9.8
9.2
Table 4 shows that some of the solid state 13C resonances elemental data match the experimental data better for the
are duplicated, one is even triplicated; this is a result of the former than for the latter (Table 6). A thermogravimetric
nucleotide moieties being nonequivalent, possibly because analysis of the solid detects a weight loss of ca. 7% up to
of chirality, packing effects, and/or polymorphism. The 13C 170 °C, prior to a more substantial and continuous loss
chemical shifts of the tin ethyl CH2 group, the pyrimidic above 190 °C. The initial low weight loss is compatible with
and the C5’ carbons deviate most from those in the liquid the expected loss of one equivalent of HCl and two equiva-
state. For the pyrimidic carbons, this variation can be ex- lents of water, an observation which again matches struc-
plained by a different H-bonding network in the solid state, ture 8 better than 9 (X ϭ Cl). Structure 8 is proposed as a
as previously seen for several nucleotides by IR spectro- hydrochloric salt with a Sn–Cl–Sn bridge, which is known
scopy.[34] The 117Sn NMR CP-MAS spectrum exhibits a for various organotin oxides,[38,39] and a hydrogen bond be-
single isotropic shift at δ ϭ –274 in the range typical for tween the chloride anion and the oxygen bridging the two
516
Eur. J. Inorg. Chem. 2000, 513Ϫ522

Interaction between the Pyrimidic Nucleotides 5’-CMP, 5’-dCMP, and 5’-UMP and Diethyltin Dichloride
FULL PAPER
Table 5. IR data of free 5’-CMP and 5’-UMP as well as IR and Raman (*) data of the mixtures 5’-CMP/Et2SnCl2 and 5’-UMP/Et2SnCl2
in the solid state in KBr pellets (cm–1)
[a]
[a]
[a]
[a]
Tentative assignment
5’-CMPNa2
5’-CMP/Et2SnCl2
5’-UMPNa2
5’-UMP/Et2SnCl2
ν˜(C2ϭO)
ν˜(ring)
1653s
1540m
1491m
1110s
1082s
978s
1730s, 1676s
1540m
1491m
1121s
1683s[b]
1472m
1708s, 1664m[b]
1466m
ν˜(PO3)asym ϩ ν˜[CO(sugar)]
ν˜(PO3)sym
1126s, 1091s
976m
1120s, 1078m
1075s
995m
1009m, 1007m, 994m, 960w
961m
ν˜(C–Sn)asym*
ν˜(C–Sn)sym*
ν˜(Sn–O–P)*
–
–
–
–
548m
–
–
–
–
548m
511s
436w
568w
511s
434w
ν˜(Sn–O–Sn)*
568w
[a]
[b]
s ϭ strong, m ϭ medium, w ϭ weak, ν˜ ϭ stretching. – ν˜(C2ϭO) ϩ ν˜(C4ϭO).
Table 6. Experimental elemental analysis data for the solid com-
pounds isolated from 5’-CMP/Et2SnCl2 and 5’-UMP/Et2SnCl2
mixtures at pH 2.2 as well as calculated data for potential struc-
tures of Figures 4 and 5; the experimental data result from the
average over three samples prepared independently with HCl or
HBr; for 5’-UMP/Et2SnCl2, identical experimental results were ob-
tained with HCl and HBr
%C %H %N %Cl %Br %P %Sn
[a]
5’-CMP/Et2SnCl2[b] 28.4 4.7 7.5 3.1
–
5.6
5.3
21.2
20.2
5’-CMP/Et2SnCl2
6
27.6 4.5 7.2 0.7
27.0 4.2 7.3
28.3 4.6 7.6 6.4
29.2 4.9 7.9
5.0
–
–
10.7 20.5
7 (X ϭ Cl)
7 (X ϭ OH)
8 (X ϭ Cl)
8 (X ϭ Br)
8 (X ϭ OH)
9 (X ϭ Cl)
9 (X ϭ Br)
9 (X ϭ OH)
5’-UMP/Et2SnCl2
10
–
5.6
5.8
5.7
5.5
5.8
5.9
5.7
6.0
5.9
5.8
6.0
6.1
21.5
22.2
21.9
21.0
22.2
22.6
21.7
23.0
23.0
22.2
23.0
23.4
–
–
28.7 4.7 7.7 3.3
–
27.6 4.5 7.4
29.2 4.9 7.9
–
–
7.1
–
29.7 4.5 8.0 3.4
–
28.5 4.3 7.7
30.2 4.7 8.1
31.0 4.3 5.5
29.2 4.7 5.2
30.2 4.5 5.4
30.7 4.4 5.5
–
–
–
–
–
–
7.3
–
–
–
11
12
–
–
[a]
[b]
Prepared with HCl. – Prepared with HBr.
provide a useful 35Cl signal since it is smeared out in the
noise. The supernatant of the solid with proposed structure
8 falls into the latter category. Provided that 8 has identical
structures in the supernatant and the solid, this structure
implies that one chloride ion has been split off from the
Figure 4. Potential structures for the solid isolated from the mixture
5’-CMP/Et2SnCl2 at pH 2.2 and assessed against the data (X ϭ
OH, Cl, or Br)
tin atoms. When prepared from the HBr solution, the com- initial Et2SnCl2, while the second is covalently bound. The
pounds appear to be of type 8 with X being either Cl or Br. absence of any visible 35Cl resonance in a solution of 8 im-
As supported by similar 13C and 117/119Sn CP-MAS and plies the presence of covalently bound chlorine (as indicated
solution NMR data, the environments of the tin atom in by elemental analysis), necessarily being in fast exchange
solution and in the solid state are proposed to be similar with the ionic chloride which must also be present.
for 5’-CMP/Et2SnCl2. Thus, the structure of the complex in
Mixtures of Et2SnCl2 with the nonphosphorylated ana-
solution should be closely related to structure 8 (X ϭ Cl), logue of 5’-CMP, cytosine, also give rise to precipitates at
even if some differences between the solution and solid state pH 2–5; this occurs to an even larger extent than for 5’-
1
31P and 13C chemical shifts of the CH2 group suggest local CMP/Et2SnCl2. Indeed, the H NMR spectra of the super-
structure differences, possibly due to different H-bridge net- natants show that only small amounts of compounds con-
work or packing effects.
taining a diethyltin moiety are present in solution, as as-
The 35Cl solution NMR data of the supernatant shed sessed from a comparison of the integrations of the H6 and
some further, albeit indirect light on this proposal. It is Et2Sn protons. The precipitate was isolated; its 13C CP-
well-known that the 35Cl NMR signal of the chloride anion MAS NMR spectrum presents only 13C resonances from
is very sharp and displays a chemical shift around δ ϭ 0.[40] the diethyltin moieties and none from the nucleobase. The
In contrast, covalently bound chlorine generally fails to 117Sn CP-MAS NMR spectrum of the cytosine-derived
Eur. J. Inorg. Chem. 2000, 513Ϫ522
517

R. Willem et al.
FULL PAPER
Table 7. Experimental elemental analysis data for the solid com-
pound isolated from cytosine/Et2SnCl2 at pH 2.2 as well as calcu-
lated data for the proposed structure
%C
%H
%Cl
cytosine/Et2SnCl2
[(OH)Et2SnOSnEt2Cl]
23.1
22.8
5.0
5.0
8.4
8.4
sample displays two isotropic 117Sn chemical shifts of com-
parable intensities at δ ϭ –164 and δ ϭ –178. In agreement
with this, the elemental analysis data of this sample as well
as the two equally intense 117Sn resonances observed match
the formula [(OH)Et2SnOSnEt2Cl] (Table 7) that corre-
sponds to intermediate 3 in the hydrolysis scheme of Fig-
ure 2. In line with Otera’s finding on 3-chloro-1-hydroxy-
tetraorganodistannoxanes (R ϭ nBu), this structure should
be dimeric.[22b] In fact this compound, which is also soluble
in CDCl3, has a 117Sn NMR spectrum which displays the
same eight resonances as species A, B, and C, albeit with
different relative populations, as in the sample isolated by
precipitation at pH 2.8 from pure aqueous Et2SnCl2. There-
fore, it probably includes intermediates 2, 3 and 4. Since
essentially the same hydrolysis products are observed as
with pure Et2SnCl2, it can be concluded that at least no
peculiar interaction between the Et2Sn moiety and cytosine
occurs (Figure 2). In particular, this rules out any interac-
tion between the Et2Sn moiety and the nucleobase func-
tionalities.
Figure 5. Potential structures for the solid isolated from the mixture
5’-UMP/Et2SnCl2 at pH 2.2 and assessed against the data; in struc-
ture 12, the ethyl groups bonded to the tin atoms have been omitted
for clarity
A solid was also isolated from a mixture of 5’-UMP and
Et2SnCl2 at pH 2.2. The 31P CP-MAS NMR spectrum ex-
hibits two resonances at δ ϭ –7 and δ ϭ –12; they are
shifted to low frequency with respect to the single one in
solution at ca. 0.2 ppm. The 117Sn CP-MAS NMR spec-
trum exhibits two isotropic resonances of equal intensity at
δ ϭ –262 and δ ϭ –281 in the range corresponding to six-
coordinate tin atoms. The vibrational data of 5’-UMP/
Et2SnCl2 also unambiguously reflect a tin–phosphate inter-
action (Table 5). In contrast to 5’-CMP/Et2SnCl2, the ele-
mental analysis data of 5’-UMP/Et2SnCl2 (Table 6) are in-
dependent of the acid used for adjusting the pH of the solu-
tion from which the solid originates, no chlorine or bromine
The tendency of the 5’-dCMP/Et2SnCl2 mixtures to pre-
cipitate is lower than that of 5’-CMP and 5’-UMP, with
precipitation only occurring at higher pH’s. Moreover, no
reproducible elemental composition and solid state NMR
data could be obtained as the solid appears to contain hy-
drolysis products of Et2SnCl2 as well as free and complexed
5’-dCMP with δ(31P) ϭ 0.6 and –9.2.
Interaction of Diethyltin Dichloride with the Nucleotides at
pH Ͼ 9.0
At pH 9.0, the 1D 119Sn NMR spectrum of the mixture
being detected. These data are fairly consistent with struc- 5’-CMP/Et2SnCl2 (2:1) exhibits a single 119Sn resonance at
ture 10 and even better for structure 11 (Table 6 and Fig- δ ϭ –137 (Figure 4), characteristic of R2SnO2 compounds
ure 5). Unlike 8, the thermogravimetric analysis of this solid in the presence of a donor giving rise to a five-coordinate
does not indicate loss of H2O or HCl, supporting structure tin.[31,41] The 13C NMR spectrum exhibits a pair of signals
11 rather than 10. However, the two different isotropic for each carbon atom of the sugar unit, one 13C resonance
chemical shifts observed in the 117Sn CP-MAS spectrum are of each pair arising from free 5’-CMP. Upon addition of
in disagreement with both structures 10 and 11; therefore, one more equivalent of diethyltin dichloride (5’-CMP/
structure 12 can be proposed (Figure 5). It contains two Et2SnCl2 ϭ 1:1) the 13C resonances corresponding to free
different tin atoms and can be formed by dimerization of 5’-CMP disappear but the 1D 119Sn NMR spectrum dis-
structure 11 with elimination of two water molecules. plays a second very small 119Sn signal at δ ϭ –202. The
Among the three structures proposed, 12 agrees best with latter increases as the amount of Et2SnCl2 is further in-
the elemental analysis data (Table 6) and is the only one to creased. The mixture 5’-UMP/Et2SnCl2 displays similar be-
fit all the NMR data.
haviour; in contrast, the mixture 5’-dCMP/Et2SnCl2 in any
518
Eur. J. Inorg. Chem. 2000, 513Ϫ522

Interaction between the Pyrimidic Nucleotides 5’-CMP, 5’-dCMP, and 5’-UMP and Diethyltin Dichloride
FULL PAPER
Figure 6. Structure proposed for the complex generated in mixture
5’-CMP/Et2SnCl2 at pH 10.5
Figure 7. 15N NMR titration curve as obtained for the N3 nitrogen
atom of 15N-enriched 5’-CMP and 5’-CMP/Et2SnCl2
Table 8. Chemical shifts (ppm) and coupling constants (Hz) for the
three nucleotides studied at pH 10.5 in the presence of Et2SnCl2
δ(119Sn) 1J(13C–119/117Sn) 2J(1H–119/117Sn)
35Cl NMR spectrum of the mixture at pH 10.5 con-
5’-CMP/Et2SnCl2
–137
667
757
665
81
A
[a]
5’-dCMP/Et2SnCl2 –202
firms that no covalently bonded chlorine is involved, as
supported by the very narrow 35Cl resonance at δ ϭ 0[40]
which indicates that this element is only present as a chlor-
ide anion. The 119Sn resonance at δ ϭ –202 starts appearing
when the tin to nucleotide ratio exceeds 1:1 in a mixture of
Et2SnCl2 with substrates which contain a diolic sugar unit,
i.e. 5’-CMP, 5’-UMP, cytidine, and -ribose. All of these
5’-UMP/Et2SnCl2
–137
79
[a]
Nonobservable due to broad signal.
ratio exhibits a single 119Sn resonance at δ ϭ –202. Further- substrates display a resonance around δ ϭ –135 to –138,
more, the 119Sn nuclei associated with the two 119Sn reson- arising from the tin diolate complex being formed first. In
ances (δ ϭ –137 and –202) are in chemical exchange, as contrast, the δ ϭ –202 signal is the only resonance observed
demonstrated by the presence of intense cross-peaks in the for a mixture of Et2SnCl2 with substrates which do not con-
2D 119Sn EXSY spectrum[42,43] recorded from the mixture tain a diolic sugar unit, i.e. 5’-dCMP and cytosine. These
5’-CMP/Et2SnCl2. A 2D 1H-119Sn HMQC spectrum of this observations unambiguously rule out that the second 119Sn
mixture reveals no correlations between the 119Sn resonance resonance at δ ϭ –202 is due to any interaction between the
at δ ϭ –202 and the proton resonances of the 5’-CMP moi- tin moiety and functional groups of the pyrimidic unit, with
ety. In contrast, the 119Sn resonance at δ ϭ –137 displays or without a hydrogen bonding network. This is confirmed
1H-119Sn HMQC correlations with the H2’ and H3’ proton by 15N NMR data obtained from 15N enriched 5’-CMP/
resonances of the sugar unit, as well as satellites corres- Et2SnCl2 mixtures (2:1, 1:1, 1:2) which did not reveal any
ponding to a 2J(119Sn–O–117/119Sn) coupling constant of deviation of the 15N chemical shift of the N3, NH2 or N1
156 Hz.
atoms in the presence of diethyltin dichloride, with respect
1
3
These H-119Sn HMQC correlations result from J(1H– to free 5’-CMP. By way of an example, the titration curve
119Sn) couplings through the pathways H2’–C2’–O2’–Sn of the 15N3 resonance of such a mixture is shown in
and H3’–C3’–O3’–Sn and can be traced to the formation Figure 7.
of bonds between the tin atom and the O(2’) and O(3’) oxy-
As in the acidic medium, the 119Sn resonance at δ ϭ
gen atoms of the sugar unit (Figure 6). The 119Sn chemical –202 can only originate from some hydrolysed species on
shift of δ ϭ –137, the J(13C–119Sn) of 667 Hz (Table 8), the hydrolysis pathway of Et2SnCl2 (Figure 2). Further
1
and the 2J(119Sn–O–117/119Sn) of 156 Hz, combined with the evidence for this is the fact that the 119Sn NMR spectrum
1H-119Sn HMQC data, are consistent with a five-coordinate of the supernatant of the mixture obtained from Et2SnCl2
tin atom in a dimeric R2SnO2 diethyldioxastannolane as in D2O at pH 10.5 exhibits the very same resonance at δ ϭ
shown in Figure 6, one oxygen atom of the diol group of –202. By comparing the 117Sn CP-MAS NMR spectrum
the sugar unit being tricoordinated.[41] Similarly, dimeric di- with the one of pure Et2SnO (δ ϭ –166), the precipitate
butylstannylene acetals have 119Sn chemical shifts between that was isolated from this mixture was identified as
δ ϭ –115 and –150 in five-membered rings with pentacoor- Et2SnO (δ ϭ –165), the anisotropy patterns being identical.
dination at the tin atom (Table II of ref.[41b]).
As expected, the supernatant of this mixture exhibits a 35Cl
Eur. J. Inorg. Chem. 2000, 513Ϫ522
519

R. Willem et al.
FULL PAPER
resonance at a slightly higher chemical shift (δ ϭ 4) than solid was isolated and structurally characterised. Among
for completely anionic chloride (δ ϭ 0, NaCl). This can several possible structures, only one satisfied all data includ-
originate from an equilibrium between intermediate 3, with ing the elemental analysis. This structure contains two tin
one chlorine atom covalently bonded to the tin atom, and atoms bridged by one oxygen and one chlorine atom, each
intermediate 4, from which chlorine can only be released as tin atom being linked to the phosphate group of the nucle-
chloride. Compound 4 is the direct precursor of Et2SnO otide. Between pH 3.0 and 5.5 the absence of visible 119Sn
which is indeed the solid detected; it can be formed directly resonances precludes a conclusion.
by simple loss of a water molecule. Therefore, the species at
Around neutral pH (5.5–9.0), no evidence for the Et2Sn
δ ϭ –202 is most likely intermediate 4 in the hydrolysis moiety interacting with the nucleotide was found, Et2SnCl2
scheme of Et2SnCl2 (Figure 2).
giving rise to several hydrolysis products that are in mutual
chemical exchange with one another.
Interaction of Diethyltin Dichloride with 5’-CMP, 5’-dCMP
and 5’-UMP at pH 5.5–9.0
In a 1:1 nucleotide/Et2SnCl2 mixture at pH Ͼ 9.0,
Et2SnCl2 reacts with the O(2’) and O(3’) oxygen atoms of
the sugar unit of the nucleotide as demonstrated by the
presence of two HMQC cross-peaks H2’–Sn and H3’–Sn.
With an excess of Et2SnCl2, a second 119Sn resonance ap-
pears at δ ϭ –202, which, however, does not correlate with
any proton resonance of the nucleotides. This species is one
of the hydrolysis products of Et2SnCl2, as further indicated
by the absence of any 15N chemical shift effect on the pyri-
midic moieties that should have originated from the pres-
ence of a tin unit in the neighbourhood of nitrogen atoms.
The corresponding pH value limits for the interactions of
the purine nucleotides 5’-IMP and 5’-GMP with Et2SnCl2
were: pH Ͻ 4 for the interaction through the phosphate
group, pH Ͼ 9.5 for the interaction through the O(2’) and
O(3’) oxygen atoms, and no observed interaction for pH
4–9.5.[11b]
These results are in contrast to the mode of interaction
of cisplatin that interacts with the nitrogen atoms of purines
(N7 of guanosine) and pyrimidines (N3 of cytidine).[44]
However, they are similar to those reported for vanadium
complexes which have been shown to interact with the
phosphate group of mononucleotides at acidic pH, to hy-
drolyse at neutral pH and to react with the O(2’) and O(3’)
oxygen atoms of the sugar unit of mononucleotides at ba-
sic pH.[45]
Several 119Sn resonances appear in the 119Sn NMR spec-
tra of 5’-CMP/Et2SnCl2 in this pH range (Figure 3). These
resonances are all very broad, needing a long acquisition
time to obtain a reasonable signal-to-noise ratio. Due to the
1
broadness of the resonances, H-119Sn HMQC correlations
could not be detected between the proton resonances of the
5’-CMP moiety and the 119Sn resonance. Attempts to re-
duce the line width of the 119Sn resonances by recording a
119Sn spectrum at lower temperatures failed, the resonances
remaining still too broad. Likewise, 5’-CMP/Et2SnCl2 mix-
tures prepared in CD3OD/D2O (50:50) turned cloudy be-
low 273 K.
Similar 119Sn resonance frequencies, at least within ex-
perimental error, are found for all three nucleotides investig-
ated in the pH range 5.5–9.0. For 5’-dCMP/Et2SnCl2, the
119Sn resonance around δ ϭ –202 is observed at pH 6.5.
These findings again suggest that all the 119Sn resonances
arise from hydrolysis products of Et2SnCl2, rather than
from Et2SnCl2 interacting with the nucleotides. The broad
119Sn resonances are in agreement with the associated spe-
cies being in moderately slow exchange on the 119Sn NMR
time scale, since no evidence for multiple species was found
1
in the H and 13C spectra (this indicates that these species,
1
whatever they maybe, are in fast exchange on the H and
13C NMR time scales but in slow exchange on the 119Sn
NMR time scale). Unfortunately, 2D 119Sn EXSY experi-
ments which were aimed at supporting the presence of ex-
change were unsuccessful; this was due to broad linewidths
and/or a low signal-to-noise ratio.
The solid isolated from the mixture 5’-CMP/Et2SnCl2
(1:1) around pH 7.0 mainly reveals a single isotropic 117Sn
chemical shift at δ ϭ –167 in its CP-MAS spectrum. A com-
parison of its chemical shift anisotropy pattern with that of
a pure sample of Et2SnO reveals that this solid is the latter
compound. However, a CP-MAS 117Sn pattern from an-
other unidentified minor species with much lower intensity
(ca. 10%) is also observed at pH 7.0.
Experimental Section
Preparation of NMR Samples: The samples were typically prepared
by dissolving the nucleotides (0.1 mmol) in D2O (0.2 mL, 99.9% D,
Aldrich). A solution of diethyltin dichloride (0.05 mmol) in D2O
(0.3 mL) was added dropwise. This resulted in a clear solution with
a pH of approximately 6. The pH was adjusted by the dropwise
addition of deuterated hydrochloric acid or sodium hydroxide solu-
tions (0.1 ). Whenever generated, the precipitate was filtered and
thoroughly washed with water and diethyl ether prior to drying in
vacuum over P2O5 and further analysis (see Results and Discus-
sion). The elemental analyses were performed by the ‘‘Laboratoire
Central d’Analyse du CNRS’’, Vernaison, France. 15N enriched
(98% isotopic purity) 5’-CMP was purchased from CIL (MA).
Conclusions
NMR Experiments: All spectra were recorded at 303 K on a Bruker
AMX500 spectrometer equipped with a digital lock and operating
at 500.13, 202.46, 125.77, 186.50, 50.68, and 49.00 MHz for 1H,
31P, 13C, 119Sn, 15N, and 35Cl nuclei, respectively. 31P chemical shifts
were referenced to an aqueous solution of H3PO4 (85%). The 119Sn
From pH 2.0–3.0, the Et2Sn moiety from Et2SnCl2 is in-
volved in bonding with the phosphate group of the three
nucleotides studied. This interaction was characterised by a
31P-1H HOESY spectrum. At the same pH, the precipitated chemical shifts were referenced to Ξ ϭ 37.290665 MHz.[46a] The
520
Eur. J. Inorg. Chem. 2000, 513Ϫ522

Interaction between the Pyrimidic Nucleotides 5’-CMP, 5’-dCMP, and 5’-UMP and Diethyltin Dichloride
35Cl chemical shifts were referenced to a solution of NaCl (1 ,
mass spectra, and Prof. Dr. Ir. B. Van Mele and Dr. Ir. H. Rahier
FULL PAPER
δ ϭ 0). The 15N chemical shifts were referenced to Ξ ϭ for access to the IR, Raman and thermogravimetric analysis facilit-
10.136767 MHz.[46b]
ies. J. C. M. is a postdoctoral research associate of the Fund for
Scientific Research Flanders (Belgium). The financial support by
the ‘‘Bourse de la Fondation David et Alice Van Buuren’’ (L. G.),
the Belgian ‘‘Fonds voor Kollektief Fundamenteel Onderzoek’’
(FKFO Grant N° 2.0094.94, R. W., M. B.), the Belgian ‘‘Nationale
Loterij’’ (Grants N° 9.0006.93 and 9.0192.98, R. W., M. B.), the
Fund for Scientific Research Flanders (Belgium, Grant N°
G.0192.98, R. W, M. B.), and the Belgian ‘‘Nationaal Fonds voor
Wetenschappelijk Onderzoek’’ (NFWO Grant N° G.0054.96, M.
G.) is gratefully acknowledged.
Broad band 1H-decoupled 13C and 119Sn spectra were recorded us-
ing the Bruker pulse sequences with standard delays. The 1D and
1
2D H-119Sn HMQC experiments consisted of gradient enhanced-
versions of the standard HMQC pulse sequence processed in the
magnitude mode and implemented as described elsewhere.[32] The
2D 31P-1H HOESY[16,17] experiments were performed by recording
128 FIDs of 1 K data points, 144 scans each, with an acquisition
time of 0.2 s, a relaxation delay of 1.6 s and a mixing time of 1 s.
The 2D 1H-15N HMQC spectra were typically obtained by record-
ing 64 FIDs of 4 K data points, 32 scans each, with an acquisition
time of 0.5 s and a relaxation delay of 1.5 s. The 2D 119Sn EXSY
spectra were typically obtained by recording 24 FIDs of 4 K data
points, 3,200 scans each, with an acquisition time of 0.1 s, a relaxa-
tion delay of 1 s and a mixing time of 10 ms.
[1]
B. K. Keppler, Metal Complexes in Cancer Chemotherapy,
VCH, Weinheim, 1993.
[2]
A. T. M. Marcelis, J. Reedijk, Recl. Trav. Chim. Pays-Bas 1983,
102, 121–129.
[3]
S. J. Lippard, Science 1982, 218, 1075–1082.
[4]
M. Bloemink, J. Reedijk, Metal Ions in Biology (Eds.: A. Sigel,
All CP-MAS NMR spectra[47] were recorded on a Bruker AC250
and, in later stages, on a Bruker DRX250 spectrometer operating
at 101.26, 62.90, and 89.15 MHz for 31P, 13C, and 117Sn nuclei,
respectively. The spectrometer was equipped with a 7 or 4 mm
MAS broad-band probe. The matching conditions for Hartmann-
Hahn cross-polarization[47] (pulse length 5 µs) and the chemical
shift reference for the 117Sn nucleus were set with (cyclo-C6H11)4Sn
[δ ϭ–97.35 relative to (CH3)4Sn]. The 13C chemical shift reference
frequency was set with adamantane. The 117Sn spectra were typic-
ally obtained by acquiring 32 K data points over a spectral width
of 166.7 kHz, a 2 ms contact time and a relaxation delay of 2 s
with 10,000 to 20,000 scans. The isotropic chemical shifts were
identified by recording two spectra with sufficiently different spin-
ning rates.[47] The 13C CP-MAS spectra were obtained by acquiring
4 K data points over a spectral width of 17 kHz, a 2 ms contact
time and a relaxation delay of 4 s with 1,000 to 10,000 scans. The
1H-decoupled 31P MAS spectra were obtained by acquiring 4 K
data points over a spectral width of 20 kHz and a relaxation delay
of 4 s with 100 to 1,000 scans.
H. Sigel), M. Dekker, New-York 1996, 32, p. 641.
[5]
T. D. Tullius, S. J. Lippard, J. Am. Chem. Soc. 1981, 103,
4620–4622.
[6]
M. Gielen, Coord. Chem. Rev. 1996, 151, 41–51.
[7]
A. J. Crowe, Drugs of the Future 1987, 12, 255–275.
[8]
R. Barbieri, A. Silvestri, J. Inorg. Biochem. 1991, 41, 31–35.
[9]
V. Piro, F. Di Simone, G. Madonia, A. Silvestri, A. M. Giuli-
ani, G. Ruisi, R. Barbieri, Appl. Organomet. Chem. 1992, 6,
537–542.
[10]
R. Barbieri, A. Silvestri, A. M. Giuliani, V. Piro, F. Di Simone,
G. Madonia, J. Chem. Soc., Dalton Trans. 1992, 585–590.
[11] [11a] Q. Li, P. Yang, H. Wang, M. Guo, J. Inorg. Biochem. 1996,
[11b]
64, 181–195. –
Z. Yang, T. Bakas, A. Sanchez-Diaz, C.
Charalambopoulos, J. Tsangaris, N. Hadjiliadis, J. Inorg. Bi-
[11c]
ochem. 1998, 72, 133–140. –
A. Jancso, L. Nagy, E. Mol-
drheim, E. Sletten, J. Chem. Soc., Dalton Trans. 1999, 1587–
1594.
[12]
[13]
[14]
[15]
D. P. Bancroft, C. A. Lepre, S. J. Lippard, J. Am. Chem. Soc.
1990, 112, 6860–6871.
F. Kayser, M. Biesemans, M. Gielen, R. Willem, J. Magn. Re-
son. 1993, A102, 249–252.
J. C. Martins, P. Verheyden, F. Kayser, M. Gielen, R. Willem,
M. Biesemans, J. Magn. Reson. 1997, 124, 218–222.
F. Kayser, M. Biesemans, M. Gielen, R. Willem, Advanced Ap-
plications of NMR to Organometallic Chemistry (Eds.: M. Gi-
elen, R. Willem, B. Wrackmeyer), Wiley, Chichester, 1996,
chapter 3, 45–86.
C. Yu, G. C. Levy, J. Am. Chem. Soc. 1983, 105, 6994–6996.
C. Yu, G. C. Levy, J. Am. Chem. Soc. 1984, 106, 6533–6537.
F. Ribot, C. Sanchez, R. Willem, J. C. Martins, M. Biesemans,
Inorg. Chem. 1998, 37, 911–917.
K. C. Molloy, Chemistry of tin (Ed.: P. J. Smith), Blackie Aca-
demic Professional, Thomson Science, London, 1998, 2nd edi-
tion, p. 138.
R. C. Poller, J. Organomet. Chem. 1965, 3, 321–329.
A. G. Davies, Organotin Chemistry, VCH, Weinheim, 1997, p.
144.
Infrared Spectroscopy: Infrared spectra were acquired from KBr
pellets in the Fourier transform mode from a Perkin–Elmer System
2000 FT-IR spectrometer. The Raman spectra were acquired in the
Fourier-transform mode from a Perkin–Elmer System 2000 NIR
FT-RAMAN spectrometer using a Raman dpy2 beam with
310 mW power.
[16]
[17]
[18]
Electrospray Mass Spectrometry: The electrospray mass spectra
were recorded in the cationic and anionic modes on a Micromass
Quattro II instrument coupled to a Masslynx system (ionisation in
an electric field of 3.5 kV; source temperature 80 °C; source pres-
sure 1 atm.; analyser pressure 10–5 mbar).[48] Cone voltages were
15 V. Calculated and experimental isotopic distributions were com-
pared using the shareware program MassCluster v.2.1. for Apple
Macintosh. The m/z ratios reported in Table 1 and Figure 3 corre-
spond to the peak with the highest intensity of the isotopic distri-
bution patterns.
[19]
[20]
[21]
[22] [22a]
D. Dakternieks, K. Jurkschat, S. van Dreumel, E. R. T.
[22b]
Tiekink, Inorg. Chem. 1997, 36, 2023–2029. –
T. Yano, K.
Nakashima, J. Otera, R. Okawara, Organometallics 1985, 4,
1501–1503.
[23] [23a]
A. Bax, M. F. Summers, J. Magn. Reson. 1986, 67, 565–
[23b]
569. –
A. Bax, R. H. Griffey, B. L. Hawkins, J. Magn.
Reson. 1983, 55, 301–315.
A. Bax, M. F. Summers, J. Am. Chem. Soc. 1986, 108, 2093–
Thermogravimetric Analysis: The thermogravimetric analysis was
made on a Perkin–Elmer TGA-7 thermogravimetric analyser by
heating the sample from 40 to 500 °C at 10 °C/min under a nitro-
gen atmosphere.
[24]
[25]
2094.
ˆ
F. Kayser, M. Biesemans, M. Boualam, E. R. T. Tiekink, A. El
Khloufi, J. Meunier-Piret, A. Bouhdid, K. Jurkschat, M. Gi-
elen, R. Willem, Organometallics 1994, 13, 1098–1113; 4126
K. J. Barnham, C. J. Bauer, M. I. Djuran, M. A. Mazid, T.
Rau, P. J. Sadler, Inorg. Chem. 1995, 34, 2826–2832.
D. Neuhaus, M. Williamson, The Nuclear Overhauser Effect in
Structural and Conformational Ananlysis, VCH Publishers,
New-York, 1984.
[26]
[27]
Acknowledgments
We thank Mrs I. Verbruggen for recording some of the NMR spec-
tra, Dr. G. Laus and Mr. M. Desmet for recording the electrospray
Eur. J. Inorg. Chem. 2000, 513Ϫ522
521

R. Willem et al.
FULL PAPER
[28]
[40]
H. Santos, A. V. Xavier, C. F. G. C. Geraldes, Can. J. Chem.
J. W. Akitt, Multinuclear NMR (Ed.: J. Mason), Plenum Press,
New-York, 1987, p. 447.
1983, 61, 1456–1464.
T. P. Lockhart, W. F. Manders, E. M. Holt, J. Am. Chem. Soc.
1986, 108, 6611–6616.
[41] [41a]
[29]
T. B. Grindley, R. E. Wasylishen, R. Thangarasa, W. P.
Power, R. D. Curtis, Can. J. Chem. 1992, 70, 205–217. – [41b] T.
B. Grindley, Adv. Carbohydr. Chem. Biochem. 1998, 53, 17–142.
I. Pianet, E. Fouquet, M. Pereyre, M. Gielen, F. Kayser, M.
Biesemans, R. Willem, Magn. Reson. Chem. 1994, 32, 617–623.
R. Willem, Progr. NMR Spectrosc. 1987, 20, 1–94.
[30]
B. Wrackmeyer, Annu. Rep. NMR Spectrosc. 1985, 16, 71–184.
[42]
[31] [31a] W. H. Howard, R. W. Crecely, W. H. Nelson, Inorg. Chem.
[31b]
1985, 24, 2204–2208. –
T. P. Lockhart, W. F. Manders,
[43]
[44]
[31c]
Inorg. Chem. 1986, 25, 892–895. –
J. Holecek, A. Lycka,
B. Lippert, Progress in Inorganic Chemistry (Ed.: S. J. Lip-
pard), John Wiley and Sons, New-York, 1989, p. 1.
Inorg. Chim. Acta 1986, 118, L15–L16.
[32]
[33]
R. Willem, A. Bouhdid, F. Kayser, A. Delmotte, M. Gielen, J.
C. Martins, M. Biesemans, B. Mahieu, E. R. T. Tiekink, Or-
ganometallics 1996, 15, 1920–1929.
[45] [45a]
L. Y. Kuo, A. H. Liu, T. J. Marks, Metal Ions in Biology
(Eds.: A. Sigel, H. Sigel), M. Dekker, New-York 1996, 33, p.
641. –[45b] S. J. Angus-Dunne, R. J. Batchelor, A. S. Tracey, F.
[45c]
E. Breitmaier, W. Voelter, Carbon-13 NMR Spectroscopy, VCH
W. B. Einstein, J. Am. Chem. Soc. 1995, 117, 5292–5296. –
Weinheim, 1987.
G. Micera, A. Dessi, D. Sanna, Inorg. Chem. 1996, 35, 6349–
[34] [34a] H. A. Tajmir-Riahi, T. Theophanides, Can. J. Chem. 1985,
6352.
[34b]
[46] [46a]
63, 2065–2072. –
H. A. Tajmir-Riahi, T. Theophanides,
J. Mason, Multinuclear NMR (Ed.: J. Mason), Plenum
[34c]
[46b]
Can. J. Chem. 1983, 61, 1813–1822. –
T. Theophanides, Inorg. Chim. Acta 1983, 80, 183–190.
H. A. Tajmir-Riahi,
Press, New York, 1987, p. 627. –
D. S. Wishart, C. G.
Bigam, J. Yao, F. Abildgaard, H. J. Dyson, E. Oldfield, J. L.
Markley, B. D. Sykes, J. Biomol. NMR 1995, 6, 135–140.
F. A. Bovey, L. Jelinski, P. A. Mirau, Nuclear Magnetic Reson-
ance Spectroscopy, Academic Press, San Diego, 1987, 2nd edi-
tion, p. 399.
[35]
[36]
F. K. Butcher, W. Gerrard, E. F. Mooney, R. G. Rees, H. A.
Willis, A. Anderson, H. A. Gebbie, J. Organomet. Chem. 1964,
1, 431–434.
H. Schumann, P. Jutzi, A. Roth, P. Schwabe, E. Schauer, J.
Organomet. Chem. 1967, 10, 71–79.
D. H. Lohmann, J. Organomet. Chem. 1965, 4, 382–391.
B. Zobel, M. Schürmann, K. Jurkschat, Organometallics 1998,
17, 4096–4104.
[47]
[48] [48a]
G. Lawson, R. H. Dahm, N. Ostah, E. D. Woodland,
[48b]
Appl. Organomet. Chem. 1996, 10, 125–133. –
G. Lawson,
[37]
[38]
[48c]
N. Ostah, Appl. Organomet. Chem. 1994, 8, 525–532. –
W.
Henderson, B. K. Nicholson, L. J. McCaffrey, Polyhedron 1998,
17, 4291–4313.
Received August 9, 1999
[I99292]
[39]
D. Dakternieks, K. Jurkschat, D. Schollmeyer, H. Wu, Organo-
metallics 1994, 13, 4121–4123.
522
Eur. J. Inorg. Chem. 2000, 513Ϫ522