Solution structures of binary and ternary metal ion complexes of 9-(5-phosphonopentyl)adenine (3′-deoxa-PEEA). A nucleotide analogue related to the antivirally active 9-[2-(phosphonomethoxy)ethyl]adenine (PMEA)

Article abstract of DOI:10.1002/ejic.200300100

The acidity constants of the twofold protonated acyclic 9-(5phosphonopentyl)adenine, H₂(dPEEA)^±, as well as the stability constants of the M(H;dPEEA)⁺ and M(dPEEA) complexes with the metal ions M²⁺ = Mg²⁺, Ca²⁺, St ²⁺, Ba²⁺, Mn²⁺, Co²⁺, Ni ²⁺, Cu²⁺, Zn²⁺ of Cd²⁺, have been determined by potentiometric pH titrations in aqueous solution at I = 0,1 M (NaNO₃) and 25 °C, Application of previously determined straight-line plots of log K^M_M[R-PO₃) versus pK^H_H(R-PO₃) for simple phosph(on)ate ligands, R-PO^2-₃, where R represents a residue which cannot participate in the coordination process, proves that the primary binding site of dPEEA^2- is the phosphonate group with all the metal ions studied. However, for the M(dPEEA) systems with Co²⁺, Ni²⁺, Cu ²⁺, Zn²⁺ and Cd²⁺ a (in part rather small) stability increase is observed which is due to macrochelate formation with the adenine residue, i.e, most likely with N7. The formation degrees of the macrochelates are 17 ± 15%, 28 ± 10%, 46 ± 12%, 42 ± 28%, and 42 ± 9% (3σ), respectively. This means that in this respect dPEEA^2- resembles the parent nucleotide adenosine 5′monophosphate (AMP^2-) more than its chain-shortened analogue 9-(4 -phosphonobutyl)adenine (dPMEA^2-); indeed, in a first approximation macrochelate formation increases for a given metal ion within the series M(dPMEA) < M(dPEEA) < M(AMP). However, the coordinating properties of all three mentioned ligands differ significantly from those of the antivirally active 9-[2-(phosphonomethoxy) ethyl] adenine (PMEA^2-) which, due to the presence of an ether oxygen in the aliphatic side chain, has a different coordination chemistry which involves five-membered chelates with the ether oxygen atom. In addition, the stability constants of the mixed ligand complexes formed between Cu(Arm)²⁺, where Arm = 2, 2′-bipyridine (Bpy) or 1,10-phenanthroline (Phen), and the H(dPEEA) ^- or dPEEA^2- species were also measured. Detailed stability constant comparisons reveal that in the monoprotonated ternary Cu(Arm)(H;dPEEA)⁺ complexes the proton is at the phosphonate group and that stacking between Cu(Arm)²⁺ and H(dPEEA)^- plays a significant role. For the Cu(Arm)(dPEEA) complexes a large increase in complex stability (compared to the stability expected on the basis of the basicity of the phosphonate group) is observed, which is due to intramolecular stack formation between the aromatic ring systems of Phen or Bpy and the purine moiety of dPEEA^2-. The formation degree of the stacked isomers is in the order of 65 to 80%. Comparisons of the Cu(Arm)(PA) systems, where PA ^2- = dPEEA^2-, dPMEA^2- of AMP^2-, reveal that here dPEEA^2- resembles its parent AMP^2- less closely than dPMEA^2- does. The biological implications of these results, including antiviral activities, are shortly discussed. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejic.200300100

FULL PAPER
Solution Structures of Binary and Ternary Metal Ion Complexes of
9-(5-Phosphonopentyl)adenine (3Ј-deoxa-PEEA).
A Nucleotide Analogue Related to the Antivirally Active
9-[2-(Phosphonomethoxy)ethyl]adenine (PMEA)
[a]
[c]
[b]
´
´
´
Raquel B. Gomez-Coca, Antonın Holy, Rosario A. Vilaplana,
[b]
[a]
´
´
Francisco Gonzalez-Vılchez, and Helmut Sigel*
Keywords: Antiviral agents / Chelates / Equilibrium constants / Nucleotides / Stacking interactions
The acidity constants of the twofold protonated acyclic 9-(5-
antivirally active 9-[2-(phosphonomethoxy)ethyl]adenine
phosphonopentyl)adenine, H2(dPEEA) , as well as the (PMEA2−) which, due to the presence of an ether oxygen in
stability constants of the M(H;dPEEA)+ and M(dPEEA) com-
plexes with the metal ions M2+ = Mg2+, Ca2+, Sr2+, Ba2+
Mn2+, Co2+, Ni2+, Cu2+, Zn2+ or Cd2+, have been determined
the aliphatic side chain, has a different coordination chem-
istry which involves five-membered chelates with the ether
oxygen atom. In addition, the stability constants of the mixed
,
by potentiometric pH titrations in aqueous solution at I = 0.1 ligand complexes formed between Cu(Arm)2+, where Arm =
M
(NaNO3) and 25 °C. Application of previously determined
2,2Ј-bipyridine (Bpy) or 1,10-phenanthroline (Phen), and the
H(dPEEA)− or dPEEA2− species were also measured. Detailed
stability constant comparisons reveal that in the monoproton-
ated ternary Cu(Arm)(H;dPEEA)+ complexes the proton is at
straight-line plots of log KMM(R-PO ) versus pKHH(R-PO ) for simple
phosph(on)ate ligands, R-PO23−, 3where R represents a residue
3
which cannot participate in the coordination process, proves
that the primary binding site of dPEEA2− is the phosphonate the phosphonate group and that stacking between
group with all the metal ions studied. However, for the
Cu(Arm)2+ and H(dPEEA)− plays a significant role. For the
Cu(Arm)(dPEEA) complexes a large increase in complex
stability (compared to the stability expected on the basis of
the basicity of the phosphonate group) is observed, which is
M(dPEEA) systems with Co2+, Ni2+, Cu2+, Zn2+ and Cd2+
a
(in part rather small) stability increase is observed which is
due to macrochelate formation with the adenine residue, i.e.
most likely with N7. The formation degrees of the macrochel- due to intramolecular stack formation between the aromatic
ates are 17 15%, 28 10%, 46 12%, 42 28%, and 42
9% (3σ), respectively. This means that in this respect
dPEEA2− resembles the parent nucleotide adenosine 5Ј-
monophosphate (AMP2−) more than its chain-shortened ana-
ring systems of Phen or Bpy and the purine moiety of
dPEEA2−. The formation degree of the stacked isomers is in
the order of 65 to 80%. Comparisons of the Cu(Arm)(PA) sys-
tems, where PA2− = dPEEA2−, dPMEA2− or AMP2−, reveal that
logue 9-(4-phosphonobutyl)adenine (dPMEA2−); indeed, in a here dPEEA2− resembles its parent AMP2− less closely than
first approximation macrochelate formation increases for a
given metal ion within the series M(dPMEA) Ͻ M(dPEEA) Ͻ
M(AMP). However, the coordinating properties of all three
mentioned ligands differ significantly from those of the
dPMEA2− does. The biological implications of these results,
including antiviral activities, are shortly discussed.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
1. Introduction
to their genetic material into two major groups: RNA and
DNA viruses. The human hepatitis A virus (HAV) and po-
lioviruses belong to the first group, whereas the hepatitis B
virus (HBV) is classified as a DNA-containing organism.[1]
Viruses and the diseases they may cause have been stud-
ied for more than 100 years. Viruses are classified according
Within the RNA-virus group special attention has been
paid to the Retroviridae family to which the human
immunodeficiency viruses (HIV-1 and HIV-2) belong.[2] It
is well-known today[3] that, at least in theory, the viral repli-
cation cycle may be interrupted in many ways. One of the
possibilities is to employ acyclic nucleotide analogues[4] and
in this category 9-[2-(phosphonomethoxy)ethyl]adenine
(PMEA; Figure 1),[5Ϫ7] also known as Adefovir,[8] appears
to be especially promising. This compound may be con-
[a]
Department of Chemistry, Inorganic Chemistry, University of
Basel, Spitalstrasse 51,
CH-4056 Basel, Switzerland
Fax: (internat.) ϩ41-61/267-1017
E-mail: Helmut.Sigel@unibas.ch
[b]
Inorganic Chemistry Department, Faculty of Chemistry,
University of Seville,
E-41071 Seville, Spain
[c]
Institute of Organic Chemistry and Biochemistry, Academy of
Sciences,
CZ-16610 Prague, Czech Republic
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947
DOI: 10.1002/ejic.200300100
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2937

´
´
´
´
R. B. Gomez-Coca, A. Holy, R. A. Vilaplana, F. Gonzalez-Vılchez, H. Sigel
FULL PAPER
chain of PMEApp4Ϫ, facilitate the correct positioning of
the two metal ions needed in the reaction and thus make it
initially an even better substrate[16,17] for DNA polymerases
than dATP4Ϫ [14,15]
.
Indeed, PMEA forms with divalent metal ions (M2ϩ),
like Mg2ϩ, Mn2ϩ or Zn2ϩ, five-membered chelates involv-
ing the ether oxygen as expressed in the intramolecular
Equilibrium (1):
Replacement of the ether oxygen by a CH2 unit gives
9-(4-phosphonobutyl)adenine, also termed 3Ј-deoxa-PMEA
(dPMEA; Figure 1).[18] This compound shows no biological
activity,[16] in agreement with the suggestions summarized
above, and it also has a coordination pattern[18] different
from that of PMEA2Ϫ [18Ϫ21]
to that of AMP2Ϫ [20,22,23]
This means, phosph(on)ate-coor-
,
yet one which in fact is similar
.
dinated metal ions may form macrochelates by interacting
with N7 of the adenine residue; this is expressed in the in-
tramolecular Equilibrium (2):[24Ϫ26]
Having a broad interest in the metal ion-binding proper-
ties of nucleotide analogues,[27Ϫ30] we decided to study 9-
(5-phosphonopentyl)adenine, which is also known as 3Ј-de-
oxa-PEEA (dPEEA; Figure 1), since it is the deoxa ana-
logue of 9-[2-(2-phosphonoethoxy)ethyl]adenine.[31] As ex-
pected, dPEEA does not show a useful biological activity
but its additional CH2 group enabled the investigation of
the effect that the lengthening of the side chain of dPMEA
has on the metal ion-binding properties. Thus, we measured
the stability constants of the binary complexes formed be-
tween dPEEA and the alkaline earth ions, the divalent ions
of the second half of the 3d series as well as Zn2ϩ and
Cd2ϩ. We compared the results regarding the position of
Equilibrium (2) with those obtained previously for the
M(dPMEA) and the M(AMP) complexes.[18,23]
Figure 1. Chemical structures of the dianions of 9-[2-(phosphono-
methoxy)ethyl]adenine (ϭ PMEA2Ϫ ϭ Adefovir), 9-(4-phosphono-
butyl)adenine (ϭ dPMEA2Ϫ ϭ 3Ј-deoxa-PMEA2Ϫ), 9-(5-phos-
phonopentyl)adenine (ϭ dPEEA2Ϫ ϭ 3Ј-deoxa-PEEA2Ϫ) and of
their parent nucleotide adenosine 5Ј-monophosphate (AMP2Ϫ).
The latter is shown in its anti conformation.
sidered as an analogue of (2Ј-deoxy)adenosine 5Ј-mono-
phosphate [(d)AMP2Ϫ; Figure 1]; it is active against
HBV and HIV.[8,9] Indeed, its bis(pivaloyloxymethyl)ester
(Adefovir dipivoxil) was very recently approved by the US
Food and Drug Administration (FDA) for use in hepatitis
B therapy.[10]
Since PMEA has a phosphonate group instead of a phos-
phate unit, it circumvents in blood plasma or in its passage
through the cellular membrane the enzyme-catalyzed de-
phosphorylation that is suffered by nucleotide analogues
which contain a monophosphate ester residue.[8,11] Instead,
PMEA is diphosphorylated to give the triphosphate-like
form PMEApp which is responsible for the therapeutic ef-
fect.[8,12] PMEApp is accepted as a substrate by DNA poly-
merases[13] and after incorporation of the nucleotidyl-like
unit into the growing nucleic acid chain leads to its termin-
ation due to the lack of a 3Ј-hydroxyl group. It has been
suggested[14,15] that the higher basicity of the phosphonyl
group compared to that of a phosphoryl one, together with
the presence of the ether-oxygen atom in the aliphatic side
Since weak interactions[32Ϫ34] like the aromatic-ring
stacking are important for the anchoring process of a sub-
strate in the active site cavity of an enzyme,[35,36] we had
previously quantified the extent of intramolecular stacking
according to Equilibrium (3)
2938
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947

Binary and Ternary Metal Ion Complexes of 9-(5-Phosphonopentyl)adenine
FULL PAPER
for the ternary complexes Cu(Arm)(AMP/dPMEA),[37,38]
Comparison of the values listed in Table 1 (entries 1 and
where Arm ϭ 2,2Ј-bipyridine (Bpy) or 1,10- phenanthroline 4Ϫ7) reveal that the first proton of H2(dPEEA)Ϯ is from
(Phen). Now the corresponding Cu(Arm)(dPEEA) com- the (N1)Hϩ site of the adenine residue and the second one
plexes were studied leading to the interesting result that from the monoprotonated phosphonate group (see entries
lengthening of the aliphatic chain by a CH2 unit diminishes 2Ϫ7). However, more comparisons are possible; a few are
intramolecular stacking [Equilibrium (3)], whereas in the given:
binary M(dPEEA) species macrochelate formation [Equi-
librium (2)] is favored.
(i) From the comparison of entries 3 and 4 with those of
2 and 5Ϫ7 it follows that protonated phosphonate groups
are less acidic than the corresponding phosphate groups;
this agrees with the lower electronegativity of a C atom
compared to an O atom.
(ii) The ether oxygen, due to its electron-withdrawing ef-
fect, makes the monoprotonated phosphonate species of
PMEA2Ϫ more acidic than those of dPMEA2Ϫ or
dPEEA2Ϫ (entries 5Ϫ7).
2. Results and Discussion
2.1 Acidity Constants of H2(dPEEA)؎
From the structure of dPEEA2Ϫ (see Figure 1) it is evi-
dent that this species can accept three protons, two at the
phosphonate group and one at the N1 site of the adenine
residue.[39] Further protonations at an adenine residue are
possible at N7 and N3, but these protons are released with
pKa Ͻ 0;[39Ϫ42] similarly, release of the first proton from a
twofold protonated phosphonate residue occurs with a pKa
of about 2.[18,43] Hence, in the present study, for which all
potentiometric pH titrations were carried out at pH Ͼ 4.0,
only the following two deprotonation reactions, in which
dPEEA2Ϫ is abbreviated as PA2Ϫ, need to be considered:
(iii) In fact, the acid-base properties of H2(dPMEA)Ϯ
and of H2(dPEEA)Ϯ are very similar, indicating that the
additional CH2 unit in PEEA has little effect (entries 6, 7).
Finally, it needs to be pointed out that due to the differ-
ent affinities of the phosph(on)ate groups for protons, the
formation degree of the free ligands, PA2Ϫ, differs consider-
ably in the physiological pH range around 7.5: Under these
conditions, for example, AMP2Ϫ is formed to about 95%
whereas dPEEA2Ϫ occurs only to about 36%.
2.2 Stability Constants of the M(H;dPEEA)؉ and
M(dPEEA) Complexes
The experimental data of the potentiometric pH ti-
trations (see Section 4.3) allow the determination of the sta-
bility constants defined by Equilibria (6a) and (7a):
Indeed, all the experimental data from the pH titrations
in aqueous solution (25 °C; I ϭ 0.1 , NaNO3) could be
excellently fitted by taking into account Equilibria (4) and
(5). The acidity constants obtained in the present study for
H2(dPEEA)Ϯ are given in Table 1 together with some re-
lated data.[44Ϫ46]
Together with Equilibria (4a) and (5a) they are sufficient
to obtain excellent fitting of the titration data provided the
evaluation is not carried into the pH range where formation
of hydroxo species occurs, which was evident from titrations
without ligand. Of course, Equilibria (6a) and (7a) are also
connected via Equilibrium (8a)
Table 1. Negative logarithms of the acidity constants of
H2(dPEEA)Ϯ [Equations (4) and (5)] together with the correspond-
ing values of some related systems in aqueous solution at 25 °C
and I ϭ 0.1  (NaNO3)[a][b]
Protonated
species
pKHH (PA)
pKHH(PA)
2
No.
(N1)Hϩ
P(O)2(OH)Ϫ
Ref.
[41]
[43]
[44]
[22]
[45]
[18]
[c]
1
2
3
4
5
6
7
H(9MeAde)ϩ
CH3P(O)(OH)2
CH3OP(O)(OH)2
H2(AMP)Ϯ
4.10 Ϯ 0.01
Ϫ
Ϫ
3.84 Ϯ 0.02
4.16 Ϯ 0.02
4.17 Ϯ 0.02
4.17 Ϯ 0.01
Ϫ
7.51 Ϯ 0.01
6.36 Ϯ 0.01
6.21 Ϯ 0.01
6.90 Ϯ 0.01
7.69 Ϯ 0.01
7.75 Ϯ 0.01
and the corresponding acidity constant [Equation (8b)] may
be calculated with Equation (9):[47]
H2(PMEA)Ϯ
H2(dPMEA)Ϯ
H2(dPEEA)Ϯ
The results are listed in Table 2; the stability constants
given for the M(H;dPEEA)ϩ complexes are in part esti-
mates since the formation degree of these species was low
(see Section 4.3). The stability constants of the M(dPEEA)
[a]
The error limits given are three times the standard error of the
mean value or the sum of the probable systematic errors, whichever
is larger.
listed (see
[b]
So-called practical, mixed or Brønsted constants are
[46]
[c]
for details). This work.
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2939

´
´
´
´
R. B. Gomez-Coca, A. Holy, R. A. Vilaplana, F. Gonzalez-Vılchez, H. Sigel
FULL PAPER
complexes show the usual trends. For the alkaline earth H(dPMEA)Ϫ or H(PMEA)Ϫ are also identical (see Table 1;
ions the stability of the complexes decreases with increasing column 3, entries 5Ϫ7) and that evidence has been provided
ionic radii indicating that metal ion binding at the phos- for the M(H;PMEA)ϩ complexes[20,21,45] that the metal ion
phonate group is (at least) in part inner-sphere. For the di- is mainly located at the nucleobase residue, one may not
valent 3d metal ions the long-standing experience[48] is con- only conclude that in the M(H;dPEEA)ϩ complexes the
firmed that the stabilities of phosph(on)ate-metal ion com- proton is at the phosphonate group, but also that the metal
plexes often do not strictly follow[18,20,27,28,44,45,49] the ion is mainly at the adenine residue. The N1 versus N7 di-
IrvingϪWilliams sequence,[50] an observation in agreement chotomy for metal ion binding to the adenine residue is
with the fact that in ligands of this kind the phosph(on)ate well-known[55] though there are indications that binding to
group is always the main stability-determining binding N7 dominates.[55,56] In any case, the fact that the stabilities
site.[19,22,24,25,51]
of
the
M(H;dPEEA)ϩ
complexes
follow
the
IrvingϪWilliams sequence[50] [in contrast to phosph(on)ate
coordinations] also supports[48a] the above conclusion that
metal ion binding in the monoprotonated species occurs
preferably to a nitrogen atom.
Table 2. Logarithms of the stability constants of the
M(H;dPEEA)ϩ [Equation (6b)] and M(dPEEA) complexes [Equa-
tion (7b)], together with the negative logarithms of the acidity con-
stants of the protonated complexes [Equations (8b) and (9)] in
aqueous solution at 25 °C and I ϭ 0.1  (NaNO3)[a]
M2ϩ
log KMM(H;dPEEA)
log KMM(dPEEA)
pKMH (H;dPEEA)
Mg2ϩ
Ca2ϩ
Sr2ϩ
0.4 Ϯ 0.3[b]
0.2 Ϯ 0.4[b]
0.1 Ϯ 0.4[b]
0.0 Ϯ 0.4[b]
0.69 Ϯ 0.18
0.86 Ϯ 0.21
1.00 Ϯ 0.15
1.92 Ϯ 0.12
1.47 Ϯ 0.22
1.38 Ϯ 0.15
1.87 Ϯ 0.05
1.59 Ϯ 0.06
1.31 Ϯ 0.07
1.25 Ϯ 0.06
2.55 Ϯ 0.03
2.36 Ϯ 0.05
2.46 Ϯ 0.03
3.86 Ϯ 0.08
2.9 Ϯ 0.2[c]
3.19 Ϯ 0.05
6.28 Ϯ 0.30
6.36 Ϯ 0.40
6.54 Ϯ 0.41
6.50 Ϯ 0.40
5.89 Ϯ 0.18
6.25 Ϯ 0.22
6.29 Ϯ 0.15
5.81 Ϯ 0.15
6.32 Ϯ 0.30
5.94 Ϯ 0.16
2.3 Evaluation of the Stabilities of the M(dPEEA)
Complexes
Ba2ϩ
Mn2ϩ
Co2ϩ
Ni2ϩ
Cu2ϩ
Zn2ϩ
Cd2ϩ
For the M(dPEEA) complexes the question arises: Does
the adenine residue also participate in metal ion binding
next to the phosphonate group? Should such an additional
interaction with the adenine residue occur, then it has to be
reflected in an increased complex stability.[57] Hence, it is
necessary to define the stability of a pure -PO32Ϫ/M2ϩ inter-
action. This can be done by applying the previously de-
fined[45] straight-line correlations which are based on log
KMM(R-PO ) versus pKHH(R-PO ) plots for simple phosphate
[a]
For the error limits see footnote [a] of Table 1. The error limits
(3σ) of the derived data, in the present case for column 4, were
[b]
calculated according to the error propagation after Gauss.
The
constants listed for these M(H;dPEEA)ϩ complexes are estimates
3
3
monoesters[58] and phosphonates;[45] these ligands are ab-
breviated as R-PO32Ϫ, where R represents a non-coordinat-
ing residue. The parameters for the corresponding straight-
line equations, which are defined by Equation (10),
(see Section 4.3). The experiments with Zn2ϩ were significantly
[c]
hampered by precipitation; i.e., the pH range accessible for the
evaluation of the constants was severely restricted (see Section 4.3).
As far as the M(H;dPEEA)ϩ complexes are concerned,
it is evident that the evaluation of potentiometric pH ti-
tration data only allows the determination of their stability have been tabulated,[19a,25,45,51] i.e., the slopes m and the
constants. Further information is required to detect the intercepts b with the y-axis. Hence, with a known pKa value
binding sites of the proton and the metal ion. At first one for the deprotonation of a -P(O)2(OH)Ϫ group an expected
may ask where the proton is located because binding of stability constant can be calculated for any phosph(on)ate-
a metal ion to a protonated ligand commonly leads to an metal ion complex.
acidification of the ligand-bound proton.[52,53] Indeed, the
The plots of log KMM(R-PO ) versus pKHH(R-PO ) according to
3
acidity constants of the M(H;dPEEA)ϩ complexes given in Equation (10) are shown in Figure 2 for the 31:1 complexes
column 4 of Table 2 are by about 1.2 to 1.9 log units smaller of Ca2ϩ, Co2ϩ, Cd2ϩ and Cu2ϩ as examples, with the data
than pKHH(dPEEA)
.
This shows that the proton in points (empty circles) of the eight simple ligand systems
M(H;dPEEA)ϩ is bound to the phosphonate group, hence, used[45] for the determination of the straight base lines. The
one may tentatively assume that the metal ion is bound four solid circles refer to the corresponding M(dPEEA)
preferentially to the nucleobase, since a monoprotonated complexes; those for the Cd2ϩ and Cu2ϩ species, and poss-
phosphonate group is only a weak binding site. Indeed, this ibly also for the Co2ϩ ones, are somewhat above their refer-
suggestion agrees with evidence obtained previously for ence lines, thus proving an increased stability for these com-
other related M(H;PA)ϩ species.[20,21,45,54]
Furthermore, the stability constants
plexes, whereas the data point for the Ca2ϩ complex fits
the within the error limits on the line. Similar properties are
of
M(H;dPEEA)ϩ complexes are, within the error limits, observed for the corresponding M(dPMEA) and M(AMP)
identical to the values determined[45] for the corresponding complexes (crossed circles). In contrast, all of the data
M(H;PMEA)ϩ and M(H;dPMEA)ϩ species.[18] Consider- points[45] of the M(PMEA) complexes (solid squares) are
ing that the basicity of the N1 sites in H(dPEEA)Ϫ and clearly above their reference lines indicating that all these
2940
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947

Binary and Ternary Metal Ion Complexes of 9-(5-Phosphonopentyl)adenine
FULL PAPER
species are more stable than is expected on the basis of the
basicity of the phosphonate group of PMEA2Ϫ
Naturally, in the complexes of dPEEA2Ϫ, dPMEA2Ϫ and
AMP2Ϫ, whose side chains or sugar residue is devoid of an
ether oxygen atom (Figure 1), Equilibrium (1) cannot exist
and therefore, any possibly observed stability increase has
to be attributed to an interaction with the adenine residue.
Stability enhancements, like those seen in Figure 2, can be
quantified by the differences between the experimentally
(exptl) measured stability constants and those calculated
(calcd) according to Equation (10); this difference is defined
in Equation (11),
.
where the expressions log KMM(PA)calcd and log KMM(PA)op are
synonymous because the calculated value equals the sta-
bility constant of the ‘‘open’’ isomer, M(PA)op [see, for ex-
ample, Equilibria (1) and (2)], in which only a -PO32Ϫ/M2ϩ
interaction occurs. In columns 2Ϫ4 of Table 3 the values
for the three terms of Equation (11) are listed. The log
∆
values for Ni(dPEEA), Cu(dPEEA) and
M/dPEEA
Cd(dPEEA) are clearly positive; the same is most likely true
for those due to Co(dPEEA) and Zn(dPEEA), especially if
one takes into account that the error limits refer to three
times the standard error. Hence, in all these instances Equi-
librium (2) needs to be considered.
Table 3. Stability constant comparisons for the M(dPEEA) com-
plexes between the measured stability constants (exptl; Table 2, col-
umn 3) and the calculated stability constants (calcd) based on the
Figure 2. Evidence for an enhanced stability of the Cd(dPEEA)
and Cu(dPEEA) and possibly also the Co(dPEEA) complexes, and
for the lack of such an enhanced stability of the Ca(dPEEA) species
(᭹), together with the data points for the corresponding metal ion
complexes of dPMEA2Ϫ, AMP2Ϫ (ᮏ) and PMEA2Ϫ (᭜) based on
the relationship between log KMM(R-PO and pKHH(R-PO for M(R-
basicity of the phosphonate group in dPEEA2Ϫ (pKHH(dPEEA)
ϭ
7.75; Table 1) and the baseline equations established pre-
viously[25,45,51] (see Equation (10) and Figure 2), together with the
stability differences log ∆M/dPEEA, as defined by Equation (11). The
previously determined stability differences, log ∆M/dPMEA, for
M(dPMEA) complexes[18] are given for comparison (aqueous solu-
tion; 25 °C; I ϭ 0.1 , NaNO3)[a]
)
)
3
3
PO3) complexes of some simple phosphate monoester and phos-
phonate ligands (R-PO32Ϫ) (᭺): 4-nitrophenyl phosphate (NPhP2Ϫ),
phenyl phosphate (PhP2Ϫ), uridine 5Ј-monophosphate (UMP2Ϫ),
-ribose 5-monophosphate (RibMP2Ϫ), thymidine [ϭ 1-(2-deoxy-
β--ribofuranosyl)thymine] 5Ј-monophosphate (dTMP2Ϫ), n-butyl
phosphate (BuP2Ϫ), methanephosphonate (MeP2Ϫ), and ethane-
phosphonate (EtP2Ϫ) (from left to right). The least-squares lines
[Equation (10)] are drawn through the corresponding 8 data sets
M2ϩ
log KMM(dPEEA)
exptl[b]
calcd
log ∆M/dPEEA log ∆M/dPMEA
[58]
[45]
(᭺) taken from
for the phosphate monoesters and from
for
Mg2ϩ 1.87 Ϯ 0.05 1.88 Ϯ 0.03 Ϫ0.01 Ϯ 0.06 Ϫ0.03 Ϯ 0.05
Ca2ϩ 1.59 Ϯ 0.06 1.65 Ϯ 0.05 Ϫ0.06 Ϯ 0.08 Ϫ0.07 Ϯ 0.07
the phosphonates. The points due to the equilibrium constants for
the M2ϩ/dPEEA systems (᭹) are based on the values listed in
Tables 1 and 2; those for the M2ϩ/dPMEA (ᮏ), M2ϩ/AMP (ᮏ),
Sr2ϩ
1.31 Ϯ 0.07 1.37 Ϯ 0.04 Ϫ0.06 Ϯ 0.08 Ϫ0.06 Ϯ 0.06
and M2ϩ/PMEA systems (᭜) are from
and [45], respectively.
[18,23]
Ba2ϩ 1.25 Ϯ 0.06 1.30 Ϯ 0.04 Ϫ0.05 Ϯ 0.07 Ϫ0.07 Ϯ 0.07
The vertical broken lines emphasize the stability differences from
the reference lines; they equal log ∆M/PA as defined in Equation
(11). All the plotted equilibrium constants refer to aqueous solu-
tions at 25 °C and I ϭ 0.1  (NaNO3).
Mn2ϩ 2.55 Ϯ 0.03 2.53 Ϯ 0.05
Co2ϩ 2.36 Ϯ 0.05 2.28 Ϯ 0.06
Ni2ϩ 2.46 Ϯ 0.03 2.32 Ϯ 0.05
Cu2ϩ 3.86 Ϯ 0.08 3.59 Ϯ 0.06
0.02 Ϯ 0.06 Ϫ0.04 Ϯ 0.06
0.08 Ϯ 0.08
0.14 Ϯ 0.06
0.27 Ϯ 0.10
0.24 Ϯ 0.21
0.24 Ϯ 0.07
0.04 Ϯ 0.08
0.10 Ϯ 0.08
0.16 Ϯ 0.09
0.1 Ϯ 0.2
0.15 Ϯ 0.11
Zn2ϩ 2.9 Ϯ 0.2
2.66 Ϯ 0.06
Cd2ϩ 3.19 Ϯ 0.05 2.95 Ϯ 0.05
[a]
[b]
The increased stability of the M(PMEA) complexes has
been proven[45] to be largely due to the involvement of the
ether oxygen atom in metal ion binding (see Figure 1) which
gives rise to the formation of five-membered chelates and
thus to the intramolecular Equilibrium (1).[15,19,45] In cer-
Regarding the error limits (3σ) see footnote [a] of Table 2.
These values are from column 3 of Table 2.
A comparison of the log ∆M/PA values given in columns
tain systems, like Ni(PMEA) and Cu(PMEA), a third iso- 4 and 5 of Table 3 for the M(dPEEA) and M(dPMEA)
mer exists which involves, in addition, N3 of the adenine complexes, respectively, reveals that their properties are
residue[15,19Ϫ21,45] as well as a fourth one involving N7 quite alike with a slightly more enhanced stability of some
which, however, occurs only to a minor extent.[18]
of the M(dPEEA) species.
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2941

´
´
´
´
R. B. Gomez-Coca, A. Holy, R. A. Vilaplana, F. Gonzalez-Vılchez, H. Sigel
FULL PAPER
2.4 Extent of Macrochelate Formation in M(dPEEA)
Systems and Comparison with That Occurring in
M(dPMEA) and M(AMP) Species
ion with N7 because N7 is known to be considerably more
basic than N3.[39Ϫ42] Furthermore, macrochelate formation
in the M(AMP) complexes occurs with certainty via
N7;[22,59] hence, it appears that M(dPEEA) and
The position of the concentration-independent Equilib-
rium (2) between the simple phosphonate-bound species,
which we designate as the ‘‘open’’ isomer, M(dPEEA)op,
and the macrochelated isomer involving the adenine resi-
due, designated as the ‘‘closed’’ species, M(dPEEA)cl, is de-
fined by the intramolecular, and hence dimensionless, equi-
librium constant KI given in Equation (12):
M(dPMEA) resemble in this respect their parent complexes,
i.e., M(AMP), more than the M(PMEA) species which con-
tain the antivirally active ligand PMEA2Ϫ and involve the
ether oxygen atom in metal ion binding (see Section 1).
With the above evidence for macrochelate formation in-
volving N7 in mind, it is interesting to compare the proper-
ties of the dPEEA2Ϫ, dPMEA2Ϫ and AMP2Ϫ ligands (Fig-
ure 1). From the results given in columns 4Ϫ6 of Table 4 it
follows that the extent of macrochelate formation increases
within the series M(dPMEA) Ͻ M(dPEEA) Ͻ M(AMP).
The changes are not dramatic, but the trend within the
series seems to be clear for each metal ion. A further inter-
esting observation is that for Cu2ϩ, Zn2ϩ and Cd2ϩ the for-
mation degrees of the M(dPEEA)cl and M(AMP)cl species
are quite alike. Hence, it seems that dPEEA2Ϫ mimics
AMP2Ϫ even better than dPMEA2Ϫ; most likely this is due
to the by one CH2 unit longer side chain of dPEEA2Ϫ, com-
pared to the one of dPMEA2Ϫ, and that this allows the
phosphonate-bound M2ϩ to reach N7 with less strain.
Values for KI can be calculated by known pro-
cedures,[22,45,57] i.e., via Equation (13):
Knowledge of KI allows then to obtain the percentage of
the closed isomer, M(dPEEA)cl, according to Equation
(14):
2.5 Stability of the Mixed Ligand Cu(Arm)(H;dPEEA)؉
and Cu(Arm)(dPEEA) Complexes
The results are summarized in Table 4 for the M(dPEEA)
complexes of the 3d ions including Zn2ϩ and Cd2ϩ. For the
complexes of the alkaline earth ions the formation degree
For the ternary complexes composed of H(dPEEA)Ϫ or
of the closed species in Equilibrium (2) is zero within the dPEEA2Ϫ, Cu2ϩ and a heteroaromatic amine (Arm), i.e.,
error limits as follows from the log ∆M/dPEEA values listed 2,2Ј-bipyridine (Bpy) or 1,10-phenanthroline (Phen), the
in column 4 of Table 3.
same evaluation procedure holds as described in the first
From the results listed in columns 2Ϫ4 of Table 4 it fol- paragraph of Section 2.2 for the binary complexes because
lows that at least for Ni2ϩ, Cu2ϩ, Zn2ϩ and Cd2ϩ, which complex formation of Cu(Bpy)2ϩ and Cu(Phen)2ϩ, due to
have the most pronounced affinity for N sites among the their high stability,[60] is complete (see also Section 4.3) be-
ten metal ions studied,[48a] Equilibrium (2) operates and fore the onset of the formation of the mixed ligand com-
that the adenine residue is involved in metal ion binding. plexes occurs. This means that in the case of the ternary
From the three nitrogen atoms available in the purine ring, systems the experimental data of the potentiometric pH ti-
i.e., N1, N3 and N7 (see Figure 1), N1 cannot be reached trations can also be fully described by taking Equilibria
by a metal ion already coordinated to the phosphonate (4a)Ϫ(7a) into account, but now M2ϩ ϭ Cu(Arm)2ϩ. The
group;[19] hence, N3 and N7 remain for macrochelate for- corresponding stability constants are given in Table 5.
mation. At this stage one cannot distinguish with absolute
A comparison of the acidity constants of H2(dPEEA)Ϯ,
certainty between these two possibilities, but we strongly pKHH (dPEEA) ϭ 4.17 and pKHH(dPEEA) ϭ 7.75 (Table 1), with
2
favor an interaction of the phosphonate-coordinated metal those of the Cu(Arm)(H;dPEEA)ϩ complexes, i.e.,
Table 4. Increased complex stability, log ∆M/dPEEA [Equation (11)], and extent of chelate formation according to Equilibrium (2) for
some M(dPEEA) complexes, as quantified by the dimensionless equilibrium constant KI [Equations (12), (13)] and the percentage of
[18]
[23]
M(dPEEA)cl [Equation (14)]. The corresponding percentages for M(dPMEA)cl
and M(AMP)cl
are given for comparison (aqueous
solution; 25 °C; I ϭ 0.1 , NaNO3)[a]
[b]
M2ϩ
log ∆M/dPEEA
KI
% M(dPEEA)cl
% M(dPMEA)cl
% M(AMP)cl
Mn2ϩ
Co2ϩ
Ni2ϩ
Cu2ϩ
Zn2ϩ
Cd2ϩ
0.02 Ϯ 0.06
0.08 Ϯ 0.08
0.14 Ϯ 0.06
0.27 Ϯ 0.10
0.24 Ϯ 0.21[c]
0.24 Ϯ 0.07
0.05 Ϯ 0.14
0.20 Ϯ 0.22
0.38 Ϯ 0.19
0.86 Ϯ 0.43
0.74 Ϯ 0.84[c]
0.74 Ϯ 0.28
5 Ϯ 13
17 Ϯ 15
28 Ϯ 10
46 Ϯ 12
42 Ϯ 28
Յ 5
15 Ϯ 11
9 Ϯ 17
21 Ϯ 15
31 Ϯ 14
21 Ϯ 37[d]
29 Ϯ 18
56 Ϯ
75 Ϯ
50 Ϯ
7
4
7
44 Ϯ 12
42 Ϯ
9
50 Ϯ
8
[a]
[b]
[c]
[d]
Regarding the error limits (3σ) see footnote [a] of Table 2.
From column 4 of Table 3.
See footnote [c] in Table 2.
The
measurements for the Zn(dPMEA) systems were also affected by precipitation.[18]
2942  2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947

Binary and Ternary Metal Ion Complexes of 9-(5-Phosphonopentyl)adenine
FULL PAPER
2.6 Proof of an Increased Stability of the Mixed Ligand
Cu(Arm)(dPEEA) Complexes
Table 5. Logarithms of the stability constants of the
M(H;dPEEA)ϩ [Equation (6b)] and M(dPEEA) complexes [Equa-
tion (7b)], where M2ϩ ϭ Cu2ϩ, Cu(Bpy)2ϩ or Cu(Phen)2ϩ, as deter-
mined by potentiometric pH titrations in aqueous solution, to-
gether with the negative logarithms for the acidity constants of
M(H;dPEEA)ϩ [Equations (8b), (9)] at 25 °C and I ϭ 0.1 
(NaNO3)[a]
One way to quantify the stability of mixed ligand com-
plexes[62,63] is to consider Equilibrium (16a) with its corre-
sponding equilibrium constant [Equation (16b)], which is
calculated from Equation (17):
M2ϩ
log KMM(H;dPEEA)
log KMM(dPEEA)
pKMH (H;dPEEA)
Cu2ϩ
1.92 Ϯ 0.12
1.87 Ϯ 0.11
2.14 Ϯ 0.12
3.86 Ϯ 0.08
4.07 Ϯ 0.07
4.36 Ϯ 0.14
5.81 Ϯ 0.15
5.55 Ϯ 0.13
5.53 Ϯ 0.18
Cu(Bpy)2ϩ
Cu(Phen)2ϩ
See footnote [a] of Table 2.
According to the general rule for complex stabilities,
K1 Ͼ K2, Equilibrium (16a) is expected to lie on the left
with negative values for ∆ log K in agreement with statisti-
pKHCu(Arm)(H;dPEEA) ഠ 5.5 (Table 5, column 4), reveals that cal considerations, i.e. ∆ log KCu/statist ഠ Ϫ0.5.[63] The re-
in these complexes, just as in the binary ones, the proton sults for ∆ log KCu/Bpy/dPEEA and ∆ log KCu/Phen/dPEEA ac-
must be located at the phosphonate group, since metal ion cording to Equation (17) are 0.21 Ϯ 0.11 and 0.50 Ϯ 0.16,
coordination must give rise to an acidification.[52,53] In fact, respectively (see Table 5). These values are clearly larger
the acidity constants of the Cu(Arm)(H;dPEEA)ϩ com- than the statistically expected value. Hence, it is clear
plexes are about 2.2 log units smaller than pKHH(dPEEA)
,
that Equilibrium (16a) is significantly displaced to the
but approximately 1.4 log units larger than right hand side and that these ternary complexes show
pKHH (dPEEA).
an increased stability. However, it is difficult to draw
2
However, where is the Cu(Arm)2ϩ unit located? Indeed, quantitative conclusions from these results because the bi-
the Cu(Arm)(H·dPEEA)ϩ species, where the notation nary Cu(dPEEA) complex itself also shows an increased
(H·dPEEA)Ϫ indicates that the proton is at the phosphon- stability due to macrochelate formation [see Section 2.4 and
ate group, may exist in two principally different forms: One, Equilibrium (2)]. Yet, despite this shortcoming the given re-
where Cu(Arm)2ϩ is stacked with the purine system of sults for Equation (17) still prove definitely an increased
(H·dPEEA)Ϫ, designated as [Cu(Arm)/(H·dPEEA)]sϩt, and stability for the Cu(Arm)(dPEEA) complexes.
another one, where Cu(Arm)2ϩ is simply coordinated either
Another way to evaluate the stability of the ternary Cu2ϩ
to the N1/N7 sites of the adenine residue (see Section 2.2), complexes, independently of that of the corresponding bi-
[(H·dPEEA)·Cu(Arm)]aϩde, or to the phosphonate group nary complexes, is analogous to the procedure used in Sec-
which already carries the proton. However, the formation tion 2.4 for the binary systems. This means, straight-line
Cu(Arm)
of the latter species with both the proton and Cu(Arm)2ϩ correlations for log K
versus pKHH(R-PO ) plots,
Cu(Arm)(R-PO3)
3
at the phosphonate group is unlikely, in agreement with the where R-PO32Ϫ represents phosphate monoester or phos-
conclusions given in Section 2.2 for the binary phonate ligands in which the residue R is unable to interact
Cu(H;dPEEA)ϩ species, where Cu2ϩ is mainly located at with Cu(Arm)2ϩ, have been defined and they are valid in
the adenine residue and Hϩ at the phosphonate group. the pKa range between 5 and 8.[51][61] These reference lines
Hence, we have to consider the [(H·dPEEA)·Cu(Arm)]ϩade are seen in Figure 3 where the stability constants log
and the [Cu(Arm)/(H·dPEEA)]ϩst species and take into ac-
K
versus the acidity constant pKHH(dPEEA) are
Cu(Arm)
Cu(Arm)(dPEEA)
count the following intramolecular Equilibrium (15):
also plotted (square symbols) together with the correspond-
ing data[38] of the Cu(Arm)2ϩ/dPMEA2Ϫ systems (circles).
The data points in Figure 3 for the two ternary
Cu(Arm)(dPEEA) complexes are far above their reference
An evaluation of the various equilibrium constants, lines proving again the increased complex stability and this
identical to the procedure described recently,[38,61] leads to must now mean[57] that, aside from the phosphonate-
the conclusion that the stacked species in Equilibrium (15) Cu(Arm)2ϩ coordination, further interactions occur.
dominate in the Cu(Phen)2ϩ system with a formation de- Such an increased stability, which in fact is even more
gree of about 70% for the [Cu(Phen)/(H·dPEEA)]ϩst species. pronounced, is also observed for the ternary
In the case of the Cu(Bpy)2ϩ system the two isomeric spec- Cu(Arm)(dPMEA) systems[38] as one can see in Figure 3.
ies of Equilibrium (15) occur in about equal amounts. These The vertical differences between the mentioned data points
results are very similar to those obtained previously[38] for and their reference lines can be defined according to Equa-
the Cu(Arm)2ϩ/(H·dPMEA)Ϫ systems. Furthermore, that tion (11) (Section 2.3) by keeping in mind that now M2ϩ
ϭ
stacking is somewhat more pronounced in the system con- Cu(Arm)2ϩ. The needed stability of the ‘‘open’’ isomers,
taining Phen is in agreement with the expectation (see also Cu(Arm)(dPEEA)op, is calculated with pKHH(dPEEA) ϭ 7.75
Section 2.7).
(Table 1, entry 7) and the straight-line parameters given
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2943

´
´
´
´
R. B. Gomez-Coca, A. Holy, R. A. Vilaplana, F. Gonzalez-Vılchez, H. Sigel
FULL PAPER
2.7 Evaluation of the Increased Stability of the
Cu(Arm)(dPEEA) Complexes
It is clear[57] that the described enhanced complex stabilit-
ies (Table 6, column 3) must originate in additional interac-
tions, i.e., beyond the coordination of Cu(Arm)2ϩ to the
phosphonate group. The ligand dPEEA2Ϫ offers only two
such
possibilities:
The
phosphonate-coordinated
Cu(Arm)2ϩ forms (i) a macrochelate with N7 of the adenine
residue, as is the case with the binary Cu(dPEEA) complex
(Section 2.4), or (ii) an intramolecular stack between the
aromatic ring systems of Bpy or Phen and the adenine moi-
ety. This latter type of interaction is known for many ex-
amples and can be very significant;[32,35Ϫ38,54,63] it is also
evident from the results indicated in Section 2.5 for the
monoprotonated Cu(Arm)(H;dPEEA)ϩ species.
That the first of these two possible interactions is of no
importance was demonstrated in detail recently for the
Cu(Arm)2ϩ systems with dPMEA2Ϫ [38] as well as for re-
lated analogues[36,54,61] and also for AMP2Ϫ itself.[37] Hence,
the enhanced complex stabilities may only be attributed to
intramolecular stack formation; in fact, for Cu(Arm)(AMP)
Figure 3. Evidence for an enhanced stability of the ternary this was also proven to occur in the solid state by
crystal structure analyses.[6b,64] Consequently, the
Cu(Bpy)(dPEEA) (᭛) and Cu(Phen)(dPEEA) (᭜) as well as the
Cu(Bpy)(dPMEA) (᭺) and Cu(Phen)(dPMEA) (᭹) complexes,
Cu(Arm)
Cu(Arm)(dPEEA) complexes occur in the two isomeric
forms indicated in Equilibrium (3): In one, Cu(Arm)2ϩ is
only coordinated to the phosphonate group and this isomer
is designated as Cu(Arm)(dPEEA)op as already indicated in
Section 2.6; in the other form, designated as
Cu(Arm)(dPEEA)st, intramolecular stack formation occurs.
based on the relationship between log K
or log
)
Cu(Arm)(R-PO3
K
and pKHH(R-PO or pKHH(PA) in aqueous solution at I ϭ 0.1
Cu(Arm)
)
Cu(Arm)
3
 (NaNO3) and 25 °C. The plotted data corresponding to the
dPEEA2Ϫ systems are from Tables 1 and 5 and those corresponding
to dPMEA2Ϫ are from ref.[38] The two reference lines represent the
log K versus pKa relationship for the ternary Cu(Arm)(R-PO3)
complexes [Equation (10)]; it should be emphasized that R-PO32Ϫ
symbolizes here phosph(on)ate ligands with a group R unable to
undergo any kind of hydrophobic, stacking or other type of A tentative structure of this latter isomer is shown in Fig-
interaction. The parameters for the straight-line equations are
ure 4. Consequently, the dimensionless equilibrium con-
stant KI/st due to the intramolecular Equilibrium (3) is de-
listed in ref.[61]
fined by Equation (18):
Cu(Bpy)
in refs.[51,61]: log K
ϭ 3.61 Ϯ 0.07 and log
Cu(Bpy)(dPEEA)op
Cu(Phen)
K
ϭ 3.62 Ϯ 0.06. Application of these val-
Values for KI/st as well as for the percentage of the
Cu(Phen)(dPEEA)op.
ues together with the corresponding experimental results stacked isomer, Cu(Arm)(dPEEA)st, are calculated in anal-
given in Table 5 (column 3) to Equation (11) leads to the ogy to Equations (13) and (14), respectively, and summar-
log ∆Cu/Arm/dPEEA values which quantify in an exact way the ized in columns 4 and 5 of Table 6.
enhanced stability of the ternary Cu(Arm)(dPEEA) com-
plexes. These values are listed in the third column of Table 6 (i) Stack formation is always more pronounced in the
(see below).
Cu(Phen)(PA) systems, where PA2Ϫ dPEEA2Ϫ
The results of Table 6 lead to the following conclusions:
ϭ
,
Table 6. Extent of intramolecular stack formation in ternary Cu(Arm)(PA) complexes, where PA2Ϫ ϭ dPEEA2Ϫ, dPMEA2Ϫ or AMP2Ϫ
,
as calculated from stability constants determined via potentiometric pH titrations: Given are the stability enhancement log ∆Cu/Arm/PA
[Equation (11)], the intramolecular and dimensionless equilibrium constant KI/st [Equations (13), (18)], and the percentage [Equation
(14)] of the stacked Cu(Arm)(PA)st species in aqueous solution at 25 °C and I ϭ 0.1  (NaNO3)[a]
No.
Cu(Arm)(PA)
log ∆Cu/Arm/PA
KI/st
%Cu(Arm)(PA)st
Ref.
[b]
1a
1b
2a
2b
3a
3b
Cu(Bpy)(dPEEA)
Cu(Phen)(dPEEA)
Cu(Bpy)(dPMEA)
Cu(Phen)(dPMEA)
Cu(Bpy)(AMP)
0.46 Ϯ 0.10
0.74 Ϯ 0.15
0.78 Ϯ 0.10
0.96 Ϯ 0.12
0.73 Ϯ 0.08
0.99 Ϯ 0.08
1.88 Ϯ 0.66
4.50 Ϯ 1.90
5.03 Ϯ 1.39
8.12 Ϯ 2.52
4.37 Ϯ 1.02
8.77 Ϯ 1.81
65 Ϯ 8
82 Ϯ 6
83 Ϯ 4
89 Ϯ 3
81 Ϯ 4
90 Ϯ 2
[b]
[29]
[29]
[31]
[31]
Cu(Phen)(AMP)
[a]
[b]
See footnote [a] of Table 2. This work.
2944
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947

Binary and Ternary Metal Ion Complexes of 9-(5-Phosphonopentyl)adenine
FULL PAPER
for the dPEEA2Ϫ and AMP2Ϫ systems. However, the metal
ion-binding properties of all three mentioned ligands differ
significantly from those of the antivirally active PMEA2Ϫ
(see Figure 1); for all its M(PMEA) complexes an ether oxy-
gen-M2ϩ interaction is of relevance which does not occur,
of course, with its deoxa analogues or with AMP2Ϫ
In the mixed ligand Cu(Arm)(PA) complexes dPMEA2Ϫ
resembles its parent AMP2Ϫ more closely than dPEEA2Ϫ
.
.
This means, the extent of intramolecular stack formation is
similar in the Cu(Arm)(dPMEA) and Cu(Arm)(AMP) sys-
tems and also more pronounced than in Cu(Arm)(dPEEA);
this is especially true for Arm ϭ Bpy. It seems that the pen-
tyl chain is somewhat too long and does therefore not favor
Figure 4. Tentative and simplified structure of a species with an
intramolecular stack for Cu(Bpy)(dPEEA) in solution. The orien-
tation of the aromatic rings may vary somewhat from one stacked an ideal fit for the aromatic ring stacks as it appears to be
species to the next; such a stacked complex in solution should not
the case with the butyl chain.
be considered as being rigid.
It has been suggested[14,15] that diphosphorylated
PMEA2Ϫ, i.e., PMEApp4Ϫ, is initially an excellent substrate
dPMEA2Ϫ or AMP2Ϫ (compare entries a with b). This re-
sult is understandable, since the overlap of the aromatic ring
systems is expected to be more pronounced with the 3-ring
system of Phen than with the 2-ring system of Bpy. (ii) The
extent of stack formation for a given Arm is within the
error limits clearly identical for dPMEA2Ϫ and AMP2Ϫ (en-
tries 2, 3) and still similar if the Phen/dPEEA2Ϫ system (en-
try 1b) is included in the comparisons (entries 2b, 3b).
However, (iii) this is different if the Bpy systems (entries 1a,
2a, 3a) are compared; here the formation degree of
Cu(Bpy)(dPEEA)st is clearly lower than the one for
Cu(Bpy)(dPMEA) and Cu(Bpy)(AMP). Most likely the
pentyl chain is too long to allow an optimal fit with one of
for DNA polymerases because of the facilitated formation of
the M(Pα)-binding mode which leads then to the also facili-
tated formation of the reactive M(Pα)-M(Pβ,Pγ) mode which
is crucial for the transfer of a nucleotidyl unit in the poly-
merase reaction.[15] This M(Pα) binding is facilitated due to
the M2ϩ interaction with the ether oxygen. At present no
data are available on the metabolism of dPMEA or dPEEA
but if it is assumed that dPMEA and dPEEA are transported
to the cell and also diphosphorylated, like PMEA,[12,13] then
it would become understandable why these two analogues
lack antiviral activity. A facilitated M(Pα) binding (via the
ether oxygen) is not possible[14] with dPMEApp4Ϫ or
dPEEApp4Ϫ and thus the transfer of a nucleotidyl unit in
the polymerase reaction[15] would also not be facilitated.
the Bpy rings. Indeed, for C6H5-(CH2)n-COOϪ/M2ϩ
/
Phen[63,65] and related systems[66] it has previously been
shown that the (CH2)n chain may become too long and
reach then beyond the rings needed for the intramolecular
interaction. In accord with this interpretation are the results
for the Cu(Phen)(dPEEA) and Cu(Phen)(dPMEA) systems
(entries 1b, 2b); Phen with its 3-ring system is less sensitive
than the 2-ring Bpy with regard to the chain length, i.e., to
an exchange of the butyl by a pentyl chain. Hence, as far
as the stacking interaction in Cu(Arm)(PA) systems is con-
cerned, dPMEA2Ϫ mimics AMP2Ϫ somewhat better than
4. Experimental Section
4.1 Materials
Twofold
protonated
9-(5-phosphonopentyl)adenine,
i.e.,
H2(dPEEA)Ϯ, was synthesized as described below in Section 4.2
with 1,5-dibromopentane, triethyl phosphite, 1,8-diazabicy-
clo[1.3.0]undec-1-ene, bromotrimethylsilane and dimethylformam-
ide, which were purchased from SigmaϪAldrich, Prague, Czech Re-
public.
dPEEA2Ϫ
.
All aqueous stock solutions for the potentiometric pH titrations
were prepared by dissolving the various compounds in deionized,
ultrapure (MILLI-Q185 PLUS; from Millipore S.A., 67120 Mols-
heim, France) CO2-free water.
3. Conclusions
The coordinating properties of dPEEA2Ϫ and dPMEA2Ϫ
do not differ dramatically. However, it is still interesting to
note that in binary complexes of divalent 3d metal ions in-
cluding Zn2ϩ and Cd2ϩ, dPEEA2Ϫ resembles the parent nu-
cleotide AMP2Ϫ somewhat better than its chain-shortened
analogue dPMEA2Ϫ (Figure 1): In other words, macrochel-
2,2Ј-Bipyridine and 1,10-phenanthroline monohydrate were from
Merck AG, Darmstadt, FRG. All the other reagents were the same
as used recently.[18]
4.2 Synthesis of 9-(5-Phosphonopentyl)adenine
ate formation between the phosphonate-coordinated metal This compound has been mentioned in the literature[67] in the con-
text of enzyme-catalyzed deamination reactions but its synthesis
has not been described and therefore it is given here.
ion and N7 of the adenine moiety seems to be somewhat
more favored in M(dPEEA) than in M(dPMEA) species. In
fact, in a first approximation, the extent of macrochelate
formation increases for a given M2ϩ within the series
M(dPMEA) Ͻ M(dPEEA) Ͻ M(AMP), though, e.g., for
The first step in the synthesis of 9-(5-phosphonopentyl)adenine [ϭ
5-(adenin-9-yl)pentylphosphonic acid] was heating of a mixture of
1,5-dibromopentane (100 g, 0.43 mol) and triethyl phosphite
Cu2ϩ and Cd2ϩ the extent is within the error limits identical (90 mL, 1.125 equiv.) for 2 days at 110 °C; the resulting ethyl bro-
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2945

´
´
´
´
R. B. Gomez-Coca, A. Holy, R. A. Vilaplana, F. Gonzalez-Vılchez, H. Sigel
FULL PAPER
mide was distilled off. The portion volatile up to a maximum of
replaced by M(NO3)2 (25 °C; I ϭ 0.1 ). The conditions corre-
100 °C/2 kPa was removed. The residue, i.e. diethyl 5-bromopen- spond to those given recently for the M2ϩ/dPMEA systems,[18] but
tylphosphonate, was utilized in the next step without further purifi-
cation.
the M2ϩ/dPEEA ratios were 111:1 (Mg2ϩ, Ca2ϩ, Sr2ϩ, Ba2ϩ
,
Mn2ϩ), 55.6:1 (Co2ϩ, Ni2ϩ), 53.3:1 (Mg2ϩ, Ca2ϩ, Sr2ϩ, Ba2ϩ), 50:1
(Mn2ϩ), 26.9:1 (Cd2ϩ), 25:1 (Co2ϩ, Ni2ϩ), 13.4:1 (Cd2ϩ), 12.3:1
The mixture of adenine (20.0 g, 0.148 mol), 1,8-diazabicy-
clo[1.3.0]undec-1-ene (30 mL) and crude diethyl 5-bromopen-
tylphosphonate (60 mL) in dimethylformamide (200 mL) was
heated at 100 °C for 5 h and the solvent evaporated. The residue
was extracted with hot chloroform (in total 500 mL) and the sol-
vent evaporated to give by chromatography on a silica gel column
(300 mL) in chloroform by extraction with chloroform-methanol
mixtures oily diethyl 5-(adenin-9-yl)pentylphosphonate (yield:
11.6 g; 23%). This compound (34 mmol) in acetonitrile (100 mL)
was stirred with bromotrimethylsilane (20 mL) till dissolution and
left to stand overnight at ambient temperature. The mixture was
evaporated in vacuo and the residue codistilled with toluene (2 ϫ
25 mL). Water (150 mL) was added and, after 20 min, the mixture
was basified with conc. aqueous ammonia and the solvents evapo-
rated. The residue in minimum water was applied to a column of
Dowex 50 ϫ 8 (Hϩ-form) (250 mL), the column was washed with
water to the loss of acidity and UV-absorption of the eluate (moni-
tored at 254 nm). Subsequent elution of the column with diluted
(1:10) aqueous ammonia gave a UV-absorbing eluate which was
evaporated to dryness in vacuo. This residue in minimum water
was basified with ammonia to pH 10 and applied onto a column
of Dowex 1 ϫ 2 (acetate form) prewashed with water (200 mL);
the column was eluted with water (1 L) and the resin was then
stirred batchwise with 1  formic acid (500 mL), filtered off and
washed with boiling water (four 500 mL portions). The combined
eluates were evaporated in vacuo, the residue codistilled with water
(five 50 mL portions) and crystallized from water. Yield, 5.1 g
(52.5%), m.p. 330 °C. C10H16N5O3P (285.2): calcd. C 42.11, H 5.65,
N 25.55, P 10.86; found C 42.25, H 5.41, N 25.28, P 11.03. 1H
NMR (in D2O, 500 MHz): δ ϭ 8.28 and 8.21 (2 s, 1 H each, H2
and H8 of Ade), 4.27 (t, J ϭ 6.9 Hz, 2 H, H1Ј), 1.90 (m, 2 H), 1.53
(m, 4 H), 1.36 (m, 2 H, H2Ј, -3Ј, -4Ј, -5Ј) ppm. 13C NMR (in D2O,
125.7 MHz): saturated solution Ϫ low concentration allowed the
detection of proton-bearing carbons only: δ ϭ 154.09 (s, C2),
145.90 (s, C8), 46.92 (s, C1Ј), 31.63 (s, C2Ј), 29.98 (d, JC,P ϭ 5.4 Hz,
C3Ј), 35.54 (d, JC,P ϭ 4.4 Hz, C4Ј), 30.58 (d, JC,P ϭ 121.1 Hz,
C5Ј) ppm.
(Zn2ϩ), 10.6:1 (Cu2ϩ
, , ,
Cu(Bpy)2ϩ Cu(Phen)2ϩ), 5.3:1 (Cu2ϩ
Cu(Bpy)2ϩ, Cu(Phen)2ϩ) and 5:1 (Zn2ϩ). Since the stability of the
Cu(Arm)2ϩ complexes is high,[60] the formation of these species is
practically complete under the experimental conditions (titrations
of solutions with HNO3 and HNO3 ϩ Cu2ϩ/Arm were identical in
the lower pH range) and therefore, the evaluation of the titration
data of the ternary systems could be done in the way described for
the binary ones.[28]
The evaluation of the Zn2ϩ/dPEEA systems was significantly ham-
pered by precipitation, and therefore the pH range accessible for
the calculations of the stability constants was severely restricted. In
fact, for Zn(dPEEA) only a formation degree of about 2% was
reached; hence, the value for Zn(dPEEA) must be considered as
an estimate.
However, several of the constants given for the M(H;dPEEA)ϩ
complexes must also be considered as estimates (see Table in Sec-
tion 2.2), since the formation degree of these species reached only
about 3%, in the maximum 18% was reached; for the ternary
Cu(Arm)(H;dPEEA)ϩ complexes a formation degree of nearly 25%
was achieved.
The individual results for the stability constants showed no depen-
dence on pH or on the excess of metal ion concentration used. The
results are in each case the average of at least five independent pairs
of titration curves.
Acknowledgments
The competent technical assistance of Mrs. Rita Baumbusch and
Mrs. Astrid Sigel in the preparation of this manuscript is gratefully
ˇˇ
´
acknowledged. Thanks are also due to Dr. M. Budesinsky (IOCB)
for measurement and interpretation of NMR spectra. This study
was supported by the Swiss National Science Foundation (H.S.)
and the Programme of Targeted Projects (S4055109) of the Acad-
emy of Sciences of the Czech Republic (A.H.) as well as within the
COST D20 programme by the Swiss Federal Office for Education
and Science (H.S.) and the Ministry of Education of the Czech
Republic (D.20.002; A.H.). This study is also part of a research
project (No. 4055905) of the Institute of Organic Chemistry and
Biochemistry (IOCB) in Prague.
4.3 Potentiometric pH Titrations. Determination of Equilibrium
Constants
The equipment for the titrations,[18] the buffers used for cali-
bration[18] and the evaluation procedures including the computers
were the same as before.[43]
[1]
Fundamental Virology (Eds.: B. N. Fields, D. M. Knipe, P. M.
Howley), 3rd Edition, Lippincott-Raven publishers, Philadel-
The acidity constants KHH2(dPEEA) and KHH(dPEEA) of H2(dPEEA)Ϯ,
where one proton is at the base moiety and the other at the phos-
phonate group, were determined by titrating 50 mL of aqueous
0.9 m  HNO3 (25 °C; I ϭ 0.1 , NaNO3) in the presence and
absence of 0.3 m dPEEA2Ϫ under N2 with 1.6 mL of 0.03 
NaOH. The differences in NaOH consumption between such a pair
of titrations were used for the calculations. These conditions corre-
spond to those used recently for the dPMEA system;[18] it needs to
be emphasized that at a ligand concentration[22] of 3 ϫ 10Ϫ4  the
well-known[32] self-association (via π-stacking) of purines is of no
significance. The results for the acidity constants of H2(dPEEA)Ϯ
are the averages of 19 pairs of independent titrations.
phia, New York, 1996, pp. 1Ϫ1339.
[2] [2a]
M. Popovic, M. G. Sarngadharan, E. Read, R. C. Gallo,
Science 1984, 224, 497Ϫ500. [2b] R. C. Gallo, S. Z. Salahuddin,
M. Popovic, G. M. Shearer, M. Kaplan, B. F. Haynes, T. J.
Palker, R. Redfield, J. Oleske, B. Safai, G. White, P. Foster, P.
D. Markham, Science 1984, 224, 500Ϫ503.
[3]
E. De Clercq, Collect. Czech. Chem. Commun. 1998, 63,
449Ϫ479.
[4]
´
ˇ´
´
´
´
A. Holy, J. Günter, H. Dvorakova, M. Masojıdkova, G. And-
rei, R. Snoeck, J. Balzarini, E. De Clercq, J. Med. Chem. 1999,
42, 2064Ϫ2086.
[5]
R. Tribolet, H. Sigel, Eur. J. Biochem. 1987, 163, 353Ϫ363.
[6] [6a]
R. B. Martin, Y. H. Mariam, Met. Ions Biol. Syst. 1979, 8,
[6b]
57Ϫ124.
K. Aoki, Met. Ions Biol. Syst. 1996, 32, 91Ϫ134.
The stability constants KMM(H;dPEEA) and KMM(dPEEA) of
M(H;dPEEA)ϩ and M(dPEEA) were determined under the same
conditions as the acidity constants, but NaNO3 was partly or fully
Abbreviations and definitions: For AMP2Ϫ
, ,
dPEEA2Ϫ
[7]
dPMEA2Ϫ and PMEA2Ϫ see Figure 1. Arm, heteroaromatic
nitrogen base like Bpy or Phen; Bpy, 2,2Ј-bipyridine; I, ionic
2946
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947

Binary and Ternary Metal Ion Complexes of 9-(5-Phosphonopentyl)adenine
FULL PAPER
H. Suezawa, T. Yoshida, Y. Umezawa, S. Tsuboyama, M.
[34] [34a]
strength of a solution; M2ϩ, general divalent metal ion also
including in part Cu(Arm)2ϩ
;
9MeAde, 9-methyladenine;
Nishio, Eur. J. Inorg. Chem. 2002, 3148Ϫ3155. [34b] L.-n. Ji, X.
Le, Chin. Sci. Bull. 2002, 47, 1Ϫ9.
PA2Ϫ ϭ dPEEA2Ϫ or any other twofold negatively charged nu-
cleotide or nucleotide analogue; Phen, 1,10-phenanthroline; R-
PO32Ϫ, simple phosphate monoester or phosphonate ligand
with R representing a non-interacting residue. Species written
without a charge either do not carry one or represent the spec-
ies in general (i.e. independent of their protonation degree);
which of the two possibilities applies is always clear from the
context. In formulas like M(H;dPEEA)ϩ, Hϩ and dPEEA2Ϫ
are separated by a semicolon to facilitate reading, yet they ap-
pear within the same parenthesis to indicate that the proton is
at the ligand without defining its location (see Sections 2.2 and
2.5). The notation (H·dPEEA)Ϫ indicates that Hϩ is at the
phosphonate group (see Section 2.5).
[35]
[36]
H. Sigel, Pure Appl. Chem. 1989, 61, 923Ϫ932.
´
´
´
R. B. Gomez-Coca, L. E. Kapinos, A. Holy, R. A. Vilaplana,
F. Gonzalez-Vılchez, H. Sigel, Metal Based Drugs 2000, 7,
313Ϫ324.
S. S. Massoud, R. Tribolet, H. Sigel, Eur. J. Biochem. 1990,
187, 387Ϫ393.
R. B. Gomez-Coca, L. E. Kapinos, A. Holy, R. A. Vilaplana,
F. Gonzalez-Vılchez, H. Sigel, J. Inorg. Biochem. 2001, 84,
39Ϫ46.
C. A. Blindauer, A. Holy, H. Dvorakova, H. Sigel, J. Chem.
Soc., Perkin Trans. 2 1997, 2353Ϫ2363.
R. L. Benoit, M. Frechette, Can. J. Chem. 1984, 62, 995Ϫ1000.
G. Kampf, L. E. Kapinos, R. Griesser, B. Lippert, H. Sigel, J.
Chem. Soc., Perkin Trans. 2 2002, 1320Ϫ1327.
´
[37]
[38]
´
´
´
´
[39]
´
ˇ´
´
[40]
[41]
´
[8]
L. Naesens, R. Snoeck, G. Andrei, J. Balzarini, J. Neyts, E. De
Clercq, Antiviral Chem. Chemother. 1997, 8, 1Ϫ23.
[9] [9a]
[42]
[43]
[44]
[45]
[46]
[47]
E. De Clercq, Collect. Czech. Chem. Commun. 1998, 63,
C. Meiser, B. Song, E. Freisinger, M. Peilert, H. Sigel, B. Lip-
pert, Chem. Eur. J. 1997, 3, 388Ϫ398.
H. Sigel, C. P. Da Costa, B. Song, P. Carloni, F. Gregan, J.
Am. Chem. Soc. 1999, 121, 6248Ϫ6257.
A. Saha, N. Saha, L.-n. Ji, J. Zhao, F. Gregan, S. A. A. Sajadi,
[9b]
480Ϫ506.
D. Colledge, G. Civitico, S. Locarnini, T. Shaw,
´ ˇ
Antimicrob. Agents Chemother. 2000, 44, 551Ϫ560.
Chemische Rundschau (CH-4501 Solothurn, Switzerland), No.
19; Oct. 8, 2002; p. 68.
[10]
´ ˇ
[11]
[12]
´
A. Holy, Il Farmaco 1991, 46 (Suppl. 1), 141Ϫ146.
B. Song, H. Sigel, J. Biol. Inorg. Chem. 1996, 1, 231Ϫ238.
ˇ
´
`
´ ˇ
´
A. Merta, I. Votruba, I. Rosenberg, M. Otmar, H. Hrebabecky,
H. Sigel, D. Chen, N. A. Corfu, F. Gregan, A. Holy, M. Stra-
´
ˇ´
R. Bernaerts, A. Holy, Antiviral Res. 1990, 13, 209Ϫ218.
sak, Helv. Chim. Acta 1992, 75, 2634Ϫ2656.
[13] [13a]
´
P. Kramata, I. Votruba, B. Otova, A. Holy, Mol. Pharma-
H. Sigel, A. D. Zuberbühler, O. Yamauchi, Anal. Chim. Acta
[13b]
ˇ
´
col. 1996, 49, 1005Ϫ1011.
G. Birkus, I. Votruba, A. Holy,
1991, 255, 63Ϫ72.
´
B. Otova, Biochem. Pharmacol. 1999, 58, 487Ϫ492.
H. Sigel, Eur. J. Biochem. 1968, 3, 530Ϫ537.
[48] [48a]
[14]
´ ˇ
H. Sigel, B. Song, C. A. Blindauer, L. E. Kapinos, F. Gregan,
H. Sigel, D. B. McCormick, Acc. Chem. Res. 1970, 3,
N. Pronayova, Chem. Commun. 1999, 743Ϫ744.
H. Sigel, Pure Appl. Chem. 1999, 71, 1727Ϫ1740.
A. Holy, E. De Clercq, I. Votruba, ACS Symp. Ser. 1989,
201Ϫ208. [48b] H. Sigel, K. Becker, D. B. McCormick, Biochim.
Biophys. Acta 1967, 148, 655Ϫ664.
S. A. A. Sajadi, B. Song, F. Gregan, H. Sigel, Inorg. Chem.
´
´
[15]
[16]
[49]
´
´ ˇ
401, 51Ϫ71.
1999, 38, 439Ϫ448.
[17]
[50] [50a]
ˇ
´
´
´
A. Holy, I. Votruba, A. Merta, J. Cerny, J. Vesely, J. Vlach, K.
Sediva, I. Rosenberg, M. Otmar, H. Hrebabecky, M. Trav-
nıcek, V. Vonka, R. Snoeck, E. De Clercq, Antiviral Res. 1990,
H. Irving, R. J. P. Williams, Nature (London) 1948, 162,
[50b]
ˇ
´
ˇ
´
´
746Ϫ747.
1953, 3192Ϫ3210.
H. Irving, R. J. P. Williams, J. Chem. Soc.
´ˇ
[51]
13, 295Ϫ311.
H. Sigel, L. E. Kapinos, Coord. Chem. Rev. 2000, 200Ϫ202,
563Ϫ594.
[18]
´
´
R. B. Gomez-Coca, L. E. Kapinos, A. Holy, R. A. Vilaplana,
F. Gonzalez-Vılchez, H. Sigel, J. Chem. Soc., Dalton Trans.
2000, 2077Ϫ2084.
H. Sigel, Coord. Chem. Rev. 1995, 144, 287Ϫ319.
Sigel, J. Indian Chem. Soc. 1997, 74, 261Ϫ271 (P. Ray Award
[52] [52a] H. Sigel, B. Lippert, Pure Appl. Chem. 1998, 70, 845Ϫ854.
´
´
[52b]
B. Song, J. Zhao, R. Griesser, C. Meiser, H. Sigel, B. Lip-
[19] [19a]
[19b]
H.
pert, Chem. Eur. J. 1999, 5, 2374Ϫ2387.
R. Griesser, G. Kampf, L. E. Kapinos, S. Komeda, B. Lippert,
J. Reedijk, H. Sigel, Inorg. Chem. 2003, 42, 32Ϫ41.
M. S. Lüth, L. E. Kapinos, B. Song, B. Lippert, H. Sigel, J.
Chem. Soc., Dalton Trans. 1999, 357Ϫ365.
H. Sigel, N. A. Corfu, L.-n. Ji, R. B. Martin, Comments Inorg.
Chem. 1992, 13, 35Ϫ59.
R. B. Martin, Met. Ions Biol. Syst. 1996, 32, 61Ϫ89.
R. B. Martin, H. Sigel, Comments Inorg. Chem. 1988, 6,
285Ϫ314.
[53]
Lecture).
[20]
[21]
[22]
[23]
[54]
´
ˇ´
´
C. A. Blindauer, A. H. Emwas, A. Holy, H. Dvorakova, E.
Sletten, H. Sigel, Chem. Eur. J. 1997, 3, 1526Ϫ1536.
[55]
´
ˇ´
´
`
C. A. Blindauer, A. Holy, H. Dvorakova, H. Sigel, J. Biol. In-
org. Chem. 1998, 3, 423Ϫ433.
[56]
[57]
`
H. Sigel, S. S. Massoud, N. A. Corfu, J. Am. Chem. Soc. 1994,
116, 2958Ϫ2971.
E. M. Bianchi, S. A. A. Sajadi, B. Song, H. Sigel, Chem. Eur.
J. 2003, 9, 881Ϫ892.
H. Sigel, Chem. Soc. Rev. 1993, 22, 255Ϫ267.
H. Sigel, B. Song, Met. Ions Biol. Syst. 1996, 32, 135Ϫ205.
R. K. O. Sigel, B. Song, H. Sigel, J. Am. Chem. Soc. 1997,
119, 744Ϫ755.
[58]
[59]
S. S. Massoud, H. Sigel, Inorg. Chem. 1988, 27, 1447Ϫ1453.
H. Sigel, S. S. Massoud, R. Tribolet, J. Am. Chem. Soc. 1988,
110, 6857Ϫ6865.
[24]
[25]
[26]
[60] [60a]
[60b]
H. Irving, D. H. Mellor, J. Chem. Soc. 1962, 5222Ϫ5237.
G. Anderegg, Helv. Chim. Acta 1963, 46, 2397Ϫ2410.
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[61]
´
´ ˇ
´
C. A. Blindauer, A. Holy, H. Sigel, Collect. Czech. Chem. Com-
D. Chen, M. Bastian, F. Gregan, A. Holy, H. Sigel, J. Chem.
Soc., Dalton Trans. 1993, 1537Ϫ1546.
H. Sigel, Angew. Chem. 1975, 87, 391Ϫ400; Angew. Chem. Int.
Ed. Engl. 1975, 14, 394Ϫ402.
R. Malini-Balakrishnan, K. H. Scheller, U. K. Häring, R. Tri-
bolet, H. Sigel, Inorg. Chem. 1985, 24, 2067Ϫ2076.
K. Aoki, J. Am. Chem. Soc. 1978, 100, 7106Ϫ7108.
H. Sigel, R. Malini-Balakrishnan, U. K. Häring, J. Am. Chem.
Soc. 1985, 107, 5137Ϫ5148.
G. Liang, R. Tribolet, H. Sigel, Inorg. Chem. 1988, 27,
2877Ϫ2887.
mun. 1999, 64, 613Ϫ632.
[62]
[63]
˚
´
C. A. Blindauer, T. I. Sjastad, A. Holy, E. Sletten, H. Sigel, J.
Chem. Soc., Dalton Trans. 1999, 3661Ϫ3671.
´
G. Kampf, M. S. Lüth, L. E. Kapinos, J. Müller, A. Holy, B.
Lippert, H. Sigel, Chem. Eur. J. 2001, 7, 1899Ϫ1908.
C. F. Moreno-Luque, R. Griesser, J. Ochocki, H. Sigel, Z.
Anorg. Allg. Chem. 2001, 627, 1882Ϫ1887.
[64]
[65]
´
´
A. Holy, I. Rosenberg, H. Dvorakova, Collect. Czech. Chem.
Commun. 1990, 55, 809Ϫ818.
[66]
[67]
O. Yamauchi, A. Odani, H. Masuda, H. Sigel, Met. Ions Biol.
Syst. 1996, 32, 207Ϫ270.
O. Yamauchi, A. Odani, S. Hirota, Bull. Chem. Soc. Jpn. 2001,
74, 1525Ϫ1545.
A. L. Margolin, D. R. Borcherding, D. Wolf-Kugel, N. Margo-
lin, J. Org. Chem. 1994, 59, 7214Ϫ7218.
Received February 19, 2003
Eur. J. Inorg. Chem. 2003, 2937Ϫ2947
www.eurjic.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2947