Development of bis(2-picolyl)amine-zinc chelates for imidazole receptors

Article abstract of DOI:10.1002/ejoc.200700926

New phenyl and phenol bis(2-picolyl)amine (Dpa) derivatives have been synthesized in order to generate zinc chelates for imidazole anion receptors. Previously, binuclear phenolic zinc and copper chelates have shown affinity for pyrophosphate and guanidine anions, respectively. Herein we report significant imidazole affinity increasing from 2.38 × 10⁶ to 2.90 × 10⁷ for phenol-bridged binuclear zinc-Dpa chelates, as evidenced by dynamic and titration ¹H NMR studies. Among the Dpa chelates investigated, the zinc-coordinated phenol group plays a crucial role in the mechanism of anion binding. Low-temperature ¹H NMR experiments suggest a σ_ν-symmetric geometry for the imidazole chelate. Computational DFT studies at the B3LYP level of theory imply that imidazole binding displaces the phenol bridge between the zinc ions. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200700926

FULL PAPER
DOI: 10.1002/ejoc.200700926
Development of Bis(2-picolyl)amine–Zinc Chelates for Imidazole Receptors
Taina Routasalo,[a] Juho Helaja,[a] Jari Kavakka,[b] and Ari M. P. Koskinen*[a]
Keywords: Amines / N,O ligands / Heterocycles / Zinc / Chelation
New phenyl and phenol bis(2-picolyl)amine (Dpa) deriva-
tives have been synthesized in order to generate zinc che-
lates for imidazole anion receptors. Previously, binuclear
phenolic zinc and copper chelates have shown affinity for
pyrophosphate and guanidine anions, respectively. Herein
we report significant imidazole affinity increasing from
2.38ϫ106 to 2.90ϫ107 for phenol-bridged binuclear zinc–
Dpa chelates, as evidenced by dynamic and titration 1H NMR
studies. Among the Dpa chelates investigated, the zinc-coor-
dinated phenol group plays a crucial role in the mechanism
of anion binding. Low-temperature 1H NMR experiments
suggest a σν-symmetric geometry for the imidazole chelate.
Computational DFT studies at the B3LYP level of theory im-
ply that imidazole binding displaces the phenol bridge be-
tween the zinc ions.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
Anion recognition with bimetallic coordination com-
plexes has been reviewed recently.[7] For this purpose bis(2-
picolyl)amines (Dpa) 1 (Figure 1) are well-known tridentate
ligands capable of donating three electron pairs to a cat-
ionic metal center.[8–10] The Dpa geometry and its confor-
mational flexibility gives this ligand a strong affinity for
biologically interesting Zn2+ and Cu2+ ions, whereas triden-
tate amine–metal chelation leaves basic coordination places
free for counterions. Accordingly, Dpa ligands have been
used as such,[11] but more widely as linked ligand units in
larger host structures.[12,13] Recently, examples of Dpa-func-
tionalized metal–anion fluorescence probes have been re-
ported in which Dpa units have been linked to a fluorescent
host molecule.[13b,14–16] Lately, it has also been demon-
strated that bimetallic double Dpa chelates can function in
molecular recognition of anionic species, for example, phos-
phate,[17] pyrophosphate,[18] or phospholipid[19] anions, as
well as histidine residues to some extent.[20]
Introduction
The development of anion receptors is a major current
challenge in the field of supramolecular chemistry. Anion
recognition can be achieved through a variety of molecular
interactions: Electrostatic interactions, hydrogen bonds,
hydrophobic effects, and metal or Lewis acid coordination,
or combinations of these interactions working together.[1]
Among these interactions, metal coordination offers a ratio-
nal approach to the design and construction of self-correct-
ing assemblies with a wide range of accessible geometries.[2]
The concept of metal-templated self-assembly in receptor
design has been inspired by various biological systems in
which proteins act as self-assembling binding sites.[3]
Among the metalloproteins, zinc enzymes offer a large
number of coordination motifs and thus provide models for
the design of zinc-based anion receptors.[4] Our model for
imidazole receptor design was the enzyme superoxide dis-
mutase (SOD), which has a bimetallic Zn,Cu imidazole
bridging moiety.[5] Previously, self-assembling and ion-spe-
cific ligands for Zn and Cu ions had been developed based
on SOD mimicry.[6] We took the Zn,Cu-bridged histidine
residue of the SOD enzyme as a model for the design of
an imidazole anion-binding receptor. We further envisioned
that an appropriately chelated metal ion-pair would be able
to contribute to imidazole anion binding in the same man-
ner as the bimetallic cation-pair does in the SOD enzyme.
Figure 1. Bis(2-picolyl)amine (Dpa).
We have studied metal–ligand structures that are capable
of zinc chelation and usable for anion recognition purposes
in biological environments. The imidazole binding sites pre-
viously reported have been mainly bimetallic imidazole
cryptand structures, including the recent work of Fabbrizzi
et al. in which fluorescent chemosensing activity was re-
ported for imidazole binding.[21] Regiospecific protein-sur-
face sensing is highly desirable in many biotechnological
and pharmaceutical processes, for example, in protein de-
tection and for analyzing enzyme activation or deactivation.
[a] Department of Chemistry, Helsinki University of Technology,
P. O. Box 6100, 02015 HUT, Finland
E-mail: ari.koskinen@tkk.fi
[b] Laboratory of Organic Chemistry, Department of Chemistry,
University of Helsinki,
A. I. Virtasen aukio 1, P. O. Box 55, 00014, Finland
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
3190
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 3190–3199

Bis(2-picolyl)amine–Zinc Chelates for Imidazole Receptors
Our ultimate goal is to construct chelates capable of histi-
dine residue recognition in proteins that will act as chemical
sensors. Dpa-based chelates form a promising tool with po-
tential applications both in metal-cation chelation and sub-
sequent chelate-based anion recognition. Herein we report
the synthesis of new Dpa chelators and the ability of their
zinc chelates to act as imidazole anion receptors. The re-
1
Scheme 1. Synthesis of the monosubstituted chelator 2a.
cognition event has been studied by H NMR and molecu-
lar modeling techniques.
The disubstituted ligand 2b was prepared in the same
manner as ligand 2a by Wohl–Ziegler bromination and sub-
stitution with Dpa (Scheme 2).
Results and Discussion
We have constructed Dpa ligands based on three dif-
ferent skeleta: A monosubstituted compound 2a and disub-
stituted compounds 2b and 2c–e. All the Dpa ligands have
a potentially linkable R group (H, CO2Me or NO2) (Fig-
ure 2). In the disubstituted ligand 2c, the phenolic oxygen
between the two ligands is capable of forming a bridge be-
tween two metal ions (3, Figure 3). Such bimetallic chelates
have been shown to participate in anion recognition.[20,21]
Dapporto and co-workers have developed dinuclear Zn or
Cu amino-phenolic ligands for small anionic molecules.[22]
Unlike Fusi and co-workers, we used the pyridine nitrogen
atoms in structures 2c, 2d, and 2e for metal coordination
instead of primary amines.
Scheme 2. Synthesis of the disubstituted chelator 2b.
Disubstituted phenolic ligands 2c–e were synthesized by
three different routes (Schemes 3, 4, and 5). The para-ester
skeleton 2c was synthesized starting from methyl 4-hy-
droxybenzoate, which was diformylated with hexamethy-
lenetetramine in TFA by using a modified Duff reaction.[24]
The product 6 precipitated in 87% yield after stirring (2–
4 h) in a large excess of water. The formyl substituents were
Figure 2. Dpa ligands based on different skeleta 2a–2e.
Scheme 3. Synthesis of the disubstituted chelator 2c.
Figure 3. Phenol-bridged structure between two metal ions.
Synthesis
The monosubstituted ligand 2a was prepared as shown
in Scheme 1. Wohl–Ziegler bromination occurs under neu-
tral conditions with N-bromosuccinimide (NBS) and a cata-
lytic amount of 2,2Ј-azobis(isobutyronitrile) (AIBN) as ini-
tiator.[23] The brominated molecule was then substituted
with the Dpa ligand under basic conditions.
Scheme 4. Synthesis of the disubstituted chelator 2d.
Eur. J. Org. Chem. 2008, 3190–3199
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3191

T. Routasalo, J. Helaja, J. Kavakka, A. M. P. Koskinen
FULL PAPER
[25]
reduced with NaBH4 and the primary alcohols were se- quently, the signals would be split, whereas in the bidentate
lectively converted into chlorides with SOCl2.[26] Final sub- case only one signal would be detected due to a higher mo-
stitution with Dpa was performed as described above.
lecular symmetry (Figure 4).
Figure 4. Determination of the coordination geometry by analysis
1
of the H NMR shifts of the imidazole. The red broken line illus-
trates the C2 symmetry axis of the imidazole.
All the studied imidazole–zinc chelates exhibited sym-
metric behavior at 30 °C, that is, only one signal was ob-
served for the δ1 protons, as listed in Table 1. However, this
could be attributed to rapid chemical exchange at this tem-
perature.
Scheme 5. Synthesis of the disubstituted chelator 2e.
In the synthesis of the nitro-substituted ligand 2d
(Scheme 4), the benzodioxine derivative was obtained from
p-nitrophenol by using paraformaldehyde under acidic con-
ditions.[27] The temperature must be carefully controlled in
order to obtain good yields of dioxine 9: Even a small devi-
Table 1. 1H NMR chemical shifts [ppm] of imidazole with different
chelates measured in CD3OD.
ation of a few degrees from the optimal +60 °C led to the Chelate system
δ1
δ2
∆δ1
∆δ2
domination of side-products. Dioxine 9 was brominated in
refluxing HBr to give compound 10. Subsequently, the bro-
mines were substituted with Dpa ligands to obtain 2d.
The synthesis of ligand 2e was similar to that of ligands
2a and 2b, that is, by Wohl–Ziegler bromination and substi-
tution with Dpa (Scheme 5). The phenolic OH was pro-
tected as the methyl ether before the radical reaction and
subsequently deprotected with BBr3.
Despite the similarities of the ligands, it was found that
general synthetic methods could not be applied to all the
ligand syntheses: For example the successful diformylation
of para-phenol ester 2c and the dioxine route to 2d gave
only negligible yields with other ligands. The overall yields
for the syntheses of 2a–e varied from 17 to 48%. Finally,
the most laborious part of the syntheses was the chromato-
graphic purification, which suffered from low capacity and
concentration-dependent retention times that were charac-
teristic of all the Dpa ligands.
Imidazole
7.05
7.28
7.29
7.37
7.52
7.53
7.46
7.67
8.06
8.16
8.36
8.76
8.77
8.60
Imidazole + Zn(NO3)2
2a + Zn(NO3)2 + imidazole
2b + Zn(NO3)2 + imidazole
2c + Zn(NO3)2 + imidazole
2d + Zn(NO3)2 + imidazole
2e + Zn(NO3)2 + imidazole
0.23
0.24
0.32
0.47
0.48
0.41
0.39
0.49
0.69
1.09
1.10
0.93
The imidazole signals could be recognized in all cases,
even at low concentrations at room temperature. Thus, the
2a–e zinc chelates were titrated against imidazole to deter-
mine the strength of the coordination (Figures 5 and 6).
Ligand 2a shows a low affinity for imidazole, as indicated
by its steeply descending curve. The curve obtained with
ligand 2b indicates that the imidazole binds this zinc chelate
with a slightly improved affinity compared with that of 2a.
Imidazole Dpa–Zinc Receptors: Geometry and Strength
We studied the coordination of imidazole to the zinc che-
lates by variable-temperature (dynamic) and titration 1H
NMR techniques. The coordination geometry and strength
were further examined by molecular modeling using DFT.
1H NMR Titration Measurements
The characteristic coordination-dependent magnetic be-
havior of imidazole protons allowed us to determine, based
on molecular symmetry, whether imidazole exhibits a
mono- or bidentate coordination to zinc ions. In the mono-
dentate case, imidazole would have a chemically nonequiva-
Figure 5. 1H NMR spectra of 2c + 2 equiv. Zn(NO3)2 titrated
lent magnetic environment for the δ1 protons. Conse- against imidazole (in CD3OD).
3192
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 3190–3199

Bis(2-picolyl)amine–Zinc Chelates for Imidazole Receptors
Figure 6. Imidazole titration curves for the 2a–e zinc chelates. The y axis represents the difference between the δ2 value of the chelated
and the free imidazole: ∆δ2 = [δ2(Zn–imidazole–2 chelate) – δ2(imidazole)]. The curve 2b/2 represents 2b in which the imidazole equiv.
number is calculated with respect to the number of Dpa units.
However, when the same comparison was performed taking
To confirm the results described above, the binding con-
into account the number of Dpa units per ligand, that is, stants (KA) for chelates 2b–e were calculated. Imidazole (S0
2a and 2b/2 curves, the result is that 2b is a weaker imid- = amount of substrate) was titrated against zinc–ligand (R0
azole binder than 2a below the concentration of 1 equiv. = amount of receptor) and the chemical shift of imidazole’s
imidazole (Dpa). This can be attributed to the stronger δ2 proton (Figure 4) was plotted versus the quantity of the
intermolecular chelation of 2a relative to the intramolecular zinc receptor. The results (Table 2) were analyzed by using
chelation of 2b. The steeper slope of 2b/2 before the ad- Equation (1)[28] on the assumption that one binding con-
dition of 1 molar equiv. (per Dpa unit) can be attributed to stant could explain the observed binding phenomena.[29]
entropic factors, that is, after saturation of the zinc–Dpa The binding constants show that 2d has the strongest coor-
chelates with imidazole the intramolecular ligand exchange dination strength, whereas 2b has a substantially weaker
becomes visible.
chelation capability compared with chelates 2c–e [Equa-
For phenol chelates 2c, 2d, and 2e, the imidazole titration tion (2)].
curves indicate significant binding activity. In these cases
δobs = δs + ∆δ/2S0{KD + S0 + R0 – [(KD + R0 + S0)2 – 4R0S0]1/2
}
the imidazole ∆δ2 value was unaffected until after more
than 0.4–0.6 equiv. imidazole had been added to the solu-
tion (Figure 6). After 1.0 equiv. of imidazole had been
added, the shapes of the curves for these chelates differed
somewhat, which can be attributed to solubility effects.
Nevertheless, the chelate with 2d exhibits the slowest de-
scent, which indicates the strongest affinity.
(1)
KA = 1/KD
(2)
Table 2. Binding constants for 2b–e zinc chelates.[a]
The coordination strength can also be determined by in-
specting the chemical shifts of the imidazole protons: In
strong imidazole–zinc coordination, the nitrogen electron-
donation is expected to make the imidazole ring electron-
deficient and thereby cause deshielding and downfield shifts
of the imidazole protons. The results correlate well with the
titration results: Phenolic chelates 2c–e are strongly de-
shielded with p-nitrophenol 2d chelated imidazole the most
affected, as shown by comparison of the ∆δ1 and ∆δ2 values
(Table 1). The fact that the δ values for the chelate of 2b
are less deshielded indicates that weaker imidazole chelation
occurs for this chelate. However, the values are still signifi-
cantly more deshielded than those of the mono-Dpa chelate
Chelate
Kα [1/]
2b
2c
2d
2e
2.38ϫ106
1.38ϫ107
2.90ϫ107
1.60ϫ107
[a] The detailed procedure for determining the binding constants is
described in the Supporting Information.
1H NMR Variable-Temperature Measurements
The coordination geometry and strength of the imid-
2a, which suggests a cooperative role for the zinc ions in azole–zinc(II) chelates were also explored through variable-
the chelation.
temperature 1H NMR measurements between –95 and
Eur. J. Org. Chem. 2008, 3190–3199
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3193

T. Routasalo, J. Helaja, J. Kavakka, A. M. P. Koskinen
FULL PAPER
+30 °C (Figures 7 and 8). The distinctive feature of these the double Dpa 2b–e chelates. This is in contrast to obser-
measurements was that the imidazole protons formed vations reported for amino-phenol-based chelates.[21] The
clearly detectable singlets, even at reduced temperatures for singlet nature of the imidazole δ1 values is an indication of
the high molecular symmetry of these chelates and suggests
bidentate imidazole coordination.
1
Inspection of the Dpa pyridine H NMR chemical shifts
at reduced temperatures gave information on the binding
geometry and strength of imidazole. In the case of the
mono-Dpa 2a–zinc chelate, lowering the temperature
caused significant signal splitting and broadening that
could not be interpreted unambiguously (Figure 7). Imid-
Figure 9. Imidazole coordinated through one nitrogen to the 2a–
Figure 7. Variable-temperature 1H NMR measurements for the 2a–
imidazole–zinc chelate.
zinc chelate (R = H) or as an anion through each nitrogen to two
chelates (R = 2a–Zn chelate).
Figure 8. The aromatic regions of the variable-temperature 1H NMR spectra measured for ligands 2b–e with Zn and imidazole in CD3OD.
δ1 and δ2 denote the imidazole chemical shifts (Figure 4). The arrows and letters a–d and aЈ–dЈ indicate pyridine protons that show
splitting at reduced temperatures.
3194
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 3190–3199

Bis(2-picolyl)amine–Zinc Chelates for Imidazole Receptors
azole clearly coordinates to the zinc ions, but the degree of Energies (∆G) based on the coalescence temperature using
mono- or bidentate coordination cannot be estimated from the Eyring equation gives energies of 12.0Ϯ0.7 kcal/mol for
this data (Figure 9).
2c and 2d, and 13.5Ϯ0.7 kcal/mol for 2e.[30] To our sur-
1
The H NMR spectra of the imidazole–zinc–2b chelate prise, the slowest chemical exchange and highest activation
showed more obscure behavior at reduced temperatures energies for 2c–e were observed for 2e, which has the weak-
than the phenolic chelates 2c–e. In the case of 2b, some est imidazole affinity according to our titration results (Fig-
chemical exchange still occurred at –70 °C. This was indi- ure 6). This indicates that the activation energies for confor-
cated by broad signals and an absence of correlations in the mational exchange and for complex formation are different
COSY spectrum. Thus, the assignment of 2b is also par- and to some extent mutually independent.
tially tentative (Figure 8). The behavior is probably con-
nected to lower chelate stability. In the case of chelate 2b,
two competing chelation models can be conspired, that is,
intra- and intermolecular ones (A and B, respectively, Fig-
ure 10), both consistent with the NMR results.
Figure 11. The aromatic part of the COSY spectrum of 2d +
2 equiv. Zn(NO3)2 + 1 equiv. imidazole measured in CD3OD at
–70 °C. The black and gray squares depict the two clearly distinct
coupling patterns for the Dpa pyridine protons a–d and aЈ–dЈ,
respectively.
Figure 10. The imidazole–zinc–2b chelate could be an internal
complex (A) or an intermolecular chelate structure (B).
Another feature observed in the variable-temperature
measurements was a clear splitting of the pyridine proton
signals, which could be detected for all phenolic chelates
2c–e by lowering the temperature (Figure 8). The splitting
of the pyridine signals into two major sets of signals (a–d,
aЈ–dЈ) was evidenced by a COSY experiment (Figure 11).
This indicates lower symmetry in the chelate: The Dpa pyr-
idine groups experience different surroundings, which is re-
flected in the δ values. Two possible σν-symmetric geome-
tries are illustrated in Figure 12. In variable-temperature ex-
periments we also looked for the coalescence points of these
sets of signals, which would reveal the energetics of the in-
terconversion. Unfortunately, sharp coalescence points
could not be detected in these spectra (Figure 8), but an
overall trend did emerge: Signals merge upon increasing the
temperature. For 2c and 2d this occurs between approxi-
mately –10 and +10 °C, but in both cases there is still some
broadening of the signal at 25 °C. For 2e, the coalescence
point lies at around +25 °C (Ϯ10 °C), signals still being
broad at 55 °C. An estimation of the related Gibbs’ Free temperature H NMR experiments.
Figure 12. Two possible σν-symmetric geometries A and B for zinc–
imidazole–2c–e chelates that can be interpreted on the basis of low-
1
Eur. J. Org. Chem. 2008, 3190–3199
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3195

T. Routasalo, J. Helaja, J. Kavakka, A. M. P. Koskinen
FULL PAPER
Molecular Modeling
The imidazole binding of 2b–e zinc chelates was studied
computationally by carrying out DFT calculations at the
B3LYP level of theory using the LACVP basis set. The pri-
mary objective of the calculations was to find theoretical
evidence for the experimentally observed bidentate imid-
azole binding of the phenyl- and phenol-type Dpa–zinc che-
lates. For computational simplicity, the nitrate was replaced
with a chloride ion as the counteranion.
Figure 14. DFT geometry-optimized nonbridged “opened” chelate
for 2e + 2 ZnCl + imidazole anion (left) and 2b + ZnCl2 + ZnCl
+ imidazole anion (right).
Imidazole-bridged and -non-bridged 2b–e zinc chelate
geometries were computationally probed by minimizing the
energy of each structure (Table 3). The starting geometry
for the imidazole-bridged 2b–zinc chelate calculation was
set simply by placing the imidazole ring in between the zinc
ions with bidentate nitrogen coordination. For this chelate,
the calculation converged smoothly to an energy minimum
at which the imidazole remained bidentate, bridging the
zinc ions (Table 3, Figures 13 and 14). The imidazole nitro-
gen–zinc bond lengths were 2.07 and 2.18 Å in the mono-
and dichloro-coordinated zinc ions, respectively. The non-
bridged, that is, “opened” imidazole–2b–zinc chelate geom-
etry was obtained by optimizing the monocoordinated
imidazole anion as the starting geometry (Figure 14). The
imidazole nitrogen–zinc ion distances were 2.03 and 5.94 Å
in the coordinated and noncoordinated ions, respectively.
Unexpectedly, this type of chelation geometry was energeti-
cally 2.22 kcal/mol more advantageous than the bridged,
monodentate imidazole anion binding.
In the case of geometry-optimized phenolic 2c–e zinc
chelates, the imidazole anion could not be placed between
the phenolic oxygen-bridged zinc cations, but was placed
next to the metals, the nitrogen atoms being located close
to the bonding distance of the zinc(II) ions. The geometry
optimizations that converged to the global minimum
showed that the symmetric phenolic oxygen bridge between
the zinc ions was opened when the imidazole anion coordi-
nated to both of the metal ions in a bidentate manner
through both of the nitrogen atoms (Figure 15). The imid-
azole nitrogen–zinc bond lengths were 2.14 and 2.05 Å for
the phenol-coordinated and non-coordinated zinc ions,
respectively. The clear energy advantage of the imidazole-
bridged bidentate chelate was a somewhat surprising result
as in the literature only monodentate imidazole coordina-
tion has been reported for phenol-bridged dinuclear zinc
chelates.[21] We also tried to minimize the energy of the phe-
nolic imidazole–zinc chelate structures in which one imid-
azole nitrogen coordinates to both zinc ions, but these were
energetically less advantageous and finally converged to
imidazole-bridged bidentate or “opened” structures.
Table 3. Molecular modeling results for the 2b–e imidazole–zinc
chelates (B3LYP LACVP).
Structure
Energy [a.u.]
Relative stability of the bridge
chelate
∆H (bridge-open) [kcal/mol]
2b–Zn bridge chelate
2b chelate open
2c–Zn bridge chelate
2c open chelate
2d–Zn bridge chelate
2d open chelate
2e–Zn bridge chelate
2e open chelate
–3304.369940
–3304.373479
–3146.573305
–3146.565529
–3123.198478
–3123.189863
–2918.764323
–2918.757787
–2.22
4.88
5.41
4.10
Figure 15. DFT geometry-optimized structure of 2e + 2ZnCl +
imidazole anion chelate (side view on the left and top view on the
right). In the energy-minimum structure, the imidazole anion is
bridging the chelated zinc ions (black) in a bidentate manner in-
stead of the phenol oxygen (gray).
A comparison between the open and closed systems, in
which the imidazole coordinates to one or two zinc ions,
respectively, is shown in Table 3. Comparison of the ener-
gies of structures 2b and 2c shows that the phenol group
stabilizes the imidazole chelation by 7.1 kcal/mol. The gene-
ral trend with respect to phenol para substitution is that
electron-withdrawing substituents notably strengthen the
imidazole chelation. The calculated difference between the
unsubstituted 2e and nitro-substituted 2d chelate models is
Figure 13. DFT geometry-optimized structure of 2b + ZnCl2
+
ZnCl + imidazole anion chelate (side view on the left and top view
on the right). The zinc ion (black) with two chlorine counter ions
(white) has a distorted-octahedral anion coordination and the zinc
ion with one chlorine counter ion has trigonal-bipyramidal anion
coordination (hydrogen atoms have been omitted for clarity).
3196
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 3190–3199

Bis(2-picolyl)amine–Zinc Chelates for Imidazole Receptors
the following exceptions: DMF and THF were distilled from ninhy-
dride and Na/benzophenone, respectively. The 1H and 13C NMR
spectra were recorded at 400.133 and 100 MHz. Chemical shifts
are reported in ppm using tetramethylsilane (1H: 0 ppm) and chlo-
roform (13C: 77.0 ppm) as internal standards. Flash column
chromatography was performed on Merck silica gel 60 (230–
400 mesh) or on Merck aluminium oxide 90 neutral (70–120 mesh).
Molecular modeling was performed with Jaguar 5.5 software via a
Maestro interface 6.0 installed on a Linux 9.0 PC workstation
equipped with a 3.4 GHz Pentium Intel processor and 2 GB RAM.
1.31 kcal/mol (Table 3), whereas for CO2Me para-substi-
tuted 2c the affinity is increased by 0.78 kcal/mol relative to
2e.
The minimum-energy structures for the phenyl and phe-
1
nol chelates do not unambiguously explain the H NMR
differences observed at reduced temperatures, although all
the calculated minima have some degree of dissymmetry
with respect to the Dpa pyridine groups. The energetics im-
ply that the chelate 2b is looser than the phenolic chelates
and thus leads to time-averaging in the NMR measure- The calculations were run for gas-phase conditions and the geome-
tries were optimized with a DIIS algorithm using several starting
geometries.
ments. Moreover, inspection of the chelate 2e geometry
(Figures 14 and 15) shows that the minimum-energy struc-
tures exhibit an approximate σν symmetry with respect to
the pyridine groups, similar to structure A in Figure 12.
The deviation between the G values estimated from the
dynamic NMR measurements and from the computed en-
thalpies arises from the entropy, that is, the computed ener-
gies are obtained in the gas phase and in nondynamic con-
ditions. However, as the studied systems are very similar it
can reasonably be expected that the entropy factors in the
systems are very similar. Hence, inspection of the relative
differences is justified. Taking this into account, the experi-
mental and theoretical relative energies correlate with high
accuracy. Overall, this study shows that phenolic 2c–e zinc
chelates have a tendency to bind imidazole in a bidentate
manner, with the electron-poor para-nitro zinc–2d chelate
exhibiting the most efficient imidazole anion receptor char-
acter.
Ligand Titration Experiments: Dpa ligands 2a–e were dissolved in
CD3OD (0.75 mL) in an NMR tube to obtain concentrations of
63, 43, 47, 50, and 43 m for ligands 2a–e, respectively. Thereafter
the ligands were chelated with zinc by adding 1 equiv. of Zn-
(NO3)2 with respect to the Dpa group in each molecule. The zinc
chelates formed were titrated against imidazole (2  in MeOH).
Methyl 4-(Bromomethyl)benzoate (4): A few crystals of AIBN were
added to a solution of methyl 4-methylbenzoate (6.33 g, 42.1 mmol,
100 mol-%) and NBS (7.63 g, 43.1 mmol, 102 mol-%) in CCl4
(40 mL). The mixture was heated to 85 °C and more AIBN was
added. After heating at reflux (45 min) the color of the solution
changed rapidly from yellow to white and the mixture was cooled
down to room temp. The solid was filtered and the solution was
concentrated in vacuo. The yellow-white crude product was recrys-
tallized from EtOAc/hexane to give 5.80 g (60%) of white crystals.
Rf = 0.80 (CH2Cl2); m.p. 54.1–54.4 °C (ref.[31] 54–55 °C). 1H NMR
(400 MHz, CDCl3): δ = 3.92 (s, 3 H), 4.50 (s, 2 H), 7.46 (d, J =
8.3 Hz, 2 H), 8.02 (d, J = 8.3 Hz, 2 H) ppm. 13C NMR (400 MHz,
CDCl3): δ = 166.3, 164.8, 122.6, 120.9, 118.8, 52.3, 32.2 ppm.
Methyl 3,5-Bis(bromomethyl)benzoate (5): Methyl 3,5-dimethylben-
zoate (6.00 g, 36.5 mmol) and NBS (13.80 g, 77.5 mmol) were sus-
pended in CCl4 (75 mL). The mixture was heated before AIBN was
added in small portions while the flask was kept at reflux for 4 h.
The white precipitate (succinimide) was filtered off. The residue
was evaporated to dryness and recrystallized from EtOAc/hexane
(8 mL/80 mL) while stored overnight in a fridge to give 4.80 g
(41%) of white microcrystals. Rf = 0.67 (CH2Cl2); m.p. 93.9–
Conclusions
We have demonstrated that zinc chelates with significant
affinity towards imidazole binding can be developed. The
largest imidazole chelation affinity was measured for phe-
nolic Dpa ligands in which the phenolic oxygen is known
to form a bridge between two zinc ions when the imidazole
is not present. The NMR and molecular modeling results
show that imidazole binds in a bidentate manner to two
zinc ions forming a bridge. Moreover, the modeling results
imply that the imidazole bridge replaces the phenolic oxy-
gen zinc bridge of an unbound chelate. Experimental and
theoretical results also imply that the phenol chelate imid-
azole affinity is increased by electron-withdrawing groups
para to the phenol group. In terms of supramolecular chem-
istry, novel cascade receptors for imidazole binding have
been generated. The Dpa-based receptor binds cationic zinc
ions, which contribute to imidazole binding.
1
94.1 °C (ref.[32] 95–97 °C). H NMR (400 MHz, CDCl3): δ = 3.93
(s, 3 H), 4.50 (s, 4 H), 7.62 (br. t, J = 1.7 Hz, 1 H), 8.00 (d, J =
1.7 Hz, 2 H) ppm. 13C NMR (400 MHz, CDCl3): δ = 165.9, 138.9,
133.8, 131.4, 130.0, 52.3, 31.8 ppm. HRMS (ESI): calcd. for
C10H10Br2O2 [M + H]+ 322.9100; found 322.9095; ∆ = 1.5 ppm.
Methyl 3,5-Diformyl-4-hydroxybenzoate (6): Methyl 4-hydroxyben-
zoate (10.21 g, 66.4 mmol, 100 mol-%) and hexamethylenetetra-
mine (38.37 g, 273.7 mmol, 412 mol-%) were dissolved in anhy-
drous THF (70 mL) and the yellow solution was heated at reflux
for 4 d. Water (400 mL) was added to the viscous dark-orange solu-
tion and the mixture was heated up to give a homogeneous solu-
We are currently studying the imidazole affinity of these tion. After cooling, the product (12.09 g, 87%) was precipitated
slowly (3 h) from solution. Rf = 0.88 (10:1 CH2Cl2/MeOH); m.p.
118–119 °C. H NMR (400 MHz, CDCl3): δ = 3.96 (s, 3 H), 8.65
chelates with biomaterial samples. Further Dpa ligands are
also being synthesized to enable us to study whether the
zinc ligands can be modified to improve their binding affin-
ity and sensing properties.
1
(s, 2 H), 10.28 (s, 2 H), 12.05 (br. s, 1 H) ppm. 13C NMR (400 MHz,
CDCl3): δ = 166.3, 164.8, 122.6, 120.8, 118.8, 52.6 ppm. HRMS
(ESI): calcd. for C10H9O5 [M + H]+ 209.0450; found 209.0439; ∆
= 5.3 ppm.
Methyl 4-Hydroxy-3,5-bis(hydroxymethyl)benzoate (7): Benzoate 6
(1.70 g, 8.2 mmol, 100 mol-%) was dissolved in MeOH (100 mL)
and NaBH4 (0.35 g, 9.3 mmol, 113 mol-%) was added at 0 °C. Af-
ter stirring for 35 min, water (10 mL) was added and the solution
Experimental Section
General Methods: All reagents and solvents were purchased from
commercial suppliers and used without further purification with
Eur. J. Org. Chem. 2008, 3190–3199
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3197

T. Routasalo, J. Helaja, J. Kavakka, A. M. P. Koskinen
FULL PAPER
was concentrated in vacuo. The residue was made acidic (pH 2) by
adding 1  HCl and extracted with EtOAc (3ϫ100 mL). The or-
ganic phase was washed with brine (3ϫ100 mL) and dried
(Na2SO4). The solvent was evaporated to give the product (1.41 g,
81%) as a yellow solid. Rf = 0.48 (10:1 CH2Cl2/MeOH); m.p. 117–
118 °C. 1H NMR (400 MHz, CD3OD): δ = 3.85 (s, 3 H), 4.73 (s, 4
%) in CCl4 (30 mL). The mixture was heated to 85 °C and more
AIBN was added. After heating at reflux (2 h) the color of the
solution changed rapidly from yellow to white and the mixture was
cooled to room temp. The solid was removed by filtration and the
solution was concentrated in vacuo. The crude product was recrys-
tallized from EtOAc/hexane to give 2.86 g (43%) of light-yellow
H), 7.86 (s, 2 H) ppm. 13C NMR (400 MHz, CD3OD): δ = 168.7, crystals. Rf = 0.78 (30:1 CH2Cl2/MeOH); m.p. 85.4–85.6 °C (ref.[33]
1
159.3, 129.7, 128.4, 122.2, 61.6, 52.3 ppm. HRMS (ESI): calcd. for
83–85 °C). H NMR (400 MHz, CDCl3): δ = 4.03 (s, 3 H), 4.56 (s,
4 H), 7.12 (t, J = 7.6 Hz, 1 H), 7.38 (d, J = 7.7 Hz, 2 H) ppm.
13C NMR (400 MHz, CDCl3): δ = 156.6, 132.2, 131.9, 125.0, 62.2,
29.5 ppm.
C10H13O5 [M + H]+ 213.0763; found 213.0732; ∆ = 14.5 ppm.
Methyl 3,5-Bis(chloromethyl)-4-hydroxybenzoate (8): A solution of
benzoate 7 (1.17 g, 5.5 mmol, 100 mol-%) and SOCl2 (5 mL) was
stirred under argon at room temp. for 3.5 h. Excess SOCl2 was
evaporated and the residue was diluted with CHCl3 (40 mL) and
washed with water (3ϫ40 mL) and brine (3ϫ40 mL). The solvent
was removed to give the product (1.17 g, 85%) as light-yellow crys-
2,6-Bis(bromomethyl)phenol (13): Boron tribromide (1.1  CH2Cl2,
7.0 mL, 7.7 mmol, 151 mol-%) was added to anisole 12 (1.50 g,
5.1 mmol, 100 mol-%) at –78 °C. Stirring was continued at room
temp. for 3 h. The reaction was quenched with water (8 mL) and
diluted with EtOAc (40 mL). The organic phase was washed with
1
tals. Rf = 0.37 (10:1 CH2Cl2/MeOH); m.p. 142–144 °C. H NMR
(400 MHz, CDCl3): δ = 3.91 (s, 3 H), 4.71 (s, 4 H), 6.26 (br. s, 1 brine (2ϫ20 mL) and dried (Na2SO4). The solvent was evaporated
H), 8.01 (s, 2 H) ppm. 13C NMR (400 MHz, CDCl3): δ = 165.9,
157.2, 132.7, 124.6, 123.0, 52.2, 41.8 ppm. HRMS (ESI): calcd. for
C10H11Cl2O3 [M + H]+ 249.0085; found 249.0101; ∆ = 6.4 ppm.
to give the product (1.27 g, 89%) as a yellow solid. Rf = 0.69 (30:1
CH2Cl2/MeOH); m.p. 101–102 °C (ref.[34] 80–82 °C). 1H NMR
(400 MHz, CDCl3): δ = 4.57 (s, 4 H), 6.90 (t, J = 7.6 Hz, 1 H),
7.27 (d, J = 7.7 Hz, 2 H) ppm. 13C NMR (400 MHz, CDCl3): δ =
153.4, 131.4, 125.2, 121.2, 29.3 ppm.
(6-Nitro-4H-benzo[d][1,3]dioxin-8-yl)methyl Acetate (9): 4-Nitro-
phenol (6.98 g, 50 mmol, 100 mol-%) was added to a mixture of
paraformaldehyde (95%, 6.36 g, 201 mmol, 402 mol-%), acetic acid
(25 mL), and sulfuric acid (11 mL) at 55 °C. After 5 min, a yellow
precipitate formed. The mixture was cooled to room temp., acetic
acid (10 mL) was added and heating was continued (60 °C). After
heating at reflux for 22 h, the mixture was cooled to room temp.
and water (50 mL) was added. The mixture was neutralized with
solid K2CO3 (0.28 g, 203 mmol, 406 mol-%) and the grey-white so-
lid was filtered and washed with cold water. The product was
recrystallized from ethanol to give 7.09 g (56%) of 9 as a grey-
Synthesis of Chelate Methyl 4-[Bis(pyridin-2-ylmethyl)aminometh-
yl]benzoate (2a). Typical Procedure: A mixture of p-(halomethyl)-
benzoate
4 (2.29 g, 10.0 mmol, 100 mol-%), Dpa (2.05 mL,
11.0 mmol, 110 mol-%), K2CO3 (1.45 g, 10.5 mmol, 105 mol-%),
and DMF (10.0 mL) was stirred under argon at room temp. for
24 h. After stirring, the solution was diluted with EtOAc (25 mL)
and washed with water (3ϫ25 mL) and saturated NaCl
(3ϫ25 mL). The organic phase was dried (Na2SO4) and concen-
trated in vacuo to give the pure product (2.16 g, 62%). Rf = 0.60
1
white solid. Rf = 0.77 (10:1 CH2Cl2/MeOH); m.p. 115–116 °C (10:1 CH2Cl2/MeOH). H NMR (400 MHz, CDCl3): δ = 3.75 (s, 2
(ref.[27] 118 °C). 1H NMR (400 MHz, CDCl3): δ = 2.08 (s, 3 H), H), 3.81 (s, 4 H), 3.90 (s, 3 H), 7.15 (ddd, J = 7.2, 5.0, 1.0 Hz, 2
4.88 (s, 2 H), 5.07 (s, 2 H), 5.29 (s, 2 H), 7.81 (d, J = 2.6 Hz, 1 H),
8.05 (d, J = 2.8 Hz, 1 H) ppm. 13C NMR (400 MHz, CDCl3): δ
= 170.5, 155.2, 141.2, 125.3, 123.2, 121.2, 121.0, 91.9, 65.9, 59.8,
20.8 ppm.
H), 7.49 (d, J = 8.2 Hz, 2 H), 7.55 (d, J = 7.9 Hz, 2 H), 7.67 (dt,
J = 7.7, 1.8 Hz, 2 H), 7.98 (dt, J = 8.2 Hz, 2 H), 8.52 (d, J = 5.0 Hz,
2 H) ppm. 13C NMR (400 MHz, CDCl3) δ = 167.0, 159.3, 149.0,
144.6, 136.4, 129.6, 128.7, 122.8, 122.0, 100.1, 60.0, 58.2, 52.0 ppm.
HRMS (ESI): calcd. for C21H21N3NaO2 [M + Na]+ 370.1531;
found 370.1569; ∆ = 10.3 ppm.[35]
2,6-Bis(bromomethyl)-4-nitrophenol (10): Dioxine
9
(2.02 g,
8.0 mmol, 100 mol-%) was dissolved in HBr (48%, 60 mL) and the
mixture was heated at reflux for 16 h. The mixture was then cooled
to room temp. and the solid was filtered and washed with cold
water. The crude product was recrystallized from chloroform, dried
with THF, and recrystallized from EtOH/hexane to give 2.17 g
(86%) of a grey-white product. Rf = 0.24 (10:1 CH2Cl2/MeOH);
Methyl 3,5-Bis[bis(pyridin-2-ylmethyl)aminomethyl]benzoate (2b):
Yield 0.95 g (57%); Rf = 0.13 (10:1 CH2Cl2/MeOH). 1H NMR
(400 MHz, CDCl3): δ = 3.74 (s, 4 H), 3.81 (s, 8 H), 3.92 (s, 3 H),
7.13 (ddd, J = 7.2, 5.0, 1.5 Hz, 4 H), 7.57 (d, J = 7.7 Hz, 4 H), 7.62
(dt, J = 7.7, 1.6 Hz, 4 H), 7.71 (br. t, J = 1.5 Hz, 1 H), 7.94 (d, J
m.p. 149–150 °C (ref.[27] 146–147 °C). 1H NMR (400 MHz, = 1.5 Hz, 2 H), 8.52 (dd, J = 5.0, 0.7 Hz, 4 H) ppm. 13C NMR
CDCl3): δ = 4.57 (s, 4 H), 6.41 (br. s, 1 H), 8.22 (s, 2 H) ppm. 13C
NMR (400 MHz, CDCl3): δ = 158.3, 141.0, 126.6, 125.7, 27.1 ppm.
(400 MHz, CDCl3): δ = 167.1, 159.5, 149.0, 139.8, 136.4, 133.8,
130.2, 128.7, 122.8, 122.0, 60.1, 58.2, 52.1 ppm. HRMS (ESI):
calcd. for C34H35N6O2 [M + H]+ 559.2821; found 559.2824; ∆ =
0.5 ppm.
2,6-Dimethylanisole (11): 2,6-Dimethylphenol (8.00 g, 65.5 mmol,
100 mol-%) was dissolved in acetone (100 mL) and MeI (28.0 g,
200 mmol, 305 mol-%) and K2CO3 (18.5 g, 134 mmol, 204 mol-%) Methyl 3,5-Bis[bis(pyridin-2-ylmethyl)aminomethyl]-4-hydroxyben-
1
were added to the solution. The reaction mixture was heated at
reflux for 3 d and then the solvent was evaporated in vacuo. The
residue was dissolved in EtOAc (100 mL), washed with saturated
NaHCO3 (3ϫ100 mL) and brine (3ϫ100 mL), and dried
(Na2SO4). The solvent was evaporated to give 7.75 g (87%) of the
product. Rf = 0.52 (10:1 cyclohexane/EtOAc). 1H NMR (400 MHz,
CDCl3): δ = 2.28 (s, 6 H), 3.72 (s, 3 H), 6.91 (t, J = 7.4 Hz, 1 H),
zoate (2c): Yield 1.06 g (80%); Rf = 0.61 (6:1 CH2Cl2/MeOH). H
NMR (400 MHz, CDCl3): δ = 3.84 (s, 4 H), 3.88 (s, 3 H), 3.89 (s,
8 H), 7.13 (ddd, J = 7.3, 4.9, 1.1 Hz, 4 H), 7.49 (d, J = 7.9 Hz, 4
H), 7.61 (dt, J = 7.7, 1.8 Hz, 4 H), 7.95 (s, 2 H), 8.52 (dd, J = 4.9,
0.7 Hz, 4 H) ppm. 13C NMR (400 MHz, CD3OD): δ = 160.7, 159.1,
148.8, 136.6, 131.1, 124.4, 122.9, 122.0, 59.7, 54.5, 51.7 ppm.
HRMS (ESI): calcd. for C34H34N6NaO3 [M + Na]+ 597.2590;
7.00 (d, J = 7.7 Hz, 2 H) ppm. 13C NMR (400 MHz, CDCl3): δ = found 597.2610; ∆ = 3.3 ppm.
157.0, 130.8, 128.8, 123.7, 59.6, 16.0 ppm.
2,6-Bis[bis(pyridin-2-ylmethyl)aminomethyl]-4-nitrophenol
(2d):
2,6-Bis(bromomethyl)anisole (12): A few crystals of AIBN were
added to a solution of 2,6-dimethylanisole (3.07 g, 22.5 mmol,
100 mol-%) and N-bromosuccinimide (8.24 g, 46.4 mmol, 206 mol-
Yield 0.69 g (77%); Rf = 0.29 (10:1 CH2Cl2/MeOH). 1H NMR
(400 MHz, CD3OD): δ = 3.84 (s, 4 H), 3.89 (s, 8 H), 7.24 (ddd, J
= 7.5, 4.9, 1.1 Hz, 4 H), 7.53 (d, J = 7.9 Hz, 4 H), 7.71 (dt, J =
3198
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 3190–3199

Bis(2-picolyl)amine–Zinc Chelates for Imidazole Receptors
[13] a) A. Mokhir, R. Krämer, H. Wolf, J. Am. Chem. Soc. 2004,
126, 6208–6209; b) P. Jiang, Z. Guo, Coord. Chem. Rev. 2004,
248, 205–229.
[14] G. K. Walkup, S. C. Burdette, S. J. Lippard, R. Y. Tsien, J. Am.
Chem. Soc. 2000, 122, 5644–5645.
7.7, 1.8 Hz, 4 H), 8.06 (s, 2 H), 8.43 (dd, J = 5.1, 0.9 Hz, 4 H) ppm.
13C NMR (400 MHz, CDCl3): δ = 162.3, 158.6, 148.7, 139.4, 136.7,
125.3, 125.0, 122.8, 122.1, 59.6, 54.0 ppm. HRMS (ESI): calcd. for
C32H31N7NaO3 [M + Na]+ 584.2386; found 584.2368; ∆ = 3.1 ppm.
2,6-Bis[bis(pyridin-2-ylmethyl)aminomethyl]phenol (2e): Yield 0.88 g
(51%); Rf = 0.57 (10:1 CH2Cl2/MeOH). 1H NMR (400 MHz,
CDCl3): δ = 3.83 (s, 4 H), 3.88 (s, 8 H), 6.78 (t, J = 7.5 Hz, 1 H),
7.11 (t, J = 6.0 Hz, 4 H), 7.23 (d, J = 7.5 Hz, 2 H), 7.50 (d, J =
7.8 Hz, 4 H), 7.59 (t, J = 7.6 Hz, 4 H), 8.52 (t, J = 4.3 Hz, 4
H) ppm. 13C NMR (400 MHz, CDCl3): δ = 159.1, 155.8, 148.7,
136.4, 129.0, 123.9, 122.8, 121.8, 118.3, 59.6, 54.5 ppm. HRMS
(ESI): calcd. for C32H33N6O [M + H]+ 517.2716; found 517.2715;
∆ = 0.2 ppm.[18]
[15] N. C. Lim, C. Brückner, Chem. Commun. 2004, 9, 1094–1095.
[16] C. J. Chang, E. M. Nolan, J. Jaworski, S. C. Burdette, M.
Sheng, S. J. Lippard, Chem. Biol. 2004, 11, 203–210.
[17] a) A. Ojida, Y. Mito-oka, K. Sada, I. Hamachi, J. Am. Chem.
Soc. 2004, 126, 2454–2463; b) A. Ojida, M. Inoue, Y. Mito-
oka, I. Hamachi, J. Am. Chem. Soc. 2003, 125, 10184–10185;
c) A. Ojida, Y. Mito-oka, M. Inoue, I. Hamachi, J. Am. Chem.
Soc. 2002, 124, 6256–6258.
[18] a) D. H. Lee, S. Y. Kim, J.-I. Hong, Angew. Chem. Int. Ed.
2004, 43, 4777–4780; b) D. H. Lee, J. H. Im, S. U. Son, Y. K.
Chung, J.-I. Hong, J. Am. Chem. Soc. 2003, 125, 7752–7753.
[19] H. Jiang, E. J. O’Neil, K. M. DiVittorio, B. D. Smith, Org.
Lett. 2005, 7, 3013–3016.
[20] Y. Mito-oka, S. Tsukiji, T. Hiraoka, N. Kasagi, S. Shinkai, I.
Hamachi, Tetrahedron Lett. 2001, 42, 7059–7062.
[21] a) L. Fabbrizzi, G. Francese, M. Licchelli, A. Perotti, A. Tagli-
etti, Chem. Commun. 1997, 581–582; b) M. A. Hortala, L. Fab-
brizzi, N. Marcotte, F. Stomeo, A. Taglietti, J. Am. Chem. Soc.
2003, 125, 20–21.
Supporting Information (see also the footnote on the first page of
1
this article): Binding constant measurements for chelates 2b–e, H
and 13C NMR spectra of compounds 2a–e, Cartesian coordinates
for the geometry-optimized zinc imidazole chelate structures 2b–e
(Table 3).
Acknowledgments
[22] a) P. Dapporto, M. Formica, V. Fusi, L. Giorgi, M. Micheloni,
P. Paoli, R. Pontellini, P. Rossi, Inorg. Chem. 2001, 40, 6186–
6192; b) P. Dapporto, M. Formica, V. Fusi, M. Micheloni, P.
Paoli, R. Pontellini, P. Rossi, Inorg. Chem. 2000, 39, 4663–4665.
[23] H. Togo, T. Hirai, Synlett 2003, 5, 702–704.
[24] L. F. Lindoy, G. V. Meehan, N. Svenstrup, Synthesis 1998,
1029–1032.
Prof. Craig S. Wilcox, University of Pittsburgh, is gratefully ac-
knowledged for his advice on the chelate titration protocol. This
work was financially supported by the Finnish National Technol-
ogy Agency (TEKES) and EU COST action D28. T. R. also thanks
the Gustaf Komppa Fund of the Alfred Kordelin Foundation.
[25] C. Pulgarin, J. Gunzinger, R. Tabacchi, Helv. Chim. Acta 1985,
[1] D. P. Beer, A. P. Gale, Angew. Chem. Int. Ed. 2001, 40, 486–
516.
[2] D. P. Beer, S. R. Bayly, Top. Curr. Chem. 2005, 255, 125–162.
[3] B. Linton, A. D. Hamilton, Chem. Rev. 1997, 97, 1669–1680.
[4] G. Parkin, Chem. Rev. 2004, 104, 699–767.
[5] R. W. Strange, S. Antonyuk, M. A. Hough, P. A. Doucette,
J. A. Rodriguez, P. J. Hart, L. J. Hayward, J. S. Valentine, S. S.
Hasnain, J. Mol. Biol. 2003, 328, 877–891.
[6] a) H. Ohtsu, S. Itoh, S. Nagatomo, T. Kitagawa, S. Ogo, Y.
Watanabe, S. Fukuzumi, Inorg. Chem. 2001, 40, 3200–3207; b)
D. Li, S. Li, D. Yang, J. Yu, J. Huang, Y. Li, W. Tang, Inorg.
Chem. 2003, 42, 6071–6080, and references cited therein.
[7] E. J. O’Neil, B. D. Smith, Coord. Chem. Rev. 2006, 250, 3068–
3080.
[8] R. G. Lacoste, A. E. Martell, Inorg. Chem. 1964, 3, 881–884.
[9] J. K. Romary, J. D. Barger, J. E. Bunds, Inorg. Chem. 1968, 6,
1142–1145.
[10] More than 1000 Dpa–transition-metal chelate structures can
be found in the Cambridge Crystal Structure Database. The
most abundant metals are Cu, Fe, Zn, Mn, Ni, Co, Rh and Ru
in decreasing order.
68, 1948–1951.
[26] S. M. Dimick, S. C. Powell, S. A. McMahon, D. N. Moothoo,
J. H. Naismith, E. J. Toone, J. Am. Chem. Soc. 1999, 121,
10286–10296.
[27] J. de Mendoza, P. M. Nieto, P. Prados, C. Sánchez, Tetrahedron
1990, 46, 671–682.
[28] C. S. Wilcox in Frontiers in Supramolecular Organic Chemistry
and Photochemistry (Eds.: H. J. Schneider, H. Dürr), VCH,
Weinheim, 1991, pp. 123–143.
[29] At least two equilibria contribute to the imidazole binding
(mono vs. bidentate).
[30] H. Günther, The Influence of Dynamic Effects on NMR Spec-
tra, 2nd ed., Wiley, Chichester, 1995, chapter 9.
[31] R. C. Fuson, H. G. Cooke, J. Am. Chem. Soc. 1940, 62, 1180–
1183.
[32] P. Liu, Y. Chen, J. Deng, Y. Tu, Synthesis 2001, 14, 2078–2080.
[33] M. Ashram, D. O. Miller, J. N. Bridson, P. E. Georgiou, J. Org.
Chem. 1997, 62, 6476–6484.
[34] J. C. Rosa, D. Galanakis, A. Piergentili, K. Bhandari, C. R.
Ganellin, P. M. Dunn, D. H. Jenkinson, J. Med. Chem. 2000,
43, 420–431.
[11] J. Wirbser, H. Vahrenkamp, Z. Naturforsch., Teil B 1992, 47,
[35] D. R. van Staveren, E. Bothe, T. Weyhermüller, N. Metzler-
Nolte, Eur. J. Inorg. Chem. 2002, 6, 1518–1529.
Received: September 28, 2007
962–968.
[12] Y. Gultneh, A. R. Khan, D. Blaise, S. Chaudhry, B. Ahvazi,
B. B. Marvey, R. J. Butcher, J. Inorg. Biochem. 1999, 75, 7–18.
Published Online: May 6, 2008
Eur. J. Org. Chem. 2008, 3190–3199
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3199