From flexible to constrained tris(tetraamine) ligands: Synthesis, acid-base properties, and structural effect on the coordination process with nucleotides

Article abstract of DOI:10.1002/ejoc.201000495

This work presents the synthesis and characterization of L2 and L3, which formed with the previously described ligand L1 a family of three tris(tetraamine) ligands based on a mesitylenyl spacer, The trimeric ligands differ from one to the other by the nat

Full text of DOI:10.1002/ejoc.201000495

FULL PAPER
DOI: 10.1002/ejoc.201000495
From Flexible to Constrained Tris(tetraamine) Ligands: Synthesis, Acid–Base
Properties, and Structural Effect on the Coordination Process with Nucleotides
Anne-Sophie Delépine,[a] Raphaël Tripier,*[a] Michel Le Baccon,[a] and Henri Handel[a]
Keywords: Amines / Macrocycles / Proton sponges / Anions / Nucleotides / N ligands
This work presents the synthesis and characterization of L2
and L3, which formed with the previously described ligand
L1 a family of three tris(tetraamine) ligands based on a mesit-
ylenyl spacer. The trimeric ligands differ from one to the
other by the nature of their tetraamine moieties: triethylene-
tetraamine (L1), cyclen (L2), and constrained cyclen (L3). The
impact of the rigidification of the structure from L1 to L3 on
the acid–base properties is investigated, and the behavior of
the ligands towards triphosphate (TriP), adenosine mono-
phosphate (AMP), adenosine diphosphate (ADP), and adeno-
sine triphosphate (ATP) anion receptors is explored by
potentiometric investigations. To characterize and under-
stand the driving forces implicated in the supramolecular as-
semblies, results are supported by 1H and 31P NMR measure-
ments performed over a wide pH range. ATP appears as the
most complexed anion and particular attention is given to
comparison with its inorganic triphosphate analogue to high-
light the contribution of π-stacking interactions. Results un-
ambiguously show the various coordination scheme induced
by the structure of the ligands.
Introduction
ters. Polycyclic or polylinear tetraamine ligands presenting
an aromatic spacer were established in our previous investi-
gations as relevant hosts for inorganic phosphate anions
and nucleotides such as ATP. Ditopic linear octaamines
proved to be highly protonated and flexible receptors in
neutral medium[6] and cyclen-based bismacrocycles proved
to be rigid but more selective ligands for binding of ATP
or polyphosphates.[7] Furthermore, polytopic proton-
sponge ligands, based on cross-bridged cyclens, were pre-
sented as constrained but efficient receptors. In continua-
tion of these works focusing on the design of innovative
anions receptors,[6,7] we present here the synthesis of tri-
topic tetraamine ligands with a mesitylenyl spacer followed
by an acid–base study and their interaction with poly-
phosphate nucleotides by NMR and potentiometric investi-
gations. The trimeric derivatives described here consist of
three parent ligands that differ from one another by the
structure of their tetraamine moieties: triethylenetetraamine
for L1, cyclen for L2, and reinforced cyclen for L3
(Scheme 1). The linear, cyclic, or constrained nature of the
tetraamine moiety confers to the ligands different structures
and consequently different acid–base properties, leading to
various interactions with anionic substrates. The aim of the
study is then to appreciate the influence, from flexible to
rigid trimeric structures, on the coordination process. The
binding interactions are studied with three nucleotides,
namely, AMP (adenosine monophosphate), ADP (adeno-
sine diphosphate), and ATP (adenosine triphosphate), and
one inorganic anion: triphosphate (P3O105–).
Anions have a great relevance from a biological point of
view since a large number of cofactors and substrates in-
volved in biology are of anionic nature.[1] Anion recognition
came up as a privileged topic with the growing importance
of supramolecular chemistry these last decades. Adenosine
5Ј-triphosphate (ATP) is a ubiquitous substrate for many
biological reactions and generally regarded as an intracellu-
lar energy donor.[2] In consequence, studies focusing on the
recognition of nucleotides and especially ATP has increased
over the last years.[3] Generally, the binding forces are
mainly based on electrostatic interactions, and the associa-
tion constants may possibly be enhanced by providing ad-
ditional binding elements as hydrogen-bonding or π-stack-
ing interactions. Among the various hosts presented in the
literature,[4] polyamines and especially protonated poly-
amines are of special interest for anionic guest complex-
ation.[5] For charged synthetic polyamine studies, it is par-
ticularly important to determine the actual degree of pro-
tonation at a given pH and the distribution of the pH cen-
[a] UMR CNRS 6521, “Chimie, Electrochimie Moléculaires et
Chimie Analytique”, Université de Bretagne Occidentale,
6 avenue Victor Le Gorgeu, 29200 Brest, France
Fax: +33-2-98017001
E-mail: raphael.tripier@univ-brest.fr
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000495.
View this journal online at
wileyonlinelibrary.com
5380
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5380–5390

From Flexible to Constrained Tris(tetraamine) Ligands
Scheme 1. From flexible to constrained tris(tetraamine) ligands.
following a previously described procedure.[9] Reaction
of tris-electrophile 1,3,5-tris(iodomethyl)benzene (1 equiv.)
with 2 and 4 leads to tris-salts 3 and 5, respectively. Ligand
L2 is then obtained after treatment of 3 with hydrazine
monohydrate, and L3 is obtained by cleavage of the bis-
aminal bridges with NaBH4 in absolute ethanol,[12] respec-
tively, in 82 and 62% overall yield. Ligand L3 was isolated
Results and Discussion
Synthesis
The three azaligands were obtained by following the
bisaminal methodology (see Scheme 2 for L2 and L3 and
ref.[8] for L1), which presents the advantage of higher yields
by following similar routes to the focused polyazaligands
3+
as its LH3 form, which is consistent with previous works
(polylinear,[8] polycyclic,[9,10] and constrained ligands[11]
)
presenting constrained ligands as proton sponges.[13]
through the selective N-alkylation of linear or cyclic tetra-
amines. Synthesis of L1 was described previously.[8] Macro-
cyclic bisaminal 2 is cleanly obtained from cyclen 1 and
leads easily to 4 by reaction with iodomethane (1 equiv.)
Ligand Protonation
The protonation equilibria of the three ligands were in-
vestigated in aqueous solution at 25 °C by potentiometric
pH (–log[H+]) measurements. The logarithms of the step-
wise protonation constants for the three studied ligands, ob-
tained from mathematical treatment of their respective
potentiometric titration data, are presented in Table 1.
Table 1. Logarithm of protonation constants of tritopic ligands
[H2O; I = 0.1  (NaCl); T = 25 °C].
Equilibrium[a]
L1[b]
L2[b]
L3[b]
L + H = LH
LH + H = LH2
10.55 (3)
9.85 (2)
9.38 (2)
8.71 (3)
8.26 (1)
6.63 (2)
5.93 (2)
4.30 (2)
4.25 (2)
0.70
11.00 (1)
9.80 (1)
9.21 (1)
8.30 (1)
8.20 (1)
6.92 (2)
3.00 (2)
–
–
–
LH2 + H = LH3
LH3 + H = LH4
LH4 + H = LH5
LH5 + H = LH6
LH6 + H = LH7
LH7 + H = LH8
LH8 + H = LH9
∆1 = logK011 – logK012
∆2 = logK012 – logK013
4.60 (1)
4.26 (1)
3.23 (2)
1.20
0.59
–
–
0.47
[a] Charges are omitted for clarity. [b] Values in parentheses are the
standard deviations in the last significant digit.
The three ligands differ from one another in the nature
of their 12 amine functions. Ligand L1 presents 6 primary,
3 secondary, and 3 ternary amines, whereas L2 possesses
9 secondary and 3 ternary amines; L3 presents 12 ternary
amines.
In the investigated pH range (2–12), for the 3 ligands,
the full set of the 12 protonation constants could not be
determined. In spite of its acyclic nature, only nine proton-
ation constants were determined for L1. Six correspond to
Scheme 2. Synthesis of L2 and L3. Reagents and conditions:
(i) Glyoxal, MeOH; (ii) MeI, THF; (iii) 1,3,5-tris(iodomethyl)-
benzene (1/3 equiv.), CH3CN; (iv) NH2NH2,H2O; (v) NaBH4,
EtOH (abs.).
Eur. J. Org. Chem. 2010, 5380–5390
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5381

A.-S. Delépine, R. Tripier, M. Le Baccon, H. Handel
FULL PAPER
strong to moderate bases and three to moderate bases; the
three last constants lower than 2 were not determined. In
contrast, L2 exhibits only six nitrogen atoms that behave as
strong to moderate bases, and one which behaves as a weak
base; the other protonation constants were not detected.
Trismacrocyclic ligand L3 behaves as a very strong base
(proton sponge) in the three first protonation steps; in fact,
deprotonation of LH33+ is not detectable by potentiometric
1
titration (and H NMR experiments as shown below). The
three protonation constants behave as moderate to weak
bases.
Observing the stepwise protonations in terms of minimi-
zation of electrostatic repulsions between positive charges
in protonated species, the first six protons occupy alternate
positions separated either by a nonprotonated amino group
or by the linker. The first set of three basicity constants
concerns the protonation of each tetraamine core, whereas
the second set corresponds to the introduction of a supple-
mentary proton on each polyaza subunit, and so on.
Ligand L1 presents high protonated species, a conse-
quence of its flexible structure, which allows full proton-
ation of the two tetraamine pendant arms without impor-
tant repulsion charge effects. The ∆1 and ∆2 values (∆1 =
logK011 – logK012 = 0.70, ∆2 = logK012 – logK013 = 0.47)
indicate that the tetraamine moieties of the ligand are rather
independent. The cyclic structure of each tetraamine of L2
leads to fewer protonated species. Moreover, the value ∆1 =
1.20 indicates that the three cyclen moieties are relatively
dependent, a consequence of the rigidity of the whole li-
gand skeleton. However, the weak value of ∆2 = 0.59 indi-
cates the independence and easy protonation of the third
cyclen core towards the others. The proton sponge behavior
of L3 generates an entirely different protonation scheme.
The fourth, fifth, and sixth protonation constants are only
observed at very low pH (logK014 = 4.60, logK015 = 4.26,
Figure 1. Species distribution diagrams for L1 (top), L2 (middle),
and L3 (bottom) as a function of pH.
1
Depending on the structure of the ligand, the H NMR
logK016 = 3.23) due to the presence and the central localiza- spectra of L1, L2, and L3 at neutral pH in D2O
tion (see below) of the three first protons in each reinforced (300.13 MHz, 298 K) display five or seven signals in the
macrocycle. This uncommon behavior is ascribed to the aliphatic area (H1 to H7); they also present resonances in
constrained architecture of the bridged cyclen cores.[11–13]
Figure 1 depicts the distribution diagrams of the proton- Ha proton.
the aromatic area near 7.3 ppm attributed to the aromatic
ated species of L1, L2, and L3. The diagrams are notably
The time-averaged C3v symmetry due to a fast proton
different. Concerning firstly triscyclen L2, one can note the exchange process in aqueous solution is kept across the
remarkably large (LH66+ + LH77+) existence domain species whole investigated pH range for both ligands L1 and L3.
that predominate at pH 2–7. Comparison with its open One can however think that steric hindrance occurs from
chain analogue L1 is interesting: in sharp contrast, the zone the open-chain skeleton to the cyclic one. Moreover, be-
6+
of the (LH6 + LH77+) forms is considerably reduced and tween pH 2 and 5, L2 NMR investigation shows two sub-
9+
the highly protonated form LH9 largely predominates in spectra (Supporting Information, Figure S1) due to a
7+
the acidic pH area. By increasing the pH, the lower proton- slower exchange process for the LH7 species. This behav-
ated species of L1 and L2 progressively appear, and at ior can be linked to the rigidity of the structure induced by
pH 12, the fully deprotonated ligands become the main spe- the high protonation state of each macrocyclic unit in acidic
cies. Concerning L3, one can note the remarkable predomi- medium. For clarity, time averaged shifts have been selected
3+
nance of LH3 at pH 5–12.
for Figure 2 in the concerned pH region.
To throw more light on the protonation sequence of the
The analysis of the pH dependence of the 1H NMR reso-
1
three ligands, H NMR measurements were carried out in nances of L1 is quite complicated due to the number of
aqueous solutions over a wide pH range. The proton attri- signals in the 2.8–3.6 ppm area. From basic to acidic me-
bution was elucidated with the help of HMQC, HMBC, dium, the H5 and H7 protons firstly present a downfield
and COSY NMR sequences. Figure 2 illustrates the pH de- shift relating the preferential protonation of the more dis-
1
pendence of the H NMR signals.
tant amines N3 and N4. The second step consists in the
5382
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5380–5390

From Flexible to Constrained Tris(tetraamine) Ligands
1
The pH dependence of the H NMR resonances of L2
also allows us to deduce its protonation pattern. At pH 12–
7, where the first six protons bind to the ligand, a remark-
ably higher downfield shift is observed for the signals of H3
and H4. Minor shifts are observed for the remaining proton
signals. This indicates that the first six protonations take
place on the N2 and N4 secondary amines. Therefore, in
6+
LH6 species, which is prevalent in aqueous solution at
pH 2–6, each macrocyclic unit contains two protonated ni-
trogen atoms, separated from one another by an uncharged
nitrogen atom (one secondary and one ternary amine), thus
achieving an optimal minimization of the electrostatic re-
pulsions between charged ammonium groups. Binding of
the seventh and higher protons below pH 4 is followed by
the downfield shift in the resonances of the H5 and H2
protons in the 1H NMR spectra. These spectral features
indicate that the last protonation steps occur on N3 then
on N1 amino groups.
Ligand L3 shows a somewhat different behavior. As
shown in Figure 2, the presence of three protons in the
1
structure leads to unusually high chemical shifts in the H
NMR resonances of the H1–7 protons, dealing with a cen-
tral place of the proton in each constrained cycle. The
analysis of the whole spectra may indeed suggest the local-
ization of the three acidic protons in the LH33+ species. The
methylene groups H7 adjacent to N2 and N4 give rise to
two signals with high chemical shifts for a CH2 group adja-
cent to the ternary nitrogen atoms. This suggests that the
acidic proton is shared between the N2 and N4 nitrogen
atoms. This clearly highlights the strong stability of the
3+
LH3 species and the fact that the following protonations
take place in quite acid medium; this confirmed the
1
potentiometric data. The H NMR spectra of the ligand in
concentrated NaOH, at pH 12 or at pH 5, are equal, indi-
3+
cating that the LH3 species is present from pH 5 to basic
medium and could not be deprotonated even in strongly
alkaline solution, that is, L3 behaves as a triple proton
sponge. Finally, bindings of the fourth, fifth, and sixth pro-
tons occur, producing once again general downfield shifts
of the H1–7 protons.
One can finally note that the aromatic Ha resonances
follow very different pathways depending on the nature of
the tetraamine moieties. For L1, no real variation can be
detected in the whole pH range, whereas for L2 and L3, the
Ha protons are sensitive to the protonation happening in
acidic medium certainly as a consequence of the nitrogen
N1 protonation.
The most interesting demonstration is that from a flexi-
ble to a constrained parent tritopic structure, the acid–base
properties of the ligands are deeply different. Ligand L1
presents more protonated species than L2 and is the most
protonated at neutral pH. The reinforced and more rigid
analogue L3 is poorly charged around pH 7 but presents a
1
Figure 2. pH dependence of the H NMR signals of L1, L2, and
L3 (300.13 MHz, D2O, 25 °C; [L] = 0.02 ). Caution: scale is dif-
ferent in the 7–8 ppm area.
3+
very stable LH3 form. In addition, this necessary prelimi-
downfield shift of H3 and H4, revealing the protonation of nary study clearly allows, for each ligand, the identification
the secondary amine N2, followed in very acidic medium of the ammonium sites at a given pH and then the potential
by the slight downfield in H2 and H6 induced by the pro- anchoring points involved in the following host–guest inter-
tonation of the ternary amine N1.
action.
Eur. J. Org. Chem. 2010, 5380–5390
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5383

A.-S. Delépine, R. Tripier, M. Le Baccon, H. Handel
FULL PAPER
Phosphate Anions Binding
Table 2. Logarithm complexation constants, logKalh, for ligands
L1, L2, and L3 with anions [H2O; I = 0.1  (NaCl); T = 25 °C].
Potentiometric and NMR investigations were performed
to evaluate and characterize the binding interactions be-
tween the ligands and the anions. The potentiometric data
of a solution containing equimolar amounts of ligand and
anion are resolved, giving the logKalh values for the species
present in solution. Computer analysis furnished the overall
equilibrium constants βalh of the complexes formed between
the anion (A) and the ligand (L), according to Equation (1)
(charges omitted for simplicity). The knowledge of the
number i of protons bound to the bis(tetraamine) ligand
L1[a]
AMP[b]
ADP[b]
ATP[b]
TriP[b]
A+LH = ALH
2.34 (1)
2.60 (1)
2.81 (1)
3.06 (1)
3.76 (1)
4.20 (2)
2.46 (1)
2.69 (1)
3.21 (1)
3.51 (1)
4.26 (1)
4.73 (2)
3.16 (2)
3.40 (3)
3.60 (3)
4.03 (3)
4.76 (5)
2.72 (3)
3.04 (2)
3.45 (2)
3.91 (4)
4.72 (2)
A+LH2 = ALH2
A+LH3 = ALH3
A+LH4 = ALH4
A+LH5 = ALH5
A+LH6 = ALH6
AH+LH5 = ALH6
AH+LH6 = ALH7
AH+LH7 = ALH8
AH+LH8 = ALH9
6.05 (4)
6.51 (4)
6.43 (4)
6.53 (5)
5.26 (3)
6.01 (4)
5.77 (5)
5.44 (4)
4.30 (2)
4.47 (2)
3.95 (3)
3.34 (3)
3.87 (3)
4.67 (2)
5.20 (2)
4.67 (3)
4.55 (3)
4.69 (3)
(L) in the general AaLlHh complex allows the complexation AH+ LH9 = ALH10
reaction in Equation (2) to be written, according to the ac-
AH2+LH8 = ALH10
5.53 (5)
ATP
3.13 (4)
TriP
tual protonation state (charges omitted) and to calculate
logKalh
L2
AMP
ADP
.
A+LH2 = ALH2
A+LH3 = ALH3
A+LH4 = ALH4
A+LH5 = ALH5
A+LH6 = ALH6
AH+LH5 = ALH6
AH+LH6 = ALH7
AH2+LH6 = ALH8
AH2+LH7 = ALH9
AH3+LH7 = ALH10
3.41 (1)
3.60 (1)
4.47 (1)
4.77 (2)
5.91 (2)
2.91 (1)
3.01 (1)
3.64 (2)
4.10 (2)
2.48 (1)
3.10 (1)
3.09 (1)
3.94 (2)
aA + lL + hH h AaLlHhβalh
(1)
3.05 (1)
3.37 (3)
4.13 (3)
AH(h–i) + LHi h ALHhKalh
(2)
4.20 (2)
5.00 (2)
5.40 (2)
5.59(2)
3.53 (2)
4.31 (2)
4.24 (3)
4.48 (3)
5.31 (3)
5.59 (3)
5.88 (2)
6.50 (1)
6.96 (1)
7.01 (1)
Protonation constants of the nucleotides, measured ac-
cording to our experimental conditions and in good agree-
ment with those of the literature, are reported in the Sup-
porting Information (Table S1). Species distribution dia-
grams for substrates AMP, ADP, ATP, and triphosphate,
deduced from their protonation constants, are presented in
the Supporting Information (Figure S2). The logarithm of
the stability constants, logKalh, of ligands L1, L2, and L3
with AMP, ADP, ATP, and triphosphate anions is pre-
sented in Table 2. The protonation constants of the anionic
guests provide several sets of ternary species: in each case,
the most probable equilibrium is kept. The overall stability
constants (logβ) of the substrates adduct with the three li-
gands are proposed in the Supporting Information
(Table S2).
Data analyses under our experimental conditions reveal
that only complexes with a 1:1 anion/ligand stoichiometry
were found in solution. The species distribution diagrams
as a function of pH for the six anion–ligand systems were
carried out and are presented in the Supporting Infor-
mation (Figure S3a–c). By comparing the binding ability of
the three ligands towards the various substrates, the sta-
bility of the adducts increases in the order AMP Ͻ ADP
= triphosphate Ͻ ATP. This order was already previously
obtained for similar ligands.[7d,11c] For L1, the stability con-
stants are quite significant from logK111 to logK114, around
logK = 3–4 considering the weak protonation degree of the
systems. The interaction process involving L2 and L3 re-
quired more protonated sites.
L3
AMP
ADP
ATP
TriP
A + LH3 = ALH3
AH + LH3 = ALH4
AH + LH4 = ALH5
AH2 + LH4 = ALH6 3.72 (2)
AH2 + LH5 = ALH7 4.04 (3)
AH2 + LH6 = ALH8 4.51 (3)
AH3 + LH8 = ALH9
2.36 (1)
2.74 (1)
3.66 (1)
2.99 (1)
3.29 (1)
4.46 (1)
4.61 (4)
4.68 (3)
5.31 (3)
3.02 (1)
3.24 (1)
4.63 (1)
4.73 (2)
4.68 (2)
5.29 (3)
6.07 (3)
2.77 (1)
2.86 (1)
3.54 (1)
3.98 (2)
4.01 (2)
5.00 (3)
5.41 (3)
[a] Charges are omitted for clarity. [b] Values in parentheses are the
standard deviations in the last significant digit.
strates form relatively stables complexes with the three re-
ceptors in a wide pH range with various maxima depending
on the nature of the ligand.
The comparison of the logKalh values at a given and
equivalent degree of protonation, between the different li-
gands is, in this study, useless, considering their high differ-
ence of protonation state at a given pH value. For that
reason, studying the pH dependence of the logarithms of
the conditional constants for the different systems is more
Figure 3. Logarithms of the conditional constant versus pH for the
three systems L1, L2, L3 – ATP, triphosphate.
Concerning L1, the stability constants increase from al-
relevant. Special attention is directed to triphosphate and kaline to slightly acidic pH values (substrates are in their
ATP anions in Figure 3. Plots of the conditional constants higher charged forms and ligands in an increasing number
versus pH for the different systems show that both sub- of ammonium sites) with a maximum between pH 6 and 7
5384
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5380–5390

From Flexible to Constrained Tris(tetraamine) Ligands
for both substrates and decrease at higher pH values. This dence interactions between the phosphate chains and the
marked decrease can be attributed to the presence in solu- polyammonium receptors. However, differences between the
tion of highly or fully protonated forms of substrates, im- behaviors of the three ligands for a given substrate clearly
plying repulsions between the highly charged tetraamine emphasize various coordination modes depending on their
moieties. For L2 and L3, logKalh increases from alkaline organic or inorganic nature.
medium to attain a maximum in very acidic medium in
In the case of complexation of ATP, Figure 4 shows a
spite of the lack of negative charge on the substrates. This downfield shift upon complexation of the signals of the
behavior, which was observed in previous work[7,8] for poly- phosphate groups in the order Pγ Ͼ Pβ Ͼ Pα. This se-
cyclic ligands, can be reasonably attributed to the lesser pro- quence is comparable to previous results[6a,7d] and suggests
tonation states of both trimeric ligands and to the rigidity that in the case of ATP binding, the ligands principally bind
induced by the cyclic nature of the tetraamine moieties. The the phosphate chain by interaction with the two terminal
situation involving a N+–H···OH interaction,[5d] effective in phosphate groups of the nucleotide. Moreover, the NMR
acidic solution can be invoked here.
shifts are higher for L1 than for L2 (especially for Pβ),
These data can be expressed in terms of the percent of whereas the phosphate groups are less influenced by the
formation of the adducts. Comparison between the presence of L3. The downfield shifts slightly persist in alka-
amounts of the ternary species formed (Supporting Infor- line medium for L1 and L2 but are mainly significant at
mation, Figure S4a–c) confirm the affinity of these ligands pH 2–9.
for the anionic substrates, but each one at different pH:
In conclusion, for ATP complexation, variations in
pH 4 Ͻ L1 Ͻ pH 8 (ATP 97.83%, triphosphate 95.30%), chemical shifts are pH dependent, being higher at pH 3–9.
pH 2 Ͻ L2 Ͻ pH 7 (ATP 98.94%, triphosphate 94.07%), This is in accordance with the potentiometric results of the
and pH 2 Ͻ L3 Ͻ pH 4 (ATP 95.78%, triphosphate same systems, which showed that a large amount of recep-
92.8%).
tor–substrate adducts are formed from slight alkaline to
As explained in numerous works investigating ATP bind- acidic pH values, that is, in the pH region where the highly
ing (Scheme 3), the differences observed in the host–guest protonated species of the receptor and the anionic species
interactions can be only partially explained in terms of elec- are simultaneously present in solution.
trostatic and hydrogen-bonding interactions. In addition to
Concerning the case of triphosphate (Figure 5), which
the electrostatic interaction between the phosphate chain obviously presents only two different phosphate groups Pα
and the polyammonium ligands, other effects, even as and Pβ, both are clearly influenced by the presence of the
minor as π-stacking interactions between the heteroarom- receptor in solution since the resonances of Pα and Pβ shift
atic linker and the adenine moiety of the nucleotides, have downfield over a wide pH range, indicating that in the ad-
to be considered. Consequently, in light of the better results ducts with triphosphate, the polyammonium receptors in-
obtained with ATP, we investigated the structural features teract equally with Pα and Pβ. However, some significant
of the complexes with ATP by comparing the behavior of differences can be pointed out depending on the pH do-
5–
the three ligands with its inorganic analogue P3O10 by main. Ligands L1 and L2 present quite similar and classical
using NMR measurements. In this way, the anion inter- diagrams with highest chemical shifts situated around
1
action was been followed by recording H and 31P NMR pH 5–6 and decreasing ones when going to the acidic and
spectra of solutions containing an equimolar amount of re- alkaline media. Ligand L3 shows an entirely different be-
ceptors and substrates in a wide pH range.
havior with chemical shifts of both the Pα and Pβ signals
being higher from basic to acidic medium.
The 31P NMR spectroscopic titrations clearly confirm
that electrostatic interactions and hydrogen bonds between
the ammonium sites of the ligands and the polyphosphate
chains of the four anionic substrates govern the host–guest
complexation process. Results definitely show important
differences in the formation of the adducts with ATP or
with triphosphate, highlighting the active role played by the
organic moiety of the nucleotides. In the aim to clarify this
point, 1H NMR spectra were carried out on a solution con-
taining an equimolar amount of the substrates and the li-
gands with ATP or triphosphate and L1–3. Concerning the
shifts relating to the tetraamine moieties, results are quite
similar for both anions. Plots involving ATP are presented
Scheme 3. ATP and triphosphate anions and labeled atoms.
Figures 4 and 5 show the pH dependence of the 31P in Figure 6 and have been compared to Figure 2 (see Fig-
NMR chemical shifts of the phosphate groups of ATP and ure S5 in the Supporting Information for a comparison in
1
triphosphate in a 1:1 molar ratio with ligands L1, L2, and a same Figure). In addition, the H NMR investigations
L3 at different pH values together with those of the free permit the localization of the ammonium sites involved in
anions. The variations in the 31P NMR chemical shifts the complexation. From alkaline to acidic medium, the pro-
clearly observable in the plots of the different diagrams evi- gressive and simultaneous shifts of the protons of L1, espe-
Eur. J. Org. Chem. 2010, 5380–5390
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5385

A.-S. Delépine, R. Tripier, M. Le Baccon, H. Handel
FULL PAPER
Figure 5. 31P NMR chemical shifts of the triphosphate phosphorus
atoms at various pH values (121.49 MHz, D2O, 25 °C; [A] = [L] =
0.02 ). Free (dotted lines) or in presence of the ligand (solid lines);
α ᭺, β ᮀ.
Figure 4. 31P NMR chemical shifts of the ATP phosphorus atoms
at various pH values (121.49 MHz, D2O, 25 °C; [A] = [L] = 0.02 ).
Free (dotted lines) or in presence of the ligand (solid lines); α ᭺, β
ᮀ, γ ᭝.
5386
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5380–5390

From Flexible to Constrained Tris(tetraamine) Ligands
cially H3, H4, H5, and H7 however indicate that the coordi- H6 at pH 2–5 indicate the additional participation of N1 in
nation responsibility is shared by all nitrogen atoms N2, this domain. The spectrum analysis is easier with L2, which
N3, and N4. Moreover, the downfield shifts of H1, H2, and shows downfield shifts of the H3 and H4 signals, indicating
that N2, and therefore N4, are more implicated in the bind-
ing system of the complex in the area 3 Ͻ pH Ͻ 12. The
downfield shift of the signal of H5 (protonation of N3),
Figure 6. Experimental 1H NMR chemical shifts for the protons of
the trismacrocycles in the presence of ATP (300.13 MHz, D2O,
25 °C; [A] = [L] = 0.02 ). Caution: scale is different in the 7–
8 ppm area.
Figure 7. Experimental 1H NMR chemical shifts for the HA1Ј,
HA2, and HA8 protons of free ATP (dotted lines) and ATP in the
presence of ligands (solid lines) (300.13 MHz, D2O, 25 °C; [A] =
[L] = 0.02 ).
Eur. J. Org. Chem. 2010, 5380–5390
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5387

A.-S. Delépine, R. Tripier, M. Le Baccon, H. Handel
FULL PAPER
which occurs when pH Ͻ 3, clearly indicate that in acidic
The protonation study clearly underlines the influence of
medium the complexation is reinforced by the participation the structure of the tetraamine moiety on the acid–base
of N3. Results obtained for L3, especially in acidic medium properties of the trimeric ligands. The tetraamine open
below pH 5, are remarkable. If between pH 5 and 12 the chain leads to the more basic receptor L1, whereas the cy-
weak host–guest interaction leads to no apparent shifts in clic structure present in L2 constitutes an evident drawback
1
the H NMR signals, strong downfield shifts are simulta- in terms of repulsions between charged nitrogen atoms
neously observed for the H3, H4, H5, and H6 signals as leading to less protonated species at a given pH. The con-
well as for H1. This indicates that all the nitrogen atoms strained structure of the cyclen moieties present in L3 leads
contribute to the coordination scheme in acid medium.
to an attractive compound that behaves as a triple proton
3+
Concerning the aromatic proton Ha, results are obvi- sponge. In aqueous solution, L3 generates LH3 species
ously different since plots presenting experimental 1H that cannot be deprotonated even in very basic medium.
NMR shifts of the free ligands and in presence of ATP or
The three ligands show significant interaction properties,
triphosphate (Supporting Information, Figure S6a–c) only especially with ATP, and their differences mainly give rise
present significant displacements in the case of adducts to comparable stepwise stability constants but in different
with ATP. The observed upfield shifts can already be inter- pH domains. Spectroscopic investigations clearly highlight
preted by interaction between the aromatic centers of the that the formation of the adducts is significantly different
1
receptors and the nucleotides. H NMR spectra can also for each receptor, showing that the lack of protonated sites
provide evidence for the contribution of π-staking interac- from L1 to L3 is partially compensated by the rigidification
tion in the stabilization of ternary species. The pH depen- of the structures. On one hand, the cyclic nature of L2 and
dence of the signals of the adenine protons HA2 and HA8 L3 leads to high interactions in acidic medium. On the
and the anomeric proton HA1Ј of the nucleotide for ATP other hand, the ability of their open chain analogue L1 to
free and in presence of the ligands is displayed in Figure 7. form highly protonated species implies strong charges re-
The significant upfield displacements observed for these pulsions in acidic medium leading to weak adducts between
three protons are consistent with the participation of inter- pH 2 and 6.
actions between the aromatic spacer and the adenine moi-
Spectroscopic titrations also allow the complexation of
ety in the stabilization of the adducts. This behavior exists triphosphate and ATP to be compared. Both lateral and
at pH 2–10 for L1 and L2 but is limited for L3 to lesser central phosphate groups of the triphosphate interact with
shifts and to a smaller pH range (2–5), which corroborates the ligands, whereas only the terminal and central ones are
the fact that the organic part of the anion is less involved implicated in nucleotide recognition. Moreover, the pres-
in the coordination scheme than for L1 and L2. This corre- ence of the adenine part necessarily permits the contri-
lates that the potentiometric data are quite similar between bution of the π-stacking interactions, which induces a dif-
triphosphate and ATP in the case of adducts formed with ferent coordination scheme stabilizing the ternary species;
L3.
this contribution is higher in the case of L2.
1
Potentiometric and 31P and H NMR titrations clearly
Taking into account that this host–guest study is per-
highlight that the three ligands are efficient receptors for formed at 25 °C and that the interactions between anionic
nucleotides and inorganic phosphate anions. Nevertheless, substrates and ammonium receptors could be temperature
the differences in the structures of L1, L2, and L3 clearly dependant, especially in the case of such ligands presenting
lead to various potentiometric results in very acidic or alk- so different rigidity, results are subject to variations, for ex-
aline media, which come from different complexation ample, in physiological samples. Nevertheless, this study
modes. For the different ligands, the coordination sites N1, clearly demonstrates that with nucleotides such as ATP the
N2, N3, and N4 are not involved in the same pH domain contribution of three important factors that have the ability
in the complexation process and the influence of π-stacking to form ammonium sites, the rigidity of the structure and
interactions is not comparable, leading obviously to dif- the possibility to form π-stacking interactions is different
ferent coordination schemes at a given pH value.
depending on the nature of the trimeric ligands and leads
however to significant interactions with the targeted sub-
strates.
Conclusions
Obtained in high overall yields following the easy to run
bisaminal methodology, the three tris(tetraamine) ligands
proposed here can be involved in host–guest investigations
to assess their acid–base properties and their ability to in-
teract with triphosphate and nucleotides. The synthetic
routes used in this paper give alternative access to already
known triscyclen[14] L2 without fastidious protection–de-
protection steps and give rise to a simple access to the new
tris-reinforced-cyclen L3.
Experimental Section
Materials: Reagents were purchased from Acros Organics and from
Aldrich Chemical Co. Cyclen and triethylenetetraamine were pur-
chased from Chematech (Dijon, France). Elemental analyses were
performed at the Service de Microanalyse, CNRS, 91198 Gif sur
Yvette, France. NMR and mass spectrometry were investigated at
the “services communs” of the University of Brest. 1H and 13C
NMR spectra were recorded with an Avance 500 Bruker
(500 MHz), Avance 400 Bruker (400 MHz), or AMX-3 300 Bruker
5388
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5380–5390

From Flexible to Constrained Tris(tetraamine) Ligands
(300 MHz) spectrometer by using MeOH as internal reference. 2D
1H–1H homonuclear and 1H–13C heteronuclear correlations and
homonuclear decoupling experiments permitted the full assignment
was extracted with chloroform (3ϫ70 mL). The combined organic
phase was finally evaporated to give L3H33+,3I– (78% yield). 1H
NMR (400.13 MHz, D2O, 298 K): δ = 2.55 (s, 3 H, Me, H6), 2.77
(t, 4 H, αCH2, H2), 2.83 (t, 4 H, αCH2, H5), 2.88 (t, 4 H, αCH2,
H4), 2.94 (t, 4 H, αCH2, H3), 3.02 (t, 4 H, αCH2, H7), 3.81 (s, 2
H, αCH2Ph, H1), 7.29 (s, 1 H, CHAr, Ha) ppm. 13C NMR
(75.47 MHz, D2O, 298 K): δ = 45.2, 49.9, 53.6, 55.1, 56.9, 57.5
(αCH2), 60.5 (αCH2-Ph), 132.4 (CHAr), 141.6 (CAr) ppm. MS
1
of the H and 13C NMR signals. Mass spectrometry analyses were
performed on an Autoflex MALDI TOF III LRF200 CID.
L1: Synthesized according to ref.[8]
Cyclen–Glyoxal 2: Synthesized according to ref.[9] starting from cy-
clen 1.
(MALDI-TOF, DHB, H2O): m/z
= 751.64, [M +
H]+.
C42H78N12·3HI·2H2O (1170.92): C, 43.08; H, 7.32; N, 14.35; I,
32.51; found C, 43.31; H, 7.02; N, 14.44; I, 32.80.
Compound 3: A solution of 1,3,5-tris(iodomethyl)benzene (4.27 g,
8.58 mmol) dissolved in dry acetonitrile (35 mL) was added drop-
wise to a stirred solution of cyclen–glyoxal 2 (5 g, 25.75 mmol) in
dry acetonitrile (40 mL). The mixture was stirred at room tempera-
ture for two weeks. The precipitate was collected by filtration,
washed with diethyl ether, and dried in vacuo to give 3 as a white
powder (95% yield). 13C NMR (100.62 MHz, D2O, 298 K): δ =
46.5, 50.5, 51.1, 51.2, 54.2, 59.8, 62.8, 64.4 (αCH2), 74.4, 86.5
(CHaminal), 133.2 (CHAr), 141.4 (CAr) ppm. MS (MALDI-TOF,
Ligands Used for Analysis: Chlorohydrated derivatives of L1, L2,
and L3 were obtained by adding concentrated HCl to an ethanolic
solution of the ligand at 0 °C. The resulting white precipitate was
filtered, washed with absolute ethanol, and dissolved in a solution
of 6  HCl. After 2 h at 80 °C, the solvent was evaporated to give
a white powder that was dried under vacuum at 80 °C for 24 h.
DHB, H2O): m/z
= 1081.25 [M +
H]+. C39H63I3N12·2H2O
Potentiometric Titrations: Potentiometric measurements were per-
formed in a jacketed cell thermostatted at 25.0 °C, kept under inert
atmosphere of purified argon, using an automatic titrator (Met-
rohm, DMS Titrino 716) connected to a microcomputer. The free
hydrogen concentrations were measured with a glass–Ag/AgCl
combined electrode (Metrohm) filled with 0.1  NaCl. The elec-
trode was calibrated to read –log[H+], designated as pH, by ti-
tration of a small quantity of diluted HCl by standardized NaOH
at 0.02  ionic strength and 25.0 °C (and determining the equiva-
lent point by the Gran’s method) followed by adjustment of the pH
meter so as to minimize the calculated pH versus observed values.
A logKw for the system, defined in terms of log([H+][OH–]), was
found to be –13.78 at the ionic strength employed and was main-
tained fixed during refinements.[15] Potentiometric measurements of
solutions containing equimolecular amounts of azaligands and the
appropriate anion were made at about 1 m concentration and
ionic strength I = 0.10  (NaCl). Each titration utilized at least 10
points per neutralization of a hydrogen ion equivalent and ti-
trations were repeated until reaching a satisfactory agreement. A
minimum of three sets of data was used in each case to calculate
the overall stability constants and their standard deviations. The
standard deviations obtained for the different stability constants
are reported in Tables 1 and 2. The range of accurate pH measure-
ments was considered to be 2–12. Equilibrium constants and spe-
cies distribution diagrams were calculated by using the program
HYPERQUAD 2003.[16] The stability constants Kalh were noted
with respect to ternary species AaLlHh where a, l, and h are the
stoichiometric number of A: anion, L: ligand, and H: proton,
respectively.
(1116.75): calcd. C 41.95, H 6.05, N 15.05; found C 42.01, H 6.12,
N 15.31.
Compound L2: Tris-salt 3 was heated at refluxed in hydrazine
monohydrate as described previously[9b] to give L2 as a white pow-
1
der (90% yield). H NMR (400.13 MHz, CDCl3, 298 K): δ = 2.59
(t, 4 H, αCH2, H2), 2.64 (t, 4 H, αCH2, H4), 2.69 (t, 4 H, αCH2,
H5), 2.80 (t, 4 H, αCH2, H3), 3.72 (s, 2 H, αCH2Ph, H1), 7.26 (s,
1 H, CHAr, Ha) ppm. 13C NMR (100.62 MHz, CDCl3; 298 K): δ
= 45.1, 46.6, 47.2, 51.0 (αCH2), 59.0 (αCH2Ph), 129.4 (CHAr),
138.9 (CAr) ppm. MS (MALDI-TOF, MeOH, dithranol): m/z =
632.61 [M + H]+. C33H67N12 (631.98): calcd. C 62.72, H 10.69, N
26.60; found C 63.01, H 10.89, N 10.37.
Compound 4: A solution of iodomethane (3.65 g; 25.75 mmol) in
dry tetrahydrofuran (20 mL) was added dropwise to a stirred solu-
tion of cyclen–glyoxal 2 (5 g; 25.75 mmol) in dry tetrahydrofuran
(40 mL). The mixture was stirred at room temperature for 2 d. The
precipitate was collected by filtration, washed with diethyl ether,
and dried in vacuo to give 4 as a white powder (90% yield). 13C
NMR (100.62 MHz, D2O, 298 K): δ = 49.9, 50.5 (2 C), 50.7, 51.0,
51.2, 54.1, 63.9, 68.3 (αCH2), 74.4, 86.3 (CHaminal) ppm. MS
(MALDI-TOF, DHB, H2O): m/z
=
335.12 [M
+
H]+.
C11H21IN4·H2O (354.23): calcd. C 37.30, H 6.54, N 15.82; found
C 37.55, H 6.76, N 16.01.
Compound 5: A solution of 1,3,5-tris(iodomethyl)benzene (2.48 g;
15.1 mmol) dissolved in dry acetonitrile (35 mL) was added drop-
wise to a stirred solution of 4 (5 g; 14.96 mmol) in dry acetonitrile
(40 mL). The mixture was stirred at room temperature for two
weeks. The precipitate was collected by filtration, washed with di-
ethyl ether, and dried in vacuo to give 5 as a white powder (88%
yield). 13C NMR (100.62 MHz, D2O, 298 K): δ = 45.6, 45.9, 49.2,
49.3, 49.7, 57.7, 62.0, 62.4, 64.2 (αCH2), 80.7, 81.4 (CHaminal),
132.8 (CHAr), 141.6 (CAr) ppm. MS (MALDI-TOF, H2O, DHB):
m/z = 1507.11 [M + H]+. C42H72I6N12·3H2O (1560.57): calcd. C
33.32, H 5.04, N 10.77; found C 33.01, H 5.12, N 11.01.
Compound L3H33+,3I–: A large excess of NaBH4 (24 equiv.) was
added in small portions over 1 h to a stirred solution of polyam-
monium salt 5 in absolute ethanol (3 g in 200 mL). The mixture
was allowed to stir at room temperature for one week. After cool-
ing to 0 °C, HI (2 ) was added to pH ≈ 3–4, and the resulting
mixture was evaporated to dryness. The resulting white solid was
then dissolved in a small quantity of water and potassium hydrox-
ide pellets were added until a basic medium was obtained. The
aqueous phase was then evaporated to dryness, and the residue
1
NMR Measurements: H and 31P NMR spectra in D2O solutions
at different pH values (adjusted with NaOD or DCl solutions) were
recorded at 298 K with Bruker spectrometers. In the 1H NMR
spectra, the reported peak positions are relative to HOD at δ =
4.79 ppm. 1H–1H and 1H–13C 2D correlation experiments were per-
formed to assign the signals. Small amounts of 0.01  NaOD or
DCl solutions were added to a solution of the chlorohydrated li-
gand to adjust the pD. The pH was calculated from the measured
pD values with the following relationship: pH = pD – 0.40.[17]
Supporting Information (see footnote on the first page of this arti-
cle): The pH dependence of the of L2; logarithm recognition con-
stants, logKalh, for the ligands with anions; species distribution dia-
grams ligands/anions as a function of pH; overall percentages of
complexed ALHh species with ligands, as a function of pH; 1H
NMR chemical shifts of aromatic Ha protons.
Eur. J. Org. Chem. 2010, 5380–5390
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5389

A.-S. Delépine, R. Tripier, M. Le Baccon, H. Handel
FULL PAPER
[6] a) A.-S. Delépine, R. Tripier, N. Le Bris, H. Bernard, A. Hon-
raedt, H. Handel, Inorg. Chim. Acta 2009, 362, 3829–3834; b)
N. Le Bris, H. Bernard, R. Tripier, H. Handel, Inorg. Chim.
Acta 2007, 3026–3032.
[7] a) S. Develay, R. Tripier, M. Le Baccon, V. Patinec, G. Serra-
trice, H. Handel, Dalton Trans. 2005, 3016; b) S. Develay, R.
Tripier, M. Le Baccon, V. Patinec, G. Serratrice, H. Handel,
Dalton Trans. 2006, 3418; c) S. Develay, R. Tripier, N. Bernier,
M. Le Baccon, V. Patinec, G. Serratrice, H. Handel, Dalton
Trans. 2007, 1038; d) A. S. Delépine, R. Tripier, H. Handel,
Org. Biomol. Chem. 2008, 6, 1743.
[8] A.-S. Delépine, R. Tripier, H. Bernard, N. Le Bris, H. Handel,
Tetrahedron Lett. 2009, 50, 2521–2524.
[9] a) J. Kotek, P. Hermann, P. Vojtisek, J. Rohovec, I. Lukes, Col-
lect. Czech. Chem. Commun. 2000, 65, 243–251; b) M. Le Bac-
con, F. Chuburu, L. Toupet, H. Handel, M. Soibinet, I.
Déchamps-Olivier, J.-P. Barbier, M. Aplincourt, New J. Chem.
2001, 25, 1168–1174.
Acknowledgments
The authors thank the “service commun” of the NMR facilities of
the University of Brest.
[1] a) A. T. Tu, M. J. Heller in Metal Ions in Biological Systems
(Ed.: H. Sigel), Marcel Dekker, New York, 1974, vol. 1, p. 1;
b) C. M. Frey, J. E. Stuehr in Metal Ions in Biological Systems
(Ed.: H. Sigel), Marcel Dekker, New York, 1974, vol. 1, p. 51;
R. Philips, Chem. Rev. 1966, 66, 501–527; c) R. M. Izatt, J. J.
Christensen, J. H. Rytting, Chem. Rev. 1971, 71, 439–481.
[2] a) B. G. Feuerstein, N. Pattabiraman, L. J. Morton, Nucleic Ac-
ids Res. 1990, 18, 1271–1279; b) A. Gasowska, L. Lomozik, R.
Jastrzab, J. Inorg. Biochem. 2000, 78, 139–147.
[3] a) J. A. Aguilar, P. Diaz, F. Escarti, E. Garcia-Espana, L. Gil,
C. Soriano, B. Verdejo, Inorg. Chim. Acta 2002, 307–316; b) C.
Bazzicalupi, A. Bencini, A. Bianchi, E. Faggi, C. Giorgi, S.
Santarelli, B. Valtancoli, J. Am. Chem. Soc. 2008, 130, 2440–
2442; c) A. Gasowska, J. Inorg. Biochem. 2003, 96, 346–356; d)
J. Alves da Silva, J. Felcman, A. L. R. Merc, A. S. Mangrich,
R. S. C. Lopes, C. C. Lopes, Inorg. Chim. Acta 2003, 356, 155–
166; e) C. Anda, C. Bazzicalupi, A. Bencini, E. Berni, A. Bian-
chi, P. Fornasari, A. Llobet, C. Giorgi, P. Paoletti, B. Valtan-
coli, Inorg. Chim. Acta 2003, 356, 167–178; f) C. Bazzicalupi,
A. Bencini, A. Bianchi, A. Danesi, E. Faggi, C. Giorgi, C. Lod-
eiro, E. Oliveira, F. Pina, B. Valtancoli, Inorg. Chim. Acta 2008,
3410–3419.
[10]
a) S. Develay, R. Tripier, F. Chuburu, M. Le Baccon, H. Han-
del, Eur. J. Org. Chem. 2003, 3047–3050; b) R. Tripier, S. De-
velay, M. Le Baccon, F. Chuburu, F. Michaud, H. Handel, New
J. Chem. 2004, 28, 173–176.
[11]
a) N. Bernier, M. Allali, R. Tripier, F. Conan, V. Patinec, S.
Develay, M. Le Baccon, H. Handel, New J. Chem. 2006, 1,
435–441; b) N. Bernier, R. Tripier, V. Patinec, M. Le Baccon,
H. Handel, C. R. Chim. 2007, 10, 832–838; c) A. Bencini, S.
Biagini, C. Giorgi, H. Handel, M. Le Baccon, P. Mariani, P.
Paoletti, P. Paoli, P. Rossi, R. Tripier, B. Valtancoli, Eur. J. Org.
Chem. 2009, 5610–5621.
[4] a) P. D. Beer, Acc. Chem. Res. 1998, 31, 71–80; b) P. D. Beer,
P. A. Gale, Angew. Chem. Int. Ed. 2001, 40, 487–516; c) A. [12] a) G. R. Weisman, M. E. Rogers, E. H. Wong, J. P. Jasinski,
Bianchi, K. Bowman-James, E. García-España, Supramolec-
ular Chemistry of Anions, Wiley, Chichester, U. K., 1997; d)
O. A. Gerasimchuk, S. Mason, A. M. Llinares, M. Song, N. W.
E. S. Paight, J. Am. Chem. Soc. 1990, 112, 8604–8605; b) G. R.
Weisman, E. H. Wong, D. C. Hill, M. E. Rogers, D. P. Reed,
J. C. Calabrese, Chem. Commun. 1996, 947–948.
Alcock, K. Bowman-James, Inorg. Chem. 2000, 39, 1371–1375; [13] a) R. W. Alder, Chem. Rev. 1989, 89, 1215–1223; R. D. Han-
e) J. Wang, A. E. Martell, R. J. Motekaitis, Inorg. Chim. Acta
2001, 322, 47–55; f) B. Dietrich, M. W. Hosseini, J.-M. Lehn,
Helv. Chim. Acta 1985, 68, 289–299; g) M. Micheloni, Com-
ments Inorg. Chem. 1988, 8, 79–100; h) M. W. Hosseini, J.-M.
Lehn, Helv. Chim. Acta 1987, 70, 1312–1319; i) P. D. Beer, P. A.
Gale, Angew. Chem. Int. Ed. 2001, 40, 487–516; j) M. W. Hosse-
ini, Perspectives in Coordination Chemistry 1992, 333–344.
[5] a) J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, Ger-
many, 1995; b) J. S. Bradshaw, Aza-crown Macrocycles, Wiley,
New York, 1993; c) J. L. Atwood, K. T. Holman, J. W. Steed,
Chem. Commun. 1996, 1401–1407; d) C. Bazzicalupi, A. Bencini,
A. Bianchi, B. Escuder, V. Fusi, E. Garcia-España, G. Giorgi,
V. Marcelino, P. Paoletti, B. Valtancoli, J. Am. Chem. Soc. 1999,
121, 6807–6815; e) Tomalia, D. A. European Patent 0115771,
1984; f) C. Bazzicalupi, A. Bencini, A. Bianchi, L. Borsari, C.
Giorgi, B. Valtancoli, J. Org. Chem. 2005, 70, 4257–4266; g) A.
Bencini, E. Berni, A. Bianchi, C. Giorgi, B. Valtancoli, D. K.
Chand, H. J. Schneider, Dalton Trans. 2003, 793–800.
cock, A. E. Martell, Chem. Rev. 1989, 89, 1875–1914; b) M.
Ciampolini, N. Nardi, B. Valtancoli, M. Micheloni, Coord.
Chem. Rev. 1992, 120, 223–236; c) A. Ingham, M. Rodopoulos,
K. Coulter, T. Rodopulos, S. Subramanian, A. McAuley, Co-
ord. Chem. Rev. 2002, 233, 255–271; d) J. Springborg, Dalton
Trans. 2003, 9, 1653–1665 and references cited therein; e) T. J.
Hubin, Coord. Chem. Rev. 2003, 241, 27–46 and references
cited therein.
[14] a) E. Kimura, S. Aoki, T. Koike, M. Shiro, J. Am. Chem. Soc.
1997, 119, 3068–3076; b) S. Aoki, M. Shiro, T. Koike, E. Ki-
mura, J. Am. Chem. Soc. 2000, 122, 576–584.
[15] a) G. Gran, Analyst (London) 1952, 77, 661; b) A. E. Martell,
R. J. Motekaitis, Determination and Use of Stability Constants,
2nd ed., Wiley, New York, 1992.
[16] P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753.
[17] P. K. Glasoe, F. A. Long, J. Phys. Chem. 1960, 64, 188–190.
Received: April 11, 2010
Published Online: August 17, 2010
5390
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5380–5390