Supramolecular catalysis of H/D exchange in pyruvate by macrocyclic polyamines involving a reactive iminium intermediate

Article abstract of DOI:10.1039/a700937b

The activity of a series of macrocyclic polyamines as catalysts for proton exchange has been investigated. Structural and physico-chemical studies demonstrate the ability of these receptors to recognize pyruvate, form a covalent iminium intermediate and catalyse H/D exchange at the CH₃ position. The reaction follows Michaelis-Menten kinetics with a rate enhancement higher than 4.8 × 10⁵ (k_cat/k_uncat). The process also represents a potential entry into the catalysis of aldol condensation reactions.

Full text of DOI:10.1039/a700937b

View Article Online / Journal Homepage / Table of Contents for this issue
Supramolecular catalysis of H/D exchange in pyruvate by
macrocyclic polyamines involving a reactive iminium intermediate
Hicham Fenniri,† Carol Dallaire, Daniel P. Funeriu and Jean-Marie Lehn*
Laboratoire de Chimie Supramoléculaire, Institut Le Bel, Université Louis Pasteur,
4, rue Blaise Pascal, 67000 Strasbourg, France
The activity of a series of macrocyclic polyamines as catalysts for proton exchange has been
investigated. Structural and physico-chemical studies demonstrate the ability of these receptors to
recognize pyruvate, form a covalent iminium intermediate and catalyse H /D exchange at the CH 3
position. The reaction follows M ichaelis–M enten kinetics with a rate enhancement higher than
4.8 × 105 (kcat/kuncat). The process also represents a potential entry into the catalysis of aldol
condensation reactions.
H
N
Introduction
Numerous biological transformations1 and organic reactions2
involve iminium and/or enamine compounds as reactive inter-
mediates. This is particularly the case for a series of reactions
co-catalysed by vitamin B6, most of which have been repro-
duced in model systems.1,3,4
Macrocyclic polyamines possess the ability to simultaneously
(1) carry a high positive charge through protonation at neutral
pH,5 so that they can serve as binding sites for anionic sub-
strates, (2) present unprotonated 5 nucleophilic sites at neutral
pH and (3) serve as a proton reservoir in a reaction on a bound
substrate. These features make macrocyclic polyamines attract-
ive systems for performing supramolecular catalysis.6 They are
involved in the catalysis of ATP hydrolysis, of phosphoryl
transfer reactions6 as well as of malonate enolization and H/D
exchange.7
N
X
N
H
NH
NH
HN
HN
NX
NX
XN
XN
NH
HN
O
O
X
N
N
H
1 X = H
2 X = Me (200)
(400)
4 (500)
3 (4000)
X
N
X
N
HN
NH
Y
N
Y
NH
NH
HN
N
N
Y
N
HN
Y
Protonated macrocyclic polyamines form stable complexes
with anionic substances.5–7 In view of its central role in bio-
logical metabolism,1,3 pyruvate was an attractive substrate
which may be bound by protonated macrocyclic polyamines
because of its negative charge at neutral pH 8 and could be
activated through the formation of an iminium/enamine
intermediate with one of the unprotonated nitrogens. The
dual nature of the acidic microenvironment provided by the
ammonium sites and the nucleophilic/basic sites provided by
the unprotonated nitrogens, should facilitate such a process
through general acid–general base catalysis. Bifunctional
catalysis of α-hydrogen exchange in carbonyl compounds by
amines has been extensively studied.11,12 Ultimately, the
iminium/enamine intermediates could function as powerful
5 X = H
6 X = H
7 X = CH3 Y = H
Y = H
Y = CH3
(8500)
(2000)
(7000)
(1200)
9 (950)
8 X = CH3 Y = CH3
CH3
H3C
N
N
NH
NH
HN
N
HN
N
electrophiles or nucleophiles for
a
second co-bound
nucleophile/electrophile in a polytopic receptor.
10 (15 000)
Fig. 1 Polyamines investigated and observed rate enhancements (in
parentheses) of H/D exchange of pyruvate measured in their presence;
[receptor] = 10 m, [pyruvate] = 50 m, pD 7, 25 ЊC, 60% (CD3)2SO–
D2O
Results and discussion
Reaction of pyruvate with macrocyclic polyamines, and
characterization of the intermediate
Two types of macrocyclic polyamine receptors were selected
(Fig. 1). The first is rather flexible and hydrophilic (1–3) and the
second is more rigid and has hydrophobic character due to
aromatic subunits (5–10). The acyclic compound 4 was used for
comparison purposes. The synthesis of 313, 514 and 915 has
already been reported, 1 and 2 are commercially available and
4, 6–8 and 10 were prepared according to the synthetic scheme
in Scheme 1. The key step in the preparation of the macrocyclic
receptors is the [2 ϩ 2] Schiff base condensation 16 yielding the
corresponding tetraimines, which were reduced to yield the
corresponding macrocyclic polyamines.
Incubation of pyruvate with any of these polyamines in
aqueous dimethyl sulfoxide [60% (CD3)2SO–D2O, pD 7, 25 ЊC]
resulted in H/D exchange of the CH3 group of pyruvate with
† Present address: The Scripps Research Institute, Beckman Center for
Chemical Sciences, 10550 North Torrey Pines Road, La Jolla, CA
92037, USA.
1
the solvent. The reaction could be followed by H NMR spec-
troscopy and the rate of reaction was dependent on the receptor
J. Chem. Soc., Perkin Trans. 2, 1997
2073

View Article Online
Table 1 Observed rate constants (kobs) and relative rate enhancements for the H/D exchange of the α-CH3 of pyruvate catalysed by various
polyamine receptors; [receptor] = 10 m, [pyruvate] = 50 m, 25 ЊC, pD 7, 60% (CD3)2SO–D2O
Assay
Polyamine
kobsa/sϪ1
t1/2/min
Relative ratesb
Chargec
d
1
2
3
4
5
6
7
8
9
kuncat
0.02
0.03
8
4.0
80
10
170
40
5.8 × 105
1
1.5
400
200
4 000
500
8 500
2 000
7 000
1 200
950
—
1
3
3
4
Diethylamine
3.9 × 105
1 444
1
2
3
4
5
6
7
8
9
2 888
144
1 155
68
289
83
481
608
39
2
3–4
3–4
3–4
3–4
3–4
3–4
140
24
19
10
11
12
10
300
15 000
a
These data are within ± 10%. b Rate enhancements relative to the reaction in absence of polyamines (assay 1). c Overall positive charge of the
polyamine upon protonation under the present experimental conditions as estimated from data obtained in H2O for some of the receptors studied
d
here, and from others.21 kuncat is the slope of the relationship V = kuncat [pyruvate] for [pyruvate] = 10, 20, 30, 40, 60, 80, 100 and 150 m.
X
N
X
N
Y
Y
H2N
NH2
11 X = H Y = CN
18 X = CH3
19 X = H
12 X = Ts
13 X = Ts
Ts
N
Y = CN
Y = CH2NH2
TsHN
NHTs
24
H
H
H
H
Br
Z
O
O
O
O
14
14 Z = CH
20 Z = N
25
X
N
N
N
N
N
TsN
NTs
XN
Z
NX
XN
2
2
2
26 X = Ts
4 X = H
15
21 X = H
22 X = CH3
23 X = CH3
Z = CH
Z = CH
Z = N
Y
N
Y
N
Y
N
Y
N
XN
NX
XN
Z
NX
2
2
5 X = H
Y = H
Z = CH
16 X = Ts Y = CN
17 X = Ts
6 X = H
8 X = CH3 Y = CH3 Z = CH
Y = CH3
Y = CH3
7 X = CH3 Y = H
10 X = CH3 Y = H
Z = CH
Z = N
Scheme 1 Synthetic schemes for the preparation of polyamine receptors 4, 6–8 and 10
used. Fig. 2 shows the 1H NMR time course of the CH3 signal
at 2.1 ppm. The reaction is first order and the observed rate
enhancements are summarized in Table 1 and Fig. 1. Under
these conditions, the chemical shifts of the 1H NMR signals of
all the receptors were not affected significantly during the
course of the reaction.
At a higher percentage of (CD3)2SO–D2O (80%) or at a
higher concentration of polyamine–pyruvate (20 m–100 m),
the 1H NMR signals corresponding to the macrocyclic
polyamine 3 developed shoulders and a new signal could be
observed around 2.94 ppm (insert in Fig. 2). These alterations
remained constant during and after complete H/D exchange of
pyruvate (5–10-fold excess) but they disappeared when the
medium was diluted with D2O, indicating that a reversible reac-
tion was taking place between the receptor and the pyruvate.
These modifications were not observed with the other
polyamines, suggesting that the corresponding intermediates
were less stable.17
was the only polyamine that showed some alterations in the 1H
NMR spectra (Fig. 2), we also studied polyamine 5 in order to
show that although it was undetectable by 1H NMR spec-
troscopy, the same intermediate can be traced by ESMS. Fig. 3
presents relevant parts of the ESMSϩ spectra for polyamine 3
(10Ϫ4 ) in the presence of pyruvate (5 × 10Ϫ4 ), in DMSO–
H2O 80% (left) or DMSO–D2O 80% (right), after incubation at
40 ЊC for 4 h at an apparent pH or pD of 7. In DMSO–H2O the
signal at m = 347.5 was assigned to the monoprotonated
macrocyclic polyamine 3 ([m/z] ϩ 1) and the signal at m = 417.5
to the monoprotonated iminium intermediate ([mi/z] ϩ 1). As
a result of the H/D exchange at the nitrogen sites and at the
pyruvate CH3, when the same experiment was performed in
DMSO–D2O an increase in the molecular mass of the free
polyamine and the intermediate by 7 and 9 units respectively
was recorded. Similar results were obtained with 5 (data not
shown). As shown in Fig. 4, the intermediate could also be the
enamine or the aminal species, since they all have the same mass
and could lead to the same mass shift in DMSO–D2O. One way
to discriminate between these species was to perform negative
The formation of a covalent intermediate was demonstrated
using electrospray mass spectrometry (ESMS) (Fig. 3). Since 3
2074
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Fig. 2 Typical 200 MHz 1H NMR spectra in the course of the H/D exchange of pyruvate in the presence of the macrocyclic polyamine 5; [5] = 10
1
m, [pyruvate] = 50 m, pD 7, 25 ЊC, 60% (CD3)2SO–D2O. Under these conditions, none of the H NMR signals of the polyamine was affected,
whereas at twice this concentration in 80% (CD3)2SO–D2O (insert) the 1H NMR spectrum of polyamine 3 showed new peaks (i) that may be
attributed to an intermediate iminium species; the letters A, B, C and D denote the methylene protons of the macrocycle 3.
Fig. 3 Electrospray mass spectra for the characterization of the iminium intermediate; [1] = 10Ϫ4 , [pyruvate] = 5 × 10Ϫ4 , 25 ЊC, pH 7, 80%
(CD3)2SO–D2O (left) or (CD3)2SO–D2O (right). The reaction media were incubated at 40 ЊC for 4 h before recording the spectra. The peaks indicated
by the letter ‘c’ are also present in the control experiments (in the absence of polyamine or pyruvate, or in the solvent alone). Similar results were
obtained with the macrocycle 5.
ESMS on the same samples. The iminium is undetectable
because of its overall charge neutrality, whereas the enamine
and the aminal should give rise to a signal corresponding to the
monoanionic species. No signal other than that corresponding
to the free pyruvate was observed (data not shown), in agree-
ment with the iminium being the major intermediate. The
enamine and aminal probably exist only as transient species.
Similar results were obtained with receptor 5 (data not shown).
In order to further support this mechanism, we performed
some experiments on the decarboxylation of acetoacetate in the
presence of 5. The reaction leading to acetone and carbon
dioxide was accelerated by an order of magnitude.18 Although
the rate enhancement is fairly modest, this result supports the
presence of the iminium/enamine intermediates.19 Further
studies to improve the catalytic performances and to elucidate
the mechanistic pathway of macrocyclic polyamines catalysed
decarboxylation processes are being pursued.
Structure–function relationships
Table 1 summarizes the observed rate constants kobs for H/D
exchange obtained with the various polyamines. The analysis of
these results gives an insight into the structural and physico-
chemical requirements for efficient catalysis.
(1) The observed rate enhancements over the uncatalysed
reaction (assay 1) vary from 1.5-fold for diethylamine (assay 2)
to a factor of about 15 000 for 10 (assay 12).20 They may result
from both pyruvate binding and facilitation of the H/D
exchange reaction by the polyamines.
(2) Comparison of the rate enhancements obtained with 1
(assay 3) and 2 (assay 4) to those obtained with the other recep-
tors shows that neither the presence of unprotonated and pro-
tonated nitrogens21 on the same receptor nor the presence or
absence of secondary nitrogens is a sufficient requirement for
catalysis.20
(3) The acyclic monotopic polyamine 4 (assay 6) which
J. Chem. Soc., Perkin Trans. 2, 1997
2075

View Article Online
Iminium
Enamine
Aminal
NH
CO2
HN
HN
NH
CO2
HN
HN
NH
CO2
HN
HN
–
–
–
H
H
H
H
H
H
+
+
+
N
+
N
N
+HN
N
+HN
CH3
NH
CH2
NH
N
Detectable by ESMS+
Detectable by ESMS+
Detectable by ESMS+
NH
CO2
HN
HN
NH
CO2
HN
HN
NH
CO2
HN
HN
–
–
–
+
N
NH
N
NH
N
NH
CH3
NH
CH2
NH
N
Undetectable by ESMS–
Detectable by ESMS–
Detectable by ESMS–
Fig. 4 Possible structures for the reactive intermediates formed between the macrocyclic polyamine and pyruvate. The top three structures are
detectable by positive electrospray mass spectrometry (positive ESMS), whereas the bottom central (enamine) and right (aminal) species are
detectable by negative ESMS.
the number of secondary nitrogen sites. In addition, besides the
geometrical and structural factors, the effect of the different
nitrogen sites on catalysis should also be related to their pKa
values and the pH of the medium.22,23
(5) The macrocyclic polyamine 10, the pyridine analogue of
7, is the most efficient catalyst in this study and is twice as
effective as 7. This might be at least in part related to the pyri-
dine units which could act as general base and participate in the
proton abstraction from the iminium intermediate.
pH profile, Michaelis–Menten kinetics and mechanistic scheme
for the catalysis of H/D exchange at CH3 in pyruvate by the
macrocyclic polyamine 10
Below pH 6 pyruvate is in equilibrium with its hydrated form.24
The ratio of hydrate–keto forms does not vary during the
course of the reaction whatever the pH. In the absence of 10,
the H/D exchange is acid catalysed with a maximum rate
around pH ≈ pKa (pyruvate) ≈ 4 (Fig. 5). This is in agreement
with acid–base catalysis taking place under the conditions
used.10
Fig. 5 pH Dependence of the observed rate constant for H/D
exchange of the CH3 group of pyruvate in the presence (᭝) and absence
(᭺) of macrocyclic polyamine 10. [10] = 10 m, [pyruvate] = 50 m,
25 ЊC, 60% (CD3)2SO–D2O. The curves represented correspond to the
equations: (᭝) in presence of polyamine, y = Ϫ7.411 ϩ 1.638x Ϫ
0.733x2 ϩ 0.173x3 Ϫ 0.017x4 ϩ 0.001x5; (᭺) in absence of polyamine,
y = Ϫ8.748 = 1.680x Ϫ 0.310x2 ϩ 0.015x3.
In the presence of 10 the observed rate is higher over the
whole pH range investigated with a maximum around pH 8
where, interestingly, the uncatalysed reaction is at its minimum
(Fig. 5). Below pH 4 pyruvate is protonated and hence it is
probably no longer bound to the polyamine. Above pH 9, the
macrocyclic polyamine becomes unprotonated and hence it
interacts less efficiently or not at all with pyruvate. At the opti-
mal pH 8 one expects pyruvate to be monoanionic and the
polyamine to carry both protonated and neutral nitrogen sites.
This coexistence should be responsible for the higher rate
enhancement, in agreement with the formation of a supra-
molecular complex between 10 and pyruvate, prior to the
chemical transformation.
mimics one site of the ditopic polyamine 5 (assay 7) is substan-
tially less effective (500- vs. 8500-fold acceleration). This result
supports a synergism between the two diethylenetriamine sub-
units for the H/D exchange of pyruvate, affecting either the
binding or the catalytic step or both. It agrees with the much
lower rate enhancement obtained with 9 (950-fold, assay 11),
where the diethylene triamine subunits are more distant, than
with its isomer 5 (8500-fold, assay 7).
(4) Methylation of the central nitrogens of 5 (assay 7)
resulted in receptor 7 (assay 9) which has similar catalytic activ-
ity (8500 vs. 7000). When the four lateral nitrogens or all the
nitrogens of 5 were methylated to give 6 (assay 8) and 8 (assay
10) respectively, the catalytic efficiency decreased appreciably
(from 8500 for 5 to 2000 for 6 and 1200 for 8). These results
indicate that, (i) nucleophilic catalysis through iminium form-
ation contributes only a part of the catalytic power of these
receptors, the other part being provided through general acid–
general base catalysis; (ii) since tertiary amines do not form an
iminium group, the relative efficiency of 5–8 may be related to
Fig. 6 shows Michaelis–Menten and Lineweaver–Burk plots
(insert) from which Kmapp, Vmax and kcat were calculated. The high
Kmapp obtained (0.5 , pD 7) indicates that the complex formed
is very weak due to the single negative charge of pyruvate. In
spite of this, the macrocyclic polyamine 10 accelerates the reac-
tion by a factor of more than 4.8 × 105 (kcat/kuncat, pD 7). This
rate enhancement constitutes a lower limit since isotope effects
from the first exchange on the second, and the effects of the
first and second on the third exchange were not taken into
account in the calculation of the rate constants. Moreover,
[2H3]pyruvate, the product of the reaction, acts as a competitive
2076
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
inhibitor, so that the actual rate enhancement is probably even
higher.
generate the most reactive iminiums.23 For these reasons, and
due also to general base catalysed proton abstraction by
another unprotonated nitrogen (aliphatic or pyridinic), this
intermediate is in rapid equilibrium with the enamine form.
Fast capture of a deuterium from the solvent or the ammonium
sites by the enamine would regenerate the iminium which may
go through one or two catalytic cycles to yield respectively the
di- or tri-deuterated iminium, or may hydrolyse to generate the
partially deuterated pyruvate. Finally, [2H3]pyruvate is pro-
duced after three catalytic cycles.
A possible mechanism for the catalytic cycle of H/D
exchange of pyruvate is illustrated for the case of 10 in Fig. 7.
Based on previously determined pKa values for several macro-
cyclic polyamines,21 10 should exist essentially as the tri- and
tetra-protonated species at neutral pH. After binding of one
molecule of pyruvate, a nucleophilic attack from one of the
lateral unprotonated nitrogens could take place leading to the
formation of the covalent iminium intermediate, which should
be more reactive towards nucleophiles and bases than the
parent ketone.22 Moreover, secondary amines are known to
Conclusions
Macrocyclic polyamines have been found to strongly activate
pyruvate towards H/D exchange through the formation of an
iminium intermediate which can in principle also participate
in decarboxylation reactions and electrophilic substitutions.
Conversely, the corresponding enamine form might act as an
electron rich intermediate for aldol condensation and other
nucleophilic reactions. Thus, a number of other important
chemical and biochemical reactions might be catalysed by
macrocyclic polyamines. Further exploration of the parent
macrocycles and of suitably designed derivatives is warranted.
Experimental
General
NMR spectra were recorded on a Bruker AC 200 (200.1 MHz
1
for H and 50.3 MHz for 13C spectroscopy), with the solvent as
internal reference. For the 1H and 13C NMR spectra in D2O, 2-
methylpropan-2-ol was used as internal reference (CH3, 1.36
ppm; HOC(CH3)3, 68.7 and 31.6 ppm). J Values are given in Hz.
Melting points were measured on a digital Thomas–Hoover
apparatus (Electrotherma). Chromatographic supports were sil-
Fig. 6 Michaelis–Menten and Lineweaver–Burk (insert) plots for H/D
exchange of the α-CH3 of pyruvate catalysed by polyamine 10; [10] = 10
m, pD 7, 25 ЊC, 60% (CD3)2SO–D2O; Km = 0.55 , kcat = 9.54 × 10Ϫ3
s
Ϫ1, Vmax = 9.54 × 10Ϫ5  sϪ1
D+
H
O
–
O
–
+
NH
N
N
N
O
O
+
+
Tautomerism
H
H
N
N
N
N
+
H
H
N
N
H
H
N
N
N
N
H
H
+
+
H
+
+
H
H
H
N
N
D+
Schiff base
formation
H+
O
D
O
–
O
–
N
N
N
N
H
O
N+
H
+
O
H
+
N
N
H
N
H
N
N
H
H
H
N
H
N
N
H
N
+
+
+
H
H
+
H
N
N
–
Pyruvate
Pyruvate
D3 via D2 and D1
N
N
H
H
N+
N
H
N
N
H
N
N
H
+
+
H
H
Fig. 7 Possible catalytic cycle for the H/D exchange of the CH3 group of pyruvate catalysed by macrocyclic polyamine 10 via formation of iminium
intermediates
J. Chem. Soc., Perkin Trans. 2, 1997
2077

View Article Online
ica Merck 60 (0.063–0.200 mm), silica flash Merck 60 (0.040–
0.063 mm), or alumina Merck Act. II–III (0.063–0.200 mm).
Elemental analyses were performed by the Service de Microana-
lyse de l’Université Louis Pasteur de Strasbourg or by the Ser-
vice Central d’Analyse du Centre National de la Recherche Sci-
entifique de Vernaison. The fast atom bombardment (FAB)
mass spectra were recorded by the Service de Spectroscopie de
Masse de l’Université Louis Pasteur de Strasbourg. ESMS
measurements were performed at the Laboratoire de Spectro-
scopie de Masse Bioorganique, Université Louis Pasteur,
Strasbourg.
(392.8) C, 55.04; H, 7.19; N, 10.70. Found: C, 54.96; H, 7.23;
N, 10.38%].
6,20-Dimethyl-3,6,9,17,20,23-hexaazatricyclo[23.3.1.111,15]-
triaconta-1(29),2,9,11(30),12,14,16,23,25,27-decaene, 22. Iso-
phthalaldehyde (14, 5.02 g, 37.2 m) and 4-methyl-1,4,7-
triazaheptane (18, 4.36 g, 37.2 m) were refluxed in methanol
(400 ml, distilled over Mg) overnight under an argon atmos-
phere. The solution was evaporated to dryness and the residual
oil was taken up in acetonitrile yielding a colloidal solution that
was left to decant for several hours. The supernatant was slowly
removed and the colloidal portion was subjected to the same
procedure several times until a yellowish solid residue was iso-
lated (C26H34N6, 430.6, 0.8 g, 10%). This compound was dried
under high vacuum and used in the next step without further
purification. No attempts were made to improve the yield.25
δH (200 MHz, CDCl3, 25 ЊC, TMS) 8.3 (s, 4H, NCH), 7.98 (s,
2H, aromatic), 7.75 (m, 4H, aromatic), 7.4 (t, 2H, aromatic),
3.75 (t, 8H, CH2NCH), 2.8 (t, 8H, CH3NCH2), 2.4 (s, 6H,
CH3N).
Synthesis of the macrocyclic polyamines
All the reagents and solvents as well as macrocyclic polyamines
1 and 2 are commercially available. 1 and 2 were converted from
the trisulfate salt to the hydrochloride salt form upon elution
on Dowex 1X4 (200–400 mesh) in the hydroxide (OHϪ) form,
followed by acidification of the free polyamine thus obtained
with 6  HCl and crystallization from methanol to yield the
1
corresponding hydrochloride salt. H and 13C NMR spectro-
6,20-Dimethyl-3,6,9,17,20,23-hexaazatricyclo[23.3.1.111,15]-
triaconta-1(29),11(30),12,14,25,27-hexaene, 7. Compound 22
(0.8 g, 1.84 m) and NaBH4 (0.53 g, 14 m) were suspended in
ethanol (20 ml), stirred at room temp. for 3 h and refluxed for 30
min. The solvent was evaporated to dryness and the residual
solid was taken up in H2O (10 ml) and extracted with CH2Cl2
(3 × 20 ml).The organic layers were combined and dried over
MgSO4, filtered and evaporated to dryness yielding the free
polyamine 7 as a white solid. It was then converted to its hydro-
bromide salt by slow addition of concentrated aqueous HBr
(48%, 5 ml). The solvent was evaporated to dryness yielding a
white solid which was taken up in H2O (5 ml) and precipitated
by addition of methanol followed by propan-2-ol. The white
solid thus obtained was filtered and recrystallized from H2O–
ethanol yielding the hydrobromide salt of 7 after drying under
high vacuum (C26H42N6 ؒ 6HBr, 924.11) as a white microcrystal-
line powder (1.26 g, 74%). δH (200 MHz, D2O, 25 ЊC, TMS) 7.78
(s, 2H, aromatic), 7.75 (s, 6H, aromatic), 4.54 (s, 8H,
NHCH2Ph), 3.55–3.37 (m, 16H, CH3NCH2CH2NH), 2.95 (s,
6H, CH3N); δC(50 MHz, D2O, 25 ЊC, TMS) 133.4, 133.2, 132.4,
131.7 (aromatic), 54.1, 53.3, 43.8 (CH2N), 42.6 (CH3N); m/z
(FAB) 601.2 [(Mϩ Ϫ 4HBr) ϩ 1] [Calc. for C26H42N6ؒ6HBr ϩ
1/2 EtOH (947.15): C, 34.24; H, 5.43; N, 8.87. Found: C, 33.97;
H, 5.74; N, 8.67%].
3,6,9,17,20,23-Hexamethyl-3,6,9,17,20,23-hexaazatricyclo-
[23.3.1.111,15]triaconta-1(29),11(30),12,14,25,27-hexaene, 8.
Macrocyclic polyamine 5 in the hexahydrobromide form (1.08
g, 1.21 m)14 was converted to the corresponding free poly-
amine by elution on Dowex 1X4 (200–400 mesh) in the OHϪ
form. After solvent evaporation, the free polyamine was
refluxed in formic acid (99%, 1.7 ml, 43.6 m) and formalde-
hyde (40%, 1.3 ml, 18.2 m) for 9 h. The reaction medium was
cooled to room temp. and concentrated HCl (12 , 0.5 ml) was
added before refluxing for an additional 3.5 h. The reaction was
cooled in an ice bath, treated with an aqueous KOH solution
(5 , 6 ml) and extracted with CH2Cl2 (3 × 20 ml). The organic
layers were combined, extracted with basic brine, dried over
MgSO4, filtered and evaporated to dryness. The free macro-
cyclic polyamine 8 (C30H50N6, 494.77) obtained as a colourless
oil was converted to the corresponding hexahydrobromide by
dissolution in concentrated aqueous HBr (48%, 5 ml), evapor-
ation to dryness, recrystallization from H2O–ethanol and dry-
ing under high vacuum (C30H50N6 ؒ 6HBr, 980.22, 0.84 g, 71%).
Free polyamine δH (200 MHz, CDCl3, 25 ЊC, TMS) 7.3–7.0
(m, 8H, aromatic), 3.4 (s, 8H, NCH2Ph), 2.5–2.35 (m, 16H,
CH3NCH2CH2NCH3), 2.19 (s, 6H, CH3N), 2.18 (s, 12H,
CH3N); hexahydrobromide form δH (200 MHz, D2O, 25 ЊC,
TMS) 8.1 (s, 2H, aromatic), 7.8 (s, 6H, aromatic), 4.82 (s, 8H,
NCH2Ph), 3.92 (s, 16H, CH3NCH2CH2NCH3), 3.10 (s, 12H,
CH3N), 3.09 (s, 6H, CH3N); free polyamine δC(50 MHz, CDCl3,
25 ЊC, TMS) 138.7, 129.7, 127.8 (aromatic), 62.6, 55.1, 54.3
scopy and microanalysis were in agreement with their struc-
tures. 3,13 514 and 915 were prepared according to literature
procedures.
1,7-Dibenzyl-1,4,7-tris(p-tolylsulfonyl)-1,4,7-triazaheptane,
26. 1,4,7-Tris(p-tolylsulfonyl)-1,4,7-triazaheptane 2413 (10 g,
17.7 m), benzyl bromide 25 (12 g, 70.8 m), and K2CO3 (12.2
g, 88.5 m) were stirred in DMF (50 ml) at 100 ЊC for 3 days.
After cooling to room temperature, the solid was filtered and
washed with CH2Cl2 (100 ml). The organic layers were evapor-
ated to a final volume of ~40 ml and the product was precipi-
tated with diethyl ether (200 ml), filtered and washed with
diethyl ether (200 ml), then dissolved in CH2Cl2 (200 ml) and
washed with H2O (2 × 150 ml). The organic layer was dried
over MgSO4, filtered and evaporated to dryness to yield com-
pound 26 (C39H43N3O6S3, 745.97) as a white microcrystalline
powder after recrystallization from CH2Cl2–diethyl ether and
drying under high vacuum (11.9 g, 90%), mp 176 ЊC. δH (200
MHz, CDCl3, 25 ЊC, TMS) 7.75 [d, 3J(H,H) 7.9, 4H, aromatic],
7.35–7.26 (m, 14H, aromatic), 7.12 (s, 4H, aromatic), 4.25 (s,
4H, PhCH2NTs), 3.17–3.12 (m, 4H, PhCH2NTsCH2CH2), 2.45
(s, 6H, CH3Ts), 2.79–2.71 (m, 4H, PhCH2NTsCH2), 2.39 (s, 3H,
CH3Ts); δC(50 MHz, CDCl3, 25 ЊC, TMS) 143.52, 143.41,
136.28, 136.06, 134.66, 129.87, 129.52, 128.81, 128.74, 127.97,
127.27, 127.02 (aromatic), 53.54, 49.12, 47.92 (CH2NTs), 21.49
(CH3Ts); m/z (FAB) 746.2 (Mϩ ϩ 1), 590.2 [(Mϩ Ϫ Ts) ϩ 1]
[Calc. for C39H43N3O6S3 (745.97): C, 62.8; H, 5.81. Found: C,
62.7; H, 5.81%].
1,7-Dibenzyl-1,4,7-triazaheptane, 4. Compound 26 (5 g, 6.7
m) was dissolved in THF (550 ml), the solution was cooled to
0 ЊC and LiAlH4 (6.9 g, 181 m) was added in small portions.
After the addition was completed, the reaction medium was
heated at 50 ЊC for 24 h. After cooling to 0 ЊC, Celite (10 g),
Na2SO4 (10 g) and ice were added. The latter was added slowly
and carefully until complete destruction of excess LiAlH4. The
mixture was stirred for 1 h at room temp., filtered over Celite
and the solid washed with CH2Cl2 (200 ml). The filtrates were
evaporated to dryness and the residual solid was taken up in
CH2Cl2 (200 ml) and extracted with a 1  aqueous solution of
NaOH (100 ml). The organic layer was dried over Na2SO4, fil-
tered and evaporated to dryness. The residual oil was taken up
in a minimum volume of ethanol (20 ml) and the desired com-
pound was precipitated in the hydrochloride form upon add-
ition of concentrated HCl. The precipitate was filtered, washed
with ethanol (50 ml) and dried under high vacuum yielding 4
(C18H25N3 ؒ 3HCl, 392.8) as a white hygroscopic solid (2.2 g,
85%), mp (dec.) 309 ЊC. δH (200 MHz, D2O, 25 ЊC, TMS) 7.63
(s, 10H, aromatic), 4.46 (s, 4H, PhCH2NH), 3.66 (s, 8H,
NHCH2CH2NH); δC(50 MHz, D2O, 25 ЊC, TMS) 131.94,
131.83, 131.35 (aromatic), 53.66, 45.49, 44.34 (CH2NH); m/z
(FAB) 284.2 [(Mϩ Ϫ 3HCl) ϩ 1] [Calc. for C18H25N3ؒ3HCl
2078
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
(CH2N), 43.1, 42.8 (CH3N); hexahydrobromide form δC(50
MHz, D2O, 25 ЊC, TMS) 135.7, 135.6, 132.5, 131.2 (aromatic),
62.2, 52.9, 51.1 (CH2N), 43.2, 42.3 (CH3N); m/z (FAB) free
polyamine: 495.3 (Mϩ ϩ 1) [Calc. for C30H50N6 ؒ 6HBr ϩ
3H2O ϩ 1EtOH (1080.34): C, 35.58; H, 6.35; N, 7.78. Found:
C, 35.73; H, 6.23; N, 7.89%].
the solvent to dryness the residual solid was taken up in an
aqueous solution of HCl (6 , 200 ml) and refluxed for 2 h. The
solvent was evaporated to dryness and the residual solid was
taken up in H2O (100 ml), basified to pH ~13 upon dropwise
addition of a 5  aqueous solution of KOH and extracted with
CH2Cl2 (2 × 200 ml). The organic layers were combined, evap-
orated to dryness and the resulting oil was taken up in ethanol
(100 ml), acidified to pH ~1 with aqueous concentrated HCl
(12 , 20 ml) and evaporated to dryness. The white solid thus
obtained was recrystallized from hot ethanol and dried under
high vacuum yielding the desired compound 13 (C11H19N3-
O2S ؒ 2HCl, 330.27) as white hygroscopic crystals (11.8 g, 46%),
mp 279 ЊC. δH (200 MHz, D2O, 25 ЊC, TMS) 7.90 [d, 3J(H,H) 8,
2H, aromatic], 7.6 [d, 3J(H,H) 8, 2H, aromatic], 3.63 [t, 3J(H,H)
6,20-Dimethyl-3,6,9,17,20,23,29,30-octaazatricyclo-
[23.3.1.111,15]triaconta-1(29),2,9,11(30),12,14,16,23,25,27-deca-
ene, 23. Pyridine-2,6-dicarboxaldehyde 20 (1.25 g, 9.25 m) in
dry acetonitrile (150 ml, distilled over CaH2) was added drop-
wise over 1 h under a nitrogen atmosphere to a solution of 4-
methyl-1,4,7-triazaheptane (18, 1.08 g, 9.25 m) in dry
acetonitrile (250 ml, distilled over CaH2) and stirred for an add-
itional 24 h. After evaporation of the solvent to dryness, the
residual oil was taken up in diethyl ether (400 ml) and the in-
soluble material was eliminated. The solvent was evaporated to
dryness and the same procedure repeated with 300 ml of diethyl
ether. The desired compound 23 (C24H32N8, 432.58, 1.21 g,
60%), obtained as a yellow solid, was dried under high vacuum
and used in the next step without further purification. δH (200
MHz, CDCl3, 25 ЊC, TMS) 8.05 (s, 4H, NCHPh), 7.85 [d,
3J(H,H) 8, 4H, aromatic], 7.56 [t, 3J(H,H) 9, 2H, aromatic], 3.73
3
5.9, 4H, CH2NH2], 3.4 [t, J(H,H) 5.8, 4H, CH2NTs], 2.55 (s,
3H, CH3Ts); δC(50 MHz, D2O, 25 ЊC, TMS) 148.08, 134.33,
132.43, 128.33 (aromatic), 50.2, 40.84 (NH2CH2CH2NTs),
22.79 (CH3Ts); m/z (FAB) 258.1 [(Mϩ Ϫ 2HCl) ϩ 1] [Calc.
for C11H19N3O2S ؒ2HCl (330.27): C, 40.00; H, 6.41; N, 12.72.
Found: C, 39.78; H, 6.61; N, 12.41%].
6,20-Bis(p-toluenesulfonyl)-3,6,9,17,20,23-hexaazatricyclo-
[23.3.1.111,15]triaconta-1(29),2,9,11(30),12,14,16,23,25,27-deca-
ene, 15. Compound 13 was converted quantitatively to the cor-
responding free polyamine (yellowish oil) upon elution on a
Dowex 1X4 (200–400 mesh) column in the hydroxide form and
evaporation. After drying under high vacuum, it was taken up
in dry acetonitrile (1040 ml, distilled over CaH2). Isophthalalde-
hyde 14 (5.2 g, 39 m) in dry acetonitrile (650 ml, distilled over
CaH2) was added dropwise under a nitrogen atmosphere to this
solution over 6 h at room temp. Stirring was maintained for an
additional 24 h. The precipitate formed was filtered, washed
with acetonitrile and dried under high vacuum. 15 (C38H42-
N6O4S2, 710.92) was obtained as a microcrystalline powder
from CH2Cl2 (24.12 g, 87%), mp 216 ЊC. δH (200 MHz, CDCl3,
25 ЊC, TMS) 8.17 (s, 4H, NCHPh), 7.93 (s, 2H, aromatic Ph),
7.76 [d, 3J(H,H) 8.3, 4H, aromatic Ts], 7.67 [d, 3J(H,H) 7.6, 4H,
3
3
[t, J(H,H) 5.5, 8H, CH2NCHPh], 2.81 [t, J(H,H) 5.5, 8H,
CH3NCH2], 2.3 (s, 6H, CH3N).
6,20-Dimethyl-3,6,9,17,20,23,29,30-octaazatricyclo-
[23.3.1.111,15]triaconta-1(29),11(30),12,14,25,27-hexaene, 10.
The tetraimine 23 (1.21 g, 2.8 m) and NaBH4 (2.1 g, 14 m)
were stirred in EtOH (50 ml) at room temp. for 20 h. The sol-
vent was evaporated and the residue was taken up in CH2Cl2–
H2O (50 ml–50 ml). The organic layer was evaporated to dry-
ness and the residual solid was taken up in ethanol (10 ml) and
the solution was acidified with concentrated aqueous HBr
(48%, 1 ml). The precipitate thus generated was filtered, dried
under high vacuum and recrystallized from H2O–methanol
yielding 10 (C24H40N8 ؒ 6HBr, 926.10) as a white microcrystal-
line powder (1.5 g, 50%), mp (dec.) 254 ЊC. δH (200 MHz, D2O,
25 ЊC, TMS) 8.08 [s, 3J(H,H) 9, 2H, aromatic], 7.61 [d, 3J(H,H)
7.8, 4H, aromatic], 4.68 (s, 8H, NCH2Ph), 3.78 (s, 16H,
CH3NCH2CH2NH), 2.97 (s, 6H, CH3N); δC(50 MHz, D2O,
25 ЊC, TMS) 152.04, 141.54, 125.16 (aromatic), 54.86, 53.05,
44.29 (CH2N), 42.32 (CH3N); m/z (FAB): 441.3 [(Mϩ Ϫ
6HBr) ϩ 1], 521.3 [(Mϩ Ϫ 5HBr) ϩ 1], 603.2 [(Mϩ Ϫ 4HBr) ϩ
1], 683.1 [(Mϩ Ϫ 3HBr) ϩ 1] [Calc. for C24H40N8 ؒ 6HBr ϩ
4H2O (998.16): C, 28.89; H, 5.45; N, 11.23. Found: C, 28.86;
H, 5.48; N, 11.26%].
3
aromatic Ph], 7.41 [7, J(H,H) 7.8, 2H, aromatic Ph], 7.30 [d,
3J(H,H) 10, 4H, aromatic Ts], 3.84 [t, 3J(H,H) 5.9, 8H,
3
CH2NCHPh], 3.54 [t, J(H,H) 6, 8H, TsNCH2], 2.43 (s, 6H,
CH3Ts); δC(50 MHz, CDCl3, 25 ЊC, TMS) 161.93 (NCHPh),
143.37, 136.42, 130.69, 130.43, 129.77, 128.96, 127.34, 126.94
(aromatic), 51.43, 61.43 (TsNCH2CH2NCHPh), 21.53 (CH3Ts);
m/z (FAB) 711.2 (Mϩ ϩ 1) [Calc. for C38H42N6O4S2 ϩ 1/2H2O
(719.93): C, 63.40; H, 6.02; N, 11.67. Found: C, 63.46; H, 5.79;
N, 11.65%].
3,9,17,23-Tetracyano-6,20-bis(p-toluenesulfonyl)-3,6,9,17,20,
23-hexaazatricyclo[23.3.1.111,15]triaconta-1(29),11(30),12,14,25,
27-hexaene 16. The tetraimine 15 (5.5 g, 7.74 m) and NaBH4
(2.9 g, 77 m) were suspended in DMSO–EtOH (75 ml–30 ml),
refluxed until complete dissolution (~2 h) and then for an add-
itional 3 h. The solvent was evaporated under reduced pressure
and the residue taken up in H2O (75 ml) and extracted with
CH2Cl2 (5 × 125 ml). The organic layers were combined,
washed with brine (200 ml), dried over MgSO4, filtered and
evaporated to dryness. The residual white solid was dried under
high vacuum yielding 16 (C38H50N6O4S2, 718.98) which was
used in the next step without further purification, m/z (FAB)
720.0 (Mϩ ϩ 1).
3,9,17,23-Tetramethyl-6,20-bis(p-toluenesulfonyl)-3,6,9,17,20,
23-hexaazatricyclo[23.3.1.111,15]triaconta-1(29),11(30),12,14,25,
27-hexaene, 17. The macrocyclic polyamine 16 was suspended in
a mixture of formic acid (90%, 7.6 ml, 186 m) and formalde-
hyde (5.3 ml, 77.4 m) and refluxed for 9 h. The suspension
dissolved slowly during the course of the reaction. After cool-
ing, concentrated HCl (12 , 1.5 ml) was added and the
medium was refluxed for an additional 4 h. The reaction mix-
ture was cooled, basified with 5  KOH and extracted with
CH2Cl2 (4 × 60 ml). The organic layers were combined, washed
with brine (2 × 75 ml), dried over MgSO4, filtered and evapor-
ated to dryness. The residual yellowish solid was chromato-
4-(p-Tolylsulfonyl)-1,4,7-triaza-hepta-1,6-diyne, 12
To a solution of p-toluenesulfonyl chloride (33.2 g, 174 m) in
pyridine (100 ml) at 0 ЊC, iminodiacetonitrile 11 (15 g, 158 m)
in pyridine (100 ml) was added dropwise over a period of 1 h.
Stirring was maintained for 3 h at 0 ЊC and 2 h at room temp.
Ice (300 ml) was added to the solution and the reaction medium
was vigorously stirred for 1 h. The precipitate formed was fil-
tered, washed with H2O (500 ml), ethanol (200 ml) and diethyl
ether (200 ml), recrystallized from CH2Cl2–hexane and dried
under high vacuum. 12 (C11H11N3O2S, 249.29) was obtained as
a white crystalline solid (30.7 g, 78%), mp 99 ЊC. δH (200 MHz,
3
CDCl3, 25 ЊC, TMS) 7.74 [d, J(H,H) 8.4, 2H, aromatic], 7.42
[d, 3J(H,H) 8.2, 2H, aromatic], 4.3 (s, 4H, CH2NTs), 2.47 (s, 3H,
CH3Ts); δC(50 MHz, CDCl3, 25 ЊC, TMS) 146.16, 132.82,
130.58, 127.71 (aromatic), 112.65 (CN), 35.94 (CNCH2NTs),
21.72 (CH3Ts); m/z (EI): 249.1 (Mϩ) [Calc. for C11H11N3O2S
(249.29): C, 52.99; H, 4.45; N, 16.86. Found: C, 53.11; H, 4.49;
N, 16.97%].
4-(pTolylsulfonyl)-1,4,7-triazaheptane, 13. Compound 12
(24.9 g, 100 m) in dry THF (100 ml, distilled over Na) and 1 
B2H6 in dry THF (400 ml) were refluxed under a nitrogen
atmosphere for 20 h. The reaction mixture was cooled in an ice
bath and the excess B2H6 was slowly and carefully destroyed by
addition of 50% aqueous THF (100 ml). After evaporation of
J. Chem. Soc., Perkin Trans. 2, 1997
2079

View Article Online
graphed (SiO2 flash, methanol–CH2Cl2 1–5%) yielding 17
(C42H60N6O4S2, 777.1) as a white solid (2.82 g, 47% from 15).
δH (200 MHz, CDCl3, 25 ЊC, TMS) 7.7 (d, 4H, aromatic), 7.2 (d,
4H, aromatic), 7.25–7.0 (m, 8H, aromatic), 3.25 (t, 8H,
NCH2Ph), 3.2 (t, 8H, TsNCH2CH2N), 2.45 (t, 8H, TsNCH-
2CH2N), 2.4 (s, 12H, NCH3), 2.13 (s, 6H, CH3Ts); δC(50 MHz,
CDCl3, 25 ЊC, TMS) 143.0, 138.9, 136.9, 129.5, 129.3, 128.1,
127.7, 127.2 (aromatic), 62.4, 55.7, 46.4 (NCH2), 42.4 (NCH3),
21.5 (CH3Ts).
corresponds to the observed rate constant of the overall H/D
exchange process. The isotope effect of the first exchange on the
second, and the first and second on the third exchange were not
taken into account for the calculation of the rate constants. The
correlation coefficients (r2) for all the curves obtained were
>0.98.
The 1H NMR acquisition parameters used on the Bruker AC
200 spectrometer were as follows: SF, 200.1328; SW, 3000; PW,
3.5 µs; O1, 3800; RG = 10–100 (depending on the sample con-
centration); AQ, 2.736 seconds; LB, 0.1; NS, 16–32 scans
(depending on the sample concentration); TD, 16K; SI, 16K;
NE, 20–50. Depending on the speed of the reaction, the time
interval between two consecutive spectra varied from 1 to 30
min. For the slowest reactions (uncatalysed) the NMR tubes
were sealed with Teflon tape and kept in a thermostatted bath at
25 ± 0.1 ЊC. The spectra were recorded once every two weeks
over several months under the same conditions.
3,9,17,23-Tetramethyl-3,6,9,17,20,23-hexaazatricyclo-
[23.3.1.111,15]triaconta-1(29),11(30),12,14,25,27-hexaene, 6. The
macrocyclic polyamine 17 (0.21 g, 0.27 m), HBr–AcOH (33%,
6 ml) and phenol (0.4 g) were refluxed for 24 h during which a
brown solid was formed. After cooling, diethyl ether (100 ml)
was added to the reaction medium and the brown solid was
filtered and washed with diethyl ether (100 ml). The residual
solid was taken up in H2O (40 ml) and washed with CH2Cl2
(3 × 40 ml). The aqueous phase was eluted on a Dowex 1X4
(200–400 mesh) column in the OHϪ form yielding the free
polyamine 6 after evaporation of the eluent (H2O). It was then
converted to the corresponding hexahydrobromide upon dis-
solution in concentrated aqueous HBr (48%, 5ml), evaporation
to dryness, recrystallization from methanol and drying under
high vacuum. The desired compound 6 (C28H46N6 ؒ 6HBr,
952.17) was obtained as a white microcrystalline powder (0.22
g, 85%). δH (200 MHz, D2O, 25 ЊC, TMS) 7.9 (s, 2H, aromatic),
7.8 (s, 6H, aromatic), 4.6 (s, 8H, NCH2Ph), 3.66 (s, 16H,
NHCH2CH2NCH3), 3.1 (s, 12H, NCH3); δC(50 MHz, D2O,
25 ЊC, TMS) 135.5, 135.4, 132.6, 131.6 (aromatic), 62.1, 52.6,
44.5 (CH2N), 42.5 (CH3N); m/z (FAB): 466 [(Mϩ Ϫ 6HBr)
ϩ 1] [Calc. for C28H46N6 ؒ 6HBr ϩ 1 CH3OH (984.21): C,
35.39; H, 5.74; N, 8.54. Found: C, 35.44; H, 5.69; N, 8.52%].
Acknowledgements
We thank A. Jaffrezik and A. Van Dorsselaer (Laboratoire de
Spectrométrie de Masse, Université Louis Pasteur, Strasbourg)
for the determination of the mass spectra and the NSERC,
Canada, for a post-doctoral fellowship to C. D.
References
1 (a) A. Fersht, Enzyme Structure and Mechanisms, W. H. Freeman,
New York, 2nd edn., 1984, pp. 69–75; (b) W. P. Jencks, Catalysis in
Chemistry and Enzymology, Dover Publications, New York, 1987,
pp. 120–128; (c) L. Stryer, Biochemistry, W. H. Freeman, New York,
3rd edn., 1988; (d) C.-H. Wong and G. M. Whitesides, Enzymes
in Synthetic Organic Chemistry, Tetrahedron Organic Chemistry
Series, vol. 12, Pergamon Press, Oxford, 1994, pp. 195–251; (e)
I. A. Rose, Methods Enzymol., 1982, 87, 84; ( f ) L. C. Kurz, S. Shah,
B. R. Crane, L. J. Donald, H. W. Duckworth and G. R. Drysdale,
Biochemistry, 1992, 31, 7899; (g) G. M. Smith and A. S. Mildvan,
Biochemistry, 1981, 20, 4340; (h) G. A. Hamilton and F. H.
Westheimer, J. Am. Chem. Soc., 1959, 81, 6332; (i) S. G. Warren,
B. Zerner and F. H. Westheimer, Biochemistry, 1966, 5, 817;
(j) E. Grazi, T. Cheng and B. L. Horecker, Biochem. Biophys. Res.
Commun., 1962, 7, 250; (k) J. C. Speck, Jr., P. T. Rowley and
B. L. Horecker, J. Am. Chem. Soc., 1963, 85, 1012; (l) B. L.
Horecker, O. Tsolas and C. Y. Lai, The Enzymes, 1972, 7, 213;
(m) O. Tsolas and C. Y. Lai, The Enzymes, 1972, 7, 259; (n) R. F.
Gould, Bioorganic Chemistry, American Chemical Society,
Washington, DC, 1971, pp. 390–412.
2 (a) The Chemistry of Enols, ed. Z. Rappoport, Wiley, New York,
1990; (b) Iminium Salts in Organic Chemistry, Advances in Organic
Chemistry, ed. E. C. Taylor, Wiley, New York, 1976.
3 (a) W. P. Jencks, Catalysis in Chemistry and Enzymology, Dover
Publications, New York, 1987, pp. 133–146; (b) H. Dugas,
Bioorganic Chemistry, Springer Verlag, New York, 2nd edn.,
1989, p. 527; (c) C. Walsh, Ann. Rev. Biochem., 1978, 47, 881;
(d) J. N. Lowe and L. L. Ingraham, An Introduction to Biochemical
Reaction Mechanisms, Foundation of Molecular Biology Series,
Prentice Hall, Englewood Cliffs, NJ, 1974.
4 (a) R. Breslow, J. W. Canary, M. Varney, S. T. Wadell and D. Yang,
J. Am. Chem. Soc., 1990, 112, 5212; (b) M. Ando, H. Kuzuhara and
J. Watanabe, Bull. Chem. Soc. Jpn., 1990, 63, 88; (c) I. Tabushi, Pure
Appl. Chem., 1986, 58, 1529; (d) Y. Murakami, J.-I. Kikuchi and
N. Shiratori, Bull. Chem. Soc. Jpn., 1989, 62, 2045; (e) M. D.
Broadhurst and D. J. Cram, J. Am. Chem. Soc., 1974, 96, 581;
( f ) Y. Murakami, J.-I. Kikuchi, Y. Hisaeda and O. Hayashida,
Chem. Rev., 1996, 96, 721; (g) E. E. Snell and S. J. Di Mari, The
Enzymes, 1970, 2, 335; (h) O. A. Gansow and R. H. Holm, J. Am.
Chem. Soc., 1969, 91, 5984; (i) S. C. Zimmerman and R. Breslow,
J. Am. Chem. Soc., 1986, 106, 1490; (j) R. Breslow, A. W. Czarnik,
M. Lauer, R. Leppkes, J. Winkler and S. C. Zimmerman, J. Am.
Chem. Soc., 1986, 108, 1969; (k) I. Tabushi, Y. Kuroda, M. Yamada,
H. Higashima and R. Breslow, J. Am. Chem. Soc., 1985, 107, 5545;
(l) F. P. Schmidtchen, in Bioorganic Chemistry in Healthcare and
Technology, ed. U. K. Pandit and F. C. Alderweireldt, Plenum
Press, New York, 1991; (m) D. A. Jaeger, M. D. Broadhurst and
D. J. Cram, J. Am. Chem. Soc., 1979, 101, 717; (n) R. Breslow and
A. W. Czarnik, J. Am. Chem. Soc., 1983, 105, 1390; (o) V. M.
Shanbhag and A. E. Martell, J. Am. Chem. Soc., 1991, 113, 6479;
Kinetic measurements
1–5 ml stock solutions (100 m in D2O) of the macrocyclic
polyamines obtained in their protonated forms were prepared
in volumetric flasks and the pH was adjusted to 6 by adding
very small volumes of NaOH (0.1–5 ) and HCl (0.1–5 ) using
a Metrohm 636 pH-Meter equipped with a microelectrode. A
few assays on a small sample (0.5 ml) were performed in order
to determine approximately the amount of base or acid that
should be added in order to reach a given pH. Care was taken
to account for the solvent molecules that could not be removed
from the polyamine solids even after drying under high vacuum
for 48 h. A 5 ml stock solution (500 m in D2O) of sodium
pyruvate was prepared in a volumetric flask and used as such.
All the solutions could be stored at ϩ4 ЊC for several months
without apparent deterioration.
For each kinetic assay, using a set of micropipettes, the
appropriate volume of the polyamine receptor that yields the
desired concentration was added to a solution containing
(CD3)2SO (60% v/v of the final volume) and the appropriate
volume of D2O so that when the pyruvate solution was added,
the final volume was 600 µl. For instance, for the preparation of
a 60% (v/v) (CD3)2SO–D2O solution of polyamine (10 m) and
pyruvate (50 m) at pH 7, 60 µl of the polyamine stock solution
were added to a mixture of (CD3)2SO (360 µl) and of D2O (120
µl) in an Eppendorf tube. 60 µl of the stock solution of pyru-
vate was added and the pH was rapidly adjusted to 7, under
magnetic stirring, by adding a few microlitres of NaOH (0.1–5
) or HCl (0.1–5 ). In any case, the total volume of base and/or
acid added to adjust the pH did not exceed 10 µl. The solution
was then immediately transferred to a 5 mm NMR tube and
the course of the H/D exchange reaction was monitored at
1
25 ± 3 ЊC by integration of the H NMR CH3 signal of pyru-
vate over several half-lives. The spectra were recorded 1–5 min
after sample mixing. The 1H NMR signal of DMSO was used
as an internal reference (2.5 ppm). The semi-logarithmic repre-
sentation of the normalized surface of the 1H NMR CH3 signal
as a function of time yields a straight line, the slope of which
2080
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
(p) J. T. Koh, L. Delaude and R. Breslow, J. Am. Chem. Soc., 1994,
116, 11234.
16 J. Jazwinski, J.-M. Lehn, R. Meric and J.-P. Vigneron, Tetrahedron
Lett., 1987, 28, 3489.
5 (a) A. Bianchi, M. Micheloni and P. Paoletti, Coord. Chem. Rev.,
1991, 110, 17; (b) B. Dietrich, M. W. Hosseini, J.-M. Lehn and
R. B. Sessions, Helv. Chim. Acta, 1983, 66, 1262; (c) E. Kimura, Top.
Curr. Chem., 1985, 128, 113; (d) B. Dietrich, Pure Appl. Chem., 1993,
65, 1457.
17 For an estimate of the lifetime of various iminium species, see:
S. Eldin and W. P. Jencks, J. Am. Chem. Soc., 1995, 117, 4851.
18 [5] = 10 m, [acetoacetate] = 50 m, pD 7, 25 ЊC, 50% (CD3)2SO–
D2O. The control experiment was performed under the same
experimental conditions in the absence of 5.
6 (a) J.-M. Lehn, Pure Appl. Chem., 1979, 51, 979; Ann. N.Y. Acad.
Sci., 1986, 471, 41; Angew. Chem., Intl. Ed. Engl., 1988, 27,
89; Supramolecular Chemistry, VCH, Weinheim, 1995, ch. 5;
(b) M. W. Hosseini, Bioorg. Chem. Front., 1993, 3, 67.
7 H. Fenniri, J.-M. Lehn and A. M. Rigault, Angew. Chem., Intl. Ed.
Engl., 1996, 35, 337.
19 J. P. Guthrie and F. Jordan, J. Am. Chem. Soc., 1972, 94, 9136.
20 Electrostatic acceleration by a factor of about 104 has been reported
for cationic ketones: J. B. Tobin and P. A. Frey, J. Am. Chem. Soc.,
1996, 118, 12 253.
21 In the aqueous–organic media used in these studies, the pKa values
of the polyamines investigated should not be affected significantly.10
For the pKa values of 1 and 2, see ref. 5a, for 3, see ref. 13, for 5, see
ref. 14, for analogues of 4 and 9 see: C. Bazzicalupi, A. Bencini,
A. Bianchi, V. Fusi, C. Giorgi, P. Paoletti, A. Stefani and
B. Valtancoli, J. Chem. Soc., Perkin Trans. 2, 1995, 275. For the
other macrocyclic polyamines, see ref. 5a,b for an overall view about
the structural effects on the pKa values and the protonation pattern.
22 The rate advantage of α-proton abstraction from an iminium
compound compared to the corresponding ketone can be as high as
108-fold and is dependent on the pH, the pKa of the amine and the
pKa of the acceptor base: R. D. Roberts, H. E. Ferran, Jr., M. J.
Gula and T. A. Spencer, J. Am. Chem. Soc., 1980, 102, 7054.
23 (a) E. H. Cordes and W. P. Jencks, Biochemistry, 1962, 1, 773;
(b) W. P. Jencks and E. Cordes, in Proceedings of a Symposium on
Chemical and Biological Aspects of Pyridoxal Catalysis, ed.
E. E. Snell, P. M. Fasella, A. Braunstein and A. Fanelli, Pergamon
Press, New York, 1963, p. 57.
8 pKa of pyruvic acid in water = 2.26;9a this value was expected to
increase by up to 3–6 pKa units in the semi-organic medium used in
this study.9b,c The pH profile (Fig. 5) indicates that the pKa is around
4. pKa of α-CH3 of pyruvate in water = 16.58.10
9 (a) A. E. Martell and R. M. Smith, Critical Stability Constants,
Plenum Press, New York, 1977, vol. 3; (b) C. Reichardt, Solvents and
Solvent Effects in Organic Chemistry, VCH, Weinheim, 2nd edn.,
1990, pp. 81–85; (c) E. Buncel and H. Wilson, Adv. Phys. Org.
Chem., 1977, 14, 133.
10 Y. Chiang, A. J. Kresge and P. Pruszynski, J. Am. Chem. Soc., 1992,
114, 3103.
11 J. Hine, F. E. Rogers and R. E. Notari, J. Am. Chem. Soc., 1968,
90, 3279; J. Hine, E. F. Glod, R. E. Notari, F. E. Rogers and
F. C. Schmalstieg, J. Am. Chem. Soc., 1973, 95, 2537; J. Hine,
M. S. Sholod and R. A. King, J. Am. Chem. Soc., 1974, 96, 835;
J. Hine and R. L. Flachskam, Jr., J. Org. Chem., 1974, 39, 863;
J. Hine and S. S. Ulrey, J. Org. Chem., 1974, 39, 3231; J. Hine, Acc.
Chem. Res., 1978, 11, 1.
12 (a) W. J. Albery, R. P. Bell and A. L. Powell, Trans. Faraday Soc.,
1965, 61, 1194; (b) J. Damitio, G. Smith, J. E. Meany and Y. Pocker,
J. Am. Chem. Soc., 1992, 114, 3081.
1
24 The ratio hydrate–ketone determined by H NMR spectroscopy is
the same within 5–10% in the absence and presence of 10. At pH 2,
3, 4, 5 and 6 the percentage of pyruvate in the hydrate form is
respectively 57, 55, 33, 8 and 2.
25 H. Adams, N. A. Bailey, D. E. Fenton, C. Fukuhara and
M. Kanesato, Supramolecular Chem., 1993, 2, 325.
13 J. Comarmond, P. Plumeré, J.-M. Lehn, Y. Agnus, R. Louis,
R. Weiss, O. Kahn and I. Morgenstern-Badarau, J. Am. Chem. Soc.,
1982, 104, 6330.
14 R. Menif, A. E. Martell, P. J. Squattrito and A. Clearfield, Inorg.
Chem., 1990, 29, 4723.
15 A. E. Martell, in Crown Compounds, Towards Future Applications,
ed. S. R. Cooper, VCH, Weinheim, 1992, pp. 99–134.
Paper 7/00937B
Received 7th February 1997
Accepted 16th June 1997
J. Chem. Soc., Perkin Trans. 2, 1997
2081