Acceleration of p-Nitrophenyl Ester Cleavage by Zn(II)-Organized Molecular Receptors

Article abstract of DOI:10.1021/jo970783f

Tris(2-aminoethyl)amine (TREN) has been functionalized by introducing phenolic residues in the tripodal ligand side arms. The resulting functionalized ligands 1-5 form stable complexes with Zn(II) ions at pH > 6-6.5. The conformation of the Zn(II) complex

Full text of DOI:10.1021/jo970783f

J . Org. Chem. 1997, 62, 7621-7628
7621
Acceler a tion of p-Nitr op h en yl Ester Clea va ge by Zn (II)-Or ga n ized
Molecu la r Recep tor s
Paolo Tecilla, Umberto Tonellato,* and Andrea Veronese
University of Padova, Department of Organic Chemistry and Centro CNR Meccanismi di Reazioni
Organiche, Via Marzolo, 1, 35131 Padova, Italy
Fulvia Felluga and Paolo Scrimin*
University of Trieste, Department of Chemical Sciences, Via L. Giorgieri, 1, 34127 Trieste, Italy
Received May 2, 1997X
Tris(2-aminoethyl)amine (TREN) has been functionalized by introducing phenolic residues in the
tripodal ligand side arms. The resulting functionalized ligands 1-5 form stable complexes with
Zn(II) ions at pH > 6-6.5. The conformation of the Zn(II) complexes is such to form an ill-defined
cavity with the metal ion occupying its bottom and the aromatic residues defining its hydrophobic
walls. In these Zn(II) complexes one of the phenolic hydroxyls is, depending on the structure of
the ligand, up to 1.3 pKa units more acidic than that of phenol itself. This enhanced acidity is
attributed to second sphere coordination to the metal center. The complexes, particularly 1‚Zn(II),
behave as molecular receptors of p-nitrophenyl esters of carboxylic acids with binding constants
g300 M-1 for those substrates capable of coordination to the Zn(II) ion (p-nitrophenyl nicotinate,
PNPN, p-nitrophenyl isonicotinate, PNPIN and p-nitrophenyl urocanoate, PNPU). At pH 8.3 they
also accelerate the cleavage of these esters with rate accelerations with respect to the uncatalyzed
hydrolysis of up to 60 times, depending on the structure of the substrate. The kinetic analysis of
the process shows that the rate effects are due to two independent mechanisms: a bimolecular
process that does not comprise binding of the substrate and a pseudointramolecular process within
the supramolecular complex made of ligand, metal ion, and substrate. In both cases the nucleophile
is one of the phenolic hydroxyls of the functionalized side arms of the TREN-based complex which,
in the first step, is acylated by the substrate and eventually slowly hydrolyzes turning over the
catalyst. Determination of second-order rate constants shows that the nucleophilicity of the phenolic
hydroxyls is higher than that of a substituted phenol of the same pKa. Comparison of the
metalloreceptor 1‚Zn(II) with cyclodextrins allowed one to highlight similarity and differences
between the two receptors.
In tr od u ction
tweezers-like ligand which forms a pseudomacrocycle
upon binding two Cu(II) ions. Furthermore self-as-
sembling supramolecular structures have been obtained
from properly designed ligands and different metal ions.
These range from helices (Lehn,7 Constable8), catenanes
(Sauvage9), macrocycles (Stang,10 Hunter,11 Fujita12), and
peptide helices bundles (Ghadiri13).
Our idea was to synthesize a ligand with properly
appended functional groups which, upon addition of
Zn(II) ions, could (a) assume a specific conformation
suitable for binding of selected substrates and (b) act as
a catalyst of a transacylation reaction. A reaction mech-
anism in which binding is portal to successive pseudo-
The Zn(II) ion in biological systems plays two relevant
roles:1 (i) stabilization of structures of proteins and (ii)
participation in the catalytic site of several enzymes.
Examples are provided by zinc finger2 proteins and
carboxypeptidase A.3 Systems are also known (liver
alcohol dehydrogenase, for instance) in which the metal
ion plays both roles within the same protein.
Synthetic systems exploiting these aspects of coordina-
tion chemistry of Zn(II) (and other transition metal ions
as well) have been reported by several laboratories. If
we focus on the aspect of the stabilization of specific
conformations of structures we may cite several ex-
amples.4 For instance, Schneider5 synthesized a diphen-
ylmethane-based molecule which forms a macrocycle as
a Cu(II) complex and we studied6 a pyridine-based,
(7) (a) Baxter, P. N. W.; Lehn, J .-M.; Fischer, J .; Youinou, M.-T.
Angew. Chem. Int. Ed. Engl. 1994, 33, 2284. (b) Hasenknopf, B.; Lehn,
J .-M.; Kneisel, B. O.; Baum, G.; Fenske, D. Angew. Chem. Int. Ed. Engl.
1996, 35, 1838.
(8) (a) Constable, E. C. Tetrahedron 1992, 48, 10013. (b) Constable,
E. C.; Heirtzler, F. R.; Neuburger, M.; Zehnder, M. J . Chem. Soc.,
Chem. Commun. 1996, 933.
(9) (a) Dietrich-Buchecker, C. O.; Sauvage, J . P. Chem. Rev. 1987,
87, 795. (b) Sauvage, J . P. Acc. Chem. Res. 1990, 23, 219. (c) Dietrich-
Buchecker, C. O.; Sauvage, J . P. Tetrahedron 1990, 46, 503.
(10) (a) Olenyuk, B.; Whiteford, J . A.; Stang, P. J . J . Am. Chem.
Soc. 1996, 118, 8221. (b) Stang, P. J .; Chen, K.; Arif, A. M. J . Am.
Chem. Soc. 1995, 117, 8793. (c) Stang, P. J .; Cao, D. H. J . Am. Chem.
Soc. 1994, 116, 4981.
* Author to whom correspondence should be addressed. Tel.: +39-
49-8275276. Fax: +39-49-8275239. E-mail: scrimin@pdchor.chor.
unipd.it.
X Abstract published in Advance ACS Abstracts, September 15, 1997.
(1) (a) Lippard, S. J .; Berg, J . M. In Principles of Bioinorganic
Chemistry; University Science Books: Mill Valley, 1994, Chapter 10.
(b) Sigel, H.; Martin, R. B. Chem. Soc. Rev. 1994, 83. (c) Vallee, B. L.;
Auld, D. S. Acc. Chem. Res. 1993, 26, 543.
(2) Klug, A.; Rhodes, D. Trends Biochem. Sci. 1987, 12, 464.
(3) Christianson, D. W.; Lipscomb, W. N. Acc. Chem. Res. 1989, 22,
62.
(4) Canary, J . W.; Gibb, B. C. Progress Inorg. Chem. 1997, 45, 1.
(5) Schneider, H.-J .; Ruf, D. Angew. Chem. Int. Ed. Engl. 1990, 29,
1159.
(11) Chi, X.; Guerin, A. J .; Haycock, R. A.; Hunter, C. A.; Sarson, L.
D. J . Chem. Soc., Chem. Commun. 1995, 2567.
(12) (a) Fujita, M.; Yazaki, J .; Ogura, K. J . Am. Chem. Soc. 1990,
112, 5645. (b) Fujita, M.; Yazaki, J .; Ogura, K. Tetrahedron Lett. 1991,
32, 5589. (c) Fujita, M.; Nagao, S.; Lida, K.; Ogata, K.; Ogura, K. J .
Am. Chem. Soc. 1993, 115, 1574.
(6) Scrimin, P.; Tecilla, P.; Tonellato, U.; Vignaga, N. J . Chem. Soc.,
Chem. Commun. 1991, 449.
(13) Ghadiri, M. R.; Soares, C.; Choi, C. J . Am. Chem. Soc. 1992,
114, 825.
S0022-3263(97)00783-4 CCC: $14.00 © 1997 American Chemical Society

7622 J . Org. Chem., Vol. 62, No. 22, 1997
Tecilla et al.
Ch a r t 1
Ch a r t 2
intramolecular reactivity is a rule in enzymatic reac-
tions14 and our system was conceived with this in mind.
The considerable interest in supramolecular systems15
stems also from the search for simple molecules that,
operating on the basis of principles similar to those of
enzymes, may allow one to develop new and fully
synthetic catalysts. We focused our attention on tris(2-
aminoethyl)amine (TREN), a strong ligand for several
transition metal cations; thus, for the formation constant
of the 1:1 complex with Zn(II) a log Kb ≈ 15 was
reported.16 An appealing aspect of the Zn(II)‚TREN
complex was that being the typical coordination geometry
a trigonal bipyramid,16a the coordination of the four
nitrogens of the ligand leaves a fifth position available
for binding of substrates functionalized with suitable
donor groups. Eventually we finalized our endeavors
toward the synthesis of ligand 1 and similarly function-
alized ligands 2-5 (Chart 1) on the assumption that a
phenolic group would be an acceptable nucleophile to test
the validity of our hypothesis. A preliminary investiga-
tion of compounds 1 and 2 gave promising results.17 This
paper reports the full analysis of the TREN-based
phenolic receptors 1-5 as transacylation catalysts of the
p-nitrophenyl ester of carboxylic acids illustrated in
Chart 2.
toxyphthalic anhydride with TREN and subsequent
reduction of phthalimide derivative with LiAlH4. Details
of the syntheses and spectroscopic characterization of
new ligands are in the Experimental Section.
The synthesis of all substrates but 4-carboxy-p-nitro-
phenyl benzoate (CPNPB) has been already reported.
CPNPB was obtained by converting commercially avail-
able 4-carboxybenzaldehyde into the corresponding p-
nitrophenyl ester and oxidizing the aldehyde to the
carboxylic acid.
Com p lex F or m a t ion w it h Zn (II). Because of the
competition with protons, the complete formation of the
1:1 complexes with Zn(II) does not occur below pH 6-6.5,
depending on the ligands. This is clearly highlighted in
the pH titration curves for ligands 1 and 2 (see Figure
S1 of the Supporting Information) by the buffer region
between pH 5.8 and 6.3 where three OH- equivalents
are taken up with the loss of three protons of the three
protonated secondary amines. At the concentration
employed for these titrations the complexes formed are
not soluble at pH > 8.5 and, hence, we cannot evaluate
the pKa of the phenolic functions from the titration
experiments. For this reason we also measured the
spectroscopic changes, as a function of pH, in the 285-
295 nm wavelength region where the strong absorbance
of the phenolate anion allows one to determine the pKa
of the phenols of the complexes. These are reported in
Table 1 along with those of simple phenols obtained from
the literature.18 It should be noted that because of the
broad pH interval associated with the change of absor-
bance due to the formation of phenolates, the pKa values
are less precise than those obtained for simple phenols.
From the analysis of Table 1 we note that (a) one of
the three phenolic -OH groups of complexes 1‚Zn(II) and
3‚Zn(II) is more acidic than the other two and more acidic
than phenol itself by more than one pKa unit; (b) the
above effect is more pronounced for 1‚Zn(II) and vanishes
in the case of complex 4‚Zn(II); (c) the acidity of the single
phenolic -OH of 5‚Zn(II) is also lower than that of phenol
although to a lower extent than that of the most acidic
Resu lts a n d Discu ssion
Syn th eses. The synthesis of ligands 1-3 was neatly
performed by condensation of TREN with three equiva-
lents of the proper aldehyde and subsequent reduction
of the imine with NaBH4. Compound 5 was obtained by
condensation of two equivalents of 3-methoxybenzalde-
hyde with Boc-monoprotected TREN, reduction, depro-
tection, and condensation with 3-hydroxybenzaldehyde
followed again by reduction. Single-arm ligand 6 was
analogously obtained by condensation/reduction of N,N-
dimethyl-1,2-diaminoethane with 3-hydroxybenzalde-
hyde. Derivative 4 was obtained by reaction of 4-ace-
(14) Fersht, A. Enzyme Structure and Mechanism: Freeman: Read-
ing, 1977.
(15) Lehn, J .-M. Angew. Chem. Int. Ed. Engl. 1990, 29, 1304.
(16) (a) Anderegg, G.; Gramlich, V. Helv. Chim. Acta 1994, 77, 685.
(b) Canary, J . W.; Xu, J .; Castagnetto, J . M.; Rentzeperis, Marky, L.
A. J . Am. Chem. Soc. 1995, 117, 11545.
(17) Scrimin, P.; Tecilla, P.; Tonellato, U.; Valle, G.; Veronese, A.
J . Chem. Soc., Chem. Commun. 1995, 1163.
(18) Stefanidis, D.; Cho, S.; Dhe-Paganon, S.; J encks, W. P. J . Am.
Chem. Soc. 1993, 115, 1650.

Accleration of p-Nitrophenyl Ester Cleavage
J . Org. Chem., Vol. 62, No. 22, 1997 7623
Ta ble 2. Ra te Con sta n tsa for th e Clea va ge of th e
Differ en t p-Nitr op h en yl Ester s w ith Zn (II) Com p lexes of
Liga n d s 1-6
a
Ta ble 1. Ap p a r en t p Ka of th e P h en olic Gr ou p s of
Com p lexes 1‚Zn (II) a n d 3-5‚Zn (II) a n d Th ose of Sim p le
P h en ols
104kψ
,
103klim
s-1 (Kb, M-1
,
c
pKa (number of protons)
entry ligandb substrate
s-1
kψ/ko
)
klim/ko
system
spectroscopicb
kineticc
1
2
3
4
5
6
7
8
9
noned
1
2
3
4
5
6
PNPIN
PNPIN
PNPIN
PNPIN
PNPIN
PNPIN
PNPIN
PNPIN
6.3
51.4
8.2
19.5
1.7f
22.5
17.0
6.3
1
-
1‚Zn(II)
8.5(1); 9.3(2)
8.7(1); 9.6(2)
9.6(3)
8.4
8.2 14.5 (380 ( 45)
23
3‚Zn(II)
1.3
3.1
6.8f
3.6
2.7
1
2.5
1
8.0
1
5.1
1
17.1
1
8.3
1
10
e
e
e
e
g
-
-
-
4‚Zn(II)
5‚Zn(II)
8.9
8.8
phenold
9.81
8.44
8.03
3,4-dichlorophenold
3,5-dichlorophenold
TREN
TRENh PNPIN
16.0
0.97
7.8
a
b
See text for comments. [ligand] ) [Zn(NO3)2] ) 0.1 mM; I )
10
11
12
13
14
15
16
17
18
19
none
1
none
1
none
1
none
1
PNPN
PNPN
PNPU
PNPU
CPNPB
CPNPB
PNPB
PNPB
PNPA
PNPA
0.05 (NaNO3); 25 °C. c From the rate vs pH profiles of Figure 4.
5.8 (300 ( 50)
60
11
d
Data from ref 18.
0.081
0.41
0.24
4.1
0.071
0.59
0.25
2.5
-
0.089 (560 ( 90)
-
e
-
e
-
e
none
1
a
Conditions: EPPS buffer 0.05 M, pH 8.3; [substrate] ) 2-3
× 10-5 M; 25 °C. 1:1 complex with Zn(NO3)2. c Pseudo-first-order
b
rate constants at [complex] ) 1 × 10-3 M. Also without Zn(NO3)2.
d
e No appreciable curvature in the kψ vs [receptor] plot; see text.
f Kinetics were run in 1:1 (v/v) DMSO/H2O due to the low solubility
of the complex in water; ko (in the absence of complex) is 2.5 ×
10-5 s-1 under this conditions. Complex forms dimers, see text.
g
h
In the presence of three equivalents of phenol.
F igu r e 1. X-ray structure of complex 2‚Zn(II), 2 Cl-. One Cl-
is omitted, while the other is coordinated to the Zn(II) ion.
-OH of 1‚Zn(II). From these data it is apparent that
the Zn(II) ion in the TREN cavity of the complexes
influences the acidity of the phenols. Direct coordination
of the phenolic -OH to the metal center is not allowed
for geometric reasons and this effect is likely due to
second-sphere (or indirect) coordination.4 The fact that
one group is more acidic than the other two may be
explained with electrostatic interactions with the first
phenolate anion. In the rigid complex 4‚Zn(II) the -OH
groups cannot get close enough to the metal center and,
consequently, their pKa is very similar to that of phenol.
The structure of 2‚Zn(II) obtained by X-ray diffracto-
metry,17 shown in Figure 1, indicates that the TREN-
based ligands form complexes the conformation of which
provide a hydrophobic cavity. On the bottom of this ill-
defined cavity is located the Zn(II) ion and the walls are
constituted by the mobile aromatic groups. However, in
aqueous solution, because of hydrophobic interaction, the
aromatic groups may be forced to approach each other
to give a more closed, calyx-like conformation.
F igu r e 2. Rate constant vs complex concentration profiles for
the cleavage of PNPIN by different complexes in 0.05 M
HEPES buffer, pH 8.3, 25 °C. Ligands (L) are: b, 1; 4, TREN,
and 3 equiv of phenol; O, 2; 0, TREN.
Kin etics. Table 2 shows the pseudo-first-order rate
constants for the cleavage of esters shown in Chart 2 in
the presence of the Zn(II)-complexes of ligands 1-6, as
determined in aqueous solutions of pH 8.3 (0.05 M EPPS
buffer) by following spectrophotometrically at 400 nm the
release of p-nitrophenolate. This pH was the highest
attainable without formation of precipitates with all
complexes. This pH is also fairly close to the pKa of the
most acidic phenolic groups linked to the TREN platform
thus permitting advantage to be taken of a sizeable
concentration of phenolate, the expected nucleophile in
the process.
to the Zn(II) complex which behaves like a molecular
receptor. Conventional analysis19 of this curve for a 1:1
complex substrate/receptor gives a binding constant, Kb,
of 380 ( 45 M-1 and a value of the observed rate constant
when all substrate is bound to the metal receptor, klim
,
of 1.45 × 10-2 s-1. However, not all substrates and Zn(II)
complexes show sizeable curvatures in analogous plots,
thus preventing such analysis. For this reason Table 2
Figure 2 shows the kψ vs [1‚Zn(II)] profile obtained for
the cleavage of p-nitrophenyl isonicotinate (PNPIN). The
curvature of the plot suggests binding of the substrate
(19) Connors, K. H. Binding Constants, The Measurement of Mo-
lecular Complex Stability; Wiley: New York, 1987.

7624 J . Org. Chem., Vol. 62, No. 22, 1997
Tecilla et al.
shows rate constants determined at [complex] ) 1 × 10-3
M, and klim and Kb only for those systems giving reliable
values. It should be pointed out, however, that because
of the very limited concentration interval we could
explore (for the low solubility of the complexes, see
above), Kb < 80 M-1 could not be determined with
confidence from the plots kψ vs [complex] because the
curves may be fitted in some cases equally well as
straight lines or saturation curves. Analysis of Table 2
allows the following observations: (a) complex 1‚Zn(II)
is the most efficient system (compare entries 2 and 7);
(b) methylation of the three phenolic hydroxyls of 1
almost completely suppresses the reactivity (compare
entries 2 and 3); (c) complexes 5‚Zn(II) and 6‚Zn(II)
bearing one phenolic -OH are only moderately reactive
(entries 6 and 7). They differ considerably, however, in
the tendency to form binuclear complexes. Dimerization
of 1,2-diaminoethane-based complexes is, in fact, well
known20 and its adverse effect on their kinetic efficiency
has been recently reported;21 (d) TREN‚Zn(II) alone is
totally ineffective; (e) TREN‚Zn(II) and three equivalents
of phenol slightly accelerate the rate but at a much lower
extent than 1‚Zn(II) (entry 9); (f) the reactivity decreases
as the phenolic -OH is moved from the meta to the para
position or the complex is made more rigid as in 4
(compare entries 2, 4, and 5); (g) significant binding of
the substrate to the metal complexes is only present with
1‚Zn(II) and esters bearing a nitrogen donor atom
(PNPIN, p-nitrophenyl nicotinate (PNPN), and p-nitro-
phenyl urocanoate (PNPU)). Surprisingly in the case of
ester 4-carboxy-p-nitrophenyl benzoate (CPNPB), which
features a free carboxylate suitable for binding to a
Zn(II)‚TREN derivative, as reported,22 kinetic data do not
indicate a Kb large enough to be appreciated.
F igu r e 3. Effect of the addition of 4-methylpyridine on the
rate constant of cleavage of PNPIN (b) and PNPB (9) by
complex 1‚Zn(II). Conditions: [1‚Zn(II)] ) 1 × 10-3 M, 0.05 M
HEPES buffer, pH 8.3, 25 °C. The white circles represent the
effect observed in the absence of the complex.
allowance is made for hydrophobic contributions and the
fact that Zn(II) is already coordinated to the TREN
platform.24 Furthermore the limiting value of kψ ob-
tained for fully bound 4-methylpyridine in the case of
PNPIN (3.5 × 10-3 s-1) reveals that a considerable
amount of the rate acceleration is not attributable to a
mechanism that comprises binding of this substrate to
1‚Zn(II). With this ester the cleavage process associated
with binding appears to be more efficient than that
occurring without complexation to 1‚Zn(II) while the
contrary holds true for PNPB. A similar behavior has
been reported recently by Tee and Hoeven25 to explain
the reactivity of â-cyclodextrin with p-nitrophenyl ac-
etate.
p H Dep en d en ce of t h e R a t e of Clea va ge of
P NP IN. We have shown above that some of the phenolic
groups in our complexes, notably 1‚Zn(II) and, to a lesser
extent 5‚Zn(II), are particularly acidic. If those pheno-
lates are the nucleophilic species in the cleavage process
we would expect a first-order dependence on hydroxide
ion of the rate of p-nitrophenol release up to a pH at
which total ionization of the nucleophile has occurred.
For this reason we run kinetics of cleavage of substrate
PNPIN in the presence of 1 × 10-3 M 1‚Zn(II) and
5‚Zn(II) as well as in the absence of any metal complex.
The data are shown in Figure 4. The basic hydrolysis of
the ester exhibits the expected unity slope in the log kψ
vs pH profile in the pH interval 8.6-11.0. Below pH 8.6
the slope decreases suggesting a change of mechanism
in the hydrolytic process. In the presence of 1‚Zn(II) and
5‚Zn(II) the plots are linear with slope close to unity up
to pH ∼ 8.3 for 1‚Zn(II) and ∼ 8.7 for 5‚Zn(II). Above
these pHs the slopes of the plots decrease until at pH
close to 11 the rate becomes only slightly faster than that
The above observations rule out an involvement of a
Zn(II)-bound hydroxyl of a deprotonated water molecule
as the nucleophile in this process and point to one of the
phenolic groups of the complex as the nucleophilic
species. This is particularly evidenced by the lack of
reactivity of the methylated complex 2‚Zn(II).
Com petitive In h ibition . In order to establish whether
binding of the substrates to the Zn(II) complexes is a key
step in the mode of cleavage of the esters we ran kinetics
in the presence of a potential inhibitor, i.e., a molecule
able to coordinate to Zn(II) and hence to displace the
substrate from the complex. Figure 3 shows the effect
on the rate of release of p-nitrophenol by PNPIN and
p-nitrophenyl benzoate (PNPB) with 1 × 10-3 M 1‚Zn(II)
upon addition of increasing amounts of 4-methylpyridine.
We clearly observe a decrease of the rate in the case of
PNPIN. However, and quite unexpectedly, we observe
acceleration in the case of PNPB. Control experiments
in the presence of TREN‚Zn(II) and phenol showed that
4-methylpyridine does not affect the rate of cleavage of
PNPIN (Figure 3) and PNPB (data not shown). Analysis
of kψ vs [4-methylpyridine] gives Ki ) 45 ( 12 M-1 and
48 ( 8 M-1 for the binding of this molecule to 1‚Zn(II)
from the rate profile of PNPIN and PNPB, respectively.
These Ki values are in fair agreement with the reported23
value of 23 M-1 for its binding to Zn(II) alone when
(20) Arenare, E.; Paoletti, P.; Dei, A.; Vacca, A. J . Chem. Soc., Dalton
Trans. 1972, 736.
(21) Scrimin, P.; Ghirlanda, G.; Tecilla, P.; Moss, R. A. Langmuir
1996, 26, 6235.
(22) De Santis, G.; Fabbrizzi, L.; Licchelli, M.; Poggi, A.; Taglietti,
A. Angew. Chem. Int. Ed. 1996, 35, 202.
(23) Smith, R. M.; Martell, A. E. Critical Stability Constants;
Plenum: New York, 1975.
(24) The first effect increases while the second decreases its binding
constant to the metal ion.
(25) Tee, O. S.; Hoeven, J . J J . Am. Chem. Soc. 1989, 111, 8318.

Accleration of p-Nitrophenyl Ester Cleavage
J . Org. Chem., Vol. 62, No. 22, 1997 7625
16b
is probably related to the fact that because its pKa is
>9, only a small fraction is deprotonated.
Mech a n ism . The apparently complicated set of ki-
netic results reported above may be simply explained by
assuming that two distinct modes of action are involved:
a) a bimolecular process (equation 1) and (b) a pseudo-
intramolecular process occurring when the substrate is
bound to the complex (eq 2). In the process depicted in
eq (1) we assume that the “inhibitor” does not affect the
reactivity of the metalloreceptor and the substrate no
longer binds to the catalyst-inhibitor complex. These
assumptions are reasonable if we consider that the effect
on the acidity of the phenol is not due to direct coordina-
tion to the metal ion and the rather weak binding of the
substrates when coordination to the Zn(II) is prevented.27
k2
substrate + Zn(II)-complex
98 P
(1)
Kb
\w x
substrate + Zn(II)-complex
kb
substrate‚Zn(II)-complex
98 P (2)
When all substrate is bound to the metal complex the
only mechanism that operates is pseudointramolecular.
Under this condition kb is klim in Table 2. If binding of
the substrate to the molecular receptor (the metal
complex in our case) leads to a supramolecular complex
in which the pseudointramolecular reaction between
nucleophile (the phenolate ion) and ester is not easily
attainable for geometric reasons without at least partial
decomplexation of the substrate, it may well occur that
the bimolecular process is more efficient than the pseudo-
intramolecular one. This appears to be the case of PNPB,
for which addition of the “inhibitor” 4-methylpyridine
increases the observed rate constant (see Figure 3); in
such a case the reaction outside the cavity of the complex
(with no binding) is faster than that inside (when bound).
On the contrary with substrate PNPIN the process
indicated in eq (2) is more efficient than that of eq (1)
and addition of 4-methylpyridine inhibits the reaction
(see Figure 3) as expected. Coordination of the pyridine
nitrogen to the metal center likely results in a geometry
of binding more appropriate for the subsequent interac-
tion of the ester group with the phenolate ion. Obviously
this control of the geometry of binding is not attainable
with PNPB. However, even with PNPIN there is a
sizeable contribution of the purely bimolecular process
(eq 1) as the rate acceleration is not completely sup-
pressed by addition of the inhibitor. This dichotomy of
behavior should not be surprising although it is often not
considered in studies of catalysis through complexation.28
The two competing mechanisms are sketched in Figure
5. Similar mechanisms appear to operate, for instance,
in the cleavage of p-nitrophenyl acetate by â-cyclodextrin,
as mentioned above.25,29
F igu r e 4. log kψ vs pH profiles for the cleavage of PNPIN at
25 °C. O, No additives; 0, with [5‚Zn(II)] ) 1 × 10-3 M; b,
with [1‚Zn(II)] ) 1 × 10-3 M. Buffers used were: MES, pH
6.0-6.7; HEPES, pH 6.7-7.7; EPPS, pH 7.7-8.5; CHES, pH
8.7-9.9; CAPS, pH 10.3-11.2.
observed for the OH- catalyzed hydrolysis. Clearly above
this pH the hydrolytic process driven by aqueous OH-
becomes dominant as this species outnumbers the phe-
nolate ions present in solution. From the curvatures of
the plots we may calculate the pKa of the nucleophile
responsible for the rate acceleration in the cleavage of
PNPIN. Unfortunately, in the pH interval 8.5-10.2
solutions of 1‚Zn(II) were slightly turbid and the accuracy
of the kinetic experiments in that pH interval (dashed
region of upper curve of Figure 4) was less than desirable.
Nevertheless, we estimate an apparent pKa of 8.4 and
8.8 for 1‚Zn(II) and 5‚Zn(II), respectively. These values
are quite close to the pKa determined spectrophotometri-
cally for the most acidic phenol of 1‚Zn(II) and for that
of 5‚Zn(II) (see Table 1); we may conclude that one of
the deprotonated phenols of the Zn(II) complex is the
actual nucleophile.
Tu r n over . If this is the case, we should expect the
formation of an acylated intermediate which may even-
tually hydrolyze, turning over the nucleophile. Under
our kinetic conditions with substrate PNPIN we failed
to detect such an intermediate. However, carrying out
the reaction in an NMR tube using a 1:1 mixture CD3CN/
D2O (pD of the D2O component 8.3, unbuffered) we clearly
detected the formation of the product of acylation (see
Figure S2 of the Supporting Information). Such an
intermediate hydrolyzes (at pH 8.3, 0.05 M EPPS buffer)
with kψ ) 3.7 × 10-4 s-1, a rate substantially slower than
klim for the transacylation process when the substrate is
fully bound to 1‚Zn(II) (8.2 × 10-2 s-1). Thus, our
complex is not a real catalyst, in analogy with what has
been observed for the cleavage of esters with cyclodex-
trins.26 The slow rate of the hydrolytic cleavage provides
indirect evidence that the Zn(II)-bound H2O in this
complex is a rather poor nucleophile at this pH and this
The “inhibition” experiments allow one to estimate,
from the limiting value of kψ obtained for fully bound
(27) Note that data by Tee and Bozzi (J . Am. Chem. Soc. 1990, 112,
7815) suggest that in the case of cyclodextrin there may be binding of
the substrate and the inhibitor with formation of a ternary complex.
However, one can hardly think that this system, in which the
hydrophobic contribution to the binding is much lower than that of
cyclodextrins, can really bind both guest molecules.
(28) This has been recently pointed out by Williams, too: Pirrin-
cioglu, N.; Zaman, F.; Williams, A. J . Chem. Soc., Perkin Trans. 2 1996,
2561.
(26) VanEtten, R. L.; Clowes, G. A.; Sebastian, J . F.; Bender, M. L.
J . Am. Chem. Soc. 1967, 89, 3253.
(29) Tee, O. S. Adv. Phys. Org. Chem. 1994, 29, 1.

7626 J . Org. Chem., Vol. 62, No. 22, 1997
Tecilla et al.
F igu r e 5. Bimolecular (a) and pseudo-intramolecular (b)
modes of nuclephilic attack of the tripodal complex 1‚Zn(II)
to the ester substrates.
4-methylpyridine, the second-order rate constants, k2, for
the bimolecular process for the two substrates PNPIN
and PNPB: they are 9.07 and 1.99 × 10-1 M-1 s-1
,
respectively, when correction for the partial formation
of phenolate ion is made.30 A Brønsted-type correlation
for these two substrates using the three phenolic deriva-
tives reported in Table 1 is illustrated in Figure 6. This
plot can be used to compare the nucleophilicity of the
phenolate anion of 1‚Zn(II) with that of a phenol of the
same pKa. Quite clearly the phenolate ion of our complex
is more nucleophilic than predicted by its acidity (see
open symbols in Figure 6) and this indicates that in the
transacylation process with the metal complex there is
some sort of assistance in the formation of the product.
Assistance may be provided by hydrogen bonding with a
Zn(II)-bound H2O or with one of the other two phenolic
groups of the metal complex. This latter hypothesis is
supported by the fact that the log k2 value obtained for
5‚Zn(II), devoid of any other free phenolic -OH, fits very
nicely in the Brønsted-type plot with no evidence of extra
nucleophilicity.
F igu r e 6. Brønsted-type plots for the cleavage of PNPIN
(circles, â ) 0.85) and PNPB (squares, â ) 0.75) by phenols of
different pKa (filled symbols). Open symbols refer to second-
order rate constants for the bimolecular process with metal
complexes 1‚Zn(II), 1 and 4, and 5‚Zn(II), 2. Point 3 is the value
obtained from klim × Kb for 1‚Zn(II). See the text for details.
We may get a further estimate of the overall efficiency
of our catalyst, including the binding step, by determining
klim × Kb which is the apparent second-order rate
constant of the fully bound substrate. This value is 14.2
s-1 M-1 for 1‚Zn(II) and PNPIN when correction is
introduced for the percentage of phenolate ion at the pH
of the kinetic experiments. It is reported in Figure 6 so
that a comparison can be made between the second-order
rate constant of the pseudointramolecular and bimolecu-
lar processes involving 1‚Zn(II) (eqs 1 and 2) and the
bimolecular process involving a phenol of the same pKa.
The pseudointramolecular process involving 1‚Zn(II) is
roughly one order of magnitude more efficient than a
bimolecular process involving a simple phenolate ion but
only slightly better than the bimolecular process involv-
ing 1‚Zn(II) itself. This is consistent with our finding of
competition between the two mechanisms under the
conditions employed for our kinetic experiments.
Com p a r ison w ith Cyclod extr in s. For many aspects
our metalloreceptors, 1‚Zn(II) in particular, show a
mechanistic behavior very similar to that of cyclodextrins
(CDs) as has already been pointed out several times in
the discussion. Table 3 compares the kinetics and
binding constants of 1‚Zn(II) with those of CDs.33 Bind-
ing constants are higher for 1‚Zn(II). With these sub-
strates the driving force for binding likely switches from
purely hydrophobic (as for CDs) to metal coordination
and, only to a lesser extent, hydrophobic with the
metalloreceptor. The interaction with the metal center
is important in our system as can be appreciated by
comparing the binding constants of PNPIN and PNPN
(380 and 300 M-1, respectively) with that of PNPB (<80
M-1). However, in CDs the binding occurs with insertion
of the p-nitrophenyl moiety as highlighted by the very
small difference in kinetic effects between PNPIN and
As for the pseudointramolecular process, the klim values
of Table 2 evaluated for 1‚Zn(II) (entries 2, 11, and 13)
allow one to compare the effectiveness of this mechanism
with the three substrates PNPIN, PNPN, and PNPU. The
best substrate appears to be PNPN with a ratio between
the catalyzed transacylation process and the uncatalyzed
hydrolysis (klim/ko) of 60. This ratio goes down to 23 with
PNPIN and 11 with PNPU. Inspection of molecular
models suggests that the ester group in meta position with
respect to the pyridine nitrogen coordinated to Zn(II) in
the supramolecular complex is closer to the phenolate ion
in the case of PNPN than in that of PNPIN, while in the
case of PNPU the spacer between the imidazole and the
ester is too extended for an effective interaction with the
nucleophile. The effective molarities,31 EM, i.e., the
nominal concentration of a phenol of the same pKa
necessary to obtain a kψ matching the klim in the pseudo-
intramolecular process, are 1.3 × 10-2 and 3.3 × 10-2
M
for PNPIN and PNPN, respectively; rather modest if
compared with those of real intramolecular processes.31
However, similar values of EM have been found for the
cleavage of simple esters by cyclodextrins31 while much
higher values have been found for substrates properly
designed to match the rigid cavity of these molecular
receptors and interact with their nucleophilic rim.32
(30) Since one of the three phenols is considerably more acidic than
the other two no correction was made for their contribution to the rate
acceleration.
(31) Kirby, A. J . Angew. Chem. Int. Ed. Engl. 1996, 35, 707.
(32) (a) Breslow, R.; Czarniecki, M. F.; Emert, J .; Hamaguchi, H. J .
Am. Chem. Soc. 1980, 102, 762. (b) Breslow, R.; Trainor, G.; Ueno, A.
J . Am. Chem. Soc. 1983, 105, 2739.
(33) Daniele, A.; Fornasier, R.; Tonellato, U. unpublished results.

Accleration of p-Nitrophenyl Ester Cleavage
J . Org. Chem., Vol. 62, No. 22, 1997 7627
Ta ble 3. Com p a r ison of Bin d in g a n d Ra te Effects by
1‚Zn (II), r- a n d â-Cyclod extr in a (r-, â-CD) on th e
Clea va ge of Selected Ca r boxyla te Ester sb
eventually result in the kinetic phenomena experimen-
tally observed.
substrate
catalyst
klim/ko
Kb, M-1
Exp er im en ta l Section
PNPIN
PNPIN
PNPIN
PNPN
PNPN
PNPN
R-CDc
3.6
7.5
23
3.7
7.5
60
0.12
5.8
11
110
65
110
170
380
90
170
300
90
210
560
90
â-CDc
Gen er a l Con sid er a tion s. 1H-NMR spectra were recorded
at 200 or 250 MHz using (CH3)4Si as internal standard.
Melting points are uncorrected. Elemental analyses were
performed by the “Laboratorio di microanalisi” of the Depart-
ment of Organic Chemistry in Padova. Kinetic traces were
recorded on multicell instruments equipped with thermostated
cell holders. Temperature control was (0.1 °C.
Ma ter ia ls. Zn(NO3)2 was analytical grade commercial
product. Metal ion stock solutions were titrated against EDTA
following standard procedure.35 Buffer solutions were pre-
pared using Milli-Q water. The buffer components36 were used
as supplied by the manufacturers. Abbreviations used are as
follows: MES, 4-morpholineethanesulfonic acid; HEPES, 4-(2-
hydroxyethyl)-1-piperazineethanesulfonic acid; EPPS, 4-(2-
hydroxyethyl)piperazine-4-(3-propanesulfonic acid); CHES,
2-(cyclohexylamino)ethanesulfonic acid; CAPS, 3-(cyclohex-
ylamino)-1-propanesulfonic acid. PNPIN,37 PNPN,37 and
PNPU33 were synthesized according to reported procedures.
4-Ca r boxy-p-Nitr op h en yl Ben zoa te (CP NP B). 4-Car-
boxybenzaldehyde (300 mg, 2 mmol) was suspended in benzene
(10 mL) containing 8 mmol of SOCl2 and the slurry refluxed
until all of the solution became limpid (ca. 3 h). Benzene was
then evaporated and the solid dissolved in CH2Cl2 (10 mL), to
which p-nitrophenol (278 mg, 2 mmol) and Et3N (0.28 mL)
were added. After the mixture was stirred at room temper-
ature for 2 h, TLC (CH2Cl2) revealed the complete formation
of the ester. The solution was washed with water (3 × 10 mL)
and CH2Cl2 evaporated to leave a white solid that was
recrystallized from CH2Cl2/Et2O. The pure CPNPB (400 mg)
has a mp 190-191 °C. 1H-NMR (200 MHz, CDCl3), δ: 7.35,
8.27 (AA′BB′, 4H); 7.93, 8.24 (AA′BB′, 4H); 10.05 (s, 1H).
The above material (200 mg, 1.22 mmol) was dissolved in 5
mL of dry pyridine to which 220 mg of KMnO4 was added and
let to stir overnight at room temperature. After evaporation
of the solvent at reduced pressure, the solid was treated with
CH2Cl2, washed with 0.1 M HCl, and the organic layer filtered
through a celite pad. The crude material obtained after
evaporation of the solvent was chromatographed down a SiO2
column (CHCl3/EtOH, 95:5) to give 40 mg of pure CPNPB with
mp 211 °C dec. 1H-NMR (200 MHz, CD3CN), δ: 7.43, 8.27
(AA′BB′, 4H); 8.12, 8.20 (AA′BB′, 4H); 8.9 (brs, 1H). Anal.
Calcd for C14H9NO6: C, 58.54; H, 3.16; N, 4.88. Found: C,
58.46; H, 3.21; N, 4.79.
1‚Zn(II)d
R-CDc
â-CDc
1‚Zn(II)d
R-CDe
PNPU
PNPU
â-CDe
PNPU
1‚Zn(II)d
R-CDc
MNPINf
MNPINf
â-CDc
160
Data from ref 33. In aqueous solutions at 25 °C. c pH 9.5.
a
b
pH 8.3, this work. e pH 10.5. f m-Nitrophenyl isonicotinate.
d
PNPN paralleled by the very large difference between
PNPIN and m-nitrophenyl isonicotinate (MNPIN). The
selectivity of the cyclodextrins toward the geometry of
the substrate appears to be much higher than that of
1‚Zn(II). The ratios between klim/ko for the two substrates
PNPIN and MNPIN are 30.5 and 8.7 for R- and â-CD,
respectively, while that between PNPIN and PNPN with
1‚Zn(II) is only 2.6. This is consistent with the much
higher rigidity of the conformation of CDs compared with
that of our metalloreceptor.
Con clu sion
We have synthesized a series of TREN-based ligands
functionalized with different phenolic groups. Coordina-
tion of Zn(II) imparts a restricted conformation to the
complexes and affects the acidity of the phenolic hy-
droxyls as a function of their proximity to the metal
center. These complexes, 1‚Zn(II) in particular, act as
metalloreceptors of p-nitrophenyl esters of carboxylic
acids, at least of those which are able to coordinate to
Zn(II) ion.
The 1‚Zn(II) complex, which we have studied in more
detail, catalyzes the cleavage of the substrates with
formation of the product of acylation of one of the phenolic
hydroxyls followed by a slower hydrolysis. The transa-
cylation process may occur via a pseudointramolecular
attack within the supramolecular complex or via a simple
bimolecular mechanism. Kinetic benefits are apparent
from both mechanisms: in the bimolecular mechanism
because of assistance by one of the phenolic hydroxyls
in the transacylation, in the associative mechanism
because of the lower entropic price to be paid for the
nucleophilic attack. The mechanistic behavior of these
metalloreceptors is reminiscent of that found with cyclo-
dextrins although the latter show an higher selectivity
which may be ascribed to their much lower conforma-
tional flexibility.
As pointed out recently by Schneider34 “supramolecular
catalytic systems lack the effectiveness but not the
complexity of an enzymatic process”. As a consequence
they must be analyzed very carefully in order to dissect
the different parameters that contribute to the observed
rate enhancements. In the present case we have shown
that a metal center not directly involved in the process
studied may in fact influence many important param-
eters (e.g., conformation of the receptor, acidity of the
nucleophilic groups, recognition of the substrate) that
Gen er a l P r oced u r e for th e Syn th esis of Com p ou n d s
1 a n d 3. TREN (3.42 mmol) and 3- or 4-hydroxybenzaldehyde
(10.26 mmol) were dissolved in acetonitrile (20 mL) and
vigorously stirred at rt. After 30 min a white solid started to
formed. After 1.5 h the solid was filtered off, washed with
acetonitrile, and dried in vacuo. 1H-NMR analysis of the
material revealed the disappearance of the signal of the proton
of the aldehyde and appearance of a new signal (imine) at ca.
8.1 δ. The tris-imine derivative was then added to a solution
of 500 mg (13.2 mmol) of NaBH4 in ethanol and the reaction
let to stir overnight at rt. Work up with concentrated HCl,
filtration of the precipitate, and evaporation of the solvent
afforded a crude material which was purified twice by elution
through a Sephadex G-75 column using water as eluent.
Lyophilization of the proper fractions gave the expected
compounds as hydrochloride salts.
N ,N ,N -T r is {2-[((3-h y d r o x y p h e n y l)m e t h y l)a m in o ]-
eth yl}a m in e-3.5HCl (1‚3.5HCl): 556 mg; 1H-NMR (250
MHz, D2O) δ 2.66 (t, J ) 6.3 Hz, 6H); 2.94 (t, J ) 6.3 Hz, 6H),
3.99 (s, 6H), 6.7-6.9 (m, 9H); 7.18 (t, J ) 7.75 Hz, 3H). Anal.
(35) Holzbecher, Z. Handbook of Organic Reagents in Inorganic
Analysis; Wiley: Chichester, 1976.
(36) Good, N. E.; Winget, G. D.; Winter, W.; Connolly, T. N.; Izawa,
S.; Smith, R. M. M. Biochemistry 1966, 5, 467.
(37) Ljungqvist, A.; Folkers, K. Acta Chem. Scand., Ser. B 1988,
B42, 408.
(34) Schneider, H.-J .; Kramer, R.; Rammo, J . J . Am. Chem. Soc.
1993, 115, 8980.

7628 J . Org. Chem., Vol. 62, No. 22, 1997
Tecilla et al.
Calcd for C27H36O3N4‚3.5HCl‚1.5H2O: C, 52.37; H, 6.92; N,
9.05; Cl, 20.04. Found: C, 52.20; H, 6.72; N, 8.93; Cl, 20.57.
N ,N ,N -T r is {2-[((4-h y d r o x y p h e n y l)m e t h y l)a m in o ]-
eth yl}a m in e-3.5HCl (1‚3.5HCl): 440 mg; 1H-NMR (200
MHz, D2O) δ 2.62 (t, J ) 6 Hz, 6H), 2.88 (t, J ) 6 Hz, 6H),
3.94 (s, 6H), 6.73, 6.77, 7.14, 7.17 (AA′BB′, 12H). Anal. Calcd
for C27H36O3N4‚3.5HCl‚1.5H2O: C, 52.37; H, 6.92; N, 9.05; Cl,
20.04. Found: C, 51.93; H, 6.70; N, 8.89; Cl, 20.26.
was added to the reaction mixture, the ethanol evaporated,
and the aqueous solution extracted with CHCl3 (30 × 3 mL).
Evaporation of the organic layer gave a crude that was purified
by column chromatography (SiO2, CHCl3/MeOH, 10:1 then
CHCl3/MeOH/NH3(aq), 10:4:1) to give 800 mg of pure material.
1H-NMR (250 MHz, CDCl3), δ: 1.42 (s, 9H), 2.59 (m, 10H),
3.14 (brm, 2H), 3.74 (s, 4H), 3.79 (s, 6H), 5.73 (brm, 1H), 6.77
(d, 2H), 6.86 (s, 2H), 6.88 (d, 2H), 7.21 (t, 2H).
N ,N ,N -Tr is{2-[((3-m e t h oxyp h e n yl)m e t h yl)a m in o]-
eth yl}a m in e-3HCl (2‚3HCl). The preparation of compound
2 differs from the above general procedure for the workup after
reduction and successive purification. Thus, the reduction
mixture was treated with H2O (10 mL), extracted with CH2Cl2
(3 × 40 mL), and the organic solution dried over Na2SO4. The
crude material obtained after evaporation of CH2Cl2 was
chromatographed through a column of SiO2 (CH2Cl2/MeOH/
NH3(aq), 90:9:1) to give 1.3 g of pure material. This was then
converted in the hydrochloride salt by treatment of an Et2O
solution with Dioxane saturated with dry HCl. The white salt
that precipitates was collected by filtration. 1H-NMR (250
MHz, D2O), δ: 2.71 (t, J ) 7.3 Hz, 6H); 2.99 (t, J ) 7.3, 6H);
3.70 (s, 9H); 4.06 (s, 6H), 6.85-7.0 (m, 9H); 7.29 (t, J ) 9.5
Hz, 3H). Anal. Calcd for C30H42N4O3‚3HCl: C, 58.49; H, 7.36;
N, 9.09. Found: C, 58.38; H, 7.31; N, 8.95.
N,N,N-Tr is{2-(6-h yd r oxyisoin d olin yl)et h yl}a m in e-
3.5HCl (4‚3.5HCl). 5-Acetoxyphthalic anhydride (2.5 g, 11.25
mmol) was dissolved in 30 mL of xylene with 510 mg (3.47
mmol) of TREN and the mixture refluxed for 1 h. After the
mixture cooled, a dark oily material separated on the bottom
of the flask. This was decanted and the residual solvent
evaporated at reduced pressure. The oily material obtained
was purified by column chromatography (SiO2, CHCl3/EtOAc,
9:1) to give 480 mg (18.2% yield) of trisphthalimido deriva-
tive: 1H-NMR (200 MHz, CDCl3), δ: 2.33 (s, 9H), 3.02 (brs,
6H), 3.82 (brs, 6H), 7.37 (d, J ) 7.5 Hz, 3H), 7.47 (s, 3H), 7.68
(d, J ) 7.5 Hz, 3H).
This material was then suspended in 20 mL of dry tetrahy-
drofuran (THF) and, to the slurry, 10 mL of a 1 M solution of
LiAlH4 in THF was added and the reaction mixture refluxed
overnight over N2. After the mixtured cooled, it was worked
up with 2 mL of 1 M NaOH (CAUTION!), evaporated of the
solvent, and the solid continuosly extracted with ethanol.
Evaporation of the solvent gave the expected compound
contaminated with some impurities. These were removed by
treating the crude with a pH 1 aqueous solution followed by
continuous extraction with CHCl3. After evaporation of the
water the residue was chromatographed twice over a Sephadex
G15 column and, subsequently, over a Sephadex G10 column
using water as eluent. The final amount of pure material
collected was 150 mg. 1H-NMR (200 MHz, D2O), δ: 3.01 (brt,
J ) 6.5 Hz, 6H), 3.53 (brt, J ) 6.5 Hz, 6H), 4.65 (s, 12H), 6.7-
6.85 (m, 6H), 7.13 (d, J ) 7 Hz, 3H). Anal. Calcd for C30H36O3-
N4‚3.5HCl‚0.5H2O: C, 56.54; H, 6.41; N, 8.79; Cl, 19.47.
Found: C, 57.02; H, 6.51; N, 8.51; Cl, 19.56.
The monoprotected bis(methoxy) derivative was then dis-
solved in CH2Cl2 (4 mL) and treated with 2 mL of trifluoro-
acetic acid. After 1 h the solvent was evaporated under
reduced pressure, the crude treated with 20 mL of a 10%
aqueous solution of Na2CO3 and extracted with CHCl3 (20 ×
3 mL). After the material was dried and the solvent was
evaporated, 580 mg of deprotected compound was obtained.
1H-NMR (250 MHz, CDCl3), δ: 2.48 (t, 2H), 2.7 (m, 10H), 3.74
(s + m, 10H), 6.85 (m, 6H), 7.23 (t, 2H).
This compound was treated, as above for the Boc-monopro-
tected TREN, with 3-hydroxybenzaldehyde to give the imine
derivative which was analogously reduced with NaBH4 to the
crude final product. This was purified by elution through a
SiO2 column (CHCl3/MeOH/NH3(aq), 100:20:0.5) to give 450
mg of pure 5. 1H-NMR (250 MHz, CDCl3), δ: 2.50 (m, 8H),
2.72 (t, J ) 6 Hz, 4H), 3.70 (s, 4H), 3.74 (s, 6H), 3.79 (s, 2H),
6.61 (d, J ) 7.5 Hz, 1H), 6.68 (d, J ) 8 Hz, 1H), 6.77 (dd, J )
8.2 Hz and 2.5 Hz, 2H), 6.88 (m, 4H), 6.99 (s, 1H), 7.14 (t, J )
8 Hz, 1H), 7.20 (t, J ) 8.2 Hz, 2H). 13C-NMR (62.5 MHz,
CDCl3), δ: 45.05, 46.05, 51.65, 52.66, 53.33, 54.15, 55.16, 112.9,
113.67, 114.01, 114.53, 118.61, 120.75, 129.5, 129.74, 139.91,
140.11, 158.22, 159.75. FAB-MS (MNBA matrix): m/ z calcd
for C29H40N4O3, 492; found, 493 (M + H+).
N,N-Dim et h yl-N′-[(3-h yd r oxyp h en yl)m et h yl]et h a n e-
1,2-d ia m in e-2HCl (6‚2HCl). 3-Hydroxybenzaldehyde (1.18
g, 9.66 mmol) and 1,2-diamino-N,N-dimethylethane (10 mmol)
were dissolved in 30 mL of CH2Cl2 containing freshly activated
4 Å molecular sieves and let to stir overnight. Evaporation of
the solvent gave an oily material that was dissolved in a 40-
mL solution of NaBH4 (730 mg, 19 mmol) in ethanol. After
the solution was stirred for 12 h, it was made acidic by adding
concentrated HCl and, subsequently, the solvent was stripped
off. Recrystallization from ethanol gave 1.1 g of pure 6‚2HCl
with mp 188-189 °C. 1H-NMR (250 MHz, CD3OD), δ: 3.02
(s, 6H), 3.35 (s, 4H), 4.26 (s, 2H), 6.91 (m, 1H), 7.03 (m, 2H),
7.33 (t, J ) 8 Hz, 1H). Anal. Calcd for C11H12N2O‚2HCl: C,
49.45; H, 7.54; N, 10.48. Found: C, 49.34; H, 7.62; N, 10.25.
Kin etics. Kinetics were started by addition of 20 µL of a 2
× 10-3 M solution of the substrate in acetonitrile to 2 mL of
buffered solution containing the metal complexes in the
appropriate concentration and followed at 400 nm (pH g 6.5)
or 317 nm (pH < 6.5) by recording the increase of absorbance
due to p-nitrophenolate or p-nitrophenol release. Rate con-
stants were determined by nonlinear fitting of the absorbance
vs time data.38 Reproducibility of independent runs was
within 2.5%.
N-{2-[((3-H yd r oxyp h en yl)m et h yl)a m in o]et h yl}-N,N-
{2,-[((3-m eth oxyp h en yl)m eth yl)a m in o]eth yl}a m in e (5).
TREN (1.46 g, 10 mmol) was dissolved in 150 mL of dry
CH2Cl2, the flask cooled to -79 °C and di-tert-butyldicarbonate
(440 mg, 2 mmol) dissolved in 50 mL of CH2Cl2 was added
dropwise in 30 min. When the addition was complete, the
solution was let to warm to rt and stirred for 12 h more. The
crude material obtained after evaporation of the solvent was
purified by chromatography (SiO2, CHCl3/MeOH/NH3(aq), 10:
4:1). The monoprotected product (Rf 0.33, 430 mg) was an oily
material used without further purification in the subsequent
step. 1H-NMR (250 MHz, CD3OD), δ: 1.48 (s, 9H), 2.56 (m,
6H), 2.73 (t, 4H), 3.16 (t, 2H).
The above compound was dissolved in 60 mL of acetonitrile,
4 Å molecular sieves (10 g), and 3-methoxybenzaldehyde (2
equivalents) were then added and the slurry stirred overnight.
Filtration and evaporation of the solvent gave the imine
derivative in quantitative yield. 1H-NMR (250 MHz, CD3OD),
δ: 1.49 (s, 9H), 2.68 (t, 2H), 3.65 (t, 4H), 3.8 (s, 6H), 5.25 (brm,
1H), 6.85 (d, 2H), 7.2 (d, 2H), 7.25 (t, 2H), 8.23 (s, 2H).
The above imine was added to a solution of NaBH4 (186 mg)
in 50 mL of ethanol and let to react overnight. Water (30 mL)
Ack n ow led gm en t. The authors are indebted to Mr.
E. Castiglione for technical assistance and Dr. A. Fava
for experimental work. Funding was provided by the
Ministry of University, Scientific, and Technological
Research (MURST) and the Italian National Research
Council (CNR)-“Progetto Strategico Tecnologie Chimiche
Innovative”.
Su p p or tin g In for m a tion Ava ila ble: Figures S1 and S2
reporting potentiometric titrations of complexes 1‚Zn(II),
2‚Zn(II), and 5‚Zn(II) and time course of the transacylation
experiment of PNPI with 1‚Zn(II) (2 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O970783F
(38) Leatherbarrow, R. J . Enzfitter; Elsevier: Amsterdam, 1987.