A new class of bioactivable self-immolative N,O-ligands

Article abstract of DOI:10.1016/j.ejmech.2012.03.017

A hexadentate ligand built on an amine-bis(phenol) skeleton with an aminal, self-immolative moiety is presented. Synthesis of the ligand is convenient and relatively high yielded. Moreover, it enables synthesis of many derivatives, both in the amino-phenol and aminal fragment (various heterocycles). Once the final hexadentate ligand is synthesized via the Katritzky reaction, it becomes prone to hydrolysis. Bioactivation by β-galactosidase cleaves the glycosylic bond and a spontaneous collapse of the aminal fragment occurs, thus leading to a pentadentate chelate. This bioactivation has been shown for pyrazole, 1,2,4-triazole and benzotriazole derivatives.

Full text of DOI:10.1016/j.ejmech.2012.03.017

European Journal of Medicinal Chemistry 52 (2012) 184e192
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
A new class of bioactivable self-immolative N,O-ligands
a
,
a
a
a
a
*
ꢀ
ꢀ
Nikodem Kuznik , Arkadiusz Chrobaczynski , Ma1gorzata Mika , Patrycja Miler , Roman Komor ,
Maciej Kubicki b
a Faculty of Chemistry, Silesian University of Technology, B. Krzywoustego 4, 44-100 Gliwice, Poland
Faculty of Chemistry, Adam Mickiewicz University, Grunwaldzka 6, 60-780 Poznan, Poland
b
ꢀ
a r t i c l e i n f o
a b s t r a c t
Article history:
A hexadentate ligand built on an amine-bis(phenol) skeleton with an aminal, self-immolative moiety is
presented. Synthesis of the ligand is convenient and relatively high yielded. Moreover, it enables
synthesis of many derivatives, both in the amino-phenol and aminal fragment (various heterocycles).
Once the final hexadentate ligand is synthesized via the Katritzky reaction, it becomes prone to
hydrolysis. Bioactivation by b-galactosidase cleaves the glycosylic bond and a spontaneous collapse of the
aminal fragment occurs, thus leading to a pentadentate chelate. This bioactivation has been shown for
pyrazole, 1,2,4-triazole and benzotriazole derivatives.
Received 22 January 2012
Received in revised form
3 March 2012
Accepted 6 March 2012
Available online 17 March 2012
Keywords:
Amino-phenol ligand
N,O-ligand
Ó 2012 Elsevier Masson SAS. All rights reserved.
Bioactivation
b-Galactosidase
Self-immolative
1. Introduction
system. Once activated by an enzyme, the system can be monitored
by appropriate instrumentation e Scheme 1.
Classic amino-phenolate ligands, with the aminobis(phenolate)
root, have a well-established chemistry [1,2] for a broad range of
metals [3e5]. They are used because of their convenience in
synthesis, with huge possible derivatives expressing high affinity to
metal chelation, which results in good product constants. Amino-
phenolates are a key fragment of the protein inhibitor [6e8] and
mimic the catechol oxygenase active site [9,10]. Conversely, great
interest has been focused on bioactivable species, which have
potential application in a system where a measurable physical or
chemical stimulus appears [11e16]. The introduction of a pendant,
bioactivable arm is one approach that has recently been applied for
smart contrast agents [17e22]. Bioactivation by enzymatic cleavage
of a non-directly coordinating fragment is followed by the self-
immolative release of a coordination site. This results in an
observable change in the relaxivity, so that enzyme activity can be
monitored non-invasively [23e25]. Here we present a new class of
bioactivable, self-immolative amine-phenol hexadentate ligand
dedicated to d-block metals. The idea was to design a delicate
system which, because of its nature, would be difficult to separate
out from the reaction mixture. Yet it could be complexed and, when
exhibiting higher stability, purified and applied to the target
In this paper we present ligand synthesis and the results of
bioactivation experiments.
2. Results and discussion
2.1. Synthesis of the pentadentate ligand 3
Synthesis of the target ligands 7ae7f was multistep (Scheme
2.). However, the main purpose was the high selectivity of each
step since separation of the potential products (i.e. mono- and di-
N-alkylation) could be very problematic. Initially, we attempted to
build the 2-hydroxybenzyl fragment by a reaction of salicylamide
reduction with LiAlH4, as reported by Parker [26]. Unfortunately,
the yield was very low and the desired product was difficult to
separate from the reaction mixture. Therefore, we abandoned this
route. Siegl et al. described the synthesis of compound 1 in
a Mannich-type reaction from phenol, HCHO and 2-aminoethanol
[27]; however, our attempt to synthesize compound 1 by this
method failed. Thus, the amine skeleton 1e3 was synthesized as
follows: N,N-bis(2-hydroxyphenylmethyl)-2-aminoethanol (1)
was obtained in a two-step one-pot reaction of the reductive
amination of 2-aminoethanol with salicylaldehyde. The reaction
conditions were similar to those described by Hinshaw [28].
NaBH3(CN) was applied as a reductive agent because of its higher
* Corresponding author. Tel.: þ48 32 2371839; fax: þ48 32 2372094.
ꢀ
E-mail address: Nikodem.Kuznik@polsl.pl (N. Kuznik).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2012.03.017

ꢀ
N. Kuznik et al. / European Journal of Medicinal Chemistry 52 (2012) 184e192
185
Scheme 1. Bioactivable, self-immolative ligand synthesis, complexation and activation e a concept.
stability than regular NaBH4, especially in an acidic environment
[29,30]. Since the synthesis is a multistep condensation followed
by reduction, relatively low yields of 24e34%, depending on the
scale, are quite satisfactory. No monobenzyl derivative was
identified in the post-reaction mixture. The purified product, 1,
was subjected to SOCl2 treatment, thus leading to chlorinated
product 2. It was crucial to use freshly distilled thionyl chloride;
otherwise, the reaction would fail. During the reaction a waxy,
Scheme 2. Synthesis of ligands. Coordination sites are bolded. Reagents and conditions: (a) MeOH, 1.5 h, HCl/MeOH, NaBH3(CN), 72 h, then the next batch of salicylaldehyde,
NaBH3(CN) 72 h, 1 34%; (b) CH2Cl2, SOCl2. 24 h, rt 10%, Na2CO3 2 90%; (c) EtOH, reflux 2 h, 3 77%; (d) Ac2O, HBr/AcOH, 4 94%; (e) MeCN, Ag2O, 4e4.5 h rt, 5a 89%, 5b 73%; (f) benzene,
reflux in DeaneStark app. 8e70 h e Table 4; (g) MeONa/MeOH, overnight, rt, CO2.

ꢀ
N. Kuznik et al. / European Journal of Medicinal Chemistry 52 (2012) 184e192
186
Table 1
Products of N-alkylation of 2-amine pyridine with product 2.
Molar ratio of 2 to
2-(aminomethyl)pyridine
1:2
1:8
70
30
8
92a
a
Actual yield of purified product was 77%.
separate phase formed. It was assigned to the N- and O-sulfonated
polar species. After reaction completion, the volatiles were
evaporated and the residue was treated with Na2CO3 aqueous
solution. This hydrolyzed the above-mentioned sulfonated inter-
mediates. Consequent workup gave product 2, which was pure
enough for further synthesis. It is interesting to note that the
similar bis(N-benzyl)-2-chloroethanamine is unstable and
undergoes intermolecular N-alkylation, which leads to an azir-
idine derivative [31]. It seems that the presence of the hydroxyl
groups stabilizes the compound. The structure of compound 2 has
been proved by X-ray of the monocrystal. We tried bromination of
compound 1 with SOBr2 according to Nagle’s work [32]; however,
we did not succeed in obtaining the expected product. Never-
theless, it turned out that chlorine was a good enough leaving
group for further synthesis: the chlorinated product 2, was used
for N-alkylation of 2-(aminomethyl)pyridine, thus leading to the
pentadentate amino-phenolic ligand 3. The reaction was
successful in boiling ethanol according to Watanabe’s work [33].
The ratio of nucleophile 2 to 2-(aminomethyl)pyridine was opti-
mized: when the ratio was 1:2, 70% of the desired mono-
substituted product and 30% of the double-substituted product
was obtained, but when the ratio was 1:8 the shares were 92e8%,
respectively (Table 1).
was successful in both cases, i.e. with and without the NO2 group on
a benzene ring.
2.4. Introduction of a bioactivable pendant fragment to the ligand e
condensation of the pentadentate ligand with O-phenylglycoside
Substitution of the secondary amine (reaction f, Scheme 2) was
considered based on Katritzky’s work [36]. Such products as 6,
formally called aminals, are very fragile and prone to hydrolysis,
thus we investigated the possibility of synthesis on a model reac-
tion e Scheme 3.
The model reaction was repeated twice on two different scales,
i.e. 10 and 2 mmol (Table 2). The smaller scale should mimic
the target conditions for the reaction of ligand 3. The reaction
was carried out in benzene in order to remove the forming water
and to shift the equilibrium to aminal 7. Water was collected in
a DeaneStark trap. When water separation was observed for the
larger scale, it was too small to be noticed on the smaller scale. It
was not possible to monitor the reaction course on a TLC silica plate
nor on an alumina plate. The product decomposed immediately on
the plates. Thus, the conversion was estimated by a rough 1H NMR
experiment e Table 2. On NMR spectra the rising signal of an
2.2. Crystal structure of compound 2
Compound 2 forms a stable, deep-brown solid which comes
directly from treatment of compound 1 with SOCl2 without any
purification. A monocrystal recrystallized from the chloroform/
heptane mixture was analyzed. It forms a triclinic unit cell with two
symmetry-related molecules. There is an intramolecular hydrogen
ꢁ
bond OeH/N from O1B to N1 (1.84 A) and there are two inter-
molecular hydrogen bonds OeH/O from O1A of one molecule to
ꢁ
O1B of the other molecule (1.98 A). There is also another pair of
oxygen atoms. The hydrogen bonds most likely stabilize the
molecule and decrease the nucleophilicity of the amine. Thus, an
intramolecular reaction of the nitrogen atom to aziridinium cation
formation is not observed in this case (Fig. 1).
2.3. Synthesis of O-phenylglycoside 4
Protected
b-D-galactose was treated with HBr in acetic acid
solution; the reaction gave crude product 4 according to Stortz’s
work [34]. Then it was directly applied to the reaction (e) with 4-
hydroxybenzaldehyde. Our first attempts under PTC conditions
[35] failed, so we tried the classic method with Ag2O [17], which
Fig. 1. X-ray structure of compound 2 with 60% probability ellipsoids.

ꢀ
N. Kuznik et al. / European Journal of Medicinal Chemistry 52 (2012) 184e192
187
also a C4eH pyrazole signal in close proximity; it was identified by
the addition of free pyrazole to the NMR sample.
During the reaction course the disappearance of the aldehyde
(CHO) and azole signals was observed with a simultaneous growth
of the methine peak. Thus, conversion was calculated based on
these three unique and well-separated signals e Table 4. In order to
eliminate deviation from the equimolar ratio due to errors in
weighing, a reference was recorded at a starting point before each
reaction. We tried to identify the forming aminals by MS, however,
we were never able to obtain the molecular peak, even at the
mildest ESI ionization. Nevertheless, fragmentation signals were
found, which proved formation of the desired aminal. Common
high mass fragmentation peaks were found: M ꢀ 92 (benzyl),
M ꢀ 106 (2-oxobenzyl), M ꢀ 176 (176: a fragment of protected
galactose); M ꢀ 226 (226: bis(2-hydroxybenzyl)amine).
Scheme 3. A model Katritzky condensation.
Table 2
Model Katritzky condensation results.
Scale [mmol]
Reaction time [h]
Conversion
10
2
9
14.5
91%
89%
Similarly as with the model reaction, here also some optimal
conversion is achieved at the 35e76% level (Table 4). A longer
reaction time leads to decomposition, which lowers the overall
yield. The reaction scale plays an important role and depends on
the accessibility of secondary amine 3 and the expected amount of
the expected complex. In the case of triazoles 6ce6f, the presence
of a nitro group (e.g. in 5b) resulted in higher conversion to ami-
nals, for pyrazole this was the reverse. Conversely, an azole-type
effect on the conversion is in agreement with the acidity trend of
the azoles. Pyrazole, as the least acidic (pKa ¼ 19.8), resulted in the
best conversion, while benzotriazole, as a relatively strong NH-
acid (pKa ¼ 8.4), gave the poorest conversion. Although not
stated directly by Katritzky, it seems that the reverse reaction of
hydrolysis is responsible for the low yield in some cases. This
delicate equilibrium of aminal formation and its hydrolysis is
decisive in the possibility of activable ligand formation and its
successful activation, i.e. controlled collapse by pendant fragment
cleavage.
aminal (methine CeH) was compared to the decreasing signals of
the aldehyde’s formyl group and to the pyrazole signal near the
aromatic region.
Although not statistically important, a similar conversion was
achieved with prolongation of the reaction time by another 5.5 h on
the smaller scale. In Table 2 the maximal conversion is given,
however, it was noticed that after multiple optimal reaction times
(up to 26 h) the conversion decreased. Nevertheless, these results
were promising for the target compounds. Therefore, condensation
was carried out for the secondary amine 3, benzaldehyde with
a galactose substituent (5ae5b) and an azole e step (f) in Scheme 2.
Conversion was measured based on NMR spectra. A new signal
near 6 ppm was observed as a singlet, or as a doublet in the case of
triazoles due to the possible isomers. The signals were assigned to
the CeH methine group of the forming aminal (Table 3). There was
Table 3
NMR identification of new aminals 6ae6f.
Azole (Y)
Entry
X
Methine CeH
Chemical shifts [ppm], in CDCl3 of the remaining groups
position [ppm]
1.98e2.22 (m, 12H, galactose e AcO), 2.70 (t, 4H, J ¼ 5.1 Hz, NCH2CH2N),
3.72e3.98 (m, 4H, AreCH2NCH2eAr), 3.98e4.27 (m, 2H, galactose C5-H, C6eH),
4.95e5.27 (m, 2H, galactose C2eH, C3eH), 5.42e5.58 (m, 2H, galactose C1eH,
C4eH), 7.11e7.59 (m, 19H, Ar).
1.95, 2.02, 2.07, 2.12 (s, 12H, galactose e OAc), 2.65e2.78 (m, 4H, NCH2CH2N),
3.20e3.82 (m, 6H, AreCH2NCH2eAr, NCH2epyridyl), 3.85e4.27 (m, 2H,
galactose C5eH, C6eH), 4.90e5.19 (m, 2H, galactose C2eH, C3eH), 5.37e5.56
(m, 2H, galactose C1eH, C4eH), 6.29 (t, 1H, J ¼ 1.8 Hz, pyrazole C4eH),
6.61e7.80 (m, 16H, Ar), 8.24 (d, 1H, J ¼ 2.1 Hz, pyridyl C6eH).
6a
6b
eH
6.23
6.17
-NO2
2.03, 2.07, 2.19 (s, 12H, galactose e OAc), 2.60e3.08 (m, 4H, NCH2CH2N),
3.28e4.03 (m, 6H, AreCH2NCH2eAr, NCH2epyridyl), 4.10e4.31 (m, 2H,
galactose C5eH, C6eH), 4.98e5.11 (m, 2H, galactose C2eH, C3eH), 5.43e5.59
(m, 2H, galactose C1eH, C4eH), 6.60e7.92 (m, 20H, Ar).
2.02, 2.08, 2.15, 2.17 (s, 12H, galactose e OAc), 2.68e2.91 (m, 4H, NCH2CH2N),
3.31e4.03 (m, 6H, AreCH2NCH2eAr, NCH2epyridyl), 4.05e4.29 (m, 2H,
galactose C5eH, C6eH), 5.01e5.25 (m, 2H, galactose C2eH, C3eH), 5.42e5.63
(m, 2H, galactose C1eH, C4eH), 6.63e8.10 (m, 20H, Ar).
6c
eH
5.84, 5.88
5.86, 5.89
6d
-NO2
1.98e2.24 (m, 12H, galactose e OAc), 2.62e2.98 (m, 4H, NCH2CH2N), 3.31e3.98
(m, 6H, AreCH2NCH2eAr, NCH2epyridyl), 3.98e4.29 (m, 2H, galactose C5eH,
C6eH), 4.97e5.21 (m, 2H, galactose C2eH, C3eH), 5.44e5.57 (m, 2H, galactose
C1eH, C4eH), 6.68e8.62 (m, 18H, Ar).
1.94; 2.01; 2.06; 2.11 (s, 12H, galactose e OAc), 2.65e2.91 (m, 4H, NCH2CH2N),
3.26e3.95 (m, 6H, AreCH2NCH2eAr, NCH2epyridyl), 3.98e4.27 (m, 2H,
galactose C5eH, C6eH), 4.93e5.29 (m, 2H, galactose C2eH, C3eH), 5.39e5.65
(m, 2H, galactose C1eH, C4eH), 6.60e8.59 (m, 18H, Ar).
6e
6f
eH
5.78, 5.80
5.69, 5.70
eNO2

ꢀ
N. Kuznik et al. / European Journal of Medicinal Chemistry 52 (2012) 184e192
188
Table 4
The Katritzky reaction between secondary amine 3, galactosylbenzaldehyde 5 and
azole resulting in aminals 6ae6f.
Azole Y
Entry, X
Scale [mmol]
Time [h]
Conversion
8
15.5
4
10
20
26
70
57%
76%
28%
31%
37%
40%
63%
6a, eH
0.42
0.12
6b, eNO2
0.44
17.5
33
11
19%
35%
41%
57%
6c, eH
0.41
0.28
6d, eNO2
26.5
6
32
19
28
0%
38%
47%
52%
6e, eH
0.28
0.28
6f, eNO2
2.5. Deprotection and bioactivation e deliberation of the pendant
fragment to the pentadentate ligand 3
Scheme 5. Mechanism of an enzymatically induced self-immolative reaction.
The final step of ligand synthesis was deprotection of the
galactose fragment in the classic method with sodium meth-
anolate. The crude products 7ae7f, obtained as part of the post-
reaction mixture, were subjected to enzymatic hydrolysis of the
glycosyl bond with a following hydrolysis of the aminal to delib-
erate pentadentate ligand 3 e Schemes 4 and 5.
In Charts 1e3 the producteazole increment was shown as
a function of the reaction time. In Table 5 a comparison of enzy-
matic hydrolysis followed by a self-immolative collapse is pre-
sented. The reaction rate (r) was calculated simply as a difference
of the azole concentration in the measure time interval. The
empirical equation y(s) was derived as a derivative of the trend line
The reaction was followed by the HPLC/UVevis analysis, in
which the growth of the final products, i.e. 4-hydroxybenzaldehyde
(or its nitro derivative) and the azole concentration, was analyzed.
The enzymatic reaction was preceded by an enzyme activity test
and the calibration curve for the analyzed species. An additional
complication was the presence of an expected product (azole and
benzaldehyde) in the post-reaction mixture since step “f” in
Scheme 2 was not completed and products 6 and 7 were not
purified due to their high instability. Nevertheless, product growth
was referenced to the starting point of the reaction.
e it has no physical meaning and is given solely for comparison
purposes.
When the reaction rates are compared it is clear that two
factors affect the kinetics: the type of azole (Y) and the presence
of the nitro group in the benzaldehyde moiety (X). Hydrolysis of
the benzotriazole derivative is the fastest, whereas hydrolysis of
the pyrazole derivative is the slowest. This trend followed the
relative acidity relation of the free azoles e benzotriazole is the
strongest NH-acid (pKa ¼ 8.4), whereas pyrazole is the weakest
acid (pKa ¼ 19.8). Two azoles, benzotriazole and the 1,2,4-triazole
Scheme 4. Enzymatic hydrolysis of 7ae7f to 3. Coordination sites are bolded.

0,6
0,5
0,4
0,3
0,2
0,1
0
nitro group, had a positive effect on kinetics; for pyrazole this was
the reverse, as was observed for the Katritzky synthesis of 6ae6f
(Table 4). A more valid explanation should be based on the
nucleophilicity of deprotonated azoles. However, a simple rela-
tion between nucleophilicity and acidity is not described in the
literature. Thus, we assume that the nucleophilicity trend follows
the basicity of the azole anions. Since the nitro group is respon-
sible for at least two contrary effects during hydrolysis, it seems
that the dominance of the effects differs among the azoles: (A)
the nitro group stabilizes the phenolate anion, thus resulting in
slower hydrolysis e and this may dominate in the case of pyr-
azole, while (B) the nitro group also increases electron deficiency
on the methine carbon atom, which facilitates the water attack on
that center. The latter effect may be valid for triazoles. The model
could be developed in other heterocyclic systems instead of
azoles. Also, other substituents (X) in the benzene ring could be
considered.
7a, X=H
7b, X=NO2
0
50
100
150
200
Reaction time [min]
Chart 1. Increase of pyrazole concentration on the self-immolative collapse.
18
16
14
12
10
8
3. Conclusions
The series of bioactivable, self-immolative N,O-ligands may
serve for chelation of d-block metals. We are currently investigating
their coordination with selected first row d-block paramagnetic
ions. From our initial experiments we can state that they are stable
enough to be purified and analyzed. Based on simple modeling, we
can assume that the ligands adopt a distorted octahedral geometry.
Extending the 2-carbon alkyl chain to a 3- or 4-carbon spacer
would probably enhance the stability of the complexes. As ligands
they are synthesized on a convenient and selective route. Once
activated, the ligand becomes pentadentate and thus changes the
inner coordination sphere of the potential complex. A new gap-free
coordination site appears which can be filled with a solvent
molecule (water?) or any other ligand. This could potentially be
monitored by NMR/MRI, IR, UV or other techniques. In biological
systems a free coordination site at a redox-active center may result
in catalytic processes. However, since our method enables the
synthesis of various derivatives, we hope that an appropriate
design would be able to control the potential redox activity of the
metal center.
6
4
7c, X=H
2
7d, X=NO2
150
0
0
50
100
Reaction time [min]
Chart 2. Increase of benzotriazole concentration on the self-immolative collapse.
3
2,5
2
4. Experimental
1,5
1
4.1. General
All syntheses were carried out in an inert atmosphere of
nitrogen. All liquid chemicals and solvents were dried and
7e, X=H
0,5
7f, X=NO2
distilled prior to use.
A sodium methanolate solution was
0
prepared each time directly before deprotection by introducing
a piece of sodium (ca. 1 g) into dry methanol (20 ml). NMR spectra
were taken on a Varian Unity Nova 300 MHz spectrophotometer
using tetramethylsilane as an internal reference at room
temperature. Electrospray-ionization mass spectrometry was
performed on a 4000 QTrap (Applied Biosystems/MDS Sciex)
mass spectrometer. High resolution mass spectra were registered
on a Micromass/Waters LCT (TOF) spectrometer at the University
of Warsaw.
0
50
100
Reaction time [min]
150
Chart 3. Increase of 1,2,4-triazole concentration on the self-immolative collapse.
Table 5
Comparison of enzyme-initiated self-immolative conversion of hexadentate ligand 7
into pentadentate ligand 3.
HPLC analysis was carried out on a Dionex-HPLC UltiMate 3000
7
Substituents
Azole (Y)
Pyrazole
y
(
s
)
r
[mmol dmꢀ3 minꢀ1
]
[mmol dmꢀ3 minꢀ1
]
with Thermo Scientific Hypersil GOLD HPLC Column 5
size, mobile phase methanol/water 1:4. UV measurements were
taken at ¼ 327 nm for
¼ 225 nm for amine (3) and azole and at
benzaldehyde or its m-nitro derivative.
-Galactosidase from Escherichia coli was purchased from
SigmaeAldrich. Stock solution was prepared by dissolving 1.32 l of
commercial 1 kU solution in 1 ml of phosphate buffer resulting in
mm particle
X
7a eH
7b eNO2 Pyrazole
7c eH
7d eNO2 Benzotriazole
7e eH
n
n
¼ ꢀ2 ꢁ 10ꢀ5s þ 0.0052 2.8 ꢁ 10ꢀ3
l
l
¼ ꢀ2 ꢁ 10ꢀ5s þ 0.0018 0.8 ꢁ 10ꢀ3
Benzotriazole
n
n
¼ ꢀ1.4 ꢁ 10ꢀ4s þ 0.1339 0.4 ꢁ 10ꢀ1
¼ ꢀ3.4 ꢁ 10ꢀ3s þ 0.3284 1.1 ꢁ 10ꢀ1
b
1,2,4-Triazole
n
¼ ꢀ2 ꢁ 10ꢀ5s þ 0.0148 1.4 ꢁ 10ꢀ2
¼ 14 ꢁ 10ꢀ5s þ 0.0013 2.1 ꢁ 10ꢀ2
m
7f eNO2 1,2,4-Triazole
n

ꢀ
N. Kuznik et al. / European Journal of Medicinal Chemistry 52 (2012) 184e192
190
a 2 U solution. Enzyme activity was determined according to Kuby’s
4.4. Crystal data for compound 2
work [37] by measuring absorbance at 410 nm of the enzymatic
cleavage of o-nitrophenyl-
b
-
D
-galactopyranoside.
The crystal chosen for X-ray analysis was a clear colorless block
with the approximate dimensions of 0.2 ꢁ 0.1 ꢁ0.1 mm.
C16H18ClNO2 (291.7 g molꢀ1) crystallizes in the triclinic system,
X-ray single crystal measurement: measurements of the
diffraction intensities were performed on a Agilent Technologies
SuperNova diffractometer with mirror-monochromatized, CuKa
space
group
P-1,
a
with
¼ 81.235(30),
(Cu
a ¼ 8.1149(16),
b ¼ 8.6333(17),
¼ 73.425(30);
2.259 mmꢀ1
and
range 3.96e73.86ꢂ; temperature of
c ¼ 11.3099(23) A,
b
¼ 85.662(30),
g
ꢁ
3
radiation,
u
/2Q
scan mode,
Q
ꢁ
the measured crystals: 100 K. Crystallographic data for 2 were
deposited with the Cambridge Crystallographic Data Centre as
supplementary publication number: CCDC 858807. A complete
listing of the atomic coordinates of x, y, and z can be obtained free of
charge, on request, from the Director, Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK [fax: þ44 1223
336 033; e-mail: deposit@ccdc.cam.ac.uk], upon quoting the
depository numbers, names of the authors and the journal citation.
Silver(I) oxide was freshly prepared according to the classic
method by precipitation from AgNO3 with NaOH, decantation of
the supernatant followed by washing of the precipitate. After
filtration it was dried in vacuo and kept for several days in
a desiccator.
V ¼ 750.13(3) A ,
Z ¼ 2,
m
Ka
)¼
,
Dcalcd ¼ 1.292 cmꢀ3. A total of 14,702 reflections were collected to
2
Qmax ¼ 73.86ꢂ (h: ꢀ10 / 10, k: ꢀ10 / 10, l: ꢀ14 /13), of which
2960 were unique. In refinements, weights were used according to
the scheme w ¼ 1/[
s
2(Fo2) þ (0.08450P)2 þ 0.10180P], where
P ¼ (Fo2 þ 2F2c)/3. The refinement of 253 parameters (data-to-
parameter ratio being 58.1) converged to the final agreement
factors R ¼ 0.0423 for 2815 reflections with Fo > 4
Rw ¼ 0.1340, and S ¼ 1.060 for all observed reflections. The electron
ꢁꢀ3
s(Fo) and
density of the largest difference peak was found to be 0.28 e A
ꢁꢀ3
,
while that of the largest difference hole was 0.39 e A
.
4.5. N,N-bis(2-hydroxyphenylmethyl)-N0-(2-pyridylmethyl)ethan-
1,2-diamine (3)
4.2. N,N-bis(2-hydroxyphenylmethyl)-2-aminoethanol (1)
Ethanol (15 ml), N,N-bis(2-hydroxyphenylmethyl)-2-chloroeth-
anamine (2) (200 mg, 0.68 mmol) and 2-(aminomethyl)pyridine
(0.56 ml, 5.4 mmol) were introduced into a round-bottomed flask.
The mixture was refluxed. Reaction progress was monitored with
TLC (chloroform/methanol 9:1). Usually, the substrate was
consumed within 2 h. After completion all of the volatiles were
removed in vacuo. Then a sodium carbonate solution (10 ml, 10%)
was added to the residue and extracted with chloroform
(2 ꢁ10 ml). The combined extracts were dried with anhydrous
MgSO4 and the solvent was evaporated in vacuo. The crude orange
oily product (325 g) was purified by flash chromatography. Elution
Methanol (15 ml) and 2-aminoethanol (0.6 ml, 10 mmol) were
introduced into a round-bottomed flask. 2-hydroxybenzaldehyde
(1 ml, 10 mmol) was added while stirring. The reaction mixture
turned yellow after 15 min. After 1.5 h hydrogen chloride (0.5 M) in
methanol was added to pH ¼ 5. Then sodium cyanoborohydride
(1 g,16 mmol) was added. After 72 h of stirring at rt the next portion
of 2-hydroxybenzaldehyde (1 ml, 10 mmol) was added followed by
sodium cyanoborohydride (1 g, 16 mmol). The reaction mixture was
light brown. After another 72 h all of the volatiles were evaporated
in vacuo. Water (30 ml) was added to the residue and was extracted
three times with chloroform (15 ml). The combined extracts were
dried with anhydrous MgSO4 and the solvent was evaporated in
vacuo. The crude brown oily product (2.094 g) was purified by
flash chromatography. Elution with a chloroform/methanol (9:1 v/v)
with
a chloroform/methanol (9:1 v/v) mixture gave clean
N,N-bis(2-hydroxyphenylmethyl)-N0-(2-pyridylmethyl)ethan-1,2-
diamine 3 (206 mg, 77% yield) as a dark orange oil. 1H NMR (CDCl3)
d
: 8.53 (d, 1H, J ¼ 4.8 Hz, py), 7.64 (td, 1H, J ¼ 7.8 Hz, 1.8 Hz, py), 7.30
mixture
aminoethanol 1 (880 mg, 34% yield) as a pale brown solid. 1H
NMR (CDCl3) : 7.17e7.03 (m, 4H, Ar), 6.85e6.76 (m, 4H, Ar), 3.86
gave
pure
N,N-bis(2-hydroxyphenylmethyl)-2-
(d, 1H, J ¼ 7.8 Hz, py), 7.20e7.10 (m, 3H, py), 7.02 (dd, 2H, J ¼ 7.5 Hz,
1.2 Hz, Ar), 6.85 (d, 2H, J ¼ 8.1 Hz, Ar), 6.76 (td, 2H, J ¼ 7.5 Hz, 1.2 Hz,
Ar), 3.95 (s, 2H, pyeCH2eN), 3.69 (s, 4H, NeCH2eC6H4), 2.90
(t, 2H, J ¼ 6.0 Hz, eNeCH2eCH2eNHe), 2.66 (t, 2H, J ¼ 6.0 Hz,
d
(t, 2H, J ¼ 5.1 Hz, NeCH2eCH2eOH), 3.79 (s, 4H, eNeCH2eAr), 2.73
(t, 2H,J ¼ 5.1 Hz, eNeCH2eCH2eOH). 13C NMR (CDCl3)
d
: 156.1,
eNeCH2eCH2eNHe). 13C NMR (CDCl3)
d: 157.9, 157.0, 149.4, 136.9,
130.4, 129.5, 122.5, 119.9, 116.3, 60.2, 56.4, 54.5. MS (ESI):
M þ 1 ¼ 274, M ꢀ 1 ¼ 272. HRMS (M þ Naþ) calculated for
C16H19NO3Na: 296.1263; found: 296.1265.
130.2, 129.4, 123.1, 122.5, 122.3, 119.2, 116.9, 56.0, 54.0, 50.8, 46.2.
MS (ESI): M þ 1 ¼364, M ꢀ 1 ¼362. HRMS calculated for
C22H26N3O2: 364.2038; found: 364.2025.
4.3. N,N-bis(2-hydroxyphenylmethyl)-2-chloroethanamine (2)
4.6. 2,3,4,6-Tetra-O-acetyl-1-bromo-
a
-
D
-galactose (4)
(4.0 g,
Dichloromethane (15 ml) and N,N-bis(2-hydroxyphenylmethyl)-
2-aminoethanol (1) (623 mg, 2.3 mmol) were introduced into
a round-bottomed flask. Freshly distilled thionyl chloride (0.8 ml,
11 mmol) was added dropwise while stirring. Dark tar separated out
from the reaction mixture. After 24 h of stirring at room temperature
all of the volatiles were removed in vacuo. After addition of sodium
carbonate solution (10 ml, 10%), the mixture was stirred for 1 h,
then extracted with dichloromethane (3 ꢁ 15 ml). The combined
extracts were dried with anhydrous MgSO4 and the solvent was
evaporated in vacuo. Crude N,N-bis(2-hydroxyphenylmethyl)-2-
chloroethanamine 2 (600 mg, 90%) was obtained of a satisfactory
1,2,3,4,6-Penta-O-acetyl- -galactose
b
-
D
10.3 mmol)
together with 33% HBr in acetic acid (11.0 ml, 58.4 mmol) were
introduced into a round-bottomed flask. The reaction mixture
was stirred and acetic anhydride (2.0 ml, 21.39 mmol) was added
after the solid had dissolved. While stirring at room temperature,
progress was monitored with TLC (MeOH/CH2Cl2 5:95) and the
reaction was completed after 2 h. Afterward, dichloromethane
(80 ml) was added followed by ice (60 g) while stirring. The
organic layer was separated and the aqueous layer was extracted
with dichloromethane (2 ꢁ 40 ml). The combined organic extracts
were washed with saturated Na2CO3 solution (40 ml), then with
water (40 ml). Following drying with anhydrous MgSO4, the
solvent was evaporated. The oily residue was dissolved in diethyl
ether and precipitated with cyclohexane. A white precipitate was
filtered and dried in vacuo yielding 3.7 g (94%) of crude product 4
purity. 1H NMR (CDCl3)
d: 7.18e7.06 (m, 4H, Ar), 6.87e6.77
(m, 4H, Ar), 3.83 (s, 4H, eNeCH2eAr), 3.63 (t, 2H,J ¼ 7.2 Hz,
eNeCH2eCH2eCl), 2.94 (t, 2H, J ¼ 7.2 Hz, eNeCH2eCH2eCl). 13C
NMR (CDCl3) d: 156.0,130.4,129.4,122.3,119.9,116.0, 56.2, 55.0, 40.6.
MS (ESI): M þ 1 ¼ 292, M ꢀ 1 ¼ 290. HRMS calculated for
applied immediately to the next reaction e synthesis of
compound 5.
C16H19NO2Cl: 292.1111; found: 292.1104.

ꢀ
N. Kuznik et al. / European Journal of Medicinal Chemistry 52 (2012) 184e192
191
4.7. 4-[1-
b
-
D
-(2,3,4,6-Tetra-O-acetylgalactosyl)]benzaldehyde (5a)
-galactose (6.8 g, 16.6 mmol)
4.10. General procedure of hexadentate ligand 6ae6f synthesis
2,3,4,6-Tetra-O-acetyl-1-bromo-
a-D
An azole (Y) (0.42 mmol) was dissolved in benzene (30 ml)
was dissolved in freshly distilled acetonitrile (170 ml) in a round-
bottomed flask and magnetic stirring was applied. Then silver(I)
oxide (17.2 g, 72.2 mmol) and 4-hydroxybenzaldehyde (1 g,
17.2 mmol) were added. The mixture turned deep brown. Stirring at
room temperature was continued for 4 h. Reaction progress was
monitored with TLC (hexane/diethyl ether 2:8). After reaction
completion the volatiles were evaporated and the brown residue
was dissolved in ethyl acetate (50 ml) and filtered through a thin
Celite layer. Then the Celite was washed twice with ethyl acetate. The
combined organic solutions were evaporated on a rotary evaporator.
The crude product (6.8 g) was obtained as a yellow oil. It was purified
on a silica gel column (hexane/diethyl ether 2:8). A total of 4.5 g of 4-
into a round-bottom flask together with 4-[1-
acetylgalactosyl)]benzaldehyde (5a) or 4-[1-
b
b
-
D
-(2,3,4,6-tetra-O-
-(2,3,4,6-tetra-O-
-D
acetylgalactosyl)]-3-nitrobenzaldehyde (5b) (0.42 mmol) and
N,N-bis(2-hydroxyphenylmethyl)-N0-(2-pyridylmethyl)ethan-1,2-
diamine (3) (0.42 mmol). The reaction mixture was refluxed under
a DeaneStark head for a given period of time (Table 4.). After
removal of the solvent in vacuo, an oily residue was left containing
the desired product with conversion as given in Table 4. NMR
characteristics are given in Table 3.
4.11. Deprotection of compounds 6ae6f to 7ae7f
[1-
b
-
D
-(2,3,4,6-tetra-O-acetylgalactosyl)]benzaldehyde (5a) in an
The entire post-reaction mixture containing compounds 6ae6f
was dissolved in methanol with sodium methanolate and stirred
overnight at room temperature. Sodium methanolate was
quenched by CO2 purging. After evaporation of the volatile fractions
a solid residue of the crude product was obtained.
89% yield was obtained. 1H NMR (CDCl3)
d: 9.94 (s, 1H, CHO), 7.86 (d,
2H, J ¼ 8.8 Hz, Ar), 7.12 (d, 2H, J ¼ 8.8 Hz, Ar), 5.47e5.56 (m, 2H,
galactose C1eH, C4eH), 5.11e5.19 (m, 2H, galactose C2eH, C3eH),
4.09e4.28 (m, 3H, galactose C5eH, C6eH2), 2.03, 2.08, 2.20 (s, 12H,
galactose e AcO).
4.12. Enzymatic hydrolysis of pendant fragment of ligands 7ae7f
The following reagents were introduced into a conical flask: crude
product 7ae7f (0.1 mmol), phosphate buffer (pH ¼ 7.2; 13.4 ml),
magnesium chloride solution (0.5 ml, 30 mM), 2-mercaptoethanol
4.8. 4-[1-b-D-(2,3,4,6-Tetra-O-acetylgalactosyl)]-3-
nitrobenzaldehyde (5b)
(0.5 ml, 3.36 mM), and b-galactosidase solution (0.1 ml, 0.2 U). The
2,3,4,6-Tetra-O-acetyl-1-bromo-a-D-galactose (2.6 g, 6 mmol)
reaction mixture was stirred at 37 ꢂC for a total time of 3 h. During
was dissolved in freshly distilled acetonitrile (60 ml) in a round-
bottomed flask and magnetic stirring was applied. Then silver(I)
oxide (6 g, 25 mmol) and 4-hydroxy-3-nitrobenzaldehyde (1.0 g,
6 mmol) were added. The mixture turned deep brown. Stirring at
room temperature was continued for 4.5 h. Reaction progress
was monitored with TLC (hexane/diethyl ether 2:8). After reac-
tion completion the volatiles were evaporated and the brown
residue was dissolved in ethyl acetate (50 ml) and filtered
through a thin Celite layer. Then the Celite was washed twice
with ethyl acetate. The combined organic solutions were evap-
this period aliquots were taken for HPLC/UVevis analysis.
Acknowledgements
This work has been supported by grant no. N N204 030935 of
the Polish Ministry of Science and Higher Education.
Appendix A. Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.ejmech.2012.03.017. These data
include MOL files and InChiKeys of the most important compounds
described in this article.
orated on a rotary evaporator. A total of 2.2 g of 4-[1-b-D-(2,3,4,6-
tetra-O-acetylgalactosyl)]-3-nitrobenzaldehyde (5b) with a 73%
yield was obtained as a crude, pale yellow solid which was pure
enough for further application. 1H NMR (CDCl3)
d: 9.98 (s, 1H,
CHO), 8.31 (d, 1H, J ¼ 1.8 Hz, Ar), 8.08 (dd, 1H, J ¼ 8.4 Hz, 1.8 Hz,
Ar), 7.49 (d, 1H, J ¼ 8.8 Hz, Ar), 5.48e5.66 (m, 2H, galactose
C1eH, C4eH), 5.08e5.30 (m, 2H, galactose C2eH, C3eH),
4.08e4.33 (m, 3H, galactose C5eH, C6eH2), 2.03, 2.09, 2.14,
2.20 (s, 12H, galactose e AcO).
References
[1] O. Wichmann, R. Sillanpää, A. Lehtonen, Coord. Chem. Rev. 256 (2012) 371e393.
[2] G. Ambrosi, M. Formica, V. Fusi, L. Giorgi, M. Micheloni, Coord. Chem. Rev. 252
(2008) 1121e1152.
[3] E. Safaei, H. Sheykhi, A. Wojtczak, Z. Jaglicic, A. Kozakiewicz, Polyhedron 30
(2011) 1219e1224.
[4] Y. Sarazin, R.H. Howard, D.L. Hughes, S.M. Humphrey, M. Bochmann, Dalton
Trans. (2005) 340e350.
[5] H. Zhou, H. Guo, Y. Yao, L. Zhou, H. Sun, H. Sheng, Y. Zhang, Q. Shen, Inorg.
Chem. 46 (2007) 958e964.
[6] F. Mu, E. Hamel, D.J. Lee, D.E. Pryor, M. Cushman, J. Med. Chem. 46 (2003)
1670e1682.
4.9. Model condensation of 4-methoxybenzaldehyde with pyrazole
and piperidine-1-N-[(4-methoxyphenyl)(1H-pyrazol)methyl]
piperidine (7)
[7] F.S. Cayabyab, L.C. Schlichter, J. Biol. Chem. 277 (2002) 13673.
[8] K.Y. Lee, D.H. Nam, C.S. Moon, S.H. Seo, J.Y. Lee, Y.S. Lee, Eur. J. Med. Chem. 41
(2006) 991e996.
[9] S. Torelli, C. Belle, S. Hamman, J.L. Pierre, E. Saint-Aman, Inorg. Chem. 41
(2002) 3983e3989.
[10] L. Que Jr., R.C. Kolanczyk, L.S. White, J. Am. Chem. Soc. 109 (1987) 5373e5380.
[11] E. Terreno, D.D. Castelli, A. Viale, S. Aime, Chem. Rev. 110 (2010) 3019e3042.
[12] L.M. De Leon-Rodriguez, A.J.M. Lubag, C.R. Malloy, G.V. Martinez, R.J. Gillies,
A.D. Sherry, Acc. Chem. Res. 42 (2009) 948e957.
[13] J.L. Major, T.J. Meade, Acc. Chem. Res. 42 (2009) 893e903.
[14] V. Jacques, J. Desreux, Top. Curr. Chem. (2002) 123e164.
[15] I. Mamedov, N.K. Logothetis, G. Angelovski, Org. Biomol. Chem. 9 (2011)
5816e5824.
Pyrazole (146 mg, 2 mmol) was dissolved in benzene (30 ml)
together with 4-methoxybenzaldehyde (0.25 ml, 2 mmol) and
piperidine (0.2 ml, 2 mmol) in a round-bottomed flask. The reac-
tion mixture was refluxed under a DeaneStark head for 14.5 h.
After removal of the solvent in vacuo an oily residue was left con-
taining 1-N-[(4-methoxyphenyl)(1H-pyrazol)methyl]piperidine 7
with 89% conversion. 1H NMR (CDCl3)
d
: 7.55 (d, 1H, J ¼ 2.0 Hz,
pyrazole C5eH), 7.48 (d, 1H, J ¼ 2.0 Hz, pyrazole C3eH), 7.25 (d, 2H,
J ¼ 8.7 Hz, Ar), 6.86 (d, 2H, J ¼ 8.7 Hz, Ar), 6.30 (t, 1H, J ¼ 2.0 Hz,
pyrazole C4eH), 5.97 (s, 1H, (NAr)(N)CHCAr), 3.89 (s, 3H, CH3O),
2.38e2.53 (m, 4H, CH2NCH2epiperidine), 1,37e1,68 (m, 6H,
CH2CH2CH2epiperidine).
[16] M. Oishi, S. Sumitani, Y. Nagasaki, Bioconjugate Chem. 18 (2007) 1379e1382.
[17] V. Stavila, Y. Stortz, C. Franc, D. Pitrat, P. Maurin, J. Hasserodt, Eur. J. Inorg.
Chem. 2008 (2008) 3943e3947.

ꢀ
N. Kuznik et al. / European Journal of Medicinal Chemistry 52 (2012) 184e192
192
[18] J.A. Duimstra, F.J. Femia, T.J. Meade, J. Am. Chem. Soc. 127 (2005)
12847e12855.
[19] T. Chauvin, P. Durand, M. Bernier, H. Meudal, B.T. Doan, F. Noury, B. Badet,
J.C. Beloeil, É Tóth, Angew. Chem., Int. Ed. 47 (2008) 4370e4372.
[20] K. Hanaoka, K. Kikuchi, T. Terai, T. Komatsu, T. Nagano, Chem.dEur. J. 14
(2008) 987e995.
[21] M. Giardiello, M.P. Lowe, M. Botta, Chem. Commun. (2007) 4044e4046.
[22] N. Kuznik, V. Stavila, M. Allali, Y. Stortz, P. Maurin, J. Hasserodt, Abstr. Paper
Am. Chem. Soc. (2008) 343-INOR.
[23] L. Shiftan, T. Israely, M. Cohen, V. Frydman, H. Dafni, R. Stern, M. Neeman,
Cancer Res. 65 (2005) 10316.
[24] A.L. Nivorozhkin, A.F. Kolodziej, P. Caravan, M.T. Greenfield, R.B. Lauffer,
T.J. McMurry, Angew. Chem., Int. Ed. 40 (2001) 2903e2906.
[25] T.J. Meade, A.K. Taylor, S.R. Bull, Curr. Opin. Neurobiol. 13 (2003) 597e602.
[26] O. Reany, T. Gunnlaugsson, D. Parker, J. Chem. Soc., Perkin Trans. 2 (2000)
1819e1831.
[28] C.J. Hinshaw, G. Peng, R. Singh, J.T. Spence, J.H. Enemark, M. Bruck,
J. Kristofzski, S.L. Merbs, R.B. Ortega, P.A. Wexler, Inorg. Chem. 28 (1989)
4483e4491.
[29] R.F. Borch, M.D. Bernstein, H.D. Durst, J. Am. Chem. Soc. 93 (1971) 2897e2904.
[30] A. Pelter, R.M. Rosser, S. Mills, J. Chem. Soc., Perkin Trans. 1 1 (1984) 717e720.
[31] K. Weber, S. Kuklinski, P. Gmeiner, Org. Lett. 2 (2000) 647e649.
[32] A.S. Nagle, R.N. Salvatore, B.D. Chong, K. Woon Jung, Tetrahedron Lett. 41
(2000) 3011e3014.
[33] T. Watanabe, I. Kinoyama, A. Kakefuda, T. Okazaki, K. Takizawa, S. Hirano,
H. Shibata, I. Yanagisawa, Chem. Pharm. Bull. 45 (1997) 996e1007.
[34] Y. Stortz, Synthèse Et Développement D’agents De Contraste Bioactifs En Vue
De La Visualisation in Vivo De L’expression De Gènes Par I.R.M, Ecole Normale
Supérieure, Lyon, France, 2006.
[35] D. Oulmi, P. Maillard, J.L. Guerquin-Kern, C. Huel, M. Momenteau, J. Org. Chem.
60 (1995) 1554e1564.
[36] A.R. Katritzky, L. Urogdi, A. Mayence, Synthesis (1989) 323e327.
[37] S.A. Kuby, H.A. Lardy, J. Am. Chem. Soc. 75 (1953) 890e896.
[27] W.O. Siegl, M.S. Chattha, EP 0,276,073, 1991.