Synthesis, physical-chemical characterisation and biological evaluation of novel 2-amido-3-hydroxypyridin-4(1H)-ones: Iron chelators with the potential for treating Alzheimer's disease

Article abstract of DOI:10.1016/j.bmc.2010.12.007

A novel class of 2-amido-3-hydroxypyridin-4-one iron chelators is described. These compounds have been designed to behave as suitable molecular probes which will improve our knowledge of the role of iron in neurodegenerative conditions. Neurodegenerative disorders, such as Alzheimer's disease (AD) and Parkinson disease (PD), can be considered as diverse pathological conditions sharing critical metabolic processes such as protein aggregation and oxidative stress. Interestingly, both these metabolic alterations seem to be associated with the involvement of metal ions, including iron. Iron chelation is therefore a potential therapeutic approach. The physico-chemical (pK_a, pFe ³⁺ and log P) and biological properties (inhibition of iron-containing enzymes) of these chelators have been investigated in order to obtain a suitable profile for the treatment of neurodegenerative conditions. Studies with neuronal cell cultures confirm that the new iron chelators are neuroprotective against β-amyloid-induced toxicity.

Full text of DOI:10.1016/j.bmc.2010.12.007

Bioorganic & Medicinal Chemistry 19 (2011) 1285–1297
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, physical–chemical characterisation and biological evaluation
of novel 2-amido-3-hydroxypyridin-4(1H)-ones: Iron chelators
with the potential for treating Alzheimer’s disease
Alessandra Gaeta a, Francisco Molina-Holgado c, Xiao L. Kong a, Sarah Salvage a, Sarah Fakih a,
Paul T. Francis b, Robert J. Williams d, Robert C. Hider a,
⇑
a Division of Pharmaceutical Science, King’s College, Franklin-Wilkins Building, Waterloo Campus, Stamford Street, London SE1 9NH, UK
b Wolfson Centre for Age Related Diseases, King’s College, Hodgkin Building, Guy’s Campus, London SE1 1UL, UK
c Health Sciences Research Centre, Roehampton University, Whitelands College, Holybourne Avenue, London SW15 4SD, UK
d Department of Biology and Biochemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 July 2010
Revised 26 November 2010
Accepted 3 December 2010
Available online 13 December 2010
A novel class of 2-amido-3-hydroxypyridin-4-one iron chelators is described. These compounds have
been designed to behave as suitable molecular probes which will improve our knowledge of the role
of iron in neurodegenerative conditions. Neurodegenerative disorders, such as Alzheimer’s disease
(AD) and Parkinson disease (PD), can be considered as diverse pathological conditions sharing critical
metabolic processes such as protein aggregation and oxidative stress. Interestingly, both these metabolic
alterations seem to be associated with the involvement of metal ions, including iron. Iron chelation is
therefore a potential therapeutic approach. The physico-chemical (pKa, pFe3+ and log P) and biological
properties (inhibition of iron-containing enzymes) of these chelators have been investigated in order
to obtain a suitable profile for the treatment of neurodegenerative conditions. Studies with neuronal cell
cultures confirm that the new iron chelators are neuroprotective against b-amyloid-induced toxicity.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Hydroxypyridin-4-ones
Neurodegeneration
Oxidative stress
b-Amyloid
1. Introduction
A chelating agent suitable for the treatment of neurodegenera-
tive disorders must fulfil critical requirements, among which
Oxidative stress, protein aggregation and redox active metal
ions can all be considered promising pharmacological targets for
the treatment of neurodegeneration.1 The critical role of iron and
copper, in both oxidative stress and protein aggregation processes,
renders chelation therapy a sensible therapeutic strategy. A chelat-
ing agent suitable for neurodegenerative disorders could have two
possible actions: (a) scavenging the free redox active metal present
in excess in the brain to form a non toxic metal complex, which is
then excreted; and (b) capping the metal at its protein binding site
blood–brain barrier (BBB) permeability is of uttermost importance.
Lipophilicity, expressed as the log P value, should represent a com-
promise between a high BBB penetration and a low liver extrac-
tion. Furthermore, with the necessity of ready permeation of the
BBB, the size of the chelator should probably be limited to less than
300 Daltons, thereby excluding most hexadentate ligands.3 The
inhibition of iron-containing enzymes, such as tyrosine hydroxy-
lase and lipoxygenase must be kept to a minimum.
There are two general classes of molecule that fit these require-
ments, 8-hydroxyquinolines and 3-hydroxypyridinones, both clas-
ses having been reported to possess potential for the treatment of
neurodegenerative disease. Thus, clioquinol (Fig. 1) has been dem-
onstrated to possess beneficial effects in both Alzheimer’s disease
and Parkinson’s disease models4,5 and also in clinical studies.6,7
Unfortunately halogenated hydroxyquinolines possess neurotoxic
side effects,8 but these can be avoided by the use of non haloge-
nated analogues, for instance VK-28 and M30 (Fig. 1).9 In a study
centred on rats with 6-hydroxydopamine induced striatal dopami-
nergic lesions, VK-28 was found to be capable of providing neuro-
protection at very low doses.10
(b-amyloid,
a
-synuclein), preventing any redox activity.2 Capping
the metal at the protein binding site would be expected to involve
additional interactions between the chelating agent and the target
protein.
Abbreviations: AD, Alzheimer’s disease; Ab, b-amyloid; BBB, blood–brain
barrier; CNS, central nervous system; DFO, desferrioxamine; FeNTA, iron nitrilotri-
acetic acid; HPO, 3-hydroxypyridin-4-one; LDH, lactate dehydrogenase; LO,
lipoxygenase; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium-bromide
mitochondrial; PD, Parkinson’s disease; TH, tyrosine hydroxylase.
Hydroxypyridin-4-ones have also been demonstrated to have
benefit in the treatment of neurodegeneration, for instance with
⇑
Corresponding author. Tel.: +44 (0) 20 7848 4882; fax: +44 (0) 20 7848 4800.
E-mail address: robert.hider@kcl.ac.uk (R.C. Hider).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.12.007

1286
A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
and has been demonstrated to be neuroprotective against b-amyloid
Cl
O
induced neurotoxicity in cortical neuron cultures.16,17
In the present work, we have synthesized a small library of
highly selective iron(III) hydroxypyridinone-based ligands with
suitable physico-chemical, toxicological properties and neuropro-
tective activity.
OH
N
CH3
I
N
The critical involvement of iron in oxidative stress and protein
aggregation renders chelation therapy a sensible strategy for the
treatment of neurodegenerative disorders in general and of AD in
particular.18 The new library of HPOs described in this work has
been specifically designed to possess potential for the treatment
of neurodegenerative diseases.
CH3
OH
Clioquinol
Deferiprone (CP20)
OH
N
2. Results and discussion
2.1. Chemistry
N
N
The synthesis of acid 9, a key building block, is summarised in
Scheme 1 and follows the procedure described by Piyamogkol et
al.19 Compounds 10a–k were prepared from the Z-protected amino
acids glycine, alanine and phenylalanine by reaction with the suit-
able amine following optimised reaction conditions, using either
EDC or DCCI as coupling reagents. Deprotection of compounds
10a–k by hydrogenation, using palladium as the catalyst, afforded
the free amines 11a–k, which were subsequentially reacted with
compound 9 using DCCI to afford intermediates 12a–k (Scheme 2).
Compound 14 (Scheme 3) was obtained by treating 12a with a large
excess of methyl iodide under reflux. Finally, compounds 12a–k and
14 were deprotected using boron trichloride to afford the desired
chelating agents 13a–l and 15. (Schemes 2 and 3).
N
N
OH
VK-28
OH
M30
Figure 1. Iron chelators reported to provide protection against neurotoxicity.
Friedrich’s ataxia.11,12 Hydroxypyridin-4-ones are suitable for this
purpose, by virtue of their relatively high selectivity for iron(III)
and their generally favorable physico-chemical properties. Deferi-
prone (Fig. 1), a 3-hydroxypyridin-4-one, is a bidentate chelator,
which is monoprotic at pH 7.4 and thus forms a neutral tris-iron(III)
complex. It has been used clinically for over fifteen years in the treat-
ment of transfusion-induced iron overload.13 It forms a stable 3:1 li-
gand iron complex, which is water soluble and readily excreted by
the kidneys.14 Furthermore, deferiprone readily crosses the BBB15
2.2. Physico-chemical properties of iron probes: iron affinity
(pM) and lipophilicity (log D and HPLC estimated log P)
Under biological conditions the pM value is a more useful param-
eter than the conventional stability constant for the assessment of
the ligand affinity for metals.20,3 The pM values of the newly
synthesised HPOs fall in a range, pFe3+ = 18.9–21.5, which is consis-
O1
C3
O2
C2
C1
C4
C5
N2
C16
C6
N1
C7
O4
C8
O3
N3
C9
C11
C10
C12
C13
C15
C14
A
B
Figure 2. (A and B) Crystal structure of compound 13f. (A) the free ligand is shown. H-bonding between N2 and O2 is highlighted; (B) compound 13f spatial arrangement. The
chelation moiety is planar and parallel to the plane of the benzyl group.

A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
1287
O
O
O
OBn
OH
OBn
a
b
N
H
O
O
maltol
2
3
OBn
OBn
OBn
OBn
c
d
OBn
OCOCH3
OBn
N
O
5
N
N
4
6
e
OBn
OBn
OBn
OBn
OBn
OH
f
OBn
OH
g
H
O
N
8
N
9
N
7
O
Scheme 1. Synthesis of Intermediate 9. Reagents and conditions: (a) Benzyl bromide, NaOH, methanol, reflux, 6 h; (b) ammonia, ethanol, reflux, overnight;
(c) triphenylphosphine, DEAD, THF, benzyl alcohol, reflux, overnight; (d) MCPBA, DCM, rt, 4 h; e) acetic anhydride, reflux, 2 h; (f) 2 N NaOH, reflux, 1 h; (g) DMSO,
py-SO3, TEA, Chloroform, rt, overnight; (h) NaClO2, NH2SO3H, acetone/water 1:1, rt, 3 h.
OH
OBn
N
OBn
OH
OBn
OBn
OH
O
a
b
H
O
H
Cl
N
N
N
H
NH
R1
NH
R1
N
9
R
H
O
R
H
O
O
13a-j
12a-j
10a, 12a, 13a, R=H, R1=CH3
O
H
10b, 12b, 13b, R=H, R1=CH2CH(CH3)2
10c, 12c, 13c, R=CH3, R1=CH3
10d, 12d, 13d, R=CH3, R1=CH2CH(CH3)2
10e, 12e, 13e, R=H, R1=Bn
N
H
Cbz
N
R
H
R1
10f, 12f, 13f, R=CH3, R1=Bn
10a-j
10g, 12g, 13g, R=Bn, R1=CH3
10h, 12h, 13h, R=Bn, R1=CH2CH(CH3)2
10i, 12i, 13i, R=Bn, R1=Bn
10j, 12j, 13j, R=H, R1=(CH3)2
OH
OH
O
H
Cl
N
N
H
N
R
H
O
13k
Scheme 2. Synthesis of iron chelators 13a–k. Reagents: (a) DCC, HOBt, DCM; (b) BCl3, DCM.
tent with high affinity iron(III) ligands (Table 1). Typically, for clini-
cally useful iron scavengers, a pFe3+ value P20 is considered to be
essential.3 When compared to deferiprone (pFe3+ = 19.4), the new
compounds show a good affinity for iron, with compounds 13g
and 13i having pFe3+ values of 21.4 and 21.5, respectively. The
potentiometric determination of one of these compounds, 13c, is
presented in Figure 3. Interestingly, compounds 13c, 13d and 13f
(alanine derivatives) have consistently lower pFe3+ values (19.1,
19.2 and 18.9, respectively), compared to the corresponding glycine
analogues 13a, 13b and 13e, (20.8, 20.6 and 19.1, respectively) and
phenylalanine analogues 13g, 13h and 13i (21.4, 20.4 and 21.5,
respectively), suggesting that the methyl group might force the

1288
A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
OBn
N
OBn
OBn
OBn
I-
a
O
O
H
H
N
N
NH
N
NH
CH3
O
CH3
O
CH3
12a
14
b
OH
CI
OH
O
O
H
N
N
NH
CH3
CH3
15
Scheme 3. Synthesis of iron chelator 15. Reagents: (a) methyl iodide, reflux; (b) BCl3, DCM
Table 1
Physico-chemical properties of novel iron chelators
A
Compound
pKa
pFe3+
log D
1.20
1.00
0.80
0.60
0.40
0.20
0.0
13a
13b
13c
13d
13e
13f
13g
13h
13i
6.3
6.2
6.0
6.0
6.2
6.0
6.1
6.0
6.0
—
20.8
20.6
19.1
19.2
19.1
18.9
21.4
20.4
21.5
—
ꢀ1.42
ꢀ0.33
ꢀ1.1
pH 7.42
ꢀ0.08
0.01
0.03
ꢀ0.11
0.39
0.90
15
ꢀ1.08a
a
pH 1.05
Log P value was estimated by HPLC.
ligand to adopt a different conformation. The crystal structure of
compound 13f shows the formation of an intramolecular hydrogen
bond between the amide NH and the ‘catechol’ hydroxyl group (
Fig. 2, Table 2). It is likely that the same hydrogen bond would form
in all the derivatives, thereby stabilising the ionised species at phys-
iological pH values and influencing the affinity for iron(III).
The lipophilicity of the novel compounds was assessed by the
shake-flask method (log D) and by HPLC (estimated log P). All the
compounds in the series have log D values (0.90–1.52) (Table 1)
falling in a range of lipophilicity, which is predicted to permit good
blood–brain barrier permeability. A good correlation was found be-
tween the determined log D values using the shake-flask method
and the log P values estimated using regression analysis of HPLC
retention times. Using this approach, the log P value for compound
15 was estimated to be –1.08 (retention time 3.87 min) as shown
in Figure 4. The log D for 15 could not be determined using the
shake-flask method due to the relative insolubility of the com-
pound in n-octanol. The estimated log P value appears to be consis-
tent with the introduction of an extra methyl group, when
compared to compound 13a, the non-methylated analogue having
a log D value of ꢀ1.52.
400
450
500
550
600
650
700
750
Wavelength (nm)
B
0.375
0.325
0.275
0.225
0.175
0.125
0.075
0.025
pH 10.98
pH 1.49
250
270
290
310
330
350
Wavelength (nm)
2.3. Lipoxygenase and tyrosine hydroxylase inhibition
Figure 3. (A) pH dependence of the spectrum of ligand 13c in the presence of
iron(III) over the pH range 1.05 and 7.42 in 0.1 M KCl at 25 °C [Fe] = 37.2
l
M
The compounds have been tested to evaluate inhibitory activity
on the iron containing enzymes lipoxygenase and tyrosine hydroxy-
[13c ] =186 lM. (B) pH dependence of the spectrum of ligand 13c over the pH range
1.49 and 10.98. [13c] = 4.68 ꢁ 10ꢀ4 M.

A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
1289
M)
Table 2
Table 3
Crystal data and structure refinement
Inhibition of lipoxygenase enzyme
Empirical formula
Formula weight (g/mol)
Crystal system
Space group
Unit cell dimensions
a (Å)
C16H18ClN3O4
351.79
orthorhombic
P212121 (No. 19)
Compound
EC50
(
lM)
%Inhibition (100
l
M)
%Inhibition (10 l
CP94
CP20
13a
13b
13c
13d
13e
13f
11
64
56
44
66
37
54
80
25
13
52
87.4a, 92
43
—
19.3
9.3
2.7
15.3
8.3
3.3
27.6
46.5
30.2
75.3a, 58
68
70
61
71
62
54
71
81
60
6.7608(1)
12.8465(1)
19.4017(2)
b (Å)
c (Å)
a
(°)
90
90
90
b (°)
13g
13h
13i
c
(°)
Volume (Å3)
1685.09(3)
Formula units/cell
8
Dc (g/cm3)
1.387
The EC50 is an indicative value, not necessarily reflecting the actual IC50, because the
Diffractometer (wavelength, Å)
BRUKER AXS SMART 6000
(Cu-Ka, k = 1.54178)
Göbel mirror
maximal inhibition is not achieved at 100
lM for all the compounds.
a
Published values.
Monochromator
Temperature (K)
293(2)
2.237
Absorption coefficient (mmꢀ1
Data collection range
Indices
)
8.26° < 2h < 142.86°
ꢀ8 6 h 6 7
ꢀ15 6 k 6 15
ꢀ21 6 l 6 23
9787
In an analogous fashion, lipophilicity is a dominant factor in
influencing the ability of HPOs to inhibit mammalian tyrosine
hydroxylase, typically hydrophilic chelators are weak inhibitors.25
Within the present HPO series, 13f and 13c were found to be weak
inhibitors at 10 lM, whereas 13g, 13h and 13j, the more lipophilic
derivatives, had a considerable inhibitory effect on the enzyme at
Reflections collected
Independent reflections
Program for structure solution
Program for structure refinement
Goodness-of-fit on F2
3118 [R(int) = 0.1417]
SHELXS-97
SHELXL-97
1.054
this concentration (Table 3 and Fig. 5).
Final R values (I > 2
r
(I))
R1 = 0.0416,
wR2 = 0.1064
R1 = 0.0421,
wR2 = 0.1070
0.301; ꢀ0.427
2.4. In vitro neuroprotective activity of new iron chelators
Final R value (all data)
(e Åꢀ3
)
The chelators have been evaluated for neuroprotective activity
in cultured primary mouse cortical neurons, against two types of
insult: the presence of either iron (FeNTA) or amyloid-b 1–40
(Ab1–40). The neuroprotective effect of iron chelation displayed
by the novel compounds in the FeNTA model was evaluated by
assessing LDH release into the media and MTT turnover within
cells and by the preservation of synaptic integrity after exposure
(D
/
q
)max; (D
/
q
)
min
of cortical neurones to FeNTA (10 lM) for 6 h prior to addition of
the drug. In the absence of drug, the exposure of neurons to FeNTA
led to a large increase in LDH levels in the media reflecting damage
to the neuronal cell membrane. This increase was significantly re-
duced by compounds 13b, 13f and 13g at concentrations of 30
lM
and above and also by 13i at 100 M (Table 4). In the absence of
l
drug, the exposure of neurons to FeNTA led to a large reduction
in MTT turnover reflecting impaired mitochondrial respiration
and reduced viable cell number. This reduction was significantly
reversed by 13a, 13b, 13f and 13g at concentrations of 30 lM
Figure 4. Regression analysis for estimation of log P for compound 15. The
retention time of the compound was determined as 3.87 min. This was plotted
against log D values determined for other compounds in the series. The estimated
log P value was calculated using the above equation.
and above (not shown). The greatest overall level of protection,
as assessed by either LDH release or MTT turnover, was observed
with compound 13f (Fig. 6) which was, therefore taken forward
for further analysis. Morphological assessment of neurons, follow-
ing exposure to FeNTA in the absence of drug, showed damage to
cell bodies and a loss of neuritis implying damage to synapses
( Fig. 7A). Damage to synapses was confirmed by a substantial
reduction in the levels of the nerve terminal marker synaptophysin
(Fig. 7B and C). Cell damage and reduction in synaptophysin levels
lase, with the second enzyme being particularly relevant in the con-
text of neurodegenerative disease. An important contribution to the
toxicityassociated with iron chelators is associated with their ability
to inhibit metallo-enzymes.21–23 In general, iron chelators do not
directly inhibit heam iron-containing enzymes due to the inaccessi-
bility of porphyrin-bound iron, however many non-heam iron-
containing enzymes are susceptible to such inhibition.24
were reversed by 13f at concentrations of 30 lM and 100 lM
(Fig. 7A–C).
Generally, hydrophobic chelators inhibit the lipoxygenase fam-
ily of enzymes,22,23 therefore the introduction of hydrophilic
groups into the chelator structure has a tendency to minimise such
potential, particularly if these substituents also create a steric
interference in the chelation at the enzyme active site.21 This con-
cept has been confirmed in this study, where none on the newly
described compounds were found to possess appreciable lipoxyge-
nase inhibitory activity. Thus, compound 13f does not show
comparable inhibition activity on LO to that of CP27, which was
used in this study as a positive control.
The neuroprotective actions of 13f were also evaluated against
synthetic human Ab1–40 induced toxicity. In this model, neurons
were exposed to 13f (10
lM, 30 lM and 100 lM) or Ab1–40
(3 M and 20 M) alone or in combination for 24 h and neuronal
l
l
cell viability assessed by Hoe33342 and propidium iodide double
labelling (Fig. 8). Exposure to Ab1–40 in the absence of 13f caused
a dramatic increase in propidium iodide staining and a reduction in
Hoe33342 indicating high levels of cell death. 13f had no effect on
cell viability alone but reversed Ab-induced toxicity at concentra-
tions of 10 lM and above.

1290
A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
Figure 5. Compounds 13a–l and 15 inhibitory activities against tyrosine hydroxylase enzyme at 10 lM are compared to potent inhibitor CP94.
Table 4
Protection of cultured cortical neurons against FeNTA-induced cell death
Compound
Control
FeNTA (10
lM)
FeNTA (10
l
M) + [Chelator] (
lM)
%Protection at 100 lM
10
30
100
13a
13b
13c
13d
13e
13f
13g
13h
13i
34.18 2.22
35.87 2.17
34.85 1.45
34.42 1.27
34.42 1.27
37.01 1.26
37.01 1.26
35.92 0.32
37.31 1.10
33.81 3.55
86.36 13.05
87.05 6.94
85.45 8.16
79.48 13.44
79.48 13.44
77.79 4.75
77.79 4.75
79.97 5.08
82.36 2.75
101.40 0.54
58.41 18.85
70.32 3.95
66.52 10.91
64.55 10.07
65.67 11.89
69.79 2.41
70.03 1.50
72.83 1.69
76.87 3.94
94.50 0.88
46.52 13.27
61.16 4.43⁄
61.96 9.95
54.73 6.37
56.62 9.46
52.49 4.05⁄⁄
48.9 2.70⁄⁄
53.8 8.74
46.10 4.15
56.44 7.46⁄
56.22 5.49
54.08 7.78
47.77 7.22
42.41 1.16⁄⁄
45.80 1.00⁄⁄
53.30 7.57
54.59 3.14⁄⁄
93.36 0.98
77
60
58
56
70
87
78
61
62
12
64.85 5.61
94.52 0.84
15
Data represents the mean release of lactate dehydrogenase into the growth medium as a % of total cellular LDH (=100%) SEM, n = 4.
⁄p <0.05; ⁄⁄p <0.01 One Way ANOVA.
This study demonstrates that in an in vitro model of neuronal
death, a range of novel 2-amido-3-hydroxypyridin-4-one iron che-
lators provide neuroprotection. Cortical neurons cultures have
proved to be useful models for studying the underlying mecha-
nisms to cell death. In this study several mechanisms of neuronal
damage have been investigated, leading to important observations.
First, FeNTA induced significant death of primary cortical neurons
in a concentration-dependent manner, which was reversed by iron
chelators in a concentration-dependent fashion. This concentra-
tion-dependent effect probably is a consequence of the neuronal
uptake of ferric iron, although an extracellular mode of action
may also be a factor. Furthermore, it is interesting to note that a
number of the chelators were capable of providing neuroprotection
for periods up to 6 h after the neuronal cultures had been exposed
to FeNTA. This finding is potentially of clinical interest, as it sug-
gests that iron chelation may provide protection post-insult, not
only in chronic neurodegeneration but also in acute conditions
involving iron, such as cerebral ischemia/stroke.26
One of the mechanisms of cytotoxicity induced by iron is the
disruption of the synaptic organization. In this study, we provide
evidence that the iron chelators 13c (not shown) and 13f preserve
synapses against iron-induced damage by preservation of the lev-
els of synaptophysin (Fig. 7B), a key synaptic structural protein.27
Recent evidence indicates that over-expression of synaptophysin
attenuates neurotoxicity.28
Despite the well established protective effects of iron chelation
in different pathologies,29 these findings, taken together with the
previous work with Deferiprone,16,17 provide the first example
where a clear connection is made between neuroprotection and
iron chelation.
processes responsible for both neuronal vulnerability in response
to an insult which requires redox active iron.
AD is associated with elevated soluble and aggregated forms of
both Ab1–40 and Ab1–42 and oxidative stress30 and there is
increasing evidence for a detrimental role of iron in the pathogenic
process.31,32 This study clearly demonstrates that Ab (3 or 10
lM)
induced neuronal cell death was abolished by treatment with low
concentrations of iron chelators, suggesting a role for iron in the
observed Ab neurotoxic effects. As a consequence, the use of iron
chelators may offer therapeutic advances in the treatment of neu-
rodegenerative disorders related to amyloid toxicity.
3. Conclusions
In summary, new iron bidentate chelators belonging to the fam-
ily of 2-amido-3-hydroxypyridin-4-one show potential as non-
toxic therapeutic agents for the treatment of neurodegenerative
disorders.
Although much needs to be understood in terms of the etio-
pathogenesis of neurodegeneration, it is clear from this study that
chelation therapy can be considered as a potential strategy for the
treatment and investigation of neurodegeneration.
4. Experimental
4.1. Chemistry
Melting points were determined using an Electrothermal IA
9100 Digital Melting Point Apparatus and are uncorrected. IR spec-
tra were performed on a Perkin–Elmer 1605 FTIR. 1H NMR spectra
were recorded on a Bruker (360 MHz) spectrometer. Chemical
Furthermore, the in vitro approach on primary neuronal
cultures using Ab1–40 enabled a specific analysis of the cellular

A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
1291
90
80
70
60
50
40
30
20
10
0
A
A
**
**
Control
0
10
30
100
B
C
FeNTA (10μM) + [13f](μM)
B
100
55kDa
31kDa
**
75
50
25
0
*
1.0
0.8
0.6
0.4
0.2
0.0
Control
0
10
30
100
FeNTA(10μM) + [13f](μM)
Figure 6. Compound 13f protects cortical neurons against FeNTA-induced neuro-
toxicity in a concentration-dependent manner. Neurons were exposed to FeNTA
(10
lM) for 6 h before the addition of 13f for a further 18 h. Neuronal viability was
then assessed by LDH release (A) and by MTT turnover (B). Compound 13f protected
neurones against FeNTA toxicity at concentrations of 30 lM and 100 l
M. ⁄p <0.05,
⁄⁄p <0.01, n = 4; One Way ANOVA.
Figure 7. Compound 13f protects cortical neurons against FeNTA-induced loss of
synaptic integrity. Upper panel (A) shows phase contrast images of neurons treated
shifts (d) are reported in ppm downfield from the internal standard
tetramethylsilane (TMS). Mass spectra (ESI) analyses were carried
out by Mass Spectrometry Facility, School of Health and Science,
Franklin-Wilkins Building, King’s College, London SE1 9NH. All
starting materials were obtained from commercial sources and
used without further purification. 3-Hydroxy-2-methyl-4H-pyr-
an-4-one (maltol, 1) was purchased from Cultor Food Science. Col-
umn chromatography was performed on silica gel 220–440 mesh
(Fluka). All the biologically tested compounds (13a–k and 15) were
tested for purity by HPLC and were found more than 95% pure.
HPLC method & column: 5–95% acetonitrile/water gradient over
with vehicle (A), FeNTA (10
with 13f (100 M). Immunoblot (B) shows levels of synaptophysin in neuronal
lysates following treatment with vehicle (Con), or FeNTA (10 M) with or without
lM) alone (B), FeNTA with 13f (30 lM) (C) or FeNTA
l
l
increasing concentrations of 13f. Arrows indicate position of molecular weight
markers. Graph (C) shows quantification of the immunoreactive bands shown in B
using image J densitometric analysis.
mp 54–56 °C. Yield 80%. 1H NMR (CDCl3) d 2.07 (3H, s, CH3), 5.15
(2H, s, CH2Ph), 6.36 (1H, d, J = 5.7 Hz, 5-H), 7.31–7.40 (5H, m,
CH2Ph), 7.59 (1H, d, J = 5.7 Hz, 6-H); m/z (ESI): 202.0.
30 min on Luna C18, 10
l
, 250 ꢁ 4.6 mm at 254 nm wavelength.
4.1.1. 2-Methyl-3-benzyloxypyran-4(1H)-one (2)
To a solution of maltol (1) (10 g, 0.079 mol) in methanol (20 mL)
was added sodium hydroxide (3.49 g, 0.087 mol, 1.1 equiv) in
water (10 mL). The reaction mixture was heated to reflux before
benzyl bromide (10.4 mL, 0.087 mol, 1.1 equiv) was slowly intro-
duced dropwise and the mixture was left to reflux for 6 h. After
the solvent was removed, the residue was taken into water and
dichloromethane. The aqueous fraction was discarded and the or-
ganic fraction washed with sodium hydroxide 5% (3ꢁ) followed
by water (2ꢁ). The combined fractions were dried over anhydrous
sodium sulfate, filtered, and evaporated under reduced pressure.
Re-crystallisation from diethyl ether afforded off-white crystals,
4.1.2. 2-Methyl-3-benzyloxypyridin-4(1H)-one (3)
To a solution of 2 (13.8 g, 0.064 mol) in ethanol (25 mL) was
added ammonia solution (50 mL) and refluxed overnight. The sol-
vent was removed under reduced pressure, then taken into water
and adjusted to pH 1 with concentrated hydrochloric acid. The
aqueous mixture was washed with ethyl acetate (3ꢁ) and the pH
was adjusted to pH 10 with sodium hydroxide (2 M.). The aqueous
phase was extracted with chloroform (3ꢁ), dried over anhydrous
sodium sulfate, filtered, and evaporated under reduced pressure.
Re-crystallisation from methanol/diethyl ether gave brown cubic
crystals, mp 162–164 °C. Yield 75%. 1H NMR (CDCl3) d 2.15 (3H,

1292
A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
1589, 1498, 1485 and 1449 (ring C@C), 1218 and 1066 (C–O–C)
cmꢀ1 1H NMR (CDCl3) d 2.43 (3H, s, CH3), 5.00 (2H, s, 3-OCH2Ph),
Control
Aβ3μM
Aβ20μM
.
5.17 (2H, s, 4-OCH2Ph), 6.79 (1H, d, J = 5.6 Hz, 5-H), 7.30–7.45
(10H, m, 3-OCH2Ph and 4-OCH2Ph), 8.13 (1H, d, J = 5.6 Hz, 6-H);
m/z (ESI): 290.8.
Aβ3μM + 13f
Aβ3μM +13f
Aβ3μM + 13f
Aβ20μM + 13f
Aβ20μM + 13f
Aβ20μM + 13f
13f(10μM)
13f(30μM)
13f(100μM)
4.1.4. 2-Methyl-3,4-dibenzyloxypyridine N-oxide (5)
A solution of m-chloroperoxybenzoic acid (MCPBA) (0.622 g,
3.63 mmol, 1.1 equiv) in dichloromethane (20 mL) was prepared
andcooledto0 °C. Asolutionof4(1 g, 3.3 mmol)indichloromethane
(5 mL) was added slowly. The reaction mixture was left to stir at
room temperature for 3 h prior to addition of dichloromethane
(20 mL) to increase the volume. The solution was washed with so-
dium carbonate (5%, 3ꢁ). The organic phase was dried over anhy-
drous sodium sulfate, filtered, and concentrated under reduced
pressure to give yellow oil. Crystallisation in the form of white fluffy
powder resulted subsequent to the addition of diethyl ether, mp
127–129 °C. Yield 77%. mmax (KBr) 3245 (ring C–H), 3041 and 2991
(aliphatic C–H), 1533 (ring C@C), 1240 and 1068 (C–O–C) cmꢀ1. 1H
NMR (CDCl3) d 2.40 (3H, s, CH3), 5.05 (2H, s, 3-OCH2Ph), 5.17 (2H, s,
4-OCH2Ph), 6.74 (1H, d, J = 7.3 Hz, 5-H), 7.32–7.41 (10H, m, 3-
OCH2Ph and 4-OCH2Ph), 8.04 (1H, d, J = 7.3 Hz, 6-H); m/z (ESI): 306.9.
100
75
4.1.5. 2-Acetoxymethyl-3,4-dibenzyloxypyridine (6)
Acetic anhydride (20 mL) was added into a flask containing 5
(1 g, 3.10 mmol) and the reaction mixture was heated to 130 °C
for 1 h. The solvent was removed under reduced pressure and
the residue dissolved in water. The pH of the solution was adjusted
to 8 with sodium hydroxide (2 M) and was then extracted with
dichloromethane (3ꢁ). The organic fractions were dried over anhy-
drous sodium sulfate, filtered, and concentrated in vacuo to yield
brown oil. Treatment with decolourising charcoal yielded yellow
oil. 1H NMR (CDCl3) d 2.07 (3H, s, OCOCH3), 5.08 (2H, s, 3-OCH2Ph),
5.18 (2H, s, 4-OCH2Ph), 5.20 (2H, s, CH2OCOMe), 6.91 (1H, d,
J = 5.6 Hz, 5-H), 7.30–7.48 (10H, m, 3-OCH2Ph, 4-OCH2Ph), 8.25
(1H, d, J = 5.6 Hz, 6-H); m/z (ESI): 305.1.
50
*
*
*
25
0
Con
Aβ
+100
+10
+30
[13f](μM)
4.1.6. 2-Hydroxymethyl-3,4-dibenzyloxypyridine (7)
To
a solution of 2-acetoxymethyl-3,4-dibenzyloxypyridine
Figure 8. Compound 13f protects cortical neurons against Ab1–40 induced toxicity.
Neurons were exposed to 13f (10 M, 30 M and 100 M) or Ab1–40 (3 M and
20 M) alone or in combination for 24 h and neuronal cell viability assessed by
Hoe33342 (blue) and propidium iodide (purple) double labelling. Data was
captured using an In Cell Analyser and % cytotoxicity calculated as number of
propidium iodide positive cells relative to the number of Hoe33342 positive cells.
Compound 13f significantly reduced Ab toxicity. ⁄p <0.05, One Way ANOVA.
(1.14 g, 3.13 mmol) in ethanol (10 mL), sodium hydroxide (2 M,
7 mL) was added and the reaction mixture refluxed for 2 h. The
product was extracted with dichloromethane (4ꢁ), dried over
anhydrous sodium sulfate, filtered, and concentrated under re-
duced pressure to give an off-white solid (81% overall yield in
two steps). Re-crystallisation from diethyl ether/petroleum spirit
gave an off-white fluffy powder, mp 83–85 °C; mmax (KBr) 3165
(br, O–H), 2954 (aliphatic C–H), 1595 (ring C@C), 1301 and 1035
l
l
l
l
l
s, CH3), 5.03 (2H, s, CH2Ph), 6.35 (1H, d, J = 6.9 Hz, 5-H), 7.25–7.31
(5H, m, CH2Ph), 7.39 (1H, d, J = 6.9 Hz, 6-H); m/z (ESI): 201.1.
(C–O–C) cmꢀ1 1H NMR (CDCl3) d 3.69 (1H, s, CH2OH), 4.65 (2H,
.
s, CH2OH), 5.06 (2H, s, 3-OCH2Ph), 5.21 (2H, s, 4-OCH2Ph), 6.89
(1H, d, J = 5.5 Hz, 5-H), 7.32–7.52 (10H, m, 3-OCH2Ph, 4-OCH2Ph),
8.19 (1H, d, J = 5.5 Hz, 6-H); m/z (ESI): 320.8.
4.1.3. 2-Methyl-3,4-dibenzyloxypyridine (4)
Triphenylphosphine (TPP) (2.9 g, 11.16 mmol, 1.2 equiv) was
slowly added to a solution of 3 (2 g, 9.30 mmol) in dry tetrahydro-
furan (20 mL), and the solution was cooled to 0 °C in ice bath. Ben-
zyl alcohol (1.2 g, 11.16 mmol, 1.2 equiv) was later introduced
dropwise followed by diethylazodicarboxylate (DEAD) (1.9 g,
11.16 mmol, 1.2 equiv) in the same manner. After refluxing the
reaction mixture overnight, the solvent was removed under re-
duced pressure and the residue was extracted with water. The mix-
ture was adjusted to pH 1 with concentrated hydrochloric acid
before washing with diethyl ether (4ꢁ). The pH of the aqueous
fraction was increased to 8 with sodium hydroxide (2 M.), followed
by extraction with ethyl acetate (4ꢁ). The combined organic frac-
tions were dried over anhydrous sodium sulfate, filtered, and con-
4.1.7. 2-Formyl-3,4-dibenzyloxypyridine (8)
To a solution of 7 (8 g, 0.025 mol) in chloroform (138 mL), was
added dimethyl sulfoxide (DMSO) (37 mL) and triethylamine (TEA)
(21 mL, 6 equiv). The reaction mixture was then cooled in an ice-
bath followed by the slow addition of sulfur trioxide pyridine com-
plex (20 g, 0.125 mol, 5 equiv). The mixture was allowed to thaw at
room temperature and left to stir overnight. Water (2ꢁ) was used
to wash the organic fraction, which was subsequently dried over
anhydrous sodium sulfate, filtered, and concentrated in vacuo.
The dark green residue obtained was loaded on to a silica gel
column (eluant: chloroform/methanol/ethyl acetate; 45:5:50 v/v)
to yield an off-white solid. Yield 62%. Recrystallisation from
chloroform/petroleum spirit yielded off-white fluffy crystals: mp
centrated under reduced pressure to give
Recrystallisation from chloroform/petroleum spirit gave white
a white solid.
crystals, mp 85–87 °C. Yield 79%. mmax (KBr) 3264 (ring C–H),
103–104 °C; mmax (KBr) 3065 and 3031 (ring C–H), 2858 (aldehyde

A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
1293
C–H), 1709 (aldehyde C@O), 1573 (ring C@C), 1251 and 1043
(C–O–C) cmꢀ1 1H NMR (CDCl3) d 5.19 (2H, s, 3-OCH2Ph), 5.23
(2H, s, 4-OCH2Ph), 7.07 (1H, d, J = 5.3 Hz, 5-H), 7.31–7.46 (10H,
m, 3-OCH2Ph and 4-OCH2Ph), 8.40 (1H, d, J = 5.3 Hz, 6-H), 10.24
(1H, s, CHO); m/z (ESI): 318.8.
Compound 10b: Yield 63%. 1H NMR (CDCl3) d 0.88 (d, 6H,
J = 6.7 Hz, –CH2CH(CH3)2), 1.75 (m, 1H, J = 6.4 Hz, 6.7 Hz,
CH2CH(CH3)2), 3.08 (t, 2H, J = 6.4 Hz, –CH2CH(CH3)2), 3.85 (d, 2H,
J = 5.7 Hz, CH2), 5.13 (s, 2H, cbzCH2), 5.47 (br s, 1H,
.
–
a
NHCH2CH(CH3)2) 6.09 (br s, 1H, NHcbz), 7.35 (m, 5H, cbzPh); m/z
(ESI): 263.9.
Compound 10c: Yield 75.6%. 1H NMR (CDCl3) d 1.36 (d, 3H,
4.1.8. 2-Carboxy-3,4-dibenzyloxypyridine (9)
Compound 8 (2 g, 6.25 mmol) was dissolved in acetone (20 mL)
and water (20 mL). To this solution was added sulfamic acid
(850 mg, 8.75 mmol, 1.4 equiv) and sodium chlorite (80%,
622 mg, 6.87 mmol, 1.1 equiv) and stirred at room temperature
for 3 h. in an open flask. Removal of acetone in vacuo yielded crude
product as a precipitate in the remaining aqueous solution. This
was collected, washed with acetone and dried to yield off-white
powder, mp 120 °C. Yield 77%. mmax (KBr) 3033 (br, O–H), 1707
(br, acid C@O), 1607 and 1499 (ring C@C), 1223 and 1026 (C–O–
J = 7.1 Hz,
aCHCH3), 2.78 (d, 3H, J = 4.8 Hz, CH3NH-), 4.21 (m, 1H,
aCHCH3), 5.18 (s, 2H, cbzCH2), 5.49 (br s, 1H, NHCH3) 6.33 (br s,
1H, NHcbz), 7.33 (m, 5H, cbzPh); m/z (ESI): 235.7.
Compound 10d: Yield 80%. 1H NMR (CDCl3) d 0.88 (d, 6H,
J = 6.7 Hz, –CH2CH(CH3)2), 1.38 (d, 3H, J = 6.9 Hz,
(m, 1H, J = 6.7 Hz, –CH2CH(CH3)2), 3.06 (m, 2H, CH2CH(CH3)2),
4.20 (m, 1H, CHCH3), 5.10 (s, 2H, cbzCH2), 5.36 (br s, 1H, NHcbz)
aCHCH3), 1.75
a
6.13 (br s, 1H, NHCH2CH(CH3)2), 7.34 (m, 5H, cbzPh); m/z (ESI):
278.3.
C) cmꢀ1 1H NMR (MeOD) d 5.15 (2H, s, 3-OCH2Ph), 5.39 (2H, s,
.
Compound 10g: Yield 85%. 1H NMR (CDCl3) d 2.71 (d, 3H,
J = 4.8 Hz, –NHCH3), 3.02 (dd, 1H, Jgem=13.6 Hz, Jvic = 7.6 Hz, –
4-OCH2Ph), 7.25–7.55 (10, m, 3-OCH2Ph and 4-OCH2Ph), 7.55 (1H,
d, J = 6.2 Hz, 5-H), 8.32 (1H, d, J = 6.2 Hz, 6-H); m/z (ESI): 334.8.
a
CHCH2Ph), 3.12 (dd, 1H, Jgem = 13.6 Hz, Jvic = 6.0 Hz, -
4.30–4.36 (m, 1H, – CHCH2Ph), 5.08 (s, 2H, cbzCH2), 5.32 (br s,
1H, NHcbz) 5.61 (br s, 1H, NHCH3), 7.17–7.38 (m, 10H, cbzPh and
aCHCH2Ph),
a
4.1.9. General procedure for preparation of compounds 10e,f,i
This procedure is illustrated for compound 10e. To a solution of
Z-glycine (550 mg, 2.4 mmol) in dry dichloromethane (10 mL) at
0 °C and under nitrogen, N-(dimethylaminopropyl)-N-ethylcarbo-
diimide hydrochloride (EDC) (690 mg, 3.6 mmol, 1.5 equiv), TEA
(364 mg, 3.6 mmol, 1.5 equiv), DMAP (293 mg, 2.4 mmol, 1 equiv)
were added. The mixture was allowed to stir for 10 min before
benzylamine (1.03 g, 9.6 mmol, 4 equiv) was added, and the reac-
tion was left to stir at room temperature for 12 h. Then, the mix-
ture was concentrated under reduced pressure, and the residue
was diluted with ethyl acetate, washed sequentially with 5% citric
acid solution (2ꢁ), saturated aqueous sodium bicarbonate (2ꢁ),
and brine, dried over anhydrous Na2SO4 and concentrated under
reduced pressure, to afford the title compound as a white solid.
-a
CHCH2Ph); m/z (ESI): 311.7.
Compound 10h: Yield 78%. 1H NMR (CDCl3) d 0.74 (d, 3H,
J = 6.5 Hz, –CH2CH(CH3)2), 0.76 (d, 3H, J = 6.4 Hz, -CH2CH(CH3)2),
1.53–1.64 (m, 1H, –CH2CH(CH3)2), 2.94–2.98 (m, 2H,
CH2CH(CH3)2), 3.02 (dd, 1H, Jgem = 13.6 Hz, Jvic = 7.8 Hz,
-
–
a
CHCH2Ph), 3.12 (dd, 1H, Jgem = 13.6 Hz, Jvic = 6.2 Hz, –
4.32–4.38 (m, 1H, – CHCH2Ph), 5.08 (s, 2H, cbzCH2), 5.39 (br s,
1H, NHcbz) 5.68 (br s, 1H, –NHCH2CH(CH3)2), 7.18–7.35 (m, 10H,
cbzPh and - CHCH2Ph); m/z (ESI): 353.9.
Compound 10j: Yield 77%. 1H NMR (CDCl3) d 2.96 (s, 3H,
N(CH3)2), 2.98 (s, 3H, N(CH3)2), 4.00 (d, 2H, J = 4.2 Hz, – CH2–),
aCHCH2Ph),
a
a
a
5.12 (s, 2H, cbzCH2), 5.83 (br s, 1H, NHcbz), 7.30–7.41 (m, 5H,
cbzPh); m/z (ESI): 237.2.
Yield 85%. 1H NMR (CDCl3) d 3.90 (d, 2H, J = 5.6 Hz,
aCH2), 4.45
Compound 10k: Yield 82%. 1H NMR (CDCl3) d 1.51–1.70 (m, 6H,
pip), 3.30 (t, 2H, J = 5.4 Hz, -pip), 3.56 (t, 2H, J = 5.5 Hz, pip), 4.00 (d,
(d, 2H, J = 5.8 Hz, NHCH2-Ph), 5.11 (s, 2H, cbzCH2), 5.36 (br s, 1H,
NHCH2Ph) 6.24 (br s, 1H, NHcbz), 7.34 (m, 10H, cbzPh and
NHCH2Ph); m/z (ESI): 298.1.
2H, J = 4.2 Hz, –aCH2–), 5.12 (s, 2H, cbzCH2), 5.87 (br s, 1H, NHcbz),
7.30–7.36 (m, 5H, cbzPh); m/z (ESI): 275.8.
Compound 10f: Yield 57%. 1H NMR (CDCl3) d 1.41 (d, 3H,
J = 7.0 Hz,
aCHCH3). 4.25 (m, 1H, aCHCH3), 4.44 (m, 2H, NHCH2-
4.1.11. General procedure for preparation of compounds 11a–k
This procedure is illustrated for compound 11a. To a solution of
compound 10a (170 mg, 0.77 mmol) in methanol (10 mL), 10% Pd/
C was added. The reaction was hydrogenated at room temperature
and atmospheric pressure for 3 h. Then the catalyst was filtered off
through Celite, and the clear solution, taken to dryness, afforded
the title compound as an oil. The reaction was monitored by TLC
and, for stability problems, the amine was directly used in the sub-
sequent reaction. Yield 97%.
Compound 11b: The reaction was monitored by TLC and, for
stability problems, the amine was directly used in the subsequent
reaction. Yield 94%.
Compound 11c: The reaction was monitored by TLC and, for sta-
bility problems, the amine was directly used in the subsequent
reaction. Yield 96%.
Ph), 5.09 (s, 2H, cbzCH2), 5.28 (br s, 1H, NHCH2Ph) 6.32 (br s, 1H,
NHcbz), 7.26–7.34 (m, 10H, cbzPh and NHCH2Ph); m/z (ESI): 311.8.
Compound 10i: Yield 92.8%. 1H NMR (CDCl3) d 3.09 (2dd, 2H,
J = 6.2 Hz, 7.6 Hz,
2H, NHCH2-Ph), 5.03 (s, 2H, cbzCH2), 5.45 (br s, 1H, NHCH2Ph)
aCHCH2Ph), 4.33 (m, 1H, aCHCH2Ph), 4.45 (m,
6.22 (br s, 1H, NHcbz), 7.06–7.34 (m, 15H, cbzPh,
NHCH2Ph); m/z (ESI): 388.0.
aCHCH2Ph and
4.1.10. General procedure for preparation of compounds
10a,b,c,d,e,h,j,k
This procedure is illustrated for compound 10a. To a stirred solu-
tion of Z-glycine (250 mg, 1.2 mmol) in dichloromethane at 0 °C,
dicyclohexylcarbodiimide (DCC) (296 mg, 1.44 mmol, 1.2 equiv)
and hydroxybenzotriazole (HOBt) (195 mg, 1.44 mmol, 1.2 equiv)
were added. The reaction mixture was maintained at 0 °C for 1 h,
and then it was allowed to warm up to room temperature. The
methylamine (112 mg, 3.6 mmol, 3 equiv) was added, and the reac-
tion mixture was stirred for 12 h. The DCU was filtered, and the or-
ganic layer was washed with 5% citric acid solution (2ꢁ), saturated
aqueous sodium bicarbonate (2ꢁ), and brine, dried and concen-
trated in vacuo, to afford a clear oil. The obtained residue was puri-
fied by flash column chromatography (EtOAc/hexane, 8:2),
affording the title compound as a white solid. Yield 64%. 1H NMR
(CDCl3) d 2.80 (d, 3H, J = 4.8 Hz, CH3NH–), 3.84 (d, 2H, J = 5.8 Hz,
Compound 11d: The reaction was monitored by TLC and, for
stability problems, the amine was directly used in the subsequent
reaction. Yield 97%.
Compound 11e: Yield 96%. 1H NMR (CD3OD) d 3.73 (s, 2H, –
CH2NH2), 4.44 (s, 2H, –NHCH2Ph), 7.25–7.46 (m, 5H, –NHCH2Ph).
Compound 11f: Yield 97%. 1H NMR (CD3OD) d 1.30 (d, 3H,
J = 6.9 Hz, –CH(CH3)NH2), 3.44–3.50 (m, 1H, -CH(CH3)NH2), 4.36
(s, 2H, –NHCH2Ph), 7.23–7.35 (m, 5H, –NHCH2Ph).
Compound 11g: Yield 96%. 1H NMR (CDCl3) d 2.65 (dd, 1H,
a
CH2), 5.12 (s, 2H, cbzCH2), 5.43 (br s, 1H, NHCH3) 6.04 (br s, 1H,
Jgem = 13.7 Hz, Jvic = 9.4 Hz, -
NHCH3), 3.27 (dd, 1H, Jgem = 13.7 Hz, Jvic = 3.9 Hz, –
a
CHCH2Ph), 2.80 (d, 3H, J = 4.9 Hz, –
NHcbz), 7.35 (m, 5H, cbzPh); m/z (ESI): 221.8.
aCHCH2Ph),

1294
A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
3.59 (dd, 1H, J = 3.9, 9.4 Hz, –
aCHCH2Ph), 7.08–7.44 (m, 5H, –
7.17–7.41 (m, 15H, 3-OCH2Ph, 4-OCH2Ph and –NHCH2Ph), 8.17
a
CHCH2Ph).
Compound 11h: Yield 94%. 1H NMR (CDCl3) d 0.89 (d, 6H,
J = 5.2 Hz, (CH3)2CHCH2NH–), 1.70–1.76 (m, 1H, (CH3)2CHCH2NH–
), 2.69 (dd, 1H, Jgem = 13.7 Hz, Jvic = 9.2 Hz, – CHCH2Ph), 3.05–3.10
(m, 2H, (CH3)2CHCH2NH–), 3.26 (dd, 1H, Jgem = 13.7 Hz, Jvic = 4.0 Hz,
CHCH2Ph), 3.60 (dd, 1H, J = 4.0, 9.2 Hz, - CHCH2Ph), 7.20–7.37
(m, 5H, – CHCH2Ph).
Compound 11i: Yield 95%. 1H NMR (CDCl3) d 3.03 (2H, m,
CHCH2Ph), 3.96 (1H, t, J = 7.1 Hz, CHCH2Ph), 4.25 (2H, dd,
J = 14.3 Hz, NHCH2Ph), 7.03–7.34 (10H, m, NHCH2Ph and
(br s, 1H, –CONH–), 8.19 (d, 1H, J = 5.3 Hz, 6-H); m/z (ESI): 495.1.
Compound 12g: Yield 31.2%. 1H NMR (CDCl3) d 2.65 (d, 3H,
J = 4.7 Hz, NHCH3), 3.04–3.16 (m, 2H, –CH2Ph), 4.82–4.88 (m, 1H,
a
–aCHCH2Ph), 5.01–5.05 (m, 2H, 3-OCH2Ph), 5.14 (s, 2H, 4-OCH2Ph),
6.45 (br s, 1H, –NHCH3), 6.96 (d, 1H, J = 5.1 Hz, 5-H), 7.15–7.44 (m,
15H, 3-OCH2Ph, 4-OCH2Ph and –CH2Ph), 8.19 (d, 1H, J = 5.1 Hz,
6-H), 8.35 (d, 1H, J = 7.9 Hz, –CONH–); m/z (ESI): 494.8.
–a
a
a
Compound 12h: Yield 43%. 1H NMR (CDCl3) d 0.73–0.76 (m, 6H,
–CH2CH(CH3)2), 1.57–1.65 (m, 1H, -CH2CH(CH3)2), 2.93–2.99 (m,
2H, –CH2CH(CH3)2), 3.06–3.19 (m, 2H, –CH2Ph), 4.81–4.83 (m,
a
a
aCHCH2Ph).
1H, –aCHCH2Ph), 5.09 (s, 2H, 3-OCH2Ph), 5.18 (s, 2H, 4-OCH2Ph),
Compound 11j: The reaction was monitored by TLC and, for sta-
bility problems, the amine was directly used in the subsequent
reaction. Yield 94%.
Compound 11k: The reaction was monitored by TLC and, for
stability problems, the amine was directly used in the subsequent
reaction. Yield 97%.
6.21 (br s, 1H, –NHCH2CH(CH3)2), 6.97–6.98 (m, 1H, 5-H),
7.18–7.69 (m, 15H, 3-OCH2Ph, 4-OCH2Ph and –CH2Ph), 8.19 (m,
1H, 6-H), 8.28 (d, 1H, J = 5.7 Hz, –CONH–); m/z (ESI): 536.9.
Compound 12i: Yield 54.5%. 1H NMR (CDCl3) d 3.16 (d, 2H,
J = 7.2 Hz, –CH2Ph), 4.27–4.39 (m, 2H, –NHCH2Ph), 4.87 (q, 1H,
J = 7.2 Hz, 8.1 Hz, –aCHCH2Ph), 5.05 (s, 2H, 3-OCH2Ph), 5.15 (s,
2H, 4-OCH2Ph), 6.43 (br s, 1H), 6.97 (d, 1H, J = 5.4 Hz, 5-H),
7.05–7.07 (m, 1H), 7.18–7.27 (m, 15H), 7.33–7.42 (m, 4H), 8.18
(d, 1H, J = 5.4 Hz, 6-H), 8.30 (d, 1H, J = 8.1 Hz, –CONH–); m/z
(ESI): 571.4.
4.1.12. General procedure for preparation of compounds 12a–k
This procedure is illustrated for compound 12a. To a stirred
solution of 9 (287 mg, 0.85 mmol) in dichloromethane at 0 °C, dicy-
clohexylcarbodiimide (DCC) (211 mg, 1.02 mmol, 1.2 equiv) and
hydroxybenzotriazole (HOBt) (138 mg, 1.02 mmol, 1.2 equiv) were
added. The reaction mixture was maintained at 0 °C for 1 h, and
then it was allowed to warm up to room temperature. Compound
11a (130 mg, 1.27 mmol, 1.5 equiv) was added, and the reaction
mixture was stirred for 12 h. The DCU was filtered, and the organic
layer was washed with 5% citric acid solution (2ꢁ), saturated aque-
ous sodium bicarbonate (2ꢁ), and brine, dried and concentrated in
vacuo, to afford a clear oil. The obtained residue was purified by
flash column chromatography (chloroform/methanol, 9:1), afford-
ing the title compound as a white solid. Yield 75.6%. 1H NMR
(CDCl3) d 2.77 (d, 3H, J = 4.9 Hz, NHCH3), 4.06 (d, 2H, J = 6.1 Hz,
Compound 12j: Yield 57.3%. 1H NMR (CDCl3) d 3.01 (s, 3H,
NH(CH3)2), 3.02 (s, 3H, NH(CH3)2), 4.23 (d, 2H, J = 4.2 Hz,
aCH2),
5.16 (s, 2H, 3-OCH2Ph), 5.17 (s, 2H, 4-OCH2Ph), 6.98 (d, 1H,
J = 5.3 Hz, 5-H), 7.37–7.47 (m, 10H, 3-OCH2Ph and 4-OCH2Ph),
8.25 (d, 1H, J = 5.3 Hz, 6-H), 8.56 (br s, 1H, –CONH–); m/z (ESI):
419.7.
Compound 12k: Yield 75.6%. 1H NMR (CDCl3) d 1.51–1.67 (m,
6H, pip), 3.38 (t, 2H, J = 5.4 Hz, pip), 3.60 (t, 2H, J = 5.5 Hz, pip),
4.22 (d, 2H, J = 4.2 Hz,
aCH2), 5.16 (s, 2H, 3-OCH2Ph), 5.17 (s, 2H,
4-OCH2Ph), 6.98 (d, 1H, J = 5.4 Hz, 5-H), 7.36–7.70 (m, 10H, 3-
OCH2Ph and 4-OCH2Ph), 8.25 (d, 1H, J = 5.3 Hz, 6-H), 8.59 (br s,
1H, -CONH-); m/z (ESI): 459.1.
aCH2), 5.14 (s, 2H, 3-OCH2Ph), 5.18 (s, 2H, 4-OCH2Ph), 6.39 (br s,
1H), 7.01 (d, 1H, J = 5.4 Hz, 5-H), 7.28–7.45 (m, 10H, 3-OCH2Ph
and 4-OCH2Ph), 8.22 (d, 1H, J = 5.4 Hz, 6-H); m/z (ESI): 405.3.
Compound 12b: Yield 45%. 1H NMR (CDCl3) d 0.87 (d, 6H,
J = 6.7 Hz, CH2CH(CH3)2) 1.75 (m, 1H, J = 6.7 Hz, CH2CH(CH3)2), 3.06
4.1.13. Preparation of compound 14
A solution of 12a in methyl iodide is stirred overnight under re-
flux condition. After cooling, ethyl acetate is added to the mixture.
The white precipitate formed is filtered off the solution and recrys-
tallised from methanol/diethylether to afford 14 as white crystals.
Yield 92.5%. 1H NMR (CD3OD) d 2.75 (s, 3H, NHCH3), 4.04 (s, 2H,
(t, 2H, J = 6.6 Hz, CH2CH(CH3)2), 4.07 (d, 2H, J = 5.9 Hz, aCH2), 5.14
(s, 2H, 3-OCH2Ph), 5.18 (s, 2H, 4-OCH2Ph), 6.48 (br s, 1H), 7.01 (d,
1H, J = 5.4 Hz, 5-H), 7.28–7.45 (m, 10H, 3-OCH2Ph and 4-OCH2Ph),
8.23 (d, 1H, J = 5.4 Hz, 6-H), 8.25 (m, 1H); m/z (ESI): 446.7.
a
CH2), 4.26 (s, 3H, RN+–CH3Iꢀ),5.18 (s, 2H, 3-OCH2Ph), 5.58 (s,
Compound 12c: Yield 75.6%. 1H NMR (CDCl3) d 1.40 (d, 3H,
2H, 4-OCH2Ph), 7.29–7.59 (m, 10H, 3-OCH2Ph and 4-OCH2Ph),
7.85 (d, 1H, J = 7.2 Hz, 5-H), 8.67 (d, 1H, J = 7.1 Hz, 6-H); m/z
(ESI): 420.3.
J = 7.1 Hz, –
a
CHCH3), 2.76 (d, 3H, J = 4.8 Hz, –NHCH3), 4.65 (q, 1H,
CHCH3), 5.12–5.18 (m, 4H, 3-OCH2Ph and 4-OCH2Ph),
J = 7.1 Hz, –
a
6.54 (br s, 1H, -NHCH3), 7.01 (d, 1H, J = 5.4 Hz, 5-H), 7.29–7.45 (m,
10H, 3-OCH2Ph and 4-OCH2Ph), 8.06 (br s, 1H, –CONH–), 8.24 (d,
1H, J = 5.4 Hz, 6-H); m/z (ESI): 419.0.
4.1.14. General procedure for preparation of compounds 13a–k
and 15
Compound 12d: Yield 60%. 1H NMR (CDCl3) d 0.85 (d, 6H,
This procedure is illustrated for compound 13a. A solution of
12a (270 mg, 0.643 mmol) in dry dichloromethane (mL) was
cooled to 0 °C before BCl3 (1 M dichloromethane solution, 2 mL,
1.97 mmol, 3 equiv) was slowly added. The reaction mixture was
left under stirring for 3 h. Then, methanol was slowly added, and
the solution was concentrated in vacuo. The following crystallisa-
tion from methanol/acetone afforded the desired compound in
the HCl salt form as a white amorphous powder. Yield 96%. 1H
NMR (CD3OD) d 2.79 (s, 3H, –NHCH3), 4.18 (s, 2H,-NHCH2CO-),
7.37 (d, 1H, J = 6.4 Hz, 5-H), 8.21 (d, 1H, J = 6.4 Hz, 6-H); m/z
(ESI): 226.00, 194.98, 167.07, 156.00.
J = 6.7 Hz, –CH2CH(CH3)2), 1.42 (d, 3H, J = 7.1 Hz,
1.69–1.79 (m, 1H, –CH2CH(CH3)2), 3.04 (m, 2H, –CH2CH(CH3)2),
4.67 (q, 1H, J = 7.1 Hz, – CHCH3), 5.12–5.16 (m, 4H, 3-OCH2Ph
–aCHCH3),
a
and 4-OCH2Ph), 6.72 (br s, 1H, –NHCH2CH(CH3)2), 6.99 (d, 1H,
J = 5.4 Hz, 5-H), 7.27–7.45 (m, 10H, 3-OCH2Ph and 4-OCH2Ph),
8.13 (br s, 1H, –CONH–), 8.22 (d, 1H, J = 5.4 Hz, 6-H); m/z (ESI):
461.4.
Compound 12e: Yield 55.3%. 1H NMR (CDCl3) d 4.02 (d, 2H,
J = 5.4 Hz,
aCH2), 4.33 (d, 2H, J = 5.8 Hz, NHCH2Ph), 5.04 (s, 2H,
3-OCH2Ph), 5.06 (s, 2H, 4-OCH2Ph), 6.88 (d, 1H, J = 5.3 Hz, 5-H),
7.13–7.37 (m, 16H, 3-OCH2Ph, 4-OCH2Ph, NH), 8.09 (d, 1H,
J = 5.3 Hz, 6-H), 8.36 (t, 1H, J = 5.4 Hz, NH); m/z (ESI): 481.0.
Compound 12f: Yield 48.4%. 1H NMR (CDCl3) d 1.43 (d, 3H,
Compound 13b: Yield 93%. 1H NMR (CD3OD) d 0.94 (d, 6H,
J = 6.7 Hz, –CH2CH(CH3)2), 1.82 (m, 1H, J = 6.7, 6.9 Hz,
-
CH2CH(CH3)2), 3.07 (d, 2H, J = 6.9 Hz, –CH2CH(CH3)2), 4.21 (s, 2H,
–NHCH2CO–), 7.34 (d, 1H, J = 6.4 Hz, 5-H), 8.18 (d, 1H, J = 6.4 Hz,
6-H); m/z (ESI): 268.07, 195.00, 167.07.
J = 6.9 Hz, –
(m, 1H, –
4-OCH2Ph), 6.97 (d, 1H, J = 5.3 Hz, 5-H), 7.08 (br s, 1H, -NHBn),
aCHCH3), 4.34–4.44 (m, 2H, –NHCH2Ph), 4.68–4.76
aCHCH3), 5.03–5.11 (m,2H, 3-OCH2Ph), 5.15 (s, 2H,

A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
1295
Compound 13c: Yield 96.5%. 1H NMR (CD3OD) d 1.50 (d, 3H,
J = 6.9 Hz, CH(CH3)), 2.79 (s, 3H, -NHCH3), 4.64 (q, 1H,
4.3. Log D Determination and log P Estimation by HPLC
–
a
J = 6.9 Hz, –NHCH(CH3)CO), 7.37 (d, 1H, J = 6.4 Hz, 5-H), 8.20 (d,
1H, J = 6.4 Hz, 6-H); m/z (ESI): 240.07, 209.00, 181.07, 156.07.
Compound 13d: Yield 96%. 1H NMR (CD3OD) d 0.94 (d, 6H,
The log D values of compounds 13 a–l were determined by the
shake-flask method using n-octanol-aqueous buffer at 25 °C. A
HPLC method was also adopted. Shake-flask method. The lipophicil-
ity, expressed as distribution coefficient at physiological pH 7.4,
was measured using n-octanol and MOPS (3-N-morpholinopro-
panesulfonic acid 0.1 mM) buffer. The absorbance of a solution of
known concentration (0.1 mM) of the test compound in the buffer
J = 6.7 Hz, –CH2CH(CH3)2), 1.52 (d, 3H, J = 6.9 Hz, –
(m, 1H, J = 6.7, 6.9 Hz, -CH2CH(CH3)2), 3.07 (d, 2H, J = 6.9 Hz, –
CH2CH(CH3)2), 4.63–4.69 (m, 1H, CHCH3), 7.30 (d, 1H,
aCHCH3), 1.82
–a
J = 6.2 Hz, 5-H), 8.16 (d, 1H, J = 6.2 Hz, 6-H); m/z (ESI): 282.07,
209.00, 181.00, 156.07.
was recorded spectrophotometrically (A289).
A given volume
Compound 13e: Yield 95%. 1H NMR (CD3OD) d 4.23 (s, 2H, –
NHCH2CO–), 4.42 (s, 2H, -NHCH2Ph), 7.29–7.34 (m, 6H, –NHCH2Ph),
8.17 (d, 1H, J = 6.4 Hz, 6-H); m/z (ESI): 302.40, 195.03, 167.07,
156.07.
(5 mL) of this solution was then shaken with the same volume of
n-octanol by vigorous stirring for 1 h. The two layers were sepa-
rated by centrifugation and the absorbance of the aqueous phase
was measured again. The decrease in absorbance of the aqueous
phase, due to partitionon into the organic phase, indicated the lipo-
phicility, which was then calculated and expressed as log D from
the ratio of the drug concentration in n-octanol and the drug con-
centration in the MOPS buffer, using the following equation:
Compound 13f: Yield 98%. 1H NMR (CD3OD) d 1.43 (d, 3H,
J = 7.0 Hz,
–
aCHCH3), 4.32 (s, 2H, –NHCH2Ph), 4.59 (q, 1H,
J = 7.0 Hz, –
aCHCH3), 7.11–7.43 (m, 6H, -CH2Ph and 5-H), 8.06 (d,
1H, J = 6.4 Hz, 6-H); m/z (ESI): 316.17, 208.94, 181.00, 156.00.
Compound 13g: Yield 97.5%. 1H NMR (CD3OD) d 2.72 (s, 3H, -
NHCH3), 3.14 (dd, 1H, Jgem = 13.7 Hz, Jvic = 7.2 Hz, -CH2Ph), 3.22
(dd, 1H, Jgem = 13.7 Hz, Jvic = 6.1 Hz, -CH2Ph), 4.84 (t, 1H, J = 6.9 Hz,
D ¼ ½Drugꢂbuffer=½Drugꢂoctanol
[Drug]buffer = A1/A0 ꢁ 0.1 and [Drug]octanol = 0.1ꢀ[Drug]buffer
,
–aCHCH2Ph), 7.22–7.33 (m, 6H, -CH2Ph and 5-H), 8.15 (d, 1H,
where A0 is the initial absorbance of the aqueous phase and A1 is
the absorbance of the aqueous phase after partition with n-octanol.
HPLC method. A Hewlet-Packard model 1090 M Series HPLC sys-
tem with an autoinjector, autosampler, diode array detector and a
reverse phase polymer HPLC column (PLRP-S 100 Å, 23 ꢁ 0.46 cm,
J = 6.4 Hz, 6-H); m/z (ESI): 316.07, 284.98, 257.04, 120.06.
Compound 13h: Yield 96.5%. 1H NMR (CD3OD) d 0.73 (d, 3H,
J = 5.1 Hz, –CH2CH(CH3)2), 0.74 (d, 3H, J = 5.1 Hz, –CH2CH(CH3)2),
1.56–1.67 (m, 1H, –CH2CH(CH3)2), 2.83 (dd, 1H, Jgem = 13.2 Hz,
Jvic = 7.1 Hz, –CH2Ph), 3.93 (dd, 1H, Jgem = 13.2 Hz, Jvic = 6.8 Hz, –
CH2Ph), 3.00–3.12 (m, 2H, –CH2CH(CH3)2), 4.77 (t, 1H, J = 6.9 Hz, -
5 lm) was used in the study. The mobile phase consisted of ion
pair buffer (5 mM 1-heptanesulfonic acid, sodium salt; pH adjusted
to 2 using HCl), which was chosen to minimise any effect derived
from the ionisation state of the analysed compounds, and acetoni-
trile. The gradient used was 2–35% acetonitrile in 20 min, the flow
rate was 1 ml/min and the analytes were e monitored at 280 nm.
Each sample was run in triplicate. The partition coefficient was cal-
culated from retention data by regression analysis using a small li-
brary of previously studied HPOs as standards.33
aCHCH2Ph), 7.11–7.22 (m, 4H), 7.45–7.57 (m, 2H), 8.05 (d, 1H,
J = 6.4 Hz, 6-H); m/z (ESI): 358.13, 285.00, 257.07, 120.07.
Compound 13i: Yield 97%. 1H NMR (CD3OD) d 3.13–3.25 (m, 2H,
aCHCH2Ph), 4.37 (dd, 2H, J = 14.9 Hz, –NHCH2Ph), 4.85 (m, 1H,
aCHCH2Ph), 7.07 (d, 1H, J = 6.4 Hz, 5-H), 7.18–7.31 (m, 10H, -
aCHCH2Ph and -NHCH2Ph), 7.98 (d, 1H, J = 6.4 Hz, 6-H); m/z (ESI):
392.13, 285.02, 257.07, 120.07, 103.07.
Compound 13j: Yield 96%. 1H NMR (CD3OD) d 3.02 (s, 3H, –
N(CH3)2), 3.10 (s, 3H, -N(CH3)2), 4.41 (s, 2H,-NHCH2CO-), 7.33 (d,
1H, J = 6.4 Hz, 5-H), 8.18 (d, 1H, J = 6.4 Hz, 6-H)); m/z (ESI): 239.2.
Compound 13k: Yield 97%. 1H NMR (CD3OD) d 1.60–1.73 (m,
6H, pip), 3.49 (t, 2H, J = 5.2 Hz, pip), 3.61 (t, 2H, J = 5.5 Hz, pip),
4.41 (s, 2H,-NHCH2CO-), 7.30 (d, 1H, J = 6.2 Hz, 5-H), 8.15 (d, 1H,
J = 6.3 Hz, 6-H)); m/z (ESI): 278.9.
4.4. Ligand pKa values and stability constant of iron(III)
complexes
Iron chloride (17.906 mM in 1% HCl, atomic absorption stan-
dard, Aldrich) was utilized in this study. Analytical grade reagent
HCl (37%) and KOH (10 M) ampoules (Fisher), HPLC grade water
and methanol (Fisher) were used in the preparation of all solutions.
The automatic titration system used in this study comprises of an
autoburette (Metrohm Dosimat 765, l mL syringe) and Mettler
Toledo MP230 pH meter with Metrohm AgCl electrode
(6.0133.100) and a reference electrode (6.0733.100). 0.1 M KCl
electrolyte solution was used to maintain the ionic strength. The
temperature of the test solutions was maintained at 25 0.1 °C
by a Techne TE-8 J temperature controller. The solution under
investigation was stirred vigorously. A Gilson mini-plus3 pump
with speed capability (20 mL/min) was used to circulate the test
solution into a Hellem quartz flow cuvette. For stability constant
determinations a 50 mm path length cuvette was used and for
pKa determinations a cuvette path length of 10 mm was used.
Automatic titration and spectral scans adopted the following strat-
egy: increase or decrease the pH value of the solution by 0.1 pH
unit by the addition of KOH or HCl from the autoburette. When
pH readings varied by <0.001 pH unit over a 3 s period, the solution
was allowed to reach equilibrium (for stability constant determi-
nations a period of 5 minutes was adopted, for pKa determinations
a period of 1 minute was adopted). The spectrum of the solution
was then recorded. The cycle was then repeated automatically
until the defined end point pH value was achieved. Following
electrode calibration, samples with an appropriate concentration
to give measurable absorbance were titrated in pH range 2–12
Compound 15: Yield 96%. 1H NMR (CD3OD) d 2.81 (s, 3H, -
NHCH3), 4.11 (s, 2H,-NHCH2CO-), 4.15 (s, 3H,-RNCH3) 7.22 (d, 1H,
J = 6.8 Hz, 5-H), 8.23 (d, 1H, J = 6.8 Hz, 6-H)); m/z (ESI): 240.7.
4.2. Crystallography
SinglecrystalX-raydiffractiondataof 13fwascollectedona BRU-
KER AXS SMART 6000 diffraction system at 293 K using Cu Ka radi-
ation (k = 1.54178 Å). The structure was solved by direct methods
using the program system SHELXS 97 [X]. All non-hydrogen atoms
were taken from a series of full-matrix least-squares refinement cy-
cles based on F2 using the program SHELXL 97) followed by difference
Fourier syntheses. All hydrogen atoms were placed on calculated
positions and were allowed to ride on their corresponding carbon
atoms. The isotropic thermal parameters for the methyl protons
were set to 1.5 times and for all other hydrogen atoms to 1.2 times
of the Ueq value of the bonding atom. All non-hydrogen atoms were
refined anisotropically. Experimental details of the X-ray diffraction
studies are listed in Table 3. CCDC676400 contains the supplemen-
tary crystallographic data for this paper. This data can be obtained
free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html
or from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44
1223 336033; email: deposit@ccdc.cam.ac.uk.

1296
A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
for pKa determination. For affinity studies, iron ligand ratio was
kept at 5:1. Insoluble iron(III) complexes were measured in 1:1
methanol/water mixtures, the results were converted to those cor-
responding to aqueous conditions. The titration data were ana-
lyzed by pHab.34,35
acid trisodium salt (Na3NTA) (1.9 g/100 ml, 100 mM) at pH 7. The
required volume of atomic absorption iron standard solution was
added to give a 5:1 molar ratio. A fresh stock solution of 3 mM
FeNTA was prepared for each experiment (75
lL NTA, 85 lL iron
and 340 L H2O), filtered and incubated for 15 min before use, in
l
Affinity is expressed as pM value pM is defined as the negative
logarithm of the metal ion concentration under the following con-
ditions: [Metal Ion]total = 10ꢀ6 M, [Ligand]total = 10ꢀ5 M at pH 7.4.
order to ensure that the compound was in the iron (III) oxidation
state before application to cell cultures. The stock solution of the
drug was prepared in sterile 100% dimethylsulphoxide (DMSO)
and stored at -20 °C until use. Final test concentrations were
obtained by diluting in the neuronal culture medium (DMEM-
F12) giving a final concentration of 1% DMSO. Human Ab[1–40]
(Cat No. 641-10; Lot: ML0810) was purchased from California
Peptides Research (Napa, CA). Ab was dissolved in DMSO before
addition to the cell cultures. Cultured cortical neurons in 24 well
4.5. Lipoxygenase and tyrosine hydroxylase inhibition
Lipoxygenase. Isolated soybean lipoxygenase (sbLPO) and its
preferential substrate linoleic acid were both purchased from
Sigma. The inhibition of the enzyme was monitored by a change
in the absorption of the product linoleic hydroperoxide at
plates were treated with different concentrations of FeNTA (3
or 10 M) or Ab (3, 10 or 20 M) for 6 h or 30 min, respectively
at 37 °C. Following this, the cells were treated with the different
chelators (10 M, 30 M or 100 M) or vehicle control. Cells were
lM
234 nm. The enzyme (1
was pre-incubated with the test compound (100
tration in the cuvette of 10 M)) for 15 min, and the reaction was
l
M solution in borate buffer 0.2 M, pH 9)
l
l
lM, final concen-
l
l
l
l
initiated by the addition of linoleic acid (1 mM). The reaction was
monitored at 234 nm for 10 min using a time based measurement.
The rates of reactions were determined using the molar absorption
subsequently incubated for a further 24 h before being assessed for
viability or other measures. All experiments were performed in
triplicate.
coefficient of hydroperoxide (
e
= 23600 Mꢀ1 cmꢀ1). None of the
M.
synthesised compounds showed inhibitory activity at 10
l
4.7. Cytotoxicity measurements
Tyrosine hydroxylase. Rat striata homogenate (1:9 v/v ratio in
0.32 M sucrose) was used as source for the enzyme. The reaction
medium contained striatal homogenate (equivalent to approxi-
mately 3 mg of tissue), potassium phosphate buffer (0.5 M, pH
6.0), m-hydroxybenzylhydrazine (DOPA decarboxylase inhibitor),
4.7.1. Lactate dehydrogenase (LDH)
Total LDH release into the culture medium by dead and dying
cells (CytoTox-96 LDH assay, Promega, Southampton, UK) was
measured. Cells were incubated untreated with 0.1% Triton X-100
for 10 min (37 °C, 5% CO2, 95% air) to induce maximal cell lyses.
Absorbance was measured at 490 nm using a Versemax plate read-
er. Treatment values were then expressed as a percentage of the
total LDH release. Background LDH release (media alone) was sub-
tracted from the experimental values. All data from LDH measure-
ments are expressed as % cytotoxicity, where 100% represents
maximum toxicity induced (n = 3 from quadruplicates).
6,7-dimethyl-5,6,7,8-tetrahydropterine
1 mM in mercaptoethanol and the test iron chelator at different
concentrations (100, 10 and 1 M). The reaction was initiated by
the addition of -tyrosine 2 mM (final concentration of 0.4 mM),
followed by 20 min incubation at 37 °C. The reaction was termi-
nated by the addition of 100 L perchloric acid 0.4 M containing
-methyldopa as internal standard and the mixture was cen-
trifuged for 10 min. -DOPA and -methyldopa were extracted
onto acid washed alumina (140–150 mg) from 150 L of the resul-
(enzymatic
cofactor)
l
L
l
5 lM a
L
a
l
4.7.2. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium-
bromide (MTT) turnover
tant supernatant. Tris buffer (pH 8.6, 1 mL) was added and the mix-
ture was shaken for 10 min. The supernatant was aspirated and the
alumina residue washed twice with water. After the second wash,
Cell viability was assessed by measuring the levels of MTT,
which is a marker of mitochondrial activity.37 Following exposure
of neuronal cultures to Fe-NTA and/or the tested compound, the
cultures were washed twice with sterile PBS before the addition
of MTT (1 mg/ml) in HEPES-buffered incubation medium (HBM)
pH 7.4 (5 mM HEPES, 154 mM NaCl, 4.6 mM KCl, 2.3 mM CaCl2,
33 mM glucose, 5 mM NaHCO3, 1.1 mM MgCl2, 1.2 mM Na2HPO4).
Following incubation (45 min; 37 °C), MTT solutions were re-
moved, and the formazan product was solubilized in DMSO, and
the absorbance recorded at 505 nm using a Versemax plate reader.
Data are presented as a percentage relative to their vehicle con-
trols. All data from MTT measurements are expressed as Formazan
blue production % of control, considered as 100%, maximum forma-
zan blue formation, (n = 3 from quadruplicates).
200
10 min to elute
resultant supernatant was used for HPLC analysis to determine
the concentration of -DOPA and -methyldopa. The assay was
performed in triplicate for each chelator. Non–enzymatic forma-
tion of -DOPA was determined by the addition of -tyrosine and
3-iodotyrosine (tyrosine hydroxylase inhibitor) instead of -tyro-
lL of perchloric acid (0.4 M) was added shaking vigorously for
L
-DOPA and -methyldopa from the alumina. The
a
L
a
L
D
L
sine. For the HPLC, the mobile phase was 0.1 M sodium phosphate
(pH 3.0) containing EDTA (1 mM), octane sulphonic acid (0.65 mM)
and 5% methanol. The flow rate was 1 mL/min and the column was
a Spherisorb ODS2 (reverse phase chromatography). Quantitation
was performed using peak height ratio with the internal standard
a-methyldopa.
4.7.3. Microscopic examination
4.6. Neuroprotection study in cultured mouse cortical neurones
All cultures were examined by phase contrast microscopy
(400ꢁ magnification Nikon Inverted Eclipse T) to assess cell body
and neurite morphology. Representative images were captured
using a digital camera (Nikon, Coolpix).
Culture reagents were obtained from Gibco-Invitrogen (UK) and
all others chemicals were obtained from Sigma-Aldrich (UK) unless
otherwise stated. Primary cultures of cortical neurons were
prepared from NIH Swiss white mouse embryos at E15-16 (Harlan,
UK). Animal procedures were in accordance with the Home Office,
Animals Scientific Procedures Act (1986, UK). Primary cultures of
cortical neurons were prepared from E15 mouse embryos on a
weekly basis as described previously.36,28 FeNTA was prepared by
mixing a solution of nitrilotriacetic acid disodium salt (Na2NTA)
(1.91 g/100 ml, 100 mM) and a solution of nitrilotriaceticacetic
4.7.4. Synaptophysin measurements
Synaptophysin levels were measured in neuronal lysates by
immunoblotting. Cell lysates (20 lg) were run on 10% SDS Poly-
acrylamide gels, transferred to nitrocellulose membranes by
semi-dry electroblotting and probed overnight with a polyclonal
anti-synaptophysin antibody (Synaptic Systems, 1/2000 dilution).
Membranes were incubated with secondary antibody (Chemicon)

A. Gaeta et al. / Bioorg. Med. Chem. 19 (2011) 1285–1297
1297
10. Ben-Shachar, D.; Kahana, N.; Kampel, V.; Warshawsky, A.; Youdim, M. B. H.
Neuropharmacology 2004, 46, 254–263.
11. Boddaert, N.; Le Quan Sang, K. H.; Rötlg, A.; Leroy-Willig, A.; Gallet, S.; Brunelle,
F.; Sidi, D.; Thalabard, J. C.; Munnich, A.; Cabantchik, Z. I. Blood 2008, 110, 401–
408.
12. Kakhlon, O.; Manning, H.; Breuer, W.; Melamed-Book, N.; Lu, C.; Cortopassi, G.;
Munnich, A.; Cabantchik, Z. I. Blood 2008, 112, 5219–5227.
13. Maggio, A.; D’Amico, G.; Morabito, A.; Capra, M.; Ciaccio, C.; Cianciulli, P.; Di
Gregorio, F.; Garozzo, G.; Malizia, R.; Magnano, C., et al Blood Cell Mol. Dis. 2002,
28, 196–208.
14. Olivieri, N.; Koren, G.; Hermann, C.; Bentur, Y.; Chung, D.; Klein, J.; Louis, S. T.
P.; Freedman, M. H.; Mcclelland, R. A.; Templeton, D. M. Lancet 1990, 336,
1275–1279.
15. Habgood, M. D.; Liu, Z. D.; Dehkordi, L. S.; Khodr, H. H.; Abbott, J.; Hider, R. C.
Biochem. Pharmacol. 1999, 57, 1305–1310.
16. Molina-Holgado, F.; Gaeta, A.; Francis, P.; Williams, R. J.; Hider, R. C. J.
Neurochem. 2008, 105, 2466–2476.
17. Molina-Holgado, F.; Hider, R. C.; Gaeta, A.; Williams, R.; Francis, P. Biometals
2007, 20, 639–654.
18. Hider, R. C.; Ma, Y.; Molina-Holgado, F.; Gaeta, A.; Roy, S. Biochem. Soc. Trans.
2008, 36, 1304–1308.
for 45 min, washed and protein bands detected using an ECL detec-
tion system with Hyperfilm ECL (GE Healthcare). Films were devel-
oped using X-ray imaging equipment Mediphot 837 (Fuji,
Germany) and quantified using Image J (NIH) .
4.7.5. Propidium iodide (PI) and HOE 33342 (HOE) staining
Propidium iodide and HOE 33342 (2’-(ethoxyphenyl)-5-(4-
methyl-1-piperazinyl)-2’,5’-bi-1H-benzimidazole HCl) were both
purchased from Calbiochem (Nottingham, UK). HOE 33342
(10 lM) and PI (10 lM) (t=15 minutes) were used to stain viable
and dead cells, respectively. HOE 33342 binds DNA in viable cells
and it is visualised as a blue colour. PI is a membrane impermeable
DNA intercalator, which is only able to permeate dying cells.
Therefore, it is useful for staining apoptotic or necrotic cells. Viable
and dead cells were quantified using an IN1000 Cell Analyser (GE
Healthcare, UK).
19. Piyamogkol, S.; Liu, Z. D.; Hider, R. C. Tetrahedron 2001, 57, 3479–3486.
20. Raymond, K. N.; Müller, G.; Matzanke, B. F. Top Curr. Chem. 1984, 58, 49–102.
21. Hider, R. C.; Singh, S.; Porter, J. B. Proc. R. Soc. Edinburgh 1992, 99B(1/2), 137–168.
22. Abeysinghe, R. D.; Roberts, P. J.; Cooper, C. E.; MacLean, K. H.; Hider, R. C.;
Porter, J. B. Journal. Biol. Chem. 1996, 271, 7965–7972.
23. Liu, Z. D.; Kayyali, R.; Hider, R. C.; Porter, J. B.; Theobald, A. E. Journal. Med.
Chem. 2002, 45, 631–639.
24. Hider, R. C. Toxicol. Lett. 1995, 82–83, 961–967.
Acknowledgments
The authors thank Professor Bernt Krebs and Dr. Michael Klos-
kowski for provision of diffractometer access and X-ray structure
determination of compound 13f.
25. Liu, Z. D.; Lockwood, M.; Rose, S.; Theobald, A. E.; Hider, R. C. Biochem.
Pharmacol. 2001, 61, 285–290.
26. Wagner, K. R.; Sharp, F. R.; Ardizzone, T. D.; Lu, A.; Clark, J. F. J. Cereb. Blood Flow
Metab. 2003, 23, 629–652.
27. Schilling, K.; Barco, E. B.; Rhinehart, D.; Pilgrim, C. J. Neurosci. Res. 1989, 24,
347–354.
28. Chen, C. X.-Q.; Huang, S. Y.; Zhang, L.; Liu, Y.-J. Neurobiol. Dis. 2005, 19, 419–426.
29. Whitnall, M.; Richardson, D. R. Semin. Pediatr. Neurol. 2006, 13, 186–197.
30. Selkoe, D. J. Ann. N.Y. Acad. Sci 2000, 924, 17–25.
31. Ong, W. Y.; Halliwell, B. Ann. N.Y. Acad. Sci. 2004, 1012, 51–64.
32. Berg, D.; Youdim, M. B. Top Magn. Reson. Imaging 2006, 17, 5–17.
33. Liu, D. Y.;Liu, Z. D.;Lu, S. L.;Hider, R. C. Journal. Chromatogr., B1999, 730, 135–139.
34. Gans, P.; Sabatini, A.; Vacca, A. Ann. Chim. 1999, 89, 45–49.
35. Gans, P.; O’Sullivan, B. Talanta 2000, 51, 33–37.
36. Crossthwaite, A. J.; Hasan, S.; Williams, R. J. J. Neurochem. 2002, 80, 24–35.
37. Carmichael, J.; DeGraff, W. G.; Gazdar, A. F.; Minna, J. D.; Mitchell, J. B. Cancer
Res. 1987, 47, 936–942.
References and notes
1. Strozyk, D; Bush, A. Neurodegenerative Diseases and Metal Ions: Metal Ions in
Life Sciences, Sigel, Sigel and Sigel, 2006, 1, 427–435.
2. Gaeta, A.; Hider, R. C. Br. J. Pharmacol. 2005, 146(8), 1041–1059.
3. Liu, Z. D.; Hider, R. C. Med. Res. Rev. 2002, 22, 26–64.
4. Cherny, R. A.; Atwood, C. S.; Xilinas, M. E.; Gray, D. N.; Jones, W. D.; Mclean, C. A.;
Barnham, K. J.; Volitakis, I.; Fraser, F. W.; Kim, Y. S. Neuron 2001, 30, 665–676.
5. Kaur, D.; Yantiri, F.; Rajagopalan, S.; Kumar, J.; Mo, J. Q.; Boonplueang, R.;
Viswanath, V.; Jacobs, R.; Yang, L., et al Neuron 2003, 37, 899–909.
6. Regland, B.; Lehmann, W.; Abedini, I.; Blennow, K.; Jonsson, M.; Karlsson, I.;
Sjögren, M.; Wallin, A.; Xilinas, M.; Gottfries, C. G. Dement. Geriatr. Cogn. Disord.
2001, 12, 408–414.
7. Bush, A. I. Trends Neurosci. 2003, 26, 207–214.
8. Oakley, G. P. JAMA 1973, 225, 395–397.
9. Amit, T.; Avramovich-Tirosh, Y.; Youdim, M. B. H.; Mandel, S. FASEB Journal.
2008, 22, 1296–1305.