Dibenzo[1,4,5]thiadiazepine: A hardly-known heterocyclic system with neuroprotective properties of potential usefulness in the treatment of neurodegenerative diseases

Article abstract of DOI:10.1016/j.ejmech.2014.04.075

In this work we describe a new family of dibenzo[1,4,5]thiadiazepines (2-12) that showed an interesting in vitro biological profile, namely neuroprotective and antioxidant properties, as well as blockade of cytosolic calcium entry. They showed no cytotoxic effects and the majority were predicted as CNS-permeable compounds. In human neuroblastoma cells they displayed good neuroprotective properties against mitochondrial oxidative stress which, in many cases, almost reached the full protection (>90%) when compounds were incubated with cells 24 h before the addition of toxic stressors. In co-incubation conditions these figures were smaller, although some compounds maintained an interesting level of neuroprotection, higher than 50%. Four selected compounds (2, 5, 8, and 11) were found to be effective antioxidant agents by sequestering mitochondrial radical oxygen species (ROS). Moreover, compound 2 showed a remarkable calcium-channel modulating activity. The interest of these compounds is increased by the fact that dibenzo[1,4,5]thiadiazepine is a barely known structure that is not difficult to synthesize and presents very few described derivatives, opening a new and broad line of research in Medicinal Chemistry.

Full text of DOI:10.1016/j.ejmech.2014.04.075

European Journal of Medicinal Chemistry 81 (2014) 350e358
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Dibenzo[1,4,5]thiadiazepine: A hardly-known heterocyclic system
with neuroprotective properties of potential usefulness in the
treatment of neurodegenerative diseases
,
1
Gema C. González-Muñoz a, Mariana P. Arce a, Concepción Pérez a, Alejandro Romero b
,
,
Mercedes Villarroya b, Manuela G. López b, Santiago Conde a
,
*
,
María Isabel Rodríguez-Franco a
*
a Instituto de Química Médica, Consejo Superior de Investigaciones Científicas (IQM-CSIC), C/ Juan de la Cierva 3, 28006 Madrid, Spain
b Instituto Teofilo Hernando and Departamento de Farmacología y Terapeutica, Facultad de Medicina, Universidad Autonoma de Madrid, C/Arzobispo
Morcillo 4, 28029 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
In this work we describe a new family of dibenzo[1,4,5]thiadiazepines (2e12) that showed an interesting
in vitro biological profile, namely neuroprotective and antioxidant properties, as well as blockade of
cytosolic calcium entry. They showed no cytotoxic effects and the majority were predicted as CNS-
permeable compounds. In human neuroblastoma cells they displayed good neuroprotective properties
against mitochondrial oxidative stress which, in many cases, almost reached the full protection (>90%)
when compounds were incubated with cells 24 h before the addition of toxic stressors. In co-incubation
conditions these figures were smaller, although some compounds maintained an interesting level of
neuroprotection, higher than 50%. Four selected compounds (2, 5, 8, and 11) were found to be effective
antioxidant agents by sequestering mitochondrial radical oxygen species (ROS). Moreover, compound 2
showed a remarkable calcium-channel modulating activity. The interest of these compounds is increased
by the fact that dibenzo[1,4,5]thiadiazepine is a barely known structure that is not difficult to synthesize
and presents very few described derivatives, opening a new and broad line of research in Medicinal
Chemistry.
Received 17 February 2014
Received in revised form
8 April 2014
Accepted 25 April 2014
Available online 12 May 2014
Keywords:
Dibenzo[1,4,5]thiadiazepines
Neuroprotection
Antioxidant
Mitochondrial oxidative stress
Calcium modulation
Alzheimer’s disease
Ó 2014 Elsevier Masson SAS. All rights reserved.
́
́
́
1. Introduction
of nine (11%) USA citizens age 65 and older [1]. The terrible
symptoms and long process of AD, the subsequent impact on the
caregivers and burden for the Public Health Systems and families
are also well-known.
The known figures of the prevalence of Alzheimer’s disease
(AD), the most extended senile dementia, are really impressive and
worrying. Although age is the highest risk factor for AD, about 5.2
million USA citizens of all ages are thought to suffer AD in 2013;
among them, approximately 4% are under 65 years old, 13% are 65e
74, 44% are 75e84 and 38% are 85 years and older. Namely, one out
AD is a multifunctional disease in which a number of physio-
logical processes change progressively into closely connected
pathological ones. The most striking hallmarks of these changes are
the occurrence of amyloid senile plaques, neurofibrillary tangles
and dysfunction of the cholinergic neurotransmission in the
framework of a massive neuronal loss, as well as other not so visible
events such as a progressive failure of the endogenous antioxidant
systems [2] and an imbalance in the cytosolic calcium concentra-
tions [3]. Thus, several neuroprotective strategies involving the
capture of mitochondrial free radicals [4,5] or the cytosolic calcium
modulation [6] have been proposed for the treatment of AD.
Continuously, new structures designed to act on one (standard
drugs) or more than one (multifunctional drugs) of the pathological
processes of AD appear published [7]. Of course, the ultimate goal
Abbreviations: AD, Alzheimer’s disease; AChE, acetylcholinesterase; BBB,
bloodebrain barrier; BuChE, butyrylcholinesterase; CNS, central nervous system;
DCFH-DA, 20,70-dichlorfluorescein-diacetate; HPLC-MS, high performance liquid
chromatographyemass spectrometry; LDH, lactate dehydrogenase; PAMPA, parallel
artificial membrane permeation assay; ROS, reactive oxygen species.
* Corresponding authors.
E-mail addresses: sconde@iqm.csic.es (S. Conde), isabelrguez@iqm.csic.es
(M.I. Rodríguez-Franco).
1
Present address: Departamento de Toxicología y Farmacología, Facultad de
Veterinaria, Universidad Complutense de Madrid, Avda. Puerta de Hierro s/n, 28040
Madrid, Spain.
http://dx.doi.org/10.1016/j.ejmech.2014.04.075
0223-5234/Ó 2014 Elsevier Masson SAS. All rights reserved.

G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
351
of all these works is to find out a new treatment to cure, or at least
alleviate, the disease. However, multi-target drugs may provide
more therapeutic advantages in the treatment of such a complex
pathology than single-target drugs [8e10]. Indeed, by applying the
multifunctional approach, in recent years we reported several
families of hybrid compounds that combine neuroprotective,
cholinergic, antioxidant and beta-amyloid reducing properties [11e
16]. Some of these hybrids have proved their usefulness in murine
models of AD and stroke [17e19].
Another rewarding strategy to discover new multifunctional
drugs is the biological evaluation of compounds proceeding from
in-house libraries towards different targets. Many groups of me-
dicinal chemistry synthesized lots of products that, in many cases in
the past, were evaluated in one activity and then, stored and
forgotten. Nowadays, the in vitro biological tests are widely
generalized and those in-house libraries are treasures of structures
frequently related to active compounds. Our group possesses a
broad in-house library of compounds which has already produced
interesting results in our line of research aimed to the study of new
drugs for neurodegenerative diseases [20e22].
Following this later approach, in a first exploratory work we
studied some biological properties of a group of compounds
selected from our library, five phenothiazines and a unique deriv-
ative of the very little-known heterocyclic system 1,4,5-
dibenzothiadiazepine [23]. The promising results obtained
encouraged us to develop two new families, taking a representative
product of each series as hit compounds. The development of the
first selected structure yielded a series of N-acylaminophenothia-
zine derivatives that displayed interesting properties as multi-
functional neuroprotective agents [24]. Phenothiazine is a well-
known heterocyclic system present in many commercial drugs
such as fluphenazine, chlorpromazine, promethazine (Phenergan)
and many others [25]. Phenothiazines are also useful as neurogenic
and neuroprotective agents [26] and as potential drugs for the
treatment of Creutzfeldt-Jakob and other prion diseases [27].
The second selected hit from our initial work was 1,4,5-(4-
chloro-5,6-dihydro-5,6-diacetyl)dibenzo[b,f]thiadiazepine 1 as a
possible bioisostere of the phenothiazine system, which showed an
interesting multifunctional profile: neuroprotection in the human
neuroblastoma cell line SH-SY5Y against oxidative stress produced
by free ROS, both of exogenous and mitochondrial origin; modu-
lation of the voltage-dependent calcium channels; not cytotoxic
effects per se; and ability to cross the bloodebrain barrier (BBB)
[23]. Compound 1 and other 1,4,5-dibenzothiadiazepine derivatives
were described in 1982 by Corral et al., from our Medicinal
Chemistry Institute [28], but they were never biologically tested
until we started our screening program.
derivatives: 2, 3 (2-Cl and 3-Cl respectively) and 4 (2-Cl basic het-
erocycle) in our library of compounds, all of them synthesized by
the Corral’s group [28] (Fig. 1). Their purity was checked by HPLC-
MS and their structures confirmed by 1H, 13C NMR and MS. Com-
pounds 2 and 3 were purified by flash column chromatography (see
Experimental Section), yielding two crystalline solids which
showed purities higher than 95% (HPLC-MS). The stored product 4
was pure enough (97%) to be used without any new purification.
With the aim to extend this initial set of compounds, new
dibenzo[1,4,5]thiadiazepines were obtained (5e12), following the
synthetic method formerly described [28]. Hydrazines were
diacylated to the hydrazides 13e15, which then yielded the
dibenzothiadiazepines 5 (60%) and 6 (40%) by cyclization in a base-
catalyzed Smiles rearrangement. All the hydrazines used in this
work, except the trifluoromethylated derivative (see Experimental
Part) had previously been obtained as common starting products
for the synthesis of N-acylaminophenothiazines [24]. The mono-
acylated product 7 (50%) was obtained instead of the expected
diacylated analog when the substituent was a nitro group. Yields
were not optimized because our chemical objective was simply to
obtain enough amounts of pure products to carry out their bio-
logical evaluation.
The same procedure, although a little more complicated, was
employed to obtain the derivatives 8e12 (Scheme 1). 2-Nitro-5-
and 2-nitro-6-chloro-20-hydrazinodiphenyl sulfides were acylated
with bromoacetyl and 2-bromopropionyl chloride to generate the
hydrazides 16e18 which were treated with selected secondary
amines to yield the new hydrazides 21e23. There was no need of
obtaining 16 and 17 because hydrazides 19 and 20 were also stored
in our in-house library, in enough purity and amount to be used as
synthetic intermediates. Finally, these compounds were subjected
to cyclization in the same conditions that the previous derivatives,
affording the expected dibenzothiadiazepines 8e12.
Compounds 2e12 were initially tested as acetylcholinesterase
(AChE) and butyrylcholinesterase (BuChE) inhibitors, following the
well-known Ellman’s method and using tacrine as a reference [29].
Enzymes of mammalian origin were used, namely AChE from
bovine erythrocytes and BuChE from horse serum, due to their high
degree of sequence identity to the human proteins [30] and their
lower cost. As it could be foreseen from the results previously ob-
tained from 1 [23], none of the compounds inhibited AChE (data
not shown) and only 9 and 10 showed a sub-micromolar inhibition
of BuChE (IC50 ¼ 0.9 and 0.7
mM respectively) (Table 1).
A basic condition of any compound to act on neurodegenerative
processes is to penetrate into the brain, that is, to be able of crossing
the bloodebrain barrier (BBB). To examine the capability of our
compounds to pass this barrier, we used a parallel artificial mem-
brane model [31]. This is a fairly easy and successful method to
predict the passive CNS permeation that our group had previously
optimized in order to be applied to investigational compounds with
limited aqueous solubility and that was employed to study many
structurally different families [11,13e16,24]. The permeabilities of
compounds 2e12 through a lipid extract of porcine brain were
determined using a mixture 70:30 of phosphate buffered saline
solution and ethanol (PBS:EtOH) (Table 1). In each experiment 10
commercial drugs were also evaluated for assay validation. The
graphic representation of experimental permeabilies vs. reported
values of such well-known drugs gave a lineal correlation, Pe
(exptl) ¼ 0.57 Pe (bibl) þ 2.36 (R2 ¼ 0.919). From this equation and
taking into account the described limits for BBB permeation, we
established that compounds with permeability values above 4.6
10ꢀ6 cm sꢀ1 could penetrate into the CNS by passive diffusion
(CNSþ), whereas products with Pe below 3.5 10ꢀ6 cm sꢀ1 could not
enter (CNSꢀ). Between these values, the CNS permeation was
considered as uncertain (CNSþ/ꢀ). As shown in Table 1, with the
In addition and as far as we know, biological properties of 1,4,5-
dibenzothiadiazepine derivatives were absolutely unexplored and
even from a chemical point of view, it is a hardly-known hetero-
cyclic system.
Thus, the structural novelty and the promising results founded
in the 1,4,5-dibenzothiadiazepine 1 prompted us to carry out a
more extensive study of the pharmacological properties of this
family of compounds. In this work, we describe a series of CNS-
permeable 1,4,5-dibenzothiadiazepines that showed neuro-
protective properties against mitochondrial oxidative stress, as well
as calcium modulatory and antioxidant properties. Some com-
pounds were taken from our library and others have been newly
synthesized.
2. Results and discussion
In addition to the original 1,4,5-(4-chloro-5,6-dihydro-5,6-
diacetyl)dibenzo[b,f]thiadiazepine 1, there were three more

352
G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
Fig. 1. Dibenzo[1,4,5]thiadiazepines studied in this work. Compounds 1e4 were taken from our library and 5e12 have been newly synthesized.
Scheme 1. Synthesis of the new dibenzo[1,4,5]thiadiazepines 5e12.
exception of compounds 6, 7 and 12 that could have some diffi-
culties to penetrate into the brain, the rest of dibenzo[1,4,5]thia-
diazepines would be able to cross BBB and reach their therapeutic
targets.
amount of lactate dehydrogenase (LDH), an enzyme that is released
to the extracellular medium when neurons die. To calculate the
percentage of neuroprotection the basal liberation of LDH was
subtracted from the value obtained in the presence of each com-
pound. Trolox, the vitamin E antioxidant substructure, was used as
a positive control and the results are shown in Table 2.
Firstly, possible cytotoxic effects of 2e12 were studied by
exposing cells to compounds at the highest concentration used in
the neuroprotection studies (3 mM) for 24 h. In all cases, LDH per-
centage was roughly the same as the basal value (data not shown),
suggesting that cell viability reached 100% and that dibenzo[1,4,5]
thiadiazepines 2e12 are non-toxic products at the highest tested
concentration.
All dibenzo[1,4,5]thiadiazepines 2e12 protected cells from the
damage induced by the mixture of rotenone and oligomycin A,
displaying very high percentages of mitochondrial protection that,
in many cases, exceeded the 90% and almost reached the 100% in
pre-incubation conditions. However, in the co-incubation experi-
ments these figures were smaller, between 10 and 60%. These re-
sults suggested that the neuroprotective properties of dibenzo
[1,4,5]thiadiazepines 2e12 are due to a mixed mechanism of action,
in which clearly predominates the activation of endogenous anti-
oxidant pathways vs. the direct capture of radical oxygen species
(ROS). However, in the case of trolox its principal mechanism seems
To explore the therapeutic potential of dibenzo[1,4,5]thiadia-
zepines 2e12 their cell viability and neuroprotective capacity
against mitochondrial oxidative stress were evaluated using the
human neuroblastoma cell line SH-SY5Y and the mixture of rote-
none and oligomycin A as the toxic insult. The combination of such
toxics induces mitochondrial ROS as a consequence of the blockade
of complex I and V of the mitochondrial electronic chain, being a
good model of endogenous oxidative stress [32]. Two different
protocols were used: (i) Pre-incubation, wherein cells were incu-
bated with each compound for 24 h before the addition of rotenone
plus oligomycin A, and then maintained for an additional 24 h
period in the presence of the mixture of toxics. This protocol is
useful to explore the possible cytoprotective effect due to the
activation of endogenous antioxidant pathways. (ii) Co-incubation,
wherein the compound and the combination rotenone plus oligo-
mycin A were added at the same time and incubated for 24 h. In this
case, a cytoprotective effect would indicate that the compound
could be acting as a free radical scavenger [33]. In both protocols,
compounds were tested at three concentrations (0.3, 1, and 3
mM)
and the percentage of cell death was determined by measuring the

G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
353
Table 1
(4
m
M) for 1 h, incubated in the presence of compounds 2, 5, 8,
Inhibition of BuChEa and prediction of the bloodebrain penetration of dibenzo[1,4,5]
thiadiazepines 2e12.
and 11 (3
mM) for 10 min and then stimulated with a concentrated
solution of potassium chloride, so that the final Kþ concentration in
the medium was 70 mM. The L-type Ca2þ-channel antagonist
nifedipine was also assayed at the same concentration as a positive
control, giving 48% inhibition of Kþ-evoked cytosolic calcium in-
crease (Table 3). Whereas the blockade of calcium entry induced by
compound 8 was under 10%, compounds 5 and 11 showed slightly
better results, around 15%. In this assay, our best compound was the
diacetylated dibenzo[1,4,5]thiadiazepine 2, which modulated the
cytosolic calcium by blocking the entry of this cation by 22%, about
a half of the nifedipine value.
Comp.
BuChE
IC50
PAMPA-BBB assay
b
(
m
M)b
Pe (10ꢀ6 cm sꢀ1
)
Prediction
2
3
4
5
6
7
8
9
>100
>100
>100
>100
>100
>100
>100
0.9 ꢁ 0.01
0.7 ꢁ 0.02
>100
10.1 ꢁ 0.1
12.3 ꢁ 0.5
12.3 ꢁ 0.2
12.6 ꢁ 0.3
4.0 ꢁ 0.1
3.9 ꢁ 0.1
5.9 ꢁ 0.2
7.3 ꢁ 0.2
14.2 ꢁ 0.1
5.3 ꢁ 0.2
3.7 ꢁ 0.1
nd
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ/ꢀ
CNSþ/ꢀ
CNSþ
CNSþ
10
11
12
Tacrine
CNSþ
CNSþ
3. Conclusions
>100
0.01 ꢁ 0.001
CNSþ/ꢀ
a
New promising neuroprotective agents derived from the hardly-
known dibenzo[1,4,5]thiadiazepine system are here described. At
micro- and sub-micromolar concentrations, these compounds
provide almost a full-protection against mitochondrial oxidative
stress when compounds were incubated with cells 24 h before the
addition of stressors. When compounds were added at the same
time than the toxic event the neuroprotection percentage was
smaller, although in some compounds was superior to 50%. These
results suggested that the neuroprotective properties of dibenzo
[1,4,5]thiadiazepines 2e12 are due to a mixed mechanism of action,
in which predominates the activation of endogenous antioxidant
pathways vs. the direct capture of ROS. Four CNS-permeable com-
pounds (2, 5, 8, and 11) were found as effective antioxidant agents
as trolox sequestering mitochondrial ROS. Moreover, compound 2
showed an interesting calcium-channel modulating activity,
blocking the entry of this cation by 22%, about a half of the nifed-
ipine value.
BuChE (EC 3.1.1.8) from horse serum.
Results are the mean of three independent experiments ꢁ SEM.
b
Table 2
Neuroprotection (%) of dibenzo[1,4,5]thiadiazepines 2e12 in the human neuro-
blastoma cell line SH-SY5Y against the combination of rotenone (30 M) and oli-
M) at the indicated concentrations, using pre- and co-incubation
m
gomycin A (10
conditionsa.
m
Comp.
Pre-incubation
0.3
Co-incubation
0.3
m
M
1
mM
3
m
M
m
M
1
mM
3 mM
2
3
4
5
6
7
8
9
64.8***
79.8***
90.3***
89.7***
76.4***
83.8**
86.0***
90.5***
75.9***
84.6***
99.3***
n.d.
76.6***
88.9***
90.2***
97.0***
91.1***
75.8*
96.4***
83.9***
92.1***
90.4***
94.6***
n.d.
81.5***
91.1***
84.4***
95.7***
92.2***
72.7*
97.5***
83.4***
96.1***
96.4***
92.4***
56.1***
49.0***
9.3*
40.5***
49.9*
20.7**
26.2*
50.6***
43.3***
8.1*
59.2***
38.3*
31.4***
50.1*
35.0***
27.8*
42.6***
38.2***
20.4*
63.2***
38.3*
38.8***
53.8*
35.8***
21.9*
35.7***
35.6***
15.9*
Such interesting profile, in addition to the almost novelty of this
heterocyclic system, highlight these dibenzo[1,4,5]thiadiazepines
as valuable candidates in the search for new treatments of neuro-
degenerative diseases.
10
11
12
Trolox
53.0***
40.7***
n.d.
55.4***
45.9***
n.d.
63.0***
32.7***
55.7***
*p < 0.05, **p < 0.01, ***p < 0.001 respect to control; n.d.: not determined.
a
Results are the mean of 4 independent experiments in triplicate.
4. Experimental
Solvents were purified and dried using standard methods,
distillation under nitrogen atmosphere and sodium wire (THF).
Reagents and anhydrous DMF were acquired from the usual com-
mercial suppliers and used without further purification. Chro-
matographic separations were performed on silica gel (Kielgel 60
Merck of 230e400 mesh) and compounds were detected with UV
to be the radical trapping, because it showed the same percentage
of protection in both conditions (around 56%).
On the basis of the above results, the CNS-permeable dibenzo
[1,4,5]thiadiazepines 2, 5, 8, and 11, which in co-incubation con-
ditions displayed a percentage of neuroprotection higher than 50%,
were selected to corroborate their capacity to capture mitochon-
drial radicals in the human neuroblastoma cell line. Thus, SH-SY5Y
cells were loaded with the fluorescent dye 20,70-dichlorfluorescein-
light (
l
¼ 254 nm). HPLC-MS analyses were performed on an
equipment composed by the separation module Alliance 2695, the
diacetate (DCFH-DA), treated with selected compounds at 0.3
and then subjected to free-radical generation by the mixture of
rotenone (30 M) plus oligomycin A (10 M). Trolox (0.3 M), a
mM
Table 3
Percentage of free-radical capture (%) and blockade of cytosolic calcium entry (%) by
compounds 2, 5, 8, and 11 in the human neuroblastoma cell line SH-SY5Y.a
m
m
m
well-known potent ROS-scavenger, was also evaluated as a positive
control. All tested compounds decreased the DCFH-DA fluorescence
with the same strength than trolox (Table 3), showing that they are
effective agents sequestering mitochondrial free radicals.
On the other hand, calcium is involved in many intracellular
signaling pathways, including learning and memory processes. It
has been described that disruptions in calcium homeostasis have
been implicated in neuronal degeneration [34], and that preventing
intracellular calcium overload could be a useful strategy to combat
AD pathology [35]. Thus, the ability of above-selected dibenzo
[1,4,5]thiadiazepines to reduce the cytosolic calcium concentration
was examined, using the human neuroblastoma cell line SH-SY5Y.
Cells were loaded with the fluorescent calcium probe Fluo-4/AM
Comp.
Free-radical capture (%)b
Blockade of cytosolic
calcium entry (%)c
2
5
8
11
24.3***
21.6**
24.3***
18.8***
22.3***
n.d.
21.5*
13.1*
9.3*
16.2*
n.d.
Trolox
Nifedipine
47.9***
a
Results are the mean of 3 independent experiments; *p < 0.05, **p < 0.01,
***p < 0.001 respect to control; n.d.: not determined.
b
Compounds at 0.3
m
M; fluorescent dye: DCFH-DA; stressor: rotenone (30
m
M)
and oligomycin A (10
m
m
M).
c
Compounds at 3
M; fluorescent dye: Fluo-4/AM (4 mM); cytosolic calcium
overload elicited by Kþ (70 mM).

354
G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
photodiode array Waters 2996, and the quadrupole mass analyser
4.1.4. N-[2-(5-Chloro-2-nitrothiophenyl)phenyl]-N,N0-bis(4-
methylpiperazin-1-yl)diacetylhydrazine (19)
ꢀ
Micromass ZQ.
A
SunFire C18 column, 91 A, 3.5
mm,
4.6 mm ꢂ 50 mm was used at a flow rate of 1 mL/min and a gradient
of (10e100)% B over 5 min. The mobile phase conditions were as
follow: eluant A: water/0.1% HCO2H; eluant B: MeCN/0.1% HCO2H.
Retention times (tR) are reported in minutes.
This intermediate remained in our chemical library as a
yellowish solid of m.p. 118e120 ꢃC. Its structure was confirmed by
MS, 1H and 13C NMR and its purity by HPLC, and was used without a
further purification. ESI-MS: m/z ¼ 576 [MH]þ, 578 [MH þ 2]þ. 1H
Melting points (uncorrected) were determined with
a
NMR (400 MHz, CDCl3):
d
(ppm) ¼ 8.92 (s, 1H), 8.23 (d, 1H,
Reichertert-Jung Thermovar apparatus. 1H NMR and 13C NMR
spectra were recorded in CD3OD or CDCl3 solutions using a Varian
J ¼ 8.9 Hz), 7.70 (dd, 1H, J ¼ 8.9 Hz, J ¼ 1.5 Hz), 7.54 (m, 2H), 7.19 (m,
2H), 6.72 (d, 1H, J ¼ 1.5 Hz), 3.15 (s, 2H), 3.05 (s, 2H), 2.22e2.59 (m,
XL-400 spectrometer. Chemical shifts are reported in
d
scale (ppm)
16H), 2.18 (s, 6H). 13C NMR (CDCl3):
d
(ppm) ¼ 170.4, 168.4, 144.3,
relative to internal Me4Si, J values are given in Hertz, and spin
multiplicities are expressed as s (singlet), d (doublet), t (triplet), or
m (multiplet). Mass spectra (MS) were obtained by electron spray
ionization (ESI) in positive mode using a HewlettePackard MSD
1100 spectrometer. Elemental analyses were carried out in a
Perkine-Elmer 240C equipment in the Centro de Química Orgánica
Manuel Lora-Tamayo (CSIC) and all results are within ꢁ0.4% of the
theoretical values.
141.2,139.3,138.4,132.6,132.1,131.9,130.2,128.1,127.2,126.8,125.8,
60.7, 59.9, 54.7 (4C), 54.5 (4C), 45.6 (2C). HPLC-MS purity: 98%,
tR ¼ 2.40 min.
4.1.5. N-[2-(6-Chloro-2-nitrothiophenyl)phenyl]-N,N0-
bis(piperidin-1-yl)diacetylhydrazine (20)
This compound remained in our chemical library as a yellowish
solid of m.p. 119e122 ꢃC, pure enough to be used as intermediate
without a further purification. ESI-MS: m/z ¼ 546 [MH]þ, 548
[MH þ 2]þ. 1H NMR (400 MHz, CDCl3):
d
(ppm) ¼ 9.33 (s, 1H), 7.72
4.1. Compounds from our in-house library: 2e4 and 19, 20.
Purification and spectroscopic data
(dd, 1H, J ¼ 7.5 Hz, J ¼ 1.0 Hz), 7.60 (m, 3H), 7.50 (td, 1H, J ¼ 8.06 Hz,
J ¼ 1.2 Hz), 7.15 (m, 2H), 3.23 (s, 2H), 3.06 (s, 2H), 2.65 (m, 8H), 1.67
(m, 8H),1.56 (m, 4H). 13C NMR (CDCl3):
d
(ppm) ¼ 169.0,168.4,141.1,
These four compounds had been stored a long time in our in-
house library and they were initially examined by HPLC. Two of
them (2 and 3) were purified by a flash column chromatography on
silica gel, while 19 and 20 kept purities higher than 98% and were
used directly as intermediates in the following synthetic step. Their
chemical structures were confirmed by MS, 1H and 13C NMR and the
data obtained are given in this work.
140.3, 132.6, 132.5, 130.3, 129.9, 129.3, 128.8, 122.1, 126.1, 121.4,
121.3, 60.9, 59.0, 55.7 (4C), 26.2 (4C), 23.5 (2C). HPLC-MS purity:
98%, tR ¼ 2.50 min.
4.2. Synthesis of hydrazines
4.2.1. 2-[(4-Trifluoromethyl-2-nitrophenyl)thio]phenylhydrazine
It was synthesized following the general procedure described in
Ref. [36], using 2-[(4-trifluoromethyl-2-nitrophenyl)thio]aniline
(0.9 g, 2.86 mmol), NaNO2 (0.21 g, 3.15 mmol) and SnCl2.2H2O
(1.41 g, 6.30 mmol). Purification involved the use of hexane/ethyl
acetate (8:1) as eluant. Yellow solid (0.40 g, 43%) of m.p.141e143 ꢃC.
ESI-MS: m/z ¼ 330 [MH]þ, 353 [MH þ Na]þ. 1H NMR (CDCl3):
4.1.1. 2-Chloro-5,6-dihydro-5,6-diacetyldibenzo[b,f][1,4,5]
thiadiazepine (2)
This compound was purified by a flash column chromatography
eluted with hexane:EtOAc (5:1) and isolated from the fractions
Rf ¼ 0.4 as a colorless solid of m.p. 148e149 ꢃC (bibl. 150e151 ꢃC
[28]). ESI-MS m/z ¼ 333 [MH]þ, 335 [MH þ 2]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 8.52 (d, 1H, J ¼ 1.0), 7.52 (m, 2H), 7.44 (dd, 1H, J ¼ 7.9 Hz,
J ¼ 1.2 Hz), 7.29 (dd, 1H, J ¼ 7.9 Hz, J ¼ 1.2 Hz), 6.89 (m, 2H), 6.12 (s,
1H), 3.38 (s, 2H). 13C NMR (75 MHz, CDCl3):
d
(ppm) ¼ 7.67e7.03 (m, 7H), 2.12 (s, 3H), 2.02 (s, 3H). 13C NMR
d
(ppm) ¼ 152.5, 145.3,
(CDCl3):
d
¼ 170.1, 168.2, 140.4, 139.6, 130.4, 129.7, 129.4, 128.7,
142.4, 138.0, 133.5, 130.2, 128.9, 127.9, 125.1, 123.9, 120.5, 112.6,
128.3, 127.6, 127.3, 126.1, 122.5, 117.8, 22.0, 21.9. HPLC-MS purity:
110.0. HPLC-MS purity: 95%, tR ¼ 2.10 min.
100%, tR ¼ 4.95 min. Anal. C16H13ClN2O2S (C, H, N, S).
4.3. Hydrazides, N,N0-diacylhydrazines
4.1.2. 3-Chloro-5,6-dihydro-5,6-diacetyldibenzo[b,f][1,4,5]
thiadiazepine (3)
N,N0-Diacylhydrazines 13e15 were synthesized following a very
general method. Derivatives 21e23 required a two-steps approach,
the acylation of the corresponding hydrazine with 2-
bromopropionyl chloride to yield 18, which was treated with the
chosen amine in order to obtain 21e23. As explained above, hy-
drazides 19 and 20 were present in our in-house library of com-
pounds and were not synthesized again.
General procedure: Acetyl chloride (1.30 mmol) was added on an
ice-cooled solution of the selected hydrazine (0.65 mmol) and
triethylamine (1.30 mmol) in anhydrous toluene (3 mL). Then, the
resulting solution was refluxed (3 h), the solvent removed under
reduced pressure and the residue purified on silica gel, using a
chromatography column and a mixture of hexane/ethyl acetate
(1:1) as eluant.
This compound was purified by a flash column chromatography
eluted with hexane:EtOAc (4:1). It was isolated from the fractions
Rf ¼ 0.5 as a colorless solid of m.p. 196e197 ꢃC (bibl. 200e202 ꢃC
[28]). ESI-MS m/z ¼ 333 [MH]þ, 335 [MH þ 2]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 7.63e7.09 (m, 7H), 2.12 (s, 3H), 2.03 (s, 3H). 13C NMR
(CDCl3):
d
(ppm) ¼ 167.2, 166.8, 140.1, 139.4, 131.0, 129.6, 129.4,
129.1, 128.6, 128.0, 127.7, 127.0, 125.9, 125.6, 21.8, 21.7. HPLC-MS
purity: 97%, tR ¼ 4.98 min. Anal. C16H13ClN2O2S (C, H, N, S).
4.1.3. 3-Chlorodibenzo[b,f][1,4,5]thiadiazepine (4)
This compound remained in our chemical library as a colorless
solid of m.p. 121e123 ꢃC (bibl. 127e128 ꢃC [28]). Its structure was
confirmed and it did not need any purification. EM: m/z ¼ 247
[MH]þ, 249 [MH þ 2]þ. 1H NMR (400 MHz, CDCl3):
d
(ppm) ¼ 7.60
4.3.1. N-[2-(4-Methoxy-2-nitrothiophenyl)phenyl]-N,N0-
diacetylhydrazine (13)
(d, 1H, J ¼ 7.4 Hz), 7.52 (d, 1H, J ¼ 7.4 Hz), 7.46 (dd, 1H, J ¼ 7.4,
J ¼ 2.0 Hz), 7.39 (dd, 1H, J ¼ 7.4, J ¼ 2.0 Hz), 7.34e7.27 (m, 3H). 13
C
Following the general procedure, this product was obtained
NMR (CDCl3):
d
(ppm) ¼ 151.6, 150.0,135.5, 132.4,131.8, 131.2, 130.2,
(48%) as a yellow solid of m.p. 96e98 ꢃC. ESI-MS: m/z ¼ 376 [MH]þ,
129.9, 129.5, 129.4, 128.7, 127.6. HPLC-MS purity: 97%, tR ¼ 5.06 min.
399 [MH
þ
Na]þ, 775 [2 MH
þ
Na]þ. 1H NMR (DMSO):
Anal. C12H7ClN2S (C, H, N, S).
d
(ppm) ¼ 10.85 (s, 1H), 7.65 (d, 1H, J ¼ 2.8 Hz), 7.51e7.25 (m, 4H),

G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
355
7.22 (dd, 1H, J ¼ 9.0 Hz, J ¼ 2.8 Hz), 6.99 (d, 1H, J ¼ 9.0 Hz), 3.81 (s,
3H), 2.04 (s, 3H), 1.72 (s, 3H). 13C NMR (DMSO):
169.1, 159.1, 149.7, 143.9, 135.2, 133.2, 131.1, 130.3, 129.6, 126.2, 121.4,
29.8, 25.5 (2C), 24.0 (2C), 23.8 (2C). HPLC-MS purity: 90%,
tR ¼ 2.60 min.
d
(ppm) ¼ 171.9,
110.1, 109.2, 56.5, 21.9, 20.7. HPLC-MS purity: 98%, tR ¼ 4.58 min.
4.3.7. N-[2-(5-Chlor-2-nitrothiophenyl)phenyl]-N,N0-di[(2-
dipirrolidin-1-yl)propionyl]hydrazine (23)
4.3.2. N-[2-(4-Trifluoromethyl-2-nitrothiophenyl)phenyl]-N,N0-
diacetylhydrazine (14)
Applying the same method, this hydrazide was obtained (65%)
as a yellow solid. ESI-MS: m/z ¼ 546 [MH]þ, 548 [MH þ 2]þ. 1H NMR
Following the general procedure, this product was obtained
(DMSO):
d
(ppm) ¼ 10.91 (s, 1H), 8.22 (d, 1H, J ¼ 9.0 Hz), 7.60 (d, 1H,
(53%) as a yellow solid of m.p. 188e190 ꢃC. ESI-MS: m/z ¼ 414
J ¼ 9.0 Hz), 7.83e7.72 (m, 1H), 7.44e6.40 (m, 4H), 3.77 (t, 2H), 3.70
[MH]þ. 1H NMR (DMSO):
d
(ppm) ¼ 10.79 (s, 1H), 8.47 (s, 1H), 7.70
(t, 2H), 2.45e2.15 (m, 12H), 1.70e1.50 (m, 8H). 13C NMR (DMSO):
(m, 1H), 7.61 (m, 3H), 6.46 (m, 1H), 7.05 (m, 1H), 2.02 (s, 3H), 1.82 (s,
d
(ppm) ¼ 173.1, 170.6, 145.1, 143.8, 140.1, 139.2, 138.2, 136.9, 132.0,
3H). 13C NMR (DMSO):
d
(ppm) ¼ 171.8, 168.5, 145.1, 144.9, 143.2,
130.7, 129.6, 128.6, 127.8, 125.8, 53.4, 53.1, 51.2, 50.4, 32.8 (4C), 23.1
138.1, 132.2, 130.6, 129.6, 129.1, 127.8, 126.1, 125.6, 125.1, 122.5, 21.7,
(4C). HPLC-MS purity: 90%, tR ¼ 2.49 min.
20.3. HPLC-MS purity: 100%, tR ¼ 4.78 min.
4.4. General method for the synthesis of dibenzothiadiazepines
4.3.3. N-[2-(2,4-Dinitrothiophenyl)phenyl]-N,N0-diacetylhydrazine
(15)
Hydrazides were dissolved in DMF and refluxed for 15 min with
an equimolecular amount of K2CO3. Then, solvent was evaporated
and the residue purified through reverse-phase chromatography
using H2O/acetonitrile (95:5) as eluant.
Following the general procedure, this product was obtained
(57%) as a yellow solid of m.p. 156e158 ꢃC. ESI-MS: m/z ¼ 391
[MH]þ. 1H NMR (DMSO):
d
(ppm) ¼ 10.83 (s, 1H), 8.87 (d, 1H,
J ¼ 2.5 Hz), 8.33 (dd, 1H, J ¼ 2.5 Hz, J ¼ 9.1 Hz), 7.62 (m, 2H), 7.48 (m,
2H), 7.06 (d, 1H, J ¼ 9.1 Hz), 2.04 (s, 3H), 1.83 (s, 3H). 13C NMR
(DMSO):
d
(ppm) ¼ 172.4, 169.5, 147.1, 145.6, 145.2, 144.7, 138.9,
4.4.1. 2-Methoxy-5,6-dihydro-5,6-diacetyldibenzo[b,f][1,4,5]
thiadiazepine (5)
137.4, 132.7, 131.2, 130.4, 129.4, 127.9, 121.7, 22.5, 21.1. HPLC-MS
purity: 100%, tR ¼ 4.32 min.
Following the general procedure, this product was obtained
(60%) as a colorless solid, m.p.: 99e101 ꢃC. ESI-MS: m/z ¼ 329
[MH]þ, 351 [M þ Na]þ, 679 [2M þ Na]þ. 1H NMR (CDCl3):
4.3.4. N-[2-(5-Chlor-2-nitrothiophenyl)phenyl]-N,N0-di(2-
bromopropionyl)hydrazine (18)
d
(ppm) ¼ 7.79e6.73 (m, 7H), 3.81 (s, 3H), 2.19 (s, 3H), 2.05 (s, 3H).
13C NMR (CDCl3):
130.9, 129.9, 129.2, 128.1, 127.3, 125.9, 112.4, 111.8, 55.9, 22.4, 22.1.
d
(ppm) ¼ 167.8, 167.6, 159.2, 139.6, 133.2, 132.2,
Following the general procedure but using 2-bromopropionyl
chloride as acylating agent and dichloromethane/methanol (9:1)
as eluant, this product was obtained (30%) as a yellow solid. ESI-MS:
HPLC-MS purity: 97%, tR ¼ 4.53 min. Anal. C17H16N2O3S (C, H, N, S).
m/z ¼ 566 [MH]þ, 568 [MH þ 2]þ. 1H NMR (DMSO):
(ppm) ¼ 11.05
d
(s,1H), 8.22 (d,1H, J ¼ 9.0 Hz), 7.66 (m, 3H), 7.50 (m, 2H), 6.76 (d,1H,
J ¼ 2.0 Hz), 3.66 (t, 2H, J ¼ 6.0 Hz), 3.54 (t, 2H, J ¼ 6.6 Hz), 3.0 (m,
4.4.2. 2-Trifluoromethyl-5,6-dihydro-5,6-diacetyldibenzo[b,f][1,4,5]
thiadiazepine (6)
2H), 2.70 (m, 2H). 13C NMR (DMSO):
d
(ppm) ¼ 171.1, 169.3, 144.6,
Following the general procedure, this product was obtained
143.8,140.1, 139.0, 137.1, 131.9, 130.0, 129.4, 127.9, 128.3, 127.3, 125.8,
36.2, 35.4, 28.8, 27.4. HPLC-MS purity: 90%, tR ¼ 5.33 min.
(40%) as a colorless solid, m.p.: 125e127 ꢃC. ESI-MS: m/z ¼ 367
[MH]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 7.82e7.27 (m, 7H), 2.21 (s, 3H),
2.09 (s, 3H). 13C NMR (CDCl3):
d
(ppm) ¼ 169.8, 168.4, 142.2, 135.2,
4.3.5. N-[2-(5-Chlor-2-nitrothiophenyl)phenyl]-N,N0-di(2-
dimethylaminopropionyl)hydrazine (21)
129.8, 129.0, 128.6, 128.2, 127.9, 127.5, 127.1, 126.5, 124.6, 123.2,
122.6, 22.0, 21.4. HPLC-MS purity: 100%, tR ¼ 5.17 min. Anal.
This derivative was synthesized following a general procedure
of amination: hydrazide 18 (0.88 mmol) reacted with dimethyl-
amine (3.52 mmol) in anhydrous toluene (6 mL) at room temper-
ature. After remaining stirred overnight, solvent was evaporated,
the residue washed (NaHCO3 5%), dried and purified on silica gel,
using a chromatography column and a mixture of dichloro-
methane/methanol (9:1) and ammonium hydroxide 2% as eluant.
The product was obtained (61%) as a yellow solid. 1H NMR (DMSO):
C17H13F3N2O2S (C, H, N, S).
4.4.3. 2-Nitro-5,6-dihydro-6-acetyldibenzo[b,f][1,4,5]thiadiazepine
(7)
Following the general procedure, this product was obtained
(50%) as a yellow solid, m.p.: 111e113 ꢃC. ESI-MS: m/z ¼ 302 [MH]þ.
1H NMR (CDCl3):
d
(ppm) ¼ 8.15 (d, 1H, J ¼ 2.4 Hz), 7.88 (dd, 1H,
J ¼ 8.9 Hz, J ¼ 2.5 Hz), 7.62 (dd, 1H, J ¼ 5.7 Hz, J ¼ 3.2 Hz), 7.44 (m,
d
(ppm) ¼ 10.84 (s, 1H), 8.22 (d, 1H, J ¼ 9.0 Hz), 7.56 (d, 1H,
J ¼ 9.0 Hz), 7.85 (m, 1H), 7.22 (m, 4H), 3.80 (t, 2H), 3.75 (t, 2H), 2.95
(s, 12H), 2.72 (t, 2H), 2.65 (t, 2H). 13C NMR (DMSO):
2H), 7.37 (m,1H), 7.32 (s,1H), 6.80 (d,1H, J ¼ 8.8 Hz), 2.02 (s, 3H). 13
C
NMR (CDCl3):
d
(ppm) ¼ 170.5, 151.2, 144.2, 141.6, 132.8,131.4, 130.6,
d
(ppm) ¼ 173.4,
129.9, 127.7, 126.4, 123.3, 117.1, 116.4, 20.6. HPLC-MS purity: 100%,
170.9, 145.1, 139.8, 139.0, 138.0, 136.8, 136.3, 131.9, 130.8, 129.2,
128.6, 127.2, 125.8, 60.1, 58.7, 47.3, 42.0, 30.7 (2C), 28.6 (2C). HPLC-
MS purity: 90%, tR ¼ 2.01 min.
tR ¼ 4.56 min. Anal. C14H11N3O3S (C, H, N, S).
4.4.4. 3-Chloro-5,6-dihydro-5,6-di(4-methylpiperazin-1-yl)
acetyldibenzo[b,f][1,4,5]thiadiazepine (8)
4.3.6. N-[2-(5-Chlor-2-nitrothiophenyl)phenyl]-N,N0-di[(2-
dipiperidin-1-yl)propionyl]hydrazine (22)
Following the general procedure, this product was obtained
Applying the same method, this hydrazide was obtained (75%)
(31%) as a colorless solid, m.p.: 174e176 ꢃC. ESI-MS: m/z ¼ 529
as a yellow solid. ESI-MS: m/z ¼ 574 [MH]þ, 576 [MH þ 2]þ. 1H NMR
[MH]þ, 531 [MH þ 2]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 7.70e7.12 (m,
(DMSO):
d
(ppm) ¼ 10.84 (s, 1H), 8.22 (d, 1H, J ¼ 9.0 Hz), 7.54 (d, 1H,
7H), 3.25e3.20 (m, 4H), 2.62e2.0 (m, 22H). 13C NMR (CDCl3):
J ¼ 9.0 Hz), 7.85e7.70 (m, 1H), 7.20e6.96 (m, 4H), 3.78 (t, 2H), 3.71
d
(ppm) ¼ 167.5, 167.2, 140.5, 139.8, 130.8, 130.2, 129.2, 128.6, 128.3,
(t, 2H), 2.30e2.10 (m, 12H), 1.50e1.20 (m, 12H). 13C NMR (DMSO):
127.8, 127.6, 127.4, 127.2, 125.9, 59.8, 59.6, 54.8 (2C), 54.7 (2C), 52.7
(2C), 52.5 (2C), 45.9 (2C). HPLC-MS purity: 100%, tR ¼ 3.03 min.
Anal. C26H33ClN6O2S (C, H, N, S).
d
(ppm) ¼ 173.4, 170.9, 145.1, 143.8, 140.1, 139.2, 138.0, 136.3, 131.9,
129.2, 128.6, 127.7, 126.6, 125.8, 54.1, 53.7, 53.6, 53.3, 30.1, 30.9, 30.7,

356
G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
4.4.5. 4-Chloro-5,6-dihydro-5,6-di(piperidin-1-yl)acetyldibenzo
[b,f][1,4,5]thiadiazepine (9)
buffered saline solution at pH 7.4 (PBS), and dodecane were pur-
chased from Sigma, Aldrich, Acros, and Fluka. Millex filter units
Following the general procedure, this product was obtained
(PVDF membrane, diameter 25 mm, pore size 0.45 mm) were ac-
quired from Millipore. The porcine brain lipid (PBL) was obtained
from Avanti Polar Lipids. The donor microplate was a 96-well filter
(39%) as a colorless solid, m.p.: 131e133 ꢃC. ESI-MS: m/z ¼ 499
[MH]þ, 501 [MH þ 2]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 7.62e7.10 (m,
7H), 3.44e3.01 (m, 4H), 2.98e2.17 (m, 20H). 13C NMR (CDCl3):
plate (PVDF membrane, pore size 0.45
microplate was an indented 96-well plate, both from Millipore. The
acceptor 96-well microplate was filled with 200 L of PBS:ethanol
(70:30) and the filter surface of the donor microplate was
impregnated with 4
L of PBL in dodecane (20 mg mLꢀ1). Com-
pounds were dissolved in PBS: ethanol (70:30) at 100
g mLꢀ1
filtered through a Millex filter, and then added to the donor wells
(200 L). The donor filter plate was carefully put on the acceptor
mm) and the acceptor
d
(ppm) ¼ 169.4, 166.5, 139.4, 134.5, 133.4, 130.6, 130.1, 128.6, 128.3,
127.9, 127.1, 126.7, 125.8, 124.9, 61.3, 60.6, 54.2 (2C), 54.0 (2C), 25.7
m
(2C), 25.6 (2C), 23.8, 23.5. HPLC-MS purity: 97%, tR ¼ 4.91 min. Anal.
C
26H31ClN4O2S (C, H, N, S).
m
m
,
4.4.6. 3-Chloro-5,6-dihydro-5,6-di(2-dimethylamino)
propionyldibenzo[b,f][1,4,5]thiadiazepine (10)
m
Following the general procedure, this product was obtained
plate to form a sandwich, which was left undisturbed for 240 min at
25 ꢃC. After incubation, the donor plate is carefully removed and
the concentration of compounds in the acceptor wells was deter-
mined by UVeVis spectroscopy. Every sample is analyzed at five
wavelengths, in four wells and at least in three independent runs,
and the results are given as the mean ꢁ standard deviation. In each
experiment, 10 quality control standards of known BBB perme-
ability were included to validate the analysis set.
(25%) as a colorless solid, m.p.: 138e140 ꢃC. ESI-MS: m/z ¼ 447
[MH]þ, 449 [MH þ 2]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 8.07e7.24 (m,
7H), 2.82e2.69 (m, 8H), 2.48e2.40 (m, 12H). 13C NMR (CDCl3):
d
(ppm) ¼ 173.1, 170.5, 141.2, 138.9, 132.2, 131.3, 131.0, 130.8, 130.4,
130.0, 129.8, 129.4, 129.0, 128.2, 55.8, 55.4, 44.7 (2C), 44.1 (2C), 30.8,
30.9. HPLC-MS purity: 96%, tR ¼ 4.20 min. Anal. C22H27ClN4O2S (C,
H, N, S).
4.4.7. 3-Chloro-5,6-dihydro-5,6-di[2-(piperidin-1-yl)]
propionyldibenzo[b,f][1,4,5]thiadiazepine (11)
4.5.3. Culture of the human neuroblastoma cell line SH-SY5Y
SH-SY5Y cells, at passages between 3 and 16 after de-freezing,
Following the general procedure, this product was obtained
were maintained in a Dulbecco’s modified Eagle’s medium
(DMEM) containing 15 non-essential amino-acids (NEAAs) and
supplemented with 10% fetal bovine serum (FBS), 1 mM glutamine,
(22%) as a colorless solid, m.p.: 151e153 ꢃC. ESI-MS: m/z ¼ 527
[MH]þ, 529 [MH þ 2]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 8.27e7.22 (m,
7H), 3.73e1.40 (m, 28H). 13C NMR (MeOH):
d
(ppm) ¼ 173.1, 170.5,
50 units/mL penicillin and 50 mg/mL streptomycin (reagents from
141.1,138.9,132.1,131.8,130.9,130.8,130.6,130.4,129.8,129.3,128.6,
128.2, 55.2, 54.8 (2C), 54.6 (2C), 53.7, 30.6, 29.7, 24.2 (2C), 24.1 (2C),
GIBCO, Madrid, Spain). Cultures were seeded into flasks containing
supplemented medium and maintained at 37 ꢃC in 5% CO2/hu-
midified air. Stock cultures were passaged 1:4 twice weekly. For cell
viability/cell death experiments, SH-SY5Y cells were sub-cultured
in 48 well plates at a seeding density of 105 cells per well and
were exposed to the compounds before confluence, in DMEM free
of serum.
22.5, 22.4. HPLC-MS purity: 100%, tR
28H35ClN4O2S (C, H, N, S).
¼
4.27 min. Anal.
C
4.4.8. 3-Chloro-5,6-dihydro-5,6-di[2-(pirrolidin-1-yl)]
propionyldibenzo[b,f][1,4,5]thiadiazepine (12)
Following the general procedure, this product was obtained
(20%) as a colorless solid, m.p.: 180e182 ꢃC. ESI-MS: m/z ¼ 499
4.5.4. Cell viability experiments
[MH]þ, 501 [MH þ 2]þ. 1H NMR (CDCl3):
d
(ppm) ¼ 8.05e7.18 (m,
(ppm) ¼ 173.1, 170.5,
To study the cytotoxic effects of compounds alone, cells were
plated at a density of 105 cells per well at least 48 h before the
toxicity measurements. Cells were exposed for 24 h to the com-
7H), 3.64e1.92 (m, 24H). 13C NMR (MeOH):
d
141.2, 138.9, 132.2, 131.3, 131.0, 130.8, 130.4, 130.0, 129.8, 129.4,
129.0, 128.2, 55.6 (2C), 55.4 (2C), 51.6, 51.4, 31.7, 31.1, 24.0 (2C), 23.9
(2C). HPLC-MS purity: 97%, tR ¼ 4.19 min. Anal. C26H31ClN4O2S (C, H,
N, S).
pound at 3 mM, and the quantitative assessment of cell death was
made by measurement of the percent of the intracellular enzyme
lactate dehydrogenase (LDH) released to the extracellular medium
(cytotoxicity detection kit, Roche). The quantity of LDH was eval-
uated in a microplate reader (Anthos 2010 or Labsystems iMES
4.5. Biological studies
Reader MS) at 492 nm (l excitation) and 620 nm (l emission).
4.5.1. Cholinesterase inhibitory activities
Acetylcholinesterase (AChE, EC 3.1.1.7) from bovine erythrocytes
(0.25e1.0 unit/mg, lyophilized powder) and butyrylcholinesterase
(BuChE, EC 3.1.1.8) from equine serum (10 units/mg protein,
lyophilized powder) were purchased from Sigma. Compounds were
measured in 100 mM phosphate buffer pH 8.0 at 30 ꢃC, using
acetylthiocholine and butyrylthiocholine (0.4 mM) as substrates,
respectively. In both cases, 5,50-dithio-bis(2-nitrobenzoic)acid
(DTNB, Ellman’s reagent, 0.2 mM) was used and the values of IC50
were calculated by UV spectroscopy, from the absorbance changes
at 412 nm [29]. Experiments were performed in triplicate.
4.5.5. Neuroprotection studies
To study the cytoprotective action of the compounds against cell
death induced by the mixture of rotenone (30 M) and oligomycin
A (10 M) in the experiments of pre-incubation, drugs were given
at time zero and maintained for 24 h. Then, the media were
replaced by fresh media still containing the drug plus the cytotoxic
insult, which was left for an additional 24 h period. Thereafter, cell
survival was assessed measuring LDH activity. For the experiments
of co-incubation the compounds and the combination rotenone
plus oligomycin A were added at the same time and incubated for
24 h.
m
m
4.5.2. In vitro bloodebrain barrier permeation assay
Prediction of the brain penetration was evaluated using a par-
allel artificial membrane permeation assay (PAMPA-BBB), in a
similar manner as previously described [11,13,14,16,31]. Pipetting
was performed with a semi-automatic robot (CyBiÒ-SELMA) and
UV reading with a microplate spectrophotometer (Multiskan
Spectrum, Thermo Electron Co.). Commercial drugs, phosphate
4.5.6. Evaluation of selected compounds as free-radical scavengers
in SH-SY5Y cells
Dihydrodichlorofluorescein diacetate (DCFH-DA) was used to
assess intracellular ROS [37]. SH-SY5Y neuroblastoma cells were
grown at confluence in 96-well black dishes. Cells were incubated
with 10
mM DCFH-DA for 45 min and basal fluorescence was

G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
357
[5] S. Benfeito, C. Oliveira, P. Soares, C. Fernandes, T. Silva, J. Teixeira, F. Borges,
Antioxidant therapy: still in search of the ’magic bullet’, Mitochondrion 13
(2013) 427e435.
measured in a fluorescence microplate reader (FLUOstar Optima,
BMG, Germany). Then, a mixture of rotenone (30 M), oligomycin A
(10 M), and the tested compound (0.3 M) were added at time
zero (t ¼ 0). After 120 min, changes in fluorescence were measured,
using 485 nm and 520 nm as wavelengths of excitation and emis-
sion, respectively.
m
m
m
[6] G.E. Stutzmann, Calcium dysregulation, IP3 signaling, and Alzheimer’s disease,
Neuroscientist 11 (2005) 110e115.
ꢁ ꢂ
[7] V. Pogacic Kramp, List of drugs in development for neurodegenerative dis-
eases: update October 2011, Neurodegener. Dis. 9 (2012) 210e283.
[8] R. Morphy, Z. Rankovic, Designed multiple ligands. An emerging drug dis-
covery paradigm, J. Med. Chem. 48 (2005) 6523e6543.
[9] M.L. Bolognesi, E. Simoni, M. Rosini, A. Minarini, V. Tumiatti, C. Melchiorre,
Multitarget-directed ligands: innovative chemical probes and therapeutic
tools against Alzheimer’s disease, Curr. Top. Med. Chem. 11 (2011) 2797e
2806.
[10] A. Cavalli, M.L. Bolognesi, S. Capsoni, V. Andrisano, M. Bartolini, E. Margotti,
A. Cattaneo, M. Recanatini, C. Melchiorre, A small molecule targeting the
multifactorial nature of Alzheimer’s disease, Angew. Chem. Int. Ed. Engl. 46
(2007) 3689e3692.
[11] M.I. Fernández-Bachiller, C. Pérez, L. Monjas, J. Rademann, M.I. Rodríguez-
Franco, New tacrinee4-oxo-4H-chromene hybrids as multifunctional agents
for the treatment of Alzheimer’s disease, with cholinergic, antioxidant, and
beta-amyloid-reducing properties, J. Med. Chem. 55 (2012) 1303e1317.
[12] M. Maroto, A.M. de Diego, E. Albinana, J.C. Fernández-Morales, A. Caricati-
Neto, A. Jurkiewicz, M. Yáñez, M.I. Rodríguez-Franco, S. Conde, M.P. Arce,
J.M. Hernández-Guijo, A.G. García, Multi-target novel neuroprotective com-
pound ITH33/IQM9.21 inhibits calcium entry, calcium signals and exocytosis,
Cell. Calcium 50 (2011) 359e369.
[13] M.I. Fernández-Bachiller, C. Pérez, G.C. González-Muñoz, S. Conde, M.G. López,
M. Villarroya, A.G. García, M.I. Rodríguez-Franco, Novel tacrine-8-
hydroxyquinoline hybrids as multifunctional agents for the treatment of
Alzheimer’s disease, with neuroprotective, cholinergic, antioxidant, and
copper-complexing properties, J. Med. Chem. 53 (2010) 4927e4937.
[14] M.P. Arce, M.I. Rodríguez-Franco, G.C. González-Muñoz, C. Pérez, B. López,
M. Villarroya, M.G. López, A.G. García, S. Conde, Neuroprotective and cholin-
ergic properties of multifunctional glutamic acid derivatives for the treatment
of Alzheimer’s disease, J. Med. Chem. 52 (2009) 7249e7257.
4.5.7. Measurement of cytosolic calcium concentration
SHSY5Y neuroblastoma cells were grown at confluence in 96-
well black dishes. Cells were loaded with Fluo-4/AM (4
mM) for
1 h at 37 ꢃC in Eagle’s minimal essential medium. Then, cells were
washed twice with Krebs Hepes solution and kept at room tem-
perature for 15 min before the beginning of the experiment.
Compounds were incubated 10 min before Kþ (70 mM) was applied
to evoke the increment of cytosolic [Ca2þ]. At the end of the
experiment, Triton X-100 (5%) and MnCl2 (1 mM) were applied to
record maximal and basal fluorescence, respectively. Fluorescence
was measured in a fluorescence microplate reader (FLUOstar Op-
tima, BMG, Germany). Wavelengths of excitation and emission
were 485 and 520 nm, respectively.
4.5.8. Measurement of LDH activity
Extracellular and intracellular LDH activity was measured by
UVevis using a cytotoxicity cell death kit (Roche-Boehringer.
Mannheim, Germany) according to the manufacturer’s indications.
Total LDH activity was defined as the sum of intracellular and
extracellular LDH activity and released LDH was defined as the
percentage of extracellular compared with total LDH activity. Data
were expressed as the mean (ꢁSEM) of at least three different
cultures in quadruplicate. LDH released was calculated for each
individual experiment considering 100% the extracellular LDH
released by the vehicle with respect to the total. To determinate
percent protection, LDH release was normalized as follows: in each
individual triplicate experiment, LDH release obtained in non-
treated cells (basal) was subtracted from the LDH released upon
the toxic treatment and normalized to 100% and that value was
subtracted from 100.
[15] M.I. Fernández-Bachiller, C. Pérez, N.E. Campillo, J.A. Páez, G.C. González-
Muñoz, P. Usán, E. García-Palomero, M.G. López, M. Villarroya, A.G. García,
A. Martínez, M.I. Rodríguez-Franco, Tacrine-melatonin hybrids as multifunc-
tional agents for Alzheimer’s disease, with cholinergic, antioxidant, and
neuroprotective properties, ChemMedChem 4 (2009) 828e841.
[16] M.I. Rodríguez-Franco, M.I. Fernández-Bachiller, C. Pérez, B. Hernández-
Ledesma, B. Bartolomé, Novel tacrine-melatonin hybrids as dual-acting drugs
for Alzheimer disease, with improved acetylcholinesterase inhibitory and
antioxidant properties, J. Med. Chem. 49 (2006) 459e462.
[17] C. Spuch, D. Antequera, M.I. Fernández-Bachiller, M.I. Rodríguez-Franco,
E. Carro,
A new tacrine-melatonin hybrid reduces amyloid burden and
behavioral deficits in a mouse model of Alzheimer’s disease, Neurotox. Res. 17
(2010) 421e431.
[18] D. Antequera, M. Bolos, C. Spuch, C. Pascual, I. Ferrer, M.I. Fernández-Bachiller,
M.I. Rodríguez-Franco, E. Carro, Effects of
a tacrine-8-hydroxyquinoline
hybrid (IQM-622) on Abeta accumulation and cell death: involvement in
hippocampal neuronal loss in Alzheimer’s disease, Neurobiol. Dis. 46 (2012)
682e691.
Acknowledgments
[19] S. Lorrio, V. Gómez-Rangel, P. Negredo, J. Egea, R. León, A. Romero, T. Dal-Cim,
M. Villarroya, M.I. Rodríguez-Franco, S. Conde, M.P. Arce, J.M. Roda,
A.G. García, M.G. López, Novel multitarget ligand ITH33/IQM9.21 provides
neuroprotection in in vitro and in vivo models related to brain ischemia,
Neuropharmacology 67 (2013) 403e411.
[20] S. Conde, D.I. Pérez, A. Martínez, C. Pérez, F.J. Moreno, Thienyl and phenyl
alpha-halomethyl ketones: new inhibitors of glycogen synthase kinase (GSK-
3beta) from a library of compound searching, J. Med. Chem. 46 (2003) 4631e
4633.
[21] D.I. Pérez, S. Conde, C. Pérez, C. Gil, D. Simon, F. Wandosell, F.J. Moreno,
J.L. Gelpi, F.J. Luque, A. Martínez, Thienylhalomethylketones: irreversible
glycogen synthase kinase 3 inhibitors as useful pharmacological tools, Bioorg.
Med. Chem. 17 (2009) 6914e6925.
The authors gratefully acknowledge the financial support of
Spanish Ministry of Economy and Competitiveness (projects
SAF2012-31035 and SAF2012-32223), Fundación de Investigación
Médica Mutua Madrileña Automovilística (AP103952012), and
Consejo Superior de Investigaciones Científicas Spain (CSIC, PIE-
201280E074). The fellowships to G.C.G.-M. and M.P.A. from CSIC
and MICINN respectively, are also acknowledged.
Appendix A. Supplementary data
[22] J.A. Morales-García, C. Susín, S. Alonso-Gil, D.I. Pérez, V. Palomo, C. Pérez,
S. Conde, A. Santos, C. Gil, A. Martínez, A. Pérez-Castillo, Glycogen synthase
kinase-3 inhibitors as potent therapeutic agents for the treatment of Parkin-
son disease, ACS Chem. Neurosci. 4 (2013) 350e360.
Supplementary data associated with this article can be found in
the online version, at http://dx.doi.org/10.1016/j.ejmech.2014.04.
075. These data include MOL files and InChiKeys of the most
important compounds described in this article.
[23] G.C. González-Muñoz, M.P. Arce, B. López, C. Pérez, M. Villarroya, M.G. López,
A.G. García, S. Conde, M.I. Rodríguez-Franco, Old phenothiazine and diben-
zothiadiazepine derivatives for tomorrow’s neuroprotective therapies against
neurodegenerative diseases, Eur. J. Med. Chem. 45 (2010) 6152e6158.
[24] G.C. González-Muñoz, M.P. Arce, B. López, C. Pérez, A. Romero, L. del
Barrio, M.D. Martín-de-Saavedra, J. Egea, R. León, M. Villarroya,
M.G. López, A.G. García, S. Conde, M.I. Rodríguez-Franco, N-acylamino-
phenothiazines: neuroprotective agents displaying multifunctional activ-
ities for a potential treatment of Alzheimer’s disease, Eur. J. Med. Chem. 46
(2011) 2224e2235.
[25] M.J. Ohlow, B. Moosmann, Phenothiazine: the seven lives of pharmacology’s
first lead structure, Drug Discov. Today 16 (2011) 119e131.
[26] A.C.-Y. Chen, J.M. Scott, B.A. Stearns, N.S. Stock, Y.P. Truong, Tricyclic com-
pounds useful as neurogenic and neuroprotective agents, PCT Int. Appl. (2013)
WO2013043744.
References
[1] Alzheimer’s Association, Alzheimer’s Disease Facts and Figures, 2013. http://
www.alz.org/downloads/facts_figures_2013.pdf.
[2] F.A. Luque, S.L. Jaffe, The molecular and cellular pathogenesis of dementia of
the Alzheimer’s type an overview, Int. Rev. Neurobiol. 84 (2009) 151e165.
[3] T.S. Anekonda, J.F. Quinn, Calcium channel blocking as a therapeutic strategy
for Alzheimer’s disease: the case for isradipine, Biochim. Biophys. Acta 1812
(2011) 1584e1590.
[4] M. Dumont, M.F. Beal, Neuroprotective strategies involving ROS in Alzheimer
disease, Free Radic. Biol. Med. 51 (2011) 1014e1026.

358
G.C. González-Muñoz et al. / European Journal of Medicinal Chemistry 81 (2014) 350e358
[27] C. Korth, B.C. May, F.E. Cohen, S.B. Prusiner, Acridine and phenothiazine de-
rivatives as pharmacotherapeutics for prion disease, Proc. Natl. Acad. Sci. U S A
98 (2001) 9836e9841.
[28] C. Corral, J. Lissavetzky, G. Quintanilla, New method for the synthesis of
chloro-substituted dibenzo[b,f][1,4,5]thiadiazepines and their 5,6-dihy-
droderivatives, J. Org. Chem. 47 (1982) 2214e2215.
[29] G.L. Ellman, K.D. Courtney, V. Andres Jr., R.M. Feather-Stone, A new and rapid
colorimetric determination of acetylcholinesterase activity, Biochem. Phar-
macol. 7 (1961) 88e95.
[30] M. Cygler, J.D. Schrag, J.L. Sussman, M. Harel, I. Silman, M.K. Gentry,
B.P. Doctor, Relationship between sequence conservation and three-
dimensional structure in a large family of esterases, lipases, and related
proteins, Protein Sci. 2 (1993) 366e382.
[31] L. Di, E.H. Kerns, K. Fan, O.J. McConnell, G.T. Carter, High throughput artificial
membrane permeability assay for blood-brain barrier, Eur. J. Med. Chem. 38
(2003) 223e232.
[32] J. Egea, A.O. Rosa, A. Cuadrado, A.G. García, M.G. López, Nicotinic receptor
activation by epibatidine induces heme oxygenase-1 and protects chromaffin
cells against oxidative stress, J. Neurochem. 102 (2007) 1842e1852.
[33] T. Valero, L. del Barrio, J. Egea, N. Cañas, A. Martínez, A.G. García, M. Villarroya,
M.G. López, NP04634 prevents cell damage caused by calcium overload and
mitochondrial disruption in bovine chromaffin cells, Eur. J. Pharmacol. 607
(2009) 47e53.
[34] J. Cooper-Knock, J. Kirby, L. Ferraiuolo, P.R. Heath, M. Rattray, P.J. Shaw, Gene
expression profiling in human neurodegenerative disease, Nat. Rev. Neurol. 8
(2012) 518e530.
[35] P.C. Trippier, K. Jansen Labby, D.D. Hawker, J.J. Mataka, R.B. Silverman, Target-
and mechanism-based therapeutics for neurodegenerative diseases: strength
in numbers, J. Med. Chem. 56 (2013) 3121e3147.
[36] A. Roe, W.F. Little, The preparation of some fluoro- and tri-
fluoromethylphenothiazines, and some observations regarding determination
of their structure by infrared spectroscopy, J. Org. Chem. 20 (1955) 1577e
1590.
[37] C.P. LeBel, H. Ischiropoulos, S.C. Bondy, Evaluation of the probe 2’,7’-dichlor-
ofluorescin as an indicator of reactive oxygen species formation and oxidative
stress, Chem. Res. Toxicol. 5 (1992) 227e231.