Synthesis and biological evaluation of amidine, guanidine, and thiourea derivatives of 2-amino-(6-trifluoromethoxy)benzothiazole as neuroprotective agents potentially useful in brain diseases

Article abstract of DOI:10.1021/jm901375r

A series of amidine, thiourea, and guanidine derivatives of 2-amino-6-(trifluoromethoxy)benzothiazole termed 2, 3, and 4, respectively, and structurally related to riluzole, a neuroprotective drug in many animal models of brain disease, have been synthesized. The biological activity of compounds 2a-e, 3a-f, and 4a,b was preliminarily tested by means of an in vitro protocol of ischemia/reperfusion injury. The results demonstrated that 2c and 3a-d significantly attenuated neuronal injury. Selected for testing of their antioxidant properties, compounds 3a-d were shown to be endowed with a direct ROS scavenging activity. Compounds 3b and 3d were also evaluated for their activity on voltage-dependent Na⁺ and Ca²⁺ currents in neurons from rat piriform cortex. At 50 μM, compound 3b inhibited the transient Na⁺ current to a much smaller extent than riluzole, whereas 3d was almost completely ineffective. 2009 American Chemical Society.

Full text of DOI:10.1021/jm901375r

734 J. Med. Chem. 2010, 53, 734–744
DOI: 10.1021/jm901375r
Synthesis and Biological Evaluation of Amidine, Guanidine, and Thiourea Derivatives of 2-Amino-
(6-trifluoromethoxy)benzothiazole as Neuroprotective Agents Potentially Useful in Brain Diseases†
Maurizio Anzini,*,‡ Alessia Chelini,‡ Alessandra Mancini,‡ Andrea Cappelli,‡ Maria Frosini,§ Lorenzo Ricci,§
Massimo Valoti,§ Jacopo Magistretti, Loretta Castelli, Antonio Giordani,^ Francesco Makovec,^ and Salvatore Vomero‡
‡Dipartimento Farmaco Chimico Tecnologico and European Research Centre for Drug Discovery and Development, Universitaꢀ di Siena, Via A.
Moro, 53100 Siena, Italy, §Dipartimento di Neuroscienze, Sezione di Farmacologia, Fisiologia e Tossicologia, Universitaꢀ di Siena, Via A. Moro,
53100 Siena, Italy, Dipartimento di Fisiologia, Sezione di Fisiologia Generale, Universitaꢀ di Pavia, Via Forlanini 6, 27100 Pavia, Italy, and
^Rottapharm SpA, Via Valosa di Sopra 7, 20052 Monza, Italy
Received September 29, 2009
A series of amidine, thiourea, and guanidine derivatives of 2-amino-6-(trifluoromethoxy)benzothiazole
termed 2, 3, and 4, respectively, and structurally related to riluzole, a neuroprotective drug in many
animal models of brain disease, have been synthesized. The biological activity of compounds 2a-e,
3a-f, and 4a,b was preliminarily tested by means of an in vitro protocol of ischemia/reperfusion injury.
The results demonstrated that 2c and 3a-d significantly attenuated neuronal injury. Selected for testing
of their antioxidant properties, compounds 3a-d were shown to be endowed with a direct ROS
scavenging activity. Compounds 3b and 3d were also evaluated for their activity on voltage-dependent
Naþ and Ca2þ currents in neurons from rat piriform cortex. At 50 μM, compound 3b inhibited the
transient Naþ current to a much smaller extent than riluzole, whereas 3d was almost completely
ineffective.
Introduction
allows them to generate a net ion flux in the presence of
glutamate. Once activated, NMDA channels allow Ca2þ to
flow into the neuron, which further increases the Ca2þ load to
neuronal cytoplasm. An excessive [Ca2þ]i increase triggers a
number of intracellular events which may eventually lead to
cell damage or death. Among such Ca2þ-dependent responses
are production of nitric oxide (NO) by neuronal nitric oxide
synthase (nNOS)3 and release of superoxide anion from
mitochondria.4 NO and superoxide then combine to form
the highly reactive species peroxynitrite, which results in
further Ca2þ influx and production of oxygen and nitrogen
free radicals.
Excitatory amino acids (EAAs)a are neurotransmitters that
play an important role in the development of both chronic
neurodegenerative disorders like Alzheimer’s (AD) and Par-
kinson’s (PD) diseases as well as amyotrophic lateral sclerosis
(ALS), and acute conditions such as brain ischemia and
trauma.1 Indeed, “excitotoxicity” is a term coined to describe
an excessive release of glutamate, and a subsequent over-
activation of excitatory amino acid receptors (NMDA,
AMPA, and kainate). Glutamate-induced depolarization
due to AMPA receptor activation is an early step in excito-
toxicity and leads to Ca2þ influx through voltage-gated
calcium channels and an intracellular Ca2þ concentration
increase, subsequently amplified by release of Ca2þ from
intracellular stores. Depolarization also promotes activation
of voltage-dependent Naþ channels, which, if uncontrolled,
may cause the intracellular Naþ concentration to abnormally
increase.2 This process can lead to osmotic swelling of the
neuron as Cl- passively enters the cell and is followed by the
entry of water. Another consequence of depolarization is the
release of the “Mg2þ block” from NMDA receptors, which
Each step of the excitotoxic cascade might be an attractive
target for the development of neuroprotective agents poten-
tially useful for treating many chronic and acute brain dis-
orders.
Riluzole (1) or 2-amino-6-(trifluoromethoxy)benzothiazole
(Chart 1) possesses neuroprotective effects in animal models
of PD,5 Huntington’s disease,6 and cerebral ischemia,7 and it
is currently the only drug that has proven to be able to modify
the course of the disease8-10 and is approved for the treatment
of ALS. Riluzole conducts multiple molecular actions in vitro,
among which those that have been documented to occur at
physiologically realistic drug concentrations, and are there-
fore most likely to be clinically relevant, are inhibition of
voltage-gated sodium channels,11-14 which can lead to re-
duced neurotransmitter release, noncompetitive inhibition of
NMDA receptors,15,16 inhibition of glutamate release,17 and
enhanced astrocytic uptake of extracellular glutamate.18
On the basis of animal models of excitotoxicity and injury,
many neuroprotective agents studied so far target a specific
pathway of the excitotoxic cascade. On the other hand, animal
†In memory of Nello Manganelli, who struggled with amyotrophic
lateral sclerosis and was the major inspirator for this work.
*To whom correspondence should be addressed. Phone: þ39 (0)577
234173. Fax: þ39 (0)577 234333. E-mail: anzini@unisi.it.
a Abbreviations: AD, Alzheimer’s disease; PD, Parkinson’s disease;
ALS, amyotrophic lateral sclerosis; NMDA, N-methyl-D-aspartate;
AMPA, R-amino-3-hydroxyl-5-methyl-4-isoxazolepropionate; L-NAME,
nitro-L-arginine methyl ester; NO, nitric oxide; nNOS, neuronal nitric oxide
synthase; LDH, lactate dehydrogenase; aCSF, artificial cerebrospinal fluid;
OGD, oxygen/glucose deprivation; HVA, high-voltage-activated currents;
ABTS, 2,20-azinobis(3-ethylbenzthiazoline-6-sulfonic acid); sem, standard
error of the mean.
r
pubs.acs.org/jmc
Published on Web 12/01/2009
2009 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 735
Scheme 1a
Chart 1. Title and Reference Compounds
a Reagents and conditions: (i) NH4SCN, PhCH2NMe3Br3, CH3CN,
room temperature, 24 h; (ii) POCl3, CH3CONRR0, dry toluene, reflux,
1-7 h; (iii) RNCS, Et3N, dry toluene, reflux, 25-54 h.
Scheme 2a
studies have shown that combination therapies have synergis-
tic effects.19 Therefore, it might be fruitful to explore novel
treatments based on molecules endowed with multiple
mechanisms of action. This study was intended to develop
novel neuroprotective agents interfering with different bio-
chemical steps of the pathways which lead to neuronal injury,
to prevent or minimize tissue damage.
a Reagents and conditions: (i) CH3I, acetone, 25 ꢀC, 60 h; (ii) NH3(g),
20 min, EtOH, room temperature, 20 h.
CH3CN as previously described25 but in higher yields (80%)
(Scheme 1). The reaction of compound 1 with the appropriate
acetamide in the presence of POCl3 in dry toluene led to the
formation of the desired acetamidines 2a-c. To prevent
degradation and to enhance their stability, we converted
compounds 1 and 2a-c to the corresponding hydrochloride
salts. As for the thiourea derivatives, synthesized as shown in
Scheme 1, compound 1 was treated with the corresponding
proper isothiocyanate upon addition of triethylamine to give
desired products 3a-f.26 In this case, a high-boiling point
solvent such as toluene guaranteed good yields.
The synthesis of compounds 2, 3, and 4 (see Chart 1) closely
related to riluzole (1) was thus performed.
In particular, we designed amidine and guanidine deriva-
tives 2 and 4, respectively (Chart 1), with the aim of conju-
gating the neuroprotective effects of riluzole with the neuro-
protective and anti-inflammatory activity of 1400W,20 and
the NOS-inhibiting properties of aminoguanidine21 and L-
NAME22 (Chart 1). Thiourea derivatives 3 (Chart 1) have
been synthesized because many thioureas were found to be
potent free radical scavengers able to prevent oxidative
damage.23
Finally, guanidines 4a,b were obtained by reaction of the
corresponding thioureas (3a,b) with methyl iodide in acetone
followed by treatment with anhydrous ammonia in ethanol27
(Scheme 2).
The title compounds were tested for their ability to counter-
act the glutamatergic excitotoxic cascade by means of an in
vitro model of cerebral ischemia, and their effects were
compared to those of riluzole as the reference drug. In
particular, the degree of neuroprotection afforded by com-
pounds 2a-e, 3a-f, and 4a,b was assessed in rat cortical brain
slices subjected to oxygen/glucose deprivation (OGD) and
reperfusion. Tissue damage and protection were assessed by
measuring the release of glutamate and lactate dehydrogenase
(LDH), the latter taken as an index of overall cellular injury.
Moreover, compounds3band 3dwereselectedandtested with
respect to their activity on voltage-dependent Naþ and Ca2þ
currents in neurons from rat piriform cortex by means of
whole-cell, patch-clamp recordings: indeed, inhibition of
voltage-gated Naþ channels is an important mechanism of
riluzole neuroprotective action, and excessive Ca2þ influx
through ion channels is a crucial step in glutamate release
and excitotoxicity. The ability of guanidine derivative 2c to
inhibit nNOS in vitro was also tested. Finally, to obtain a
direct evidence that thiourea derivatives have a ROS scaven-
ging capacity in vitro, the total equivalent antioxidant capa-
city (TEAC assay)24 of compounds 3a-d with respect to that
of Trolox, a water-soluble vitamin E analogue, was measured.
Results and Discussion
Neuroprotective Effects of the New Compounds. Neuro-
protection exerted by compounds 2a-e, 3a-f, and 4a,b was
evaluated in an in vitro model that mimics neuronal ischemia
and excitotoxicity, namely rat cortical brain slices subjected
to OGD and reperfusion. The compounds were added to the
reoxygenation buffer in a concentration range of 0.001-100
μM, and their effects were compared with those of riluzole,
taken as the reference drug. As shown in Table 1, riluzole
significantly reduced the levels of OGD- and reoxygenation-
induced LDH and glutamate release. Its effects, however,
followed a “U-shaped” concentration-response curve
typical of a hormetic response, with an “efficacy window”
between 0.1 and 25 μM (Figure 1).
The most effective riluzole concentrations turned out to be
1-25 μM, which were capable of decreasing the release of
glutamate from 0.65 ( 0.04 nmol/mg of wet tissue (OGD) to
0.27 ( 0.02 nmol/mg of tissue (P < 0.01; n=9) and that of
LDH from 18.0 ( 0.9 units/mg of protein (OGD) to 5.38 (
1.01 units/mg of protein (P < 0.01; n = 11). Some of the
tested compounds were effective in reducing OGD-induced
LDH and glutamate release, although with different poten-
cies. In particular, 2c and 3b were the most interest-
ing compounds (Figure 1, panels B and D, respectively).
Chemistry
2-Amino-6-(trifluoromethoxy)benzothiazole (1) was pre-
pared via a one-pot procedure based on the condensation of
4-(trifluorometoxy)aniline with 1 molar equiv of ammonium
thiocyanate and benzyltrimethylammonium tribromide in

736 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Anzini et al.
Table 1. Effects of Riluzole and Compounds 2c and 3a-d on OGD- and Reperfusion-Induced Release of Glutamate (GLU) and Lactate
Dehydrogenase (LDH) in Rat Cortical Brain Slices
GLU release
efficacy windowa (μM)
LDH release
efficacy windowa (μM)
compound
MECb (% ( sem)c
MECb (% ( sem)c
riluzole
2c
3a
1-25
0.1-10
1
1-10 (100.0 ( 5.1)
0.1 (100.0 ( 4.3)
1 (80.1 ( 4.1)
0.01 (94.2 ( 3.8)
0.1 (80.3 ( 3.9)
1 (100.0 ( 3.2)
0.1-100
0.1-25
0.1-10
0.01
10-25 (100.0 ( 4.5)
0.1 (56.9 ( 2.8)
0.1 (73.1 ( 3.5)
0.01 (60.9 ( 2.9)
0.01 (50.0 ( 2.7)
0.01 (32.0 ( 2.2)
3b
3c
0.01-10
0.01-1
0.01-1
0.01
0.01
3d
a The efficacy windows represent the interval of concentrations at which a significant reduction of OGD-induced GLU and LDH release was
observed. b MEC (minimal effective concentration) is the micromolar concentration at which the highest level of reduction was observed. c The value
between parentheses represents the percent of reversion exerted at such concentration; 100% is taken as the return to basal values (control).
Indeed, in the concentration range of 0.1-25 μM (2c) or
0.001-0.1 μM (3b), they fully antagonized OGD-induced
glutamate release while partially preventing that of LDH.
Considering that several pathways leading to cell death are
activated in cerebral ischemia,28 these results suggest that
compounds 2c and 3b interfere selectively with the crucial
steps promoting glutamate release and not with those leading
to LDH leakage.
Compound 3a reduced LDH efflux caused by OGD and
reperfusion by ∼70% (P < 0.01) in the concentration range
between 0.1 and 10 μM, whereas it was less efficient against
glutamate release (Figure 1C). Finally, 3c and 3d completely
reversed glutamate release in the concentration range
between 0.01 and 1 μM but were largely ineffective toward
LDH efflux (panels E and F of Figure 1, respectively).
On the other hand, compounds 2a,b,d,e, 3e,f, and 4a,b
were not able to significantly antagonize OGD-induced
glutamate and LDH release, when added during a 90 min
reperfusion phase. Moreover, concentrations of such com-
pounds between 50 and 100 μM promoted a significant
glutamate release (data not shown).
nNOS Inhibition and Antioxidant Activity. To gain insight
into the mechanism of action of the new compounds, some
potential pharmacological effects expected on the basis of
the molecules’ chemical design were then evaluated. In parti-
cular, the ability of amidine derivative 2c to inhibit nNOS in
vitro was also assessed by using partially purified rat brain
enzyme.29 nNOS activity was determined by measuring the
conversion of oxyhemoglobin to methemoglobin by NO at a
λ of 401 nm.30 However, our results indicate that 2c did not
significantly inhibit nNOS activity (data not shown),
suggesting that its neuroprotective activity was not related
to this effect.
Then, the total antioxidant activity of compounds 3a-d
was evaluated by means of the ABTS cation decolorization
assay.24 This method is based on the ability of antioxidant
molecules to quench the stable radical cation, ABTS•þ, a
blue-green chromophore with characteristic absorption at
734 nm. The concentration of antioxidant giving the same
percent ABTS•þ absorbance inhibition of 1 mM Trolox was
calculated (Trolox equivalent antioxidant activity, TEAC).
A TEAC value close to 1.00 indicated that the antioxidant
had a scavenging activity similar to that of Trolox.
strongly indicate that compounds 3a-d can directly sca-
venge ROS and suggest that this activity might contribute to
their neuroprotective effects.
Taken together, the results reported above indicate that
compounds 2c and 3a-d displayed interesting neuroprotec-
tive properties but different potencies. Their global effects,
however, are inferior to those of riluzole, as indicated by the
biochemical assay of LDH release taken as an index of
overall cellular injury in this study. Indeed, none of them
was able to completely antagonize the efflux of the intracellu-
lar enzyme. In the light of these considerations, it could be
worth further characterizing the molecular mechanism(s)
underlying the effects of compounds 2c and 3a-d toward
ischemia/reperfusion injury as well as the receptor(s) poten-
tially involved.
Effects of Compounds 3b and 3d on Voltage-Dependent
Naþ and Ca2þ Currents. Inhibition of voltage-dependent
Naþ currents through a mechanism involving stabilization
of the Naþ channel’s inactivated state is a well-established
effect of riluzole11-14 and is believed to be an important basis
of this drug’s neuroprotective action. Moreover, in peri-
pheral sensory neurons riluzole also exerts an inhibitory
effect on high-voltage-activated (HVA) Ca2þ currents.32
Indeed, inhibition of voltage-dependent Ca2þ influx under
conditions of metabolic distress and depolarization could
contribute to reducing the level of glutamate release, thus
partly justifying the effects of riluzole and compounds like 3b
and 3d. Therefore, 3b and 3d were tested to evaluate their
activity on neuronal voltage-dependent Naþ and Ca2þ cur-
rents and compare it with that of riluzole.
Patch-clamp33 recordings of voltage-dependent Naþ
currents were performed using neurons of rat piriform cortex
layer II in brain slices as an experimental model (see the
Experimental Section for all details of the patch-clamp
experiments). Differently from acutely dissociated neurons
(see below), this preparation allowed for optimal Naþ
current stability, as required to evaluate the activity of the
drugs examined. The transient Naþ current (INaT) was
studied by applying a current-voltage (I/V) protocol con-
sisting of 19 ms depolarizing step pulses at -75 to 20 mV in
5 mV increments, starting from a holding potential of -80
mV (not shown). During drug application, the INaT ampli-
tude was also monitored by repetitively commanding (once
every 10 s) a single 19 ms step at -20 mV. Riluzole, 3b, or 3d
was applied through the bath perfusion at 50 μM. After a
saturating effect on the INaT amplitude was reached, the
drug’s effect was quantified by measuring INaT inhibition at a
voltage level equal to the I/V peak plus 20 mV (Figure 2A),
rather than at the peak itself, to minimize the consequences
of space-clamp artifacts (see ref 34). Whereas 50 μM riluzole
All the tested compounds exhibited a TEAC value ranging
between 0.98 and 0.50 [0.88 ( 0.17 for 3a, 0.50 ( 0.05 for 3b,
0.64 ( 0.02 for 3c, and 0.98 ( 0.14 for 3d (Table 2)], whereas
riluzole was ineffective. Interestingly, the antioxidant ebse-
len, which protects neuronal cells against oxidative stress by
acting at multiple steps,31 had a TEAC value much lower
than those of the compounds mentioned above. These results

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 737
Figure 1. Effects of riluzole (A) and compounds 2c (B), 3a (C), 3b (D), 3c (E), and 3d (F) on oxygen/glucose deprivation- and reoxygenation-
induced release of GLU and LDH, in rat cortical slices. Drugs were added to reoxygenation buffer after OGD for 30 min. Data are means ( sem
of at least four different experiments. Statistical analysis was performed by using ANOVA followed by a post hoc Dunnett test. One asterisk
indicates P < 0.05, and two asterisks indicate P < 0.01 vs OGD.
reduced the INaT amplitude by more than 50%, 3b and 3d had
much weaker effects (∼15 and ∼0% inhibition, respectively)
(Figure 2B). The differences were statistically highly signi-
ficant (P < 0.001 with respect to riluzole in both cases).
Hence, the attachment of the 1-propylthioureidic group and,
even more, the 1-butylthioureidic group to the amino group
in position 2 of riluzole’s benzothiazole ring causes a drastic
decrease in the drug’s potency in interacting with voltage-
gated Naþ channels. Therefore, the effects of 3b and 3d
on voltage-dependent Naþ currents were not further char-
acterized.
Table 2. Relative Antioxidant Activities (TEAC)
compound
TEAC valuea
Trolox
ebselen31
riluzole
3a
1.00 ( 0.00
0.32 ( 0.06
not active
0.88 ( 0.17
0.50 ( 0.05
0.64 ( 0.02
0.98 ( 0.14
3b
3c
3d
a Values are presented as means ( sem of at least three experi-
ments.

738 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Anzini et al.
Figure 2. Inhibitory action of compounds 3b and 3d on the voltage-dependent, transient Naþ current (INaT) in rat piriform cortex neurons.
(A) INaT current tracings recorded in three different, representative neurons under control conditions and in the presence of 50 μM 3b (A1), 3d
(A2), or riluzole (A3). The currents shown here were recorded at a test potential equal to the peak of the I/V relationship plus 20 mV (-5
or -10 mV). (B) Average values of INaT amplitude percent inhibition caused by 50 μM 3b (n = 4), 3d (n = 9), or riluzole (n = 6).
Voltage-dependent Ca2þ currents were studied in acutely
dissociated neurons from rat piriform cortex layer II,35 in
which they proved to be highly stable and optimal clamp
conditions could be achieved. Ba2þ was used instead of Ca2þ
as the charge carrier, and Ba2þ currents (IBa) were recorded.
High-voltage-activated (HVA) currents were elicited by
commanding depolarizing step or ramp protocols starting
from a conditioning potential of -60 mV. 3b and 3d were
applied to the recorded neuron using a focal perfusion
system.35 Riluzole was also tested as a control. The effects
of compounds 3b and 3d on HVA currents were monitored
by repetitively commanding (once every 7 s) a voltage-clamp
protocol consisting of a 40 ms depolarizing ramp from -60
to 30 mV, which returned an “instantaneous” I/V relation-
ship for total HVA currents (see Figure 3B). This instanta-
neous I/V closely matched that obtained with standard step
I/V protocols (see Figure 3B and the inset). Current ampli-
tude was measured at the peak of the instantaneous I/V.
Both drugs reduced HVA current amplitude in a con-
centration-dependent manner (Figure 3B). This inhibitory
effect fully developed after drug perfusion for ∼1 min
(in Figure 3A, the horizontal bar indicates the time periods
during which control solution and the various drug concen-
trations were perfused) and appeared to be largely irrever-
sible after prolonged drug washout (not shown).
The potency of the two drugs could therefore be somewhat
underestimated in these experiments.
In dorsal-root-ganglion neurons, the inhibitory effect of
riluzole on HVA currents was found to be voltage-indepen-
dent.32 Riluzole was also reported to accelerate the activa-
tion and deactivation kinetics of HVA currents, although, as
far as deactivation is concerned, only at relatively positive
potentials.32 Therefore, the effects of 3d, which appeared to
be the more active compound with respect to HVA currents,
were characterized in more depth and compared with those
of the parental molecule by also analyzing HVA current
voltage dependence and kinetics. The voltage dependence of
HVA currents was studied by applying the I/V protocol
illustrated in Figure 4 (panel A1). The I/V plots obtained in
the presence of riluzole (n=3; not shown) or 3d (Figure 4B)
were largely superimposable with those obtained under
control conditions. The voltage dependence of HVA channel
activation was further quantified by fitting I/V plots with
a combination of a single Boltzmann function and the
Goldman equation (see the Experimental Section, eq 1).
Average values of the fitting parameters thus obtained are
listed in Table 3. Both half-activation potential (V1/2) and the
slope factor (k) were not significantly different in the pre-
sence of riluzole or 3d as compared to control. Therefore,
neither riluzole nor 3d significantly modifies the voltage
dependence of activation of HVA Ca2þ channels in the
central neurons studied here.
Riluzole and 3d also had no consistent effect on HVA
current activation speed, quantified by measuring the cur-
rent rise time of 10-90% (RT10-90) over the whole voltage
range explored [n = 3 for riluzole, and n = 4 for 3d (not
shown)]. Instead, both drugs induced some significant modi-
fication of HVA current deactivation, although at relatively
positive potentials only. This was revealed by the study of the
decay kinetics of tail currents elicited upon repolarization
(Figure 5A-C). As shown in Figure 5 (panels D and E1), 3d
(60 μM) significantly accelerated the “fast” deactivation time
constant (τd1) at the most positive repolarization potential
examined (-20 mV), but not at more negative voltages.
The percent HVA current inhibition as a function of drug
concentration is illustrated in Figure 3C. Experimental data
points were fitted with a Hill function of the form y=100 ꢀ
n
[D]n/(IC50 þ [D]n), where [D] is the drug molar concentra-
tion and n is the Hill coefficient. Fitting parameters were as
follows: IC50=112.9 μM, n=0.58 (3b); IC50=24.1 μM, n=
0.82 (3d); IC50 =34.2 μM, n=0.82 (riluzole). It should be
noted that both 3b and 3d, but not riluzole, tended to
precipitate when added to the extracellular Ca2þ current
recording solution at the highest concentrations employed
(60 and 200 μM), thus yielding actual concentrations lower
than nominal ones. This could explain the poor steepness of
the dose-response curve observed especially for 3b, in which
case the Hill coefficient returned by fitting was as low as 0.58.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 739
Figure 3. Inhibitory action of compounds 3b and 3d on high-voltage-activated Ca2þ currents in rat piriform cortex neurons. (A) Time course
of the effects of increasing 3d concentrations on the peak amplitude of total HVA currents in a representative neuron. HVA currents
were evoked by ramp depolarizations (see the text). The filled circles correspond to the peak amplitudes of the representative current tracings
shown in panel B. (B) Representative currents recorded in the same neuron of panel A under control conditions and in the presence of a
steady inhibition by increasing 3d concentrations. The inset shows the currents recorded in the same neuron by applying, under control
conditions, a step I/V protocol (the voltage protocol is illustrated in the top part of the inset). Peak current amplitudes of these currents are
indicated by the filled squares in the main panel. Calibration bars in the inset are for 10 ms and 100 pA. (C) Dose-response curve for the
inhibitory effect exerted by 3b, 3d, and riluzole on the total HVA current peak amplitude. Smooth lines are fittings with Hill functions (see the
text).
Figure 4. Compound 3d does not modify the voltage dependence of activation of HVA Ba2+ currents. (A) Ba2+ currents evoked by a step I/V
protocol (A1) in a representative neuron under control conditions (A2) and in the presence of 60 μM 3d (A3). (B) Average I/V relationships of
Ba2+ currents recorded, in response to the protocol illustrated in A1, in the absence and presence of 60 μM 3d. In each cell, the peak current
amplitude was measured at every test potential and then normalized for the maximal values observed under control conditions (normally at
0 mV). Current values thus normalized were then averaged among cells, and the resulting average plots are shown in panel B1. Peak current
values were also normalized for the maximal values observed under each experimental condition (control, 3d) and averaged among cells: the
resulting average plots are shown in panel B2. Each data point is the average from five cells.

740 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Anzini et al.
A similar effect was caused by riluzole (60 μM) (Figure 5,
panel E2). Hence, 3d modifies HVA Ca2þ channel deactiva-
tion kinetics in a manner similar to that of riluzole in the
central neurons studied here.
Overall, our results indicate that compounds 3b and 3d do
act on HVA Ca2þ channels, but at concentrations consider-
ably higher than those required to antagonize OGD-induced
glutamate and LDH release, and that their activity, if any, on
voltage-gated Naþ channels is much weaker than that of
riluzole. This suggests that at the basis of the neuroprotective
effects of these new compounds there are mechanisms prob-
ably independent of voltage-gated Naþ and Ca2þ channels.
Conclusions
A series of amidine, guanidine, and thiourea derivatives of
2-amino-6-(trifluoromethoxy)benzothiazole have been syn-
thesized, and their ability to counteract the excitotoxic cas-
cade has been evaluated. In an in vitro protocol of ischemia/
reperfusion injury, some of them significantly attenuated
neuronal injury in a concentration range between 0.1 and
25 μM. Furthermore, thioureas 3b and 3d were also tested to
evaluate their activity on voltage-dependent Naþ and Ca2þ
currents in neurons from rat piriform cortex. At 50 μM, 3b
inhibited Naþ currents to a much lesser extent than riluzole,
whereas 3d was almost inactive. Both compounds inhibited
high-voltage-activated Ca2þ currents in a concentration-de-
pendent manner, with an IC50 of 24.1 μM for 3d and 112.9 μM
Table 3. Parameters of HVA Current Voltage Dependence of Activa-
tion under Control Conditions and in the Presence of Compound 3d or
Riluzole
V1/2 (mV)a
Pb
k (mV)a
Pb
n
control
riluzole (60 μM)
control
-6.8 ( 2.0
-6.5 ( 2.2
-7.2 ( 1.8
-7.5 ( 1.7
7.9 ( 0.6
8.3 ( 0.5
7.5 ( 0.4
7.9 ( 0.6
0.75
0.55
0.16
0.17
3
5
3d (60 μM)
a These parameters were returned by fitting the I/V relationships of
HVA currents with a combination of the Boltzmann function and the
Goldman equation (see the Experimental Section, eq 1). b Level of
statistical significance of the differences observed between drug and
control (t test for paired data).
Figure 5. Effects of compound 3d and riluzole on HVA IBa deactivation kinetics. (A and B) Voltage-clamp protocol applied to evoke tail
currents upon repolarization (A) and Ba2+ currents thus recorded in a representative neuron under control conditions (B1) and in the presence
of 60 μM 3d (B2). (C) Detail of the tail currents recorded at the repolarization potential of -20 mV. The currents’ decay phase has been fitted
with a second-order exponential function (see the Experimental Section, eq 2) (blue lines). Fitting parameters are as follows: A1 = -406.8 pA,
τ
d1 = 0.823 ms, A2 = -206.8 pA, τd2 = 2.62 ms, C = -374.4 pA (control); A1 = -308.0 pA, τd1 = 0.647 ms, A2 = -152.5 pA, τd2 = 2.79 ms,
C = -336.2 pA (3d). (D) Further detail of tail currents shown in panel C, after zeroing of the offset level and normalization for peak amplitude.
Note the slightly faster decay kinetics in the presence of 3d. (E) Plot of the average, “weighted” fast deactivation time constant (τd1) (see the
Experimental Section, eq 3, for details) as a function of repolarization potential, under control conditions and in the presence of 60 μM 3d (E1)
or riluzole (E2) (n = 4 in both cases). One asterisk indicates P < 0.05 (t test for paired data).

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 741
for 3b. These observations strongly suggest that at the basis of
neuroprotective effects of 3b and 3d are mechanisms that
probably do not importantly involve voltage-gated Naþ and
Ca2þ channels.
Finally, compounds 3a-d showed additional antioxidant
properties, measured as Trolox equivalent activity, suggesting
that they are also endowed with a direct ROS scavenging
ability probably responsible for their ability to counteract
ischemia-induced injury in vitro.
Further studies will be needed to characterize the neuro-
protective effects of these compounds and to determine
whether they could provide a basis for new therapeutic app-
roaches to brain diseases.
(i) N0-[6-(Trifluoromethoxy)benzothiazol-2-yl]acetamidine (2a).
Reflux for 6 h:light-yellow solid(67%yield); 1H NMR(CDCl3) δ
2.35 (s, 3H), 5.62 (br s, 2H), 7.20 (m, 2H), 7.48 (d, 1H, J=8.9);
MS-ESI m/z 276 (M þ Hþ).
(ii) N0-[6-(Trifluoromethoxy)benzothiazol-2-yl]acetamidine
Hydrochloride. White square platelets from a methanol/diethyl
ether mixture: mp 202-206 ꢀC.
(iii) N-Methyl-N0-[6-(trifluoromethoxy)benzothiazol-2-yl]aceta-
midine (2b). Reflux for 7 h: yellow solid (21% yield); 1H NMR
(CDCl3) δ 2.21 (s, 3H), 3.11 (d, 3H, J=5.1), 7.18 (d, 1H, J=8.3),
7.58 (m, 2H), 10.68 (br s, 1H); MS-ESI m/z 290 (M þ Hþ).
(iv) N-Methyl-N0-[6-(trifluoromethoxy)benzothiazol-2-yl]aceta-
midine Hydrochloride. Light-yellow square platelets from metha-
nol: mp 202 ꢀC.
(v) N,N-Dimethyl-N0-[6-(trifluoromethoxy)benzothiazol-2-yl]-
acetamidine (2c). Reflux for 1 h: light-yellow solid (34% yield);
1H NMR (CDCl3) δ 2.25 (s, 3H), 3.06 (s, 6H), 7.14 (d, 1H, J=
8.7), 7.48 (s, 1H), 7.61 (d, 1H, J=8.8); MS-ESIm/z 304 (M þ Hþ).
(vi) N,N-Dimethyl-N0-[6-(trifluoromethoxy)benzothiazol-2-yl]-
acetamidine Hydrochloride. Light-orange square platelets from
methanol: mp 219-222 ꢀC.
Experimental Section
Chemistry. All chemicals used were of reagent grade. Yields
refer to purified products and are not optimized. Melting points
were determined in open capillaries on a Gallenkamp apparatus
and are uncorrected. Microanalyses were conducted by means
of a Perkin-Elmer 240C or a Perkin-Elmer Series II CHNS/O
Analyzer 2400 to confirm the purity (g95%) of the target
compounds used in the biological testing. Merck silica gel 60
(230-400 mesh) was used for column chromatography. Merck
(vii) N,N-Diethyl-N0-[6-(trifluoromethoxy)benzothiazol-2-yl]-
acetamidine (2d). Reflux for 3 h: light-yellow oil (58% yield);
1H NMR (CDCl3) δ 1.21 (t, J=7.1, 6H), 2.31 (s, 3H), 3.49 (br s,
4H), 7.15 (d, J=8.8, 1H), 7.49 (s, 1H), 7.63 (d, J=8.8, 1H); MS-
ESI m/z 332 (M þ Hþ).
1
TLC plates (silica gel 60 F254) were used for TLC. H NMR
(viii) N,N-Dipropyl-N0-[6-(trifluoromethoxy)benzothiazol-2-yl]-
acetamidine (2e). Reflux for 3 h: yellow oil (52% yield); 1H NMR
(CDCl3) δ 0.92 (t, J=7.2, 6H), 1.66 (m, 4H), 2.30 (s, 3H), 3.36 (br
d, 4H), 7.15 (d, J=9.0, 1H), 7.49 (s, 1H), 7.63 (d, J=8.9, 1H);
MS-ESI m/z 360 (M þ Hþ).
spectra were recorded with a Bruker AC 200 spectrometer in the
indicated solvent (TMS as internal standard): the values of the
chemical shifts (δ) are expressed in parts per million and the
coupling constants (J) in hertz. Mass spectra were recorded on
either a Varian Saturn 3 spectrometer or a ThermoFinnigan
LCQ-deca instrument.
General Procedure for the Preparation of Thiourea Derivatives
(3a-f). To a solution of 1 (2.7 mmol) in dry toluene (20 mL) were
added triethylamine (0.43 mmol) and the suitable amount of
isothiocyanate (3.78 mmol). The reaction mixture was refluxed
for the suitable time and then cooled. The solvent was removed
under reduced pressure, and the residue, diluted with CH2Cl2,
was washed with water, dried over Na2SO4, filtered, and eva-
porated under reduced pressure. The resulting residue was
purified by flash chromatography with a petroleum ether/ethyl
acetate (65:35, v/v) mixture to give a solid that after recrystal-
lizzation from ethyl acetate and diethyl ether afforded the
required product.
2-Amino-6-(trifluoromethoxy)benzothiazole (1). To a mixture
of 4-trifluoromethoxyaniline (2.1 mmol) and ammonium thio-
cyanate (2.0 mmol) in CH3CN (15 mL) was added benzyltri-
methylammonium tribromide (2.0 mmol) at room temperature.
The reaction mixture was stirred for 24 h, neutralized with
aqueous NaHCO3, and extracted with CH2Cl2. The combined
organic layers were washed with water, dried over Na2SO4,
filtered, and evaporated in vacuo. The crude material was
purified by flash chromatography on silica gel eluting with a
petroleum ether/ethyl acetate (1:1, v/v) mixture to give a solid
that after recrystallization from n-hexane and dietyl ether gave
light-yellow square platelets in 80% yield.
(i) 1-Ethyl-3-[6-(trifluoromethoxy)benzothiazol-2-yl]thiourea
(3a). Reflux for 30 h: light-yellow needles (36% yield); mp
1
The analytical data, melting point, and H NMR spectrum
1
were consistent with those reported in the literature.36
199-200 ꢀC; H NMR (DMSO-d6) δ 1.17 (t, 3H, J = 7.20),
3.55 (m, 2H), 7.37 (d, 1H, J=8.4), 7.70 (d, 1H, J=8.7), 8.02 (s,
2-Amino-6-(trifluoromethoxy)benzothiazole Hydrochloride.
To a solution of compound 1 (1.3 mmol) in methanol (10 mL)
was carefully added a concentrated 37% HCl solution
(1 mL). The resulting solid was recrystallized from methanol
to give white needles in quantitative yield. Crystals were
filtered, washed with dry diethyl ether, and dried: mp 214-
216 ꢀC.
General Procedure for the Preparation of Acetamidine Deri-
vatives (2a-e). To an ice-cooled solution of phosphorus oxy-
chloride (0.85 mmol) in dry toluene (20 mL) was added the
suitable amount of acetamide (0.47 mmol), and the mixture was
stirred for 30 min at room temperature. At the end of the
reaction, compound 1 (0.42 mmol) dissolved in dry toluene
was added dropwise and the reaction mixture was refluxed for
the proper time. The solution was then cooled, carefully poured
into ice-water, and made alkaline with a 1 N NaOH solution.
The organic layer was extracted with CHCl3, washed to neu-
trality with water, dried over Na2SO4, filtered, and then evapo-
rated in vacuo. The crude material was purified by flash chro-
matography on silica gel eluting with a petroleum ether/ethyl
acetate (4:6, v/v) mixture, and the expected compound was
transformed into the corresponding hydrochloride salt [as des-
cribed for 2-amino-6-(trifluoromethoxy)benzothiazole hydro-
chloride] and then crystallized from the indicated solvent.
1H), 9.54 (br s, 2H); MS-ESI m/z 320 (M - Hþ).
(ii) 1-Propyl-3-[6-(trifluoromethoxy)benzothiazol-2-yl]thiourea
(3b). Reflux for 25 h: yellow needles (40% yield); mp 226-228
ꢀC; 1H NMR (DMSO-d6) δ 0.90 (t, 3H, J=7.3), 1.57 (m, 2H),
3.48 (m, 2H), 7.35 (d, 1H, J=8.5), 7.79 (d, 1H, J=44.6), 8.01 (s,
1H), 9.62 (br s, 1H), 11.82 (br s, 1H); MS-ESI m/z 334 (M - Hþ).
(iii) 1-Isopropyl-3-[6-(trifluoromethoxy)benzothiazol-2-yl]thiourea
(3c). Reflux for 54 h: white needles (23% yield); mp 242 ꢀC; 1H
NMR (DMSO-d6) δ 1.21 (d, 6H, J=6.0), 4.33 (m, 1H), 7.33 (d,
1H, J=8.6), 7.69 (d, 1H, J=8.5), 7.99 (s, 1H), 9.36 (br s, 1H),
11.70 (br s, 1H); MS-ESI m/z 334 (M - Hþ).
(iv) 1-Butyl-3-[6-(trifluoromethoxy)benzothiazol-2-yl]thiourea
(3d). Reflux for 52 h: white needles (27% yield); mp 211 ꢀC; 1H
NMR (DMSO-d6) δ 0.88 (t, 3H, J=7.2), 1.33 (m, 2H), 0.92 (m,
2H), 3.50 (m, 2H), 7.33 (d, 1H, J=8.5), 7.66 (d, 1H, J=8.5), 7.98
(s, 1H), 9.53 (br s, 1H), 11.80 (br s, 1H); MS-ESI m/z 348 (M -
Hþ).
(v) 1-Phenyl-3-[6-(trifluoromethoxy)benzothiazol-2-yl]thiourea
(3e). Reflux for 30 h: white needles (22% yield); mp 234 ꢀC; 1H
NMR (DMSO-d6) δ 7.15 (m, 1H), 7.33 (m, 3H), 7.63 (m, 3H),
7.98 (s, 1H), 10.76 (br s, 2H); MS-ESI m/z 368 (M - Hþ).
(vi) 1-(4-Fluorophenyl)-3-[6-(trifluoromethoxy)benzothiazol-
2-yl]thiourea (3f). Reflux for 42 h: light-yellow needles (20%

742 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Anzini et al.
yield); mp 207 ꢀC; 1H NMR (DMSO-d6) δ 7.14 (m, 2H), 7.40 (m,
1H), 7.62 (m, 3H), 7.97 (s, 1H), 10.72 (br s, 2H); MS-ESI m/z 386
(M - Hþ).
are reported as mean ( sem, and n is defined as the number of
samples. For glutamate and LDH activity, data are expressed as
nanomoles per milligram of tissue and units per milligram of
tissue, respectively. One unit of LDH activity is defined as that
which gives rise to 1 μmol of lactate/min.
Statistical analysis was performed by using one-way ANOVA
followed by the post hoc Dunnett test (GraphPad INSTAT
version 3.00, GraphPad Software, San Diego, CA).
Interaction of the Compounds with Nitric Oxide Synthase. The
interactions of studied compounds with neuronal nitric oxide
synthase were performed on rat brain partially purified prepara-
tions according to the method reported by Chen et al.40 Briefly,
rat brain homogenized in buffer (5 mL/g of tissue) containing
(final concentrations) 320 mM sucrose, 20 mM HEPES [N-
(2-hydroxyethyl)piperazine-N0-2-ehanesulfonic acid], 1 mM
EDTA (ethylenediaminetetraacetic acid), 1 mM dithiothreitol,
10 μg/mL soybean trypsin inhibitor, and 10 mg/mL PMSF
(phenylmethanesulfonyl fluoride) (pH 7.4) was centrifuged at
45000 rpm for 30 min and the supernatant collected. All pro-
cedures were performed at 4 ꢀC. To remove endogenous L-
arginine from the preparation, the supernatant was mixed with
Dowex 50W (200-400 mesh 8% cross-linked) for 10 min and
shaken. After that, the mixture was centrifuged at 9000 rpm for
5 min at 4 ꢀC and the supernatant aliquoted out and kept at
-80 ꢀC until it was used. The protein content was determined
according to the method of Lowry et al.41
Determination of nNOS Activity. The reaction was started by
adding 0.02 mM L-arginine to the assay mixture [50 mM
potassium phosphate buffer, pH 7.2 (final concentration)
1.15 μM oxyhemoglobin, 0.2 mM CaCl2, 0.1 mM NADPH,
and rat brain NOS preparations (2.0 mg/mL)] and followed by
reading the change in absorbance versus time at 401 nm at the
initial linear phase of the reaction.30
Oxyhemoglobin was purified from commercial hemoglobin
(Sigma) by dissolving 20 mg in 1 mL of distilled water and
adding sodium dithionite and oxygen over surface. Purification
and desalting were carried out by eluting 300 μl of hemoglobin
solution from a sephadex G-25 column (1 ꢀ 17 cm) using 1 M
NaCl solution as eluent. The concentration of the hemoglobin
was determined by measuring the absorbance at 415 nm (ε=131
mM-1 ꢀ cm-1).
General Procedure for the Preparation of Guanidine Deriva-
tives (4a,b). To a solution of the suitable thiourea (0.76 mmol) in
acetone (40 mL) was added methyl iodide (0.84 mmol), and the
mixture was stirred at room temperature under a nitrogen
atmosphere. At the end of the reaction, the solvent was removed
under reduced pressure, and the resulting residue was dissolved
in ethanol and anhydrous ammonia bubbled for 20 min. The
reaction flask was sealed, and the solution was stirred overnight.
The solution was then concentrated in vacuo, and the residue
was diluted with CH2Cl2 and washed with water. The combined
organic layers were dried over Na2SO4, filtered, and concen-
trated in vacuo. The product was purified by flash chromato-
graphy using a petroleum ether/ethyl acetate (65:35, v/v) mix-
ture as eluent. Recrystallization from n-hexane and diethyl ether
gave the expected compound.
(i) 1-Ethyl-3-[6-(trifluoromethoxy)benzothiazol-2-yl]guanidine
1
(4a). White pearly crystals (20% yield): mp 127-129 ꢀC; H
NMR (CDCl3) δ 1.27 (m, 3H), 3.28 (m, 2H), 6.28 (br s, 3H), 7.11
(d, 1H, J=8.9), 7.24 (s, 1H), 7.47 (d, 1H, J=8.8); MS-ESI m/z
305 (M þ Hþ).
(ii) 1-Propyl-3-[6-(trifluoromethoxy)benzothiazol-2-yl]guanidine
1
(4b). White pearly crystals (22% yield): mp 110-111 ꢀC; H
NMR (CDCl3) δ 1.03 (t, 3H, J=7.4), 1.70 (m, 2H), 3.19 (q, 2H,
J=6.2), 6.22 (br s, 3H), 7.11 (d, 1H, J=7.2), 7.24 (s, 1H), 7.46 (d,
1H, J=8.8); MS-ESI m/z 319 (M þ Hþ).
Neuroprotection Experiments. (i) Preparation of Animals and
Slices. All experiments were performed in strict compliance with
the recommendation of the EEC (86/609/CEE) for the care and
use of laboratory animals, and the protocols were approved by
the Animal Care and Ethics Committee of the University of
Siena. Male Sprague-Dawley rats (350-450 g) (Charles River
Italia, Calco, Italy) were sacrificed after anesthesia (ip injection
of 30 mg of ketamine hydrocloride/kg of body weight and 8 mg
of xylazine chloride/kg body weight), and the whole brains were
rapidly removed and placed in aCSF (120 mM NaCl, 2.5 mM
KCl, 1.3 mM MgCl2, 1.0 mM NaH2PO4, 1.5 mM CaCl2, 26 mM
NaHCO3, and 11 mM glucose, saturated with a 95% O2/5%
CO2 mixture, with a final pH of 7.4, and osmolality of 285-290
mosM). The cortex was dissected and cut into 400 μm thick slices
by using a manual chopper (Stoelting Co., Wood Dale, IL).
Afterward, slices were maintained in oxygenated aCSF enriched
with 400 μM ascorbic acid for 1 h at room temperature to allow
maximal recovery from slicing trauma.37
Antioxidant Activity. The antioxidant activities of the studied
compounds were assayed by measuring spectrophotometrically
the quenching of the stable cation radical ABTS3 þ as reported
by Re et al.24 Briefly, the cation radical was obtained by reacting
ABTS with ammonium persulfate (1:0.35) in a water solutioþn
for 24 h in the dark at room temperature. Further, the ABTS3
solution diluted with ethanol to an absorbance value of 0.70 (
0.02 at 751 nm was incubated in the presence of different
concentrations (1-100 μM) of 3a-d compounds or Trolox, as
a reference compound. The decrease in absorbance at a λ of
751 nm was recorded 5 min after ABTS and compound(s) had
been initially mixed and followed for 10 min. The data are
reported as Trolox equivalent antioxidant capacity (TEAC).
TEAC represents the ratio between the estimated concentration
of the compound that promotes the decrease of radical cation
ABTS•þ absorbance equal to 1 mM Trolox. The absorbance
value for Trolox or compounds was estimated from the inhibi-
tion curve obtained by plotting the concentration versus absor-
bance (GraphPad Software) to determine the TEAC.
Patch-Clamp Experiments. Whole-cell, patch-clamp record-
ings were conducted in layer II pyramidal neurons of rat
piriform cortex (PC), either in slices or after acute dissociation.
Young (P15-P22) Wistar rats of either sex were used, follow-
ing a protocol that conformed with the rules established
by the University of Pavia for the use of animals in experimental
studies, in compliance with the guidelines of the Italian
Ministry of Health, the national laws on animal research (d.l.
116/92), and the EU guidelines on animal research (N. 86/609/
CEE).
(ii) In Vitro Ischemia-like Conditions. Cortical slices (≈4-5,
total wet weight of 33.6 ( 2.6 mg, n=10) were placed in covered
incubation flasks, containing 2 mL of aCSF continuously
bubbled with a 95% O2/5% CO2 mixture and incubated at
37 ꢀC for an additional period of 30 min. Afterward, oxygen/
glucose deprivation was conducted by incubating slices for
30 min in aCSF in which glucose was replaced with an equimolar
amount of saccharose and continuously bubbled with a 95% N2/
5% CO2 mixture. After the oxygen/glucose deprivation period,
the ischemic solution was replaced with fresh, oxygenated aCSF
for an additional 90 min (reoxygenation). In treated samples,
oxygen/glucose deprivation was followed by reoxygenation with
aCSF containing riluzole or the compounds under study.
(iii) Assessment of Neuronal Injury. Neuronal damage was
assessed quantitatively by measuring the amount of both gluta-
mate and LDH released into the aCSF during 90 min reoxy-
genation period. In particular, glutamate was measured fluori-
metrically (excitation at 366 nm; emission at 450 nm) using the
conversion of NADþ to NADH by glutamate dehydrogenase,38
whereas LDH activity was determined spectrophotometrically
by measuring the rate of the decrease in absorbance at 340 nm
via the oxidation of NADH to NADþ.39
(iv) Data Analysis. All the experiments were performed by
using brain slices derived from at least four different rats. Data

Article
(i) Recordings of Voltage-Dependent Naþ Currents from Neu-
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2 743
significant effects on Ca2þ currents when applied at the same
concentrations used to dissolve drugs in aqueous solution
(0.02-0.67%, v/v).
rons in Slices. The procedures followed and the solutions used
for extracting and sectioning the brain were the same as those
described previously.42 The rostral part of the brain, containing
the anterior piriform cortex, was glued on the stage of a
vibratome, and 350 μm thick coronal sections were cut. The
slices were kept submerged in the incubation chamber at room
temperature for at least 1 h before the recording was begun.
The experimental apparatus employed and the recording
procedures adopted for patch-clamp experiments in slices were
also identical to those described elsewhere.42 Slices were per-
fused with an extracellular solution suitable for the isolation of
Naþ currents and containing 100 mM NaCl, 26 mM NaHCO3,
19.5 mM tetraethylammonium chloride (TEA-Cl), 3 mM KCl,
2 mM MgCl2, 2 mM CaCl2, 2 mM BaCl2, 0.5 mM CdCl2, 4 mM
4-aminopyridine (4-AP), and 11 mM D-glucose (pH 7.4 by
saturation with a 95% O2/5% CO2 mixture). Patch pipettes
were filled with an intracellular solution containing 104 mM
CsF, 50 mM TEA-Cl, 2 mM MgCl2, 10 mM N-(2-hydro-
xyethyl)piperazine-N0-2-ethanesulfonic acid (HEPES), 10 mM
ethylene glycol bis(β-aminoethyl ether) N,N,N0,N0-tetraacetic
acid (EGTA), 2 mM adenosine 50-triphosphate (ATP)-Na2, and
0.2 mM guanosine 50-triphosphate (GTP)-Na (pH adjusted to
7.2 with CsOH).
The general holding potential of voltage-clamp recordings
was -80 mV. Current signals were low-pass filtered and digi-
tized at cutoff and sampling frequencies of 5 and 50 kHz,
respectively. Currents were always online leak subtracted
via a P/4 routine. Tetrodotoxin (TTx) (Alomone Laboratories,
Jerusalem, Israel) was applied in the bath with the superfusing
solution at the end of all recordings, and currents recorded in the
presence of 1 μM TTx were always subtracted from those
recorded under control conditions and in the presence of drugs
to abolish residual, unsubtracted capacitive and/or leakage
currents.
In experiments for recording Naþ currents in slices, the
drug-containing extracellular solution was applied through the
bath perfusion. In experiments for recording Ca2þ currents in
acutely dissociated neurons, drugs were applied through a local-
perfusion system consisting of a multibarrel pipet (diameter at
the tip of ∼150 μm), each barrel of which was connected to a
separate perfusion channel. When each channel was opened,
controlled by operation of on remote-commanded electrovalves
(Sirai, Milano, Italy), the drug-containing solution flowed by
gravity, thus forming a laminar-flux cone. The tip of the
perfusion pipet was positioned in the proximity of the recording
site, so that the recorded cell was fully drenched by the laminar-
flux cone.
(iv) Data Analysis. Whole-cell current signals were analyzed
using Clampfit from pClamp version 8.2 (Axon Instruments,
Union City, CA). Naþ currents were normally refiltered off-line
at 3.5 kHz. Current amplitude was measured at the peak of each
tracing. Current-voltage (I/V) plots for IBa were fitted by apply-
ing the following function, which is a combination of the
Boltzmann function and the Goldman equation:
IBa ¼ ðPBaðmaxÞ=f1 þ exp½ðVm -V1=2Þ=kꢁgÞ
ꢀ ð4F2VmÞ=ðRTÞ
ꢀ f½Ba2þꢁi -½Ba2þꢁo
ꢀ exp½ð -2FVmÞ=ðRTÞꢁg=f1 -exp½ð -2FVmÞ=ðRTÞꢁg
(1Þ
in which the nominal intra- and extracellular Ba2þ concentra-
tion values (0 and 5 mM, respectively) were introduced and
where PBa(max) is the maximal Ba2þ permeability, Vm is the
membrane potential, V1/2 and k are the half-maximal activation
potential and slope factor, respectively, of the Boltzmann
activation function, and F, R, and T have their usual meanings.
Tail-current decay could be consistently best fitted with a
second-order exponential function:
(ii) Recordings of Voltage-Dependent Ca2þ Currents from
Acutely Dissociated Neurons. The procedures followed for ex-
traction of the nervous tissue and dissection of PC layer II were
described previously.35 We acutely dispersed neurons by apply-
ing a mechanical and enzymatic dissociation procedure also
described previously.35After being seeded in the recording
chamber, cells were perfused with an oxygenated extracellular
solution suitable for isolating Ba2þ currents conducted through
Ca2þ channels, containing 88 mM choline-Cl, 40 mM TEA-Cl,
3 mM KCl, 2 mM MgCl2, 5 mM BaCl2, 3 mM CsCl, 10 mM
HEPES, 5 mM 4-AP, and 25 mM D-glucose (pH 7.4 with HCl).
Patch pipettes were filled with an intracellular solution contain-
ing 78 mM Cs methanesulfonate, 40 mM TEA-Cl, 10 mM
HEPES, 10 mM EGTA, 20 mM phosphocreatine di-Tris salt,
2 mM ATP-Na2, and 20 units/mL creatine phosphokinase (pH
adjusted to 7.2 with CsOH). Tight seals (>5 GΩ) and the whole-
cell configuration were obtained according to the standard
technique.33 All recording conditions were as previously de-
scribed.35
I ¼ A1 ꢀ expð -t=τd1Þ þ A2 ꢀ expð -t=τd2Þ þ C
(2Þ
Because at relatively positive repolarization potentials the
slower exponential component could be influenced by channel
inactivation, the analysis was focused on the faster deactivation
time constant (τd1). Since the importance of τd1 in governing the
deactivation process is determined not only by its value but also
by its relative weight, to obtain an overall description of the fast
component of tail current decay we calculated the following
parameter:
τd1 ¼ τd1=½A1=ðA1 þ A2Þꢁ
namely, the fast time constant divided by its relative amplitude.
(3Þ
Hence, the lower τd1 is and the higher its relative weight, the
smaller τd1, which represents a “weighted” version of the fast
h
The general holding potential of voltage-clamp recordings
was -70 mV. Either step or ramp protocols were commanded
to activate voltage-dependent Ca2+ currents. In both cases, a 2 s
conditioning prepulse at -60 mV preceded each test depolariza-
tion. Current signals were filtered at 5 kHz and digitized at
50 kHz. Step protocols were online leak subtracted via a P/4
protocol.
deactivation time constant.
Data fittings with eqs 1 and 2 were conducted using Origin
version 6.0 (MicroCal Software, Northampton, MA) and
Clampfit version 8.2, respectively.
Average values were expressed as arithmetic means ( sem.
Statistical significance was evaluated by applying the two-tail
Student’s t test for paired or unpaired data.
(iii) Drug Application. Concentrated (30 mM) stock solutions
of riluzole and compounds 3b and 3d were prepared in DMSO,
divided into small aliquots, and stored at -20 ꢀC. The aliquots
were then diluted to the final concentrations in one of the
extracellular solutions described above. Because an insoluble
precipitate formed when 3b and 3d stocks were dissolved in this
solution at high concentrations (60-200 μM), 3b- and 3d-
containing solutions were always filtered before application.
Preliminary control experiments indicated that DMSO had no
Acknowledgment. We thank Rottapharm SpA (Monza,
Italy) for financial support of this research and Prof. Stefania
D’Agata D’Ottavi for the careful reading of the manuscript.
This work has been performed with the financial support of
the University of Siena, Piano di Ateneo per la Ricerca (PAR)
2006.

744 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 2
Anzini et al.
Supporting Information Available: Microanalytical data of
compounds 2-4. This material is available free of charge via the
Internet at http://pubs.acs.org.
(23) Kim, S. G.; Kim, H. J.; Yang, C. H. Thioureas differentially induce
rat hepatic microsomal epoxide hydrolase and rGSTA2 irrespec-
tive of their oxygen radical scavenging effect: Effects on toxicant-
induced liver injury. Chem.-Biol. Interact. 1999, 117, 117–134.
(24) Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M.;
Rice-Evans, C. Antioxidant activity applying an improved ABTS
radical cation decolorization assay. Free Radical Biol. Med. 1999,
26, 1231–1237.
(25) Jordan, A. D.; Luo, C.; Reitz, A. B. Efficient conversion of
substituted aryl thioureas to 2-aminobenzothiazoles using benzyl-
trimethylammonium tribromide. J. Org. Chem. 2003, 68, 8693–
8696.
References
(1) Dong, X. X.; Wang, Y.; Qin, Z. H. Molecular mechanisms of
excitotoxicity and their relevance to pathogenesis of neurodegene-
rative diseases. Acta Pharmacol. Sin. 2009, 30, 379–387.
(2) Besancon, E.; Guo, S.; Lok, J.; Tymianski, M.; Lo, E. H. Beyond
NMDA and AMPA glutamate receptors: Emerging mechanisms
for ionic imbalance and cell death in stroke. Trends Pharmacol. Sci.
2008, 29, 268–275.
(3) Brown, G. C. Mechanisms of inflammatory neurodegeneration:
iNOS and NADPH oxidase. Biochem. Soc. Trans. 2007, 35, 1119–
1121.
(26) Farkas, L.; Kasztreiner, E.; Andrasi, F.; Borsi, J.; Kosary, J.;
Csanyi, E.; Elek, S. Thiazolylurea derivatives and process for
the preparation thereof. Patent GB 1437895, 1976-06-03. EC:
C07D417/04; C07D417/14.
(4) Nicholls, D. G. Mitochondrial dysfunction and glutamate excito-
toxicity studied in primary neuronal cultures. Curr. Mol. Med.
2004, 4, 149–177.
(5) Fernandez-Espejo, E. Pathogenesis of Parkinson’s disease: Pro-
spects of neuroprotective and restorative therapies. Mol. Neuro-
biol. 2004, 29, 15–30.
(6) Bonelli, R. M.; Wenning, G. K.; Kapfhammer, H. P. Huntington’s
disease: Present treatments and future therapeutic modalities. Int.
Clin. Psychopharmacol. 2004, 19, 51–62.
(7) Ates, O.; Cayli, S. R.; Gurses, I.; Karabulut, A. B.; Yucel, N.;
Kocak, A.; Cakir, C. O.; Yologlu, S. Do sodium channel blockers
have neuroprotective effect after onset of ischemic insult? Neurol.
Res. 2007, 29, 317–323.
(27) Latli, B.; D’Amour, K.; Casida, J. E. Novel and potent 6-chloro-
3-pyridinyl ligands for the R4β2 neuronal nicotinic acetylcholine
receptor. J. Med. Chem. 1999, 42, 2227–2234.
(28) Doyle, K. P.; Simon, R. P.; Stenzel-Poore, M. P. Mechanisms of
ischemic brain damage. Neuropharmacology 2008, 55, 310–318.
(29) Chen, Q. M.; LaBella, F. S. Inhibition of nitric oxide synthase by
straight chain and cyclic alcohols. Eur. J. Pharmacol. 1997, 321,
355–360.
(30) Salter, M.; Knowels, G. R. Assay of NOS activity by the measure-
ment of conversion of oxyhemoglobin to methemoglobin by NO.
Methods in Molecular Biology; Nitric Oxide Protocols; Titheradge, M.
A., Ed.; Humana Press: Totowa, NJ, 1998; Vol. 100, Chapter 6, pp
61-66.
(31) Satoh, T.; Ishige, K.; Sagara, Y. Protective effects on neuronal cells
of mouse afforded by ebselen against oxidative stress at multiple
steps. Neurosci. Lett. 2004, 371, 1–5.
(8) Bensimon, G.; Lacomblez, L.; Meininger, V. A controlled trial of
riluzole in amyotrophic lateral sclerosis. ALS/Riluzole study
group. N. Engl. J. Med. 1994, 330, 585–591.
(9) Bryson, H. M.; Fulton, B.; Benfield, P. Riluzole: A review of its
pharmacodynamic and pharmacokinetic properties and therapeu-
tic potential in amyotrophic lateral sclerosis. Drug Eval. 1996, 52,
549–563.
(10) Miller, R. G.; Mitchell, J. D.; Lyon, M.; Moore, D. H. Riluzole for
amyotrophic lateral sclerosis (ALS)/motor neuron disease (MND).
Cochrane Database Syst. Rev. 2 2002, CD001447.
(11) Benoit, E.; Escande, D. Riluzole specifically blocks inactivated Na
channels in myelinated nerve fibre. Pfluegers Arch. 1991, 419, 603–
609.
(12) Hebert, T.; Drapeau, P.; Pradier, L.; Dunn, R. J. Block of the rat
brain IIA sodium channel R subunit by the neuroprotective drug
riluzole. Mol. Pharmacol. 1994, 45, 1055–1060.
(13) Song, J. H.; Huang, C. S.; Nagata, K.; Yeh, J. Z.; Narahashi, T.
Differential action of riluzole on tetrodotoxin-sensitive and tetro-
dotoxin-resistant sodium channels. J. Pharmacol. Exp. Ther. 1997,
282, 707–714.
(14) Zona, C.; Siniscalchi, A.; Mercuri, N. B.; Bernardi, G. Riluzole
interacts with voltage-activated sodium and potassium currents in
cultured rat cortical neurons. Neuroscience 1998, 85, 931–938.
(15) Debono, M. W.; Le Guern, J.; Canton, T.; Doble, A.; Pradier, L.
Inhibition by riluzole of electrophysiological responses mediated
by rat kainate and NMDA receptors expressed in Xenopus oocytes.
Eur. J. Pharmacol. 1993, 235, 283–289.
(32) Huang, C. S.; Song, J. H.; Nagata, K.; Yeh, J. Z.; Narahashi, T.
Effects of the neuroprotective agent riluzole on the high voltage-
activated calcium channels of rat dorsal root ganglion neurons.
J. Pharmacol. Exp. Ther. 1997, 282, 1280–1290.
(33) Hamill, O. P.; Marty, A.; Neher, E.; Sakmann, B.; Sigworth, F. J.
Improved patch-clamp techniques for high-resolution current re-
cording from cells and cell-free membrane patches. Pfluegers Arch.
1981, 391, 85–100.
(34) Castelli, L.; Nigro, M. J.; Magistretti, J. Analysis of resurgent
sodium-current expression in rat parahippocampal cortices and
hippocampal formation. Brain Res. 2007, 1163, 44–55.
(35) Magistretti, J.; Castelli, L.; Taglietti, V.; Tanzi, F. Dual effect of
Zn2þ on multiple types of voltage-dependent Ca2þ currents in rat
palaeocortical neurons. Neuroscience 2003, 117, 249–264.
(36) Jimonet, P.; Audiau, F.; Barreau, M.; Blanchard, J.-C.; Boireau,
A.; Bour, Y.; Coleno, M. A.; Doble, A.; Doerflinger, G.; Do Huu,
C.; Donat, M. H.; Duchesne, J. M.; Ganil, P.; Gueremy, C.;
Honore, E.; Just, B.; Kerphirique, R.; Gontier, S.; Hubert, P.;
Laduron, P. M.; Le Blevec, J.; Meunier, M.; Miquet, J. M.;
Nemecek, C.; Pasquet, M.; Piot, O.; Pratt, J.; Rataud, J.; Reibaud,
M.; Stutzmann, J. M.; Mignani, S. Riluzole series. Synthesis and in
vivo “antiglutamate” activity of 6-substituted-2-benzothiazola-
mines and 3-substituted-2-imino-benzothiazolines. J. Med. Chem.
1999, 42, 2828–2843.
(16) Doble, A. The pharmacology and mechanism of action of riluzole.
(37) Brahma, B.; Forman, R. E.; Stewart, E. E.; Nicholson, C.; Rice, M.
E. Ascorbate inhibits edema in brain slices. J. Neurochem. 2000, 74,
1263–1270.
(38) Eilers, H.; Kindler, C. H.; Bickler, P. E. Different effects of volatile
anesthetics and polyhalogenated alkanes on depolarization-
evoked glutamate release in rat cortical brain slices. Anesth. Analg.
(Hagerstown, MD, U.S.) 1999, 88, 1168–1174.
(39) Gay, R. J.; McComb, R. B.; Bowers, G. N., Jr. Optimum reaction
conditions for human lactate dehydrogenase isoenzymes as they
affect total lactate dehydrogenase activity. Clin. Chem. 1968, 14,
740–753.
(40) Chen, Q. M.; LaBella, F. S. Inhibition of nitric oxide synthase by
straight chain and cyclic alcohols. Eur. J. Pharmacol. 1997, 321,
355–360.
Neurology 1996, 47 (Suppl. 4), S233–S241.
(17) Martin, D.; Thompson, M. A.; Nadler, J. V. The neuroprotective
agent riluzole inhibits release of glutamate and aspartate from slices
of hippocampal area CA1. Eur. J. Pharmacol. 1993, 250, 473–476.
(18) Frizzo, M. E.; Dall’Onder, L. P.; Dalcin, K. B.; Souza, D. O.
Riluzole enhances glutamate uptake in rat astrocyte cultures. Cell.
Mol. Neurobiol. 2004, 24, 123–128.
(19) Ginsberg, M. D. Neuroprotection for ischemic stroke: Past, pre-
sent and future. Neuropharmacology 2008, 55, 363–389.
(20) Zhu, Y.; Nikolic, D.; Van Breemen, R. B.; Silverman, R. B. Mecha-
nism of inactivation of inducible nitric oxide synthase by amidines.
Irreversible enzyme inactivation without inactivator modification.
J. Am. Chem. Soc. 2005, 127, 858–868.
(21) Iravani, M. M.; Liu, L.; Rose, S.; Jenner, P. Role of inducible nitric
oxide synthase in N-methyl-D-aspartic acid-induced strio-nigral
degeneration. Brain Res. 2004, 1029, 103–113.
(22) Babbedge, R. C.; Hart, S. L.; Moore, P. K. Anti-nociceptive
activity of nitric oxide synthase inhibitors in the mouse: Dissocia-
tion between the effect of L-NAME and L-NMMA. J. Pharm.
Pharmacol. 1993, 45, 77–79.
(41) Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J.
Protein measurement with the Folin phenol reagent. J. Biol. Chem.
1951, 193, 265.
(42) Magistretti, J.; Castelli, L.; Forti, L.; D’Angelo, E. Kinetic and
functional analysis of transient, persistent, and resurgent sodium
currents in rat cerebellar granule cells in situ: An electrophysio-
logical and modeling study. J. Physiol. 2006, 573, 83–106.