Benzothiazepine CGP37157 and its 2′-isopropyl analogue modulate Ca2+ entry through CALHM1

Article abstract of DOI:10.1016/j.neuropharm.2015.02.016

CALHM1 is a Ca²⁺ channel discovered in 2008, which plays a key role in the neuronal electrical activity, among other functions. However, there are no known efficient blockers able to modulate its Ca²⁺ handling ability. We herein desc

Full text of DOI:10.1016/j.neuropharm.2015.02.016

Neuropharmacology 95 (2015) 503e510
Contents lists available at ScienceDirect
Neuropharmacology
journal homepage: www.elsevier.com/locate/neuropharm
Benzothiazepine CGP37157 and its 20-isopropyl analogue modulate
Ca2þ entry through CALHM1
,
Ana J. Moreno-Ortega a b, Francisco J. Martínez-Sanz a, Rocío Lajarín-Cuesta a,
b
,
, *
Cristobal de los Rios **, María F. Cano-Abad a
ꢀ
a
ꢀ
ꢀ
ꢀ
Instituto Teofilo Hernando, Departamento de Farmacología y Terapeutica, Universidad Autonoma de Madrid, C/ Arzobispo Morcillo, 4, 28029, Madrid,
Spain
b
ꢀ
ꢀ
Servicio de Farmacología Clínica, Instituto de Investigacion Sanitaria, Hospital Universitario de la Princesa, C/ Diego de Leon, 62, 28006, Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
CALHM1 is a Ca2þ channel discovered in 2008, which plays a key role in the neuronal electrical activity,
among other functions. However, there are no known efficient blockers able to modulate its Ca2þ
handling ability. We herein describe that benzothiazepine CGP37157 and its newly synthesized analogue
ITH12575 reduced Ca2þ influx through CALHM1 at low micromolar concentrations. These results could
serve as a starting point for the development of more selective CALHM1 ligands using CGP37157 as a hit
Article history:
Received 3 October 2014
Received in revised form
16 January 2015
Accepted 11 February 2015
Available online 20 April 2015
compound, which would help to study the physiological role of CALHM1 in the control of [Ca2þ
excitable cells, as well as its implication in CNS diseases.
]
in
cyt
Keywords:
© 2015 Elsevier Ltd. All rights reserved.
Benzothiazepines
Calcium homeostasis modulator 1
(CALHM1)
CGP37157
ITH12575
Neurodegenerative diseases
Chemical compounds studied in this article:
CGP37157
PubChem CID: 2688
Dimebolin
PubChem CID: 197033
Carbonyl cyanide-p-
trifluoromethoxyphenylhydrazone
FCCP
PubChem CID: 3330
1. Introduction
the organism (Ma et al., 2012). CALHM1 has been proposed to be
the pore-forming subunit of a cation channel, which would
modulate the extracellular Ca2þ-controlled neuronal excitability
and the Ca2þ homeostasis, in the so-called Ca2þ add-back condi-
tions (Ma et al., 2012). CALHM1 is located in plasma membrane and
endoplasmic reticulum (ER), where its alteration could lead to Ca2þ
homeostasis disruption, ER stress, and cell damage (Gallego-Sandin
et al., 2011). In summary, it seems that CALHM1 plays an essential
role in the neuronal electrical activity by controlling the intracel-
lular Ca2þ-derived signal transduction (Dreses-Werringloer et al.,
2013). Heterologous expression of human CALHM1 in Caeno-
rhabditis elegans has revealed its physiological role in bio-
locomotion, though its overexpression causes neurodegeneration
through a Ca2þ-dependent necrotic-like mechanism (Tanis et al.,
Since the discovery of the channel CALHM1 (Ca2þ Homeostasis
Modulator 1) in 2008 (Dreses-Werringloer et al., 2008), several
papers have discussed whether this channel is implicated in the
progression of certain CNS (central nervous system) diseases, not
only Alzheimer's disease (AD) (Koppel et al., 2011), but also tem-
poral lobe epilepsy (Lv et al., 2011) and Creutzfeldt-Jakob's disease
(Calero et al., 2012). Six human CALHM1 homologous have been
characterized, with different splicing and expression throughout
* Corresponding author.
** Corresponding author.
E-mail address: maria.cano@uam.es (M.F. Cano-Abad).
http://dx.doi.org/10.1016/j.neuropharm.2015.02.016
0028-3908/© 2015 Elsevier Ltd. All rights reserved.

504
A.J. Moreno-Ortega et al. / Neuropharmacology 95 (2015) 503e510
2.2. HeLa cells source, culture, and transfection
2013). Furthermore, CALHM1 might have peripheral physiological
functions, as it is expressed in taste buds and it has been hypoth-
esized that it controls the taste perception (Taruno et al., 2013;
Tordoff et al., 2014). From a physiopathological point of view,
HeLa cells were grown in plastic flasks in DMEM supplemented with 10% fetal
bovine serum, 2 mM glutamine, 25 U/mL penicillin and 25 mg/mL streptomycin (all
products were obtained from Lonza (Basel, Switzerland)). The experiments were
performed with cells seeded on 13-mm diameter coverlips, in 24-well plates and
grown to 60e70% confluence after 24 h in the incubator at 37 ꢀC and 5% CO2.
Transient co-transfections with the genetically encoded photoprotein aequorin,
targeted to the cytosol (cyt_AEQ) and CALHM1 or P86L-CALHM1, in a ratio 1:1, were
achieved by using Metafectene (Biontex, Martinsried/Planegg, Germany) (Diaz-
CALHM1 controls cytosolic Ca2þ concentration ([Ca2þ
]
cyt) and am-
) levels in depolarizing conditions (Dreses-
Werringloer et al., 2008). When the polymorphism P86L-CALHM1
yloid beta peptide (A
b
(rs2986017) is present, a deregulation of such [Ca2þ
]
cyt
and A
b
Prieto et al., 2008). Experiments to measure [Ca2þ
36e48 h after transfection. The two recombinant proteins were expressed in the
same subset of cells.
]
cyt
changes were performed
levels are found (Dreses-Werringloer et al., 2008), generating A
b
accumulation as well as mitochondrial Ca2þ ([Ca2þ
]
mit) overload
(Moreno-Ortega et al., 2010). Despite these findings, there is no
agreement in the role played by CALHM1 in AD progression, as
some papers describe a lack of correlation between the poly-
morphism P86L-CALHM1 and the risk of suffering AD (Giedraitis
et al., 2010; Inoue et al., 2010; Tao et al., 2014). By contrast,
Lambert et al., in an international meta-analysis, established that
P86L-CALHM1 is not a genetic factor, but it might modulate the age
of AD onset (Lambert et al., 2010). Studies focused on clarifying the
role of CALHM1 have to recruit molecular biology techniques
(Dreses-Werringloer et al., 2013) due to the lack of drugs capable of
regulating its activity. Some metals and metal-organic complexes at
2.3. Measurements of [Ca2þ
]
with aequorin
cyt
HeLa cells expressing cyt_AEQ were reconstituted by incubating cells 1.5 h with
mM wild type coelenterazine before the experiment. The cell monolayer was
5
continuously superfused with KrebseHEPES buffer (KHB) of the following compo-
sition (mM): 125 NaCl, 5 KCl, 1 Na3PO4, 1 MgSO4, 5.5 glucose, 20 HEPES, with pH
7.4 at room temperature (24 2 ꢀC); the zero Ca2þ solution contained 0.5 mM EGTA
(ethylene glycol tetraacetic acid). To induce Ca2þ entry, KHB deprived of Ca2þ with
EGTA was changed by another solution containing 1 mM CaCl2, as specified in figure
legends. Light emission was measured in a purposed-built luminometer and cali-
brated in terms of [Ca2þ]. At the end of the experiment, cells were lysed by super-
fusing them with KHB containing 10 mM CaCl2 and 100 mM digitonin, in order to
expose cells to excess Ca2þ to burn out the aequorin remaining at the end of each
experiment.
high concentrations, e.g. Co2þ (100
(10 M), 2-aminoethoxydiphenyl borate (2-APB;
M), Gd3þ (100
1 mM) or ruthenium red (20
M), block Ca2þ entry through
m m
M), Zn2þ (20 M), Ni2þ
m
m
2.4. Data analysis
m
Data are presented as means SEM. Peak increases of the [Ca2þ
]
cyt
elevations
CALHM1 (Dreses-Werringloer et al., 2013; Ma et al., 2012).
In this paper, we present the first organic compound capable to
block the Ca2þ uptake through CALHM1 at low micromolar con-
centrations, the benzothiazepine CGP37157 (7-chloro-5-(2-
chlorophenyl)-3,5-dihydro-4,1-benzothiazepin-2-(1H)-one, Fig. 1),
and areas under the curve (AUC) were determined by using Origin v. 5.0 (OriginLab
Corporation, Northampton, USA). Statistical differences were assessed by ANOVA
test followed by Bonferroni's or NewmaneKeuls' post hoc analysis. Statistical dif-
ferences were taken as significant when p ꢁ 0.05. All statistical analyses were per-
formed using a Prism software (GraphPad) version 5.0 for Mac (OS X).
which has been widely used as
a
mitochondrial Naþ/Ca2þ
2.5. Chemical synthesis of ITH12575
exchanger (mNCX) blocker (Chiesi et al., 1988; Pei et al., 2003). This
compound and some of its analogues have shown a neuro-
protective profile in several in vitro models of AD and stroke
(Gonzalez-Lafuente et al., 2012; Nicolau et al., 2009, 2010). In
addition, we herein prove the CALHM1 blocking activity of one of
its derivatives, synthesized in our research group, ITH12575 (7-
chloro-5-(2-isopropylphenyl)-3,5-dihydro-4,1-benzothiazepin-2-
(1H)-one, Fig. 1), which has shown the most interesting neuro-
protective properties (unpublished results). Our results will help to
clarify the actual contribution of CALHM1 to physiological/patho-
logical events implicating cell Ca2þ dynamics.
Reactions were monitored by thin layer chromatography using precoated silica
gel plates. Detection was made with light UV at 254 nm. Pre-charged silica gel
columns were used in a Biotage chromatography station (230e400 mesh). Melting
points were determined in a Stuart apparatus (SMP-10) and are not corrected. MS
spectra were obtained in a QSTAR de ABSciex apparatus. 1H and 13C NMR spectra
were carried out with a Bruker AVANCE 300 MHz. Elemental analyses were used as a
purity criteria and performed on a LECO CHNS-932 station. Compounds described
had a purity of 95% or more. For the determination of absolute configuration, single
crystal X-ray diffractions have been performed, using a Bruker Kappa Apex II
apparatus with a Mo source and graphite monochromater.
2.5.1. Synthesis of 4-chloro-N-tert-butoxycarbonylaniline (1)
To a solution of 4-chloroaniline (1 g, 7.84 mmol) in dry tetrahydrofurane (THF)
(65 mL), tert-butyl dicarbonate (1.88 g, 8.62 mmol) and triethylamine (872 mg,
1.2 mL, 8.62 mmol) were added. The reaction mixture was stirred at room tem-
perature for 48 h, after then solvent was evaporated and ethyl acetate (300 mL) was
added. The solution was washed with saturated NH4Cl (2 ꢂ 100 mL) and water
(2 ꢂ 100 mL). The organic layer was dried over MgSO4, filtered and evaporated,
obtaining a solid white showing spectral data according to 1 (yield 99%) (Roosen
et al., 2012). It was used with no further purification.
2. Material and methods
2.1. Reagents
Coelenterazine was purchased from Biotium (Hayward, USA). Dimebolin was
purchased from Biotrend Chemikalien GmbH (Cologne, Germany). Metafectene was
purchased from Biontex (Martinsried/Planegg, Germany). Digitonin and CoCl2, and
other general chemicals were purchased from Sigma (Madrid, Spain). CGP37157 was
purchased from Tocris (Madrid, Spain).
2.5.2. Synthesis of o-isopropylbenzaldehyde (2)
To a solution of 1-bromo-2-isopropylbenzene (388 mg, 298
m
L,1.95 mmol) in dry
THF, n-butyllithium (2.5 M in hexanes, 858
mL, 2.14 mmol) was added dropwise
at ꢃ78 ꢀC. After 20 min at ꢃ78 ꢀC, dimethylformamide (157 mg, 166
mL, 2.14 mmol)
was added and the reaction mixture was stirred at ꢃ78 ꢀC for 20 min and allowed to
heat up to ꢃ10 ꢀC and stirred for 3 h more. Reaction was stopped by adding water
(2.5 mL) and extracted with diethyl ether (3 ꢂ 100 mL). Combined organic layers
were dried over anhydrous MgSO4, filtered, and evaporated, affording 2 with
quantitative yield as a colorless oil (277 mg) showing spectral data according to its
structure (Ewing, 1974). It was used with no further purification.
2.5.3. Synthesis of 4-chloro-2-[hydroxy(2-isopropylphenyl)methyl]-N-tert-
butoxycarbonylaniline (3)
To a solution of 1 (496 mg, 2.18 mmol) in dry THF (15 mL), tert-butyllithium
(1.7 M, 3.49 mL, 5.93 mmol) was added dropwise at ꢃ78 ꢀC and the reaction was
stirred at this temperature for 15 min, then it was stirred for 2 h at ꢃ20 ꢀC. After this
time, reaction was cooled down to ꢃ78 ꢀC and 2 (358 mg, 2.55 mmol) in dry THF
(5 mL) was added dropwise and the reaction was allowed to stir 2 h at this tem-
perature. Reaction was terminated by addition of water (15 mL) and once it reached
room temperature, it was extracted with diethyl ether (3 ꢂ 30 mL) and the
Fig. 1. Chemical structures of the benzothiazepines CGP37157 and ITH12575.

A.J. Moreno-Ortega et al. / Neuropharmacology 95 (2015) 503e510
505
cells do not express plasmalemmal Naþ/Ca2þ exchanger or voltage-
gated Ca2þ channels (VGCC). Thus, they are an excellent model for
the study of the expression and function of Ca2þ channels such as
CALHM1. Either wild type CALHM1 or its polymorphism P86L-
CALHM1 were transiently co-transfected with aequorins, in a ra-
tio 1:1. Luminescence experiments were performed as previously
described (Moreno-Ortega et al., 2010). Cells were first perfused
with a 0 Ca2þ/EGTA solution for 2 min, after which it was replaced
by another solution containing 1 mM Ca2þ (no EGTA). HeLa cells
transfected with CALHM1 showed an apparent increase in the Ca2þ
entry when reintroducing the extracellular Ca2þ, compared with
that found in control cells. Similarly to previous observations by
other authors (Dreses-Werringloer et al., 2008), the increases in
both maximal peak and the area under the curve (AUC) were
combined organic layer washed with brine (3 ꢂ 50 mL), dried over anhydrous
MgSO4, filtered and evaporated, to obtain a yellow oil that was purified by flash
chromatography (ethyl acetate:hexanes, 1:18), affording compound 3 with a 67%
yield. 1H NMR (300 MHz, CDCl3)
d
7.85 (d, 1H, J ¼ 8.6 Hz, H6), 7.42 (s, 1H, NH),
7.38e7.10 (m, 6H, HAr), 6.90 (d, 1H, J ¼ 2.0 Hz, H3), 6.10 (s, 1H, CH), 3.00 [heptet,
J ¼ 7.0 Hz, 1H, CH(CH3)3], 2.45 (bs, 1H, OH), 1.41 (s, 9H, CH3), 1.23 (d, J ¼ 7.0 Hz, 3H,
CH3), 1.18 (d, 3H, J ¼ 7.0 Hz, CH3).
2.5.4. Synthesis of 7-chloro-5-(20-isopropylphenyl)-3,5-dihydro-4,1-benzothiazepin-
2-(1H)-one (ITH12575)
To a solution of methyl thioglycollate (172 mg, 148
mL, 1.62 mmol) in trifluoro-
acetic acid (431 mg, 289 L, 3.78 mmol) compound 3 (100 mg, 0.27 mmol) was
m
added. The reaction mixture was stirred at 85 ꢀC for 24 h and it was dissolved in
CH2Cl2 (30 mL). The crude was washed with brine (30 mL), 1N NaOHaq (30 mL), and
brine (30 mL), dried over Na2SO4, filtered, and evaporated, to obtain a brown oil that
was purified by flash chromatography (ethyl acetate:hexanes, 1:3), affording com-
pound 4 as a white solid, with a 62% yield. Melting point: 188e190 ꢀC. 1H NMR
blocked by Co2þ 100
m
M, used as a standard, non-specific Ca2þ
(300 MHz, CDCl3)
d
8.16 (s, 1H, NH), 7.75 (dd, 1H, J ¼ 8.5, 2.0 Hz, H8), 7.43e7.31 (m,
3H, HAr), 7.29e7.21 (dd, 1H, H50), 7.07 (d, J ¼ 8.5, 1H, H9), 6.81 (d, J ¼ 2.0 Hz, 1H, H6),
6.10 (s, 1H, H5), 3.40 and 3.01 (AB, J ¼ 12.3, CH2), 2.82 [heptet, J ¼ 6.8 Hz, 1H,
CH(CH3)3], 1.22 (d, J ¼ 6.8 Hz, 3H, CH3), 0.87 (d, J ¼ 6.8 Hz, 3H, CH3). Anal.
(C18H18ClNOS) C, H, N, and S similar to the experimentally found data.
channels blocker, in 87 and 89%, respectively, (Fig. 2). Regarding to
the P86L-CALHM1-transfected cells, they underwent a lower in-
crease in both [Ca2þ
]
peak and AUC, together with a slower rise
cyt
rate, as reported by our group (Moreno-Ortega et al., 2010) In these
cells, Co2þ 100 M completely abolished peak of [Ca2þ
cyt and AUC
increases (Fig. 2).
2.6. Chiral separation procedure of ITH12575
m
]
High Performance Liquid Chromatography with Ultraviolet and Mass detection
(HPLC/MS) was used for the development of conditions for the resolution of the
racemic mixture of ITH12575, as well as for the determination of the enantiomeric
excess (ee) of the two enantiomers isolated by semipreparative chiral HPLC. Chro-
matographies were performed on CHIRALPAK AD [amylose tris (3,5-dimethyl-
phenyl carbamate)] units. The column dimensions were 100
250 ꢂ 20 mm for analytical and semi-preparative work, respectively. In both units,
the enantioselective phase was coated onto silica-gel substrate (10 m). Experi-
ments were carried out at room temperature. Flow rate was set up at 0.2 and 65 mL/
min for analytical and semipreparative work, respectively. Mobile phase consisted of
0.2% IPAm/methanol, for analysis, and CO2 (A)/methanol-DMEA (0.2% v/v) (B) Iso-
cratic 10% B, for purification. In all cases, the wavelength of UV detection was
monitored from 200 to 400 nm although chromatograms were recorded at 220 and
254 nm signals. Mass spectra were recorded using API-APCI ionization (full scan in
positive/negative modes simultaneously). Retention time at analytical scale for the
two target enantiomers are 1.08 and 1.30 min. Chiral quality control of isomers
isolated (AT1 and AT2) by semi-preparative HPLC confirmed ee >98% for the two
3.2. Antihistaminic, and antiamyloidogenic drug dimebolin does
not affect Ca2þ permeability through CALHM1 or P86L-CALHM1 in
HeLa cells
ꢂ
4.6 mm and
m
In the search for new chemical entities capable of modulating
CALHM1-dependent [Ca2þ
]
homeostasis, we aimed to study
cyt
neuroprotective drugs of our own chemical library that had been
described to block ionic channels, preferentially VGCC, following
the hypothesis previously by other authors (Dreses-Werringloer
et al., 2008; Ma et al., 2012), who had focused on a huge variety
of ionic channels blockers in order to modulate the CALHM1 signal.
Among several compounds studied, we present dimebolin as an
example, which is a non-selective antihistamine drug with multiple
mechanisms of action including L-type VGCC blockade, protection
target compounds. AT1 (Ret. time ¼ 1.08 min) showed an [
a
]
D
of ꢃ411.6 (c 0.61,
CH2Cl2). AT2 (Ret. time ¼ 1.30 min) showed an [ D of þ425.9 (c 0.69, CH2Cl2).
a
]
against the Ab-induced neurotoxicity and the reduction of the
mitochondrial swelling induced by toxic insults (Bachurin et al.,
2003; Cano-Cuenca et al., 2014; Lermontova et al., 2001). In all
2.7. Single crystal X-Ray diffraction data for (ꢃ)-ITH12575 (AT1) and (þ)-ITH12575
(AT2)
Diffraction data were collected on
a Bruker Kappa Apex II diffractometer
the experimental conditions performed, dimebolin 20 mM had no
equipped with Mo source and a graphite monochromator. The software package
SHELXTL was used for space group determination, structure solution, absorption
correction and refinement1. The structures were solved by direct methods,
completed with difference Fourier syntheses, and refined with anisotropic
displacement parameters. In both structures, carbon atoms C17 and C18, from the
terminal methyl groups, were found at two alternate positions. A statistical disorder
model was made to account for this situation, and the occupation factor refined for a
final value of 56% for part A and 44% for part B in AT1; and 55% for part A and 45% for
part B in AT2.
effect on Ca2þ signaling, neither on peak nor AUC increases (Fig. 3).
3.3. Neuroprotectant and mitochondrial Naþ/Ca2þ exchanger
blocker CGP37157 reduces Ca2þ permeability through CALHM1 but
not through P86L-CALHM1 in HeLa cells at low micromolar
concentrations
CCDC 1021365 and 1021366 contain the supplementary crystallographic data for
compounds AT1 and AT2. These data can be obtained free of charge at www.ccdc.
cam.ac.uk/conts/retrieving.html [or from the Cambridge Crystallographic Data
Centre, 12, Union Road, Cambridge CB2 1EZ, UK; fax: þ44 1223/336-033; e-mail:
deposit@ccdc.cam.ac.uk].
By contrast, when we tested the mNCX blocker CGP37157, a
reduction of the CALHM1-controlled Ca2þ permeability was
observed (Fig. 4). When applied to HeLa cells transfected with
CALHM1 at 0.1
m
M, CGP37157 blocked by 57% and 44% the maximal
peak of [Ca2þ
]
and the AUC, respectively, induced by the Ca2þ
cyt
reintroduction. Similar blockades in [Ca2þ
]
peak and AUC were
cyt
3. Results and discussion
found at 0.3
respectively) (Fig. 4D and E). CGP37157 at 3
[Ca2þ
cyt signal in a significant manner, but when applied at 10 and
30 by 26% and
M, CGP37157 reduced maximal peak of [Ca2þ
mM (50% and 40%, respectively) and 1
mM (36 and 32%,
3.1. Co2þ blocks Ca2þ entry through both CALHM1 and P86L-
CALHM1 but not in blank-transfected HeLa cells
mM did not affect the
]
m
]
cyt
39%, respectively, but lacking statistical significance. The weird
behavior of CGP37157 could be due to the lack of target selectivity
when doses reach higher micromolar levels. HeLa cells transfected
with the P86L-CALHM1 presented a different sensitivity to the
administration of CGP37157, as no significant reduction of Ca2þ
To study the pharmacological modulation of CALHM1, we used
HeLa cells transfected with the genetically encoded photoprotein
aequorin targeted to the cytosol (cyt_AEQ), (Moreno-Ortega et al.,
2010) achieved with Metafectene (Diaz-Prieto et al., 2008). HeLa
signals was observed at any concentration, but even 10 and 30
CGP37157 evoked an augmentation in either AUC and maximal
peak of [Ca2þ
cyt, respectively (Fig. 4D and E).
mM
1
SHELXTL version 6.12, Structure Determination Package, Bruker, Madison,
Wisconsin, USA, 2001.
]

506
A.J. Moreno-Ortega et al. / Neuropharmacology 95 (2015) 503e510
Fig. 2. Effect of Co2þ on the CALHM1-controlled Ca2þ permeability. Cells were transiently transfected with blank (Control), CALHM1, or P86L-CALHM1. Cells were perfused with
Co2þ at 100 M for 2 min before and during the pulse of 1 mM Ca2þ. Panel A shows the mean of peak increases of the [Ca2þ
m ]cyt elevations and Panel B shows the area under the curve
(AUC) of cytosolic Ca2þ in HeLa cells. Mean of at least 9 experiments for each transfection protocol, made with cells from 3 different cultures. Data are means SEM. *p < 0.05,
**p < 0.01, ***p < 0.001.
3.4. Reduction of Ca2þ permeability through CALHM1 elicited by
CGP37157 is independent to its effect on the mitochondrial Ca2þ
buffering capacity
FCCP and CGP37157 1
m
M, a similar reduction in [Ca2þ
]
]
cyt signal was
maintained, (expressed as an augmentation of [Ca2þ
peak and
cyt
AUC (Fig. 5D and E, respectively), comparing HeLa cells exposed
only to FCCP with cells exposed to FCCP plus CGP37157, expressed
These results confirm that CGP37157 is the first organic com-
pound modulating Ca2þ entry through CALHM1 at very low con-
centrations. Nevertheless, we wondered whether this regulation is
actually due to an indirect effect on the mNCX. To answer this
question, we carried out similar experiments in the presence of the
ionophore FCCP (carbonyl cyanide-p-trifluoromethoxyphenylhydr
azone) (Mitchell and Moyle, 1967), thus abolishing the regulatory
as an augmentation of [Ca2þ
]
peak and AUC (Fig. 5D and E,
cyt
respectively). Again, P86L-CALHM1 polymorphism seems insensi-
tive to CGP37157 in terms of [Ca2þ
]
oscillations. These data
cyt
suggest that pharmacological actions of CGP37157 were not due to
its blockade of mNCX. However, further experiments need to be
addressed in order to point out a direct interaction of CGP37157 on
CALHM1.
action on the [Ca2þ
]
cyt
performed by mitochondria (Fig. 5).
3.5. CGP37157 derivative ITH12575 slightly reduces Ca2þ
permeability through CALHM1 in an enantiospecific fashion
Fig. 5A, B, and C show an additional transient elevation of
[Ca2þ during the 0 Ca2þ interval when FCCP 1
]
mM is adminis-
cyt
tered. This [Ca2þ
]cyt oscillation is due to the emptying of the stored
Ca2þ in organelles by the protonophore FCCP. In any case, when
Ca2þ permeability of CALHM1 is achieved by the reintroduction of
extracellular Ca2þ 1 mM, plus preincubation with the cocktail of
Additional studies were carried out with the CGP37157 deriv-
ative ITH12575 (Fig. 1), synthesized in our laboratory (see Mate-
rials and methods), which has shown an improved
Fig. 3. Effect of dimebolin on the CALHM1-controlled Ca2þ permeability. Panels A, B, and C show typical [Ca2þ
]
cyt
registers when cells were transiently transfected with blank,
CALHM1, or P86L-CALHM1, respectively. Cells were perfused with Dimebolin at 20
[Ca2þ cyt elevations and Panel E shows the area under the curve (AUC) of the Ca2þ transients in HeLa cells transfected with blank (Control), CALHM1 or P86L-CALHM1. Mean of at
least 9 experiments for each transfection protocol, made with cells from 3 different cultures. Data are means SEM. No significant differences were found.
m
M for 2 min before and during the 1 mM Ca2þ pulse. Panel D shows the mean of peak of the
]

A.J. Moreno-Ortega et al. / Neuropharmacology 95 (2015) 503e510
507
Fig. 4. Intracellular Ca2þ responses induced by CALHM1 and P86L-CALHM1. Concentrationeresponse curves of CGP37157. Panels A, B, and C show typical registers of [Ca2þ
]
cyt when
cells were transiently transfected with blank, CALHM1, or P86L-CALHM1, respectively. CGP37157, at 0.1e30
m
M, was perfused for 2 min before and during the 1 mM Ca2þ pulse.
Panel D shows the mean of peak increases of the [Ca2þ signal and panel E shows the area under the curve (AUC) of the Ca2þ transients in HeLa cells transfected with blank
]
cyt
(Control), CALHM1 or P86L-CALHM1. Mean of at least 6 experiments for each transfection protocol, made with cells from 3 or more different cultures. Data are means SEM.
*p < 0.05, **p < 0.01, ***p < 0.001.
neuroprotective profile (data not shown). ITH12575, at 1
m
M, did
CH2Cl2)) corresponded to the (þ)-S-enantiomer of ITH12575 (see
Reports and 3D, CIF structures in supplementary material). Hence,
when testing each enantiomer separately, we observed that
not block CALHM1 or P86L-CALHM1 permeability. However,
considering that it is a racemic compound, a possibility exists that
one enantiomer could preferentially interact with CALHM1 or
P86L-CALHM1. Thus, after HPLC-conducted chiral separation, we
resolved the absolute configuration of both enantiomers by single
crystal X-ray diffraction (Supplementary data), reporting that
enantiomer with lesser retention time in HPLC (AT1, 1.08 min) and
(þ)-S-ITH12575 blocked both increase of [Ca2þ
]
peak and AUC
cyt
of Ca2þ entry through the wild type CALHM1 by 47 and 46%,
respectively. Additional experiments were carried out to evaluate
the concentration-dependency of the S-enantiomer of ITH12575.
Unfortunately, the other concentrations assayed (0.1 and 10 mM)
levo rotatory optical deviation ([
a
]
of ꢃ411.6 (c 0.61, CH2Cl2))
were ineffective to block CALHM1 (see Supplementary material,
Fig. S3). Once more, P86L-CALHM1 was insensitive to the expo-
sure of ITH12575, neither racemic nor optically pure compounds
(Fig. 6).
D
corresponded to the (ꢃ)-R-enantiomer of ITH12575, whilst the
enantiomer with greater retention time in HPLC (AT2, 1.30 min)
and dextro rotatory optical deviation ([
a
]
of þ425.9 (c 0.69,
D

508
A.J. Moreno-Ortega et al. / Neuropharmacology 95 (2015) 503e510
Fig. 5. CALHM1 blockade with CGP37157, upon ablation of mitochondria Ca2þ buffering with FCCP. Panels A, B, and C show typical [Ca2þ
]
registers when cells were transiently
cyt
transfected with blank, CALHM1, or P86L-CALHM1, respectively, in presence of CGP37157 1 mM and/or FCCP 1 m ]cyt elevations and
M. Panel D shows the mean of peak of the [Ca2þ
panel E shows the AUC of the Ca2þ transients in HeLa cells transfected with blank (Control), CALHM1 or P86L-CALHM1. Mean of at least 9 experiments for each transfection protocol,
made with cells from 4 or more different cultures. Data are means SEM. *p < 0.05, ***p < 0.001.
4. Conclusions
(SERCA), and the subsecuent Ca2þ cell dynamics alteration when
administered at higher micromolar concentrations (Ragone et al.,
2013). Moreover, though we have discarded that CCGP37157 ex-
erts its modulatory effect on CALHM1 indirectly through mito-
chondria, further experiments need to be addressed to confirm a
direct action on CALHM1. Nevertheless, observing the pharmaco-
logical tools available up to date to study CALHM1, this work would
represent a significant breakthrough, as CALHM1 functionality can
In summary, we have found an organic compound acting as a
CALHM1 channel regulator, the benzothiazepine CGP37157, able to
reduce Ca2þ entry through CALHM1 at very low concentrations.
Previously described blockers, such as Co2þ, have to be adminis-
tered at much higher concentrations. CGP37157 reduced Ca2þ
influx with no concentration-dependent manner. A possibility
exists to ascribe this behavior to additional interactions of
CGP37157 with other biological targets expressed in HeLa cells that
are also contributing to the cell Ca2þ handling and thus modifying
the Ca2þ uptake via CALHM1. The pharmacological actions of
CGP37157 over CALHM1 should be more deeply studied, to explain
its lack of concentration-dependent activity. These effects are
presumably due to the interaction of CGP37157 with other bio-
logical targets, e.g. mNCX or sarco-endoplasmic Ca2þ ATPase
be more finely dissected by applying CGP37157 at 0.3
mM, for
instance. This work should serve as a starting point for the
development of better CALHM1 ligands, using CGP37157 as a hit
compound. The optimization of the pharmacological activity of
CGP37157 over CALHM1 will help to get a better comprehension of
the role of CALHM1 in the control of [Ca2þ
]
in excitable cells, as
cyt
well as the implication of its mutated form in the development of
CNS like AD.

A.J. Moreno-Ortega et al. / Neuropharmacology 95 (2015) 503e510
509
Fig. 6. Effect of ITH12575 and its enantiomers on the CALHM1-induced Ca2þ entry. Cells were transiently transfected with blank (A), CALHM1 (B), or P86L-CALHM1 (C), and a Ca2þ
add-back protocols were carried out in presence of the rac-, S- or R-ITH12575 1 M. Compounds were perfused at 1
M for 2 min before and during the 1 mM Ca2þ pulse. Panel D
shows the mean of peak of the [Ca2þ cyt elevations and panel E shows the AUC of the Ca2þ transients in HeLa cells transfected with blank (Control), CALHM1, or P86L-CALHM1. Mean
of at least 8 experiments for each transfection protocol, made with cells from 4 or more different cultures. Data are means SEM. *p < 0.05.
m
m
]
Acknowledgment
Appendix A. Supplementary data
We thank the generous gifts of cDNAs for CALHM1 and P86L-
CALHM1 to Dr. Philippe Marambaud, Litwin-Zucker Research
Center for the study of Alzheimer's disease (N.Y.), and cDNA for
cyt_AEQ to Prof. Tullio Pozzan, University of Padova (Italy). This
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.neuropharm.2015.02.016.
ꢀ
work supported by the following grants: to M.F.C.A.: Consolidacion
References
de Grupos UAM-CAM (ref. 1004040047); to C.d.l.R: Proyectos de
ꢀ
Investigacion en Salud (PI13/00789, IS Carlos III). A.J.M.O. is granted
Bachurin, S.O., Shevtsova, E.P., Kireeva, E.G., Oxenkrug, G.F., Sablin, S.O., 2003.
Mitochondria as a target for neurotoxins and neuroprotective agents. Ann. N. Y.
Acad. Sci. 993, 334e344. Discussion 345e339.
Calero, O., Bullido, M.J., Clarimon, J., Hortiguela, R., Frank-Garcia, A., Martinez-
Martin, P., Lleo, A., Rey, M.J., Sastre, I., Rabano, A., de Pedro-Cuesta, J., Ferrer, I.,
by Ministry of Economy (FPU program, ref. AP2009/0343). R.L.-C.
thanks Universidad Autonoma de Madrid for predoctoral fellow-
ship (FPI-UAM2012).

510
A.J. Moreno-Ortega et al. / Neuropharmacology 95 (2015) 503e510
Calero, M., 2012. Genetic variability of the gene cluster CALHM 1-3 in sporadic
Creutzfeldt-Jakob disease. Prion 6, 407e412.
Cano-Cuenca, N., Solis-Garcia del Pozo, J.E., Jordan, J., 2014. Evidence for the efficacy
of latrepirdine (Dimebon) treatment for improvement of cognitive function: a
meta-analysis. J. Alzheimers Dis. 38, 155e164.
Chiesi, M., Schwaller, R., Eichenberger, K., 1988. Structural dependency of the
inhibitory action of benzodiazepines and related compounds on the mito-
chondrial Naþ-Ca2þ exchanger. Biochem. Pharmacol. 37, 4399e4403.
Diaz-Prieto, N., Herrera-Peco, I., de Diego, A.M., Ruiz-Nuno, A., Gallego-Sandin, S.,
Lopez, M.G., Garcia, A.G., Cano-Abad, M.F., 2008. Bcl2 mitigates Ca2þ entry and
mitochondrial Ca2þ overload through downregulation of L-type Ca2þ channels
in PC12 cells. Cell Calcium 44, 339e352.
Marambaud, P., Seripa, D., Galimberti, D., Tanzi, R.E., Bertram, L., Lendon, C.,
Lannfelt, L., Licastro, F., Campion, D., Pericak-Vance, M.A., Soininen, H., Van
Broeckhoven, C., Alperovitch, A., Ruiz, A., Kamboh, M.I., Amouyel, P., 2010. The
CALHM1 P86L polymorphism is a genetic modifier of age at onset in Alz-
heimer's disease: a meta-analysis study. J. Alzheimers Dis. 22, 247e255.
Lermontova, N.N., Redkozubov, A.E., Shevtsova, E.F., Serkova, T.P., Kireeva, E.G.,
Bachurin, S.O., 2001. Dimebon and tacrine inhibit neurotoxic action of beta-
amyloid in culture and block L-type Ca(2þ) channels. Bull. Exp. Biol. Med.
132, 1079e1083.
Lv, R.J., He, J.S., Fu, Y.H., Shao, X.Q., Wu, L.W., Lu, Q., Jin, L.R., Liu, H., 2011.
A polymorphism in CALHM1 is associated with temporal lobe epilepsy. Epilepsy
Behav. 20, 681e685.
Dreses-Werringloer, U., Lambert, J.C., Vingtdeux, V., Zhao, H., Vais, H., Siebert, A.,
Jain, A., Koppel, J., Rovelet-Lecrux, A., Hannequin, D., Pasquier, F., Galimberti, D.,
Scarpini, E., Mann, D., Lendon, C., Campion, D., Amouyel, P., Davies, P.,
Foskett, J.K., Campagne, F., Marambaud, P., 2008. A polymorphism in CALHM1
influences Ca2þ homeostasis, Abeta levels, and Alzheimer's disease risk. Cell
133, 1149e1161.
Dreses-Werringloer, U., Vingtdeux, V., Zhao, H., Chandakkar, P., Davies, P.,
Marambaud, P., 2013. CALHM1 controls Ca2þ-dependent MEK/ERK/RSK/MSK
signaling in neurons. J. Cell Sci. 126, 1199e1206.
Ewing, D.F., 1974. Carbon-13-hydrogen coupling constants. III. Conformation
dependence of 1J(13CH) fo r the formyl proton in aromatic aldehydes. Org.
Magn. Reson. 6, 298e300.
Gallego-Sandin, S., Alonso, M.T., Garcia-Sancho, J., 2011. Calcium homoeostasis
modulator 1 (CALHM1) reduces the calcium content of the endoplasmic retic-
ulum (ER) and triggers ER stress. Biochem. J. 437, 469e475.
Ma, Z., Siebert, A.P., Cheung, K.H., Lee, R.J., Johnson, B., Cohen, A.S., Vingtdeux, V.,
Marambaud, P., Foskett, J.K., 2012. Calcium homeostasis modulator 1 (CALHM1)
is the pore-forming subunit of an ion channel that mediates extracellular Ca2þ
regulation of neuronal excitability. Proc. Natl. Acad. Sci. U. S. A. 109,
E1963eE1971.
Mitchell, P., Moyle, J., 1967. Acid-base titration across the membrane system of rat-
liver mitochondria. Catalysis by uncouplers. Biochem. J. 104, 588e600.
Moreno-Ortega, A.J., Ruiz-Nuno, A., Garcia, A.G., Cano-Abad, M.F., 2010. Mitochon-
dria sense with different kinetics the calcium entering into HeLa cells through
calcium channels CALHM1 and mutated P86L-CALHM1. Biochem. Biophys. Res.
Commun. 391, 722e726.
Nicolau, S.M., de Diego, A.M., Cortes, L., Egea, J., Gonzalez, J.C., Mosquera, M.,
Lopez, M.G., Hernandez-Guijo, J.M., Garcia, A.G., 2009. Mitochondrial Naþ/
Ca2þ-exchanger blocker CGP37157 protects against chromaffin cell death eli-
cited by veratridine. J. Pharmacol. Exp. Ther. 330, 844e854.
Giedraitis, V., Glaser, A., Sarajarvi, T., Brundin, R., Gunnarsson, M.D., Schjeide, B.M.,
Tanzi, R.E., Helisalmi, S., Pirttila, T., Kilander, L., Lannfelt, L., Soininen, H.,
Bertram, L., Ingelsson, M., Hiltunen, M., 2010. CALHM1 P86L polymorphism
does not alter amyloid-beta or tau in cerebrospinal fluid. Neurosci. Lett. 469,
265e267.
Gonzalez-Lafuente, L., Egea, J., Leon, R., Martinez-Sanz, F.J., Monjas, L., Perez, C.,
Merino, C., Garcia-De Diego, A.M., Rodriguez-Franco, M.I., Garcia, A.G.,
Villarroya, M., Lopez, M.G., de Los Rios, C., 2012. Benzothiazepine CGP37157 and
its Isosteric 20-methyl analogue provide neuroprotection and block cell calcium
entry. ACS Chem. Neurosci. 3, 519e529.
Inoue, K., Tanaka, N., Yamashita, F., Sawano, Y., Asada, T., Goto, Y., 2010. The P86L
common allele of CALHM1 does not influence risk for Alzheimer disease in
Japanese cohorts. Am. J. Med. Genet. B Neuropsychiatr. Genet. 153B, 532e535.
Koppel, J., Campagne, F., Vingtdeux, V., Dreses-Werringloer, U., Ewers, M.,
Rujescu, D., Hampel, H., Gordon, M.L., Christen, E., Chapuis, J., Greenwald, B.S.,
Davies, P., Marambaud, P., 2011. CALHM1 P86L polymorphism modulates CSF
Abeta levels in cognitively healthy individuals at risk for Alzheimer's disease.
Mol. Med. 17, 974e979.
Lambert, J.C., Sleegers, K., Gonzalez-Perez, A., Ingelsson, M., Beecham, G.W.,
Hiltunen, M., Combarros, O., Bullido, M.J., Brouwers, N., Bettens, K., Berr, C.,
Pasquier, F., Richard, F., Dekosky, S.T., Hannequin, D., Haines, J.L., Tognoni, G.,
Fievet, N., Dartigues, J.F., Tzourio, C., Engelborghs, S., Arosio, B., Coto, E., De
Deyn, P., Del Zompo, M., Mateo, I., Boada, M., Antunez, C., Lopez-Arrieta, J.,
Epelbaum, J., Schjeide, B.M., Frank-Garcia, A., Giedraitis, V., Helisalmi, S.,
Porcellini, E., Pilotto, A., Forti, P., Ferri, R., Delepine, M., Zelenika, D., Lathrop, M.,
Scarpini, E., Siciliano, G., Solfrizzi, V., Sorbi, S., Spalletta, G., Ravaglia, G.,
Valdivieso, F., Vepsalainen, S., Alvarez, V., Bosco, P., Mancuso, M., Panza, F.,
Nacmias, B., Bossu, P., Hanon, O., Piccardi, P., Annoni, G., Mann, D.,
Nicolau, S.M., Egea, J., Lopez, M.G., Garcia, A.G., 2010. Mitochondrial Naþ/Ca2þ
exchanger, a new target for neuroprotection in rat hippocampal slices. Biochem.
Biophys. Res. Commun. 400, 140e144.
Pei, Y., Lilly, M.J., Owen, D.J., D'Souza, L.J., Tang, X.Q., Yu, J., Nazarbaghi, R., Hunter, A.,
Anderson, C.M., Glasco, S., Ede, N.J., James, I.W., Maitra, U., Chandrasekaran, S.,
Moos, W.H., Ghosh, S.S., 2003. Efficient syntheses of benzothiazepines as an-
tagonists for the mitochondrial sodium-calcium exchanger: potential thera-
peutics for type II diabetes. J. Org. Chem. 68, 92e103.
Ragone, M.I., Torres, N.S., Consolini, A.E., 2013. Energetic study of cardioplegic
hearts under ischaemia/reperfusion and [Ca(2þ)] changes in cardiomyocytes of
guinea-pig: mitochondrial role. Acta Physiol. (Oxf.) 207, 369e384.
Roosen, P.C., Kallepalli, V.A., Chattopadhyay, B., Singleton, D.A., Maleczka Jr., R.E.,
Smith 3rd, M.R., 2012. Outer-sphere direction in iridium C-H borylation. J. Am.
Chem. Soc. 134, 11350e11353.
Tanis, J.E., Ma, Z., Krajacic, P., He, L., Foskett, J.K., Lamitina, T., 2013. CLHM-1 is a
functionally conserved and conditionally toxic Ca2þ-permeable ion channel in
Caenorhabditis elegans. J. Neurosci. 33, 12275e12286.
Tao, Q.Q., Sun, Y.M., Liu, Z.J., Yang, P., Li, H.L., Lu, S.J., Wu, Z.Y., 2014. Lack of asso-
ciation between CALHM1 p.P86L variation and Alzheimer's disease in the Han
Chinese population. Neurobiol. Aging 35 (1956), e1913e1954.
Taruno, A., Vingtdeux, V., Ohmoto, M., Ma, Z., Dvoryanchikov, G., Li, A., Adrien, L.,
Zhao, H., Leung, S., Abernethy, M., Koppel, J., Davies, P., Civan, M.M.,
Chaudhari, N., Matsumoto, I., Hellekant, G., Tordoff, M.G., Marambaud, P.,
Foskett, J.K., 2013. CALHM1 ion channel mediates purinergic neurotransmission
of sweet, bitter and umami tastes. Nature 495, 223e226.
Tordoff, M.G., Ellis, H.T., Aleman, T.R., Downing, A., Marambaud, P., Foskett, J.K.,
Dana, R.M., McCaughey, S.A., 2014. Salty taste deficits in CALHM1 knockout
mice. Chem. Senses 39, 515e528.