Functional inhibition of intestinal and uterine muscles by non-permeant triphenylethylene derivatives

Article abstract of DOI:10.1016/j.ejphar.2005.11.031

We have previously shown that the triphenylethylene antiestrogen tamoxifen reversibly inhibited spontaneous contractile activity in isolated duodenal muscle. Now, we have synthesized different quaternary ammonium salts of tamoxifen by changing the substit

Full text of DOI:10.1016/j.ejphar.2005.11.031

European Journal of Pharmacology 532 (2006) 115–127
www.elsevier.com/locate/ejphar
Functional inhibition of intestinal and uterine muscles by non-permeant
triphenylethylene derivatives
a
b
a
a,
⁎
Jorge Marrero-Alonso , Benito García Marrero , Tomás Gómez , Mario Díaz
a
Laboratorio de Fisiología Animal, Departamento de Biología Animal, Universidad de La Laguna, 38206 Tenerife, Spain
b
Instituto de Productos Naturales y Agrobiología. CSIC. 38206 Tenerife, Spain
Received 7 July 2005; received in revised form 7 November 2005; accepted 14 November 2005
Available online 7 February 2006
Abstract
We have previously shown that the triphenylethylene antiestrogen tamoxifen reversibly inhibited spontaneous contractile activity in isolated
duodenal muscle. Now, we have synthesized different quaternary ammonium salts of tamoxifen by changing the substituents on the nitrogen of the
alkylaminoethoxy side-chain, to obtain plasma membrane impermeable compounds. Synthesized molecules were N-desmethyl-tamoxifen-
hydrochloride, ethylbromide-tamoxifen and butylbromide-tamoxifen, which differed in the size of their ionic side-chain. All compounds rapidly
2
and reversibly inhibited spontaneous and CaCl -induced contractions in mouse duodenum and uterus. Dose-response analyses revealed a
structure-activity relationship where the larger the side-chain the higher the inhibitory potency. Fourier analyses on triphenylethylene-relaxed
duodenal tissues showed that harmonic components of contractile activity were readily recovered upon exposure to the L-type calcium channel
agonist 1,4-dihydro-2,6-dimethyl-5-nitro-4-[2-(trifluoromethyl)phenyl]-pyridine-3-carboxilic acid methyl ester (BAY-K644). Likewise, BAY-
K644 completely reversed triphenylethylene-induced effects on uterine tonic tension. Our experiments suggest that impermeant tamoxifen
derivatives relax visceral smooth muscle through a membrane-mediated non-genomic mechanism that involves inhibition of L-type calcium
channels.
©
2005 Elsevier B.V. All rights reserved.
2+
Keywords: Tamoxifen derivatives; Impermeant triphenylethylene salts; Non-genomic effects; Uterine muscle; Intestinal muscle; Ca channels
1
. Introduction
al., 1987; Batra, 1990) multidrug resistance P-glycoprotein (Kirk
et al., 1994), ligand-gated cationic channels (Allen et al., 1998),
+
+
The non-steroidal triphenylethylene antiestrogen tamoxifen
voltage-dependent Na and K currents (Hardy et al., 1998,
2
+
has been proven to be a useful clinical adjuvant in the treatment of
estrogen receptor-positive breast cancer in postmenopausal
women. Its ability to displace estrogens from their intracellular
receptor is the generally accepted antagonistic mode of action of
tamoxifen. However, there is now considerable evidence to
suggest that tamoxifen may act through estrogen receptor-
independent processes. Indeed, ithas been reportedthat tamoxifen
is able to modify the function of a number of plasma membrane
proteins. These include neurotransmitters dopamine, histamine
and muscarinic receptors (Hiemke and Ghraf, 1984; Brandes et
Smitherman and Sontheimer, 2001), Ca channels (Song et al.,
1996; Dick et al., 1999) and volume-sensitive chloride currents
(Zhang et al., 1994) and Maxi-Cl channels (Díaz et al., 2001). In
addition, crucial enzymes involved in cellular transduction, like
protein kinase C (PKC, Horgan et al., 1986), calmodulin (Lopes et
al., 1990), calmodulin-dependent cAMP-phosphodiesterase
2
+
(Lam, 1984) or Ca -ATPase (Malva et al., 1990) have also
been shown to be inhibited by tamoxifen.
Although many of the cellular proteins inhibited by tamoxifen
play crucial roles in the excitation–contraction coupling from
contractile cells, it is remarkable the low number of clinical acute
side effects reported for antiestrogens (Jordan and Murphy, 1990;
Trump et al., 1992). This is particularly relevant on smooth
muscle cells, which appear to be especially sensitive to
triphenylethylene antiestrogens. Thus, in vitro studies on isolated
⁎
Corresponding author. Laboratorio de Fisiología Animal, Departamento de
Biología Animal, Universidad de La Laguna, 38206 Tenerife, Spain. Tel.: +34
922 318343; fax: +34 922 318311.
E-mail address: madiaz@ull.es (M. Díaz).
0014-2999/$ - see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejphar.2005.11.031

116
J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
uterine (Lipton et al., 1984; Cantabrana and Hidalgo, 1992;
Kostrzewska et al., 1997), vascular (Song et al., 1996; Figtree et
al., 2000), detrusor (Ratz et al., 1999) and recently on isolated
duodenal smooth muscle (Díaz, 2002) from different species,
including human, have shown that tamoxifen rapidly inhibits
agonist-induced contractile activity, although the precise mechan-
isms of action are still poorly understood. Though tamoxifen is a
highly lipophilic molecule, most of its acute effects on smooth
muscle have been explained by interaction with targets other than
estrogen receptors that might be located at the plasma membrane;
DC amplifier (Letica, Spain) connected to a multichannel
polygraph (S460, Goerz Metrawatt. Germany) and to the A/D
interface (LabPC+, National Instruments). Voltage signals were
digitised at a sampling rate of 20 Hz (duodenal tissues) or 4 Hz
(uterine tissues) using the A/D card and visualised on a computer
screen using a data acquisition and analysis program written by
one of the authors (MD, PHYSCAN software). Data was low-
pass filtered at 1 Hz (uterine) or 5 Hz (duodenum) and analysed
using computer routines included in the acquisition software.
+
such is the regulatory β1 subunit of the Maxi K channel (Dick et
2.3. Synthesis of tamoxifen derivatives
al., 2001) or the L-type calcium channels (Díaz, 2002; Song et al.,
1
996). The aim of the present work was twofold: first, to
The N-desmethyl-tamoxifen hydrochloride was obtained
from tamoxifen with vinyl chloroformate (Olofson et al., 1977).
The free base, N-desmethyl-tamoxifen, was obtained from the
suspended salt in sodium hydroxide aqueous solution by
extraction with hexanes. The ethyl- and butyl-bromide salts
were obtained by refluxing tamoxifen in acetone in the presence
of a fourfold molecular excess of the corresponding bromide
(ethyl- or butyl-), therefore following a different method to that
reported by Jarman et al. (1986), for the ethylbromide salt. All
synthesized compounds exhibited melting points, ultraviolet,
infrared nuclear magnetic resonance and high resolution mass
spectroscopy spectra consistent with the structures assigned to
them. The microanalyses results were within ±0.4% of the
calculated values for molecular formulas. At the time of
submitting the present article, both the procedures and the
molecules were subjected to patenting processes.
synthesize different halide salts of quaternary derivatives of
tamoxifen that are unable to cross the plasma membrane. These
compounds have proven a powerful tool to unravel whether the
rapid effects of tamoxifen on smooth muscle cells involve plasma
membrane targets or if, alternatively, are due to modulation of
intracellular signalling molecules such as calmodulin or protein
kinase C. The second objective of our work was to assess the
effects of these triphenylethylene derivatives on the contractile
properties of uterine and duodenal muscles and explored their
putative mechanism of action.
2
. Materials and methods
2
.1. Tissue preparation and solutions
Duodenal segments were dissected from male mice weighing
2
6–30 g following diethyl ether anaesthesia. Strips of duodenal
2.4. Experimental procedures
smooth muscle (1.0 cm long) were immediately placed in cold
physiological salt solution (PSS), containing (in mM) NaCl,
Once mounted in the organ baths, tissues were equilibrated at
a resting tension of 0.5 g (duodenum) or 0.5 g (uterine). Muscle
resting tension was readjusted every 5 min during the 30 min
equilibration period. The maximum contraction produced by
addition of 60 mM KCl and the minimal tension produced by
calcium removal were recorded for each muscle preparation at
the beginning of each experiment. These values were used to
check the dynamic contractile response (100% RA) and to
adjust the amplifier gain. Bath solutions were replaced every
20 min and muscle tissues were washed at least three times after
application of drugs.
1
26; KCl, 4.5; MgSO .7H O, 1.0; CaCl , 2.0; NaH PO , 0.56;
4 2 2 2 4
Na HPO , 1.44; (adjusted to pH 7.4) and glucose 15.0.
2
4
Duodenal strips were incubated in aerated PSS at 37 °C in a
water-jacketed 25 ml organ bath.
Uterine tissues were obtained from female mice (weighing
2
3–27 g) one week after being subjected to bilateral
ovariectomy. Female mice were ovariectomized under ketamine
100 mg/kg) and xilazine (10 mg/kg) anaesthesia. Once
(
isolated, both uterine horns were cleaned of adherens and
placed in tyrode solution, containing (in mM), NaCl, 120; KCl,
4
.7; MgSO .7H O, 1.2; CaCl , 1.6; NaH PO , 1.0; NaHCO ,
4 2 2 2 4 3
2
5.0 (pH 7.4) and glucose 12.0. Uterine tissues were incubated
2.5. CaCl -induced contractions
2
in tyrode at 37°C in a water-jacketed 25 ml organ bath and
continuously bubbled with a 95% O /5% CO gas mixture.
In these experiments, tissues were first incubated in the
presence of Ca -free bathing solutions for 4 min. For uterine
2
2
2
+
2+
+
2+
Ca -free solutions and Ca - and K -free solutions were
prepared by replacing CaCl (supplemented with 25 μM EGTA)
tissues, bath solution was made depolarizing by adding 60 mM
2
and KCl from the corresponding saline solution.
All experimental procedures were performed in accordance
with the European Community and Local guidelines for the care
of laboratory animals.
KCl. Small volumes of 1 M CaCl were added to the bath to
2
yield the desired calcium concentration while the isometric
tension was continuously recorded. To obtain the concentra-
tion–response curves for calcium dependence, transient peak
(duodenum) or tonic (uterus) contractions were measured in
2
.2. Recording of isometric tension
response to graded concentrations of CaCl . After each
2
concentration was tested, tissues were washed three times
2
+
The isometric tension of isolated tissues was measured using
isometric force transducers (TRI110, Letica, Spain) through a
with Ca -free solutions and left for 4 min before the next
calcium concentration was assayed.

J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
117
2
.6. Effects of tamoxifen derivatives on spontaneous activity
carboxilic acid methyl ester (BAY-K8644) were obtained
from Sigma (Biosigma, Spain). 7α-[9-[(4,4,5,5,5,-pentafluor-
openty)sulphinyl]nonyl]-estra-1,3,5(19)-triene-3,17β-diol
(ICI182,780) was a gift from Astra Zeneca (Madrid, Spain).
Nifedipine was purchased from Alomone Labs (Israel).
Tamoxifen and their derivatives were dissolved in ethanol and
stored at 4 °C as 20 mM stock. BAY-K8644, verapamil and
nifedipine were dissolved in dimethyl sulfoxide (DMSO) and
stored as stock solutions at −20 °C. Solvent concentrations in the
bath never exceeded 0.12%.
Muscle preparations were incubated in PSS (duodenum) or
tyrode solution (uterus) and spontaneous activity was recorded
for 10 min after the equilibration period. Tamoxifen and
derivatives were then added in small volumes (10–12 μl)
directly to the bath solution while the time-course of muscle
activity was recorded. In the case of duodenum, comparison of
contractile activity was assessed by linear frequency analysis
(
see below) of data segments taken just before the addition of
the drug and also during the steady-state of drug effects. In some
experiments, the effects of carbachol, acetylcholine, tetraethy-
lammonium chloride, nifedipine, verapamil and BAY K8644,
on contractile activity were assessed by adding small volumes
of appropriate stock solutions directly to the bath.
3. Results
3.1. Spontaneous activity and CaCl -induced contraction
2
Duodenal tissues studied displayed characteristic spontane-
ous peristaltic activity that was entirely dependent on the
presence of extracellular calcium, i.e., calcium removal
completely abolished spontaneous activity and relaxed duode-
nal muscle (Fig. 1A). Spontaneous activity was not affected by
addition of maximal volumes of vehicles ethanol (Fig. 1B) or
DMSO (not shown). Despite some variability between tissues,
analyses using the fast Fourier transform revealed that
spontaneous activity was consistent with a principal harmonic
component corresponding to frequencies between 0.49–
0.61 Hz, which were preserved in the presence of vehicles
(Fig. 1B and D). The frequency spectrum obtained remained
similar all along the experiment with little variations over the
time. In the absence of calcium, the power density spectrum was
notably affected, and all frequency components were sup-
pressed (Fig. 1B). These observations are well in agreement
with our previous findings on this same tissue (Díaz, 2002; Díaz
et al., 2004). Progressive addition of calcium to the incubation
bath resulted in increased basal tone and wider peristaltic
amplitude (Fig. 1E). Using non-linear fitting of experimental
data to the logistic equation, we have previously shown that
half-maximal activity was obtained at 196 μM and maximal
activity observed around 1.0 mM (Fig. 1F).
2
.7. Effect of triphenylethylene derivatives on CaCl -induced
2
contraction
2
+
Smooth muscle preparations were incubated in Ca -free
2
+
+
(
duodenum) or Ca - and K -free (uterus) solutions for 4 min,
and then exposed to 2.0 mM CaCl (duodenum) or 3.0 mM
CaCl + 60 mM KCl (uterus). Both the resulting peak
2
2
(
duodenum) and tonic (uterus) contractions were used as
control values (Ttest) for subsequent effects. Afterwards, the
2
+
solution was replaced with a Ca -free solution and left for
min to stabilise. Tissues were then exposed to different
4
concentrations of triphenylethylene derivatives at the desired
concentration for another 5 min. At the end of this period, the
peak (duodenum) and tonic (uterus) contraction (Tdrug) elicited
by a second application of CaCl was measured. Bath solutions
were then replaced with fresh PSS or tyrode (in which
spontaneous activity was checked) and then left for another
2
1
0–20 min until complete recovery.
2
.8. Statistics and mathematical analyses
One-way analysis of variance (ANOVA) and t Student–
Newman–Keuls test were used to determine differences between
sample means. Values of Pb0.05 were considered significant.
Dose-response curves were fitted to logistic equation using non-
linear regression analysis tools provided in SigmaPlot software
Uterine tissues often displayed spontaneous transient
contractions which were irregular in both amplitude and
frequency (Fig. 2A). However, under depolarizing conditions
(high-potassium, 60 mM), uterine tissues always responded
with an initial transient peak contraction which was followed
by a plateau (Fig. 2A, right panel). The plateau (tonic) phase
was stable and maintained with little or no decay for up to 15
min as long as sufficient calcium ions were present in the
bathing solution (shown in Fig. 2A, right panel). Concentra-
tion-dependence experiments for uterine isometric contraction
against calcium and potassium concentrations are illustrated in
(
Jandel Scientific, San Rafael, CA). Frequency analyses of
duodenal activity were assessed using the fast Fourier transform
(
(
FFT) algorithm implemented in the acquisition software
PHYSCAN). Analyses were performed on 512 data segments
taken from data windows of 25.6 s from the steady-state phases of
each experimental condition. Linear trends were removed from
each data segment and the spectral coefficients of the power
spectra were then smoothed to reduce their variance and
assembled to obtain the average power spectrum.
+
Fig. 2B. The results showed that in the presence of 60 mM K
2
+
and 3 mM Ca tissues displayed reproducible plateau phases.
Fitting of data from steady-state plateau tension to logistic
equations demonstrated an EC 's value of 2.17 mM for
2
.9. Drugs
5
0
calcium (Fig. 2C). Given that the response to KCl was complex
and only triggered a plateau phase at high concentrations (Fig.
2B, lower panel), dose-response analysis could not be
performed.
Tamoxifen, carbachol, acetylcholine, EGTA, tetraethylam-
monium chloride (TEA-Cl), verapamil and 1,4-dihydro-2,6-
dimethyl-5-nitro-4-[2-(trifluoromethyl) phenyl]-pyridine-3-

118
J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
Fig. 1. Characterisation of duodenal contractile activity. (A) Recording of isometric tension in mouse duodenal muscle incubated in physiological saline solution (left
trace) and following replacement of calcium ions from the extracellular solution (right trace). (B) Fourier analysis of duodenal spontaneous activity determined under
control (left panel) and calcium-free conditions (right panel). Ordinates represent power spectral density (PSD) and abscises the harmonic components. Note that the
single dominant component observed at 0.507 Hz under control conditions vanishes upon calcium replacement. (C) Spontaneous isometric signals were not
substantially affected by addition of maximal vehicle volume (ethanol, 0.1%) either in time or frequency domains (D). (E) Original recording showing contractile
2 2
responses induced by cumulative addition of CaCl . (F) Concentration–response curve for CaCl -induced contraction for mouse duodenum. Values are expressed as
mean±S.E.M. (n=3).
3
.2. Synthesis of tamoxifen derivatives
We have previously shown that tamoxifen and its relevant
absolute requirement of this nitrogen for tamoxifen effects. And
third, the modification of the size of substituents on the nitrogen
of the aminoethoxy side chain (N-desmethyl-, N-ethyl- and N-
butyl-) to perform structure-activity analyses. Molecular
structures of tamoxifen and all synthesized molecules, i.e., N-
desmethyl-tamoxifen, N-desmethyl-tamoxifen hydrochloride,
ethylbromide-tamoxifen and butylbromide-tamoxifen are
shown in Fig. 3.
physiological metabolite 4-OH-tamoxifen exerted a potent
relaxing effect on duodenal peristaltic activity by an estrogen-
receptor independent mechanism likely involving an interaction
with a plasma membrane target (Díaz, 2002). In order to gain a
better understanding of the mechanism of tamoxifen action on
smooth muscle cells, we have proceeded to synthesize different
membrane-impermeant derivatives of tamoxifen, namely N-
desmethyl-tamoxifen hydrochloride, ethylbromide-tamoxifen
and butylbromide-tamoxifen. Within these molecules, the
alkylaminoethoxy side chain of tamoxifen has been modified
to hold quaternary nitrogen atoms. The rationale for the
synthesis of the different derivatives was threefold. First, the
generation of non-permeable molecules, either negatively or
positively charged in aqueous solution, therefore rendering
them impermeant to cell membranes. Second, suppression of
the unshared pair of electrons in the nitrogen of the side chain,
which are non-available in quaternary salts, to determine the
3.3. Effects of tamoxifen and membrane-impermeant deriva-
tives on duodenal contractile activity
We have explored the effects of different chemically related
membrane-impermeant tamoxifen derivatives on the spontane-
ous peristaltic activity of duodenal tissues. As can be seen in
Fig. 4A, tamoxifen, N-desmethyl-tamoxifen hydrochloride,
ethylbromide-tamoxifen and butylbromide-tamoxifen, all inhib-
ited spontaneous peristaltic activity (Fig. 4A). Measurements of
basal tone in response to tamoxifen and its derivatives showed a
significant reduction (Pb0.01) compared to respective control

J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
119
triene-3,17β-diol (ICI182,780, 10 μM) was unable to produce
any appreciable change on either peak-to-peak amplitude
(
(
98.72±9.32% vs 100% in control tissues) or basal tone
92.34±11.03% vs 100% in control tissues), paralleling our
previous findings on duodenal tissues (Díaz, 2002; Díaz et al.,
004).
In order to quantify the response of intestinal muscle to the
2
molecules under study, we have assessed their effects on
calcium-induced contractions. As can be seen in Fig. 5A,
ethylbromide-tamoxifen produced a concentration-dependent
reduction of peak contractions, with maximal inhibition being
reached around 3 μM. Similar results were observed in
response to preincubation with the other tamoxifen derivatives
(
not illustrated). Dose-response curves for non-permeant
triphenylethylene derivatives (Fig. 5B) showed that butylbro-
mide-tamoxifen was more potent (IC =0.099 μM) than
5
0
ethylbromide-tamoxifen (IC =0.37 μM), while N-desmethyl-
5
0
tamoxifen hydrochloride was the least effective triphenylethyl-
ene (IC =1.42 μM) towards inhibition of calcium-induced
5
0
contractions. Under this same experimental paradigm, N-
desmethyl-tamoxifen and tamoxifen exhibited IC50 values
of 2.46 μM (this study) and 0.85 μM (Díaz, 2002). Ac-
cordingly the inhibitory potency of triphenylethylene mole-
cules (Table 1) follows a sequence: butylbromide-tamoxifenN
ethylbromide-tamoxifenNtamoxifenNN-desmethyl-tamoxifen
hydrochloride NN-desmethyl-tamoxifen, which in turn,
matches the molecular size of side chains (see Fig. 3).
3.4. Effects of tamoxifen and membrane-impermeant deriva-
tives on uterine contractile activity
Tamoxifen and its derivatives also reduced spontaneous
activity and tonic contraction of uterine tissues observed under
depolarizing conditions (see methods). Results illustrated in
Fig. 6 shows that micromolar concentrations of tamoxifen or its
impermeant derivatives rapidly and reversibly reduced the tonic
contraction of uterine tissues induced by calcium in the
presence of high potassium. The relaxing effect of tripheny-
lethylenes was fastest for butylbromide-tamoxifen (t ≈5 s) and
Fig. 2. Characterisation of uterine contractile activity. (A) Illustrative recordings
of isometric tension from mouse uterine muscle incubated in tyrode saline
solution (spontaneous activity, left trace) and following depolarization with KCl
in the presence of high calcium to induce tonic contractile response (right trace).
5
0
(
B) Original traces showing the calcium-dependence (upper traces, bath KCl
concentration was 60 mM) and depolarization-dependence (lower traces, bath
CaCl concentration was 3.0 mM) of uterine contractile activity. (C)
Concentration–response curve for CaCl - and depolarization-induced contrac-
slowest for N-desmethyl-tamoxifen hydrochloride (t ≈50 s)
5
0
and was not mimicked by the steroidal antiestrogen 7α-[9-
2
2
tion in mouse uterus. Values are expressed as mean±S.E.M. (n=3) for
percentage fraction of 100% maximal activity.
periods (Fig. 4A, lower right panel). The inhibitory effects were
rapidly achieved within 10–180 s after exposure to micromolar
concentrations of the different triphenylethylenes derivatives
reversible upon washout in PSS (Fig. 4B). Spectral analyses
using the fast Fourier transform algorithm revealed that
exposure to tamoxifen derivatives abolished all frequency
components in the bandwidth of peristaltic activity observed in
the corresponding control periods which, in turn, reappeared
with minor changes after washout (shown for butylbromide-
tamoxifen in Fig. 4C). These effects are specific for tamoxifen
and its derivatives since the pure steroidal antiestrogen 7α-[9-
Fig. 3. Molecular structures of tamoxifen, N-desmethyl-tamoxifen and non-
[
(4,4,5,5,5,-pentafluoropenty)sulphinyl]nonyl]-estra-1,3,5(19)-
permeant derivatives.

120
J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
Fig. 4. Effects of triphenylethylene derivatives on duodenal activity. (A) Representative traces for the effects of tamoxifen and non-permeant tamoxifen derivatives on
spontaneous activity of isolated mouse duodenum. All compounds were applied at 10 μM. The lower right panel shows the effect of tamoxifen and non-permeant
tamoxifen derivatives on basal tone. Results are expressed as mean±S.E.M. of, at least, three different preparations. ** Statistically different from control conditions
with a probability value of Pb0.01. (B) The inhibitory effect of tamoxifen and derivatives was reversible upon washout (shown here for 5 μM butylbromide-
tamoxifen, BBTx). Tissues recovered basal tone and peristaltic activity once triphenylethylene molecules were removed from the bath. (C) Spectral density spectrum
of duodenal spontaneous activity determined under control conditions (left panel), 5 min after butylbromide-tamoxifen (BBTx) application to the bath (10 μM, middle
panel) and 10 min after butylbromide-tamoxifen (BBTx) washout (right panel). Ordinates represent the power spectral density (PSD) and abscises the frequency
components (Hz). Abscises have been scaled to illustrate frequencies associated with contractile activity. Similar responses were obtained for another 6 preparations.
[
(4,4,5,5,5,-pentafluoropenty)sulphinyl]nonyl]-estra-1,3,5(19)-
(vehicle) and after 5 min preincubation with the corresponding
molecule. The results demonstrated that tamoxifen and its
derivatives produced a dose-dependent reduction of both peak
and tonic contractions, with maximal inhibitions being reached
around 10 μM (shown in Fig. 7A for ethylbromide-tamoxifen)
except for N-desmethyl-tamoxifen hydrochloride that caused
maximal inhibition above 30 μM. Dose-response curves for
triphenylethylene derivatives showed that butylbromide-
triene-3,17β-diol (ICI182,780, not shown). In general, the
slower the inhibition by triphenylethylenes the longer the
recovery time, suggesting a certain degree of toxicity.
In order to quantify the relaxing effect of tamoxifen and
derivatives under study on uterine tissues, measurements were
obtained for peak and plateau contractions elicited by CaCl2
under depolarizing conditions (60 mM KCl), both in control

J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
121
+
3
.5. Effects of KCl, K channel blockers, acetylcholine and
carbachol on triphenylethylene-relaxed tissues
In an attempt to elucidate the mechanisms underlying the
relaxing action of triphenylethylene derivatives on duodenal
smooth muscle, several experiments were designed to deter-
mine the effects of several agents known to affect smooth
muscle activity on triphenylethylene-relaxed tissues. In the
presence of tamoxifen derivatives, addition of KCl (33 mM) did
not cause any appreciable change on isometric tension of
duodenal tissues (Fig. 8A, left panel), which is indicative that
relaxed tissues have lost the ability to respond to depolarization.
Similarly, addition of 5 mM tetraethylammonium chloride
+
(
TEA-Cl), a general Maxi K channel blocker failed to reverse
triphenylethylene-induced relaxation (Fig. 8A, right traces). On
the contrary, application of the muscarinic agonist carbachol
(1–10 μM) or acetylcholine (10 μM, not shown) to tripheny-
lethylne-treated duodenum was followed by a phasic peak
contraction that returned to baseline after 10–30 s after the
agonist pulse (Fig. 8B).
Similar experiments were performed on uterine tissues.
Application of TEA-Cl (10 mM) to triphenylethylene-relaxed
uteri failed to induce any significant change on isometric
tension (Fig. 8C). Conversely, addition of acetylcholine (10–
50 μM) to triphenylethylene-relaxed uteri brought about an
increase of isometric tension, (Fig. 8C).
3
.6. Effects of triphenylethylene derivatives on calcium-
dependence curves
Following similar protocols to those used in Fig. 1F for
duodenal tissues and Fig. 2C for uterine tissues, we have
assessed the effects of triphenylethylene derivatives on the
concentration–response curves for CaCl -induced contractions.
2
Fig. 5. Effects of non-permeant tamoxifen derivatives on duodenal CaCl
2
All compounds, used at concentrations around the IC50 values
shown in Table 1, caused a right shift in the calcium-
dependence curves that was accompanied by a considerable
reduction of the maximal asymptotes in both duodenal and
uterine tissues (Fig. 9). The analyses also showed that the larger
the inhibitory potency the wider the displacement of the
calcium-dependence curve to the right.
−
induced contraction. (A) Illustrative recordings showing the inhibitory effect of
2
ethylbromide-tamoxifen (EBTx) on CaCl -induced contractions in isolated
duodenal muscle strips. Calcium pulses are indicated by black rectangles under
the traces. (B) Cumulative concentration–response curves for N-desmethyl-
tamoxifen hydrochloride (NDTHCl), ethylbromide-tamoxifen (EBTx) and
butylbromide-tamoxifen (BBTx). Values are expressed as mean±S.E.M of, at
least, 4 different preparations for each concentration.
3.7. Effects of BAY K8644 on triphenylethylene-relaxed tissues
tamoxifen was more potent (IC =0.40 μM) than ethylbromide-
5
0
tamoxifen (IC =0.58 μM), while N-desmethyl-tamoxifen
Finally, we explored the effects of the L-type calcium
5
0
hydrochloride was the least effective triphenylethylene
channel agonist BAY K8644 on the relaxation induced by
tamoxifen derivatives. The relaxing effect of all tamoxifen
derivatives could be significantly counteracted by the addition
of 0.5–1 μM BAY K8644 to the bath and, both basal tone and
peristaltic amplitude returned to control values (Fig. 10A–Bd).
Interestingly, this same response was obtained when tissues
were relaxed with the L-type calcium channel antagonist
nifedipine (1 μM, not shown), a finding that may be interpreted
as the result of competition between dihydropyridines.
Frequency analysis using the fast Fourier transform algorithm
revealed that even in the continuous presence of ethylbromide-
tamoxifen, application of BAY K8644 readily restored the main
(
IC =23.42 μM) towards inhibition of calcium-induced
5
0
contractions. Interestingly, tamoxifen and N-desmethyl-
tamoxifen were poor inhibitors towards relaxation of uterine
muscle and, even at the highest concentration used, maximal
inhibition (as ascertained from the logistic minimal asymp-
tote) were never above 48% (Table 1), therefore were not
used in the structure-activity assessments. Similarly to the
aforementioned results for duodenal tissues, the inhibitory
potency sequence (Table 1) for non-permeant derivatives
was butylbromide-tamoxifenNethylbromide-tamoxifenNN-
desmethyl-tamoxifen hydrochloride.

122
J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
Table 1
Pharmacological comparison of the effects of tamoxifen and its derivatives on the contractile activity of visceral smooth muscles
Tissue
Compound
Tx
NDTx
2.46±0.18
0.34
−0.81±2.84
100.81
NDTHCl
EBTx
BBTx
Duodenum
IC50
RP
Asymptote
0.85±0.25
1.00
2.11±5.67
97.89
1.42±0.30
0.59
0.86±8.00
99.14
0.37±0.04
2.29
2.81±3.49
97.19
0.099±0.03
8.58
3.06±5.68
96.94
%
Maximal inhibition
Uterus
IC50
RP
1.17±0.12
1.00
2.15±1.00
0.54
23.44±2.73
0.05
0.58±0.18
2.01
0.40±0.19
2.92
Asymptote
64.90±6.28
35.1
51.88±6.19
48.12
4.71±37.2
95.29
25.01±4.58
74.99
21.34±3.84
78.66
%
Maximal inhibition
IC50 values are given in μM. IC50 and Asymptote values are expressed as mean±S.E.M. RP: relative potency compared to tamoxifen (Tx). NDTx: N-desmethyl-
tamoxifen; NDTHCl: N-desmethyl-tamoxifen hydrochloride; EBTx: ethylbromide-tamoxifen; BBTx: butylbromide-tamoxifen.
frequency components in the bandwidth of spontaneous activity
featuring the corresponding control period (Fig. 10B). Similar
results were obtained for triphenylethylenes butylbromide-
tamoxifen and N-desmethyl-tamoxifen hydrochloride (not
shown) and, as we have previously reported, for tamoxifen
ed for ethylbromide-tamoxifen and N-desmethyl-tamoxifen
hydrochloride in Fig. 10C), suggesting that all compounds
share a common relaxation mechanism. Similarly, exposure of
precontracted uterine tissues to the dihydropyridine nifedipine
(1 μM) or the phenothyacine verapamil (100 μM), well-known
blockers of L-type calcium channels, both caused a dose-
dependent inhibition of tonic isometric contraction (not shown),
mimicking the effect observed here for triphenylethylene
derivatives.
(
Díaz, 2002). All these data strongly suggest a crucial
involvement of L-type calcium channels in triphenylethylene-
induced duodenal muscle relaxation.
With regard to uterine tissues, experiments performed on
triphenylethylene-relaxed tissues showed that exposure to BAY
K8644 (0.5 μM) effectively recover tonic contraction in a dose-
dependent manner. Indeed, the L-type calcium channel agonist
readily increases (and restores) tonic isometric tension of tissues
relaxed with tamoxifen, N-desmethyl-tamoxifen hydrochloride,
ethylbromide-tamoxifen and butylbromide-tamoxifen (illustrat-
4. Discussion
Data reported here show that tamoxifen derivatives, N-
desmethyl tamoxifen hydrochloride, ethylbromide-tamoxifen
and butylbromide tamoxifen, three non-permeable salts of
tamoxifen, are able to inhibit spontaneous activity and calcium-
induced contractions in mice duodenal and uterine muscles.
These compounds share the central triphenylethylene core of
tamoxifen but differ in that the nitrogen of the basic
alkylaminoethoxy side chain contains a positively charged
quaternary ammonium (see structures in Fig. 3). Unlike
tamoxifen, which is a highly lipophilic molecule and can
traverse cell membranes, the ionic nature of these tamoxifen
derivatives makes them impermeable to plasma membrane.
We have observed that the effects of tamoxifen and
derivatives on contractile activity were rapidly achieved within
1
0–180 s after exposure and completely reversible upon
washout. On the contrary, steroidal antiestrogen 7α-[9-
(4,4,5,5,5,-pentafluoropenty)sulphinyl]nonyl]-estra-1,3,5(19)-
[
triene-3,17β-diol (ICI182,780) had no discernible effect on
either duodenal or uterine contractile activity, ruling out the
involvement of either canonic estrogen receptor α or mem-
brane-associated estrogen receptor α-like receptors, in the
inhibitory effect of triphenylethylene compounds. These
findings, together with the rapid onset of relaxation of duodenal
and uterine smooth muscles by triphenylethylene derivatives,
and the impermeable nature of these molecules rule out the
involvement of classic genomic pathways for the effects
reported here. In fact, tamoxifen and ethylbromide-tamoxifen
have been reported to behave as estrogen receptor antagonists
Fig. 6. Effects of tamoxifen and non-permeant tamoxifen derivatives on uterine
CaCl −induced contraction. Representative traces showing the inhibitory effect
2
of tamoxifen, N-desmethyl-tamoxifen hydrochloride (NDTHCl), ethylbromide-
tamoxifen (EBTx) and butylbromide-tamoxifen (BBTx) (all applied at 10 μM)
on CaCl - and depolarization-induced contraction in isolated uterine muscle
2
strips. Recordings were obtained from four different animals. High potassium
high calcium pulses were applied at the times indicated by the arrows under the
traces.
3
and to displace H-estradiol binding to estrogen receptors in in
vitro studies (Jarman et al., 1986). However, the inability of the

J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
123
inhibitory efficiency was related to the size of the alkylami-
noethoxy side chain. Accordingly, the larger the ionic side chain
of triphenylethylene derivatives, the higher the inhibitory
potency. It is noteworthy the low IC50 value obtained for
butylbromide-tamoxifen (0.099 μM in duodenum and 0.40 μM
in uterus), which is nearly two orders of magnitude smaller than
values reported for tamoxifen and ethylbromide-tamoxifen in
the in vitro rat uterus (Cantabrana and Hidalgo, 1992). Our data
also show that acute effects of tamoxifen on smooth muscle are
largely independent on the presence of the unshared pair of
electrons on the lateral chain of tamoxifen, which are required
for its antiestrogenic action (Murphy and Jordan, 1989), since
they were not prevented by formation of quaternary ammonium
salts.
Evidence accumulated in the last decades has demonstrated
that triphenylethylene antiestrogens are able to exert a number
of non-genomic actions by modulating the function of several
proteins, including key molecules of intracellular signalling
cascades. Studies on rat uterine myometrium have shown that
the spasmolytic effect of tamoxifen is due to inhibition of
calmodulin (Lipton and Morris, 1986). In the case of duodenal
muscle, this would produce a complete inhibition of peristaltic
activity and muscle tone because of the impairment of myosin
2
+
light chain kinase activation by the Ca -calmodulin complex
Allen and Walsh, 1994). Also, because of its ability to bind
(
calmodulin, tamoxifen can increase cyclic AMP levels by
inhibiting cyclic AMP-dependent phosphodiesterase (Lam,
1984), which, in turn, would relax smooth muscle. However,
this does not seem to be the case of tamoxifen derivatives on
either duodenum or uterus. Besides their inability to traverse
cell membrane, the fact that acetylcholine (and carbachol)
induced contractile responses in triphenylethylene-relaxed
tissues is indicative that the contractile machinery remained
functionally active. If calmodulin were inhibited then all steps
of the contractile machinery located downwards calmodulin
activation would be impaired and the smooth muscle would
remain in an inactive state. Clearly, these experiments rule out a
possible inhibition of calmodulin or activation of protein kinase
A by non-permeant tamoxifen analogues, and demonstrates that
the mechanism of relaxation by triphenylethylenes is initiated at
the plasma membrane.
Fig. 7. Concentration-dependence curves for non-permeant tamoxifen deriva-
tives on uterine CaCl -induced tonic contraction. (A) Recordings showing the
inhibitory effect of 5 min preincubation of uterine tissues with ethylbromide-
tamoxifen (EBTx) on tonic contractions induced by CaCl under depolarizing
conditions. Calcium pulses are indicated by the arrows under the traces. (B)
Concentration–response curves for N-desmethyl-tamoxifen hydrochloride
2
2
It has been reported that tamoxifen, in the range of
concentrations used here, negatively modulates different types
of potassium channels in different excitable cells (Hardy et al.,
1998; Smitherman and Sontheimer, 2001). However, it is well
(
NDTHCl), ethylbromide-tamoxifen (EBTx), butylbromide-tamoxifen (BBTx)
and tamoxifen (Tx). Values are expressed as mean±S.E.M. At least four
different preparations were used for each concentration.
+
known that both K channel blockade or depolarization by high
extracellular potassium causes a considerable increase of
contraction force generated by gastrointestinal muscles
(Meiss, 1987). Hence, it cannot be expected a blocking effect
quaternary derivative ethylbromide-tamoxifen to cross the cell
membrane and binding intracellular targets prove that tamox-
ifen but not ethylbromide-tamoxifen can inhibit estrogen-
dependent proliferation in MCF-7 cells (Jarman et al., 1986).
+
of tamoxifen derivatives on K channels since it would result in
a depolarization of smooth muscle cells, which, in turn, would
activate muscle contraction. In agreement with this hypothesis,
comparative studies using tamoxifen and ethylbromide-tamox-
ifen have demonstrated that unlike tamoxifen, ethylbromide-
Dose-response analyses on CaCl -induced (and depolariza-
2
tion-induced for uterine tissues) contractions revealed an
inhibitory potency sequence: N-desmethyl-tamoxifen hydro-
chloridebethylbromide-tamoxifenbbutylbromide-tamoxifen
for both duodenal and uterine tissues, indicating that the
+
tamoxifen fails to block voltage-dependent K channels in
NG108-15 neuroblastoma cells (Allen et al., 2000) and colonic
myocytes (Dick et al., 2002).

124
J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
+
Fig. 8. Effects of KCl, K channel blockers, acetylcholine and carbachol on triphenylethylene-relaxed muscles. (A) Illustrative recordings showing the effects of KCl
+
(
60 mM) and K channel blockade with TEA-Cl (5 mM) on the relaxation induced by tamoxifen derivatives (10 μM). (B) Representative traces showing the effects of
carbachol (CCH, 10 μM) on the relaxation induced by triphenylethylene derivatives (10 μM) on isolated duodenal muscle. Results are representative of another 4
experiments for each condition and tamoxifen derivative. (C) Sequential effects of TEA-Cl (5 mM) and acetylcholine (ACh, 10 μM) on triphenylethylene-relaxed
uteri. Traces are representative of, at least, four different preparations. Concentration of tamoxifen derivatives was 10 μM in all records. Similar results were obtained
for ethylbromide-tamoxifen (EBTx).
+
A putative activation of potassium channels and consequent
hyperpolarization of smooth muscle cells would provide a
feasible explanation for the mechanism of relaxation by
triphenylethylene derivatives. Indeed, recent studies on the
modulation of Maxi K(Ca) channels (BK) in smooth muscle cells
have shown that tamoxifen and ethylbromide-tamoxifen
increase the channel open probability by interacting with the
regulatory β1 subunit (Dick et al., 2001, 2002), and it has been
suggested that the activation of BK channels reduces the
Maxi K β1 subunit was not expressed in mouse duodenum
(Díaz et al., 2004). These observations indicate that the relaxing
effect of tamoxifen derivatives must have been occurred on the
coupling of the oscillating membrane potential and the voltage-
dependent calcium influx.
This hypothesis was tested by using the dihydropiridine
derivative BAY K8644, a well-known calcium channel agonist,
which increases the current through the L-type calcium
channels of nerve and muscle cells (Schramm et al., 1983;
Coruzzi and Poli, 1985). Our present results demonstrate that
inhibition by triphenylethylene derivatives could be counter-
acted by reactivating calcium influx through L-type calcium
channel with the dihydropiridine BAY K8644. Interestingly, the
isometric tension developed by duodenal tissues followed a
peristaltic pattern, which exhibited a frequency spectrum similar
to that present in control tissues before the addition of
triphenylethylenes. This result strongly points to a blockade
of L-type calcium channels by tamoxifen derivatives being
2
+
voltage-dependent calcium influx and [Ca ] through tonic
i
hyperpolarization of smooth muscle cells (Lohn et al., 2001).
However, in our case, the fact that high KCl-induced
depolarisation of duodenal muscle and the blockade of
+
duodenal and uterine K channels with TEA, did not induce
any noticeable change in the recorded tension of duodenal or
uterine tissues (see Fig. 7), strongly suggest that the effect of
+
tamoxifen derivatives was not related to activation of K
channels. On the other hand, we have previously observed that

J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
125
2+
Fig. 9. Effects of non-permeant tamoxifen derivatives on calcium-dependence curves. Concentration–response curves for Ca -induced contraction in uterine (A) and
duodenal (B) tissues. Data were obtained in controls and after 5 min preincubation of duodenal and uterine tissues with butylbromide-tamoxifen (BBTx),
ethylbromide-tamoxifen (EBTx) and N-desmethyl-tamoxifen hydrochloride (NDTHCl). In duodenal tissues the concentrations of triphenylethylene derivatives were
1
2
00 nM, 1.2 and 1.4 μM, for BBTx, EBTx and NDTHCl, respectively. In uterine tissues the concentrations of BBTx, EBTx and NDTHCl were 0.4 μM, 0.58 μM and
3.0 μM, respectively. Values are expressed as mean±S.E.M. At least four different preparations were used for each tissue and tamoxifen derivative.
responsible for the spasmolytic action of tryphenylethylene
derivatives on mouse duodenum. Likewise, the ability of BAY
K8644 to fully recover the tonic contraction in triphenylethy-
lene-relaxed uterine tissues demonstrates that L-type calcium
channels play a central role in the mechanism of relaxation.
Accordingly, analyses of calcium-dependence curves on
duodenal and uterine tissues measured at the IC50 for each
tryphenylethylene derivative, showed both a right shift in the
concentration-dependence curve and a significant reduction of
maximal contraction that was more pronounced for BBTx,
paralleling its relative inhibitory potency. In agreement with our
hypothesis, several studies have demonstrated that tamoxifen
inhibits calcium entry through L-type calcium channels in A7r5
and aortic smooth muscle cells (Song et al., 1996), colonic
myocytes (Dick et al., 1999), rabbit detrusor (Ratz et al., 1999),
and non-muscle cells like clonal pituitary cells (Sartor et al.,
inward calcium currents in small intestine smooth muscle
cells, the depolarisation phase of ICC is insensitive to L-type
calcium channel blockers and abolished by hyperpolarization
(Lee et al., 1999). On the other hand, despite the existence of
spontaneous contractile activity in uterine tissues, ICC-like
pacemaking cells have not been found in uterus and the
spontaneous activity appears to be a property of smooth
muscle cells (Wray et al., 2001). Second, the power density
spectrum in the presence of BAY K8644 after triphenylethy-
lene relaxation showed similar frequency components in the
control periods in the same tissues (see Fig. 10). This finding
can only be explained if the mechanisms responsible for
generating the pace within the ICC remained unaffected by
quaternary tamoxifen analogues.
In summary, non-permeant tamoxifen derivatives relax
duodenal and uterine smooth muscles in a rapid and reversible
fashion by a common mechanism most likely involving
inhibition of L-type calcium channels. Since quaternary
derivatives studied here are unlikely to fully partition into a
lipid bilayer because of its charged quaternary group, the
binding site for non-steroidal quaternary compounds is most
likely on the extracellular domain of the channel, likely the
dihydropiridine binding site, rather than the intracellular
hydrophobic membrane spanning region. Studies using fluo-
rescent calcium probing and electrophysiological experiments
1
988) and PC12 neurosecretory cells (Greenberg et al., 1987).
Furthermore, antiestrogens tamoxifen and clomiphene have
3
been shown to compete H-nitrendipine binding in membrane
fractions of human and rabbit urinary bladder and myometrium
(
Batra, 1990).
Finally, two pieces of evidence suggest that the main target
for the actions of tamoxifen derivatives are likely the smooth
muscle cells themselves rather than the interstitial Cajal
pacemaker cells (ICC). First, it has been shown that unlike

126
J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
Fig. 10. Effect of BAY K8644 on tamoxifen derivatives-relaxed tissues. (A) Representative recording showing the effects of BAY K8644 (1 μM) on tamoxifen
derivatives-relaxed duodenal muscles. Traces were obtained in different preparations and were mimicked by tamoxifen and butylbromide-tamoxifen. Similar results
were obtained in another 4 preparations. (B) Frequency analysis of duodenal spontaneous activity determined under conditions indicated above. Ordinates represent
power spectral density (PSD) and abscises the harmonic components. Tension traces and frequency spectra are representative of 3 different experiments for each
compound. (C) Effects of application of BAY K8644 (0.5 μM) on triphenylethylene-induced inhibition of uterine tonic contraction. Concentration of non-permeant
derivatives was 10 μM. Similar effects were produced in ethylbromide-tamoxifen-relaxed tissues. Data are representative of, at least, three different experiments.
on calcium channels in isolated duodenal and uterine smooth
muscle cells are being carried out to ascertain such hypothesis.
Overall, these data add new insights in the rationale for the
development of a new generation of more specific and selective
estrogen receptor modulators.
a fellowship from convenio ULL-CajaCanarias (Spain). We
wish to thank Mari Carmen González-Montelongo for her
collaboration in caring the animals and cooperation in some of
the experiments and to Dr. Raquel Marín for critical reading of
the manuscript.
References
Acknowledgements
Allen, M.C., Gale, P.A., Hunter, A.C., Lloyd, A., Hardy, S.P., 2000. Membrane
impermeant antioestrogens discriminate between ligand-and voltage-gated
cation channels in NG108-15 cells. Biochim. Biophys. Acta 1509, 229–236.
Allen, B.G., Walsh, M.P., 1994. The biochemical basis of the regulation of
smooth-muscle contraction. TIBS 19, 362–368.
Supported by research grant numbers SAF2001-3614-C03-
2 (Ministerio de Ciencia y Tecnología, Spain), PI042460 (FIS,
Ministerio de Sanidad y Consumo, Spain) and the Spanish
Network for Neurological Disorders (CIEN, Spain). JMA holds
0

J. Marrero-Alonso et al. / European Journal of Pharmacology 532 (2006) 115–127
127
Allen, M.C., Newland, C., Valverde, M.A., Hardy, S.P., 1998. Inhibition of
ligand-gated cation-selective channels by tamoxifen. Eur. J. Pharmacol. 354,
Kostrzewska, A., Laudanski, T., Batra, S., 1997. Potent inhibition by tamoxifen
of spontaneous and agonist-induced contractions of the human myometrium
and intramyometrial arteries. Am. J. Obstet. Gynecol. 176, 381–386.
Lam, H.Y., 1984. Tamoxifen is a calmodulin antagonist in the activation of
cAMP phosphodiesterase. Biochem. Biophys. Res. Comm. 118, 27–32.
Lee, J.C., Thuneberg, L., Berezin, I., Huizinga, J.D., 1999. Generation of slow
waves in membrane potential is an intrinsic property of intersticial cells of
Cajal. Am. J. Physiol. 277, G409–G423.
Lipton, A., Morris, I.D., 1986. Calcium antagonism by the antiestrogen
tamoxifen. Cancer Chemother. Pharmacol. 18, 17–20.
Lipton, A., Vinijsanum, A., Martin, L., 1984. Acute inhibition of rat myometrial
responses to oxytocin by tamoxifen steroisomers and estradiol. J.
Endocrinol. 103, 383–388.
261–269.
Batra, S., 1990. Interaction of antiestrogens with binding sites for muscarinic
cholinergic drugs and calcium channel blockers in cell membranes. Cancer
Chemother. Pharmacol. 26, 310–312.
Brandes, L.J., McDonald, L.M., Bogdanovic, R.P., 1987. Histamine and growth:
interaction of antiestrogen binding site ligands with a novel histamine site
that may be associated with calcium channels. Cancer Res. 47, 4025–4031.
Cantabrana, B., Hidalgo, A., 1992. Effects of nonsteroidal antiestrogens in the in
vitro rat uterus. Pharmacology 45, 329–337.
Coruzzi, G., Poli, E., 1985. The stimulatory action of the calcium channel
agonist Bay K8644 on isolated duodenal muscle. Antagonism by nifedipine
and verapamil. Naunyn-Schmiedeberg's Arch. Pharmacol. 331, 290–292.
Díaz, M., 2002. Triphenylethylene antiestrogen-induced acute relaxation of
mouse duodenal muscle. Possible involvement of calcium channels. Eur. J.
Pharmacol. 445, 257–266.
Lohn, M., Lauterbach, B., Haller, H., Pongs, O., Luft, F.C., Gollasch, M., 2001.
2+
Beta(1)-subunit of BK channels regulates arterial wall [Ca ] and diameter
in mouse cerebral arteries. J. Appl. Physiol. 91, 1350–1354.
2+
Lopes, M.C.F., Vale, M.G.P., Carvalho, A.P., 1990. Ca -dependent binding of
tamoxifen to calmodulin isolated from bovine brain. Cancer Res. 50,
2753–2758.
Díaz, M., Bahamonde, M.I., Lock, H., Muñoz, F., Hardy, S.P., Posas, F.,
−
Valverde, M.A., 2001. Okadaic acid-sensitive activation of Maxi Cl
channels by triphenylethylene antiestrogens in C1300 neuroblastoma cells.
J. Physiol. 536, 79–88.
Díaz, M., Ramirez, C.M., Marin, R., Marrero-Alonso, J., Gomez, T., Alonso, R.,
Malva, J.O., Lopes, M.C.F., Vale, M.G.P., Carvalho, A.P., 1990. Actions of
2+
2+
+
2+
antiestrogens on the (Ca Mg )-ATPase and Na /Ca exchange of brain
cortex membranes. Biochem. Pharmacol. 40, 1877–1884.
2
004. Acute relaxation of mouse duodenun by estrogens. Evidence for an
Meiss, R.A., 1987. Mechanical properties of gastrointestinal smooth muscle. In:
Johnson, L.R. (Ed.), Physiology of the Gastrointestinal Tract. Raven Press,
New York, NY, pp. 273–329.
estrogen receptor-independent modulation of muscle excitability. Eur. J.
Pharmacol. 501, 161–178.
Dick, G.M., Kong, I.D., Sanders, K.M., 1999. Effects of anion channel
Murphy, C.S., Jordan, V.C., 1989. Structural components necessary for the
antiestrogenic activity of tamoxifen. J. Steroid Biochem. 34, 407–411.
Olofson, R.A., Schnur, R.C., Bunes, L., Pepe, J.P., 1977. Selective N-
Dealkylation of tertiary amines with vinyl chloroformate: an improved
synthesis of naloxone. Tetrahedron Lett. 18, 1567–1570.
Ratz, P.H., MacCammon, K.A., Altstatt, D., Blackmore, P.F., Shenfeld, O.Z.,
Schlossberg, S.M., 1999. Differential effects of sex hormones and
phystoestrogens on peak and steady state contractions in isolated rabbit
detrusor. J. Urol. 162, 1821–1828.
−
antagonists in canine colonic myocytes: comparative pharmacology of Cl ,
2+
+
Ca and K currents. Br. J. Pharmacol. 127, 1819–1831.
Dick, G.M., Rossow, C.F., Smirnov, S., Horowitz, B., Sanders, K.M., 2001.
Tamoxifen activates smooth muscle BK channels through the regulatory
beta 1 subunit. J. Biol. Chem. 276, 34594–34599.
Dick, G.M., Hunter, A.C., Sanders, K.M., 2002. Ethylbromide tamoxifen, a
membrane-impermeant antiestrogen, activates smooth muscle calcium-
activated large-conductance potassium channels from the extracellular side.
Mol. Pharmacol. 61, 1105–1113.
Figtree, G.A., Webb, C.M., Collins, P., 2000. Tamoxifen acutely relaxes
coronary arteries by an endothelium-, nitric oxide-, and estrogen receptor-
dependent mechanism. J. Pharmacol. Exp. Ther. 295, 519–523.
Greenberg, D.A., Carpenter, C.L., Messing, R.O., 1987. Calcium channel
antagonist properties of the antineoplastic antiestrogen tamoxifen in the
PC12 neurosecretory cells. Cancer Res. 47, 70–74.
Hardy, S.P., deFelipe, C., Valverde, M.A., 1998. Inhibition of voltage-gated
cationic channels in rat embrionic hypothalamic neurones and C1300
neuroblastoma cells by triphenylethylene antiestrogens. FEBS Lett. 434,
Sartor, P., Vacher, P., Mollard, P., Dufy, B., 1988. Tamoxifen reduces calcium
currents in a clonal pituitary cell line. Endocrinology 123, 534–540.
Schramm, M., Thomas, G., Towart, R., Franckowiak, G., 1983. Activation of
calcium channels by novel 1,4-dihydropyridines. A new mechanism for
positive inotropics of smooth muscle stimulants. Arzneimittelforsch 33,
1268–1272.
+
+
Smitherman, K.A., Sontheimer, H., 2001. Inhibition of glial Na and K currents
by tamoxifen. J. Membr. Biol. 181, 125–135.
Song, J., Standley, P.R., Zhang, F., Joshi, D., Gappy, S., Sowers, J.R., Ram, J.L.,
1996. Tamoxifen (estrogen antagonist) inhibits voltage-gated calcium
current and contractility in vascular smooth muscle from rats. J. Pharmacol.
Exp. Ther. 277, 1444–1453.
Trump, D.L., Smith, D.C., Ellis, P.G., Rogers, M.P., Schold, S.C., Winer, E.P.,
Panella, T.J., Jordan, V.C., Fine, R.L., 1992. High-dose oral tamoxifen, a
potential multidrug-resistance-reversal agent: phase I trial in combination
with vinblastine. J. Natl. Cancer Inst. 84, 1811–1816.
239–240.
Hiemke, C., Ghraf, R., 1984. Interaction of non-steroidal antiestrogens with
dopamine receptor binding. J. Steroid Biochem. 21, 663–667.
Horgan, K., Cooke, E., Hallet, M.B., Mansel, R.E., 1986. Inhibition of protein
kinase C mediated signal transduction by tamoxifen. Biochem. Pharmacol.
35, 4463–4465.
Jarman, M., Leung, O.T., Leclercq, G., Devleeschouwer, N., Stoessel, S.,
Coombes, R.C., Skilton, R.A., 1986. Analogues of tamoxifen: the role of the
basic side-chain. Applications of a whole-cell oestrogen-receptor binding
assay to N-oxides and quaternary salts. Anti-Cancer Drug Des. 1, 259–268.
Jordan, V.C., Murphy, C.S., 1990. Endocrine pharmacology of antiestrogens as
antitumor agents. Endocr. Rev. 11, 579–610.
Kirk, J., Syed, S.K., Harris, A.L., Jarman, M., Roufogalis, B.D., Stratford, I.J.,
Carmichael, J., 1994. Reversal of P-glycoprotein-mediated multidrug
resistance by pure anti-oestrogens and novel tamoxifen derivatives.
Biochem. Pharmacol. 48, 277–285.
Wray, S., Kupittayanant, S., Shmygol, A., Smith, R.D., Burdyga, T., 2001. The
physiological basis of uterine contractility: a short review. Exp. Physiol. 86,
239–246.
Zhang, J.J., Jacob, T.J.C., Valverde, M.A., Hardy, S.P., Mintenig, G.M.,
Sepúlveda, F.V., Gill, D.R., Hyde, S.C., Trezise, A.E.O., Higgins, C.F.,
1994. Tamoxifen blocks chloride channels: a possible mechanism for
cataract formation. J. Clin. Invest. 94, 1690–1697.