Pharmacological inhibition of outwardly rectifying Cl- currents in rat peritoneal mast cells: A comparison of different stilbene derivatives

Article abstract of DOI:10.1016/j.ejphar.2003.08.010

Diethylstilbestrol and other stilbene derivatives can provide some inhibition of the outwardly rectifying Cl^- current (I _(Cl-,OR)) in rat peritoneal mast cells. In order to elucidate structure-activity relationships of diethylstilbestrol, 12 stilbenes as well as 17β-estradiol and hexestrol were tested in rat peritoneal mast cells using the nystatin-perforated patch approach of the whole-cell patch-clamp technique. Since trans-stilbene showed no effect, the substituents of diethylstilbestrol must be of importance. The introduction of only one hydroxy group in trans-stilbene produced potent inhibition of the I_(Cl-,OR) (IC ₅₀: 3.3 μM). But in contrast, resveratrol with hydroxy groups at positions 4, 3′, and 5′ as well as methoxy substituted stilbene derivatives and 17β-estradiol were ineffective. On the other hand, hexestrol potently inhibited I_(Cl-,OR) indicating that the aromatic ring systems can also be connected by an ethyl bridge. In summary, a hydroxy group at position 4 (or 4′) is a prerequisite for diethylstilbestrol- mediated inhibition of I_(Cl-,OR).

Full text of DOI:10.1016/j.ejphar.2003.08.010

European Journal of Pharmacology 478 (2003) 81–88
www.elsevier.com/locate/ejphar
À
Pharmacological inhibition of outwardly rectifying Cl currents in rat
peritoneal mast cells: a comparison of different stilbene derivatives
a
Tim Roloff , Albrecht Ziegler , Dieter Heber , J o¨ rg Seebeck
a
b
a,*
a
Institute of Pharmacology, University of Kiel, Hospitalstr. 4, D-24105 Kiel, Germany
b
Pharmaceutical Institute, University of Kiel, Gutenbergstr. 76, D-24118 Kiel, Germany
Received 19 May 2003; received in revised form 1 August 2003; accepted 8 August 2003
Abstract
À
Diethylstilbestrol and other stilbene derivatives can provide some inhibition of the outwardly rectifying Cl current (I(ClÀ,OR)) in rat
peritoneal mast cells. In order to elucidate structure–activity relationships of diethylstilbestrol, 12 stilbenes as well as 17h-estradiol and
hexestrol were tested in rat peritoneal mast cells using the nystatin-perforated patch approach of the whole-cell patch-clamp technique. Since
trans-stilbene showed no effect, the substituents of diethylstilbestrol must be of importance. The introduction of only one hydroxy group in
trans-stilbene produced potent inhibition of the I(ClÀ,OR) (IC50: 3.3 AM). But in contrast, resveratrol with hydroxy groups at positions 4, 3V,
and 5Vas well as methoxy substituted stilbene derivatives and 17h-estradiol were ineffective. On the other hand, hexestrol potently inhibited
I
(ClÀ,OR) indicating that the aromatic ring systems can also be connected by an ethyl bridge. In summary, a hydroxy group at position 4 (or 4V)
is a prerequisite for diethylstilbestrol-mediated inhibition of I(ClÀ,OR)
D 2003 Elsevier B.V. All rights reserved.
.
À
Keywords: Stilbene; Mast cell; Cl channel; Exocytosis; Diethylstilbestrol; Hexestrol
1
. Introduction
cyanatostilbene-2,2V-disulfonic acid), e.g. SITS (4V-acet-
amido-4V-isothiocyanatostilbene-2,2V-disulfonic acid), and
diethylstilbestrol (E-a,h-diethyl-4,4V-stilbenediol) (Mat-
thews et al., 1989; Roloff et al., 2001). The structural
properties which enable stilbene molecules to inhibit
I(ClÀ,OR) are unknown. In order to address this question,
a panel of 14 different stilbene and structurally related
non-stilbene molecules were tested in the present study
regarding their ability to inhibit I(ClÀ,OR) in rat peritoneal
mast cells. Because diethylstilbestrol acts as an exception-
ally potent inhibitor of I(ClÀ,OR), this molecule was taken
as a starting point and lead compound for structural
variation (Roloff et al., 2001). Chemically, diethylstilbes-
trol is a trans-stilbene whose aliphatic ethene part is
substituted by two ethyl groups. The phenyl rings of
diethylstilbestrol carry each a hydroxy group at (para)-
positions 4 and 4V. The stilbene derivatives assessed in the
present study can be grouped into molecules either
modified at their aromatic rings (DIDS, resveratrol, SITS,
trans-4-hydroxystilbene, trans-4-hydroxy-4V-methoxystil-
bene), or at their aliphatic ethene part (tamoxifen, trans-
diethylstilbene, cis-diethylstilbene), or both (clomiphene,
chlorotrianisene, diethylstilbestrol). In addition, trans-stil-
bene, hexestrol (a diethylstilbestrol related non-stilbene)
Rat peritoneal mast cells express a non-voltage gated
À
outwardly rectifying Cl channel (Matthews et al., 1989;
Penner et al., 1988). In the open state, this channel allows
À
Cl to enter the cell interior and thereby to achieve a
negative membrane potential (V ). The membrane poten-
m
tial in rat peritoneal mast cells is of physiological impor-
tance since it influences the electrochemical driving force
2
+
for transmembranous Ca currents (Matthews et al., 1989;
Penner et al., 1988). The molecular identity of the out-
À
wardly rectifying Cl channel (ORCC) expressed in rat
peritoneal mast cells is currently unclear. A recent study
À
has provided evidence that type 7 Cl channel (ClC-7)-
mRNA is present in rat peritoneal mast cells (Kulka et al.,
2
002). However, the presence of ClC-7-protein within the
plasma membrane of these cells remains to be shown.
A characteristic feature of outwardly rectifying chan-
nel-mediated Cl currents (I(ClÀ,OR)) is a potent inhibition
À
by certain stilbene derivatives, e.g. DIDS (4,4V-diisothio-
*
Corresponding author. Tel.: +49-431-597-3504; fax: +49-431-597-
3522.
E-mail address: joerg.seebeck@web.de (J. Seebeck).
0014-2999/$ - see front matter D 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejphar.2003.08.010

82
T. Roloff et al. / European Journal of Pharmacology 478 (2003) 81–88
and 17h-estradiol (a steroid estrogen receptor agonist)
were tested.
Germany) as previously described (Mousli et al., 1989;
Roloff et al., 2001). In brief, rats in deep ether anesthesia
were decapitated. Then, 20 ml of extracellular solution
containing NaCl 137, KCl 2.7, CaCl2 2.0, MgCl2 5.0,
HEPES 10.0, ascorbic acid 1.0, and glucose 5.6 mM (pH
7.3) were injected into the peritoneal cavities. After a
short abdominal massage, the lavage fluid was reaspirated
using a wide-bore pipette through a small incision in the
abdominal wall. Aliquots (50 Al) of the resulting cell
suspension were plated on glass cover-slips and kept in
the dark at 22 jC until use (within 120 min). Cover slips
were placed below an inset in 35-mm culture dishes
(fluid volume c 500 Al) and mounted on the stage of
an inverted microscope (Axiovert 135, Zeiss, G o¨ ttingen,
Germany). Before starting an experiment, cover-slips
were superfused with extracellular solution in order to
remove non-adherent cells and particles. The peritoneal
mast cells were identified under the light microscope by
their characteristic morphological features (round shape,
granular appearance, small prominent nucleus).
Previous studies have shown that some stilbene deriva-
tives (e.g. DIDS, diethylstilbestrol) at higher concentrations
(
>10 AM) are capable to impair agonist-induced exocytotic
responses in rat peritoneal mast cells via I(Cl ,OR)-indepen-
dent mechanisms (Roloff et al., 2001; Vliagoftis et al.,
992). Interestingly, this secretostatic effect depends on
the type of stimulus used. Secretagogues whose action
À
1
2
+
depends on the presence of extracellular [Ca ]e (nerve
growth factor [NGF] in the presence of lysophosphatidyl-
serine [lysoPS], antigen/immunoglobulin E [Ag/IgE]) are
2
+
inhibited more potently than (Ca )e-independent acting
stimuli (compound 48/80, mastoparan) (Hill et al., 1996;
Pearce and Thompson, 1986; Roloff et al., 2001). There-
fore, the stilbene derivatives chosen were also assessed
regarding their ability to inhibit NGF/lysophosphatidylser-
ine- and mastoparan-stimulated exocytotic responses. As a
3
measure of exocytosis the release of [ H]5-hydroxytrypta-
3
3
mine ([ H]5-HT) from [ H]5-HT-loaded rat peritoneal mast
cells was used in the present study. Furthermore, the effect
3
of the stilbene derivatives on unstimulated [ H]5-HT-release
2.2. Electrophysiological recordings
was tested, in order to monitor for putative toxic/membra-
nolytic actions.
For nystatin-perforated patch recordings, the pipette or
internal solution always contained KCl 137, NaCl 2.7,
MgCl 3.0, CaCl 1, HEPES 10.0 mM (pH 7.3). Imme-
2
2
2
. Materials and methods
diately before starting the recordings, nystatin was dis-
À 1
solved in dimethylsulfoxide (DMSO) at 250,000 U ml
.
The compounds used were purchased from different
Aliquots of this stock solution were mixed with internal
solution to a yield a final nystatin concentration of 1250
manufacturers (as indicated) and of the highest available
degree of purity: chlorotrianisene, clomiphene, DIDS, dieth-
ylstilbestrol, 17h-estradiol, HEPES, hexestrol, lysophospha-
tidylserine, nerve growth factor 7S (mouse submandibular
gland), nystatin, SITS, tamoxifen, trans-resveratrol, trans-
stilbene were all received from Sigma-Aldrich Chemie (Dei-
À 1
U ml . Whole-cell currents were measured in voltage-
clamp experiments using the nystatin-perforated-patch
approach of the whole-cell patch-clamp technique (Hamill
et al., 1981; Hill et al., 1996; Horn and Marty, 1988). All
electrophysiological experiments were performed at room
temperature (20–24 jC). Patch pipettes were pulled from
borosilicate glass (Hilgenberg, Malsfeld, Germany) and
fire-polished to a resistance of 2.0–3.5 MV when filled
with the nystatin-containing internal solution. The tips of
the patch-pipettes were filled with nystatin-free buffer
solution. As previously described (Hill et al., 1996),
when challenged with step depolarisations, perforation
of the membrane patch was indicated by characteristic
changes of the capacitance transient and by a continuos
decline of the input resistance. To standardise experi-
ments, recordings were started when the input resistance
fell to a value of 25 MV or less; this point was reached
usually 5–10 min after gigaseal formation. Whole-cell
perforated-patch recordings were acquired by a computer-
based patch-clamp amplifier system (EPC-9, HEKA Elec-
tronics, Germany). Capacitance transients were cancelled
before each series of test pulses using the automatic
compensation of the EPC-9. Currents were filtered using
an 8-pole Bessel filter at 2.9 Hz and digitized at 100 As.
Series of incremental voltage pulses (from À 80 to + 80
mV in 20-mV steps, duration 100 ms) were applied every
senhofen, Germany). Mastoparan was obtained from Bachem
(
3
Heidelberg, Germany). 5-[1,2- H[N]]-hydroxytryptamine
3
creatine sulfate ([ H]5-HT) was purchased from NEN (Co-
logne) Germany. The synthesis of the E- and Z-isomeric
forms of diethylstilbene was accomplished by reductive
coupling of propiophenone using titanium chloride/zinc/
pyridine; the mixture of the isomers was separated and
purified on a silica gel column (Leimner and Weyerstahl,
1
982). Trans-4-hydroxybenzaldehyde was obtained by the
Wittig reaction of benzyltriphenylphosphonium chloride and
-hydroxybenzaldehyde in the presence of sodium ethanolate
4
and trans-4-hydroxy-4V-methoxystilbene prepared using the
same method with corresponding educts (Friedrich and Hen-
1
ning, 1959). All products were characterized by H and IR
spectra and also by comparison of the melting points with
literature data.
2
.1. Preparation of rat peritoneal mast cells
Rat peritoneal mast cells were obtained by peritoneal
lavage from male Wistar rats (Charles River, Sulzfeld,

T. Roloff et al. / European Journal of Pharmacology 478 (2003) 81–88
83
3
second from holding potential 0 mV (standard pulse
protocol). A complete series of test pulses was repeated
at 60- or 30-s intervals. All currents shown were cor-
rected by leak subtraction. Leak currents were determined
by applying six short voltage steps of alternating polarity
release of [ H]5-HT was always below 5%. All experimental
conditions were determined in duplicate.
2.4. Log P-values
(
leak holding potential À 100 mV) at the start of each
As a measure of lipophilicity, Log P-values of the
compounds were determined. The Log P-values were cal-
culated by using the online version of the program LogKow/
KowWin (http://esc.syrres.com/interkow/kowdemo.htm).
test pulse. Steady-state currents were determined by
calculating the mean of current signals obtained over
the last 10 ms of single sweeps. For analysis of drug
effects, the current responses at F 80 mV over time were
used. Test compounds were applied by superfusion after
the current maximum has been reached. Per experiment
only a single concentration of each test compound was
assessed.
2.5. Statistics and curve fitting
Data are means F S.E.M. Concentration–response
curves were fitted to a four parameter logistic equation
through computer-assisted curve fitting (Prism 2, GraphPad
software, San Diego, USA). The equation fitted was:
2
.3. Secretion assay
(
log IC50 À X)nH
Response = Imin+(Imax À Imin)/(1 + 10
), where
Imin is the control response, I is the maximal inhibition,
For secretion assays, a modified Krebs–Ringer–Hense-
leit buffer (MKRH) containing 0.2% bovine serum albumin
and NaCl 137, KCl 2,7, CaCl 0,3, MgCl 1, NaH PO 0,4,
max
X is the test compounds concentration and nH the Hill
coefficient.
2
2
2
4
HEPES 10, and glucose 5,6 mM (pH 7.3) was used. Experi-
ments were performed as previously described (Mousli et al.,
1
989). Briefly, peritoneal lavages of three rats were com-
3. Results
bined and the containing mast cells were enriched using a
bovine serum albumin-gradient centrifugation procedure to
at least 90% purity. Consecutively, the enriched mast cells
solution was incubated for 60 min, at 37 jC, in the presence
3.1. Inhibition of I(
À
(see also Table 1)
Cl ,OR)
In a few whole-cell recording experiments, transient
openings of large conductance ion channels were observed
(see Fig. 2). However, the contribution of this currents to
the whole cell current was negligible since no shift of the
reversal potential (c 0 mV) could be observed after inhi-
3
of 1 ACi/ml 5-[1,2- H[N]]-hydroxytryptamine creatine sul-
fate. After washing, the cells were resuspended in buffer and
stimulated in the absence or presence of different test
compounds in a shaking water bath (37 jC), as indicated.
The amount of the total cellular radioactivity released into
the supernatant was determined and, expressed in [% of
total], taken as a measure of secretion. Spontaneous release
of radioactivity increased with incubation time; within the
first 20 min, in the absence of test compounds, spontaneous
bition of I(
À
(Table 1).
Cl ,OR)
3.1.1. Trans-stilbene (Fig. 1A)
Trans-stilbene represents the basic stilbene motif pres-
ent in all stilbene derivatives tested in the present study
Table 1
The table summarises the results obtained for different stilbene derivatives and hexestrol regarding their ability to inhibit outwardly rectifying chloride currents
(ClÀ,OR)) and agonist-induced exocytosis in rat peritoneal mast cells
(I
3
[ H]5-HT secretion (IC50)
Substance
Log P
I
(Cl
À
,OR)
IC50 [AM]
Inhibition at 10 AM
NGF/lyso-phosphatidyl-serine
Mastoparan
[
% of control]
(AM)
(AM)
Clomiphene
6.7
6.7
6.3
6.2
5.6
5.6
4.5
4.1
5.1
4.1
85
82
89
5
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
Trans-diethylstilbene
Tamoxifen
3.5
Chlorotrianisene
Diethylstilbestrol
Hexestrol
>10
1.6
98
92
26
9
1.1
Trans-stilbene
Trans-4-hydroxy-4V-
methoxystilbene
Trans-4-hydroxystilbene
DIDS
>10
>10
4.0
3.1
3.1
0.9
3.2
1.4
79
>10
96
>10
>100
>10
100
2.3
58
Resveratrol
>10
9.2
>10
72
SITS
>100
The compounds were ordered by their respective Log P-values.

84
T. Roloff et al. / European Journal of Pharmacology 478 (2003) 81–88
(
with the exception of cis-diethylstilbene). Superfusion of
AM, resveratrol only slightly reduced I(
À
by about
Cl ,OR)
rat peritoneal mast cells with 10 AM of trans-stilbene
reduced steady-state Cl currents elicited by a test poten-
2.3 F 2.3% (n = 3). Superfusion of rat peritoneal mast cells
with 100 AM of resveratrol caused a suppression of the
À
tial of + 80 mV by 25.9 F 9.1 [(% of control); n = 3]. This
small inhibitory effect could be fully reversed by super-
fusion with stilbene-free buffer solution.
Trans-stilbene derivatives substituted at their phenyl
nucleus.
I(
inhibitory effect was completely reversible.
À
by about 46.6 F 7.0 [(% of control); n = 6]. This
Cl ,OR)
3.1.5. DIDS (Fig. 1E) and SITS (Fig. 1G)
The anions of the sulfonic acid derivatives DIDS and
SITS were used. In contrast to all other stilbene derivatives
tested the extent of the inhibitory effect of DIDS and SITS
3
.1.2. Trans-4-hydroxystilbene (Fig. 1B)
Trans-4-hydroxystilbene concentration-dependently
on I(
À
varied with the test pulse-duration, i.e., current
Cl ,OR)
inhibited I(
À
in rat peritoneal mast cells (IC : 3.2
50
inhibition was minimal at the beginning and increased up to
a plateau during the course of a 100 ms-test pulse. With
Cl ,OR)
AM). When applied at a concentration of 3 AM, the
inhibitory effect of trans-4-hydroxystilbene was partly
reversible (n = 3); at a concentration of 10 AM (n = 9)
DIDS, a complete inhibition of I(
À
was achieved at 10
Cl ,OR)
AM (n = 15; IC50 1.4 AM). After short periods of incubation
with DIDS, its inhibitory effect was partly reversible,
whereas the inhibitory effect of SITS could be fully reversed
even after minutes of incubation when the superfusion
medium was changed to a SITS-free buffer solution. SITS
suppression of I(
À
could not be reversed, even by
Cl ,OR)
extensive superfusion with stilbene-free buffer solution.
3
.1.3. Trans-4-hydroxy-4V-methoxystilbene (Fig. 1C)
In comparison with trans-4-hydroxystilbene, trans-4-hy-
was less potent in inhibiting I(
À
(IC : 9.2 AM; n = 3).
50
Cl ,OR)
droxy-4V-methoxystilbene is characterised by an additional
methoxy substituent at position 4Vof the other phenyl ring
systems. At a concentration of 10 AM, trans-4-hydroxy-4V-
3.1.6. Trans-stilbene derivatives substituted at their vinyl
group
methoxystilbene induced a reduction of I(
À
elicited by
Cl ,OR)
standard test pulses ( F 80 mV) by about 8.7 F 8.3 [(% of
control); n = 3]. This effect was reversible.
3.1.6.1. Trans-diethylstilbene (Fig. 1H). Superfusion of
rat peritoneal mast cells with 10 AM of trans-diethylstil-
bene effectively reduced steady state I(
À
elicited by
Cl ,OR)
3
.1.4. Resveratrol (Fig. 1D)
Resveratrol represents a phytostilbene with two differ-
ently hydroxylated aromatic rings. At a concentration of 10
test potentials of F 80 mV to 81.8 F 4.3 [(% of control);
IC : 4.1 AM; n = 3]. Upon the start of superfusion, trans-
5
0
diethylstilbene-mediated inhibition of I(
À
occurred
Cl ,OR)
Fig. 1. Depiction of structural formulas of the test compounds used in the present study. Left column (trans-stilbene derivatives substituted at the aromatic ring
system): (A) trans-stilbene, (B) trans-4-hydroxystilbene, (C) trans-4-hydroxy-4V-methoxystilbene, (D) resveratrol, (E) DIDS, (F) SITS; middle column
(
stilbene derivatives substituted at their ethene moiety): (G) cis-diethylstilbene, (H) trans-diethylstilbene, (I) diethylstilbestrol, (J) hexestrol; right column
(
triphenylethene derivatives): (K) clomiphene, (L) tamoxifen, (M) chlorotrianisene.

T. Roloff et al. / European Journal of Pharmacology 478 (2003) 81–88
85
after a delay of 1–2 min. The suppression of I(
À
by
Superfusion of rat peritoneal mast cells with hexestrol led
À
Cl ,OR)
trans-diethylstilbene could not be reversed by superfusion
with stilbene-free buffer solution. Cis-diethylstilbene (Fig.
to rapid, potent, and partially reversible inhibition of I(
(IC : 1.1 AM; n = 3; Fig. 2).
Cl ,OR)
5
0
1
G) at a concentration of 10 AM was completely ineffective
data not shown).
(
3.1.9. Triphenylethene derivatives
3
.1.7. Diethylstilbestrol (Fig. 1I)
In comparison with trans-diethylstilbene, the estrogen
3.1.9.1. Tamoxifen (Fig. 1K). Like trans-diethylstilbene,
the triphenylethene derivative tamoxifen can be regarded as a
stilbene substituted at the ethene bridge (see Fig. 1K).
receptor agonist diethylstilbestrol possesses two additional
hydroxy groups at positions 4 and 4V of both phenyl
rings. Superfusion of mast cells with diethylstilbestrol led
Tamoxifen concentration-dependently inhibited the I(
À
Cl ,OR)
in rat peritoneal mast cells (IC : 3.5 AM; n = 3). At a
5
0
to a rapid and potent suppression of I(
concentration of 10 AM, a complete inhibition of I(
À
. At a
À
concentration of 10 AM, tamoxifen effectively inhibited
Cl ,OR)
I
of I was only partially reversible. The onset of current
À
(Cl ,OR)
by about 89 F 6.2 [(% of control); n = 3]. Inhibition
Cl ,OR)
could be observed (IC : 1.6 AM; n = 3). Suppression of
À
5
0
(Cl ,OR)
I(
À
achieved by diethylstilbestrol at a concentration
inhibition after start of the tamoxifen superfusion occurred
with a delay of about 1–2 min.
Cl ,OR)
of 2 AM was completely reversible.
3
.1.8. Hexestrol (Fig. 1J)
The only structural difference between hexestrol and
3.1.10. Clomiphene (Fig. 1L)
Despite being a stilbene substituted at the ethene bridge,
clomiphene also exhibits a modification at one of its two
phenyl rings. Similarly to tamoxifen, upon superfusion, the
diethylstilbestrol is the ethane instead of the ethene bridge
between the phenyl nuclei. Thus, hexestrol does not belong
to the stilbenes.
onset of the inhibitory effect of clomiphene on I(
À
Cl ,OR)
Fig. 2. Representative whole-cell, patch-clamp experiment (nystatin perforated patch), showing the time-dependent development of an outwardly rectifying
current in rat peritoneal mast cell challenged by step hyper- and depolarisations and its partly reversible block by 3 AM hexestrol. (A) Time course (holding
potential 0 mV); test potentials (100 ms) + 80 mV (solid circles), À 80 mV (open circles). (B–D) Original current traces from step protocol ( À 80 F 80 mV, in
20 mV increments), (B) before addition of 3 AM hexestrol; (C) at the start, and (D) end of hexestrol wash-out; the corresponding current–voltage relationships
are represented in (E).

86
T. Roloff et al. / European Journal of Pharmacology 478 (2003) 81–88
occurred with a delay of about 1–2 min. Clomiphene
concentration-dependently inhibited I( (IC : 5.1
incubating rat peritoneal mast cells for 60 min with 1–
À
300 AM DIDS (IC : 96 AM) or SITS (IC : 72 AM), a
Cl ,OR)
50
50
50
AM; n = 3); at a concentration of 10 AM, the inhibitory effect
amounted to 85.2 F 8.9 [% of control].
complete inhibition of the NGF/lysophosphatidylserine-
induced exocytotic response could be achieved (n = 3).
Under identical experimental conditions, DIDS and SITS
also reduced the exocytotic response induced by masto-
paran, but with a c 10 fold lower potency (data not
shown).
3
.1.11. Chlorotrianisene (Fig. 1M)
In comparison with tamoxifen and clomiphene, the more
hydrophilic triphenylethene derivative chlorotrianisene, at
the maximal test concentration of 10 AM (n = 3), only
Tamoxifen and clomiphene (10 AM) partially reduced the
NGF/lysophosphatidylserine induced exocytotic response in
rat peritoneal mast cells in a time-dependent manner. After a
maximal incubation time of 60 min, exocytosis was reduced
minimally inhibited I(
control), n = 3].
À
by about 4.6 F 12.6 [(% of
Cl ,OR)
3
3
.1.12. b-Estradiol (Fig. 1N)
At a concentration of 10 AM (n = 3), the steroid estrogen
receptor agonist 17h-estradiol inhibited I( by about
by about 55.5 F 1.6 and 17.1 F 3.8 ([ H]5-HT release [% of
control]; n = 3), respectively.
À
Cl ,OR)
2
1.8 F 8.3 [(% of control), n = 3].
4. Discussion
3
3
.2. Secretion studies
The stilbene derivative diethylstilbestrol has previously
been shown to act as a potent inhibitor of the outwardly
rectifying chloride current (I( ) in rat peritoneal mast
3
.2.1. Spontaneous [ H]5-HT release
Incubation with either 10 AM tamoxifen or clomiphene
increased the spontaneous release of [ H]5-HT from
À
Cl ,OR)
3
cells (Roloff et al., 2001). To characterise structural prop-
erties enabling diethylstilbestrol to inhibit I( , differ-
3
[
H]5-HT-loaded rat peritoneal mast cells in a time-
dependent manner; after a maximum incubation time of
0 min, the respective values amounted to: tamoxifen/
control 23.4 F 8.0/12.7 F 8.1 (n = 3), clomiphene/control
À
Cl ,OR)
ent diethylstilbestrol-related stilbene and non-stilbene
derivatives were tested in the present study in rat perito-
neal mast cells. In comparison with diethylstilbestrol, the
unsubstituted stilbene, trans-stilbene, did not inhibit chlo-
ride currents in concentrations up to 10 AM. Thus, the
structural moiety discerning diethylstilbestrol from trans-
stilbene must be of importance. Diethylstilbestrol is sub-
stituted at the aromatic phenyl rings by two hydroxy
groups in positions 4 and 4V, and, at the aliphatic ethene
bridge, by two ethyl groups. Accordingly, in order to
analyse the structure–activity relationships of diethylstil-
bestrol different stilbene derivatives, either modified at
their aromatic and/or their aliphatic part, were tested in
the present study. In addition, the non-stilbene molecule
hexestrol was assessed.
6
3
17.0 F 3.8/3.2 F 1.0 (n = 3) ([ H]5-HT release [% of to-
tal]). The other stilbene derivatives studied in the present
investigation did not increase the spontaneous release of
3
[
H]5-HT, when present at a concentration of 10 AM for
up to 60 min (data not shown).
3
.3. Agonist-induced [ H]5-HT-release (Table 1)
3
The ability of different stilbene derivatives to impair
the NGF/lysophosphatidylserine- or mastoparan-induced
3
3
exocytotic release of [ H]5-HT from [ H]5-HT-loaded rat
peritoneal mast cells was tested in a series of time course
experiments (incubation time 1–60 min). Mastoparan was
Hexestrol differs from diethylstilbestrol only by lacking
the double-bond within its aliphatic part. Interestingly,
À 5
always used at a final concentration of 2 Â 10
M; the
final concentrations for NGF and lysophosphatidylserine
amounted to 1 Ag/ml and 1 AM, respectively. At a test
concentration of 10 AM, the following compounds did not
hexestrol inhibited I(
À
as potent as diethylstilbestrol
Cl ,OR)
(IC -value 1.1 vs. 1.6 AM). Thus, the conjugated double
5
0
bond within the aliphatic part of the trans-stilbene scaffold
of the diethylstilbestrol molecule probably is not essential
3
inhibit the agonist-induced [ H]5-HT release: trans-stil-
bene, trans-diethylstilbene, trans-4-hydroxystilbene, trans-
for inhibition of I(
À
.
Cl ,OR)
4
incubation of cells for 5 min, which is a maximal-
-hydroxy-4V-methoxystilbene, and chlorotrianisene. After
Diethylstilbene differs from diethylstilbestrol by the
absence of the hydroxy groups at both phenyl rings. Cis-
diethylstilbene at a concentration of 10 AM showed no effect
effective time-period, diethylstilbestrol reduced the NGF/
lysophosphatidylserine-, but not the mastoparan-induced
on I(
À
(data not shown). In contrast, trans-diethylstil-
Cl ,OR)
3
[
H]5-HT release by 29.2 F 5.3 [(% of control); n = 3].
bene inhibited the I(
À
nearly as potent as diethylstil-
Cl ,OR)
Agonist-induced exocytotic responses were not impaired
by incubation of cells with 10 AM DIDS or SITS for up
to 30 min. After an incubation period of 60 min, a
reduction of the NGF/lysophosphatidylserine-induced re-
bestrol (IC : 4.1 vs. 1.6 AM). Consequently, the presence of
5
0
this special phenolic moiety seems not to be essential for
diethylstilbestrol mediated inhibition of I( . In accor-
À
Cl ,OR)
dance with these findings, trans-4-hydroxystilbene, charac-
terized by only a single hydroxy group at position 4 potently
inhibited the chloride current (IC : 3.3 AM).
3
lease of [ H]5-HT by DIDS (43.8 F 9.3) and SITS
[
31.9 F 7.7 (% of control); n = 3] became apparent. By
5
0

T. Roloff et al. / European Journal of Pharmacology 478 (2003) 81–88
87
Stilbene derivatives modified at positions 4 and/or 4V
with methoxy substituents, as in the case of trans-4-
hydroxy-4V-methoxystilbene or chlorotrianisene, did not
The data received for inhibition of I(
À
with different
Cl ,OR)
estrogen receptor agonists (e.g. 17h-estradiol, diethylstil-
bestrol) or antagonists (e.g. tamoxifen) did not correlate
with their established ability to affect estrogen receptors
(Riggs and Hartmann, 2003). Therefore, an involvement of
estrogen receptors appears unlikely.
inhibit I(
À
. This finding indicates that methoxy-sub-
Cl ,OR)
stituents, in contrast to hydroxy groups, might sterically
prevent an interaction of these molecules with a specific
binding site at the ORCC-protein. But also other mecha-
nisms could underlie this failure of action (see below).
Resveratrol differs from trans-4-hydroxystilbene by two
hydroxyl groups localised at positions 3Vand 5Vof the phenyl
ring and by the absence of an OH-group at position 4V. In
comparison with trans-4-hydroxystilbene, resveratrol is
more hydrophilic (Log P 3.08 vs. 4.04). In concentrations
up to 10 AM, resveratrol exerted no inhibitory effect on
The profile of stilbene derivatives capable to inhibit
I(
À
corresponds with previous structure–activity data
Cl ,OR)
regarding diethylstilbestrol-induced inhibition of certain
ATPases. Future studies will reveal, if structural homologies
between these ATPases and the mast cell ORCC-protein
may be the reason for this analogy (Martinez-Azorin et al.,
1992).
In summary, the aim of the present study was to assess the
importance of different structural features of the diethylstil-
bestrol molecule for inhibition of an outwardly rectifying
I(
À
. One possible explanation for this observation may
Cl ,OR)
be that the hydroxy groups at positions 3V and 5V of the
phenyl ring for steric reasons do not allow an interaction of
resveratrol with a specific binding site at the ORCC-protein.
Alternatively, the hydrophilicity of resveratrol may cause
insufficient insertion into the mast cell plasma membrane,
thus preventing the molecule from reaching its target site at
chloride current I(
À
in rat peritoneal mast cells. Dieth-
Cl ,OR)
ylstilbestrol represents a modified stilbene derivative. Be-
cause trans-stilbene was ineffective to inhibit I( , the
À
Cl ,OR)
chemical substituents discerning diethylstilbestrol from the
simple trans-stilbene must be of importance. An analysis of
action of different test compounds revealed that a hydroxy
group at position 4 or 4Vof the phenyl rings of diethylstil-
À
the Cl channel. The finding that DIDS and SITS, two other
hydrophilic stilbenes, potently inhibit I(
argument against this hypothesis since DIDS and SITS, as
À
is not an
Cl ,OR)
bestrol represents a prerequisite for inhibition of I(
À
. In
Cl ,OR)
shown before in previous studies, inhibit I(
À
by plug-
ging the pore of the outwardly rectifying Cl channel
contrast, trans-stilbene molecules modified at position 4 or 4V
by methoxy substituents were incapable to inhibit I(
Cl ,OR)
À
À
.
Cl ,OR)
(
mechanism of action is due to the fact that DIDS and SITS,
Dietrich and Lindau, 1994; Lewis et al., 1993). This
Additionally, the absence of the double bond within the
aliphatic part of the stilbene structure as well as a modifica-
tion of the aliphatic part by certain substituents did not
À
like Cl , are anions and capable to enter the pore of the ion
channel. Moreover, DIDS and SITS have been shown to
impair the ability of molecules to inhibit I(
peritoneal mast cells. After identification of the gene coding
À
in rat
Cl ,OR)
inhibit I(
À
via allosteric mechanisms. Due to this
Cl ,OR)
À
complexity of action, data obtained with DIDS and SITS
are not suitable for interpreting the structure–activity rela-
tionships of diethylstilbestrol.
for the rat peritoneal mast cell outwardly rectifying Cl
channel protein, the structure–activity data generated in the
present study are hoped to facilitate the development of
À
A modification of the aliphatic part of trans-stilbene
with large aromatic substituents, as in the case of the
triphenyl derivatives tamoxifen and clomiphene, does not
optimised outwardly rectifying Cl channel-inhibitors.
impair the ability of these molecules to inhibit I(
À
. The
Cl ,OR)
References
putative reason for the ineffectiveness of chlorotrianisene,
another triphenyl derivative, has been described above.
In the present study, with the exception of DIDS and
SITS, none of the compounds capable to inhibit the
Dietrich, J., Lindau, M., 1994. Chloride channels in mast cells: block by
DIDS and role in exocytosis. J. Gen. Physiol. 104, 1099–1111.
Friedrich, K., Henning, H., 1959. Darstellung stilbenartiger Verbindungen
nach der Methode von G. Wittig. Chem. Ber. 92, 2944–2952.
Hamill, O.P., Marty, A., Neher, E., Sakmann, B., Sigworth, F.J., 1981.
Improved patch-clamp techniques for high-resolution current record-
ing from cells and cell-free membrane patches. Pflugers Arch. 391,
85–100.
I(
À
impaired the process of agonist-induced exocyto-
Cl ,OR)
sis. Exocytosis involves a highly regulated fusion of
secretory granules with the plasma membrane (Pinxteren
et al., 2000). Assumingly, compounds influencing the
biophysical properties of plasma membranes (e.g. mem-
brane fluidity, etc.) would also alter the process of agonist-
induced exocytosis. According to this assumption, the
inability of the stilbene derivatives to impair exocytosis
argues against an unspecific alteration of membrane prop-
erties as a putative mechanism for inhibition of chloride
currents. Complementary, this result favours a more spe-
cific mechanism of action, which probably may involve the
interaction with a specific binding site at the ORCC-
protein.
Hill, P.B., Martin, R.J., Miller, H.R., 1996. Characterization of whole-cell
currents in mucosal and connective tissue rat mast cells using ampho-
tericin-B-perforated patches and temperature control. Pflugers Arch.
4
32, 986–994.
Horn, R., Marty, A., 1988. Muscarinic activation of ionic currents meas-
ured by a new whole-cell recording method. J. Gen. Physiol. 92,
1
45–159.
Kulka, M., Schwingshackl, A., Befus, A.D., 2002. Mast cells express
chloride channels of the ClC family. Inflamm. Res. 51, 451–456.
Leimner, J., Weyerstahl, P., 1982. Stereochemie und Nebenprodukte der
relativen Kopplung von Arylketonen zu 1,2-Dialkyl-1,2-diarylethenen.
Chem. Ber. 115, 3697–3705.

88
T. Roloff et al. / European Journal of Pharmacology 478 (2003) 81–88
Lewis, R.S., Ross, P.E., Cahalan, M.D., 1993. Chloride channels activated
by osmotic stress in T lymphocytes. J. Gen. Physiol. 101, 801–826.
Martinez-Azorin, F., Teruel, J.A., Fernandez-Belda, F., Gomez-Fernandez,
J.C., 1992. Effect of diethylstilbestrol and related compounds on the
Penner, R., Matthews, G., Neher, E., 1988. Regulation of calcium influx by
second messengers in rat mast cells. Nature 334, 499–504.
Pinxteren, J.A., O’Sullivan, A.J., Larbi, K.Y., Tatham, P.E., Gomperts,
2
+
B.D., 2000. Thirty years of stimulus-secretion coupling: from Ca to
GTP in the regulation of exocytosis. Biochimie 82, 385–393.
Riggs, B.L., Hartmann, L.C., 2003. Selective estrogen-receptor modula-
tors—mechanisms of action and application to clinical practice. N.
Engl. J. Med. 348, 618–629.
(2+)
Ca -transporting ATPase of sarcoplasmic reticulum. J. Biol. Chem.
67, 11923–11929.
2
Matthews, G., Neher, E., Penner, R., 1989. Chloride conductance activated
by external agonists and internal messengers in rat peritoneal mast cells.
J. Physiol. 418, 131–144.
Roloff, T., Wordehoff, N., Ziegler, A., Seebeck, J., 2001. Evidence against
À
Mousli, M., Bronner, C., Bueb, J.L., Tschirhart, E., Gies, J.P., Landry, Y.,
the functional involvement of outwardly rectifying Cl channels in
1
989. Activation of rat peritoneal mast cells by substance P and mas-
agonist-induced mast cell exocytosis. Eur. J. Pharmacol. 431, 1–9.
Vliagoftis, H., Dimitriadou, V., Boucher, W., Rozniecki, J.J., Correia, I.,
Raam, S., Theoharides, T.C., 1992. Estradiol augments while tamox-
ifen inhibits rat mast cell secretion. Int. Arch. Allergy Immunol. 98,
398–409.
toparan. J. Pharmacol. Exp. Ther. 250, 329–335.
Pearce, F.L., Thompson, H.L., 1986. Some characteristics of histamine
secretion from rat peritoneal mast cells stimulated with nerve growth
factor. J. Physiol. 372, 379–393.