Department of Anesthesiology (S.M.T., M.P., Y.M.N.,C.J.L.),
Department of Molecular Biology and Pharmacology (K.R.N., M.H.,
D.F.C.), and
Department of Psychiatry (C.F.Z.), Washington University
School of Medicine, St. Louis, Missouri 63110
A number of steroids seem to have anesthetic effects resulting
primarily from their ability to potentiate currents gated by 
-aminobutyric acidA (GABAA) receptor
activation. One such compound is
(3
,5,17
)-3-hydroxyandrostane-17-carbonitrile [(+)-ACN]. We
were interested in whether carbonitrile substitution at other ring
positions might result in other pharmacological consequences. Here we
examine effects of (3,5,17)-17-hydroxyestrane-3-carbonitrile [(+)-ECN] on GABAA receptors and Ca2+
channels. In contrast to (+)-ACN, (+)-ECN does not potentiate GABAA-receptor activated currents, nor does it directly
gate GABAA-receptor mediated currents. However, both
steroids produce an enantioselective reduction of T-type current.
(+)-ECN blocked T current with an IC50 value of 0.3 µM with a maximal block of 41%. (+)-ACN produced a
partial block of T current (44% maximal block) with an
IC50 value of 0.4 µM. Block of T current
showed mild use- and voltage-dependence. The (
)-ECN enantiomer was
about 33 times less potent than (+)-ECN, with an IC50 value
of 10 µM and an amount of maximal block comparable to
(+)-ECN. (+)-ECN was less effective at blocking high-voltage-activated Ca2+ current in DRG neurons (IC50 value of 9.3 µM with maximal block of about 27%) and hippocampal
neurons. (+)-ECN (10 µM) had minimal effects on
voltage-gated sodium and potassium currents in rat chromaffin cells.
The results identify a steroid with no effects on GABAA
receptors that produces a partial inhibition of T-type Ca2+
current with reasonably high affinity and selectivity. Further study of
steroid actions on T currents may lead to even more selective and
potent agents.
 |
Introduction |
Because
low-voltage-activated, or T-type, calcium (Ca2+)
currents are activated at potentials as negative as 60 mV, they are thought to play a key role in the initiation of regenerative
depolarizing inward current (reviewed by Huguenard, 1996
). The
properties of T currents and their distribution in particular cell
types suggest a critical role in the regulation of excitability in both
neurons (Llinas, 1988; Huguenard and Prince, 1992) and other excitable cells (Matteson and Armstrong, 1984; Hirano et al., 1989). T
currents have also been proposed to contribute to initiation of seizure activity in thalamic neurons (Huguenard and Prince, 1994; Tsakiridou et al., 1995). Thus, physiological regulation of T-type
current is likely to be of profound significance to the regulation of neuronal activity.
In contrast to the abundance of peptide toxins that have proven
useful in identifying the physiological roles of HVA variants of
Ca2+ current (review by DeWaard et
al., 1996), there is an absence of highly potent and selective
antagonists for T-type channels. Except for recent reports of the
T-current blocking effects of mibefradil (Mishra and Hermsmeyer, 1994),
a compound which also affects HVA types of Ca2+
current at somewhat higher concentrations (Bezprozvanny and Tsien, 1995), most other T current blockers are of relatively weak potency and
selectivity. However, T-type currents are blocked at high concentrations by a variety of compounds within concentrations that are
perhaps clinically relevant. This includes anesthetics (Herrington
et al., 1991; Study, 1994; Todorovic and Lingle,
1998) and also some anticonvulsants [e.g., succinimides (Coulter
et al., 1989a, 1989b; but see Leresche et al.,
1998) and phenytoin (Todorovic and Lingle, 1998)]. Selective blockers
of T-currents are therefore likely to have interesting and important
consequences on neuronal excitability and may be of unique clinical use.
The modulation of ion channel function by steroids is an area of
increasing interest. Much of this interest has been focused on the
modulation of GABAA receptor function by steroids
(reviewed by Lambert et al., 1995). With regard to
Ca2+ channel modulation, some steroids have been
reported to modulate Ca2+ currents through
G-protein-mediated pathways (ffrench-Mullen et al., 1994).
Additionally, we have shown that (+)-ACN (Fig. 1), a steroid that powerfully potentiates
and gates GABAA receptors (Wittmer et
al., 1996), exerts somewhat selective, direct blocking effects on
particular components of HVA Ca2+ currents
(Nakashima et al., 1998). As part of ongoing
structure-activity studies, we are examining other steroids for effects
on Ca2+ currents. Here we describe the effects of
(+)-ECN. (+)-ECN is a 5-reduced steroid without a C-19 methyl group
(a 19-norsteroid). Relative to (+)-ACN, the ring positions of the
carbonitrile and hydroxyl groups are reversed. The stereochemical
relationship between these two groups is also different.
The main finding of this study is that (+)-ECN (Fig. 1), which has no
effect on GABA receptors at 10 µM, is a potent,
enantioselective, partial blocker of T-type current in rat DRG sensory
neurons. Furthermore, we compare the blocking effects of (+)-ECN to the action of two other steroids that exhibit anesthetic effects, (+)-ACN
and alphaxalone. Although all three compounds share similarities in
their effects on T type currents, (+)-ECN is unique in lacking any
effect on GABAA receptors. Steroid analogues that
exhibit relatively selective, potent, and reversible effects on T
currents, without any effects on GABA receptors, may provide useful
tools for examining the role of T currents in neuronal excitability and
aid the potential development of compounds that may mediate anesthetic,
analgesic, or anticonvulsant effects.
| |
Materials and Methods |
Preparation of cells.
Acutely dissociated DRG neurons from
adult male Sprague-Dawley rats (100-300 g) were obtained using
enzymatic treatment as described elsewhere (Todorovic and Lingle,
1998). Glass coverslips with adherent DRG cells were transferred to a
standard culture dish with a total volume <1 ml. Most results from DRG
neurons were obtained from smaller diameter cells with no visible
processes. Average uncompensated
Rs was 6.6 ± 2.5 M
(mean ± standard deviation) and average
Cm was 13.5 ± 4 pF for 217 neurons.
Microisland cultures of neonatal rat hippocampal neurons were prepared
as described previously (Mennerick et al., 1995). Cells used
for recordings of HVA Ca2+ currents were used
after 2-5 days in culture. Chromaffin cells were prepared from adult
rat adrenal glands as described elsewhere (e.g., Solaro et
al., 1995).
Electrophysiological methods, solution application, and current
isolation procedures.
Currents were recorded using standard
whole-cell patch-clamp methods (Hamill et al., 1981).
Solutions were applied to cells through multiple independently
controlled glass capillary tubes, and solution was removed from the
other end of the chamber with the use of constant suction. Solution
application was accomplished by manually controlled valves. Test
solutions were maintained in all-glass syringes and allowed to fall by
gravity. Changes in Ca2+ current amplitude in
response to rapidly acting drugs or ionic changes were typically
complete in 10-20 sec. Switching between separate perfusion syringes,
each containing control saline, resulted in no changes in
Ca2+ current. For all steroids examined here, no
dependence on the order of presentation or desensitization with
repeated applications was observed.
The intracellular saline for recording of T current consisted of:
135-140 mM tetramethylammonium hydroxide, 10 mM EGTA, 40 mM HEPES, and 2 mM
MgCl2. The intracellular saline was usually titrated to pH 7.15-7.20 with HF, although in some experiments HCl or
methanesulfonic acid was used. HVA currents were blocked by procedures
described previously (Todorovic and Lingle, 1998). Specifically,
experiments on T currents were done on smaller DRG neurons that express
L- and N-type HVA current almost exclusively (Scroggs and Fox, 1992).
Thus, in most experiments, the addition of F
to
the intracellular solution was used to abolish L-type HVA current as
described previously (Herrington and Lingle, 1992; Todorovic and
Lingle, 1998). In addition, such cells were preincubated with 1 µM GVIA to abolish N-type HVA current. For generation of concentration-response curves, T currents were elicited by voltage steps to 30 mV from a holding potential of 90 mV. This resulted in
T current with minimal HVA current contamination (e.g., Todorovic and
Lingle, 1998).
For recording of HVA currents, cells were held at 60 mV and inward
currents were elicited by a test step to 10 mV. For HVA currents, the
intracellular solution contained: 110 mM Cs-methane sulfonate, 14 mM phosphocreatine, 10 mM HEPES,
9 mM EGTA, 5 mM Mg-ATP, and 0.3 mM
Tris-GTP, pH adjusted to 7.15-7.20 with CsOH (standard osmolarity: 300 mOsM). To verify that the composition of the intracellular
solution did not influence the sensitivity of T currents to steroid
action, in some experiments, the internal saline used for recording HVA
currents was also used for recording of T current. In such cases, to
isolate T current, HVA current was blocked by preincubation of cells
with 1 µM GVIA, and by also including 2 µM
MVIIC and 5 µM nifedipine in the external solution, to
block N-, P-, Q- and L-types of HVA current, respectively. The blocking
effects of steroids on T current were identical with all of the
procedures used to isolate T current, regardless of whether the
intracellular anion was F, methanesulfonic
acid, or Cl.
The standard extracellular saline for recording of T-type
Ca2+ currents contained: 152 mM
tetraethylammonium-Cl, 10 mM HEPES, and 10 mM
BaCl2, adjusted to pH 7.4 with
tetraethylammonium-OH, osmolarity 316 mOsM. For recording
of HVA Ca2+ currents in DRG neurons, a 5 mM Ba2+ solution was used. Recordings
of HVA Ba2+ current in cultured hippocampal
neurons followed procedures described previously (Nakashima et
al., 1998).
Recordings of GABA currents on cultured hippocampal neurons were done
as described previously (Mennerick et al., 1995; Wittmer et al., 1996). The extracellular recording solution
contained: 140 mM NaCl, 4 mM KCl, 2 mM MgCl2, 2 mM
CaCl2, 10 mM HEPES, pH 7.3. Recording
pipettes were filled with a solution containing: 140 mM
CsCl, 4 mM NaCl, 5 mM EGTA, 0.5 mM
CaCl2, 4 mM
MgCl2, and 10 mM HEPES, pH 7.3. In
studies examining autaptic currents, CsCl was replaced by KCl
and MgCl2 was replaced with 2 mM
Mg-ATP and 0.5 mM Na-GTP in the intracellular solution.
GABA and steroids were applied for 500 msec using a pressure (20 p.s.i.
air) ejection drug delivery system with a patch pipette positioned
approximately 5 µm from the neuron. The concentrations reported here
are those in the pipette and are an upper limit for the concentrations
reaching the cell. Autaptic responses were evoked from a holding
potential of 70 mV using 1.5-msec voltage steps to +20 mV applied
every 30 sec.
To record Na+ and K+
currents from chromaffin cells, the external saline contained: 140 mM NaCl, 5.4 mM KCl, 10 mM HEPES,
1.8 mM CaCl2 and 2.0 mM
MgCl2 titrated to pH 7.4 with
N-methylglucamine. For recording of
Na+ currents, the internal saline was identical
to the one for HVA Ca2+ currents used for DRG
neurons. The internal saline for recording of K+
currents contained: 140 mM KCl, 20 mM KOH, 10 mM HEPES (H+), 5 mM
HEDTA with added CaCl2 to make 10 µM [Ca2+]i
as defined by the EGTAEC program (E. McCleskey, Vollum Institute, Portland, OR).
Analysis of steroid effects on current amplitude and
properties.
The percent reduction in peak inward current carried
by Ba2+ ions at a given steroid concentration was
used to generate concentration-response curves. For each of these
curves, all points are averages of multiple determinations obtained
from at least five different cells. Only cells where at least two
different concentrations of the same steroid were tested were used to
construct concentration-response curves. Because the level of maximal
block by different steroids was somewhat variable from cell to cell,
only those cells in which a drug concentration producing a near maximal
effect was tested were taken for analysis. On all
concentration-response curves, vertical bars indicate standard errors.
Mean values on all concentration-response curves were fit to the
following function:
![PB([<UP>Steroid</UP>])=PB<SUB><UP>max</UP></SUB>/(1+(<UP>IC</UP><SUB>50</SUB>/[<UP>Steroid</UP>])<SUP>n</SUP>)](/content/vol54/issue5/fulltext/918/img001.gif)
|
(1)
|
where PBmax is the maximal percent
block of peak T current, IC50 is the
concentration that produces 50% of maximal inhibition, and
n is the apparent Hill coefficient for blockade. Fitted
values are reported with 95% linear confidence limits. The
voltage-dependence of steady state inactivation was described with
Boltzmann distribution:
![<UP>I</UP>(<UP>V</UP>)=<UP>I<SUB>max</SUB></UP>/(1+<UP>exp</UP>[<UP>−</UP>(<UP>V</UP>−<UP>V</UP><SUB>0.5</SUB>)/k])](/content/vol54/issue5/fulltext/918/img002.gif)
|
(2)
|
where Imax represents maximal activatable
current, V0.5 represents the voltage where half
of the current is inactivated, and k (units of millivolts)
represents the voltage dependence of the distribution.
Effects of (+)-ECN on current activation and inactivation time
constants (
m and h,
respectively) were determined from the fit of the following form of a
Hodgkin and Huxley (1952) current activation equation:
![<UP>I</UP>(t)=A ∗ [1−<UP>exp</UP>(<UP>−</UP>t/&tgr;<SUB>m</SUB>)]<SUP>n</SUP> ∗ <UP>exp</UP>(<UP>−</UP>t/&tgr;<SUB>h</SUB>)](/content/vol54/issue5/fulltext/918/img003.gif)
|
(3)
|
where I(t) is the current (I) as a function of time
(t), A is the maximal activatable current,
m is the activation time constant,
h is the inactivation time constant, and
n is a term for the sigmoidicity in the activation process. For T current at potentials more positive than 20 mV,
n was typically constrained to 1.0. In the absence of such
constraint, for currents between 60 and 0 mV, n varied
from about 1.8 to about 1.0.
Drugs and chemicals.
The synthesis of (+)-ECN and ()-ECN
will be described elsewhere. The compounds had spectroscopic data (IR,
NMR) consistent with the assigned structures and were shown to have the
correct elemental composition by combustion analysis for C, H, and N. The preparation of (+)-ACN (Hu et al., 1993) and ()-ACN
(Hu et al., 1997) have been described previously.
Alphaxalone was obtained from Sigma (St. Louis, MO). All steroids were
dissolved in DMSO to make 10-30 mM stock solutions.
Aliquots of the stock solutions were added to the standard external
solution to make the final concentrations given in the text. The final
concentration of DMSO was less than 0.6% in these experiments; this
concentration of DMSO did not affect IBa (data
not shown, n = 5 cells) or
GABAA currents. At steroid concentrations of 60 µM or higher, the steroids seem to crystallize out of
solution, thus reducing the effective steroid concentration in
solution. This has prevented us from using higher concentrations of
steroids in experiments where maximal blockade could not be determined reliably.
GVIA (RBI, Natick, MA; Sigma) and MVIIC (RBI; Sigma) were dissolved in
distilled water to make stock solutions of 0.5, 0.5 and 0.2 mM, respectively. Nifedipine (RBI) was dissolved in DMSO at
5 mM as a stock solution. GDPS and GTPS were obtained
from Sigma and, when used, replaced GTP in the pipette solution.
| |
Results |
(+)-ECN does not potentiate GABA currents in rat hippocampal
neurons.
Fig. 1 displays the structures of the various steroids
used in this investigation. A feature of many neuroactive steroids is
their ability to potentiate GABAA-receptor
mediated Cl currents (Lambert et
al., 1995; Wittmer et al., 1996). Both (+)-ACN (Wittmer
et al., 1996) and alphaxalone (Sear, 1996) are anesthetic steroids whose effects are thought to be mediated by
GABAA-receptor potentiation.
The effects of (+)-ECN on GABAA-receptor-mediated
currents were examined in cultured neonatal hippocampal neurons grown
in microisland cultures (Fig. 2A).
(+)-ECN (10 µM) was totally without effect on currents
activated by 2 µM GABA. For comparison, 10 µM (+)-ACN produces a large potentiation of currents
activated by 2 µM GABA (Fig. 2B; Wittmer et
al., 1996). ()-ECN does not produce any significant potentiation
of GABAA-mediated currents at 10 µM
(not shown), whereas ()-ACN produces some potentiation (Wittmer
et al., 1996).
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 2.
A, Traces show currents activated by
application of 2 µM GABA to a hippocampal neuron either
with or without 10 µM (+)-ECN. B, Traces
show currents activated by 2 µM GABA, but in the presence
and absence of 10 µM (+)-ACN. Neurons shown in A and B
were held at 60 mV. C, Stimulation of a hippocampal neuron grown in
microisland culture resulted in an inhibitory autaptic current. (+)-ECN
(10 µM) had no effect on the amplitude of the evoked
inhibitory current. D, Stimulation of a hippocampal neuron resulted in
an excitatory autaptic current. (+)-ECN (10 µM) had no
effect. In C and D, stimulus artifacts have been truncated for
clarity.
|
|
To ascertain other potential targets of (+)-ECN action, we also
examined the effects of 10 µM (+)-ECN on autaptically
evoked synaptic currents in the hippocampal microisland cultures
(Mennerick et al., 1995). (+)-ECN (10 µM) had
no effect on either GABA-mediated inhibitory synaptic currents (Fig.
2C; two experiments) or glutamate-mediated excitatory synaptic currents
(Fig. 2D; four experiments).
(+)-ECN inhibits T-type Ca2+ current in rat DRG
neurons.
T-type Ca2+ currents were isolated
as described in Materials and Methods and typically monitored with
voltage-steps to 30 mV from a holding potential of 90 mV. (+)-ECN
reversibly depressed the amplitude of T current as seen in Fig.
3A without apparent effects on current
activation or inactivation kinetics. Blockade by (+)-ECN was
concentration-dependent (Fig. 3B) from 0.1 to 10 µM. In
most cells, blockade by 30 µM (+)-ECN was
indistinguishable from blockade produced by 10 µM
(+)-ECN. The percent block of peak T current by 10 µM
(+)-ECN was 41.6 ± 10.7% for 36 cells. The blocking effect was
strongly enantioselective, as shown in Fig. 3C, where 10 µM (+)-ECN blocked about 2-fold more T-type current than
the same concentration of ()-ECN. No apparent desensitization was
obvious when cells were exposed sequentially to the same concentration of steroid. From such experiments, concentration-response curves for
both agents were generated as depicted in Fig. 3D. In each cell,
responses to any application of steroid were normalized to blockade
produced by 10 µM (+)-ECN. For (+)-ECN, the
IC50 for blockade of T-type
Ca2+ current was 0.3 ± 0.02 µM with a Hill coefficient of 0.98 ± 0.07 (n = 15 cells). ()-ECN was about 30-fold less potent
with an IC50 value of 10 ± 1.6 µM and a Hill coefficient of 1.2 ± 0.2 (n = 10 cells). At concentrations producing maximal
block, both compounds were about equally efficacious.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 3.
(+)-ECN is a potent, enantioselective antagonist of
T-type Ca2+ current in rat DRG cells. A,
Traces show inward T-type Ca2+ currents
activated from a holding potential of 90 mV to a test potential of
30 mV before, during and after application of 10 µM
(+)-ECN. Note that current activation and inactivation rates are not
obviously altered by (+)-ECN, despite the ~37% reduction in current
amplitude. All current amplitudes are measured from the peak current to
the current amplitude at the end of test pulse. Cm, 20 pF;
Rs, 10 M. B, The peak T current amplitude is plotted
over the course of an experiment in which three different
concentrations of (+)-ECN were applied to a DRG neuron (Cm,
17 pF; Rs, 11 M). Horizontal bars, times
of steroid application. Note that 30 µM steroid did not
depress peak current amplitude more than that by 10 µM.
C, Block of T current by identical concentrations of the (+) and ()
enantiomers of ECN is compared. At these concentrations, ()-ECN was
about half as effective. The similarity in response to both
applications of 10 µM (+)-ECN indicates that
desensitization between applications does not occur (Cm, 21 pF; Rs, 5 M). D, Concentration-response curves to (+)-
and ()-ECN are plotted. Points (open
symbols, ()-ECN;  , (+)-ECN) are averages of at least five
different cells and are normalized to effect of 10 µM
(+)-ECN within the same cell. Solid line, best fit from
eq. 1 (see Materials and Methods); vertical lines,
mean ± standard error. For (+)-ECN, the IC50 value
was 0.3 ± 0.02 µM with a Hill coefficient of
0.98 ± 0.07 (15 cells), whereas for ()-ECN, the
IC50 value was 10 ± 1.6 µM with a Hill
coefficient of 1.2 ± 0.2 (10 cells).
|
|
The partial blockade of T current by (+)-ECN raises the question of
whether the blocking effect of (+)-ECN is directly on the T channel or
via some modulatory pathway regulating T-current behavior. Four
observations suggest that (+)-ECN does, in fact, block T current
directly. First, T current inhibition by a given concentration of
(+)-ECN is readily reversible and reproducible over sequential
applications. Inhibition of Ca2+ currents by
G-protein mediated pathways often exhibits a characteristic desensitization (Ikeda and Schofield, 1989; Shapiro and Hille, 1993).
Second, the blocking effect of (+)-ECN was observed with an
intracellular saline that lacked either ATP or GTP, two constituents deemed necessary for maintenance of second-messenger mediated signaling
pathways. Third, the addition of either 100 µM GTPS or
2 mM GDPS to the pipette saline did not alter the
ability of (+)-ECN to inhibit T current. With GTPS, 10 µM (+)-ECN blocked 34.5 ± 3.2% (four experiments)
of the T current. With GDPS, 10 µM ECN inhibited
36.3 ± 3.2% (three experiments) of the T current. Neither
GTPS or GDPS resulted in appreciable run-down or run-up of T
current. Finally, the presence or absence of F
in the intracellular saline, an anion which stimulates many G-proteins, did not influence the blocking actions of (+)-ECN. However, we did
observe some variability among cells in the maximal blocking effect of
(+)-ECN on T current. This might occur if T current were partially
contaminated by inactivating, (+)-ECN-resistant HVA current.
Alternatively, the blocking mechanism may involve state-dependent
features, perhaps influenced by modulatory pathways, which may exhibit
cell-to-cell variability.
Blockade by (+)-ECN produces little change in T current kinetic
behavior but exhibits mild voltage- and use-dependence.
Many
compounds are thought to inhibit ion channels either by plugging the
ion permeation or by producing allosteric changes in channel gating,
such that inactivated or closed states are favored. Such effects are
often revealed by kinetic alterations in the channel gating behavior.
To provide initial clues concerning possible mechanisms of (+)-ECN
action, we next examined the effects of (+)-ECN on several aspects of
T-current behavior.
Effects of (+)-ECN on T-current deactivation were examined at
potentials from 160 through 60 mV following a 15-msec depolarizing step to 30 mV. Tail currents were reasonably well described by single
exponential functions over this range and 10 µM (+)-ECN had no obvious effect on current deactivation (Fig.
4A). Current activation time constants
were determined from fits of a Hodgkin-Huxley model (eq. 3) to T
currents activated during a 380-msec depolarizing step to potentials
between 65 mV and +30 mV. Values for n ranged from
2.0 to near 1.0, being near 1.0 at potentials of 20 mV and more
positive. Thus, the Hodgkin-Huxley term m
approximates a single exponential fit to the rising phase of the
current. (+)-ECN (10 µM) had no obvious effect on the
rates of T current activation.
View larger version (39K):
[in this window]
[in a new window]
|
Fig. 4.
Effects of (+)-ECN on kinetic properties of T
currents in DRG neurons. A, Tail currents were examined at potentials
from 160 mV to 60 mV after current activation at 30 mV.
Deactivation values are the mean decay time constants for 4 cells
(mean ± standard deviation). Activation values are mean ± standard deviation from four cells of
m values from eq. 3 (Hill
coefficient ~ 1.0 to 1.4). Solid
symbols, control saline;  , 10 µM (+)-ECN. B,
Symbols plot mean values for four cells of
th values from eq. 3. C, Percent recovery of
peak inactivating current is plotted as a function of recovery duration
at 90 mV. Symbols, mean ± standard deviation for
four cells. Solid line is single exponential fit, whereas dotted line
is two exponential fit. Non-zero initial recovery reflects incomplete T
current inactivation during initial 100-msec step to 30 mV and/or
some unblocked, noninactivating HVA current contamination of the peak
current amplitude. For single exponential fits, recovery time constants
were 448.9 ± 59.1 msec (control) and 619.5 ± 61.8 msec [10
µM (+)-ECN]. From the fit of a two exponential function,
in control saline the fast time constant (f) was
186 ± 25.2 msec (59.1%) with a slow time constant
(s) of 1171 ± 232.8 msec. In 10 µM
(+)-ECN, f was 407.9 + 129 msec (72.5%) and
s was 2123 ± 1450 msec. However, in 10 µM (+)-ECN, a reasonable fit could also be obtained by
constraining f and s to the values
obtained in control saline with the relative amplitude of
f reduced to 45.8%. D, Fractional recovery at 130 mV
with and without 10 µM (+)-ECN is plotted. Single
exponential time constants were 489.3 ± 82.6 msec and 556.5 ± 59.3 msec, for control and 10 µM (+)-ECN,
respectively. For two exponential components, in control saline
f was 95.8 ± 67.2 msec (41.8%) and
s was 830.8 + 227 msec, whereas in 10 µM
(+)-ECN f was 103.5 ± 59.5 msec (29.0%) and
s was 760.2 ± 128.6 msec.
|
|
Potential effects of (+)-ECN on the rate of current inactivation were
also determined from the value of h in the fit
of eq. 3 to the current waveforms. Values plotted in Fig. 4B indicate that 10 µM (+)-ECN had no significant effect on the time
constant of current inactivation.
Recovery from inactivation was examined with a paired-pulse protocol in
which a 100-msec step to 30 mV was first used to inactivate most T
current. After a variable recovery interval (25 to 10,000 msec) at
either 90 mV or 130 mV, a second test step to 30 mV was used to
determine the amount of T current that had recovered from inactivation
during the recovery period. The percent recovery in the presence and
absence of 10 µM (+)-ECN for four cells was then plotted
as a function of recovery duration at either 90 mV (Fig. 4C) or 130
mV (Fig. 4D). Recovery time courses with and without (+)-ECN were best
fit with two exponential components with values given in the legend of
Fig. 4. The time constants of recovery are similar both with and
without (+)-ECN. However, the relative amplitude of the fast recovery
component is somewhat smaller in (+)-ECN, resulting in a somewhat
slower overall recovery.
We next determined whether (+)-ECN might alter T current availability
at different conditioning potentials. T currents were evoked by a
voltage-step to 30 mV after a 5-sec conditioning step at potentials
from 110 to 55 mV in the presence and absence of 10 µM (+)-ECN (Fig. 5A). This
procedure defines the voltage-dependence of T-current fractional
availability (Todorovic and Lingle, 1998). Fig. 5A shows that 10 µM (+)-ECN reduced T current elicited from negative
potentials by almost 50%. The normalized maximal current elicited from
each conditioning potential is plotted as a function of the
conditioning potential in Fig. 5B for a set of eight cells. Fig. 5B,
solid lines, represent the best fits from the Boltzmann equation (eq. 2); for control conditions, half availability occurred at
78 mV with slope factor of 8 mV, whereas in the presence of 10 µM (+)-ECN, the V0.5 was 85.5 mV
with a slope factor of 8.8 mV. These experiments indicate that (+)-ECN
exerts a somewhat stronger blocking effect at more positive
conditioning potentials, but the effect is rather small. The slight
slowing of recovery from inactivation observed in Fig. 4, C and D,
might contribute to the effect of (+)-ECN on steady state inactivation.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 5.
Availability of T current for activation is
influenced by (+)-ECN. A, Traces show currents activated
by voltage steps to 10 mV after a 5-sec step to potentials from 110
through 55 mV either in control saline (top) or 10 µM (+)-ECN (bottom). Cm, 16 pF; Rs, 10 M. B, The average fractional availability of
T current as a function of voltage is plotted for control and 10 µM (+)-ECN for eight cells. Error bars,
mean ± standard error; solid lines, best fit of
eq. 1. For control saline, half inactivation occurred at 78 mV with a
slope factor of 8 mV; in the presence of 10 µM (+)-ECN,
the V0.5 was 85.5 mV with a slope factor of 8.8. C, T
currents were elicited once every 5 sec or once every 20 sec, in the
absence and presence of 10 µM (+)-ECN. The change in
stimulus frequency has no effect on T current amplitude under control
conditions, but in the presence of (+)-ECN, peak T current amplitude is
reduced at higher stimulus frequencies (Cm, 13 pF;
Rs, 12 M).
|
|
The dependence of the fractional block of T current by (+)-ECN on T
current stimulation frequency was also examined. In control conditions,
when T currents are activated by 250-msec depolarizations applied every
20 or 5 sec, no change in peak T current amplitude is noted (Fig. 5C).
However, in the presence of 10 µM (+)-ECN, activation of
T current at 1 per 5 sec increases the amount of blockade by (+)-ECN by
about 25% relative to blockade at 1 per 20 sec. The average increase
in block (when cells are stimulated every 20 versus every 5 sec) was
25 ± 8% (mean ± standard deviation) (n = 7 cells) for 10 µM (+)-ECN. This result is consistent with the somewhat stronger blockade of T current by (+)-ECN at more positive
potentials. At higher stimulation frequencies, the recovery time at
90 mV is insufficient to allow full recovery from the blockade
developed at 10 mV.
(+)-ECN is relatively ineffective at blocking HVA current in rat
DRG neurons.
A number of steroids have been reported to inhibit
HVA types of Ca2+ currents (ffrench-Mullen and
Spence, 1991; Spence et al., 1991; ffrench-Mullen et
al., 1994). Recently, we have shown that (+)-ACN, another
neuroactive steroid, blocks N-, Q-, and R-type HVA currents but not L-
or P-type currents in DRG and hippocampal neurons, with
IC50 values in the range of 5-20
µM (Nakashima et al., 1998). To examine the
effectiveness of (+)-ECN on HVA current, cells were held at 60 mV and
largely noninactivating currents were evoked by depolarizing steps to
10 mV. HVA current in these cells was composed primarily of
nifedipine-sensitive L-type current and GVIA-sensitive N-type current
(Scroggs and Fox, 1992; Todorovic and Lingle, 1998). In Fig.
6A, a cell that exhibited both T-type and
HVA current is depicted. T-type current was initially evoked by a test
step to 40 mV from a holding potential of 90 mV; after return to a
holding potential of 50 mV, a step to 0 mV resulted in activation of
a largely noninactivating HVA current. (+)-ECN (1 µM)
produced a partial inhibition of the peak T current but had no effect
on current activated by the subsequent step to 0 mV. Fig. 6B
illustrates traces of HVA currents from the same cell before, during,
and after application of 30 µM (+)-ECN, which produced a
reversible, 22% reduction of HVA current. Fig. 6C illustrates the time
course of HVA current blockade in another rat DRG cell. (+)-ECN (3, 30, and 60 µM) reversibly reduced the peak HVA current amplitude in a concentration-dependent manner with a maximal block of
about 27%. In this cell, 1 µM GVIA irreversibly blocked
about 16% of the total HVA current indicative of N-type current
blockade, whereas 5 µM Nifedipine (an "L" type
antagonist) blocked most of the remaining HVA current in this cell. For
this cell, this result indicates that at least most of the current
blocked by 60 µM (+)-ECN is primarily L-type current.
Fig. 5D displays the concentration-response curve for blockade of total
HVA current by (+)-ECN in rat DRG cells. All points are an average of
at least five cells (total n = 11 cells). In Fig. 5D,
the solid line is a best fit of eq. 1, yielding an
IC50 value of 9.3 ± 2.7 µM,
with a Hill coefficient of 1.2 ± 0.3 and a fitted maximal
block of 27.6 ± 3%.
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 6.
Effects of (+)-ECN on HVA currents in rat DRG
neurons. A, Traces show currents in the absence and
presence of 1 µM (+)-ECN. Currents were activated with
the voltage protocol shown on the top. A step to 40 mV was used to
activate and inactivate T current; after repolarization to 50 mV, a
step to 0 mV was used to activate HVA current. (+)-ECN (1 µM) blocks about 20% of the inactivating current at 30
mV but has no effect on current activated at 0 mV. Note the different
time bases used for acquisition of LVA and HVA currents.
Vertical calibration bar, time at which the sampling
interval was changed from 0.8 msec to 0.1 msec. B,
Traces show currents activated from a voltage step to
10 mV from a holding potential of 60 mV before, during, and after
application of 30 µM (+)-ECN from the same cell used in
A. About 22% of the total HVA current was blocked. Cm, 10 pF; Rs, 5 M. In C, a time record of peak HVA current
amplitude from another DRG cell shows the relative blocking effect of
3, 30, and 60 µM (+)-ECN. Horizontal
bars, times of steroid application. Note that 60 µM steroid blocked only slightly more current than 30 µM. Comparing the effect of GVIA and nifedipine indicates
that most HVA current in this cell was of L-type.
Cm, 17 pF; Rs, 9 M. D, The
concentration-dependence of blockade of total HVA current by (+)-ECN is
displayed. Smaller size DRG cells containing primarily N- and
L-type Ca2+ currents were used (Scroggs and
Fox, 1992). Points, averages of at least five different
cells; vertical lines, mean ± standard error.
Solid line, best fit of eq. 1, yielding an
IC50 value of 9.3 ± 2.7 µM, a Hill
coefficient of 1.2 ± 0.3, and maximal block of 27.6 ± 3%
(8 cells).
|
|
In other experiments, the effect of (+)-ECN was examined on isolated N-
or L-type HVA current. For these experiments, small DRG neurons, which
express predominantly N- and L-type HVA currents (Scroggs and Fox,
1992), were used. For five DRG neurons in which N-type current was
abolished with 1 µM GVIA, (+)-ECN blocked a maximum of 37 ± 8% of the residual, predominantly
L-type current with an IC50 value of
12.7 ± 6.6 µM (n = 1.6 ± 0.8). For five DRG neurons in which L-type current was
abolished by a combination of intracellular F
and application of 5 µM nifedipine, (+)-ECN blocked a
maximum of 44 ± 4.6% of the residual, predominantly N-type
current with an IC50 value of 8.7 ± 2.1 µM (n = 1.5 ± 0.4). Thus, both N-
and L-type current are only weakly sensitive to (+)-ECN.
We also examined the effects of (+)-ECN on total HVA current in
cultured neonatal rat hippocampal neurons. In such neurons, HVA current
is typically composed of at least 5 distinguishable components. (+)-ACN
has previously been shown to block N-, Q-, and R-types of HVA current
in these cells, with IC50 values of 10-25
µM, but does not affect L- and P-type current
(Nakashima et al., 1998). Here, we simply compared the block
produced by 30 µM (+)-ECN with that produced by 30 µM (+)-ACN. 30 µM (+)-ECN blocked an
average of 21.1 ± 3.5% of total HVA current (n = 5) (data not shown), whereas effects of 10 µM (+)-ECN on
HVA current were difficult to discern. Blockade by 30 µM
(+)-ECN was 50.2 ± 4.5% of the blockade produced by 30 µM (+)-ACN. Thus, in sum, (+)-ECN seems to produce
partial blocking effects on some HVA current components, but with
relatively weak effects at less than 10 µM.
The effects of (+)-ECN on voltage-gated Na+ and
K+ currents in rat chromaffin cells.
We next examined
the effects of (+)-ECN on several other potential ion channel targets
found in cultured adult rat adrenal chromaffin cells. Rat chromaffin
cells express a robust, tetrodotoxin-sensitive voltage-dependent
Na+ current. Fig.
7A shows voltagedependent
Na+ current before and during application of
escalating concentrations of (+)-ECN. (+)-ECN (30 µM but
not 1 and 10 µM) produces a slight reduction (~14%) in
Na+ current amplitude. Fig. 7B plots the time
course of Na+ current amplitude from the same
experiment.
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 7.
Lack of effect of (+)-ECN on voltage-dependent
Na+ current (INa) and K+ currents
in rat adrenal chromaffin cells. A, Traces show
INa activated by the indicated voltage protocol.
Traces in 1 and 10 µM (+)-ECN are
indistinguishable from control, whereas 30 µM produced a
~14% reduction in INa. B, Peak INa amplitude
during the course of an experiment indicates that any effects of
(+)-ECN are rather minor. C, Traces show outward
currents evoked by steps to 90 mV from a holding potential of 110 mV
with 10 µM pipette Ca2+. The inactivating
component of current is BK-type Ca2+dependent
K+ current, whereas the residual noninactivating current
reflects at least one other voltage-dependent K+
conductance. Neither current component was affected by either 10 or 30 µM (+)-ECN. D, Peak current amplitude is plotted as a
function of experimental time illustrating the lack of effect of
(+)-ECN on both peak () and steady state () K+
current components.
|
|
Rat chromaffin cells also express a robust BK-type
Ca2+- and voltage-dependent
K+ current which exhibits inactivation (Solaro
et al., 1995). This current can be observed in relative
isolation by voltage steps to +90 mV when the recording pipette
contains 10 µM Ca2+ (Fig. 7C).
After inactivation of the BK current at +90 mV, there is also a
persistent voltage-dependent K+ current. Neither
10 nor 30 µM (+)-ECN had any effect on either the
inactivating or sustained component of K+
current. The lack of effect of (+)-ECN on either
K+ current is also shown in Fig. 7D.
From the above experiments, we conclude that at 10 µM
(+)-ECN, a concentration maximally effective at blocking T current in rat DRG cells, there is no effect upon several other voltage-gated currents. Even at 30 µM (+)-ECN, effects on
K+ currents are nonexistent with only minimal
effects on Na+ current.
The effects of alphaxalone and (+)-ACN on T current in rat DRG
cells.
The enantioselectivity in the blocking effect of (+)-ECN on
T-type Ca2+ current indicates that particular
structural requirements are necessary for the blocking effect. Although
a more thorough examination of the structural requirements of T-current
block will be required, here we have examined the ability of three
other steroids, alphaxalone, (+)-ACN, and ()-ACN, to block T-type
calcium current in rat DRG cells.
Alphaxalone is the only steroid anesthetic that has been widely used in
human medicine (Sear, 1996). Fig. 8A
shows traces of T current before, during, and after application of 30 µM alphaxalone, which in this cell blocked about 50% of
peak T current. Alphaxalone, in contrast to (+)-ECN, is also a potent
GABAergic agent (Lambert et al., 1995). We were therefore
concerned that the apparent reduction in outward current observed with
alphaxalone might result from a GABAA-receptor
mediated activation of a superimposed outward current. To test this
possibility, DRG neurons were stimulated with voltage ramps from 90
to 90 mV (data not shown) in the presence of cadmium to completely
block inward current. Subsequent application of alphaxalone failed to
evoke any inward or outward current, indicating that the apparent
effects of alphaxalone on T-current do not arise from coincidental
activation of a Cl current.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 8.
Effects of alphaxalone on T currents in rat DRG
cells. A, Traces show T current in a DRG cell before,
during, and after application of 30 µM alphaxalone. B,
The effect of alphaxalone on fractional availability of DRG T current
is illustrated. Solid lines, fits of eq. 2. Alphaxalone
shifted the V0.5 from 79.5 mV (control, open
symbols) to 88.5 mV (alphaxalone, filled
symbols) with slope factors of 7.7 mV (control) and 9.4 mV
(alphaxalone) (n = 5 cells). Error
bars, mean ± standard error. C, Concentration-response
curves for percent inhibition of T current by alphaxalone are plotted.
Points, average of at least five different cells;
bars, mean ± standard error. Solid
line, best fit of eq. 1 with an IC50 value of
1.3 ± 0.3 µM, a hill coefficient of 0.92 ± 0.14, and a fitted maximal block of 62 ± 4.5%
(n = 20 cells). D, HVA currents were activated by
steps to 10 mV from 60 mV in the presence and absence of 10 µM alphaxalone. Alphaxalone had no effect.
|
|
The effect of alphaxalone on the T-current steady state inactivation
curves was also examined. As shown on Fig. 8B, the
V0.5 for T-current availability was shifted from
79.5 mV to 88.5 mV, with slope factors of 7.7 mV and 9.4 mV in the
absence and presence of this steroid, respectively (n = 5 cells). The magnitude of this effect, although not profound, is
comparable with the effect of (+)-ECN on DRG T current. Also similar to
the effect of (+)-ECN, there was an increase in the fractional blockade
by alphaxalone (26 ± 9%; n = 4 cells) as the
frequency of stimulation of T current was increased from every 20 to
every 5 seconds (data not shown). Alphaxalone also shares with (+)-ECN
a lack of any discernible effect on T current activation or
inactivation kinetics (data not shown).
Blockade of T current by alphaxalone was concentration-dependent (Fig.
8C) with an IC50 value of 1.3 ± 0.3 µM, a Hill coefficient of 0.92 ± 0.14, and maximal
block of 55 ± 12% (n = 15 cells). Alphaxalone at
10 µM had no effect on total HVA current in DRG cells
(Fig. 8D; n = 2 cells) and had minimal effect upon
total HVA current in hippocampal neurons (only 7% block of total HVA current at 30 µM; Nakashima et al., 1998).
(+)-ACN is another steroid that, unlike (+)-ECN, has significant
effects upon GABAA receptors (Fig. 2; Wittmer
et al., 1996). Potentiation of currents activated by 2 µM GABA occurs with an EC50 value
of 1.4 µM (+)-ACN, whereas direct gating of
GABAA receptor current by (+)-ACN occurs with an
EC50 value of 5 µM (Wittmer et al., 1996). In addition, (+)-ACN, in contrast to (+)-ECN
and alphaxalone, exhibits blocking effects on specific subtypes of HVA
Ca2+ currents in the range of 5-20
µM (Nakashima et al., 1998).
In rat DRG cells, we found that (+)-ACN also produces enantioselective
blockade of T currents. Maximal block was incomplete being about 40%
and no increase in block was observed between 10 and 30 µM (+)-ACN (Fig. 9A).
Effects of 10 µM (+)-ACN on steady state T current
availability were similar to effects seen with (+)-ECN and alphaxalone
(Fig. 9B). Blockade by (+)-ACN exhibited marked enantioselectivity
(Fig. 9C). At 10 µM, (+)-ACN blocked 44 ± 13% of
the T current (n = 36), with an
IC50 value of 0.4 ± 0.07 µM
and n of 1 ± 0.2. ()ACN was about 50 times less
potent with an IC50 value of 23.5 ± 11 µM, n of 1.4 ± 0.33 and a fitted amount
of maximal T current blockade comparable with that produced by 10 µM (+)-ACN.
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 9.
Effects of (+)-ACN on T currents in rat DRG cells.
A, Traces show inhibition of T current in a DRG cell by
10 and 30 µM (+)-ACN. Maximal block is achieved with
concentrations of about 10 µM. B, The effect of 10 µM (+)-ACN on availability of DRG T current is plotted.
Solid lines, fits of eq. 2. For control currents, the
V0.5 was 68.7 ± 0.7 mV with a slope factor of
6.7 ± 0.6 mV. In 10 µM (+)-ACN, the
V0.5 was 77.3 ± 1.1 mV with a slope factor of
7.6 ± 1.0 mV with a limiting maximal availability of 71.8 ± 2.7%. C, Concentration-response curves show inhibition of peak T
current by (+)-ACN and its enantiomer, ()-ACN. In each cell studied
with ()-ACN, responses were normalized to the block produced by 10 µM (+)-ACN obtained in the same cell.
Points, averages of multiple determinations (at least
five cells), error bars, mean ± standard error;
solid lines, best fits of eq. 1. The IC50
value for block by (+)-ACN was 0.4 ± 0.2 µM with a
Hill coefficient of 1.1 ± 0.6. For ()-ACN, assuming a
comparable maximal block, the IC50 value was 23.9 ± 2.4 µM with a Hill coefficient of 1.3 ± 0.2. D,
Peak T current over the course of an experiment is plotted to
illustrate the lack of additivity of the blocking effects of (+)-ACN
and (+)-ECN. (+)-ACN (10 µM) and 10 µM
(+)-ECN each produce a similar blocking effect, which is also
comparable to block by the simultaneous application of 10 µM (+)-ACN and 10 µM (+)-ECN.
|
|
These results suggest that a number of steroids can block T current in
DRG neurons with a high degree of enantioselectivity. Furthermore, the
partial blockade of T current by these compounds, the lack of kinetic
alterations by these compounds, and similar small shifts in steady
state inactivation curves suggests that each of these compounds may
block T current with a similar mechanism.
If different steroids were acting at different sites and by different
mechanisms to produce blockade of T currents, some additivity in their
blocking effects might be expected. To test this possibility, concentrations of (+)-ACN and (+)-ECN yielding near maximal blocking effects (10 µM in each case) were coapplied on the same
cell (n = 7 cells; Fig. 9D) and compared with
responses to a 10 µM concentration of each steroid alone.
The amount of block when they were given together was not additive,
which suggests that these two steroids may act in a similar fashion,
perhaps at the same site, to block neuronal T current.
| |
Discussion |
A novel neuroactive steroid, (+)-ECN, produces a potent blockade
of T-type Ca2+ current in rat DRG neurons with
50% of the maximal blocking effect occurring at 0.3 µM.
This effect is strongly enantioselective; ()-ECN is more than 30 times less potent. Maximal blockade by (+)-ECN is only about 40% of
total T current. Similarly, for all steroids studied here that do
inhibit T current, maximal blockade was incomplete.
A number of other T current blockers have also been reported to produce
an incomplete block at concentrations producing a maximal effect. For
example, the anticonvulsants phenytoin and -methyl--phenyl-succinimide also block less than 50% of DRG T
current (Todorovic and Lingle, 1998). Partial block of other Ca2+ channel variants has also been described
and, in the case of blockade of P-type current by 
-agatoxin IIIA, it
has been proposed that a partial reduction of the rate of ion
permeation through the P-type channel may account for the partial
blocking effects (Mintz, 1994). In the case of T current block, the
mechanism underlying the partial blockade produced by any compound
remains unknown.
The anticonvulsant drug ethosuximide has been reported to block only
about 40% of T current in thalamic neurons (Coulter et al.,
1989a, 1989b). However, recent work has failed to identify any effect
of 0.5 mM ethosuximide on T current in thalamic neurons (Leresche et al., 1998). In fact, other work indicates that
T current can be maximally blocked by ethosuximide in both GH3 cells (Herrington and Lingle, 1992) and DRG neurons (Todorovic and Lingle, 1998), but only at concentrations (IC50 ~ 20-30 mM) that greatly exceed those used clinically.
Selectivity in blockade by (+)-ECN.
In contrast to alphaxalone
and (+)-ACN, (+)-ECN seems to be relatively selective in its ability to
block T current and exerts little effect on other targets at comparable
concentrations. Although maximal block of T current by (+)-ECN is only
partial, this block is of relatively high affinity, producing half
maximal block at about 0.3 µM. In contrast, at 10 µM, (+)-ECN has only small effects on HVA current in both
rat DRG and hippocampal neurons. Providing additional support for the
idea that (+)-ECN is relatively ineffective against HVA currents, we
have observed that (+)-ECN has weak blocking effects on cloned human
1E Ca2+ channels expressed in HEK cells
(Nakashima Y, Pereverzev A, Schneider T, Covey DF, and Lingle CJ.
Blockade of Ba2+ current through human 1E
channels by two steroid analogs, (+)-ACN, and (+)-ECN; submitted for
publication.), blocking up to about 80% of the 1E current with an
IC50 value of about 19 µM. Thus, HVA Ca2+ currents seem to be largely unaffected
by (+)-ECN at concentrations (~1 µM) producing a near
maximal effect on T currents. This apparently marked selectivity of
(+)-ECN is also supported by the lack of effect on voltage-gated
Na+ and voltage dependent
K+ current and
Ca2+-dependent K+ current
in rat chromaffin cells. (+)-ECN therefore seems to exhibit a
combination of potency and selectivity that may allow it to be of
potential use in the pharmacological evaluation of T currents.
Blockade of T current by both (+)-ECN and (+)-ACN also exhibits strong
enantioselectivity. This implies that the site affected by these
steroids has quite specific structural requirements. Given the
disparity in structure among (+)-ECN, (+)-ACN, and alphaxalone, is it
possible that the blocking effects observed here represent effects on
more than one target site? It is difficult to exclude this possibility.
However, the fact that the blocking effects of (+)-ACN and (+)-ECN are
not additive implies that, at least for these two structurally distinct
steroids, there may be a common site and mechanism of action.
Furthermore, several features of the block of T current produced by
(+)-ECN, alphaxalone, and (+)-ACN support this view. In particular, all
three compounds produce similar changes in steady state inactivation
curves, each produces a partial block at maximal concentrations, and
each has essentially no effect on kinetic properties of T currents.
Although it is possible that each compound acting at distinct sites
might result in this identical set of blocking characteristics, the
simplest view at the present time is that they are acting at the same site.
Despite the similarity in action of (+)-ECN, (+)-ACN, and alphaxalone
on T-type current, the lack of effect of (+)-ECN on GABA receptors
seems particularly remarkable. There are multiple differences in the
structures of (+)-ACN and (+)-ECN that might contribute to the
selectivity of (+)-ECN in producing T channel inhibition, while leaving
many other steroid-sensitive targets unaffected. These differences
include: 1) the positions of the hydroxy and carbonitrile groups; 2)
the relative stereochemistry between these groups; 3) the presence or
absence of a C-19 methyl group; and 4) the distances between the oxygen
and nitrogen atoms. Each of these differences needs to be evaluated
more fully in future studies to understand its contribution to the ion
channel selectivity observed in this study for (+)-ECN.
(+)-ECN, (+)-ACN, and alphaxalone also show interesting differences in
their ability to inhibit HVA Ca2+ currents.
Whereas both (+)-ECN and alphaxalone have relatively small effects on
HVA Ca2+ currents, (+)-ACN seems to inhibit N-,
Q-, and R-type currents with IC50 values in the
range of 5-20 µM (Nakashima et al., 1998). Thus, (+)-ECN and alphaxalone seem to share similar effects on T-type
current and HVA currents, but differ in their effects on GABAA receptors. The lack of effect of (+)-ECN on
HVA currents is also consistent with its lack of effect on inhibitory
or excitatory synaptic currents.
Does T current inhibition result in interesting
clinical/behavioral effects?.
Until recently, T current inhibition
has been the primary proposed explanation for the anticonvulsant
actions of the succinimides (Macdonald and McLean, 1986; Coulter
et al., 1989a, 1989b). As noted above, this hypothesis has
now been challenged by work that has failed to observe inhibition of T
current by appropriate concentrations of ethosuximide (Leresche
et al., 1998). Yet, an important role of T current in
convulsant activity is also suggested by the role of T current in burst
generation in thalamic neurons (Huguenard and Prince, 1992) and the
fact that increases in T current amplitude seem to favor epileptic
discharges (Tsakiridou et al., 1995).
At present, whether inhibition of T currents may contribute to other
clinically or behaviorally important alterations remains unknown.
However, the present results with alphaxalone may support this
possibility. Alphaxalone remains the only anesthetic steroid that has
been widely used in human medicine. Interestingly, whereas T current
blockade by alphaxalone occurs with an IC50 value
of 1.3 µM, the reported values for alphaxalone in plasma
during anesthesia in humans is in the range of 6.5-13 µM
(Sear and Prys-Roberts, 1979). This suggests that, in mammals,
alphaxalone affects T current in subanesthetic concentrations and,
thus, T current inhibition is occurring during the production of
anesthesia. On the other hand, it would seem unlikely that T current
inhibition per se would contribute to the production of anesthesia.
It is interesting to consider several other aspects of the clinical
action of alphaxalone in relation to a possible role of T current
blockade. Alphaxalone has been reported to be a more efficacious agent
in treatment of intractable status epilepticus than classic
GABAergic agents like barbiturates (Chin et al., 1979). It
is also more effective in suppressing epileptic activity in
experimental models than thiopental and diazepam (DeRiu et al., 1987). In addition, alphaxalone has been reported to have stronger analgesic effects than propofol and pentobarbital (Gilron and
Coderre, 1996). It is possible that these clinical effects of
alphaxalone, which distinguish it from other general anesthetics, may
result from effects on novel ion channel targets, perhaps T currents.
Thus, T current inhibition by particular steroids may contribute both
to anticonvulsant effects and analgesic consequences.
In conclusion, we have shown that several steroids inhibit T type
Ca2+ currents at submicromolar concentrations.
Furthermore, one of these compounds, (+)-ECN, produces these effects
while exerting essentially no effects on GABAA
receptors. The strong enantioselectivity in the blocking action of
(+)-ECN indicates that T channels probably contain a steroid binding
site with well-defined structural features. Over the range of
concentrations effective on T current, (+)-ECN has essentially no
effect on HVA Ca2+ currents, voltage-dependent
Na+ current, and some K+
currents at concentrations affecting T currents. (+)-ECN and related
compounds may prove useful in clarifying physiological and behavioral
roles of T currents. Further work may lead to identification of
compounds with even more potency and selectivity in blocking T currents.
We wish to thank A. Evers for comments on the manuscript and for
helpful discussions during various portions of this work.
This work was supported by National Institutes of Health Grants
GM47969 (C.F.Z., D.F.C., and C.J.L.) and MH00964 (C.F.Z.).
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
-Aminobutyric
Acid-Modulatory Activity
Top
Summary
Introduction
