![[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}
|
|
|
|
|
||
Vol. 54, Issue 3, 559-568, September 1998
-Hydroxy-5-androstane-17
-carbonitrile Blocks N-, Q-, and
R-Type, but Not L- and P-Type, High Voltage-Activated
Ca2+ Current in Hippocampal and Dorsal Root Ganglion
Neurons of the Rat
Departments of Anesthesiology (Y.M.N., S.M.T., C.J.L.) and Molecular Biology and Pharmacology (D.F.C.), Washington University School of Medicine, St. Louis, Missouri 63110
 |
Summary |
|---|
Top
Summary
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
High voltage-activated (HVA) Ca2+ current (ICa)
was recorded from neonatal rat hippocampal and adult rat dorsal root
ganglion neurons. In both cell types,
(+)-3
-hydroxy-5-androstane-17
-carbonitrile [(+)-ACN], a
neuroactive steroid, had no effect on nifedipine- (L-type) or
-agatoxin IVA- (P-type) sensitive ICa. Selective blockade of N-type current with -conotoxin GVIA and of Q-type current with -conotoxin MVIIC indicated that (+)-ACN inhibits both
N- and Q-type current components in both cell types. Current persisting
after blockade of all other current components (R-type) was also
sensitive to (+)-ACN. Half-blockade of (+)-ACN-sensitive HVA current
occurred in the range of 3-25 µM, with N-type current somewhat more sensitive than Q- or R-type. The (+)-ACN enantiomer, (
)-ACN, and pregnanolone were somewhat less effective at inhibiting total HVA current than (+)-ACN, whereas several steroid analogs, including alfaxalone, were relatively ineffective at inhibiting total
HVA current. Neither guanosine-5'-O-(2-thio)diphosphate nor guanosine-5'-O-(3-thio)triphosphate altered the
ability of (+)-ACN to inhibit HVA current in dorsal root ganglion
neurons, indicating that (+)-ACN acts directly on Ca2+
channels. The partial selectivity exhibited by (+)-ACN among different
HVA current components suggests that manipulations of steroid analogues
may be a useful strategy in the generation of more selective, more
potent, small-molecular-weight HVA channel blockers.
| |
Introduction |
|---|
|
Top
Summary
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Five
distinguishable subtypes (L, N, P, Q, and R) of HVA
ICa have been discerned in central nervous system
neurons (Ishibashi et al., 1995
; Randall and Tsien, 1995; De
Waard et al., 1996). Many HVA currents share similarity in
activation voltages and kinetic properties. As a consequence,
identification of different current components has depended on
particular peptide toxins, including -CgTx GVIA, -CmTx MVIIC, and
-Aga IVA, that exhibit some degree of selectivity in their blocking
actions. However, the usefulness of these peptides is somewhat limited
for several reasons. First, some toxins block more than one type of HVA
current, in some cases with overlapping potency (Randall and Tsien,
1995; McDonough et al., 1996). For example, the inhibition
by -CmTx MVIIC and -Aga-IVA of both P- and Q-type current has
contributed to the suggestion that P- and Q-type channels may share
similar molecular components (Stea et al., 1994; De Waard
et al., 1996). Although careful consideration of the
concentration dependence of block may allow differential effects on
different current components to be discerned (McDonough et
al., 1996), overlap in sensitivities may limit the usefulness of
such agents in evaluation of physiological roles. Second, the
irreversible or slowly reversible nature of blockade (Randall and
Tsien, 1995; McDonough et al., 1996) of some currents by
different peptides limits their usefulness in experiments in which the
demonstration of reversibility of an effect would be desirable.
Finally, although the potential physiological roles of different HVA
current components can be evaluated in in vitro systems,
peptides are not particularly suitable for in vivo
evaluation of consequences of selective blockade of central nervous
system HVA currents. Therefore, the availability of
small-molecular-weight HVA current blockers with reasonable potency,
selectivity, and reversibility would be of great value not only as
tools for identification and definition of the physiological roles of
particular HVA current components but also as a starting point for the
development of a clinical pharmacology of central nervous system
Ca2+ channels.
Steroids are interesting compounds for evaluation of pharmacological
actions because they provide a rigid structural template on which both
diverse and subtle structural variations can be introduced (Hu et
al., 1993; Han et al., 1996). The anesthetic effects of
some neurosteroids have been well documented, and these effects are
thought to involve, to some extent, potentiation, gating, or both of
GABAA-mediated responses (Harrison and Simmonds, 1984; Wittmer et al., 1996). However, qualitative
differences in action among different steroids in both behavioral
assays and clinical situations suggests that steroids may also exert
effects on other ion channel targets (Kavaliers and Wiebe, 1987;
Wieland et al., 1995). In particular, some steroids have
been reported to produce inhibitory effects on
Ca2+ channels (ffrench-Mullen and Spence, 1991),
although these effects have been reported to involve G protein
mediation (ffrench-Mullen et al., 1994). The availability of
some new, particularly potent, anesthetic steroid analogs (Wittmer
et al., 1996) has prompted our interest in evaluating the
ability of these steroids to inhibit different components of
ICa. Here, using both cultured neonatal rat
hippocampal neurons and acutely dissociated rat dorsal root ganglion
neurons, we examined the sensitivity of different HVA current
components to the novel neurosteroid (+)-ACN. The results show that
although (+)-ACN does not affect L- and P-type current, it does block
N-, Q-, and R-types of current with similar potency in both cell types.
Thus, (+)-ACN exhibits partial selectivity in its blocking actions
among ICa components. These effects of (+)-ACN
seem to involve direct action on Ca2+ channels.
Because the blocking effects of (+)-ACN on HVA
ICa occur at concentrations outside the range of
those effective at potentiating GABAA currents
(Wittmer et al., 1996), blockade of HVA
ICa by (+)-ACN is unlikely to participate in its
anesthetic effects. However, the partial selectivity in action of
(+)-ACN in blocking Q- and N-type current over P- and L-type current
suggests that manipulation of steroid structures is a promising
strategy for the development of more potent, selective HVA
Ca2+ channel antagonists.
| |
Materials and Methods |
|---|
|
Top
Summary
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Cell culture.
Hippocampal neurons were cultured from
1-2-day-old albino rats as described previously (Rodgers-Neame
et al., 1992). Specifically, neurons were grown in
microisland cultures, a procedure that minimizes arborizations
associated with each neuron (Mennerick et al., 1995). Furthermore, neurons were used after 3-7 days in culture (usually 
5)
to help minimize space-clamp problems associated with extensive processes.
). Cells
were kept at room temperature and used for electrophysiology within 4 hr from dissociation. For recordings from DRG neurons, neuronal cell
bodies were plated onto a glass coverslip and placed in a 35-mm culture
dish. All experiments were carried out at room temperature (~25°).
Solutions and drugs. The control Tyrode's solution contained 150 mM NaCl, 4 mM KCl, 2 mM CaCl2, 2 mM MgCl2, and 10 mM HEPES, with pH adjusted to 7.4 with NaOH. The standard external solution used to isolate IBa contained 5 mM BaCl2, 160 mM tetraethylammonium chloride, 10 mM HEPES, and 0.1 mM EGTA, with pH adjusted to 7.4 with tetraethylammonium hydroxide (standard osmolarity, 310 mOsM). The composition of standard internal solution was 110 mM Cs-methane sulfonate, 14 mM phosphocreatine, 10 mM HEPES, 9 mM EGTA, 5 mM Mg-ATP, and 0.3 mM Tris-GTP. The pH was adjusted to 7.3 with CsOH (standard osmolarity, 300 mOsM).
(+)-ACN, ()-ACN (Hu et al., 1993; Wittmer et
al., 1996), pregnanolone (3-hydroxy-5-pregnan-20-one; Sigma,
St. Louis, MO), pregnenolone (3-hydroxypregn-5-en-20-one; Sigma),
pregnenolone sulfate (Sigma), (+)-3-hydroxy-5-pregnan-20-one, and
alfaxalone (3-hydroxy-5-pregnane-11,20-dione; Sigma) were each
dissolved in DMSO to make 10 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 <0.6% in these experiments; 1% DMSO did
not affect IBa (data not shown, four
experiments). At steroid concentrations of 
30 µM, the
steroids seem to crystallize out of solution, thus reducing the
effective steroid concentration in solutions kept for >~10 min (data
not shown, 10 experiments). To avoid this problem, all solutions
containing concentrations of 30 µM were prepared just before each application (<30 sec). Solutions were applied to cells with the "Y-tube" technique, in which the previous solution in the
tubing can be quickly flushed out and the fresh steroid solution quickly positioned for the next solution application.
-CgTx GVIA (RBI, Natick, MA; and Sigma), -CmTx MVIIC (RBI and
Sigma), -Aga IVA (gift from Pfizer, Groton, CT), and
CdCl2 were dissolved in distilled water to make
stock solutions of 0.5, 0.5, 0.2, and 200 mM, respectively.
Nifedipine (RBI) was dissolved in DMSO at 5 mM as a stock
solution. GDPS and GTP
S were obtained from Sigma and, when used,
replaced GTP in the pipette solution.
In some experiments, P- and Q-type IBa was
blocked before initiation of recording (see Fig. 4). To accomplish this
preblockade procedure, cells were incubated in an external solution
containing 2 µM -CmTx MVIIC and 0.7 µM
-Aga IVA for >30 min. After washing out this solution, recovery
from block of P- and Q-type currents is quite slow (McDonough et
al., 1996), and block of P-type current is essentially
irreversible over the time course of our experiments in the absence of
strong depolarizing voltage steps (e.g., Mintz et al.,
1992). To ensure the persistence of blockade of P- and Q-type currents
using this procedure, in some cells -CmTx MVIIC was reapplied just
before the initiation of recording. In a few tests, -CmTx MVIIC was
reapplied after recording was initiated. In such cases, additional
block of IBa was minimal.
Methods for isolation of (+)-ACN-mediated inhibition of
IBa.
As described in the text, coincident activation
of the GABAA receptor by (+)-ACN or other agents
may confound efforts to define effects of (+)-ACN on
IBa. The standard procedure used for all experiments on hippocampal neurons described here was to include 100 µM picrotoxinin (Sigma) and 50 µM
bicuculline (Sigma) in all solutions. This solution had no direct
effect on IBa, although in most cells, the
combination of picrotoxinin/bicuculline blocked an outward current.
Although the picrotoxinin/bicuculline combination fully blocked the
ability of -aminobutyric acid to activate any current in the
hippocampal neurons, in some cells, we observed a small direct
activation of a presumed ICl by (+)-ACN in the presence of both antagonists (Fig. 1A).
In the presence of 200 µM Cd2+, the
size of this potentially contaminating current was never larger than a
few pA near 0 mV. A contaminating current of this magnitude does not
influence any of the key observations or conclusions of this work. In
DRG neurons, (+)-ACN did not activate any current in the presence of
Cd2+ (five experiments).
|
Electrical recording technique.
Single hippocampal or DRG
neurons were voltage-clamped using the whole-cell configuration of the
patch-clamp technique (Hamill et al., 1981). The resistance
of the patch electrode was 2-4 M
when filled with the internal
solution. After a gigaohm seal was established, a strong negative
pressure was briefly applied to the pipette interior to rupture the
patch membrane. In hippocampal neurons, the experiments were initiated
after >5 min, at which time the amplitude of IBa
had stabilized. Typical series resistance (Rs) values at such times were ~10
M with a typical whole-cell capacitance of 10 pF. In DRG neurons,
for the set of smaller neurons (37 experiments),
Rs = 7.8 ± 2.8 M and
Cm = 11.3 ± 3.1 pF. For larger
neurons (36 experiments), Rs = 6.5 ± 2.2 M and Cm = 26.8 ± 8.7 pF.
10 mV for 25 msec were applied every 15 or 20 sec. In hippocampal neurons, a holding
potential (Vh) of 80 mV was used;
for DRG neurons, the holding potential was 60 mV to abolish
activation of T-type current (e.g., Todorovic and Lingle, 1998). For
I-V relationships, a ramp pulse (dV/dt = ± 1 V/sec) was applied
from Vh of 80 mV, and voltage
extremes were set to 80 and +100 mV. The I-V relations were measured
from the depolarizing portion of the ramp pulses. The current and
voltage were recorded using a patch-clamp amplifier (Axopatch-1C, Axon Instruments, Foster City, CA) and pClamp software (Axon Instruments) and stored in the computer.
Because these cultured hippocampal neurons have long processes,
voltage-clamp conditions were less than ideal, even if neurons with
shorter processes are selected. Because of this, rapid components of
current, such as tail currents, are not likely to reflect the true
amplitude and time course of IBa behavior.
However, all measurements of amplitudes from holding, peak, and steady
state currents are made at time points sufficient to ensure reasonably
well-clamped current conditions. Example traces from hippocampal
neurons typically show currents after subtraction of traces obtained in
Cd2+. For currents from DRG neurons, traces show
raw currents without leak subtraction. No correction was made for
Rs.
Statistics. The concentration-response curves for the percent block of ICa by (+)-ACN were drawn according to the equation
|
(1) |
| |
Results |
|---|
|
Top
Summary
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
(+)-ACN reduces HVA IBa in cultured hippocampal and
acutely dissociated DRG neurons.
A number of compounds reported to
have effects on ICa also are known to produce
direct gating of GABAA receptors; this includes a
number of anesthetics such as propofol, isoflurane, and pentobarbital (Gross and MacDonald, 1988; Jones et al., 1992; Hara
et al., 1994; Olcese et al., 1994; Study, 1994).
The steroid, (+)-ACN, also activates GABAA-gated
Cl
channels in rat hippocampal neurons at
concentrations in excess of 1 µM (Wittmer et
al., 1996). We therefore were concerned that despite the usual
solutions used to isolate IBa, coincident
activation of ICl might contaminate records of
IBa.
10 mV from 80 mV, before, during, and
after the application of 10 µM (+)-ACN. At concentrations
of 10 µM, (+)-ACN consistently reduces the magnitude of
inward current. To illustrate that these traces are not confounded by
the simultaneous activation of an outward ICl,
Fig. 1B shows I-V curves obtained from voltage ramps applied before and
after the application of (+)-ACN in the presence of the
bicuculline/picrotoxinin cocktail. In addition, a third trace was
generated in the presence of 200 µM
Cd2+. The similar intersection of current traces
obtained before and during 30 µM (+)-ACN with that
obtained during blockade of IBa by
Cd2+ indicates that the I-V relationship is not
significantly contaminated by the coincident activation of
ICl. The activation of ICl
by (+)-ACN would have substantially shifted the apparent
IBa reversal potential to much more negative
potentials. Thus, Fig. 1, A and B, shows that (+)-ACN produces
appreciable inhibition of HVA IBa in hippocampal
neurons.
IBa in these hippocampal cells consists of
multiple, pharmacologically distinct components (Ishibashi et
al., 1995). Fig. 1C demonstrates the pharmacological
identification of specific subtypes of IBa in
these cells. We observed that 1 µM -CgTx GVIA, 0.1 µM -Aga IVA, and 5 µM nifedipine,
relatively selective blockers of N-, P-, and L-type
Ca2+ channels, blocked 29 ± 8%, 11 ± 7%, and 26 ± 5% of the total IBa,
respectively (13 experiments). These fractions are comparable to
earlier work on hippocampal CA1 neurons, in which N, P-, L-, Q-, and
R-type current components comprised 27 ± 3%, 13 ± 1%, 38 ± 4%, 9 ± 2%, and 13 ± 2% of total
IBa,respectively (Ishibashi et
al., 1995). In addition, in our experiments, after blockade of N-,
P-, and L-type -, -CmTx MVIIC (1-2 µM) blocked an
additional 18 ± 4% of ICa (four
experiments), whereas ~16% of Cd2+-sensitive
current was resistant to all Ca2+ channel
blockers. We will refer to this resistant current as R-type (Randall
and Tsien, 1995), although the identity of such currents remains
unclear and may to some extent involve residual unblocked channels of
the other Ca2+ channel subtypes.
To provide an additional test of the effects of (+)-ACN on HVA
ICa, the effects of (+)-ACN on HVA current in
acutely dissociated DRG neurons also were examined. As shown in Fig.
1D, (+)-ACN markedly reduces total IBa in these
cells. The identity of different HVA current components in DRG neurons
depends to some extent on the diameter of the underlying cells (Scroggs
and Fox, 1992). Small-diameter neurons typically seem to be selectively
enriched in L- and N-type HVA currents (Scroggs and Fox, 1992). In
contrast, larger DRG neurons express a larger portion of non-L, non-N
current, which has been reported to be of P-type (Mintz et
al., 1992), Q-type (Rusin and Moises, 1995), and a component
unblocked by any known blocker (Rusin and Moises, 1995), presumably
R-type. Therefore, for examination of the effects of (+)-ACN on L- and
N-type current DRG neurons, we selected smaller-diameter cells
(generally <21 µm), whereas larger cells (>23 µm) were used for
examination of the effects of (+)-ACN on other currents.
L- and P-type IBa are insensitive to (+)-ACN.
To
determine whether (+)-ACN may selectively block particular components
of IBa, the amplitude of current blocked by 30 µM (+)-ACN was compared before and after blockade of
particular HVA current components. Fig.
2A plots the absolute amplitude of
current activated by depolarizing steps to 10 mV during the time
course of an experiment. Fig. 2B shows the original
IBa traces. (+)-ACN (30 µM) blocked
~150 pA of IBa. After blockade of ~60 pA of
current by the addition of 0.1 µM -Aga IVA to the
external solution, 30 µM (+)-ACN still blocked ~150 pA
of current. In 15 hippocampal cells, (+)-ACN did not block P-type
current. This suggests that under conditions that result in complete
blockade of P-type current, the current blocked by (+)-ACN is
unaffected. Similar results were obtained in DRG neurons in which L-
and N-type currents were blocked by nifedipine and -CgTx GVIA,
respectively. As shown in Fig. 2C, 30 µM (+)-ACN blocks
~1000 pA of current in this DRG neuron. After blockade of ~300 pA
of IBa by 0.1 µM -Aga IVA, 30 µM (+)-ACN still blocks ~1000 pA of current. Fig. 2D
displays traces indicating that (+)-ACN blocks the same amount of
current before and after blockade of current by -Aga IVA. The lack
of effect of -Aga IVA on blockade by (+)-ACN was observed in seven DRG neurons (also see Fig. 5C).
|
|
-CgTx GVIA-blockable current (N-type) is blocked by
(+)-ACN.
To examine the effects of (+)-ACN on N-type current in
hippocampal neurons, P- and Q-type currents were first minimized by prior blockade with -Aga IVA and -CmTx MVIIC as described in Materials and Methods. -CmTx MVIIC also blocks N-type current, but
the block of N-type is rapidly reversible. Nifedipine then was applied
continuously to remove L-type current. Fig.
4A illustrates the effect of (+)-ACN
after blockade of P-, Q-, and L-type current. In this cell, 10, 30, and
60 µM (+)-ACN were sequentially applied, and 60 µM (+)-ACN blocked ~200 pA of current, which was almost all of the remaining Cd2+-blockable current.
After blockade of ~150 pA of N-type current with -CgTx GVIA, 60 µM (+)-ACN blocked <50 pA of current. Representative traces from this experiment are shown (right), indicating
that in this cell almost all the -CgTx GVIA-blockable current also is blocked by (+)-ACN.
|
-CgTx GVIA was
large relative to that persisting after block of N-type current. Thus, block of the -CgTx GVIA-resistant current by 60 µM
(+)-ACN is only a small portion of the block produced by 10 and 30 µM (+)-ACN before -CgTx GVIA application. Thus, in
this cell, the amount of current blocked by 10 µM (+)-ACN
is somewhat more than half of the total -CgTx GVIA-blockable
current, with some uncertainty due to some residual block by 60 µM (+)-ACN or perhaps current run-down. Thus, 10 µM (+)-ACN may be blocking a little more than half of the
N-type current in this cell.
A second example of block of N-type current in a hippocampal neuron is
shown in Fig. 4B. Again, P-, Q-, and L-type currents were blocked as
above; 10, 30, and 60 µM (+)-ACN then were applied sequentially, and the amplitude of the current blocked by 60 µM (+)-ACN was reduced from ~750 pA to ~250 pA after
blockade of N-type current by -CgTx GVIA. In this cell, presumed
R-type current is a larger fraction of the total
IBa. Taking into account that a larger portion of
the current blocked by 10, 30, and 60 µM (+)-ACN is
non-N-type current, in this cell, it seems that 10 µM
(+)-ACN is blocking less than half of the total -CgTx GVIA-sensitive current. Overall, in seven cells studied with this method, the absolute
amount of combined N/R-type current blocked by (+)-ACN was markedly
reduced after the application of -CgTx GVIA. A substantial portion
of the current blocked by (+)-ACN was removed by the irreversible blocking action of -CgTx GVIA on N-type current. Because both the N
and the non-N component are sensitive to (+)-ACN, it is difficult to
assess the relative concentration dependence of the N-type current in
hippocampal neurons. However, it is clear that 10 µM
(+)-ACN can block a substantial fraction of N-type current. Furthermore, the persistent R-type current also exhibits sensitivity to
(+)-ACN.
In DRG neurons, small-diameter neurons express predominantly L- and
N-type HVA current (Scroggs and Fox, 1992). Selection of such cells
allows relatively unambiguous examination of N-type current after
blockade of L-type current. In the DRG neuron shown in Fig. 4C, 30 µM (+)-ACN blocks almost 2800 pA of HVA current. After
blockade of L-type current with nifedipine, the amplitude of the
current blocked by (+)-ACN is unchanged. Subsequent blockade of N-type
current with -CgTx GVIA almost completely blocks any inhibitory
effect of (+)-ACN on the residual IBa. Thus, as
shown in Fig. 4C (right traces), in a cell with
predominantly N-type current, removal of N-type current by -CgTx
GVIA almost completely removes the (+)-ACN-blockable current. N-type
current in DRG neurons seems to be strongly blocked by (+)-ACN.
An -CmTx MVIIC-sensitive current is also blocked by
(+)-ACN.
A similar strategy was used to assess whether (+)-ACN
might produce inhibition of Q-type current. After preblockade in a
hippocampal neuron of L-, N-, and P-type currents with nifedipine,
-CgTx GVIA, and -Aga-IVA, respectively, sequential applications
of 10, 30, and 60 µM (+)-ACN showed that up to ~70% of
the remaining Q/R-type current could be blocked (Fig.
5A). Fig. 5B (top traces) shows representative currents before (+)-ACN application and during the
application of 60 µM (+)-ACN. Subsequent application of 2 µM -CmTx MVIIC then blocked almost 200 pA of current,
resulting in a marked reduction in the amount of current that was
subsequently blocked by (+)-ACN (Fig. 5B). In nine of nine cells, the
absolute amount of combined Q/R-type current inhibited by (+)-ACN was
markedly reduced after the application of -CmTx MVIIC. This result
shows that most or all of the current blocked by -CmTx MVIIC is also blocked by (+)-ACN, after prior blockade of L-, N-, and P-type currents. We do not know how much of the current persisting after -CmTx MVIIC application is residual unblocked Q- or R-type current.
|
-CgTx GVIA, and 5 µM nifedipine was continuously applied. (+)-ACN blocks
~700 pA of the remaining current in this cell. After blockade of
~250 pA of current by -Aga IVA, (+)-ACN still blocks a similar
amount of IBa. However, after a ~5-min
application of 2 µM -CmTx MVIIC, which blocked ~600
pA of current, (+)-ACN only blocked ~150-200 pA of additional
current. This suggests that (+)-ACN blocks a Q-type current (i.e.,
current sensitive to -CmTx MVIIC that persists after application of
100 nM -Aga IVA).
Finally, the persistence of some (+)-ACN-sensitive current after block
of all other current components in both DRG and hippocampal neurons
again argues that R-type current also is sensitive to (+)-ACN. However,
it is possible that in experiments in which P- and Q-type currents were
preblocked by -Aga IVA and -CmTx MVIIC, the residual,
(+)-ACN-sensitive current might reflect some recovery from the toxin
block of P- and Q-type current. To exclude this possibility, in several
hippocampal cells an additional application of -CmTx MVIIC was used
to show that little recovery from blockade of P- and Q-type current had
occurred. Furthermore, the remaining (+)-ACN-sensitive current was
larger in amplitude than any residual -CmTx MVIIC-sensitive
component of current, indicating that R-type current is, in fact,
blocked by (+)-ACN. Similarly, in DRG neurons, (+)-ACN blocked residual
current in all 10 neurons where N-, L-, P-, and Q-type currents were
pharmacologically removed.
Concentration-dependence of block of IBa by (+)-ACN. Because we know of no selective inhibitors of R-type current, for the hippocampal neurons it is not currently possible to define the pharmacological sensitivity of isolated Q- or N-type current to (+)-ACN in the absence of concomitant R-type current activation. As a consequence, three separate concentration-response curves were generated to assess the concentration dependence of (+)-ACN inhibition. First, the concentration-dependent blockade of IBa by (+)-ACN was evaluated after simultaneous blockade of L-, P-, N-, and Q-type currents. This allows definition of the sensitivity of R-type current to (+)-ACN (Fig. 6A). Second, the effects of (+)-ACN on combined Q/R-type current were determined (Fig. 6B). Third, the effects of (+)-ACN on combined N/R-type current were defined (Fig. 6C). It should be kept in mind that each of these concentration-response curves is limited by assumptions about the pharmacological specificity of various blockers and, when the preblocking procedure was used, by the extent to which block persists. However, these curves provide an important first step in clarifying the relative specificity of (+)-ACN in blocking native ICa.
|
-CgTx
GVIA-sensitive current. This suggests the N-type current sensitivity in
hippocampal cells is not too dissimilar from that in DRG neurons. One
explanation for the differences in Fig. 6C is that the lower affinity
IC50 value of the combined N/R
concentration-response curve for hippocampal neurons may be skewed by a
large contribution of R-type current in most cells and the weaker
sensitivity of the R-type current (Fig. 6A).
Inhibition of HVA current by (+)-ACN is not G protein
mediated.
Several aspects of the block of different HVA current
components by (+)-ACN seem inconsistent with the involvement of a G protein-mediated pathway. First, (+)-ACN seems able to block N-, Q-,
and R-type currents almost completely. Second, the amount of inhibition
seems to be quite stable over repeated applications of (+)-ACN. Third,
we have observed none of the temporal alterations of HVA current time
course often observed for some types of G protein-mediated inhibition
(Mintz and Bean, 1993; Ikeda, 1996). To address this issue more
directly, we examined the ability of alterations of G protein-mediated
signaling to interfere with the inhibitory action of (+)-ACN on HVA
current in DRG cells.
S into the recording pipette.
After activation of G proteins, GDPS, an antagonist of G protein
activation (Eckstein et al., 1979; Holz et al.,
1986), will displace other guanine nucleotides from the GDP/GTP binding site on the G protein subunit. Subsequent activation of G protein coupled receptors will fail to result in G protein activation. Such an
experiment is illustrated in Fig. 7, in
which inclusion of GDPS results in abolition of a muscarinic
receptor-mediated inhibition of HVA current (Fig. 7B) without affecting
inhibition produced by 10 µM (+)-ACN (Fig. 7, A and B).
In these experiments, GDPS also produced a characteristic run-up of
HVA current, perhaps indicative of the removal of some tonic inhibition
of HVA current under our experimental conditions. In these experiments,
because small DRG neurons were used and L-type currents were blocked
with 5 µM nifedipine, the HVA current was predominantly
of N-type. In the presence of 2 mM GDPS, 10 µM (+)-ACN inhibited 80.5 ± 6.1% (mean ± standard deviation; four experiments) of the HVA current. With 300 µM GTP in the pipette, 10 µM (+)-ACN
inhibited 83 ± 10% (mean ± standard deviation; five
experiments) of the HVA current persisting after blockade of L-type
current.
|
S. With GTPS, HVA current exhibited a rapid
run-down to ~10-20% of its initial level as seen in other systems
(Ikeda and Schofield, 1989). Once this steady state level of HVA
current was achieved, 10 µM (+)-ACN still blocked
76.3 ± 6.6% (mean ± standard deviation; six experiments)
of the residual HVA current. In two of these six cells, -CgTx GVIA
was applied after the reduction in HVA current by GTPS had
approached a steady state level (Fig. 7C). In both cases, most of the
residual current blocked by (+)-ACN was also blocked by -CgTx GVIA,
indicating that N-type current is still blocked by (+)-ACN after G
protein activation.
Initial structure-activity relationships for blockade of
ICa by steroids.
There have been other reports of
blocking effects of steroids on ICa
(ffrench-Mullen and Spence, 1991; Spence et al., 1991; ffrench-Mullen et al., 1994). To assess to what extent the
effects of (+)-ACN may resemble the actions of these other steroids, we examined several steroid analogs, including some previously reported to
block IBa. To provide a simple comparison of the
relative effectiveness of these agents, the ability of 30 µM concentration of each compound to inhibit total
IBa was determined and compared in the same cells with the ability of 30 µM (+)-ACN to inhibit total
IBa. These results are summarized in Table
1. ()-ACN, the enantiomer of (+)-ACN,
was somewhat weaker than (+)-ACN in inhibiting total HVA
IBa. The anesthetic steroid pregnanolone also
exhibited an effectiveness comparable to (+)-ACN in inhibiting total
IBa. In contrast, the anesthetic steroids
alfaxalone and (+)-3-hydroxy-5-pregnan-20-one produced only small
effects on total IBa at 30 µM.
Furthermore, we observed that 30 µM pregnenolone was
essentially ineffective at producing inhibition of HVA
IBa. In contrast, 30 µM PS was observed to produce a small, but consistent, increase in total HVA
IBa. Voltage-ramp elicited currents showed that
the increase produced by PS was not associated with any change in
outward leak current or change in the apparent reversal potential for
IBa (data not shown). Thus, the results suggest
that PS may produce a direct enhancement of IBa.
|
)-ACN to block N-,
Q/R-, and R-type current in DRG neurons. Similar to the results on
total HVA current in hippocampal neurons, (+)-ACN is slightly more
effective than ()-ACN in blocking each of the individual current
components. For N-type current, the fractional blockade by 30 µM ()-ACN was 0.76 ± 0.09 (five experiments) of
that elicited by 30 µM (+)-ACN. For combined Q/R-type
currents, the fractional blocking effect of ()-ACN was 0.61 ± 0.08 (four experiments) of that of (+)-ACN. For R-type current, the
current blocked by ()-ACN was 0.65 ± 0.08 (three experiments)
of that blocked by (+)-ACN. This suggests that the relative sensitivity
of each HVA current to (+)-ACN and ()-ACN is similar and that
blockade of total HVA current by the two enantiomers involves the same
set of ICa subtypes. Furthermore, there is clear enantioselectivity in the effects of (+)-ACN. It is worth noting that
assuming similarly shaped concentration-response curves for both
(+)-ACN and ()-ACN, a reduction in block in the range of 0.61-0.76
could correspond to shifts in IC50 values of as
much as 3-8-fold.
| |
Discussion |
|---|
|
Top
Summary
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
One of the key findings of this work is that (+)-ACN, an
anesthetic steroid, completely and reversibly blocks Q-, N-, and R-type
HVA IBa without affecting L- and P-type HVA
current in two different cells, neonatal rat hippocampal neurons and
acutely dissociated DRG neurons. Although N- and Q-type currents are
somewhat more sensitive to (+)-ACN blockade than R-type current, the
IC50 value for blockade in each case is in the
range of ~3-30 µM (+)-ACN. The effects of (+)-ACN do
not involve G protein-mediated inhibitory pathways, suggesting that
inhibition may result from specific binding sites on
Ca2+ channels. A number of other steroids were
much less effective at reducing HVA current, indicative that compounds
must meet specific structural requirements to produce blockade. This
suggests that selective structural alterations of particular steroids
may result in compounds with more specificity, greater potency, or
both. Specific, reversible, small-molecular-weight blockers of HVA
ICa subtypes would provide valuable aides in the
investigation of the physiological roles of different central nervous
system Ca2+ channel variants. Because the effects
of (+)-ACN on IBa occur at concentrations higher
than those that produce effects on GABAA receptors, the anesthetic effects of (+)-ACN (Wittmer et
al., 1996) are unlikely to involve any effects on HVA
ICa.
The selectivity in block of HVA current components by (+)-ACN.
Our assertion that (+)-ACN selectively blocks N-, Q-, and R-type
currents over P- and L-type currents requires that different components
of HVA current can be clearly distinguished. Separation of L- and
N-type currents can be relatively easily accomplished with nifedipine
and -CgTx GVIA. We observed no reduction in the blocking effects of
(+)-ACN after blockade of P- and L-type currents (Figs. 2 and 3). The
pharmacological effects of -Aga IVA and -CmTx MVIIC are more
complicated. -Aga-IVA seems to inhibit both P- and Q-type current at
higher concentrations (1 µM) (Randall and Tsien, 1995),
whereas for sufficiently brief applications at 0.2 µM,
the effects of -Aga IVA are thought to be primarily on P-type
current (Mintz et al., 1992; Randall and Tsien, 1995). Similar separations between -Aga IVA- and -CmTx MVIIC-sensitive currents have been reported in rat DRG neurons (Rusin and Moises, 1995)
and in a number of rat central and peripheral neurons (McDonough et al., 1996). However, Tottene et al. (1996),
using rat cerebellar Purkinje cells, did not find a current
distinguishable by -Aga IVA and -CmTX MVIIC. Our observations
support the former view and provide additional evidence for the idea
that P and Q components are pharmacologically distinguishable.
Specifically, (+)-ACN has no effect on that current removed by brief
treatments with 0.1-0.2 µM -Aga IVA (Fig. 2), whereas
(+)-ACN does block current also blockable by -CmTx MVIIC. In the
case of -CmTx MVIIC, its ability to inhibit N-, P-, and Q-type
currents can complicate interpretation of its actions. Similar to
previous work (McDonough et al., 1996), we observed that a
rapid and reversible block of IBa produced by
-CmTx MVIIC corresponded to the -CgTx GVIA-sensitive N-type current. We did not evaluate whether -CmTx MVIIC could slowly block
P-type current, but we did observe that a slowly blocked, -CmTx
MVIIC-sensitive current persisted after blockade by 0.1 µM -Aga-IVA (Figs. 2 and 5). Overlap of the -CmTx
MVIIC-sensitive current with current blocked by (+)-ACN strongly
supports the view that part of the action of (+)-ACN involves blockade
of a Q-type current (Fig. 5).
-Aga IVA and
-CmTx MVIIC has led to the suggestion that there may be some similarity in molecular components of these channels (Stea et al., 1994; De Waard et al., 1996). One proposal is that
differences between P- and Q-type currents may arise from the subunits that are coassembled with the 1A subunits (Moreno et
al., 1997). The current results suggest that (+)-ACN
exhibits strong specificity in its ability to distinguish between Q-
and P-type currents in both hippocampal and DRG neurons. This suggests
that Q- and P-type channels must contain at least some different
molecular components.
(+)-ACN also has significant blocking effects on presumed R-type
current. After prior blockade of all other current components, an
inhibitory action of (+)-ACN on the remaining current persisted (Figs.
4 and 5). This assertion is tempered by the fact that some of the
residual current may reflect unblocked channels of other types (e.g.,
some Q-type current). R-type current is now suspected to arise from
1E subunits (Randall and Tsien, 1997). In support of the idea that
R-type current is blocked by (+)-ACN, current arising from 1E
Ca2+ channel subunits is also blocked by (+)-ACN
with a concentration dependence similar to the block of R-type current
observed here (Nakashima et al., in preparation).
Ca2+ channels as possible targets of anesthetics.
(+)-ACN is a potent anesthetic in both tadpole and mouse anesthesia
assays (Wittmer et al., 1996). As with other anesthetics (Jones et al., 1992; Hara et al., 1994), the
ability of (+)-ACN to potentiate GABAA-activated
currents probably accounts for its anesthetic effects. The potential
role, if any, of ICa inhibition in the actions of
anesthetics remains unclear. Both L- and T-type ICa have been reported to be somewhat affected by
anesthetics at concentrations that may occur clinically (Herrington
et al., 1991; Study, 1994; Todorovic and Lingle, 1998),
whereas several general anesthetics were reported to be ineffective at
blocking P-type current in rat Purkinje neurons (Hall et
al., 1994). Thus, different types of Ca2+
channels may exhibit differential sensitivities to various anesthetic compounds. However, in most cases, the effects of anesthetics on
ICa are probably secondary compared with the
ability of the same compounds to potentiate
GABAA-mediated currents (Franks and Lieb, 1994).
For (+)-ACN, its relative lack of effect on any HVA current at
concentrations of 1 µM argues that the behavioral effects of (+)-ACN do not involve effects on HVA
ICa. However, we cannot exclude the possibility
that for other steroids, Ca2+ channel inhibition
may occur at concentrations producing anesthesia.
Steroids as potential probes of Ca2+ channels and the mechanism of (+)-ACN action. (+)-ACN does not strongly distinguish among N-, Q-, and R-type currents, although N-type current seems to be the most sensitive to (+)-ACN. Yet, (+)-ACN seems to be somewhat unique in its apparently absolute specificity between Q- and P-types of ICa. We consider this a useful starting point for an attempt to identify reversible, small-molecular-weight blockers of specific subtypes of Ca2+ channels.
A comparison of the relative effectiveness of various steroids on inhibition of total HVA ICa with ability to potentiate GABAA responses (Table 1) indicates a different rank order of potency for the two effects. This comparison must be taken with some caution because, except for (+)-ACN and ()-ACN, we do not have any information about the specificity of
action of other Ca2+ channel-blocking steroids
among different HVA current components. However, because some compounds
have essentially no effect on Ca2+ channels, we
can conclude that the structural requirements of the site involved in
steroid inhibition of some Ca2+ channels must be
different from that on GABAA channels.
It has been reported that inhibition of some components of
ICa in guinea pig hippocampal neurons by
pregnenolone and PS involves a G protein-mediated pathway
(ffrench-Mullen et al., 1994). Because in the cultures of
rat hippocampal neurons used here, we have been unable to find any
blocking effects of pregnenolone or PS, we believe that these earlier
reports are unrelated to the blocking effects of (+)-ACN. Furthermore,
the current results argue strongly that the blocking effects of (+)-ACN
do not involve a G protein-mediated pathway. Interference of G
protein-mediated pathways with either GTPS or GDPS had no effect
on the ability of (+)-ACN to inhibit HVA current in DRG neurons.
Although we cannot rigorously exclude the possibility that Q- and
R-type current inhibition might involve G protein pathways, we consider
this unlikely for two reasons. First, inhibition by (+)-ACN seems to
approach 100% for all current components. G protein-mediated
inhibition typically results in only partial inhibition of an
ICa (e.g., Shapiro and Hille, 1993; Viana and
Hille, 1996). Second, the reproducibility and reliability of the
inhibition by (+)-ACN contrast with the often-desensitizing nature of G
protein-mediated inhibition of ICa (Ikeda and
Schofield, 1989; Shapiro and Hille, 1993).
In summary, (+)-ACN blocks N-, Q-, and R-type
Ca2+ channels with essentially no effects on P-
and L-type channels. Inhibition of Ca2+ channels
by (+)-ACN seems to involve direct interaction of steroids with the
Ca2+ channels. The structural requirements for
steroid blockade of Ca2+ channels differ from the
requirements for effects on GABAA receptors. Thus, modifications of steroid structures may be a fruitful approach to
the development of new small-molecular-weight, potent, specific, and
reversible blockers of HVA ICa.
| |
Acknowledgments |
|---|
We are grateful to the Zorumski lab (Washington University School of Medicine, St. Louis, MO) for preparation of hippocampal microislands.
| |
Footnotes |
|---|
Received January 29, 1998; Accepted May 18, 1998
1 Current affiliation: Surgical Operating Theater, Kyushu University Hospital, Fukuoka, 812-8582, Japan.
This work was supported by National Institutes of Health Grant GM47969 (D.F.C. and C.J.L.).
Send reprint requests to: Dr. Chris Lingle, Washington University School of Medicine, Dept. of Anesthesiology, Box 8054, St. Louis, MO 63110. E-mail: clingle{at}morpheus.wustl.edu
| |
Abbreviations |
|---|
HVA, high voltage activated;
(+)-ACN, (+)-3-hydroxy-5-androstane-17-carbonitrile;
IBa, Ba2+ current;
ICa, Ca2+ current;
ICl, Cl current;
DMSO, dimethylsulfoxide;
I-V, current-voltage;
PS, pregnenolone sulfate;
CgTx, conotoxin;
Aga, agatoxin;
DRG, dorsal root ganglion;
GDPS, guanosine-5'-O-(2-thio)diphosphate;
GTPS, guanosine-5'-O-(3-thio)triphosphate;
EGTA, ethylene
glycol bis(-aminoethyl
ether)-N,N,N',N'-tetraacetic
acid;
HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid.
| |
References |
|---|
|
Top
Summary
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
subunits.
Nature (Lond)
380:
255-258[Medline].-hydroxy-5-pregnan-20-one, and possible modes of action in mice.
Brain Res
415:
393-398[Medline].-conotoxin MVIIC.
J Neurosci
16:
2612-2623-aminobutyric acid-gated chloride currents in cultured postnatal hippocampal neurons.
Mol Pharmacol
42:
952-957[Abstract].1A calcium channel reflects similarities to neuronal Q- and P-type channels.
Proc Natl Acad Sci USA
91:
10575-10580.-OH, 5 -pregnan-20-one and 3 -OH,5 -pregnan-20-one in rodents.
Psychopharmacology
118:
65-71[Medline].This article has been cited by other articles:
|
|
 |
|
  S. M. Todorovic, S. Pathirathna, B. C. Brimelow, M. M. Jagodic, S.-H. Ko, X. Jiang, K. R. Nilsson, C. F. Zorumski, D. F. Covey, and V. Jevtovic-Todorovic 5{beta}-Reduced Neuroactive Steroids Are Novel Voltage-Dependent Blockers of T-Type Ca2+ Channels in Rat Sensory Neurons in Vitro and Potent Peripheral Analgesics in Vivo Mol. Pharmacol., November 1, 2004; 66(5): 1223 - 1235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Takei, H. Saegusa, S. Zong, T. Murakoshi, K. Makita, and T. Tanabe Anesthetic Sensitivities to Propofol and Halothane in Mice Lacking the R-Type (Cav2.3) Ca2+ Channel Anesth. Analg., July 1, 2003; 97(1): 96 - 103. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Todorovic, V. Jevtovic-Todorovic, S. Mennerick, E. Perez-Reyes, and C. F. Zorumski Cav3.2 Channel Is a Molecular Substrate for Inhibition of T-Type Calcium Currents in Rat Sensory Neurons by Nitrous Oxide Mol. Pharmacol., September 1, 2001; 60(3): 603 - 610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Todorovic, M. Prakriya, Y. M. Nakashima, K. R. Nilsson, M. Han, C. F. Zorumski, D. F. Covey, and C. J. Lingle Enantioselective Blockade of T-type Ca2+ Current in Adult Rat Sensory Neurons by a Steroid That Lacks gamma -Aminobutyric Acid-Modulatory Activity Mol. Pharmacol., November 1, 1998; 54(5): 918 - 927. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
