Department of Molecular and Cellular Pharmacology, University of
Miami School of Medicine, Miami, Florida
Mercuric chloride exerted a biphasic modulatory effect on rat neuronal
nicotinic acetylcholine receptors (nAChRs) expressed in Xenopus
laevis oocytes as heteromers of the 
3 or 4 and 
2 or
4 subunits. The degree of modulation was subunit-dependent, with
4-containing receptors displaying greater potentiation and 4-containing receptors displaying greater inhibition. Thus, 44 receptors displayed both robust potentiation and robust inhibition. During prolonged coapplication of HgCl2, first potentiation
then inhibition of the acetylcholine (ACh) response was observed. Upon coapplication of 1 µM HgCl2, a 2-fold increase in
ACh-induced current was achieved in 55 ± 1 s. With continued
HgCl2 application, the ACh response was slowly inhibited
until, after 5 min, less than 10% of the initial response remained. By
measuring potentiation at its peak and inhibition 5 min after the start
of HgCl2 coapplication, we obtained EC50 and
IC50 values of 262 ± 75 and 430 ± 72 nM, respectively. HgCl2 potentiation was voltage-dependent,
increasing at more positive holding potentials. Upon washout of mercury
chloride, potentiation reversed with a t1/2
of 4.6 min. Inhibition reversed more slowly, with less than half the
initial response recovered after 15 min of wash. Although free cysteine
residues are common targets for mercury, elimination of all free
cysteines located in the extracellular domains of the 4 and 4
subunits did not alter the effects of mercuric chloride. Potentiation
and inhibition of neuronal nAChRs may occur through action at a
transmembrane or cytoplasmic location after passive diffusion of
mercuric chloride across the plasma membrane.
 |
Introduction |
Nicotinic
acetylcholine receptors (nAChRs) belong to a superfamily of
neurotransmitter-gated ion channels that includes glycine, 
-aminobutyric acid (GABA), and 5-hydroxytryptamine receptors (Corringer et al., 2000
). Nicotinic receptors are found throughout the
central and peripheral nervous systems and at neuromuscular junctions.
Similar to neuromuscular nAChRs, neuronal nAChRs are thought to be
pentameric assemblies of subunits surrounding a central ion pore (Anand
et al., 1991; Cooper et al., 1991). To date, nine (2-10) and
three (2-4) subunits of neuronal nAChRs have been identified
(Corringer et al., 2000; Elgoyhen et al., 2001). Experiments examining
nAChRs natively expressed in the nervous system, as well as nAChRs
expressed in exogenous systems, suggest that neuronal nAChRs can be
formed as homomers of a single subunit type (e.g., 7), as simple
heteromers of two subunit types (e.g., 42), or as more complex
heteromers (e.g., 345) (Whiting et al., 1991; Flores et al.,
1992; Conroy and Berg, 1995; Chen and Patrick, 1997; Corringer et al.,
2000; Drisdel and Green, 2000). Although these subunits are homologous
with one another, the different receptor subunit combinations can vary considerably in their pharmacological and biophysical properties (Role,
1992; Corringer et al., 2000).
Inorganic and organic mercury compounds are ubiquitous environmental
contaminants and potent neurotoxic agents. The toxic effect of
mercurial compounds are due in part to the ability to penetrate into
the central and peripheral nervous systems through lipid layers and the
blood-brain barrier (Chang, 1977). Highly reactive methylmercury and
mercury (II) chloride (mercuric chloride) can interact with various
functional groups of many proteins, such as receptors, channels, and
enzymes, effectively disrupting excitable membrane function and
synaptic transmission within the central and peripheral nervous systems
(Sirois and Atchison, 1996). Several types of receptors and ion
channels have been reported to be directly affected by mercury
compounds. Methylmercury blocks the binding of ACh to the Torpedo
californica electric organ nAChRs and the binding of nicotine to
nAChRs expressed in rat brain (Eldefrawi et al., 1977). Methylmercury
also suppresses nicotinic responses in neuroblastoma cells (Quandt et
al., 1982). Mercuric chloride has been shown to suppress the function
of kainate receptors expressed in Xenopus laevis oocytes
(Umbach and Gundersen, 1989). Na+ and
Ca2+ channels are inhibited by mercuric chloride
(Shafer and Atchison, 1991; Hisatome et al., 2000). Mercuric chloride
has also been shown to interact with sulfhydryl groups in the
extracellular domain of GABAA receptors in rat
dorsal root ganglion neurons and potentiate GABA-induced currents
(Huang and Narahashi, 1996).
Several other heavy metals have been shown to modulate neuronal
nicotinic receptor function. Lead inhibits 34 and 42
receptors at submicromolar concentrations and potentiates 32
receptors at high micromolar concentrations (Zwart et al., 1995). Zinc
and cadmium potentiate and inhibit various neuronal nAChRs, depending on subunit combination and zinc concentration (Palma et al., 1998; Garcia-Colunga et al., 2001; Hsiao et al., 2001). To extend this work,
we examined the effect of mercury chloride on rat heteromeric neuronal
nAChRs expressed in X. laevis oocytes. We show that mercuric chloride exerts biphasic modulatory effects, both potentiating and
inhibiting, on neuronal nAChRs.
| |
Experimental Procedures |
Materials.
X. laevis frogs were purchased from
Nasco (Fort Atkinson, WI). The care and the use of X. laevis
frogs in this study were approved by the University of Miami Animal
Research Committee and meet the guidelines of the National Institutes
of Health. RNA transcription kits were obtained from Ambion (Austin,
TX). Collagenase B was purchased from Roche Diagnostics (Indianapolis, IN). Mercury (II) chloride was abtained from Aldrich Chemical Co. (Milwaukee, WI). All other reagents were from Sigma-Aldrich (St.
Louis, MO).
Site-Directed Mutagenesis.
The 4C75S mutation was
generated using the GeneEditor in vitro site-directed mutagenesis
system (Promega, Madison, WI) following protocols supplied by the
manufacturer. The mutation was confirmed by sequencing.
nAChR Expression in X. laevis
Oocytes.
m7G(5')ppp(5')G-capped cRNA
transcripts encoding nAChR subunits were prepared by in vitro
transcription from linearized template DNA encoding the rat 3,
4-1, 4-2, 2, 4, and 4C75S subunits. In this study,
unless otherwise noted, 4 refers to 4-1. Mature X. laevis frogs were anesthetized by submersion in 0.1%
3-aminobenzoic acid ethyl ester. The oocytes were surgically removed
and treated with collagenase B for 2 h at room temperature to
remove follicle cells. Stage V oocytes were manually injected with 2 to
10 ng of each cRNA in 13 to 50 nl of water and incubated at 18°C in modified Barth's saline (88 mM NaCl, 1 mM KCl, 2.4 mM
NaHCO3, 0.3 mM CaNO3, 0.41 mM CaCl2, 0.82 mM MgSO4,
100 µg/ml gentamicin, and 15 mM HEPES, pH 7.6) for 2 to 7 days.
Electrophysiological Recordings and Data Analysis.
Oocytes
were perfused at room temperature (20-25°C) with perfusion solution
(115 mM NaCl, 1.8 mM CaCl2, 2.5 mM KCl, 0.1 µM atropine, and 10 mM HEPES, pH 7.2) in a chamber constructed from 1/8-inch i.d. Tygon tubing. Perfusion was maintained at a constant rate
of ~20 ml/min. ACh and mercuric chloride were diluted in the
perfusion solution and then applied to oocytes using solenoid valves.
HgCl2 stock solutions were prepared fresh weekly.
Current responses were measured under two-electrode voltage clamp,
using a TEV-200 voltage-clamp unit (Dagan, Minneapolis, MN).
Micropipets were filled with 3 M KCl and had resistances of 0.3 to 2.0 M
. Current responses were sampled at 100 Hz, filtered at 20 Hz, and
recorded, stored, and analyzed on a Macintosh PowerPC 7100 computer
using Axodata 1.2.2 and AxoGraph 4.6 software (Axon Instruments, Inc.,
Union City, CA). All experiments, except voltage dependence studies,
were performed at a holding potential of 
70 mV.
A typical mercury chloride application experiment was performed as
follows. ACh alone was applied for 30 s followed by a 5-min coapplication of ACh and mercury chloride. At high concentrations of
mercury chloride (3 and 10 µM) where inhibition occurred more rapidly, the coapplication period was shorter. Subsequently, the oocytes were washed, and the response to ACh alone was measured at
various times during the washout period. When no desensitization was
evident (experiments involving 44), defined as a current decrease
of less than 5% over 30 s, control current in response to ACh
alone was determined from a 1-s average beginning 29 s after ACh
application. To measure potentiation, a 1-s average was taken at the
peak of potentiation (which occurred 10-120 s after the start of
mercury chloride coapplication, depending on mercury chloride
concentration; see Results) and compared with the control
current value. To measure inhibition, a 1-s average was taken 10 s
before the end of mercury chloride and ACh coapplication and compared
with the control current value.
When desensitization was evident (experiments involving 42,
32, and 34), a different analysis method was used to
determine the effects of mercury chloride. For each receptor subunit
combination, we first examined desensitization during a 5-min
application of ACh alone. The desensitizing phase of the response was
fit by a dual exponential decay equation. During mercury chloride
application experiments, the initial 30-s ACh response was fit to a
dual exponential decay equation, with the first exponential allowed to
vary during the fitting process and the second exponential fixed to the
value derived from the fit of the response to 5 min of ACh alone
(2.1 ± 0.1 min for 42, 1.2 ± 0.7 min for 32,
and 6.3 ± 0.8 min for 34; mean ± S.E.M.,
n = 3 in each case). The fit was projected over the
next 5 min, during which both ACh and mercury were applied. The degree
of modulation was measured by comparing a 1-s average in the presence
of mercury (at the peak of potentiation and 10 s before the end of
coapplication) with a 1-s average of the projected fit of the response
to ACh alone at the same time period.
Due to the slow reversibility of the effects of mercury, only a single
mercury application could be performed with each oocyte. We did not
observe any adverse effects of mercury application on the oocytes
during the recording period.
Because nAChRs differ in their agonist sensitivity, it was important to
ensure that the effects of mercuric chloride application were measured
with ACh concentrations that elicit a similar fraction of maximal
response. It was also important to use low concentrations of ACh to
minimize the effects of receptor desensitization. For each receptor, an
ACh concentration between the EC2 and
EC10 was chosen and used throughout the study (4 µM for 32, 17 µM for 34, 10 µM for 42, and 1 µM for 44 and 4-24C75S). The EC2 and EC10 values were calculated from previously
published data (Harvey et al., 1996) or, in the case of 4-24C75S,
from a concentration-response curve constructed as previously described
(Harvey et al., 1996).
Data from 44 mercury concentration-response data were analyzed by
fitting to concentration-response and concentration-inhibition equations I = Imax/(1 + (EC50/X)n) and
I = Imax/(1 + (X/IC50)n), respectively, where
I represents the current response at a given mercury
concentration X, Imax is
the maximal current, EC50 is the concentration of
mercury yielding half-maximal potentiation, n is the Hill
coefficient, and IC50 is the concentration of
mercury yielding half-maximal inhibition. Data are presented as
mean ± S.E.M. GraphPad Prism software (GraphPad Software, San
Diego, CA) was used to fit data and assess statistical significance
using a two-tailed unpaired t test.
| |
Results |
A Biphasic Effect of Mercuric Chloride on Neuronal nAChRs.
Simple rat heteromeric neuronal nAChRs were expressed in X. laevis oocytes by injection of cRNA transcripts encoding the 3 or 4 subunits with the 2 or 4 subunits. Receptors were first activated by a 30-s application of ACh (at or below the
EC10 for each receptor; see Experimental
Procedures), followed by a 5-min coapplication of
HgCl2 and ACh. The effect of mercury varied with receptor subunit composition. When coapplied with ACh, micromolar mercuric chloride concentrations exerted both potentiating and inhibiting effects on 44, 34, and 42. Inhibition, but
little or no potentiation, of 32 receptors was observed. Typical
current responses, showing the effect of mercuric chloride application on these receptors, are shown in Fig. 1.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 1.
Mercury chloride potentiates and inhibits neuronal
nAChRs. A, current responses of 44-expressing oocytes to 1 µM
ACh alone and during coapplication of 1 µM (left trace) and 10 µM
(right trace) HgCl2. B, current responses of
42-expressing oocytes to 10 µM ACh alone and during
coapplication of 1 µM (left trace) and 10 µM (right trace)
HgCl2. C, current responses of 34-expressing oocytes
to 17 µM ACh alone and during coapplication of 1 µM (left trace)
and 10 µM (right trace) HgCl2. D, current responses of
32-expressing oocytes to 4 µM ACh alone and during
coapplication of 1 µM (left trace) and 10 µM (right trace)
HgCl2. The current scale is indicated for each trace, and
the time scale is 1 min for all traces. Application of ACh and
HgCl2 is indicated by bars above each trace.
|
|
Potentiation developed first but was transient, with the peak of
potentiation followed by a gradual decrease in current, in some cases
to near baseline. For the 44 receptor where desensitization was
insignificant, potentiation was measured by comparing the peak of
potentiation to the ACh response immediately prior to mercury
application. Similarly, the extent of inhibition was measured by
comparing the current remaining near the end of mercury coapplication with the ACh response immediately prior to mercury application. For the
other receptors (42, 34, and 32), receptor
desensitization was taken into account (see Experimental
Procedures). Briefly, the initial response to ACh alone was fit to
a dual exponential decay equation and projected over the time period
during which both ACh and mercury were coapplied. The peak of
potentiation and extent of inhibition near the end of mercury exposure
could then be compared with the projected response to ACh alone.
The potentiating effects of 300 nM, 1 µM, and 10 µM
HgCl2 on the four different receptors are shown
in Fig. 2A. Receptors containing the 4
subunit (44 and 34) were potentiated more effectively than
2-containing receptors. Mercury chloride (1 µM) potentiated
44 receptors to 205 ± 17% and 34 receptors to
209 ± 30% of the response to ACh alone. 42 receptors were potentiated to 150 ± 3%, whereas 32 receptors displayed
almost no potentiation (106 ± 1%). The inhibiting effects of 300 nM, 1 µM, and 10 µM HgCl2 on the four
different receptors are shown in Fig. 2B. Receptors containing the 4
subunit were more potently inhibited than 3-containing receptors.
Mercury chloride (1 µM) application resulted in a nearly complete
block of the ACh responses of the 44 and 42 receptors
(7 ± 3% and 6 ± 1% of the response to ACh alone,
respectively). Inhibition of the 34 and 32 receptors by 1 µM mercury chloride was much less extensive (77 ± 6% and 78 ± 7% of the response to ACh alone, respectively). Our data suggest that potentiation of neuronal nAChRs by
HgCl2 is associated mainly with the 4 subunit,
whereas inhibition is associated mainly with the 4 subunit.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 2.
Potentiation and inhibition of neuronal nAChRs by
mercury chloride is dependent on subunit composition. A, potentiation
by 300 nM (white bars), 1 µM (hatched bars), and 10 µM (black bars)
mercuric chloride was measured at the peak of potentiation and is
presented as a percentage of the response to ACh alone (mean ± S.E.M., n = 3-4). Compared with the potentiation
of 44 receptors by 1 µM HgCl2, potentiation of
34 receptors by 1 µM HgCl2 was not significantly
different (p > 0.05), whereas potentiation of
42 and 32 were significantly different
(p < 0.05 and p < 0.01, respectively). B, inhibition by 300 nM (white bars), 1 µM (hatched
bars), and 10 µM (black bars) mercuric chloride was measured as
described under Experimental Procedures and is presented
as percentage of the response to ACh alone (mean ± S.E.M.,
n = 3-4). Compared with the inhibition of 44
receptors by 1 µM HgCl2, inhibition of 42 receptors
by 1 µM HgCl2 was not significantly different
(p > 0.05), whereas inhibition of 34 and
32 were significantly different (p < 0.001).
|
|
Because the 44 receptor displayed both robust potentiation and
robust inhibition, we used this receptor for more detailed investigation of the effects of mercury chloride during the remainder of this study. As shown in Fig. 3, we
measured the potentiation and inhibition of 44 over a wide range
of mercury chloride concentrations to obtain an
EC50 for potentiation of 262 ± 75 nM
(nH = 1.6 ± 0.6) and an
IC50 for inhibition of 430 ± 72 nM
(nH = 2.3 ± 0.9).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 3.
Modulation of 44 receptors by mercury chloride.
A, potentiation of 44 receptors by a range of mercuric chloride
concentrations was measured at the peak of potentiation and is
presented as a percentage of the response to ACh alone. Each point is
the mean ± S.E.M. of three to six oocytes. The data were fit as
described under Experimental Procedures. B, inhibition
of 44 receptors by a range of mercuric chloride concentrations
was measured as described under Experimental Procedures
and is presented as percentage of the response to ACh alone. Each point
is the mean ± S.E.M. of three to five oocytes. The data were fit
as described under Experimental Procedures.
|
|
The potentiating effect of low concentrations of mercury chloride
developed quite slowly. This is particularly evident for 44
receptors, where potentiation by 1 µM mercury chloride reached a
maximum 55 ± 1 s (n = 3) after initiation of
mercury application (Fig. 1A). This slow development of potentiation
was not an artifact of our perfusion system. The initial response to
ACh alone develops much more quickly, reaching a maximum within a few
seconds. Also, when additional ACh is coapplied instead of mercury
chloride, the increase in current in response to the increased
(doubled) ACh concentration is as fast as the initial ACh response
(data not shown). At a lower mercury chloride concentration (300 nM), the development of potentiation was even slower, reaching a peak in
118 ± 14 s (n = 3). At a higher mercury
chloride concentration (10 µM), the development of potentiation was
more rapid, reaching a peak in 10 ± 1 s (n = 3).
The Effects of Mercury Chloride Reverse Slowly.
While
conducting the experiments shown in Figs. 1 through 3, we found that
the effects of mercury were only slowly reversible. This necessitated
discarding each oocyte after a single mercury application. In Fig.
4, we examine this slow reversibility in more detail.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4.
Slow reversal of potentiation and inhibition of
44 receptors by mercury chloride. A, oocytes were exposed to 1 µM mercury chloride for 30 s or 5 min in the presence or absence
of 1 µM ACh and were then washed for 5 min. Responses to 1 µM ACh
alone after the wash are presented as a percentage of the response to
ACh alone prior to mercury chloride exposure (mean ± S.E.M.,
n = 3-5). B, responses to 1 µM ACh alone were
measured at various times during the washout, after a 30-s exposure to
1 µM mercury chloride and ACh, and are presented as a percentage of
the response to ACh alone prior to mercury chloride exposure (mean ± S.E.M., n = 6). The data were fit to a single
exponential decay equation.
|
|
To examine the persistence of mercury potentiation of the 44
receptors, we first applied ACh for 30 s, followed by a 30-s coapplication of 1 µM mercury chloride and ACh. The oocytes were then
washed for 5 min. The response to ACh alone measured after the wash was
141 ± 4% of the ACh response preceding the mercury incubation
(Fig. 4A). By measuring the ACh response after washes of various
lengths, we obtained an off rate of 0.15 min
1
and a half-time for loss of potentiation of 4.6 min (Fig. 4B).
The slow reversibility of mercury inhibition was also examined (Fig.
4A). The oocytes were exposed to ACh alone for 30 s, followed by a
5-min coapplication of ACh and 1 µM mercury chloride. At the end of
the coapplication period, the response was 4.8 ± 2.2% of the
response to ACh alone. After a 5-min wash, the response to ACh
recovered to 38 ± 12% of the response before mercury chloride application. An additional 10-min wash period did not yield additional recovery of the ACh response amplitude (data not shown).
Because the concentration of ACh used in these experiments was low (1 µM, the EC2 for 44), it was unlikely that
the presence of ACh was required for mercury to exert its modulatory
effects. However, the slow reversibility of both potentiation and
inhibition offered the opportunity to test this idea (Fig. 4A).
Following measurement of a response to ACh alone, 44-expressing
oocytes were incubated with 1 µM mercury chloride in the absence of
ACh for 30 s or 5 min. The oocytes were then washed for 5 min, and the response to ACh alone was measured again. ACh currents of oocytes
exposed to mercury for 30 s remained potentiated (126 ± 5%
of the response to ACh before mercury incubation), whereas those
exposed to mercury for 5 min remained inhibited (34 ± 6% of the
response to ACh before mercury incubation). Thus, the presence of ACh
(and open channels) is not required for mercury chloride to potentiate
and inhibit neuronal nAChRs.
Potentiation by Mercury Chloride Is Voltage-Dependent.
We
measured potentiation of 44 receptors by 1 µM mercury chloride
at several different holding potentials (Fig.
5). At a holding potential of 40 mV,
the ACh response was potentiated to 258 ± 31% of the response to
ACh alone. At 90 mV, potentiation was 180 ± 3% of the response
to ACh alone.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 5.
Potentiation of 44 receptors by mercury
chloride is voltage-dependent. The potentiating effect of 1 µM
HgCl2 on 44 receptors at several holding potentials
was measured at the peak of potentiation and is presented as a
percentage of the response to ACh alone (mean ± S.E.M.,
n = 3).
|
|
The Effects of Mercuric Chloride Are Not Dependent on Extracellular
Free Cysteine Residues.
Inorganic and organic mercury compounds
are highly reactive toward the sulfhydryl groups of cysteine residues
(Webb, 1966). Therefore, we tested the potential role of the
extracellular free cysteine residues of 4 and 4 subunits in
mediating the effects of mercury. The 4 and 4 subunits each have
a single extracellular free cysteine residue. As the result of
alternative RNA splicing, the 4 subunit exists in two forms, 4-1
and 4-2 (Goldman et al., 1987). In this study, we used 4-1, which
has an extracellular, C-terminal cysteine residue (position 594). The
4-2 subunit is identical to 4-1 except for the last few residues.
The C-terminal Ala-Cys of 4-1 is replaced with Gly-Met-Ile in 4-2
(Goldman et al., 1987). The 4 subunit contains a free cysteine at
position 75 (Duvoisin et al., 1989). To test the role of these two
cysteine residues, we examined the effects of mercury chloride on the
receptor formed by 4-2 and a mutant 4 subunit (4C75S). Typical
current responses, showing the effect of 1 and 10 µM mercuric
chloride on the 4-24C75S receptor, are shown in Fig.
6A. The extent of potentiation and
inhibition by 1 µM mercury chloride, measured with several oocytes,
is shown in Fig. 6, B and C. Potentiation and inhibition of the
4-24C75S receptor by mercury chloride were not significantly
different from potentiation and inhibition of the 4-14 receptor.
Thus, extracellular free cysteine residues are not required for mercury
chloride to potentiate and inhibit neuronal nAChRs.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 6.
Extracellular free cysteine residues are not required
for potentiation and inhibition of 44 by mercury chloride. A,
current responses of 4-24C75S-expressing oocytes to 1 µM ACh
alone and during coapplication of 1 µM (left trace) and 10 µM
HgCl2 (right trace). The current scale is indicated for
each trace, and the time scale is 1 min for both traces. B,
potentiation by 1 µM mercuric chloride was measured at the peak of
potentiation and is presented as a percentage of the response to ACh
alone (mean ± S.E.M.). When compared with the potentiation of
44 receptors (black bar, 205 ± 17%, n = 3), potentiation of 4-24C75S receptors by 1 µM
HgCl2 (white bar, 234 ± 2%, n = 3) was not significantly different (p > 0.05). C,
inhibition by 1 µM mercuric chloride was measured as described under
Experimental Procedures and is presented as percentage
of the response to ACh alone (mean ± S.E.M.). When compared with
the inhibition of 44 receptors (black bar, 7 ± 3%,
n = 3), inhibition of 4-24C75S receptors by 1 µM HgCl2 (white bar, 7 ± 2%, n = 3) was not significantly different (p > 0.05).
|
|
| |
Discussion |
Mercuric chloride potentiates and inhibits neuronal nAChRs in a
concentration-, subunit-, and time-dependent manner. Receptors containing the 4 subunit were potentiated to a greater extent than
2-containing receptors, whereas 4-containing receptors were
inhibited to a greater degree than receptors containing the 3
subunit. Thus, the 44 receptor was strongly potentiated and inhibited, whereas the 32 receptor displayed little or no
potentiation and only moderate inhibition. When 1 µM mercury chloride
was coapplied with ACh to 44 receptors, potentiation developed
slowly, reaching a maximum 55 ± 1 s after application.
Inhibition of the ACh response then developed over a period of several
minutes. Potentiation and inhibition in response to mercury chloride
concentrations <1 µM developed more slowly, whereas potentiation and
inhibition in response to concentrations >1 µM developed more
quickly. As we discuss below, one possible explanation for this slow
development of potentiation and inhibition is that the site of action
may be on the cytoplasmic face of the receptor, requiring mercury chloride to diffuse across the membrane before acting. The increasingly rapid development of potentiation and inhibition observed as the concentration of mercury chloride was increased is presumably due to
potentiating, and especially inhibiting, concentrations being achieved
more quickly. A possible explanation for the observation that
potentiation always precedes inhibition is that the potentiation site
is more sensitive to mercury chloride than the inhibition site. Thus,
during mercury chloride application, potentiating concentrations would
be achieved before inhibiting concentrations.
Prior to this study, there has been little information about the
effects of mercurial compounds on neuronal nAChRs. Eldefrawi et al.
(1977) reported that 10 µM methylmercury inhibited binding of
[3H]nicotine to rat brain membranes by 60%.
Methylmercury (10 µM) was also shown to suppress depolarizing
responses to nicotine in N1E-115 neuroblastoma cells (Quandt et al.,
1982). More detailed information is available regarding the effects of
mercurials on other ligand-gated ion channels. Both potentiation and
inhibition of GABAA receptors by mercury chloride
have also been reported. Brief application (several seconds) of 1 and
10 µM mercury chloride potentiated, whereas 100 µM mercury chloride
suppressed, GABA-induced currents in rat dorsal root ganglion neurons
(Arakawa et al., 1991). Prolonged exposure (several minutes) to 10 µM
mercury chloride resulted in reduction of GABA-induced currents in
these neurons (Huang and Narahashi, 1996). Mercury chloride also caused
slowly reversible inhibition of human kainate receptors expressed in X. laevis oocytes, with an IC50 of 70 nM (Umbach and Gundersen, 1989).
The effects of mercury chloride on neuronal nAChRs reversed slowly,
persisting for several minutes after washing with mercury-free solution
(Fig. 4). If mercury chloride was applied for 30 s, achieving near
maximal potentiation, the receptors remained potentiated for several
minutes after washout of the mercury. Potentiation reversed with a
t1/2 of 4.6 min. When mercury chloride
was applied for 5 min, achieving maximal inhibition, the receptors then
remained inhibited for at least 15 min after washout of the mercury.
Mercury chloride has been shown to have slowly reversible, or in some cases, apparently irreversible, effects on other channels and receptors. The reduction of GABA-induced current responses on rat
dorsal root ganglion neurons after prolonged coapplication of GABA and
mercury chloride was only slightly reversed after a 15-min wash (Huang
and Narahashi, 1996). Mercuric chloride block of
K+ channels in human B-lymphocytes was also
slowly reversible (Gallagher et al., 1995), and apparently irreversible
block of Ca2+ channels has been observed with
bovine chromaffin cells and rat dorsal root ganglion neurons (Pekel et
al., 1993; Weinsberg et al., 1995).
Organic and inorganic mercury compounds are well known for their strong
affinity for the sulfhydryl groups of cysteine residues (Webb, 1966;
Sirois and Atchison, 1996; Hisatome et al., 2000). The potent
neurotoxic effects of mercurials have been attributed in part to their
ability to bind and form strong complexes with free cysteine residues
of various receptors and ion channels, causing modulation or disruption
of their function (Sirois and Atchison, 1996). Mercuric chloride has
been shown to modulate the function of GABAA
receptors, D2 dopamine receptors, and sodium channels through binding to one or more cysteine residues (Scheuhammer and Cherian, 1985; Huang and Narahashi, 1996; Hisatome et al., 2000).
Mercuric chloride-cysteine interactions in glutamate receptors result
in enhanced agonist binding affinity (Terramani et al., 1988). In
several studies, chemical modification of cysteine residues has been
shown to effectively abolish the modulating effects of mercury
(Scheuhammer and Cherian, 1985; Huang and Narahashi, 1996). With these
studies in mind, we tested the potential involvement of the only two
extracellular free cysteine residues present in 4 and 4 (4C594
and 4C75) in mediating the effects of mercury chloride on 44
neuronal nAChRs. We found that these two cysteine residues are not
required for mercury chloride to modulate the receptor (Fig. 6).
Although interaction with the sulfhydryl group of cysteine is most
common, it is possible that mercury chloride is interacting with other
residues in the extracellular domain of the receptor. Mercurials can
interact with the imidazole nitrogens of histidines. However, it is
unlikely that an extracellular histidine residue is involved in
mediating the effects of mercury chloride, because lowering the pH from
7.2 to 6.0 (thus changing the degree of protonation of imidazole
nitrogens) had no effect on either potentiation or inhibition (data not
shown). Interaction with other extracellular residues is also possible,
but as we discuss below, mercury chloride may be interacting with the
receptor within its transmembrane or cytoplasmic domains.
When examining the effects of mercury in a biological system, it is
important to consider the various mercury species present in solution.
In physiological saline solutions, the interaction of a protein with
mercuric chloride is unlikely to involve an interaction with
Hg2+ cations but rather with one or more of a
variety of neutral and negatively charged mercury complexes (Hietanen
and Sillen, 1952; Webb, 1966; Huang and Narahashi, 1996).
Hg2+ cations have high affinity for and form
strong complexes with chloride and hydroxide ions in aqueous solution.
As a result, the composition of mercury complexes is pH and chloride
concentration-dependent. Only at very low chloride concentrations (1 nM
or less) and acidic pH are Hg2+ cations a
predominant species. In our solutions (containing 121 mM
Cl), mercury is present mainly as a mixture of
HgCl42 and
HgCl4(OH)3, with lesser
amounts of HgCl2 and HgCl
(Webb, 1966; Gutknecht, 1981). In a study examining mercuric chloride potentiation of GABAA receptors, these neutral
and negatively charged mercury complexes were shown to be much more
potent than Hg2+ in causing potentiation (Huang
and Narahashi, 1996). Thus, it is likely that one or more of these
neutral and negatively charged mercury species is interacting with the
neuronal nAChRs in our experiments.
Our data suggest that the interaction between mercury chloride and the
receptor might not be in the extracellular domain but may be in either
the transmembrane or cytoplasmic domains of the receptor. The mild
voltage dependence of potentiation seen in Fig. 5, with potentiation
increasing at more depolarized potentials, suggests that a negatively
charged species of mercury might be interacting with the receptor near
or within the electric field of the membrane. However, open channels
are not required for this interaction because mercury chloride
initiated both potentiation and inhibition in the absence of ACh (Fig.
4A). It is also possible that mercury chloride interacts with the
receptor in the cytoplasmic domain. An interaction with intracellular
sulfhydryls after entry through Na+ and
Ca2+ channels was proposed to underlie the
irreversible toxicity of mercury at motor nerve terminals (Miyamoto,
1983). Again, because the effects of mercury chloride in our
experiments could occur in the absence of ACh, entry through open
channels is not required. However, in our solutions, as much as a third
of the mercury exists in an uncharged complex with chloride (Webb,
1966; Gutknecht, 1981). HgCl2 has been shown to
diffuse easily across lipid bilayers (Gutknecht, 1981) and would
equilibrate with the oocyte cytoplasm within a few seconds. Once in the
cytoplasm, the mercury would re-equilibrate among the various neutral
and negatively charged complexes. One or more of these species may then
interact with the cytoplasmic domain of the receptor. Both 4 and
4 have numerous cysteine residues in their cytoplasmic domains that
are potential targets for the mercury complexes. The ability of
HgCl2 to move quickly across the cell membrane
would allow the rapid potentiation and inhibition that we observed at
high mercury chloride concentrations. More difficult to explain is the
slow time course of potentiation and inhibition at low mercury chloride
concentrations. It may be that cytoplasmic proteins within the oocyte
have a considerable binding capacity for mercury chloride, slowing the
approach to active concentrations. It also may be that the target
residues on the receptor are relatively inaccessible. Identification of the sites of interaction between mercury chloride and neuronal nAChRs,
whether in the transmembrane or cytoplasmic domains, can be approached
through a more extensive mutagenesis program. These sites may represent
a new target for drug development.
This work was supported by a grant (to C.W.L.) from the
National Institute on Drug Abuse (DA08102). A.M. was supported by a
postdoctoral fellowship from the American Heart Association, Florida/Puerto Rico Affiliate.
![[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}
Top
Abstract
Introduction
