| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1-mediated by nonselective cation
channels
1 The Wenner-Gren Institute, The Arrhenius Laboratories F3, Stockholm University, SE-106 91 Stockholm, Sweden; and 2 Klinik für Neurologie, Universität Magdeburg, D-39120 Magdeburg, Germany
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
The
nature of the sustained norepinephrine-induced depolarization in brown
fat cells was examined by patch-clamp techniques. Norepinephrine (NE)
stimulation led to a whole cell current response consisting of two
phases: a first inward current, lasting for only 1 min, and a sustained
inward current, lasting as long as the adrenergic stimulation was
maintained. The nature of the sustained current was here investigated.
It could be induced by the 
1-agonist cirazoline but not
by the 
3-agonist CGP-12177A. Reduction of extracellular
Cl
concentration had no effect, but omission of
extracellular Ca2+ or Na+ totally eliminated
it. When unstimulated cells were studied in the cell-attached mode,
some activity of 
30 pS nonselective cation channels was observed. NE
perfusion led to a 10-fold increase in their open probability (from
0.002 to 0.017), which persisted as long as the perfusion was
maintained. The activation was much stronger with the
1-agonist phenylephrine than with the
3-agonist CGP-12177A, and with the Ca2+
ionophore A-23187 than with the adenylyl cyclase activator forskolin. We conclude that the sustained inward current was due to activation of
30 pS nonselective cation channels via 1-adrenergic
receptors and that the effect may be mediated via an increase in
intracellular free Ca2+ concentration.
cirazoline; calcium; patch-clamp technique
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
NOREPINEPHRINE HAS WIDE-RANGING EFFECTS on brown fat cells, spanning from acute stimulation of thermogenesis to regulation of cellular proliferation, growth, and differentiation (for review, see Ref. 19). Among the least understood effects of norepinephrine on brown fat cells are those on the cell membrane potential. That norepinephrine has such effects is well documented (5-7, 9, 10, 15, 30). However, no definite physiological role (in, for example, thermogenesis) has as yet been ascribed to these membrane potential alterations.
There is general agreement that three phases can be discerned in the
brown fat cell membrane response to adrenergic stimulation (as earlier
summarized by Refs. 2 and 8). 1) The first phase, a rapid
depolarization, is mediated via 1-adrenergic
receptors and results from a transient Cl efflux
mediated by an increase in intracellular free Ca2+
concentration ([Ca2+]i) (4, 6, 15,
21). 2) The second phase, a short-lived hyperpolarization, is also evoked (directly or indirectly) via an
1-adrenergic pathway and results from the activation of
apamin-sensitive Ca2+-activated K+ channels
(13, 15, 17, 18) and probably also of voltage-sensitive K+ channels (14, 23). 3) The third
phase, the sustained depolarization, has until now generally been
suggested to be mediated via a -adrenergic pathway (9,
15); the channels involved have until now not been identified.
The sustained depolarization was observed already in the first
recording of cell membrane potentials in brown fat cells, performed
with sharp penetrating microelectrodes (7), and it has
been observed in several subsequent studies with such techniques
(5, 6, 9, 10, 30) but not with patch-clamp techniques.
Because the nature of this sustained, and therefore potentially most
important, phase of the membrane depolarization events has remained
unresolved, we have attempted here to characterize it with patch-clamp
techniques, both at the cellular and at the single-channel level.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Cell Isolation and Maintenance
The brown fat cells used for this study were prepared principally as earlier described (11, 12, 23, 26, 28). Young male (80-200 g) Sprague-Dawley rats (reared at 21°C with free access to food and water) were killed by CO2 followed by decapitation. The interscapular fat depots were excised, and the pooled and minced brown fat depots were dispersed into single cells by collagenase digestion (Sigma, type II, 5 mg/ml) for 35 min in extracellular solution (see below) in a shaking water bath at 37°C. The floating cells (i.e., the mature brown fat cells) were separated from the collagenase suspension by centrifugation (150 g, 10 min) and kept in 6-well dishes in an incubator at 37°C, in an 8% CO2-92% air atmosphere in DMEM, supplemented with newborn-calf serum (10% vol/vol), insulin (4 nM), Na-ascorbate (25 µg/ml), glutamine (4 mM), penicillin (50 IU/ml), and streptomycin (50 µg/ml). After addition of the cells to a well, a Heraeus-Biofoil-25 membrane was placed on the surface of the medium. The floating brown fat cells spontaneously adhered to the hydrophilic side of this membrane. The medium was exchanged after 24 h, and the cells on the membrane were then examined, as we will describe. Routinely, cells were examined during the first 1-3 days, but occasionally they were kept in the well for as many as 7 days; we did not observe any difference in response during this time, nor did the cells change their multilocular appearance.Patch-Clamp Techniques
For both the perforated-patch studies and the cell-attached-mode studies, the biofoil with the attached brown fat cells was placed in a chamber with a volume of300 µl and was perfused continuously with
extracellular solution (see below) at room temperature (22-25°C)
with a flow rate of ~300 µl/min; the perfusate was removed at the
same rate by a peristaltic pump. The perfusion system consisted of a
4-channel electromagnetic valve-controlled local application holder,
fed by gravity. The extracellular solution consisted of (in mM): 134 NaCl, 6 KCl, 1.2 MgCl2, 1.2 CaCl2, 5 glucose,
and 10 HEPES (pH 7.4 adjusted with NaOH, corresponding to 5 mM
Na+). Where indicated, modified solutions were used, and/or
agents were dissolved in this medium.
Pipettes were pulled from borosilicate glass and had resistances
between 6 and 3 M
. The pipette solution consisted of (in mM): 60 KCl, 80 K gluconate, 10 NaCl, 1 CaCl2, 1 EGTA, and 10 MOPS
(pH 7.2 adjusted with KOH, corresponding to 5 mM K+).
The choice of the rather high level of Cl was based on
implications from 36Cl efflux experiments
indicating that the cytosolic Cl level is higher than the
equilibrium level for the resting membrane potential (4)
(which would be ~50 mM) and from experiments (21)
showing that the Cl currents induced by norepinephrine
stimulation had a reversal potential of 
20 mV, corresponding to an
estimated cytosolic concentration of 70 mM. The free
Ca2+ concentration in the pipette was calculated to be 10 µM, according to the computer program BAD3; the presence of 80 mM
gluconate, which may have Ca2+-chelating properties, may
have lowered this level further. To the pipette solution,
0.24-0.32 µg of amphotericin B (dissolved in DMSO) was added per
milliliter. In a few cases, pluronic F-127 (0.08%) was also added to
the pipette solution (to facilitate perforation), but no clear effects
were seen. The reference electrode was an AgCl-pellet electrode
connected to the chamber through a 150 mM NaCl agar bridge (50 k).
A few minutes after a giga seal had formed, the membrane potential was
read off in the current-clamped mode. The voltage read off was about
20 mV, i.e., identical to the zero-current voltage determined in the
voltage ramps (see below). The pipette was then voltage-clamped to 50
mV, and the capacitance (5-7 pF) of the pipette was electronically
compensated. Access resistance was monitored from frequent 40-ms 5-mV
hyperpolarizing steps. The recording was started after 15-35 min,
when the access resistance had stabilized and had usually dropped below
50 M. Access resistance was not compensated in the results shown.
The junction potential of the pipette solution vs. the standard
extracellular solution was experimentally determined to be +6 mV; i.e.,
the potential is 6 mV more negative than that indicated on the axes.
The junction potentials in the buffers with altered Cl
concentration were not markedly different. Routinely, the cell capacitance in the resting state (30 pF) was compensated by the circuitry in the amplifier, and no further capacitance compensation was
performed during the experiment. Each cell responded qualitatively in
the same way to repeated agonist stimulation for up to 2 h, but in
general there was a tendency to a decrease in the magnitude of the
responses on successive stimulations.
For the cell-attached-mode studies, the pipettes had resistances
between 8 and 12 M, and the pipette medium was here identical to the
extracellular solution. Each experiment was finalized by excising the
patch and studying it as an inside-out patch; these excised patches
were thus formed and studied in the extracellular solution.
Recordings and Data Analysis
The currents were recorded with an L/M EPC 7 patch-clamp amplifier. During the whole cell experiments, voltage ramps were frequently run under manual initiation. In control experiments (not shown), we found that the current at each voltage tested was stable with time, except for those at very high depolarizations (more positive than 0 mV), which became inactivated, in accordance with our earlier observations (23). Therefore, we surmise that the voltage-dependent conductance changes observed during voltage ramps represented time-independent values. In each voltage ramp, the holding potential of50 mV was increased to 100 mV for 10 ms; during the
next 600 ms, it was linearly decreased to +20 mV and then directly
returned to 50 mV. The command voltage ramps were generated and saved
together with the resulting currents by the Clampex program and were
later analyzed by the Clampfit of the pCLAMP 6.03 program. The data
were used for calculation of cellular zero current potentials (i.e.,
membrane potentials) and conductances. Conductances at 50 mV
(Gm,50) were calculated from linear fits
between 60 and 40 mV, and conductances at +15 mV
(Gm,+15) were calculated from linear fits
between +10 and +20 mV. Agonist-induced current characteristics were
determined from voltage ramp-induced currents in the presence of
agonist minus the currents in the absence of agonist, as indicated.
After the current signal had been digitized with a modified pulse code modulator, the data were stored on video cassettes. For the later off-line analysis, the current signal from the cassettes was in the
whole cell studies sampled at a frequency of 1 or 3 kHz and was
low-pass filtered with a 3-dB frequency of 200 Hz by an 8-pole filter
with Bessel characteristics. In the cell-attached mode, the current
signal from the cassettes was sampled at a frequency of 5 kHz and was
low-pass filtered with a 3-dB frequency of 0.5 kHz. The signals were
transferred by a 12-bit interface to a computer and analyzed with the
pCLAMP 6.03 program (Axon Instruments).
In the recordings in the cell-attached mode, the number of
simultaneously open channels was low (1-3), making it
possible routinely with the pCLAMP program to calculate the open and
closed times and the observed open probability
(Po,obs) directly from lists of events,
according to a 50% amplitude threshold crossing criterion. To obtain
the Po,obs values, data were collected over 
60
s in the control state (to yield Po,obs,basal)
and 240 s during stimulation (to yield
Po,obs,stim). Each observation was finalized with the excision of the patch, which led to the spontaneous activation of several channels. The maximal number of channels observed in the
excised inside-out patch was denoted Ntot. The
Po,tot values were obtained by dividing the
Po,obs values by the
Ntot/Nobs value. The
current amplitudes of the single channels were calculated from
amplitude histograms fitted to Gaussian distributions by the pCLAMP program.
In the studies in the cell-attached mode, potentials applied to the pipette are stated as holding potentials, relative to the rest of the outside of the cells (because the true cell membrane potential was not simultaneously measured). In the displayed recordings, downward deflections denote currents corresponding to cellular inward currents.
Chemicals
EGTA, K-gluconate, L-norepinephrine bitartrate, Na-aspartate, and L-phenylephrine HCl were directly dissolved in the extracellular solution, and amphotericin B, forskolin, and A-23187 were stock-dissolved in DMSO (0.1% in final solution); all were from Sigma. CGP-12177A was a gift from Ciba-Geigy, cirazoline was from Research Biochemicals International, and pluronic F-127 (dissolved in DMSO) was from Molecular Probes. The adrenergic agonists were protected from light in the perfusion system.| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
In the present experiments, the response of brown fat cells to sustained adrenergic stimulation was analyzed first as whole cell currents, with the perforated-patch technique. To directly identify the channels responsible for the currents observed, the cell-attached patch-clamp technique was subsequently used.
Whole Cell Currents
In the perforated-patch whole cell studies, the cell membrane potential was clamped at50 mV (except during the voltage ramps); this is a voltage similar to that observed in resting brown fat cells
at 37°C (5, 30) but somewhat hyperpolarized when
compared with the actual resting potential of the brown fat cells
studied under the present conditions (see below).
The pipette solution was composed so as to be close to expected
cytosolic ion levels. Because amphotericin makes the membrane permeable
to Na+, K+, and Cl ions
(22), the monovalent ions from the pipette probably
diffused sufficiently to in reality determine the intracellular levels. [Because amphotericin does not make the patch permeable to
Ca2+ (22), the pipette Ca2+
concentration was not expected to influence cytosolic Ca2+
levels.] The expected equilibrium potentials were therefore 82 mV
for K+, +68 mV for Na+, and 18 mV for
Cl.
Whole cell currents were determined when a stable access membrane
resistance had been attained. The resting cell capacitance was 35 ± 1 pF (n = 26). During the experiments, we frequently followed the current response to voltage ramps (as is seen, for example, in Fig. 1). These ramps allowed
for determination of membrane potential (i.e., the zero current
potential) and were also used to measure total cellular conductance at
50 mV (referred to as Gm,50) (i.e., the
conductance at the clamped potential) and at +15 mV
(Gm,+15).
|
The unstimulated state.
In unstimulated cells, a stable resting current (Fig. 1A)
was observed. On the basis of voltage ramps performed during this period (as exemplified in Fig. 1B, current 1),
the mean resting membrane potential was estimated to be 21 mV (Table
1); this was also the voltage observed in
the current clamp mode immediately before the setup was switched to the
voltage clamp mode. The membrane conductance at the holding potential
of 50 mV (Gm,50) was 0.7 nS (Table 1). As is
also seen from the current ramp in Fig. 1B, the membrane
conductance at positive membrane potentials was much higher; at +15 mV
(i.e., Gm,+15), it was 14 nS.
|
Effects of norepinephrine. After initiation of perfusion with 1 µM norepinephrine, the cells responded with characteristic successive alterations in transmembrane current, despite the constant presence of norepinephrine. Two major response types could be distinguished. In ~75% of the cells, a relatively simple type of response, such as that illustrated in Fig. 1, was observed. In the other cells, an oscillatory response was observed (not shown). The present analysis concerns only the response pattern of the majority of cells.
CURRENTS. In the current response elicited by norepinephrine, there were two phases (Fig. 1). Both phases consisted of inward currents, i.e., both currents would probably be depolarizing in the unclamped cell. The "first inward current" (referred to throughout in this way) was transient, with a rapid initiation and a rather slow inactivation. It was large, with a mean maximal current value of 367 pA (Table 1). The mean total length of this current phase, i.e., the time from the current deviation from resting level to its attainment of new plateau corresponding to the next phase, was 58 ± 5 s (n = 13). This current was followed by a "sustained inward current" (referred to throughout in this way), which persisted unabated and nonoscillatory for as long as the norepinephrine stimulation lasted. This sustained inward current amounted to 44 pA. When norepinephrine was washed out, the cell membrane current returned to the initial resting levels. Because the sustained phase (which is the one under investigation here) of necessity always was preceded by the first inward current, our results concerning the first inward current are also briefly reported here, but a full analysis is not attempted. MEMBRANE POTENTIALS. On the basis of the voltage ramps, the alterations in membrane potential (Em) under the present conditions and in conductances were followed during the entire response. Membrane potential changes are depicted in Fig. 2A and summarized in Table 1. As seen, after norepinephrine had reached the cell, a depolarization occurred during the first inward current, from21 mV down to 6 mV,
i.e., a depolarization of ~15 mV. After this,
Em slowly but steadily repolarized, until it
reached and became stable at about 13 mV, i.e., still ~8 mV
depolarized compared with the resting state. Em
fully returned to initial values upon norepinephrine washout (Table 1).
|
50 mV
(Gm,50). However, maximal depolarization was
obtained already at the time of the first ramp, when the increase in
conductance was only about threefold, and the further increase in
conductance (seen here during the subsequent three voltage ramps) was
actually occurring while the membrane potential was repolarizing toward
the resting level. During the sustained inward current, the
Gm,50 was still twice as high as it was in the
resting state. Gm,50 remained elevated until
norepinephrine had been washed out.
The conductance at +15 mV (Gm,+15), which
probably results mainly from voltage-gated K+ channels
(14, 23), was marginally altered during the norepinephrine response (Fig. 2C). Although not clearly evident from Fig.
2C, Gm,+15 was significantly
increased during the first inward current (Table 1). There was a
tendency toward a reduced Gm,+15 during the
sustained inward current in many experiments (as is evident in Fig.
2C) (Table 1).
VOLTAGE RAMP CURRENTS.
For a first analysis of the ionic character of the changes in membrane
conductance, the character of the currents during the voltage ramps was studied.
As seen in Fig. 1B , the currents of the ramp obtained in
the resting state (trace 1) reversed at fairly low
potential, here at about 18 mV, i.e., far from the expected
K+ equilibrium potential of 83 mV. At positive membrane
potentials, large outward currents of the type earlier described for
voltage-gated K+ channels in these cells (14,
23) were observed.
The norepinephrine-induced currents were obtained by subtracting this
resting-state voltage ramp current from the subsequent ones. In Fig.
1C, a family of norepinephrine-induced voltage ramp currents
obtained during the first inward current is shown; these curves thus
represent 
-currents. These induced-current curves were
complex, implying that the induced currents consisted of several ionic
fluxes, and they were not further analyzed here. The induced curves
(i.e., the -curves) obtained during voltage ramps during the
sustained inward current phase (exemplified in Fig. 1D) were
comparably simple to analyze. The curves were time independent (not
shown) and close to ohmic when negative potentials were applied. Zero
current was reached between 10 mV and 0 mV [mean value 0.4 ± 1.1 mV (n = 15)]. That the zero current potential was
close to 0 mV eliminated the possibility that this current represented
currents through specific Na+ channels but would
principally be in agreement with characteristics of currents through
nonselective cation (NSC) channels. Therefore, activated NSC channels
could be responsible for the inward current during the sustained
current phase. The absence of outward current at positive membrane
potential (Fig. 1D) is, however, not in accordance with
general characteristics of NSC channels (which are conducting also at
positive membrane potentials). We suggest that this deviation may be
due to an unrelated norepinephrine-induced decrease in the activity of
the voltage-dependent K+ channels (cf. Fig. 2C).
This decrease may be of sufficient magnitude (a <10% inhibition of
the K+ current would be needed) to more than counteract the
expected outward current of activated NSC channels at positive membrane potentials. (Attempts to eliminate this K+ current by
exchanging K+ with Cs+ in the pipette were
principally successful but had to be abandoned because the ion
substitution was otherwise poorly tolerated by the cells.)
Effects of the adrenergic subtype-selective agonists cirazoline and
CGP-12177A.
To determine which subtype of adrenergic receptor was responsible for
the different phases in the norepinephrine-induced current response, we
studied the effects of the 1-selective agonist
cirazoline and the 3-specific agonist CGP-12177A, known
selective activators of the indicated adrenergic subtypes in brown fat
cells (31). We compared the responses to each of these
selective agonists with the responses to the endogenous agonist
norepinephrine described above; norepinephrine is expected in itself to
be able to activate all types of adrenergic receptors.
1-adrenergic agonist cirazoline could qualitatively
mimic the action of norepinephrine; both the first and the sustained inward currents could clearly be identified (not shown, but compare with Table 2). Quantitatively, however,
cirazoline did not seem to be able to mimic fully the effect of
norepinephrine. As seen in Table 2, the cirazoline-induced first inward
current was only two-thirds of that induced by norepinephrine, and the
induced membrane depolarization during this phase was only 10 mV, i.e., also only two-thirds of that induced by norepinephrine. However, the
sustained inward current was the same whether it was induced by
cirazoline or by norepinephrine, as was the depolarization during this
phase (8 mV). Thus, whereas a -adrenergic component of the first
inward current cannot be excluded, 1-adrenergic stimulation seemed to be competent to induce the sustained inward current to the same extent as did norepinephrine.
|
3-stimulation induced by CGP-12177A. In cells that
responded well to norepinephrine or cirazoline, 1 or 10 µM CGP-12177A
was in some cases fully without effect, and in other cases it induced membrane current oscillations (not shown). However, the sustained inward current was always totally absent in the response induced by
CGP-12177A.
Thus 1-receptors were primarily responsible for the
overall current response. Through 1-stimulation, most of
the first inward current response and the entire sustained inward
current response could be elicited. 3-Receptors may
participate in the first inward current response but could not induce
at all the sustained inward current.
Effects of extracellular Ca2+ removal during
1-stimulation.
1-Adrenergic stimulation is associated with an increase
in [Ca2+]i in brown fat cells (1, 13,
27, 29), and this increase is believed to be one of the
intracellular mediators of 1-adrenergic stimulation. To
investigate to what extent the observed 1-effects were
dependent on Ca2+, we examined the responses to adrenergic
stimulation in a buffer that did not contain Ca2+ but
instead contained 1 mM of the Ca2+ chelator EGTA. Such
conditions lead to a diminished norepinephrine-induced increase in
[Ca2+]i (see Ref. 29).
50 was halved, down to the normal resting
level (cf. Table 1), whereas Gm,+15 was
statistically unchanged.
|
|
1-receptor and the channel carrying the sustained inward current.
Identification of ions involved.
To establish which ion(s) carry the inward current, especially the
sustained inward current, experiments were performed with omission of
extracellular Cl or Na+.
REMOVAL.
Extracellular Cl was reduced from 144.8 mM to 30 mM by
partially substituting it with aspartate. The calculated
Cl reversal potential would now be +31 mV, i.e., an
expected shift of +49 mV. The resting membrane potential was not
altered by partial Cl substitution (24 mV before
substitution, 25 mV after; mean from 2 experiments), implying that
Cl permeability in this state was low compared with that
of other ions. Accordingly, there was no marked effect of
Cl substitution on Gm,50 or
Gm,+15 in the resting state (not shown).
However, the first inward current was diminished (not shown), and the
first depolarization was now only ~8 mV (not shown). In contrast, the
sustained inward current was not diminished by Cl
substitution, and the induced current obtained from the voltage ramps
at negative voltages was not markedly affected by Cl
substitution (not shown), implying that this phase of the response was
not mediated by Cl currents.
EFFECTS OF NA+ REMOVAL.
If Na+ was replaced by
N-methyl-D-glucamine (NMDG+) already
at the start of the experiment, the resting membrane potential was increased from 20 to 40 mV, and the
Gm,50 was halved (not shown). In the absence
of Na+, the first inward current could still be observed
(Fig. 4A). The voltage
ramp-induced current had characteristics similar to those observed in
normal buffer (not shown). However, Na+ omission completely
eliminated the norepinephrine-induced sustained inward current (Fig.
4A). Similarly, when Na+ was temporarily omitted
during the time of the expected sustained inward current (Fig.
4B and Table 4), it acutely
eliminated the inward current; thus the absence of inward current in
Na+-free buffer was not due to secondary effects of
prolonged omission of Na+. The sustained inward current
reappeared when NMDG+ was again replaced with
Na+; similarly, the membrane potential depolarized 8 mV
when Na+ was reintroduced, and the
Gm,50 was doubled. In the absence of
extracellular Na+, there was no norepinephrine-induced
current during the voltage ramps (contrast the -curve in Fig.
4C with that in Fig. 1D), but it may be noted that there was
still an apparent induced inward current at positive membrane
potentials, again probably a manifestation of an independent inhibitory
effect of norepinephrine on K+ channels (cf. Fig.
1C). This effect was thus clearly not mediated through an
Na+ influx.
|
|
A Large-Conductance NSC Channel
During norepinephrine perfusion, we often (in 14 of the 21 cells tested) observed currents apparently resulting from single channel activity. Such currents are quite prominent in Fig. 3B and are discernible in Fig. 1A. In Fig. 5A, we show an enlargement of such current events. The events clearly represented inward currents from a single but large ion channel. In most cases, only one active channel per cell was detected, maximally two. A preliminary characterization of these channels could be made from their response to the conditions tested above.
|
The large-conductance channel currents were never observed in
unstimulated cells but were frequently observed during norepinephrine and cirazoline stimulation. CGP-12177A (tested in 4 cells) was not able
to activate the channel, even in a cell in which both norepinephrine and cirazoline could activate it. The activity disappeared upon washout of the adrenergic agonists. It was
occasionally possible to observe activity during a voltage ramp (Fig.
5B); when it opened during the sustained inward current
phase, it temporarily doubled the current. Extrapolation to zero
current voltage indicates that the cell depolarized ~10 mV during
each such opening. It is also clear from this and similar recordings
(not shown) that the activity was not observable at potentials more
positive than 10 mV.
In some experiments, we recorded the conductance and the activity of the large-conductance channel at different membrane potentials. An example of this is given in Fig. 5A, and data from one such study are collected in Fig. 5, C and D. In Fig. 5C, the current-voltage relationship is shown. The activity corresponded to a single-channel conductance of 270 pS, which was fully ohmic within the voltages tested and which had a reversal potential very close to 0 mV. This would indicate that this large-conductance channel was an NSC channel. The open probability (Po) showed an unusual dependence on the membrane potential, being, if anything, activated by hyperpolarization (Fig. 5D).
Omission of extracellular Ca2+ did not affect the
conductance of the channel (the current was 13 pA at 50 mV both
before and during Ca2+ removal) but impressively increased
the open probability sevenfold, from 0.05 to 0.33 (not shown but cf.
Fig. 3B). Reduction of extracellular Cl to 30 mM did not
affect the single-channel amplitude or the activity of the channel (not
shown). However, removal of extracellular Na+ reversibly
abolished the large-conductance channel activity (not shown),
demonstrating that it was a Na+-conducting channel.
Thus the large-conductance channel was identified as an NSC channel with unusual regulatory properties. The physiological role of this type of channel in brown fat cells remains unknown, but a similar type of channel has been observed in rat basophilic leukemia cells; that channel was suggested to regulate exocytosis (20).
The properties of the large-conductance channel, especially the unusual Ca2+ dependence characteristics, clearly indicated that it was not this channel that carried the current during the sustained inward current phase. We therefore proceeded to the cell-attached mode to identify the ion channels carrying the depolarizing current.
Adrenergic Activation of the 27 pS NSC Channel
To identify the ion channels responsible for the sustained depolarization of brown fat cells during adrenergic stimulation, patch-clamp experiments in the cell-attached mode were performed. In the whole cell studies above, the current to be sought was established to be elicited by1-adrenergic stimulation and have a
delay of 1 min, but then to be persistent for as long as the adrenergic stimulation persisted. It was a Ca2+-dependent
Na+ current and was implied to have characteristics
compatible with it being mediated via a channel that is nonselective in
character (e.g., with a reversal potential close to 0 mV). Because the
membrane potential in these cells is only about 20 mV (see above), we chose to clamp the holding potential to +20 mV to enhance the single-channel current through the NSC channel we searched. The expected transmembrane potential across the patch was therefore 40 mV.
Effects of norepinephrine on single channel activity.
NOREPINEPHRINE ACTIVATES SINGLE CHANNEL ACTIVITY.
Figure 6A shows a trace of the
current through the cell-attached pipette after a gigaohm seal had been
formed. The recording was started while the cell was perfused with only
extracellular solution; as seen, some spontaneous ion channel activity
was observed during this time. Such channel activity in unstimulated
brown fat cells has earlier been observed and demonstrated to be
mediated via 27 pS NSC channels (28) (and see below).
During the time indicated by the horizontal line, the cell was perfused
with the same solution, now containing 1 µM norepinephrine; as seen,
this resulted in a dramatic increase in channel activity in the patch.
It was considered likely that this prominent activity could be
responsible for the sustained inward current; further analysis detailed
below supported this view. Because the perfused norepinephrine did not
come in direct contact with the channels investigated, it was clear
that the activation must have been mediated through an intracellular mediator.
|
channels and Ca2+-activated K+ channels) in the
brown fat cells, but we did not observe prominent channel activity from
any other channel types. This may, however, be understandable on the
basis of characteristics of these channels: they are probably
low-conductance channels and were here monitored under conditions
fairly close to their respective equilibrium potential.
After the delay of ~1 min, the ion channel activity in the patch
became more intense, although not fully uniform; it had a tendency to
fluctuate with an interval of 1-2 min between periods of highest
Po. The response was, however, persistent and
did not show any tendency to inactivate during the routine recordings, which were continued for 5 min. Thus the observed ion channel activity had the current direction and time course that would be
required for a current mediating the sustained inward current.
In Fig. 7, a quantification of results is
shown from a series of studies performed like those illustrated in Fig.
6, A and B. Because of the fluctuating behavior
of the channel, the Po was calculated for an
extended time period (4 min). As seen, norepinephrine very
significantly increased the observed open probability
(Po,obs) from 0.005 to 0.038; the mean increase
was >10-fold when calculated from each experiment.
|
|
|
30 pS.
The single channel current was measured in several experiments at the
holding potential of +20 mV before and during norepinephrine perfusion
(Table 5). The mean current observed before norepinephrine perfusion was 1.36 pA. During norepinephrine perfusion, the current was
reduced to 1.19 pA, i.e., by 0.17 pA. On the basis of a conductance of
30 pS, the driving force for the current was in the mean reduced by 6 mV, i.e., the cells had become depolarized. The depolarization determined in this way is in remarkably good agreement with the depolarization (change in zero current potential, 8 mV) measured during
the sustained inward current phase in the whole cell study above (Table
1).
The extrapolated reversal potential observed in Fig. 8 and similar
studies was in the range of +15 mV in the control state and +10 mV in
the norepinephrine-stimulated cells. These values are very close (but
with reversed sign) to the measured cell membrane potentials in the two
conditions (Table 1). This means that the experienced reversal
potential of these currents in the patch would be very close to 0 mV,
as would be expected for a NSC channel. There is, therefore, good
reason to conclude that the 30 pS conductance channel activated by
norepinephrine stimulation of brown fat cells represents an NSC channel.
Thus the norepinephrine-induced single channel activity observed in
cell-attached patches had the time course and reversal potential
required for the channel mediating the sustained membrane depolarization in brown fat cells.
Effects of selective 1- and
3-adrenergic agonists on ion channel activity.
On the basis of the data shown (above) from whole cell current studies,
the channel responsible for the sustained inward current should be
1-adrenergically activated. To examine whether this was
the case for the single channel activity observed here, we examined the
ability of two subtype-selective adrenergic agonists, the
3-agonist CGP-12177A and the 1-agonist
phenylephrine, to elicit a channel activation similar to that elicited
by norepinephrine.
27 pS (Table 5)
but, in contrast to the case for norepinephrine, the single channel
current was not lower than in unstimulated cells (if anything, it was
higher). Thus, in agreement with our observations in whole cell
studies, CGP-12177A perfusion did not seem to lead to measurable
membrane depolarization. Because the 10-fold increase in channel
activity observed with norepinephrine (Table 5) led only to a doubling
in total cell conductance (Table 1), the doubling in channel activity
observed with CGP-12177A would only be expected to lead to a <10%
increase in total cellular conductance, which would hardly be
observable as a depolarization.
PHENYLEPHRINE.
Phenylephrine, just like norepinephrine, increased channel activity
about an order of magnitude (Fig. 7). The channel activity was, if
anything, higher than that observed during norepinephrine perfusion.
The mean open time (Table 5) was even longer than that seen during
norepinephrine perfusion, but the estimated single channel conductance
was the same (27 pS) (Table 5). In accordance with the tendency to a
more intensive effect of phenylephrine than of norepinephrine, the
single channel current was also more reduced during phenylephrine
perfusion, by 0.22 pA, corresponding to a membrane depolarization of 8 mV.
Thus much more dramatic effects on channel activity were seen after
phenylephrine (i.e., 1-stimulation) than after
CGP-12177A (i.e., 3-stimulation); in fact, phenylephrine
could fully mimic the effect of norepinephrine. This is thus in
accordance with the adrenergic receptor requirements from the whole
cell studies for the channel activity mediating the sustained membrane
depolarization in these cells.
Identity of the second messenger regulating the ion channel
activity.
On the basis of the fact that the sustained inward current observed in
the whole cell studies was 1-adrenergic and
Ca2+ dependent, it would be expected that the single
channel activity responsible for the mediation of this current would be
activated, directly or indirectly, through an increase in
[Ca2+]i. To investigate whether this was the
case, we examined whether an increase in cAMP, or, as would be
expected, an increase in [Ca2+]i, would be
able to increase the single channel activity. We used the adenylyl
cyclase activator forskolin to increase intracellular cAMP levels and
the ionophore A-23187 to increase [Ca2+]i.
1/Ca2+ pathway had a much better ability
than the /cAMP pathway to activate the channel activity, an
observation fully in accordance with the demands for the channel
mediating the sustained whole cell current.
Identity of the channel mediating the sustained inward current.
It is clear from the patch-clamp studies in the cell-attached mode that
the single channel activity observed fulfilled the criteria for being
the channel activity behind the sustained inward current observed in
the whole cell studies: time course of activation, nonselective nature
of current, adrenergic receptor involvement, and Ca2+
dependence of activation. With respect to Ca2+ dependence,
and especially on consideration of the observed channel size of 30
pS, it is very notable that, in excised patches from these cells, an
NSC channel with exactly these properties has been well characterized
(11, 12, 23, 26, 28). It is therefore very likely that it
is this NSC channel, the 27 pS channel, that mediates the sustained
inward current.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
We show here that norepinephrine induced two inward current phases
in brown fat cells: a transient current, which lasted for ~1 min, and
a sustained current that persisted as long as norepinephrine stimulation persisted. The experiments conducted had as their main goal
to further the understanding of the nature of the sustained current,
which we could characterize fairly well, and which we conclude is an
1-adrenergically induced Ca2+-dependent
activation of 27 pS NSC channels.
Concerning the first inward current phase, it was, of course,
unavoidable to obtain experimental results during the experiments conducted here; they were also briefly reported above. The elucidation of these events was not the prime goal of the present investigations, and a full understanding of the events has not been reached here. However, on the basis of these observations and literature data, a
tentative interpretation of the plasma membrane events after norepinephrine stimulation of brown fat cells is given below, in which
we discuss in succession the resting state, the first inward current,
and the sustained inward current (Fig.
9).
|
The Resting State
The mean capacitance of the brown fat cells investigated here was 35 pF (Fig. 8A). Based on the general value of 1 µF/cm2 cell membrane, this capacitance corresponds to the surface of a sphere with a diameter of 33 µm.The Gm,50 value of 0.7 nS was much smaller
than whole cell conductances observed in classical microelectrode
experiments in brown adipose tissue [~700 nS (7)] and
even somewhat smaller than those earlier observed with patch-clamp
techniques, in classical whole cell studies (14), or in
perforated-patch studies (2-8 nS) (15).
On the basis of the zero current during voltage ramps, the resting
membrane potential was estimated to be about 21 mV. Because the cells
studied probably had a clamped intracellular ion level determined by
the pipette concentration, it may be questioned whether this potential
represents that present under innate conditions. However, the estimates
in the cell-attached mode, in which intracellular ion concentrations
were not affected, confirmed this rather low value under these conditions.
The identity of the ion permeabilities that determine the resting
membrane potential is not immediately evident. The measured potential
is far from the expected K+ equilibrium potential of 83
mV. This deviation could be due to fairly high permeabilities for
Cl or Na+. Alterations in extracellular
Cl concentration did not markedly influence the resting
membrane potential, but the membrane potential became more negative
when extracellular Na+ was exchanged with
NMDG+. However, in the presence of mefenamic acid, which
fully inhibits the 27 pS NSC channel (11), the observed
resting Em did not come closer to 85 mV. This
also implies that the low basal activity of the NSC channels (cf. Table
5) is not sufficient to significantly influence the resting membrane
potential. Thus the identities of the ion permeabilities responsible
for the low resting membrane potential are not known, but a
Na+ permeability not mediated by the 27 pS NSC channels is
likely involved.
The First Inward Current
Adrenergic receptor type involved.
The selective 1-adrenergic agonist cirazoline was able
to induce the first inward current in a manner qualitatively similar to
norepinephrine, but the response was only two-thirds of that observed
with norepinephrine (Fig. 8, B and C). The
difference may be due to effects mediated via
3-receptors, because CGP-12177A generally elicited some
membrane currents during this phase. The 1-induced
inward current became much smaller in Ca2+-free buffer,
implying that the 1-signal is intracellularly mediated via an increase in Ca2+ levels.
Ionic characteristics of the first inward current.
The first inward current phase probably involves currents of more than
one ion species, mediated through more than one channel. The fact that
the currents during voltage ramps were changed as an effect of
reduction of extracellular Cl levels is in agreement with
results from earlier ion flux (4) and patch-clamp
(21) studies concluding that Cl currents
probably were induced in this phase.
1-adrenergic, Ca2+-dependent K+
current is in agreement with earlier ion flux and and
electrophysiological studies (15, 17, 18). In our
voltage-clamped cell system, the putative K+ current never
became sufficiently large to induce a net outward current, but under
nonclamped conditions, the voltage-sensitive K+ channels
(14, 23) may also be activated secondarily to the depolarization caused by the increased Cl permeability,
leading to the hyperpolarization observed in non-voltage-clamped cells
(9, 14).
A tentative interpretation of the events during the first inward
current in the voltage-clamped cell is therefore that
1-adrenergic stimulation, via increases in cytosolic
Ca2+, first activates a Cl conductance that
is responsible for the immediate depolarization observed (Fig.
9B). However, nearly at the same time, but with a slight
delay, there is an activation of K+ channels, probably also
Ca2+ dependent; thus the total conductance of the cell
increases vastly (Fig. 9C). The activation of K+
channels tends to counteract the depolarization and help in the repolarization event; thus the time of maximal depolarization precedes
the time of maximal conductance. Both the Cl and
K+ currents are inactivated (through unknown mechanisms)
after ~1 min.
That both Cl and K+ currents are involved
during the first inward current is supported by the earlier
observations of ionic fluxes from these cells. Within the first minute
after 1-adrenergic stimulation, both Cl
(4) and K+ (17, 18) ions are lost
from the cells. If only Cl conductance had increased, the
cell would have depolarized, but because a new electrochemical
equilibrium would have been reached, a large
36Cl efflux could hardly have been expected.
Thus a net loss of KCl from the cell probably occurs during the 1st min
of adrenergic stimulation. This may mean that volume changes occur in
the cells in this phase.
Ion channels involved.
The identities of the ion channels involved in the first inward current
have not been established. A 50 pS Cl channel has been
observed in brown fat cells (24), but there is no
functional evidence to associate it with the norepinep
