Reversible blockade of the calcium-activated nonselective cation ...

Pfltigers Arch (1993) 425:549-551

Journal of Physiology 9 Springer-Verlag 1993

Short communication

Reversible blockade of the calcium-activated nonselective cation channel in brown fat cells by the sulfhydryl reagents mercury and thimerosal Ari Koivisto 1, Detlef Siemen 2, J a n N e d e r g a a r d 1 1 The Wenner-Gren Institute, The Arrhenius Laboratories F3, Stockholm University, S-106 91 Stockholm, Sweden 2 Institut ftir Zoologie, Universit~t Regensburg, D-93040 Regensburg, Germany Received June 23, 1993/Received after revision September 24, 1993/Accepted September 28, 1993 Abstract. We have used patch-clamp techniques to study the effect of the sulfhydryl group oxidizing agents mercury and thimerosal on calcium-activated nonselective cation channels from brown adipose tissue. 100 nmol/l mercury and 50 /~mol/1 thimerosal induced a complete block. Blockade could be reversed by reduction of the mercaptide by dithiotreitol (DTT). Mercury was found to be the most potent blocker (ICs0-value 2 1 x 1 0 -9 mol/l), whereas thimerosal (IC50-value 1.5 x 10-6 mol/l) was as effective as 3',5-dichlorodiphenylamine-2-carboxylic acid (DCDPC). The DCDPC effect, however, could not be reversed by DTT, indicating different blocking mechanisms. It is concluded that SH-groups are involved in gating of the calcium-activated nonselective channel.

indicates that the channel has -SH groups which are essential for its activity.

Materials and methods Materials. Mercuric chloride (HgCl2), thimerosal (mercury-[(o-earboxy-

phenyl) thio]-ethyl sodium), and dithiothreitoi (D'I'T) were from Sigma. 3',5-dichiorodiphenyl-amine-2-earboxylie acid (DCDPC) was kindly provided by Dr H.J. Lang (Hoechst AG, Frankfurt, Germany). All other chemicals were of analytical grade and were either from Sigma or Merck. DCDPC was dissolved in dimethyl sulfoxide (DMSO), the final concentration of DMSO was 0.05 %. DMSO alone had no effect on the single channel recordings. Methods. Brown fat cells from young male (80-140 g) Sprague-Dawley

Key words: Brown fat; DCDPC; Mercaptide; Mercury; Nonselective cation channel; Sulfhydryl reagent; Thimerosal

Introduction Nonselective cation (NSC) channels have been found in the plasma membranes of a wide variety of tissues, but in many cases their physiological role is still enigmatic. In brown fat, as in many other tissues, NSC channels are activated by calcium ions and blocked by purine nucleotides, acting from the intracellular side of the plasma membrane (for review see [6]). In some types o f ion channels, alteration of the oxidation state of suifhydryl (-SH) groups leads to altered gating [2,3,7,8,10]. The status of these -SH groups may depend on the redox potential of the cell which could therefore be important in the regulation of channel activity [11]. We report here that the sulfhydryl oxidizing agents mercury and thimerosal induce rapid and reversible blockade of the calcium-activated NSC channel from brown fat cells. This

rats were isolated by eollagenase (5 mg/ml) digestion and dispersed into single cells by shaking in buffer at 37~ for 35 min. Cells were collected by eentrifugation (150 x g, 10 min) and kept 1-7 days in an incubator at 37~ with a water-saturated atmosphere of 8% C 0 2 - 92% air. The medium was Dulbeecos modified Eagles medium supplemented with 100 ml newborn calf serum per 0.9 1 medium, insulin (4 nmol/l), Na-ascorbate (126 #mol/l), glutamine (4 mmol/l), penicillin (50 IU/ml) and streptomycin (50 #g/ml). Since mature fat cells float in the culture medium, a prerequisite for patch-clamping was that the cells adhered to the hydrophilic side of Heraeus-Biofoil-25R (Iteraeus GmbH, Hanau, Germany), which was placed on the surface of the medium. Biofoil with attached cells was placed on a petri-dish, and filled with extracellular solution. We used the inside-out mode of the patch-clamp technique. Pipettes were pulled from borosilicate glass (Clark GC-150, Pangbourne, England) and had resistances between 8 and 12 ME. Channel activity was measured at either -40 mV or +40 mV membrane potential (Era). In the displayed recordings, single-channel currents are shown according to the convention with upward deflections denoting outward currents. Currents were recorded using an L/M EPC 7 patch-clamp amplifier (List Medical, Darmstadt, Germany). Data were stored on a video recorder after the signal had been digitized with a modified pulse code modulator (Sony 601 ESD). The data were analysed off-line with a 486-processor based computer. Open time probability (Po) was calculated from the all-points amplitude histograms by the pSTAT program (ANON Instruments, Foster City, CA) using the data of 30 s at minimum. In all experiments, the extracellular and the patch-pipet solutions contained (in mmol/l): 134 NaC1, 6 KCI, 1.2 MgCI2, 1.2 CaCI2, and 10

Correspondence to: A. Koivisto

HEPES (pH 7.4 with NaOH). Chemicals were dissolved into a solution

550 containing: 130 KCI, 10 NaCI, 1.2 MgC12, 1.2 CaCI2, and 10 HEPES (pH 7.2 with KOH). After excision, the patches were perfused by these test solutions using a system of 4 gravity-driven pipettes in parallel controlled by electromagnetic valves. All experiments were performed at room temperature (22~ The current signal was low-pass filtered with a -3 dB frequency of 0.5 kHz by using an 8-pole Bessel filter (Frequency Devices, Haverhill, USA). Figures were made by using the pCLAMP 5.5 program (Axon Instruments Inc. Foster City, USA) and a Laserjet Ill printer. A

50 pmol/I thimerosal

2 mmol/I DTT

4

B

100 nmol/I HgCI2

2 mmol/I DTT 10s

C

is

..............

~ 2

~'-,12.A

........................................

[

D 1.0-

0.5-

Results The blocking effect of mercury and thimerosal, highly specific sulfhydryl reagents [9], on the calcium-activated NSC channel was complete at 100 nmol/1 and 50 #mol/l, respectively (Fig. 1A,B). Higher concentrations of both substances also blocked completely in all cases tested. Concentration dependence is shown in Fig. 1D. The IC50-values for mercury and thimerosal were (in mol/l): 21 x 10-9 and 1.5 x 10.6 at +40 mV (cf. Fig. 1D) and 15 x 10-9 (n=33) and 0 . 9 x 10.6 (n=21) at -40 mV, respectively. As the IC50-values did not significantly differ between positive and negative membrane potentials, the sulfhydryl oxidation-induced blockade was not voltage-dependent in this range. Inspection of single-channel recordings showed that both substances acted by increasing the mean closed time of the channel. The amplitude histograms were not shifted towards the baseline and the peaks corresponding to open channels were not broadened. Thus, the blockade observed was of the "slow" type. The blocking action of mercury and thimerosal was not spontaneously reversed, even after extensive washout (up to 4 min) with mercurial-free solution. However, perfusion with 2 mmol/l of the disulfidereducing reagent DTT was able to restore substantial channel activity in all cases tested (n= 15, n = 11). The blockade and its reversal by DTT could be repeated several times. DTT itself had no effect on open time probability (n=3) (Fig. 2A). DCDPC is a well-known blocker of calcium-activated NSC channels [1]. Even 2 mmol/l DTT was not able to reverse the complete blocking action of 50 gmol/l DCDPC (n=4), demonstrating that mercurials and DCDPC act through different mechanisms. In contrast to inhibiton by mercurials, DCDPC inhibition was rapidly reversed upon washout with DCDPC-free solution (Fig. 2B). A

0.O .

r [tool/I] 10-

.

.

10 - s

.

.

10 .7

.

.

10 - s

i "'

10 -5

9

10 -4

ls

I Fig. 1A-D. A: Blockade o f calcium-activated NSC channels from brown adipose tissue by 50 #mol/l thimerosal added to the cytoplasmic side of the membrane and recovery after addition of 2 mmol/l D T r . B: Blockade of NSC channels by 100 nmol/l HgCI 2. There was no recovery upon washout but 2 mmol/l DTT restored activity. Upward shift of the baseline is an artefact due to switching to D T r solution. C: Absence of effect of extracellular thimerosal. Thimerosal (100 gmol/l) was present in the patch-pipette throughout the recording. D: Dependence of the open probability (Po) on the concentration (c) of mercury (&) and thimerosal (0). Continuous curves calculated by Po = Po,m~x / (1 + c / ICs0) where Po,max is the control value and 1C50 is the half maximal concentration. This equation assumes 1:1 binding. Number of measurements per point (with increasing concentrations of the sulthydryl reagent): 7,3,4,4,8 (&) and 5,2,5,5,3,6 (0). Recording from inside-out patches, Era: + 4 0 mV (A,B,D) and -40 mV (C), Number of channel open levels is indicated at the left side o f Figs.

pA

B

50 pmol/l DCDPC + 2 mmol/I DTT

0 ............

,,

~'Y~--~.~s--~

- " =" . . . . . . . . . . . . .

120A

12

""~'~ 'I . . . . . . . . '~v

"i;

Fig. 2A,B. Effect of D T r and DCDPC. A: D T r (2 retool/l) alone was without effect on NSC channel activity. D T r was present at the intracellular side of the patch throughout this experiment. Era= + 4 0 mV. B: 50 #mol/l DCDPC in the presence of 2 mmol/l D T r was able to block NSC channels completely. The blocking effect was spontaneously reversed after washout. Era=-40 InV. Different time scale in A and B.

551 As mercurials are hydrophilic and therefore only weakly membrane permeant, it is likely that their action was on SH groups on the cytoplasmic side o f the plasma membrane. This was confirmed in experiments where patchpipettes were filled either with 100 gmol/l thimerosal (Fig. 1C) or 10 /xmol/1 mercury. The extracellular side o f the patch was thus continuously exposed to the substances, but this did not have any noticeable effect on channel activity.

Discussion

In the present investigations, we have shown an effect o f sulfhydryl oxidation on calcium-activated NSC channel activity in excised membrane patches from the brown fat cell. Mercury and thimerosal blocked channel activity in a concentration-dependent manner with ICs0-values o f 21 • 10 -9 mol/l and 1.5 • 10 -6 mol/l and with a complete blocking effect occurring at 100 nmol/l and 50 #mol/l. The blockade was o f the slow type. The higher potency o f mercury as compared with thimerosal could be due to the fact that mercury oxidizes both mono- and dithiols, whereas thimerosal oxidizes only monothiols. The effects o f both substances were substantially reversed by perfusion with an excess o f the disulfide-reducing reagent DTT, but not by washing away the substance. This demonstrates that the action o f mercury or thimerosal was not due to a nonspecifie effect, but was caused by a sulfhydryl oxidation. The fact that D T T itself had no effect, indicates that -SH groups were in the reduced form in the intact channels. Thimerosal, although structurally quite dissimilar from D C D P C , seems to be as potent an inhibitor o f the channel as is D C D P C , which was known as the most potent blocker o f calcium-activated N S C channels [1]. D T T had no effect on the inhibitory action by D C D P C , showing that sulfhydryl oxidation is not involved in the blocking mechanism o f DCDPC. This may make it possible to develop new types o f N S C channel blockers based on sulfhydryl oxidation, and thereby to avoid problems connected with D C D P C and its derivatives [1,5], which, as weak lipophilic acids, can act as protonophores in biological membranes [4]. The sulfhydryl oxidation effect was not dependent on membrane voltage. In the inside-out patches used in this study, mercurials were effective only when added to the cytoplasmic side. This was not surprising, as mercurials are hydrophilic and weakly membrane permeant (in contrast to D C D P C , which is very lipophilic and blocks from both sides) [1]. These results demonstrate therefore the presence o f -SH groups on the cytoplasmic side o f the calcium-activated NSC channel in the brown fat cell. F r o m our results w e conclude that free -SH groups are critically important for the gating o f the N S C channel. There have been some reports showing that modification o f sulfhydryl groups leads to changes in ion channel gating in other channel types. Such results have been obtained with intracellular Ca 2+ channels [7,8,10], the N M D A

receptor-channel complex [3], and the ATP-regulated K + channel [2]. In pancreatic acinar cells, extracellularly applied thimerosal was able to release Ca 2+ from intracellular stores and to evoke inward cation currents via N S C channels [8]. As these currents were considerably smaller in the presence o f 50 #mol/l thimerosal (their Fig. 2) than at 5 /zmol/l thimerosal (their Fig. 1), we take this data as additional support o f the idea that thimerosal blocks N S C channels. The fact that mercury is such a potent blocker could indicate that the channel is indeed regulated by some physiological factor. Whether some o f the acute toxic effects o f mercury on principal target tissues, s u c h as kidney and neurons, can be correlated with the blockade o f the NSC-channels in these tissues, remains to be tested.

Acknowledgements. Financial support was obtained from the Swedish Natural Science Research Council and the Hasselblad foundation. We thank Barbara Cannon for stimulating discussions. Technical support from Stig Sundelin is gratefully acknowledged.

References 1.

2.

3.

4. 5.

6. 7.

8.

9.

10.

11.

G6gelein H, Pfannmiiller B (1989) The nonselective cation channel in the basolateral membrane of rat exoerine pancreas. Inhibition by 3',5-dichlorodiphenylamine-2-carboxylic acid (DCDPC) and activation by stilbene disulfonates. Pfliigers Arch 413:287-298 Islam MS, Berggren P-O, Larsson O (1993) Sulfhydryl oxidation induces rapid and reversible closure of the ATP-regulated K + channel in the pancreatic fi-cell. FEBS Lett 319:128-132 Lei SZ, Pan ZH, Aggarwal SK, Chen HS, Hartman J, Sueher NJ, Lipton SA (1992) Effect of nitric oxide production on the redox modulatory site of the NMDA receptor-channel complex. Neuron 8:1087-1099 McLaughlinSGA, Dilger JP (1980) Transport of protons across membranes by weak acids. Physiol Revs 60:825-863 Poronnik P, Ward MC, Cook DI (1992) Intracellular Ca2+ release by flufenamie acid and other blockers of the non-selective cation channel. FELLSLett 296:245-248 SiemenD, Hescheler J (eds) (1993) Nonselective Cation Channels - Pharmacology, Physiology and Biophysics, Birk.h/iuser,Basel Swann K (1991) Thimerosal causes calcium oscillations and sensitizes calcium-induced calcium release in unfertlized eggs. FEBS Lett 278:175-178 Thorn P, Brady P, Llopis J, Gallacher DV, Petersen OH (1992) Cytosolic Cax+ spikes evoked by the thiol reagent thimerosal in both intact and internally perfused single pancreatic acinar cells. Pfliigers Arch 422:173-178 Torchinsky YM (1981) The chemical properties of SH groups. Sulfhydryl reagents. In: Torchinsky YM. Sulphur in Proteins. PergamonPress, Oxford, pp 3-71. Trimm JL, Salama G, Abramson JJ (1986) Sulfhydryl oxidation induces rapid calcium release from sacroplasmic reticulum vesicles. J Biol Chem 261:16092-16098 Ziegler DM (1985) Role of reversible oxidation-reduction of enzyme thiols-disulfides in metabolic regulation. Ann Rev Biochem 54:305-329