Molecular Membrane Biology, September Á/October 2004, 21, 307 Á/313
pH modulation of large conductance potassium channel from adrenalchromafﬁn granules
channel gene CLCN7 leads to a severe osteopetrotic
phenotype because osteoclasts fail to resorb bone and
they cannot acidify the lacuna . Mitochondrial potassiumchannel has been suggested as a trigger and effectormyocardial ischemic preconditioning .
$ Department of Biophysics, Agriculture University SGGW,
Ion channels have also been reported in the membrane of
159 Nowoursynowska St., 02-776 Warsaw, Poland
chromaffin granules from adrenal medulla [8 Á/12]. Thechromaffin granules are involved in catecholamine synthesis
% Laboratory of Intracellular Ion Channels, Nencki Institute of
and traffic, both within and outside the cell . The uptake of
Experimental Biology, Polish Academy of Sciences, 3
hormones, driven by pH gradient (DpH), from the cytosol into
chromaffin granules is catalyzed by a catecholamine carrier
§ Laboratory of Protein Chemistry, Institute of Bioorganic
. Secretion of hormones occurs as a result of the fusion of
Chemistry, Belarus National Academy of Sciences,
chromaffin granule vesicles with the plasma membrane of
The chromaffin granule ion channels have been investi-
gated after fusion of granule membranes with bilayermembrane (BLM), followed by single channel recordings.
We report here that large conductance K selective channel inadrenal chromaffin granules is controlled by pH. We measured
Several different cation selective channels were described
electrogenic influx of 86Rb into chromaffin granules prepared
after incorporation of intact chromaffin granules, but only two
from bovine adrenal gland medulla. The 86Rb influx was
types of highly selective K' channels could be reconstituted
inhibited by acidic pH. Purified chromaffin granule membranes
from preparation of chromaffin granule ‘‘ghosts’’ [11,12]. A
were also fused with planar lipid bilayer. A potassium channelwith conductance of 4329
K' selective, large conductance ( Â/160 pS in symmetrical
observed after reconstitution into lipid bilayer. The channel
400 mM KCl) channel was described by Arispe et al . . It
activity was unaffected by charybdotoxin, a blocker of the
was insensitive to charybdotoxin, a blocker of the Ca2'-
Ca2-activated K channel of large conductance. It was
activated K' channel of large conductance . The channel
observed that acidification to pH 6.4 cis side of the membranelowered the channel open probability and single channel
activity was also unaffected by Ca2' and potential across
conductance. Whereas only weak influence on the single
the bilayer . It was also reported that the chromaffin
channel current amplitude and open probability were observed
granule K' channel was controlled by both inhibitory and
upon lowering of the pH at the trans side. We conclude that a
stimulatory heterotrimeric GTP-binding proteins . A
pH-sensitive large conductance potassium channel operates inthe chromaffin granule membrane.
similar channel, highly selective for potassium, but with adifferent conductance ( Â/400 pS in symmetric 450 mM KCl),
Keywords: intracellular chromafﬁn granule potassium channel,
was described by Ashley et al . . The channel was
adrenal chromafﬁn granules, bilayer lipid membranes.
insensitive to both Ca2' and charybdotoxin, and wasblocked by TEA'.
A key problem concerning single channel recordings of
BLM, blacklipid membrane technique; I, single-channel current
chromaffin granules in planar bilayer membranes is the purity
amplitude; U, potential; Urev, reversal potential; to, mean lifetime Á/
of the applied membrane preparation. Therefore, to study
open time; tc, mean lifetime Á/ closed time; Popen, open-probability;
chromaffin granule potassium transport we applied both
single channel recordings and flux measurements using86
Rb , a K' analog, as described by others [16,17].
Previously we have successfully used this approach to
show that electrogenic K' transport in chromaffin granules
Potassium and chloride selective channels exist in mem-
is blocked by sulfhydryl reagents , various potassium
branes of organelles such as mitochondria, sarco/endoplas-
channel blockers  and by ATP . Transport measure-
mic reticulum, endosomes, synaptic vesicles and secretory
ments were performed under such conditions that only an
granules [1 Á/4]. They are involved in intracellular ion traffic
electrogenic influx of 86Rb' into chromaffin granules was
and play a vital role in cellular function. For example, loss of
measured. This simple and convenient flux assay, combined
endosome-associated chloride channel, in Dent’s disease,
with marker enzyme estimations, forms a valuable method
strongly inhibits endocytosis of low molecular weight proteins
for measuring K' channel activity of chromaffin membrane
in kidney proximal tubular cells . Mutations in the Cl(
vesicles. This kind of approach together with single channelmeasurements after reconstitution into planar lipid bilayerallows us to study new properties of the chromaffin granules
*To whom correspondence should be addressed.
e-mail: [email protected]
ISSN 0968-7688 print/ISSN 1464-5203 online # 2004 Taylor & Francis Ltd
In this paper we report that the large conductance
potassium channel present in chromaffin granule mem-
[K+] > > [K+]
branes is regulated by pH. The potassium channel was
investigated both by 86Rb' ion flux measurements and
single channel recordings after reconstitution into blacklipidmembrane (BLM). Both techniques showed that the chro-maffin granule potassium channel is inhibited by the acidic
pH, probably from the intragranular side.
[K+] = [K+]
[% of total radioactivity]
Regulation of 86Rb' uptake into chromaffin granules by pH
The principle of the applied flux assay was originally
described by others [16,17]. In brief, we prepared chromaffingranule vesicles containing an inner concentration of
100 mM KCl. Shortly before the assay external K' wasreplaced with Tris'. As a result of a K' gradient, an
electrical diffusion potential was established in vesicles
containing active K' channels. The addition of
isotope, a K' analog, to the external solution, led to the
uptake of Rb' due to its equilibration with the membrane
potential, but not affecting the level of the potential itself. It
is important to note that 86Rb' accumulation occurs selec-tively into the vesicles containing open K' channels thus
[fold of stimulation
enhancing the sensitivity of the transport measurements.
Figure 1 (a) presents the time course of 86Rb' uptake into
chromaffin granule vesicles (expressed as the percentage of
total radioactivity present in the sample). Addition of 30 mM
KCl, which caused depolarization of the diffusion potential,promoted a rapid efflux of 86Rb' from the vesicles. In the
86Rb' uptake into chromafﬁn granules and its regulation
absence of a K' gradient (no diffusion potential was created)
by pH. (a) Time course of 86Rb' uptake into chromafﬁn granules.
After addition of 86RbCl, accumulation of radioactivity was measured
accumulation of 86Rb' was low (Figure 1 (a)). This result
as described in ‘‘Materials and Methods’’ (j). At the time indicated
suggests that the K' transport pathway operates by an
by an arrow, 30 mM KCl was added to the reaction mixture (m).
electrogenic rather than electroneutral mechanism.
Accumulation of radioactivity without removal of external potassium
Figure 1 (b) demonstrates the effect of the pH of the
is also shown ('). Values are means9/S.D. for triplicate determina-
incubation medium on 86Rb' uptake into chromaffin granule
tion. Measurements were performed at 208C. (b) Effect of pH on
86Rb' uptake into chromafﬁn granules. The uptake was measured
vesicles. A large inhibition of 86Rb' uptake was observed in
at different pH values as described under Materials and Methods.
a medium of pH below 7.0. In order to verify whether the
The 86Rb' uptake at pH 7.0 was taken as unity. Values are means9/
observed effect was specific to chromaffin granule mem-
branes, a similar experiment was performed with beef heartsubmitochondrial particles (SMP), known to have only aslightly pH-dependent electrogenic potassium transport. In
KCl after addition of the vesicles into the trans side. Anion
fact, no inhibition of 86Rb' transport was observed in beef
selective channels were observed in only Â/2% of all
experiments. The potassium channels were usually similarin amplitude and gating behavior from experiment to experi-
Reconstitution of chromaffin granule membranes into planar
ment. Examples of single-channel recordings of the potas-
sium channel are illustrated in Figure 2. We calculated thecurrent/voltage (I/V) relation from the mean amplitude of the
Figure 2 shows current changes upon reconstitution of
channels currents at different potentials as 3609/7 pS (n 0/
chromaffin granules into a planar lipid bilayer. The quality
22) in a KCl gradient (450/150 mM KCl) (Figure 2 (a)) and
of the lipid bilayers was checked at 0 mV and 9/50 mV before
addition of chromaffin granule membranes suspension and
9/9 pS (n 0/7) in symmetric 450 mM KCl (data not
showed no channel-like activity. Electrically silent mem-
shown), both at pH 7.0. Channel had an ohmic behavior
between '/70 and (/70 mV, both in symmetrical solution
of 10 pS. We routinely observed a positive current at 0 mV,
(open squares) and in presence of 450/150 mM KCl ionic
gradient (closed squares) (Figure 2 (b)). The selectivity of the
/50 minutes after addition of chromaffin granule ‘‘ghosts’’
to the trans -bilayer chamber. These could be identified as
observed fluxes through this channel was investigated with
due to a cation selective channel by the direction of current
asymmetric KCl solutions. The obtained reversal potential
flow in the presence of an ion gradient 450 mM KCl/150 mM
was Urev 0/(/289/2 mV, indicating that the channel is highly
Calculated probability of opening for 50 mV in symmetrical450 mM KCl was Popen0/0.519/0.15 whereas for (/50 mVwas only Popen0/0.119/0.07. Such a result clearly indicatedthat potassium channel with large conductance is voltagedependent. We also performed gating analysis of singlechannel recording at 50 mV and (/50 mV in symmetrical450 mM KCl, pH 7.0. Single current amplitudes at 9/50 mVobtained from histograms had the same values ( Â/229/1 pA)whereas strong differences were observed in the closed- andopen-time distributions. The calculated open-times for thelarge conductance potassium channel investigated are:to 0/9.749/0.37 ms at 50 mV but only to 0/3.859/0.31 ms at
(/50 mV. The effect was more evident for the closed-times:
tc0/10.189/0.38 ms at 50 mV and tc0/31.989/0.92 ms at
(/50 mV, suggesting that the channel is voltage dependent.
The minimum number of open and closed states entered byKCG channel can be estimated determining the number ofexponential components necessary to fit the observe open-and closed- time distribution . Open- and closed-timehistograms of the KCG channel in 90% of experiments clearlyrequired only one time constant to fit the points. In remaining10% of experiments second lifetimes-lower than 3 ms weregenerated but they determined not significant (only Â/1%)per cent of all. Such a result points that investigated channelhas only one population of open- and closed times.
Charybdotoxin has no effect on the potassium channel
activity upon application to either the cis or trans side (datanot shown).
Figure 3 shows KCG channel activity at holding potential
40 mV in control experimental conditions and after additionTEA' to the trans and cis sides. Experiments wereperformed in ion gradient concentration of 450 mM KCl/150 mM KCl, pH 7.0. 15 mM TEA' applied to the cis sideclearly blocked the channel. The same TEA' concentration
Reconstitution of chromafﬁn granule large conductance
potassium channel into planar lipid bilayer. (a) Single channelrecordings at different holding potentials in asymmetric ionic condi-tions (450/150 mM KCl, pH 7.0 cis /trans ). Closed levels are markedas c. Recordings were low Á/pass ﬁltered at 200 Hz. (b) Current/voltage relation for single channel recordings in symmetric andasymmetric ionic conditions. The stright line was ﬁtted to experi-mental data; in symmetric 450 mM KCl (I), and in asymmetricconditions 450/150 mM KCl cis/trans (j). The observed single-channel conductances are 4329/9 pS for symmetric and 3609/7 pSfor asymmetric ionic conditions.
Effect of TEA' on single channel activity. Channel
selective for cations (the K' Nernstian potential for this ionic
recordings at holding potential of 40 mV under control conditions in
asymmetric 450 mM KCl/150 mM KCl, pH 7.0 cis /trans ) and with
15 mM TEA' applied to the trans and cis sides as indicated. Closed
Probability of opening was lower for single channel
levels are marked as c. Recordings were low Á/pass ﬁltered at
activity when negative potential was applied to the bilayer.
placed into the trans -bilayer chamber results in current
increased to tc0/8.89/0.3 ms but after lowering it in the cis
amplitude reduction at 20% without changing open prob-
side it increased to tc0/30.59/2.9 ms.
ability and gating. These findings pointed on the fact that
Lowering of the pH in the cis side resulted in diminishing of
TEA'-sensitive channel was located in resealed ‘‘ghost’’
both open probability and the amplitude of opening. Only
membranes and its’ sensitivity is higher from the cis side
weakinfluence on single channel current amplitude and open
named as an intragranular side. These results are consistent
probability were observed when pH 6.4 was applied from the
with one presented by Ashley et al . (1994).
We performed gating analysis of single channel recording
Figure 5 shows the effect of pH on the open probability of
in control and after addition 15 mM TEA' to the trans -bilayer
the large conductance potassium channel. The significant
chamber. As occurred, the calculated open- and closed-
difference in probability of opening at the level PopenB/0.05
times stayed unchanged. Open- and closed- times for the
(test T-Student) was observed only after lowering pH from
large conductance potassium channel at 40 mV in gradient
the cis side when compared with open probability at pH 7.0
at both the cis and trans sides as a control. Upon changing
o 0/14.339/0.24 ms and tc 0/13.789/1.78 ms in
the pH from 7.0 to 6.4 from the cis side the open probability
o 0/14.439/0.26 ms and tc 0/9.329/1.74 ms after
15 mM TEA' addition to the trans -bilayer chamber.
decreased from Popen0/0.619/0.21 to Popen0/0.139/0.11.
Regulation of chromaffin granule large conductancepotassium channel by pH
Protons influence the behavior a variety of ion channelsincluding potassium channels [23 Á/26]. Unitary conductance
Figure 4 shows single channel recordings at holding potential
of BK channels is also reduced by protons [27,28]. The
of 30 mV after reconstitution of chromaffin granule mem-
magnitude of K' current probably is controlled by glutamate
branes into lipid bilayers in the gradient of 450/150 mM KCl
residues present there [25,29]. Recently, it was shown that
(cis /trans ) and kinetic analysis performed for activity of KCG
low pH reduced the unitary current in a voltage dependent
channel in such experimental conditions. Figure 4 (a) shows
manner-increasing with the membrane depolarization .
channel recordings after lowering pH from 7.0 to 6.4 only at
Heterologously expressed hSlo1 BK channels were also pH-
the cis side and again change to the pH 7.0. Open- and
closed- time histograms for various pH values of the
The value of the presented results and interpretation
experimental solution in the trans and cis sides are
of the data are due to application of two different transport
presented in Figure 4 (b). Considerable differences in the
methods reflecting K' conductance in chromaffin granule
lifetimes were observed at pH 6.4 at the cis side. In the
membranes. The radioactive flux assay, applied in this
control pH 7.0 in both compartments the closed lifetime was
report, was successfully used to show amiloride-block-
tc0/7.79/0.3 ms. After lowering the pH in the trans side it
able Na' channels in toad bladder microsomes, and
Regulation of chromafﬁn granule large conductance potassium channel by pH. (a) Single channel recording at a holding potential of
30 mV from ‘‘ghost’’ membranes incorporated into bilayer in the presence of gradient 450/150 mM (cis /trans ) KCl and at pH 7.0 or 6.4. A changein channel kinetics at pH 6.4 at the cis side. The pH at the trans side was always 7.0. The closed levels, corresponding to the current through thelipid bilayer, are indicated with c. Recordings were low Á/pass ﬁltered at 200 Hz. (b) Gating analysis of large conductance potassium channelrecordings at different pH. Open- and closed- time analysis of the channel recordings at various pH values at the trans and cis sides. ThepH values at the trans and cis sides are marked below the diagram. The signiﬁcant difference in the time constant values are marked by asterisk.
Mean open- (to) and closed (tc) lifetimes are indicated in ms. Data are means9/SD (n 0/7).
both insensitive to ChTX and Ca2'. Our results indicate thatthe channel activity observed in the present paper wassimilar to one described by Ashley et al . (1994). Similar to86Rb' flux experiments we observed a strong inhibition ofthe K' channel by low pH. Interestingly, this strong effectwas observed only from the cis side. The inhibition of the K'channel by lowering pH was observed from the same side asTEA' inhibition. Previously it was shown that a chromaffingranule K' channel is blocked by TEA' from the intragra-nular side . This suggests that the effect of pH is alsofrom the intragranular side. The findings of the present studymay be important for our understanding of the physiologicalrole of potassium conductance in chromaffin granules.
The chromaffin granule membrane contains a vacuolar-
type (V-type) H'-ATPase which generates an electrochemi-cal proton gradient, acidifying the granule interior . Thepotassium channel may play an important physiological role
Effect of pH on the open probability of chromafﬁn granule
large conductance potassium channel. Changes of the pH values at
by compensating for the electric charge transfer produced by
the trans and cis sides are marked below the diagram. Columns and
the V-ATPase . This would enable formation of a
error bars indicate means9/SD (n0/7). The signiﬁcant difference in
membrane potential and DpH, sufficient to drive catechola-
the probability of opening is marked by asterisk. PopenB/0.05 (test T-
mine uptake into the chromaffin granules. This hypothesis is
Student) compared with open probability at pH 7.0 at both the cisand trans sides.
also supported by experiments on the effects of intra-granular cation composition on ATP-dependent acidification
veratridine-activated tetrodotoxin-blockable Na' channels in
of chromaffin granules . In fact, a much higher DpH was
rat brain synaptic membranes [16,17]. The 86Rb' flux
observed with K' inside than with TEA . Our present
method, was previously used to study K' transport in
observation on the pH-dependence of K' transport points to
the fact that low pH should blockthis ‘‘charge compensation’’
‘‘concentrative uptake’’, was also applied to measure the
mechanism. Blockage of K' channels by low pH would block
activity of Cl( channels from Torpedo californica elektroplax
further acidification of chromaffin granules. This channel be
plasma membrane . The principle of the assay is as
involved in protective mechanism to prevent over-acidifica-
follows. Chromaffin granule vesicles are prepared to contain
tion of the granular lumen. Regulation of chromaffin granuleK' transport by pH could be also important during granule
a high concentration of KCl. Shortly before the assay, the
swelling, playing a role, e.g., in the fusion of the granule with
external potassium is replaced by the relatively impermeant
the plasma membrane. In fact, chromaffin granule swelling
Tris' ion. As a consequences of the potassium gradient
has been observed to be regulated by internal pH .
created, an electrical diffusion potential is set up, the
In conclusion, the results of our investigation support the
magnitude of which is determined by the permeabilities of
concept of the existence of an electrogenic, pH-regulated K'
K', Cl( and Tris' through the membrane. Only in the
transport system in chromaffin granules. Our results support
vesicles containing active potassium channels is the K'
the previous observations that K' channels are present in
permeability likely to be much greater than the Cl( and Tris'
chromaffin granule membranes [11,12]. Such a system
permeabilities, and hence in these vesicles only a potassium
appears to fulfill an important physiological role in the
diffusion potential, interior negative, is formed. An isotope
creation of transmembrane potential (DC) and a proton
that permeates through the channel (in our case 86Rb')
concentration gradient (DpH) across the granule membrane
when added to the exterior solution will tend to equilibrate
with the membrane potential and thus will accumulate in thevesicles that have formed a membrane potential (DC).
We observed that 86Rb' transport into chromaffin gran-
ules is strongly pH-dependent. A large inhibition of 86Rb'
uptake in a medium of pH below 7.0 was observed. Thisresult strongly indicated an effect of pH on the K' channel
Subcellular fractionation of adrenal glands
present in chromaffin granules. However a contribution of
Bovine adrenal medullas were fractionated essentially as previously
86Rb'/K' electroneutral exchange or changes of chloride
described by . Purification of chromaffin granules and the purity
conductance in the observed effect cannot be excluded.
of membrane preparations were confirmed by a marker enzymeestimation as previously described .
Hence, further studies on single channel activities of thechromaffin granule K' channels and their regulation by pHwere performed.
Reconstitution of chromaffin granule membranes revealed
The isotope flux through ion-conducting pathways was performed
the presence of a potassium channel with a conductance of
essentially as described by Garty et al . [16,17]. Application of 86Rb'
Â/430 pS in symmetric 450 mM KCl. Previously, two different
flux for K' transport measurements in chromaffin granules experi-
K' channels were described in chromaffin granules [11,12],
ments was described previously .
Granule membrane marker cytochromu b561 activity was also
measured by the difference between dithionite-reduced and oxidizedstates at 429 nm .
 Kourie, J. I., 1997, Chloride channels in the sarcoplasmic
reticulum of muscle. Prog. Biophys. Mol. Biol. , 68, 263 Á/300.
 Szewczyk, A., 1998, The intracellular potassium and chloride
channels: properties, pharmacology and function. Mol. Membr.
Biol. , 15, 49 Á/58.
The planar lipid membrane was formed by spreading phospholipid
 O’Rourke, B., 2000, Pathophysiological and protective roles of
solution (painted bilayer). Planar phospholipid bilayers were formed
mitochondrial ion channels. J. Physiol. , 529, 23 Á/36.
in a 250 mm diameter hole which separated two chambers (cis 2 and
 The´venod, F., 2002, Ion channels in secretory granules of the
trans 3 ml internal volume). The chambers contained 450/150 or
pancreas and their role in exocytosis and release of secretory
450/450 mM KCl, 5 mM Hepes, pH 7.0 (adjusted with KOH). The
proteins. Am. J. Physiol. , 283, C651 Á/672.
outline of the aperture was coated with lipid suspension and dried
 Jentsch, T. J., Friedrich, T., Schriever, A. and Yamada, H.,
with N2 prior to bilayer formation to improve membrane stability.
1999, The CLC chloride channel family. Pﬂu¨gers Arch. , 437,
Planar phospholipid bilayers were painted using L-a-Lecithin in
n-decane at a final concentration of 30 mg of lipid/ml. Formation
 Kornak, U., Kasper, D., Bosl, M. R., Kaiser, E., Schweizer, M.,
and thinning of the bilayer were monitored by capacitance measure-
Schulz, A., Friedrich, W., Delling, G. and Jentsch, T. J., 2001,
ments. Final capacitance values ranged from 110 to 200 pF.
Loss of the ClC-7 chloride channel leads to osteopetrosis in
Electrical connections were made using Ag/AgCl electrodes and
mice and man. Cell , 104, 205 Á/215.
agar salt bridges (3 M KCl) to minimize liquid junction potentials.
 Szewczyk, A. and Marban, E., 1999, Mitochondria: a new target
Voltage was applied to the cis compartment of the chamber and the
for K' channel openers? Trends Pharmacol. Sci. , 20, 157 Á/
trans compartment was grounded. Suspensions of chromaffin
granules in 300 mM sucrose, 10 mM Hepes, pH 7.2 (adjusted with
 Pazoles, C. J. and Pollard, H. B., 1978, Evidence for stimulation
KOH) were added to the trans compartment. Changes of pH in the
of anion transport in ATP-evoked transmitter release from
cis and trans chambers were obtained by addition of a fixed amount
isolated secretory vesicles. J. Biol. Chem. , 253, 3962 Á/3969.
of HCl or KOH. All measurements were carried out at room
 Pollard, H. B., Stopak, S. S., Pazoles, C. J. and Creutz, C. E.,
1979, A simpliﬁed, one-step method for radiometric analysis ofphenylethanolamine-N-methyltransferase in adrenal chromafﬁncells. Anal Biochem. , 99, 281 Á/282.
 Picaud, S., Marty, A., Trautmann, A., Grynszpan-Winograd, O.
and Henry, J. P., 1984, Incorporation of chromafﬁn granule
The current was measured using a Bi-layer Membrane Admittance
membranes into large-size vesicles suitable for patch-clamp
Meter (model ID 562, IDB, Gwynadd, UK). Signal was filtered at
0.2 kHz (Low Pass Bessel Filter 4 Pole, Warner Instrument Corp.),
 Arispe, N., Pollard, H. B. and Rojas, E., 1992, Calcium-
digitized (A/D converter 1401, Cambridge Electronic Design, UK)
independent K'-selective channel from chromafﬁn granule
and transferred to a PC for off-line analysis by CED Electrophysiol-
ogy Package V6.41 and pClamp6 software (Axon Instruments) and
 Ashley, R. H., Brown, D. M., Apps, D. K. and Phillips, J. H.,
1994, Evidence for a K' channel in bovine chromafﬁne granule
Channel conductances were obtain as follows. In experiments all
membranes: single-channel properties and possible bioener-
observable current steps were measured for each voltage and their
getic signiﬁcance. Eur. Biophys. J. , 23, 263 Á
amplitudes were plotted versus voltage. Channel conductances were
 Kirshner, N., Corcoran, J. J., Caughey, B. and Korner, M., 1987,
then derived from the slopes of the linear regression lines to the data
Chromafﬁn vesicle function in intact cells. Ann. N. Y. Acad. Sci. ,
points. The conductances obtained for each individual experiment
were then averaged over the number of experiments performed
 Angeletti, R. H., Nolan, J. A. and Zaremba, S., 1985, Catecho-
under the same conditions. Results are expressed as 9/S.Ds.
lamine storage vesicles: topography and function. Trend.
Ion selectivity was measured under 450/150 mM KCl (cis/trans )
asymmetrical conditions. The reversal potential U
 Arispe, N., De Mazancourt, P. and Rojas, E., 1995, Direct
a voltage for which the corresponding linear regression intersected
control of a large conductance K'-selective channel by
the horizontal axis of I Á/V curves, and was compared to the value
G-proteins in adrenal chromafﬁn granule membranes. J.
calculated for K' according to Nernst equation .
Histograms were made from more than 5000 events (open or
 Garty, H., Rudy, B. and Karlish, S. J. D., 1983, A simple and
closed states of the channel). Open probability of the channel (Popen)
sensitive procedure for measuring isotope ﬂuxes through ion-
was determined experimentally by calculating the mean fraction of
speciﬁc channels in heterogenous populations of membrane
vesicles. J. Biol. Chem. , 258, 13094 Á/13099.
 Garty, H. and Karlish, S. J. D., 1989, Ion channel-mediated
ﬂuxes in membrane vesicles: selective ampliﬁcation of isotope
uptake by electrical diffusion potential. Methods Enzymol. , 172,
86RbCl, with a specific radioactivity of 20 Ci/mmol, was purchased
from Polatom (Poland). L-a-Lecithin from soybean, potassium
 Szewczyk, A., Lobanov, N. A., Nowotny, M. and Nalecz, M. J.,
chloride, tetraethylammonium chloride, calcium chloride, n Á
1997, Interaction of sulfhydryl reagents with K' transport in
and chloroform were from Sigma (USA). All other chemicals were of
adrenal chromafﬁn granules. Acta Neurobiol. Exp. , 57, 329 Á/
the highest purity commercially available.
 Lobanov, N. A., Szewczyk, A., Wojcik, G., Nowotny, M. and
Nalecz, M. J., 1997, Effects of K' channel inhibitors onpotassium transport in bovine adrenal chromafﬁn granules.
Biochem. Mol. Biol. Int. , 41, 678 Á/686.
 Szewczyk, A., Lobanov, N. A., Kicinska, A., Wojcik, G. and
This study was supported partially by the Nencki Institute of
Nalecz, M. J., 2001, ATP-sensitive K' transport in adrenal
Experimental Biology, grant No. 6 P203 003 04 from the State
chromafﬁn granules. Acta Neurobiol. Exp. , 61, 1 Á/12.
Committee for Scientiﬁc Research to Adam Szewczykand by the
 Brocklehurst, K. W. and Pollard, H. B., 1990, Cell biology of
fellowship from UNESCO/MCBN to Nikolai A. Lobanov. This study
secretion. In J. C. Hutton and K. Siddle, eds Peptide Hormone
was also supported by a grant from the Agricultural University
Secretion. A Practical Approach (Oxford, New York, Tokyo,
 Heinemann, S. H., 1995, Guide to data acquisition and analysis.
 Brelidze, T. I., Niu, X. and Magleby, K. L., 2003, A ring of eight
In B. Sakmann and E. Neher, eds Single channel recording
conserved negatively charged amino acids doubles the con-
(Plenum Press, New York), pp. 53 Á/90.
ductance of BK channels and prevents inward rectiﬁcation.
 Geiger, D., Becker, D, Lacombe, B. and Hedrich, R., 2002,
Proc. Natl. Acad. Sci. U S A. , 22, 9017 Á/9022.
Outer pore residues control the H('/) and K('/) sensitivity of the
 Avdonin, V., Tang, X. D. and Hoshi, T., 2003, Stimulatory action
Arabidopsis potassium channel AKT3. Plant Cell , 14, 1859 Á/
of internal protons on Slo1 BK channels. Biophys. J. , 84, 2969 Á/
 Lopes, C. M., Gallagher, P. G., Buck, M. E., Butler, M. H. and
 Goldberg, A. F. and Miller, C., 1991, Solubilization and
Goldstein, S A., 2000, Proton blockand voltage gating are
functional reconstitution of a chloride channel from Torpedo
potassium-dependent in the cardiac leakchannel Kcnk3. J. Biol.
californica electroplax. J. Membr. Biol. , 124, 199 Á/206.
 Forgac, M., 1989, Structure and function of vacuolar class of
 Nimigean, C. M., Chappie, J. S. and Miller, C., 2003, Electro-
ATP-driven proton pumps. Physiol. Rev. , 69, 765 Á/796.
static tuning of ion conductance in potassium channels.
 Ornberg, R. L., Furuya, S., Goping, G. and Kuijpers, G. A.,
1995, Granule swelling in stimulated bovine adrenal chromafﬁn
 Brelidze, T. I. and Magleby, K. L., 2004, Protons blockBK
cells: regulation by internal granule pH. Cell. Tissue Res. , 279,
channels by competitive inhibition with K('/) and contribute to
the limits of unitary currents at high voltages. J. Gen. Physiol. ,
 Zinder, O., Hoffman, P. G., Bonner, W. M. and Pollard, H. B.,
1978, Comparison of chemical properties of puriﬁed plasma
 Laurido, C., Candia, S., Wolff, D. and Latorre, R., 1991, Proton
membranes and secretory vesicle membranes from bovine
modulation of a Ca(2'/)-activated K' channel from rat skeletal
adrenal medulla. Cell. Tissue Res. , 188, 153 Á/70.
muscle incorporated into planar bilayers. J. Gen. Physiol. , 98,
 Hille, B., 1992, Ionic Channels of Excitable Membranes
(Sinauer Associates, Sunderland, MA).
 Habartova, A., Krusek, J. and Zemkova, H., 1994, Sensitivity of
high-conductance potassium channels in synaptosomal mem-branes from the rat brain to intracellular pH. Eur. Biophys. J. ,23, 71 Á/77.
Received 4 May 2004; and in revised form 24 June 2004.
The British Journal of Diabetes & Vascular Disease Metformin European Association for the Study of Diabetes, Amsterdam, The Netherlands, September 2007 British Journal of Diabetes & Vascular Disease 2007 7: 247The online version of this article can be found at: can be found at: The British Journal of Diabetes & Vascular Disease Additional services and information fo
For Faces Almost Lost to History, a Chance to Speak The 19th and 20th centuries saw the invention of a lot of new ways to do away with people. But the same 200 years ensured that more people might be remembered, thanks to new recording devices, photography most of all. Many who might once have completely disappeared from view now leave behind visible traces of themselves. But what do these