membranes
Article
Investigating Sterol and Redox Regulation of
the Ion Channel Activity of CLIC1 Using
Tethered Bilayer Membranes
Heba Al Khamici 1 , Khondker R. Hossain 1,2 , Bruce A. Cornell 3 and Stella M. Valenzuela 1, *
1
2
3
*
School of Life Sciences, University of Technology Sydney, Sydney 2007, Australia;
[email protected] (H.A.K.); [email protected] (K.R.H.)
Australian Centre for Neutron Scattering, Australian Nuclear Science and Technology
Organisation (ANSTO), NSW 2234, Australia
Surgical Diagnostics Pty Ltd., Sydney 2069, Australia; [email protected]
Correspondence: [email protected]; Tel.: +61-2-95141917
Academic Editor: Terry Hébert
Received: 31 October 2016; Accepted: 5 December 2016; Published: 8 December 2016
Abstract: The Chloride Intracellular Ion Channel (CLIC) family consists of six conserved proteins
in humans. These are a group of enigmatic proteins, which adopt both a soluble and membrane
bound form. CLIC1 was found to be a metamorphic protein, where under specific environmental
triggers it adopts more than one stable reversible soluble structural conformation. CLIC1 was found
to spontaneously insert into cell membranes and form chloride ion channels. However, factors that
control the structural transition of CLIC1 from being an aqueous soluble protein into a membrane
bound protein have yet to be adequately described. Using tethered bilayer lipid membranes and
electrical impedance spectroscopy system, herein we demonstrate that CLIC1 ion channel activity is
dependent on the type and concentration of sterols in bilayer membranes. These findings suggest
that membrane sterols play an essential role in CLIC1’s acrobatic switching from a globular soluble
form to an integral membrane form, promoting greater ion channel conductance in membranes.
What remains unclear is the precise nature of this regulation involving membrane sterols and
ultimately determining CLIC1’s membrane structure and function as an ion channel. Furthermore,
our impedance spectroscopy results obtained using CLIC1 mutants, suggest that the residue Cys24 is
not essential for CLIC1’s ion channel function. However Cys24 does appear important for optimal ion
channel activity. We also observe differences in conductance between CLIC1 reduced and oxidized
forms when added to our tethered membranes. Therefore, we conclude that both membrane sterols
and redox play a role in the ion channel activity of CLIC1.
Keywords: CLIC; chloride intracellular ion channel proteins; tethered lipid membranes;
cholesterol; ergosterol
1. Introduction
The Chloride Intracellular Ion Channel (CLIC) family members contain no obvious
transmembrane domain in their protein structure; nevertheless, they are capable of inserting
into phospholipid membranes directly from their soluble state, where they can function as ion
channels [1,2]. This has allowed for their ease of study using a range of artificial lipid membrane
systems including tip-dip and tethered bilayer membranes (tBLMs) [3,4]. Reported conductance levels
from single channel recordings of the ion channel activity of recombinant CLIC1 (rCLIC1) added
to artificial membrane systems and from cells over-expressing rCLIC have been varied [5,6]. For
example, single channel conductance recordings of rCLIC1 added to bilayer membranes made from
Membranes 2016, 6, 51; doi:10.3390/membranes6040051
www.mdpi.com/journal/membranes
Membranes 2016, 6, 51
2 of 13
asolectin, phosphatidylethanolamine (PE) and phosphotidylserine (PS) lipids were 60 pS and 120 pS.
Its conductance in membranes containing PC lipids was 28 pS [4,7], while single channel conductance
levels recorded from CLIC1-transfected CHO-K1 cells were 8 pS and 16 pS [4,7–9].
Similar variations in the conductance levels of CLIC4 were also reported following patch clamp
measurements. Reported CLIC4 single channel activity was higher from bilayer membrane recordings
using reconstituted brain microsomes [10,11], as compared to recordings obtained from adding
rCLIC4 to artificial membranes containing neutral lipids only [11–14]. Therefore, it has been proposed
that CLIC protein channel activity is sensitive to membrane lipid composition in addition to previously
demonstrated external membrane factors, including redox environment and pH [7,10]. The localization
of the protein within membranes was also varied, with CLIC4 found to localize to cholesterol rich
micro-domains called caveolae [15–17], which led to the proposal that CLIC proteins may only be
functional in membranes containing cholesterol. In a study using phospholipid vesicle chloride
efflux assays by Tulk et al., 2002 [7], it was found that CLIC1 demonstrated no ion channel activity
in membranes containing pure neutral lipid mixtures, while activity was greater in membranes
containing 10% of a negatively charged phospholipid such as phosphotidylserine (PS) or phosphatidic
acid (PA). Increasing cholesterol concentration to 30 mol % in these membranes caused the channel
activity of CLIC1 to be suppressed [7]. Conversely, a study by Singh, et al. [10] using Langmuir-film
monolayers and patch clamping techniques, shows that membranes containing POPE, POPS and
cholesterol in a molar ratio of 4:1:1 induced CLIC1 membrane insertion and ion channel conductance.
However, in both these studies the role of cholesterol in CLIC1’s membrane insertion or its channel
conductance activity was not explored. Subsequently, Valenzuela et al., [3] confirmed that CLIC1 ion
channel conductance is optimal when inserted into membranes containing cholesterol. This increased
conductance can be correlated with an increase in membrane insertion of the protein, as we have
previously demonstrated by Langmuir monolayer studies that indicate both sterol and phospholipid
membrane composition, regulate CLIC1 membrane insertion [18].
Here-in, we further investigate the effect of two different sterols (mammalian sterol, cholesterol
and the fungal sterol, ergosterol) on the membrane conductance activity of CLIC1. Membrane sterols
are known to affect the stability of ion channels and pore proteins in membranes which is an important
factor in maintaining the rate of ion transport across membranes [19]. The effect of these sterols on
the ion channel conductance of CLIC1 can indicate a potential role for CLIC1 as an anti-fungal agent,
acting in a manner similar to nystatin A and amphotericin B. The approach we have employed in
this study is to use tethered bilayer lipid membranes (tBLMs) and Electrical Impedance Spectroscopy
(EIS). tBLMs have increasingly become the tools of choice for the study of protein and membrane
interactions. This is due to their ease of use, stability and the ability to tailor their lipid composition
in order to mimic natural cell membranes [20–22]. The influence of the oxidation state of the protein
CLIC1, on its conductance activity was also assessed, using monomer (reduced) and dimer (oxidized)
versions of the protein.
2. Results and Discussion
2.1. Conductance Properties of CLIC1 Reduced Monomer and CLIC1 Oxidised Dimer in tBLMs
In order to study the effect of cholesterol on CLIC1’s ion channel function, rCLIC1 (reduced and
oxidised forms) were added to sealed tBLMs containing phospholipids and varied concentrations
of cholesterol, where the initial conductance of all the freshly formed membranes was stabilised to
a baseline value of less than 1 µS and with a capacitance ranging between 20 and 23 nF. Both the
conductance and capacitance of the tBLMs were measured using EIS, as previously described [3].
Figure 1 shows the results for CLIC1 reduced monomer (protein had been purified and
subsequently stored in the presence of 0.5 mM TCEP) and CLIC1 dimer (protein purified and
subsequently stored in the presence of 2 mM H2 O2 ). Both showed little to no conductance in
membranes containing no cholesterol. Conduction levels were similar to the control, in which no
Membranes 2016, 6, 51
3 of 13
protein was added to membranes containing 50 mol % cholesterol. These data are in agreement with
our previous study using monomer rCLIC1 [3]. In membranes containing 6.25 mol % cholesterol,
both rCLIC1 monomer and dimer samples showed increases in their conductance. The conductance
appears to increase
proportionally as the level of cholesterol increased from 12.5 mol % to
25 mol % and
Membranes 2016, 6, 51
3 of 12
was highest in membranes containing 50 mol % cholesterol. ANOVA and regression analysis confirm
to increase
proportionally
the level
of cholesterol
increased
from 12.5 mol
% to 25
mol % monomer
a significantappears
(p value
< 0.005)
difference as
exists
among
all of the
tests samples:
CLIC1
reduced
and was highest in membranes containing 50 mol % cholesterol. ANOVA and regression analysis
and CLIC1 confirm
oxidised
dimer samples in membranes with different cholesterol concentrations, compared
a significant (p value < 0.005) difference exists among all of the tests samples: CLIC1 reduced
to CLIC1 conductance
measured
in membranes
containing
no cholesterol
Figurecholesterol
1). However, there
monomer and
CLIC1 oxidised
dimer samples
in membranes
with (see
different
concentrations,
compared
to CLIC1the
conductance
measured inand
membranes
no cholesterol containing
was no significant
difference
between
CLIC1 monomer
dimercontaining
within membranes
(see Figure 1). However, there was no significant difference between the CLIC1 monomer and dimer
varying amounts
of cholesterol.
within membranes containing varying amounts of cholesterol.
Figure 1. Conductance of CLIC1 in tBLMs containing varying amounts of cholesterol. 20 µ g of CLIC1
Figure 1. Conductance
of pre-incubated
CLIC1 in tBLMs
containing
of cholesterol.
oxidised dimer was
with 2 mM
H2O2; 20 µvarying
g of CLIC1amounts
reduced monomer
was pre- 20 µg of
incubated
withwas
0.5 mM
TCEP in 100 µ Lwith
of HEPES/KCl
(pH
6.5)
prior
to
adding
to
tethered
CLIC1 oxidised
dimer
pre-incubated
2 mM Hbuffer
O
;
20
µg
of
CLIC1
reduced
monomer was
2 2
bilayer membranes containing 0, 6.25, 12.5, 25 and 50 mol % cholesterol concentrations. Control is
pre-incubated
with 0.5 mM TCEP in 100 µL of HEPES/KCl buffer (pH 6.5) prior to adding to tethered
membrane with 0 mol % cholesterol containing 100 µ L HEPES/KCl buffer (pH 6.5) with no protein
bilayer membranes
0, 6.25,
25 and
mol independent
% cholesterol
concentrations.
added. The containing
error bars represent
the12.5,
standard
error 50
of three
impedance
spectroscopy Control is
membrane conductance
with 0 molmeasurements
% cholesterol
containing 100 µL HEPES/KCl buffer (pH 6.5) with no protein
(n = 3).
added. The error bars represent the standard error of three independent impedance spectroscopy
In Figure 2, monomeric CLIC1 is seen to have a more linear relationship between it’s
conductance measurements (n = 3).
conductance versus protein concentration. At lower concentrations (10 µ g and 20 µ g) CLIC1 reduced
monomer conduction rate is steeper (slope is 0.0769 µ S/s and R2 of 0.9978) compared to CLIC1 dimer
(slope
0.0487 µ S/s and
R2 of is
0.9664).
would
suggestlinear
that atrelationship
lower concentrations
the CLIC1
In Figure
2, =monomeric
CLIC1
seen This
to have
a more
between
it’s conductance
reduced monomer form can more readily enter the membrane and/or oligomerise to form ion
versus protein
concentration. At lower concentrations (10 µg and 20 µg) CLIC1 reduced monomer
conductive channels in artificial membranes containing cholesterol. Of note, is that the data was fitted
1/2. This suggests
conductionusing
ratea Gm
is steeper
is 0.0769
µS/s
andconduction
R2 of 0.9978)
compared
to CLIC1 dimer
vs. [CLIC](slope
a model
for the
that may arise
via a mechanism
other than
of an
multimeric
channel.
Insteadthat
an alternate
model
suggests that the the CLIC1
(slope = 0.0487
µS/sthat
and
R2assembled
of 0.9664).
This ion
would
suggest
at lower
concentrations
protein interacts with the membrane via a “detergent-like” action that modifies the diameter of prereduced monomer
form can more readily enter the membrane and/or oligomerise to form ion
existing toroidal pore ion pathways across the tBLM. Confirmation of this however requires further
conductiveinvestigation
channels in
membranes
containing
andartificial
is beyond the
scope of the current
study. cholesterol. Of note, is that the data was
1/2
fitted using a Based
Gm vs.
[CLIC]
. This
a knowledge
model for
may arise via
on our
current results
and suggests
applying the
wethe
haveconduction
of the distinct that
structural
conformations
adopted
and reduced
forms of CLIC1
[5], we speculate
that an
the slower
a mechanism
other than
that by
ofthe
anoxidised
assembled
multimeric
ion channel.
Instead
alternate model
conduction rate of the CLIC1 oxidised dimer compared to reduced monomeric CLIC1, is due to its
suggests that the protein interacts with the membrane via a “detergent-like” action that modifies the
dimer form that is stabilized by an intramolecular disulphide bond formed between Cys24 and
diameter ofCys59.
pre-existing
toroidal
pore
ion pathways
tBLM.
Confirmation
of this however
It is postulated
that the
transmembrane
form across
of CLIC1the
would
resemble
an oxidized version
of the investigation
protein, where the
CLIC1
reducedthe
form
first undergoes
a shapestudy.
change exposing a large
requires further
and
is beyond
scope
of the current
surface, that facilitates insertion into the membrane, as previously also suggested by
Based hydrophobic
on our current
results and applying the knowledge we have of the distinct structural
Goodchild et al., 2009 [23]. In the absence of a membrane, the oxidized protein forms reversible nonconformations
adopted
oxidised
and reduced
forms
of CLIC1
[5],that
wethe
speculate
that
covalent
dimersby
viathe
its exposed
hydrophobic
surface.
Our results
suggest
slower rate
of the slower
conductionconductance
rate of the
dimer
compared
to reduced
monomeric
CLIC1,
by CLIC1
oxidised oxidised
dimeric CLIC1
compared
to the reduced
monomeric
CLIC1 is due
to the is due to
fact that
when
in the dimerby
form,
hydrophobic surfaces
are masked
andformed
less likelybetween
to interact Cys24 and
its dimer form
that
is stabilized
anthese
intramolecular
disulphide
bond
with the membrane. Of note, our experiments were performed in air thus it is expected that oxidation
Cys59. It is postulated that the transmembrane form of CLIC1 would resemble an oxidized version
of the reduced protein would occur over time.
of the protein, where the CLIC1 reduced form first undergoes a shape change exposing a large
hydrophobic surface, that facilitates insertion into the membrane, as previously also suggested by
Goodchild et al., 2009 [23]. In the absence of a membrane, the oxidized protein forms reversible
non-covalent homodimers via its exposed hydrophobic surface. Our results suggest that the slower
rate of conductance by oxidised dimeric CLIC1 compared to the reduced monomeric CLIC1 is due to
Membranes 2016, 6, 51
4 of 13
the fact that when in the dimer form, these hydrophobic surfaces are masked and less likely to interact
with the membrane. Of note, our experiments were performed in air thus it is expected that oxidation
of the reduced
protein would occur over time.
Membranes 2016, 6, 51
4 of 12
Membranes 2016, 6, 51
4 of 12
Figure 2. Conductance of different concentrations of CLIC1 in tBLMs containing 25 mol % cholesterol.
Figure 2. Conductance
different
of reduced
CLIC1 monomer
in tBLMs
containingwith
25 mol
% cholesterol.
Concentrations of
of 0,
10, 20, 40concentrations
and 60 µ g of CLIC1
(pre-incubated
0.5 mM
TCEP)
and
CLIC1
oxidised
dimer
(pre-incubated
with
H
2
O
2
)
in
100
µ
L
of
HEPES/KCl
buffer
(pH
6.5)
Concentrations of 0, 10, 20, 40 and 60 µg of CLIC1 reduced monomer (pre-incubated with 0.5 mM
added
into membranes
containing 25 mol %
cholesterol
theµL
conductance
was measured
TCEP) and were
CLIC1
oxidised
dimer (pre-incubated
with
H2 O2where
) in 100
of HEPES/KCl
buffer (pH 6.5)
and linear fitting (as indicated in black for CLIC1 reduced monomer and red for oxidised dimeric
were addedCLIC1)
into membranes
containing
25
mol
%
cholesterol
where
the
conductance
was
measured
and quadratic polynomial fits were generated using Microsoft Excel 2010 (y = −0.0005x2 +
2
2
2
and linear 0.0924x
fitting+(as
indicated
in
black
for
CLIC1
reduced
monomer
and
red
for
oxidised
dimeric
0.3639, R = 0.9985 for CLIC1 monomer and y = −0.0001x + 0.0399x + 0.5761, R = 0.9857). The
bars represent
the standard
error
of three
experimental
repeats
(n = 3). Excel 2010 (y = −0.0005x2 +
CLIC1) anderror
quadratic
polynomial
fits
were
generated
using
Microsoft
0.0924x + 0.3639, R2 = 0.9985 for CLIC1 monomer and y = −0.0001x2 + 0.0399x + 0.5761, R2 = 0.9857).
Investigations into why certain proteins tend to associate with membranes containing higher
The error
bars represent
the standard
error
of three experimental
repeats
(n = 3).
cholesterol
concentration,
indicate the
involvement
of specific segments
or motifs
within the proteins
Figure
Conductance
of different concentrations
CLIC1 in tBLMs
containing 25such
mol %as
cholesterol.
themselves
that2.facilitate
interactions
with specificofmembrane
components
cholesterol at the
Concentrations
of 0,For
10,example,
20, 40 and the
60 µ interaction
g of CLIC1 reduced
monomer
(pre-incubated
with 0.5and
mMcaveolins
membrane
interface
[24].
of
the
scaffolding
protein
flotillin
InvestigationsTCEP)
intoandwhy
proteins
tend
toH2associate
membranes
CLIC1certain
oxidised dimer
(pre-incubated
with
O2) in 100 µ L ofwith
HEPES/KCl
buffer (pH 6.5) containing higher
with cholesterol rich domains in membranes [24]. The cholesterol recognition amino acid consensus
were added indicate
into membranes
25 mol % cholesterol
where segments
the conductanceor
was
measuredwithin the proteins
cholesterol (CRAC)
concentration,
thecontaining
involvement
of specific
motifs
motif
is located near the trans-membrane helix of some proteins and is represented by the
and linear fitting (as indicated in black for CLIC1 reduced monomer and red for oxidised dimeric
themselvesamino
that facilitate
with
specific
membrane
such
cholesterol
at the
acid
sequence
L/VXXXXXR/K
or
where
the components
X represents
amino
[13].
2as
CLIC1)
andinteractions
quadratic
polynomial
fitsYXXXXXR/K,
were generated
using
Microsoft
Excel 2010 (yany
= −0.0005x
+ acid
2 that
2 + 0.0399x +
CRAC
hasfora CLIC1
major
role in sequestering
into
cholesterol
0.0924x
+segment
0.3639,
= 0.9985
monomer
and y =of
−0.0001x
0.5761,
R2 = 0.9857).rich
The domains,
membrane Another
interface
[24].
For Rexample,
the interaction
theproteins
scaffolding
protein
flotillin
and caveolins
error bars represent
standard
error of
repeats (nof
= 3).
is the tryptophan
residuethemotif
found
inthree
the experimental
fusgenic protein
HIV glycophorin-41 or gp41,
with cholesterol
rich domains
in membranes
[24].
The cholesterol
recognition amino acid consensus
represented by the sequence LWYIK [25]. When the Leucine residue was substituted with Isoleucine,
Investigations
into
why
certain
proteins
tend
to
associate
with
membranes
containing
(CRAC) motif
is located
near the
helix
some
proteins
and ishigher
represented
by the
the interaction
of protein
withtrans-membrane
cholesterol was found
to be of
fully
supressed.
Replacement
of
Leucine
cholesterol concentration, indicate the involvement of specific segments or motifs within the proteins
Alanine or
Valine, resulted in or
both
mutants having weak
cholesterol
binding compared
to
wild acid [13].
amino acidwith
sequence
L/VXXXXXR/K
YXXXXXR/K,
where
the
X
represents
any
amino
themselves that facilitate interactions with specific membrane components such as cholesterol at the
type protein
[26].
Most has
human
CLIC
proteins
onlyofcontain
a GXXXG
motif,
previously
speculated
membrane
interface
[24]. For
example,
the interaction
the scaffolding
protein
flotillin
and caveolins
Another CRAC
segment
that
a major
role
innot
sequestering
proteins
into
cholesterol
rich domains, is
to bewith
involved
as the
binding site
they also
(except amino
for CLIC3)
contain the
cholesterol
richcholesterol
domains in membranes
[24].[3],
Thebut
cholesterol
recognition
acid consensus
the tryptophan
residue
motif
found
inthethe
fusgenic protein
of HIV
glycophorin-41
or
gp41, represented
conserved
motif
adjacent
to their PTMD
domain
Further
investigations
(CRAC)
motifL35WLKG
is located near
trans-membrane
helix
of some(Figure
proteins3).and
is represented
by the are
by the sequence
LWYIK
[25].
When
the Leucine
residuewhere
was
substituted
Isoleucine,
amino
acid sequence
L/VXXXXXR/K
or YXXXXXR/K,
the
X represents
any
amino
[13]. the interaction
required
however,
in order
to confirm
the
contribution
of
these
motifs
to thewith
changes
inacid
conductance
Another
CRAC segment
that hasto
acholesterol.
major
role in supressed.
sequestering proteins
into cholesterol
domains,with Alanine or
of CLIC1
in membranes
of protein with
cholesterol
wascontaining
found
be fully
Replacement
of rich
Leucine
is the tryptophan residue motif found in the fusgenic protein of HIV glycophorin-41 or gp41,
Valine, resultedrepresented
in both mutants
having
weak
cholesterol
toIsoleucine,
wild type protein [26].
by the sequence
LWYIK
[25]. When
the Leucine binding
residue wascompared
substituted with
the interaction
of protein
with contain
cholesterolawas
found to motif,
be fully supressed.
Replacement
of Leucine
Most human CLIC
proteins
not only
GXXXG
previously
speculated
to be involved as
with Alanine or Valine, resulted in both mutants having weak cholesterol binding compared to wild
the cholesterol binding
they
also
(except
for
CLIC3)
contain
conserved
motif L35WLKG
type proteinsite
[26].[3],
Mostbut
human
CLIC
proteins
not only
contain
a GXXXG
motif, the
previously
speculated
to be
involved
as the cholesterol
site [3], but
they also (except for
containhowever,
the
adjacent to their
PTMD
domain
(Figurebinding
3). Further
investigations
areCLIC3)
required
in order
motif L35WLKG adjacent to their PTMD domain (Figure 3). Further investigations are
to confirm the conserved
contribution
of
these
motifs
to
the
changes
in
conductance
of
CLIC1
in
membranes
required however, in order to confirm the contribution of these motifs to the changes in conductance
containing cholesterol.
of CLIC1 in membranes containing cholesterol.
Figure 3. Amino Acid Sequence Alignment of Human CLIC proteins showing the CRAC motif.
Highlighted in red is the GXXXG motif and in Green highlighted the LWLK motif in human CLICs.
CLIC1 (accession number: CAG46868.1), CLIC2 (accession number: CAA03948.1), CLIC3 (accession
number: NP_004660.2), CLIC4 (accession number: CAG38532.1), CLIC5 (accession number:
AAF66928.1), CLIC6 (accession number: NP_444507.1). The alignment was produced using Clustalw.
Figure 3. Amino Acid Sequence Alignment of Human CLIC proteins showing the CRAC motif.
Figure 3. Amino
Acid Sequence Alignment of Human CLIC proteins showing the CRAC motif.
Highlighted in red is the GXXXG motif and in Green highlighted the LWLK motif in human CLICs.
number: CAG46868.1),
CLIC2
CAA03948.1),
CLIC3 (accession
Highlighted inCLIC1
red (accession
is the GXXXG
motif and
in(accession
Green number:
highlighted
the LWLK
motif in human CLICs.
number: NP_004660.2), CLIC4 (accession number: CAG38532.1), CLIC5 (accession number:
CLIC1 (accession
number: CAG46868.1), CLIC2 (accession number: CAA03948.1), CLIC3 (accession
AAF66928.1), CLIC6 (accession number: NP_444507.1). The alignment was produced using Clustalw.
number: NP_004660.2), CLIC4 (accession number: CAG38532.1), CLIC5 (accession number:
AAF66928.1), CLIC6 (accession number: NP_444507.1). The alignment was produced using Clustalw.
Membranes 2016, 6, 51
5 of 13
2.2. Defining the Role of Redox Sensitive Residues within CLIC1
Membranes 2016, 6, 51
5 of 12
CLIC1 contains a total of six cysteine residues, with Cys24 and Cys59 known to form an
2.2. Defining
the Role ofbridge
Redox Sensitive
withinof
CLIC1
intramolecular
disulphide
uponResidues
oxidation
the protein. CLIC1 is also the only member
of the CLIC family
to contain
Cys59.
measuring
CLIC1
CLIC1 contains
a total
of six Experiments
cysteine residues,
with Cys24 the
and conductance
Cys59 known toofform
an mutants
intramolecular
disulphide bridge
upon
oxidation of
protein. CLIC1
alsoof
theredox
only member
of activity.
(CLIC1-C24A
and CLIC1-C59A)
were
performed
tothe
determine
the isrole
in CLIC1
the CLIC family to contain Cys59. Experiments measuring the conductance of CLIC1 mutants
In addition, the CLIC-like protein EXC-4 was also used and compared to wild type CLIC1. In Figure 4,
(CLIC1-C24A and CLIC1-C59A) were performed to determine the role of redox in CLIC1 activity. In
all proteinsaddition,
assessed,
the three
CLIC
mutants
and
EXC-4 were
conductive
in membranes
the including
CLIC-like protein
EXC-4
was also
used and
compared
to wildmore
type CLIC1.
In Figure
4,
containingallcholesterol,
compared
to the
(25 mutants
mol % cholesterol
in membrane,
withinno protein
proteins assessed,
including
the control
three CLIC
and EXC-4 were
more conductive
membranes
containing
cholesterol,conductance
compared to theofcontrol
(25wild
mol %type
cholesterol
in membrane,
with mutants
added). Statistical
analysis
comparing
CLIC1
monomer
to all CLIC1
no protein added). Statistical analysis comparing conductance of CLIC1 wild type monomer to all
revealed a significant difference in the conductance of the tested proteins as indicated by one way
CLIC1 mutants revealed a significant difference in the conductance of the tested proteins as indicated
ANOVA test
with
a ANOVA
p valuetest
of <0.0001.
Further
statistical
were performed
using
the Tukey’s
by one
way
with a p value
of <0.0001.
Further analyses
statistical analyses
were performed
using
multiple comparisons
statistical
test. Thisstatistical
revealed
noThis
significant
in conductance
the Tukey’s multiple
comparisons
test.
revealeddifference
no significant
difference in between
conductance
between
the two CLIC1and
mutants
CLIC1-C24ASimilarly,
and CLIC1-C24S.
Similarly,
CLIC1monomer
wild
the two CLIC1
mutants
CLIC1-C24A
CLIC1-C24S.
CLIC1
wild type
and
type monomer and EXC-4 conductance values were not significantly different as assessed by Student
EXC-4 conductance values were not significantly different as assessed by Student t-test.
t-test.
Figure 4. Conductance of CLIC1 mutants and EXC-4 in membranes containing 25 mol % cholesterol.
Figure 4. Conductance
of CLIC1
mutantsCLIC1-C59A;
and EXC-4 EXC-4
in membranes
25 in
mol
cholesterol.
20 µ g of CLIC1-C24A;
CLIC1-C24S;
and CLIC1containing
(WT) proteins
100%µ L
20 µg of HEPES/KCl
CLIC1-C24A;
CLIC1-C59A;
EXC-4
and
CLIC1 containing
(WT) proteins
buffer CLIC1-C24S;
(pH 6.5) were reconstituted
in tethered
bilayer
membranes
25 mol %in 100 µL
cholesterol
the6.5)
conductance
was measured and
analysis was
performed
using excel
2010 and 25 mol %
HEPES/KCl
bufferand
(pH
were reconstituted
in tethered
bilayer
membranes
containing
Graph pad prism 6. Control sample is buffer only containing 0.5 mM TCEP with no protein added to
cholesterol and the conductance was measured and analysis was performed using excel 2010 and
membrane containing 25 mol % cholesterol. The error bars represent the standard error of three
Graph padindependent
prism 6. repeats
Control
sample is buffer only containing 0.5 mM TCEP with no protein added
of conductance measurements (n = 3).
to membrane containing 25 mol % cholesterol. The error bars represent the standard error of three
Structural
of the soluble
form of the CLIC
proteins have demonstrated that they are
independent
repeatsstudies
of conductance
measurements
(n = 3).
members of the Glutathione-S-Transferase (GST) fold family of proteins and contain a monothiol,
single cysteine redox active site [27]. We recently demonstrated that members of the CLIC family
Structural
studies of the
soluble
form
of the
proteins
have
demonstrated
have oxidoreductase
activity
in their
soluble
formCLIC
and that
Cys24 in
CLIC1
was critical for that
this they are
activity [28]. Due to the presence
and conservation
of of
thisproteins
active cysteine
in a
themonothiol,
members enzymatic
of the Glutathione-S-Transferase
(GST)
fold family
and residue
contain
structure of all human CLIC proteins (Cys24 in CLIC1), the activity of the CLICs is redox sensitive
single cysteine redox active site [27]. We recently demonstrated that members of the CLIC family have
[5,28]. Experiments investigating the activity of CLIC proteins within lipid bilayers have shown that
oxidoreductase
in theirofsoluble
form in
and
that Cys24
in dependent
CLIC1 was
critical
this enzymatic
the ion activity
channel activity
CLIC proteins
membranes
is also
upon
redox for
processes
activity [28].
Due
to
the
presence
and
conservation
of
this
active
cysteine
residue
in
the
structure
of all
[5,23]. Mutation of the residue Cys24 to alanine in CLIC1 resulted in a reduction of its single
ion
channel
conductance,
compared
to
the
wild
type
protein
[10].
However
when
Cys24
was
replaced
by
human CLIC proteins (Cys24 in CLIC1), the activity of the CLICs is redox sensitive [5,28]. Experiments
a serine residue, the ion channel activity of the protein was completely eliminated [5]. This same
investigating
the activity of CLIC proteins within lipid bilayers have shown that the ion channel
study also showed a complete abolition of channel activity for the mutant C59S [5].
activity of CLIC
proteins
in membranes
also mutants,
dependent
redox
processes
[5,23].
Mutation of
In the
current study
using the twoisCLIC1
C24S upon
and C24A,
we found
that both
mutants
the residue
Cys24equally
to alanine
CLIC1
resulted
in a reduction
its single
channel
conduct
well inin
tBLMs
containing
cholesterol.
All three of
CLIC1
mutantsion
(C24A,
C24S conductance,
and
compared to the wild type protein [10]. However when Cys24 was replaced by a serine residue, the ion
channel activity of the protein was completely eliminated [5]. This same study also showed a complete
abolition of channel activity for the mutant C59S [5].
In the current study using the two CLIC1 mutants, C24S and C24A, we found that both mutants
conduct equally well in tBLMs containing cholesterol. All three CLIC1 mutants (C24A, C24S and C59A)
were found to have lower conductance levels compared to wild type CLIC1 protein in membranes
Membranes 2016, 6, 51
6 of 13
containing 25 mol % cholesterol. This suggests that the protein likely adopts a transmembrane structure
which does not depend upon the formation of a disulphide bond between Cys24 and Cys59. The lower
activity noted for both Cys24 mutants compared to WT CLIC1, supports previous findings of reduced
or abolished activity for C24 mutants [5,10]. It also supports structural studies that place the Cys24
residue at the start of the putative transmembrane domain of the protein, which is predicted to span
residues 24–46 [5,10]. Hence, it is not unexpected that mutation of this critical residue would impact
channel conductance and/or gating. Similarly, the CLIC-like protein EXC-4 from C. elegans, contains
an aspartic acid residue at the equivalent position to Cys24 found in CLIC1. As seen in Figure 4,
Exc4 also formed functional ion channels in the tBLMs containing 25 mol % cholesterol. Under the
specific conditions used for this study, it appears that redox may not be involved in regulating CLIC1
conductance or its activity once located within the membrane. There is clearly a need to further
investigate redox control of CLIC1 and to determine the role of other critical residues lining the pore
and the transmembrane domain of the channel.
2.3. Regulation of CLIC1 Conductance by Sterols in tBLMs
Given that the optimal functioning of a number of membrane proteins including CLIC1, have
been shown to be dependent upon the presence of particular membrane lipids and specifically sterols,
it was important to further explore this as a regulatory mechanism for CLIC1 channel activity. tBLMs
made with varying amounts of cholesterol or ergosterol (ranging between 0 mol % and 50 mol %)
were initially characterised. As seen in Figure 5, there were no significant changes to the tBLMs’
conductance or capacitance, made using a range of sterol concentrations. Statistical analysis (using
two way ANOVA followed by Benferroni’s multiple comparison test) for the impedance spectroscopy
measurements of membranes containing cholesterol or ergosterol, confirmed this.
Interestingly membranes containing higher ergosterol concentration become slightly better sealed
(Figure 5A). Although these differences were not found to be statistically significant, they do support
previous findings showing membranes containing ergosterol to be more rigid and tighly packed
compared to cholesterol containing membranes [19]. Also it was previously reported that cholesterol
increases the acyl chain order of phospholipids and therefore results in an increased thickness of
membranes. Our capacitance measurements of membranes containing 0 to 50 mol % cholesterol or
ergosterol, revealled no significant differences in membrane thickness (Figure 5B).
Published studies by others have shown that cholesterol concentrations higher than 20 mol %
cause an increased disorder of artificial membranes made from PC lipids [29]. This could help explain
some of the apparent fluctuations in our results. Of note is the low conductance value at 40 mol %
cholesterol, which is likely anomalous, given that 30 and 50 mol % are similar. As such, a final
concentration of between 20 and 25 mol % of cholesterol or ergosterol was routinely employed in this
study. This level also correlates with the physiological sterol content found in most animal cells [30].
The results in Figure 6 show that both rCLIC1 monomer and rCLIC1 dimer were surprisingly
greatly more conductive when added to tBLMs containing 25 mol % ergosterol compared to tBLMs
containing equivalent levels of cholesterol. When quantified this translated a 3.7 fold increase
in conductance for CLIC1 monomer and 2.8 fold increase in conductance for CLIC1 dimer when
incorporated into tBLMs containing 25 mol % ergosterol, compared to membranes containing 25 mol %
cholesterol. It has been demonstrated that at temperatures higher than 15 ◦ C the transition of POPC
membranes into the liquid ordered phase occurs at lower concentrations of ergosterol than cholesterol
and indicates that ergosterol may be a more effective promoter of raft-like domains in POPC membrane
compared to cholesterol [31,32]. This raises the possibility that sterol-raft domains in membranes aid
in the initial binding of CLIC1 to the membrane, which likely involves the structural unfolding of
CLIC1. Furthermore, sterols in membranes may play a role in the oligomerisation and final quaternary
structure of the membrane protein configuration.
Membranes 2016, 6, 51
7 of 13
Membranes 2016, 6, 51
Membranes 2016, 6, 51
7 of 12
7 of 12
Figure 5. Conductance and capacitance of membranes containing different concentrations of
Figure 5. Conductance and capacitance of membranes containing different concentrations of cholesterol
cholesterol or ergosterol. Bilayer lipid membranes were formed using 10% tethered lipids on a gold
or ergosterol. Bilayer lipid membranes were formed using 10% tethered lipids on a gold electrode
electrode representing the monolayer of membrane and zwitterionic lipids dissolved in ethanol
representing
the
ofand
membrane
and
zwitterionic
lipids
dissolved
in ethanol containing
Figure
5.monolayer
Conductance
capacitance
of membranes
containing
different
containing
different
concentrations
of cholesterol
or ergosterol
was used
as theconcentrations
second layeroffor the
cholesterol or ergosterol.
Bilayer lipid
membranes
were
formed
using
10%
tetheredlayer
lipidsfor
on athe
gold
different
concentrations
of
cholesterol
or
ergosterol
was
used
as
the
second
membrane.
membrane. Membranes were then rapidly flushed with HEPES/KCl buffer (pH 6.5) in
order
to
electrode representing the monolayer of membrane and zwitterionic lipids dissolved in ethanol
Membranes
were
then
rapidly
flushed
with
HEPES/KCl
buffer
(pH
6.5)
in
order
to
remove
remove the ethanol by solvent exchange method and enhance the formation of the bilayer lipid the
containing different concentrations of cholesterol or ergosterol was used as the second layer for the
ethanol
bymembrane.
solvent
exchange
method
enhance
the
formation
of measured
the(pH
bilayer
lipid
membranes.
membranes.
TheMembranes
conductance
and
theand
capacitance
membranes
were
impedance
were
then
rapidly
flushedofwith
HEPES/KCl
buffer
6.5)using
in order
to
spectroscopy.
(A)
Represents
conductance
at different
cholesterol
or ergosterol
The conductance
and
the
capacitance
of membranes
were
measured
using
spectroscopy.
remove the
ethanol
by solventmembrane
exchange
method
and
enhance
the formation
ofimpedance
the bilayer
lipid
concentrations
and
(B)conductance
Capacitance
ofdifferent
membrane
with different
cholesterol
or ergosterol
membranes.
The conductance
and theatcapacitance
ofcholesterol
membranes
were
measured
using
impedance
(A) Represents
membrane
or ergosterol
concentrations
and (B)
spectroscopy.
Represents
membrane
conductance
at
different
cholesterol
or ergosterol
concentrations.
The(A)error
barsdifferent
represent
the
standard
error
of three
independent
impedance
Capacitance
of membrane
with
cholesterol
or ergosterol
concentrations.
The
error bars
concentrations
and (B) Capacitance of membrane with different cholesterol or ergosterol
spectroscopy
conductance
represent
the standard
error ofmeasurements.
three independent impedance spectroscopy conductance measurements.
concentrations. The error bars represent the standard error of three independent impedance
spectroscopy conductance measurements.
Figure 6. Conductance
of CLIC1
in tBLMscontaining
containing 2525
% ergosterol.
20 µ g of20
monomer
Figure
6. Conductance
of of
CLIC1
inintBLMs
mol
ergosterol.
of
CLIC1
monomer
Figure
6. Conductance
CLIC1
tBLMs containing
25mol
mol
%%
ergosterol.
20 µ gCLIC1
ofµg
CLIC1
monomer
(pre-incubated with 0.5 mM TCEP) and dimer (pre-incubated with 2 mM H2O2) in 100 µ L of
(pre-incubated
with
0.5
mM
TCEP)
and
dimer
(pre-incubated
with
2
mM
H
2
O
2
)
in
100
µ
L ofµL of
(pre-incubated with 0.5 mM TCEP) and dimer (pre-incubated with 2 mM H2 O2 ) in 100
HEPES/KCl buffer (pH 6.5) were incorporated into membranes containing zwitterionic lipids and 25
HEPES/KCl
buffer
(pH
6.5)
were
incorporated
into
membranes
containing
zwitterionic
lipids
and
HEPES/KCl
(pH The
6.5)error
were
into membranes
containingexperimental
zwitterionic25lipids
mol buffer
% ergosterol.
barsincorporated
represent the standard
error of three independent
mol
%
ergosterol.
The
error
bars
represent
the
standard
error
of
three
independent
experimental
and 25 molrepeats
% ergosterol.
The error bars represent the standard error of three independent
experimental
(n = 3).
repeats
(n
=
3).
repeats (n = 3).
The stability of channels in membranes is also an important factor in maintaining the rate of ion
transport
across
membranes
this further
that the
high in
conductance
of CLIC1
The
stability
of channels
in [19,33];
membranes
is alsosuggests
an important
factor
maintaining
the rateinof ion
membranes
ergosterol
is
due to
thefurther
ability
of
ergosterol
to increase
the
stability
of channels
The
stability
ofwith
channels
in[19,33];
membranes
is also
an important
factor
in maintaining
the in
rate of
transport
across
membranes
this
suggests
that
the high
conductance
of CLIC1
formed
by CLIC1
via increasing
the
of protein
aggregateto
structures.
This
also suggest
ion transport
across
membranes
[19,33];
this
further
suggests
that
the high
conductance
of CLIC1
membranes
with
ergosterol
is due
to rigidity
the
ability
of ergosterol
increase
the may
stability
of channels
that CLIC1 monomeric protein aggregates experience a higher rigidity in the presence of ergosterol
formed by with
CLIC1ergosterol
via increasing
theto
rigidity
of protein
aggregate to
structures.
also suggest
in membranes
is due
the ability
of ergosterol
increaseThis
the may
stability
of channels
than in cholesterol membranes and therefore CLIC1 showed higher conductance with faster initiation
that
CLIC1
monomeric
protein
aggregates
experience
a
higher
rigidity
in
the
presence
of
ergosterol
formed byrates
CLIC1
via increasing
the rigidity
of protein
aggregate
structures. This may also suggest
in membranes
with ergosterol
than membranes
containing
cholesterol.
than in monomeric
cholesterol membranes
and therefore
CLIC1 showed
higher
conductance
faster initiation
that CLIC1
protein aggregates
experience
a higher
rigidity
in thewith
presence
of ergosterol
rates in membranes with ergosterol than membranes containing cholesterol.
than in cholesterol membranes and therefore CLIC1 showed higher conductance with faster initiation
rates in membranes with ergosterol than membranes containing cholesterol.
Others have published cell based studies investigating the interaction between proteins and
sterols in membranes. These were performed via hemolysis assays where Cholesterol Dependant
Membranes 2016, 6, 51
8 of 13
6, 51forming proteins were incubated with free cholesterol prior to addition
8 of 12
CytolysinsMembranes
(CDC)2016,
pore
to the
cells. The result was an inhibition of the cytolytic activity of the toxins against erythrocytes [34,35].
Others have published cell based studies investigating the interaction between proteins and
Using a similar
set-up,
CLIC1
was via
pre-incubated
with
1% Cholesterol
free-cholesterol
or ergosterol
sterols experimental
in membranes. These
were
performed
hemolysis assays
where
Dependant
prior to addition
to(CDC)
tBLMs
50 mol
cholesterol
or cholesterol
ergosterol.
CLIC1
monomer
Cytolysins
porecontaining
forming proteins
were%
incubated
with free
prior
to addition
to the and the
cells. The result
was an
inhibition
of the pre-incubated
cytolytic activity of
the toxins
against erythrocytes
[34,35].
CDC Listeriolysin-O
(LLO)
that
were each
with
cholesterol,
showed low
conductance
Using a similar experimental set-up, CLIC1 was pre-incubated with 1% free-cholesterol or ergosterol
levels as expected (Figure 7). Interestingly, CLIC1 monomer that was pre-incubated with 1% ergosterol
prior to addition to tBLMs containing 50 mol % cholesterol or ergosterol. CLIC1 monomer and the
also showed
low
conductance
when
to CLIC1 monomer
not pre-incubated
with ergosterol or
CDC
Listeriolysin-O
(LLO)
thatcompared
were each pre-incubated
with cholesterol,
showed low conductance
as expected containing
(Figure 7). Interestingly,
CLIC1 monomer
that was (Figure
pre-incubated
with
1%
cholesterollevels
in membranes
50 mol % cholesterol
or ergosterol
7). The
pre-incubation
also showed
conductanceresulting
when compared
to CLIC1 monomer
not pre-incubated
withactivity in
of CLIC1 ergosterol
with cholesterol
orlow
ergosterol
in inhibition
of chloride
ion channel
ergosterol or cholesterol in membranes containing 50 mol % cholesterol or ergosterol (Figure 7). The
membranes, likely occurs by the sterol preventing either the initial binding and/or insertion of the
pre-incubation of CLIC1 with cholesterol or ergosterol resulting in inhibition of chloride ion channel
protein onto
theinmembrane,
and/or
located
withinand/or
the membrane.
This
activity
membranes, likely
occursits
by oligomerisation
the sterol preventing once
either the
initial binding
insertion
of the protein
the membrane,
and/or
its oligomerisation
once
located
within
the membrane.
This
strongly suggests
thatonto
CLIC1
membrane
interaction
proceeds
with
initial
binding
to the membrane
via
strongly
CLIC1 membrane
interaction
proceeds
with initial
binding to
membrane
cholesterol,
whichsuggests
acts asthat
a receptor
or docking
site for
the protein,
followed
bytheoligomerisation
and
via cholesterol, which acts as a receptor or docking site for the protein, followed by oligomerisation
full assembly into functional ion channels.
and full assembly into functional ion channels.
Figure 7. Conduction of pre-incubated CLIC1 monomer with sterols in tBLMs containing 50 mol %
Figure 7. Conduction
of pre-incubated
monomer
with
sterols
tBLMs containing
cholesterol or ergosterol.
CLIC1 (WT)CLIC1
monomeric
protein (20
µ g in
100 µ L in
of HEPES/KCl
buffer, pH 50 mol %
cholesterol6.5)
orwas
ergosterol.
CLIC1
(WT) monomeric
protein
µg in
100 µL
of HEPES/KCl
buffer,
incubated with
1% cholesterol
or 1% ergosterol
for ~1(20
h prior
addition
to tethered
bilayer
lipid
membranes
containing
25
mol
%
cholesterol
or
ergosterol.
Conductance
of
pre-incubated
CLIC1
pH 6.5) was incubated with 1% cholesterol or 1% ergosterol for ~1 h prior addition to tethered bilayer
monomer with sterols was then measured with impedance spectroscopy and compared to Controls:
lipid membranes containing 25 mol % cholesterol or ergosterol. Conductance of pre-incubated CLIC1
CLIC1 monomer not pre-incubated with sterols added into membranes containing 1% of cholesterol
monomer or
with
sterols
was then measured
spectroscopy
and
to Controls:
ergosterol,
listeriolysin-O
(20 µ g of LLOwith
in 100impedance
µ L of HEPES/KCl
buffer, pH 6.5)
wascompared
also incubated
CLIC1 monomer
pre-incubated
with
sterols
added
membranes
containing
1% of cholesterol
or
with 1%not
cholesterol
under the
same
conditions
asinto
CLIC1
monomer followed
by addition
to
with 50 (20
mol µg
% cholesterol.
The
error
represent the standard
three
repeats
ergosterol,membranes
listeriolysin-O
of LLO in
100
µLbars
of HEPES/KCl
buffer,error
pHof6.5)
was
alsoofincubated
experimental measures (n = 3).
with 1% cholesterol
under the same conditions as CLIC1 monomer followed by addition to membranes
with 50 mol
%
cholesterol.
error
barscombined
representwith
the standard
of three repeats
of experimental
In conclusion, ourThe
model
tBLMs
impedanceerror
spectroscopy
have allowed
us to
measures
(nthe
= 3).
probe
role of both sterol and redox environment as regulators of CLIC1 spontaneous membrane
insertion and ion channel activity. The high sterol-dependent conductance of CLIC1 in tBLMs
strongly points to cholesterol as a receptor or initial membrane binding site for CLIC1, that may also
In conclusion,
our model tBLMs combined with impedance spectroscopy have allowed us to
aid in the protein’s unfolding, oligomerisation and formation of functional ion channels in
probe the role
of both
sterol and
redox
environment
as regulators
spontaneous
membrane
membranes.
Similarly,
the effects
of redox
on CLIC1 transition
betweenof
itsCLIC1
soluble to
membrane form
suggests
the proteinactivity.
adopts a The
structure
membranes that resembles
the reduced
protein state.
The strongly
insertion and
ion channel
highinsterol-dependent
conductance
of CLIC1
in tBLMs
redox reactive
Cys24or
located
at membrane
the start of the
putativesite
TMD
CLIC1 appears
not to
be aid in the
points to cholesterol
asresidue
a receptor
initial
binding
forinCLIC1,
that may
also
essential for its ion channel activity, but does influence the channel’s conductance and potentially its
protein’s unfolding,
oligomerisation and formation of functional ion channels in membranes. Similarly,
gating properties.
the effects of redox on CLIC1 transition between its soluble to membrane form suggests the protein
first adopts a structure that likely resembles a version of its oxidized protein state, exposing a large
hydrophobic surface that facilitates its membrane interaction. The redox reactive residue Cys24 located
at the start of the putative TMD in CLIC1 appears not to be essential for its ion channel activity, but
does influence the channel’s conductance and potentially its gating properties.
Membranes 2016, 6, 51
9 of 13
3. Materials and Methods
3.1. Preparation of His-Tagged Recombinant CLIC1 WT, CLIC1-C24A and C59A Proteins
The following annotations will be used to refer to each mutant, with each containing a single
amino acid substitution (CLIC1-C24A, CLIC1-C24S and CLIC1-C59A).
Protein purification was performed as previously described [23,36]. Briefly, E. coli bacterial
cells, BL21 (DE3) containing the His-tag pET28a vector system were grown in 2xYT media
containing 50 µg/mL Kanamycin antibiotic (Sigma Aldrich, Carlsbad, Carlifonia, USA). Cells then
were induced with 1 mM IPTG (Sigma Aldrich) and left to grow further at 20 ◦ C for about 16 h with
shaking at 200 rpm. Then the cells were harvested and lysed using Sonication on ice for 6 times
with 10 s pulses at 80% output. Cell lysate was then ran through His-tag Ni-NTA high affinity
chromatography column in the presence of 0.5 mM TCEP and the His-tagged protein was cleaved
off the resin by overnight incubation with 30 NIH units of bovine plasma thrombin (Sigma Aldrich)
per litre of cell culture. The eluted recombinant CLIC1 protein from the Ni-NTA resin was again
incubated with 0.5 mM TCEP and ran through size exclusion chromatography using Superdex-75 prep
grade high performance chromatography column (GE healthcare, Piscataway, NJ, USA) in order to
obtain 99% purity of monomeric protein. The chromatography column was initially equilibrated in
column sizing buffer with reducing agent (100 mM KCl, 0.5 mM TCEP, 1 mM NaN3 , and 20 mM
HEPES pH 7.5) and the purification was performed at 4 ◦ C. The purified proteins were quantified
using the BCA protein assay (Thermo Scientific, Sydney, Australia) and the purity and oligomeric state
of proteins was further investigated by running SDS-PAGE.
3.2. Recombinant CLIC1 Dimeric Protein
Dimeric CLIC1 was provided by Dr Louise Brown from Macquarie University, Australia. It was
prepared as previously described [23,36].
3.3. Preparation of Recombinant EXC-4 and CLIC1-C24S by GST Gene Fusion System
Glutathione S-Transferase (GST) Gene fusion system (AMRAD-Pharmacia, Melbourne, Australia)
was used for the expression and purification of fusion proteins in E. coli bacteria. E. coli bacteria
strain, BL21 (DE3) containing pGEX-4T-1 vector (Novagen, ON, Canada) was left to grow in 2xYT
media containing 100 µg/mL Carbenicillin on a shaker at 180 rpm, at 37 ◦ C for 2.5 h or until an OD
of 600 was achieved. Cells then were induced with 1 mM IPTG and returned to incubation for
another 4.5 h at 37 ◦ C with 180 rpm shaking. Cells were harvested and lysed by sonication as described
in the previous section. Then purification of the cells lysate was achieved by running it through the
glutathione-sepharose 4B resin (Amersham Biosciences, Sydney, Australia) in the presence of 0.5 mM
TCEP where the GST-tagged proteins were cleaved off from the resin beads by incubation with 30 NIH
units per 1 L of cells culture of bovine plasma thrombin (Sigma Aldrich) as described in the section
above. Then the recombinant proteins were further purified by size exclusion chromatography in the
presence of 0.5 mM TCEP as the reducing agent in order to obtain reduced monomeric proteins with
size exclusion chromatography profile containing one peak. The protein concentration and purity was
determined as was described in the section above.
3.4. Formation of Tethered Bilayer Lipid Membranes (tBLM)
Artificial membranes of ~4 nm thickness capable of incorporating proteins of up to 40 kDa
were formed using methods reported in [3,22,37,38]. The monolayer tethering coating was
prepared by coating freshly deposited, 100 nm pattern gold electrodes, on 25 mm × 75 mm × 1 mm
polycarbonate slides, with two benzyl disulphide families, one being a spacer molecule containing
a four oxygen-ethylene glycol spacer, terminated with an OH group (90%), and the second
being a tethering group comprising an eleven oxygen–ethylene glycol linker group with a single C20
hydrophobic phytanyl chain (10%) as the hydrophobic tether. The gold electrode was then assembled
Membranes 2016, 6, 51
10 of 13
onto a 6 well polyethylene cartridge, possessing a 2 mm2 active area and a flow cell chamber
of 100 µm in height. 8 µL of 3 mM mobile lipid phase (MLP) containing 70% zwitterionic C20
diphytanyl-ether-glycero-phosphatidylcholine: 30% C20 diphytanyl-diglyceride ether lipids dissolved
in 99% (v/v) ethanol was added into each of the 6 wells and allowed to incubate at room temperature
for ~2 min. For making the second layer of membranes, cholesterol or ergosterol (both from
Sigma Aldrich) that were dissolved in 95% (v/v) ethanol were mixed with the MLPs to make
different concentrations (0 mol %–50 mol %). This process was then followed by rinsing the lipids
with 3 × 100 µL of HEPES/KCl buffer: 0.1M KCl, 0.1 mM HEPES and 0.01 mM CaCl2 , pH 6.5.
Rinsing away the ethanol solvent and replacement with an aqueous buffer drives the rearrangement
of the dissolved lipids to form a lipid bilayer, which is detected by the changes in impedance
spectroscopy measurements.
3.5. Formation of tBLM Using Yeast and Bacterial Lipids
Lipid extracts of yeast (Saccharomyces cerevisiae from BioAustralis Pty Ltd., Sydney, Australia) and
E. coli bacterial cells (provided by Dr Charles Cranfield from the Victor Chang Institute for Medical
Research, Sydney, Australia), were dissolved separately in 95% (v/v) ethanol with the aid of heating in
a 50 ◦ C water bath followed by continuous vortexing for at least 15 min. 3 mM of yeast or E. coli lipid
solution was added to the first layer of membrane in the coated gold electrode in place of the mobile
lipid solution and tBLM formation, as described in the section above.
3.6. Incorporation of CLIC1 WT, Mutants and EXC-4 into tBLMs Containing Cholesterol
Recombinant CLIC1 (WT) monomeric and dimeric proteins; CLIC1-C24A; CLIC1-C59A and
EXC-4 (WT) were diluted to a concentration of 20 µg/100 µL (7.4 µM) in HEPES/KCl buffer
(0.1 M KCl, 0.1 mM HEPES and 0.01 mM CaCl2 of pH 6.5). Each protein was incubated with 0.5 mM
TCEP for ~1 h. They were then applied to pre-prepared tethered bilayer lipid membranes with
or without cholesterol or ergosterol that were equilibrated for ~1 h with 100 µL of HEPES/KCl
buffer containing 0.5 mM TCEP when measuring the conductance of monomeric reduced CLIC1
or membranes were equilibrated with 2 mM H2 O2 when conducting experiments with oxidized or
dimeric CLIC1.
3.7. Pre-Incubation of CLIC1 with Cholesterol or Ergosterol
CLIC1 (WT) monomeric protein (20 µg in 100 µL of HEPES/KCl buffer) was incubated for
approximately 1 h with 2 µL of 13.3 mg of cholesterol or ergosterol dissolved in 1 mL of 95% (v/v)
ethanol prior addition to tBLMs with or without sterols.
3.8. Pre-Incubation of Listeriolysin-O with Cholesterol
Listeriolysin-O (Sigma Aldrich), 2 µM in 100 µL as a final volume of HEPES/KCl buffer was
incubated with 2 µL of 13.3 mg of cholesterol dissolved in 95% (v/v) ethanol before application to
membranes with or without cholesterol.
4. Conclusions
This study has employed an artificial tethered lipid bilayer system to demonstrate the presence
of sterols (cholesterol or ergosterol) within the bilayer is critical for the ion channel conductance
activity of the protein CLIC1. Furthermore, the oxidation state of the protein also serves to regulate
its channel conductance activity. While the redox active residue Cysteine 24 in CLIC1 although
essential for its oxidoreductase enzymatic activity, does not appear to be essential for its ion channel
conductance activity.
Acknowledgments: Many thanks to Louise Brown from Macquarie University for providing purified dimeric
CLIC1 (WT) protein.
Membranes 2016, 6, 51
11 of 13
Author Contributions: Experimental design: Bruce A. Cornell, Heba Al Khamici, Stella M. Valenzuela,
Khondker R. Hossain; Protein expression and purification: Heba Al Khamici; Performed experiments:
Heba Al Khamici; Analyzed the data: Heba Al Khamici and Bruce A. Cornell.
Contributed
reagents/materials/analysis tools: Bruce A. Cornell and Stella M. Valenzuela.
Conflicts of Interest: Bruce A. Cornell is employed by Surgical Diagnostics Pty Ltd.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Valenzuela, S.M.; Martin, D.K.; Por, S.B.; Robbins, J.M.; Warton, K.; Bootcov, M.R.; Schofield, P.R.;
Campbell, T.J.; Breit, S.N. Molecular cloning and expression of a chloride ion channel of cell nuclei.
J. Biol. Chem. 1997, 272, 12575–12582. [CrossRef] [PubMed]
Littler, D.R.; Harrop, S.J.; Goodchild, S.C.; Phang, J.M.; Mynott, A.V.; Jiang, L.; Valenzuela, S.M.; Mazzanti, M.;
Brown, L.J.; Breit, S.N.; et al. The enigma of the CLIC proteins: Ion channels, redox proteins, enzymes,
scaffolding proteins? FEBS Lett. 2010, 584, 2093–2101. [CrossRef] [PubMed]
Valenzuela, S.M.; Alkhamici, H.; Brown, L.J.; Almond, O.C.; Goodchild, S.C.; Carne, S.; Curmi, P.M.;
Holt, S.A.; Cornell, B.A. Regulation of the membrane insertion and conductance activity of the metamorphic
chloride intracellular channel protein CLIC1 by cholesterol. PLoS ONE 2013, 8. [CrossRef] [PubMed]
Warton, K.; Tonini, R.; Fairlie, W.D.; Matthews, J.M.; Valenzuela, S.M.; Qiu, M.R.; Wu, W.M.; Pankhurst, S.;
Bauskin, A.R.; Harrop, S.J. Recombinant CLIC1 (NCC27) assembles in lipid bilayers via a ph-dependent
two-state process to form chloride ion channels with identical characteristics to those observed in chinese
hamster ovary cells expressing CLIC1. J. Biol. Chem. 2002, 277, 26003–26011. [CrossRef] [PubMed]
Littler, D.R.; Harrop, S.J.; Fairlie, W.D.; Brown, L.J.; Pankhurst, G.J.; Pankhurst, S.; DeMaere, M.Z.;
Campbell, T.J.; Bauskin, A.R.; Tonini, R. The intracellular chloride ion channel protein CLIC1 undergoes a
redox-controlled structural transition. J. Biol. Chem. 2004, 279, 9298–9305. [CrossRef] [PubMed]
Singh, H.; Cousin, M.; Ashley, R. Functional reconstitution of mammalian ‘chloride intracellular channels’
CLIC1, CLIC4 and CLIC5 reveals differential regulation by cytoskeletal actin. FEBS J. 2007, 274, 6306–6316.
[CrossRef] [PubMed]
Tulk, B.M.; Kapadia, S.; Edwards, J.C. CLIC1 inserts from the aqueous phase into phospholipid membranes,
where it functions as an anion channel. Am. J. Physiol. Cell Physiol. 2002, 282, C1103–C1112. [CrossRef]
[PubMed]
Littler, D.R.; Assaad, N.N.; Harrop, S.J.; Brown, L.J.; Pankhurst, G.J.; Luciani, P.; Aguilar, M.I.; Mazzanti, M.;
Berryman, M.A.; Breit, S.N. Crystal structure of the soluble form of the redox-regulated chloride ion channel
protein CLIC4. FEBS J. 2005, 272, 4996–5007. [CrossRef] [PubMed]
Singh, H.; Ashley, R.H. CLIC4 (p64H1) and its putative transmembrane domain form poorly selective,
redox-regulated ion channels. Mol. Membr. Biol. 2007, 24, 41–52. [CrossRef] [PubMed]
Singh, H.; Ashley, R.H. Redox regulation of CLIC1 by cysteine residues associated with the putative channel
pore. Biophys. J. 2006, 90, 1628–1638. [CrossRef] [PubMed]
Duncan, R.R.; Westwood, P.K.; Boyd, A.; Ashley, R.H. Rat brain p64H1, expression of a new member of
the p64 chloride channel protein family in endoplasmic reticulum. J. Biol. Chem. 1997, 272, 23880–23886.
[CrossRef] [PubMed]
Proutski, I.; Karoulias, N.; Ashley, R. Overexpressed chloride intracellular channel protein CLIC4 (p64H1) is
an essential component of novel plasma membrane anion channels. Biochem. Biophys. Res. Commun. 2002,
297, 317–322. [CrossRef]
Luetterforst, R.; Stang, E.; Zorzi, N.; Carozzi, A.; Way, M.; Parton, R.G. Molecular characterization of caveolin
association with the golgi complex: Identification of a Cis-Golgi targeting domain in the caveolin molecule.
J. Cell Biol. 1999, 145, 1443–1459. [CrossRef] [PubMed]
Palmer, M. Cholesterol and the activity of bacterial toxins. FEMS Microbiol. Lett. 2004, 238, 281–289.
[CrossRef] [PubMed]
Suginita, W.; Karoulias, N.; Aitken, A.; Ashley, R.H. Chloride intracellular channel protein CLIC4 (p64H1)
binds directly to brain dynamin I in a complex containing actin, tubulin and 14-3-3 isoforms. Biochem. J.
2001, 359, 55–64. [CrossRef]
Bretscher, M.S.; Munro, S. Cholesterol and the golgi apparatus. Science 1993, 261, 1280–1281. [CrossRef]
[PubMed]
Membranes 2016, 6, 51
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
12 of 13
Kucerka, N.; Nieh, M.-P.; Pencer, J.; Sachs, J.N.; Katsaras, J. What determines the thickness of a biological
membrane. Gen. Physiol. Biophys. 2009, 28, 117–125. [CrossRef] [PubMed]
Hossain, K.R.; Al Khamici, H.; Holt, S.A.; Valenzuela, S.M. Cholesterol promotes interaction of the protein
CLIC1 with phospholipid monolayers at the air–water interface. Membranes 2016, 6. [CrossRef] [PubMed]
Coutinho, A.; Silva, L.; Fedorov, A.; Prieto, M. Cholesterol and ergosterol influence nystatin surface
aggregation: Relation to pore formation. Biophys. J. 2004, 87, 3264–3276. [CrossRef] [PubMed]
Andersson, J.; Köper, I. Tethered and polymer supported bilayer lipid membranes: Structure and function.
Membranes 2016, 6. [CrossRef] [PubMed]
Jackman, J.A.; Knoll, W.; Cho, N.-J. Biotechnology applications of tethered lipid bilayer membranes. Materials
2012, 5, 2637–2657. [CrossRef]
Cornell, B.A.; Krishna, G.; Osman, P.D.; Pace, R.; Wieczorek, L. Tethered-bilayer lipid membranes as a
support for membrane-active peptides. Biochem. Soc. Trans. 2001, 29, 613–617. [CrossRef] [PubMed]
Goodchild, S.C.; Howell, M.W.; Cordina, N.M.; Littler, D.R.; Breit, S.N.; Curmi, P.M.; Brown, L.J. Oxidation
promotes insertion of the CLIC1 chloride intracellular channel into the membrane. Eur. Biophys. J. 2009, 39,
129–138. [CrossRef] [PubMed]
Poveda, J.; Fernández, A.; Encinar, J.; Gonzalez-Ros, J. Protein-promoted membrane domains. Biochim. Biophys.
Acta (BBA) Biomembr. 2008, 1778, 1583–1590. [CrossRef] [PubMed]
Epand, R.F.; Thomas, A.; Brasseur, R.; Vishwanathan, S.A.; Hunter, E.; Epand, R.M. Juxtamembrane protein
segments that contribute to recruitment of cholesterol into domains. Biochemistry 2006, 45, 6105–6114.
[CrossRef] [PubMed]
Vishwanathan, S.A.; Thomas, A.; Brasseur, R.; Epand, R.F.; Hunter, E.; Epand, R.M. Hydrophobic
substitutions in the first residue of the CRAC segment of the gp41 protein of HIV. Biochemistry 2008, 47,
124–130. [CrossRef] [PubMed]
Harrop, S.J.; DeMaere, M.Z.; Fairlie, W.D.; Reztsova, T.; Valenzuela, S.M.; Mazzanti, M.; Tonini, R.; Qiu, M.R.;
Jankova, L.; Warton, K.; et al. Crystal structure of a soluble form of the intracellular chloride ion channel
CLIC1 (NCC27) at 1.4-Å resolution. J. Biol. Chem. 2001, 276, 44993–45000. [CrossRef] [PubMed]
Al Khamici, H.; Brown, L.J.; Hossain, K.R.; Hudson, A.L.; Sinclair-Burton, A.A.; Ng, J.P.; Daniel, E.L.;
Hare, J.E.; Cornell, B.A.; Curmi, P.M.; et al. Members of the chloride intracellular ion channel protein family
demonstrate glutaredoxin-like enzymatic activity. PLoS ONE 2015, 10. [CrossRef] [PubMed]
Verma, S.P.; Wallach, D.F. Multiple thermotropic state transitions in erythrocyte membranes a laser-raman
study of the CH-stretching and acoustical regions. Biochim. Biophys. Acta (BBA) Biomembr. 1976, 436, 307–318.
[CrossRef]
Alberts, B.; Bray, D.; Hopkin, K.; Johnson, A.D.; Johnson, A.; Lewis, J.; Raff, M.; Roberts, K.; Walter, P.
Essential Cell Biology, 3rd ed.; Garland Science: New York, NY, USA; London, UK, 2010.
De Almeida, R.F.; Fedorov, A.; Prieto, M. Sphingomyelin/phosphatidylcholine/cholesterol phase diagram:
Boundaries and composition of lipid rafts. Biophys. J. 2003, 85, 2406–2416. [CrossRef]
Xu, X.; Bittman, R.; Duportail, G.; Heissler, D.; Vilcheze, C.; London, E. Effect of the structure of natural
sterols and sphingolipids on the formation of ordered sphingolipid/sterol domains (rafts) comparison of
cholesterol to plant, fungal, and disease-associated sterols and comparison of sphingomyelin, cerebrosides,
and ceramide. J. Biol. Chem. 2001, 276, 33540–33546. [PubMed]
Baginski, M.; Resat, H.; Borowski, E. Comparative molecular dynamics simulations of amphotericin
B-cholesterol/ergosterol membrane channels. Biochim. Biophys. Acta (BBA) Biomembr. 2002, 1567, 63–78.
[CrossRef]
Giddings, K.S.; Johnson, A.E.; Tweten, R.K. Redefining cholesterol’s role in the mechanism of the
cholesterol-dependent cytolysins. Proc. Natl. Acad. Sci. USA 2003, 100, 11315–11320. [CrossRef] [PubMed]
Tweten, R.K. Cholesterol-dependent cytolysins, a family of versatile pore-forming toxins. Infect. Immun.
2005, 73, 6199–6209. [CrossRef] [PubMed]
Goodchild, S.C.; Angstmann, C.N.; Breit, S.N.; Curmi, P.M.; Brown, L.J. Transmembrane extension and
oligomerization of the CLIC1 chloride intracellular channel protein upon membrane interaction. Biochemistry
2011, 50, 10887–10897. [CrossRef] [PubMed]
Membranes 2016, 6, 51
37.
38.
13 of 13
Cranfield, C.G.; Cornell, B.A.; Grage, S.L.; Duckworth, P.; Carne, S.; Ulrich, A.S.; Martinac, B. Transient
potential gradients and impedance measures of tethered bilayer lipid membranes: Pore-forming peptide
insertion and the effect of electroporation. Biophys. J. 2014, 106, 182–189. [CrossRef] [PubMed]
Cranfield, C.G.; Berry, T.; Holt, S.A.; Hossain, K.R.; Le Brun, A.P.; Carne, S.; Al Khamici, H.; Coster, H.;
Valenzuela, S.M.; Cornell, B. Evidence of the key role of H3 O+ in phospholipid membrane morphology.
Langmuir 2016, 32, 10725–10734. [CrossRef] [PubMed]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).