Content uploaded by Anaclet Ngezahayo
Author content
All content in this area was uploaded by Anaclet Ngezahayo on Nov 08, 2018
Content may be subject to copyright.
Cataract-associated D3Y mutation of human connexin46
(hCx46) increases the dye coupling of gap junction channels
and suppresses the voltage sensitivity of hemichannels
Barbara Schlingmann &Patrik Schadzek &Stefan Busko &
Alexander Heisterkamp &Anaclet Ngezahayo
Received: 22 May 2012 /Accepted: 8 July 2012 / Published online: 28 July 2012
#Springer Science+Business Media, LLC 2012
Abstract Connexin46 (Cx46), together with Cx50, forms
gap junction channels between lens fibers and participates in
the lens pump-leak system, which is essential for the ho-
meostasis of this avascular organ. Mutations in Cx50 and
Cx46 correlate with cataracts, but the functional relationship
between the mutations and cataract formation is not always
clear. Recently, it was found that a mutation at the third
position of hCx46 that substituted an aspartic acid residue
with a tyrosine residue (hCx46D3Y) caused an autosomal
dominant zonular pulverulent cataract. We expressed EGFP-
labeled hCx46wt and hCx46D3Y in HeLa cells and found
that the mutation did not affect the formation of gap junction
plaques. Dye transfer experiments using Lucifer Yellow
(LY) and ethidium bromide (EthBr) showed an increased
degree of dye coupling between the cell pairs expressing
hCx46D3Y in comparison to the cell pairs expressing
hCx46wt. In Xenopus oocytes, two-electrode voltage-
clamp experiments revealed that hCx46wt formed voltage-
sensitive hemichannels. This was not observed in the
oocytes expressing hCx46D3Y. The replacement of the
aspartic acid residue at the third position by another nega-
tively charged residue, glutamic acid, to generate the mutant
hCx46D3E, restored the voltage sensitivity of the resultant
hemichannels. Moreover, HeLa cell pairs expressing
hCx46D3E and hCx46wt showed a similar degree of dye
coupling. These results indicate that the negatively charged
aspartic acid residue at the third position of the N-terminus
of hCx46 could be involved in the determination of the
degree of metabolite cell-to-cell coupling and is essential
for the voltage sensitivity of the hCx46 hemichannels.
Keywords hCx46 .N-terminus .Voltage sensitivity .Dye
transfer .Hemichannels .Cataract
Introduction
Connexins are transmembrane proteins that are encoded in
humans by a gene family composed of 21 members (Sohl
and Willecke 2004). In the membrane, all connexins adopt
the same topology, with four transmembrane domains (TM)
and two extracellular loops. Both the N- and C-terminus are
localized in the cytoplasmic space, where a loop between
TM2 and TM3 is also found (Falk et al. 1994). During
trafficking to the cell membrane, connexins hexamerize
and form the so-called connexons, or hemichannels. In the
plasma membrane, the connexons of adjacent cells interact
through the extracellular loops of the connexins and form
gap junction channels, which are assembled in gap junction
plaques that are found in the contact regions of adjacent
cells. Gap junction channels allow the exchange of ions and
metabolites between neighboring cells and are therefore
essential for the formation of functional physiological units
in tissue.
In the lens, three connexins, Cx43, Cx46 and Cx50,
participate in the development and maturation of lens fibers
(Gerido and White 2004; Goodenough 1992; White and
Bruzzone 2000). Cx46 is mainly expressed in lens fibers,
B. Schlingmann :P. Schadzek :S. Busko :A. Ngezahayo (*)
Institute of Biophysics, Leibniz University Hannover,
Herrenhäuserstr. 2,
30419 Hannover, Germany
e-mail: ngezahayo@biophysik.uni-hannover.de
B. Schlingmann :A. Ngezahayo
Center for Systems Neuroscience Hannover,
University of Veterinary Medicine Hannover Foundation,
Buenteweg 2,
30559 Hannover, Germany
A. Heisterkamp
Institute of Applied Optics, Friedrich-Schiller-University-Jena,
Froebelstieg 1,
07743 Jena, Germany
J Bioenerg Biomembr (2012) 44:607–614
DOI 10.1007/s10863-012-9461-0
where it forms, together with Cx50, gap junction channels,
which are part of the pump-leak system that maintains the
homeostasis of the lens (Donaldson et al. 2001; Mathias et al.
1997,2010). The importance of gap junctions for lens
physiology is attested by the malformations and cataract
development that are correlated with mutations in both Cx50
and Cx46 (Berthoud and Beyer 2009); one such correlation
was recently shown for an autosomal dominant zonular
pulverulent cataract. This form of cataract was found to be
related to a mutation in the N-terminus (NT) of hCx46, in
which the aspartic acid residue at the third position was
replaced by a tyrosine residue (hCx46D3Y) (Addison et al.
2006). To understand the relationship between the D3Y mu-
tation and the development of the autosomal dominant zonular
pulverulent cataract, the functional changes in the gap junc-
tion channels caused by the D3Y mutation must be studied.
Mutation of connexins can affect the formation of gap
junctions at different levels. Some mutations affect connexin
synthesis, connexin trafficking to the plasma membrane or
docking of connexons in adjacent cells (Arora et al. 2006,
2008; Berthoud et al. 2003; Lichtenstein et al. 2009; Minogue
et al. 2005; Thomas et al. 2008). Other mutations affect regu-
latory mechanisms, such as voltage-dependent gating
(Minogue et al. 2009; Pal et al. 2000). For the voltage sensi-
tivity, studies of Cx26 and Cx32 revealed that the six N-termini
of a hemichannel lined the pore and contained charged amino
acid residues that formed a part of the voltage sensor and are
involved in ion permeation (Gonzalez et al. 2007; Maeda et al.
2009;Ohetal.1999,2000,2004; Purnick et al. 2000a,b;
Verselis et al. 1994). From analysis of the crystal structure,
NMR and circular dichroism spectroscopy of Cx26 and Cx37,
a structural model of the NT was proposed: residues 1–10 of
Cx26 adopt a helical conformation, line the pore and form a
funnel structure. The funnel structure is stabilized by a circular
network of hydrogen bonds between the aspartic acid residue at
the second position (D2) in one Cx-subunit and the threonine
residue at the fifth position (T5) in the adjacent Cx-subunit
(Maeda et al. 2009; Maeda and Tsukihara 2011). For Cx37, the
residues between D3 and H16 may form a helical structure
(Kyle et al. 2009). Further, a homology model predicted hy-
drogen bonds between the D3 of one subunit and both the
phenylalanine at position six (F6) and the leucine at position
seven (L7) of the adjacent subunit (Beyer et al. 2012). In the
model for Cx26, the six N-termini form helices that move in an
electrical field because of their charged residues. The move-
ment of the N-termini can open or close the channel, depending
on the direction of movement (Maeda and Tsukihara 2011).
For Cx46, it was shown that a replacement of the aspartic
acid residue at the third position by an asparagine residue
(D3N) changed the polarity of voltage gating in Cx46.
Moreover, additional mutations in the NT of Cx46, such as
D3A, D3C and D3G, altered the formation of functional
hemichannels when expressed in oocytes (Srinivas et al.
2005). These findings provide evidence that D3 could be a
part of the voltage sensor of Cx46.
In this report, HeLa cells and Xenopus oocytes were used
as expression systems to compare the functional expression of
the gap junction channels and the biophysical properties of
hCx46wt and hCx46D3Y hemichannels. In HeLa cells, both
hCx46wt and the mutant were equally expressed. LY and
EthBr transfer experiments revealed that the degree of dye
coupling was larger for cells expressing hCx46D3Y than for
those expressing hCx46wt. In Xenopus oocytes, we found that
the mutant hCx46D3Y did not form voltage-dependent
hemichannels, indicating that the mutation induced a loss of
the voltage-dependent regulation of the channels.
Materials and methods
Molecular biology
The pSP64TII vector that contained the hCx46 gene, which
was kindly provided by Dr. Viviana Berthoud, was used as
template for further cloning steps. For in vitro transcription,
the hCx46 gene was inserted into the pGEMHE vector using
EcoRI restriction sites (Walter et al. 2008). For the fluores-
cence imaging and dye transfer experiments, the hCx46
gene was PCR amplified and cloned into the pEGFP N1
vector (Clontech Laboratories, Mountain View, CA). The
first two primers, which are described in Table 1, were used
to generate EcoRI restriction sites, mutate the stop codon
and fuse the gene in frame before EGFP with the addition of
a 19-amino acid polylinker. The D3Y and D3E mutations
were introduced by site-directed mutagenesis using the last
six primers, which are described in Table 1.Eschericha coli
XL10-Gold (Stratagene, Waldbronn Germany) was used to
host the plasmids containing the different genes. All of the
cloned vectors were verified by sequencing (Seqlab,
Göttingen, Germany). The plasmid DNA for the transfection
Table 1 Primers used for the construction of expression vectors and
for site-directed mutagenesis
Primer 5′-3′sequence
pEGFP hCx46 fw tccgaattcactagtgagccgccatgggcgactggag
pEGFP hCx46 rev ctagagaattcgatttcctccgatggccaagtcctccgg
Mut D3Y pGEMHE fw ctagtgatttgcaatgggctattggagctttctgggaagac
Mut D3Y pEGFP fw tccgaattcactagtgagccgccatgggctattggag
Mut D3Y pGEMHE/
pEGFP rev
gtcttcccagaaagctccaatagcccattgcaaatcactag
Mut D3E pGEMHE fw tagtgatttgcaatgggcgaatggagctttctgg
Mut D3E pGEMHE rev ccagaaagctccattcgcccattgcaaatcacta
Mut D3E pEGFP fw gagccgccatgggcgaatggagctttc
Mut D3E pEGFP rev gaaagctccattcgcccatggcggctc
608 J Bioenerg Biomembr (2012) 44:607–614
experiments was purified using the QIAprep Spin Miniprep
Kit (QIAGEN, Hilden, Germany). For in vitro transcription,
the coding region, including the T7 promotor, was amplified
and purified using a PCR purification kit (QIAGEN, Hilden,
Germany). The cRNA of the different hCx46 variants was
prepared using a synthesis kit containing T7 RNA polymerase
and CAP analogue, which was purchased from Ambion
(Austin, USA). The transcript concentration was estimated
spectrophotometrically and analyzed on agarose gels.
Expression in HeLa cells
HeLa cells were cultivated using Dulbecco´s Modified
Eagle´s Medium (DMEM/Ham’s F12, 1:1) (Biochrom,
Berlin, Germany) supplemented with 10 % fetal calf serum
(FCS) (PAA, Pasching, Austria), penicillin and streptomycin
(100 U/ml and 10 mg/ml, respectively). The cultures were
maintained at 37 °C in a humidified atmosphere containing
5%CO
2
.
For the transfection experiments, cover slips (Ø 10 mm)
coated with rat collagen I (150 μg/ml) (Cultrex,
Gaithersburg USA) were placed into a 24-multiwell plate.
7×10
4
cells/well were seeded. OptiMEM® (Gibco,
Invitrogen) containing 0.5 μg plasmid DNA and 1.5 μl
FuGENE HD Transfection reagent (Roche, Penzberg,
Germany) was used. The examination of gap junction
plaque formation and the dye transfer experiments were
performed 24–48 h after transfection. For each hCx46 variant,
cells of at least three different passages were transfected.
For all of the transfected hCx46 variants, the transfection
efficiency was between 50 and 70 %.
Formation of gap junction plaques
The fluorescence images were acquired using an inverted
Nikon Eclipse TE2000-E confocal laser scanning micro-
scope with a 60× water immersion objective (Nikon,
Düsseldorf, Germany). The EZ-C1 3.80 software (Nikon,
Düsseldorf, Germany) was used to record the images.
During the different experiments, the settings for gain,
brightness, contrast, and pixel dwell time remained constant.
The images were taken at a resolution of 2048 × 2048 pixels.
To improve cell selection, the nuclei of the transfected cells
were stained with Hoechst 33342 (1 μg/ml) (Sigma Aldrich)
and the cell membranes were stained with Wheat Germ
Agglutinin Alexa555 (5 μg/ml)(MolecularProbes,
Eugene, OR, USA). The cells were fixed with 3.7 % form-
aldehyde. To quantify the plaque formation of the hCx46
variants, the expressing cell pairs were counted in a double-
blind randomized fashion. The plaque formation was quan-
tified using ImageJ (http://rsbweb.nih.gov/ij/docs/menus/
analyze.html#plot). For each transfection, five cover slips
were evaluated. Four images from different sections of a
cover slip were taken. For each construct, the average ratio
of the number of cell pairs that formed gap junction plaques
to the total number of cell pairs expressing the
corresponding construct is given. The error bars are the
SEM, and the statistical significance of changes was evalu-
ated by Student’st-test (** for p≤0.01 and * for p≤0.05).
Functional formation of gap junction channels of the hCx46
variants
Formation of functional gap junction channels was evaluated
using dye transfer experiments. Cover slips with transfected
cells were transferred to a perfusion chamber containing
500 μl of a bath solution composed of (in mM) 121 NaCl,
5.4 KCl, 6 NaHCO
3
, 5.5 glucose, 0.8 MgCl
2
, 1.6 CaCl
2
,
and 25 HEPES at pH 7.4 and mounted on a Zeiss inverted
fluorescence microscope (Oberkochen, Germany). For the
dye transfer experiments, a whole-cell patch-clamp config-
uration was established on a cell of a pair expressing the
EGFP-labeled hCx46 variant using a EPC7 patch-clamp
amplifier (List Medical, Darmstadt, Germany). Lucifer
Yellow lithium salt (LY) (1 mg/ml) (Biotium, Hayward
CA) or ethidium bromide (EthBr) (1 mg/ml) (Sigma
Aldrich) was diluted in a pipette medium containing (in
mM) 135 K-Gluconate, 5 KCl, 10 HEPES, 1 MgCl
2
,
1CaCl
2
, 2 glucose, 5 Na
2
ATP, 5 EGTA, 0.1 cAMP,
0.1 cGMP at pH 7.4. A Polychrome II monochromator
(T.I.L.L. Photonics GmbH, Planegg, Germany) equipped
with a 75 W XBO xenon lamp was used to excite the
fluorescent molecules (EGFP-labeled hCx46 variant at
488 nm, LY at 410 nm and EthBr at 350 nm). The images
were taken with a digital CCD camera (C4742-95,
Hamamatsu Photonics K.K.; Japan) using Aquacosmos
software (Hamamatsu Photonics K.K.; Japan). For each
variant, the degree of dye coupling was estimated as the
ratio of the number of coupled pairs to the total number of
tested pairs expressing the particular variant. The results are
given as the average values. The error bars represent the
SEM. The significance of the difference was evaluated by
Student’st-test (** for p≤0.01 and * for p≤0.05).
Expression of hCx46 in Xenopus oocytes
The expression of the hCx46 variants in Xenopus oocytes was
performed as previously described (Walter et al. 2008). Follicles
containing oocytes were harvested form anesthetized frog. The
oocytes were isolated by a treatment of the follicles with colla-
genase type II in modified Barth medium without Ca
2+
(mM:
88 NaCl, 1 KCl, 0.82 MgCl
2
×6 H
2
0, 5 glucose, 2.4 NaHCO
3
,
15 Hepes at pH 7.4). The oocytes were injected with 23 nl of
cRNA solution (1 μg/μl) and the antisense to the endogenous
Cx38 (400 ng/μl). The antisense DNA (AS38) with the se-
quence C*T*GACTGCTCGTCTGTCCACAC*A*G*
J Bioenerg Biomembr (2012) 44:607–614 609
(* indicates phosphothioate modifications) was purchased
from Eurofins MWG Operon (Ebersberg, Germany). For
the control experiments, 23 nl of AS38 (400 ng/μl) was
injected. For expression, the injected oocytes were incubated
for at least 12 h at 18 °C in modified Barth medium containing
2.4 mM CaCl
2.
Electrophysiological measurements
The recordings of the macroscopic currents with two-electrode
voltage-clamp technique were performed on single Xenopus
oocytes 12–36 h after cRNA injection. The oocytes were
transferred into a perfusion chamber containing nominal
Ca
2+
-free modified Barth solution at room temperature. From
a constant holding voltage of -90 mV, test voltage pulses
rangingfrom-100mVto+30mVwereappliedfor3sin
10 mV steps. The voltage-evoked currents were filtered at
1 kHz and sampled at 5 kHz. The data acquisition and
analysis were performed as previously described (Walter
et al. 2008). For the steady-state current voltage plots
(I(V)), the amplitude of the steady-state currents at the
end of each voltage pulse was measured and plotted against
the corresponding voltage. To estimate the activation
parameters of the connexons, the macroscopic conductance
for the different voltage pulses was calculated and plotted
against the corresponding voltages in a G(V) plot. The data
points were fitted with a simple Boltzmann equation, as
previously described (Walter et al. 2008).
Results
Recently, it was shown that a mutation in the human connexin46
that changed the third amino acid residue in the NT from the
negatively charged aspartic acid to the neutral tyrosine correlates
with cataract (Addison et al. 2006). We used recombinant sys-
tems to investigate the functional consequences of this mutation.
To test whether the D3Y mutation suppressed the ability of
connexins to form gap junction channels, plasmids encoding
EGFP-labeled hCx46wt and hCx46D3Y were transfected into
HeLa cells. Confocal microscopy imaging followed by gap
junction plaque counting revealed that both connexin variants
were transported to the membrane and formed gap junction
Fig. 1 Amino acid sequence of hCx46. The transmembrane domains, as
predicted by TMHMM software (version 2.0), are underlined (Hansen et
al. 2006; Krogh et al. 2001). The mutations introduced by site-directed
mutagenesis are indicated as bold letters above the respective amino acid
Fig. 2 hCx46 gap junction
formation in HeLa cells. The
cells were transfected with
different hCx46-EGFP variants.
After 24 h expression time, the
cell nuclei were stained with
Hoechst. The cells were fixed
with 3.7 % formaldehyde, and
the membranes were stained
with WGA conjugated to Alexa
Fluor 555. (a) Representative
images of each hCx46 variant
are shown. The arrows indicate
gap junction plaques. (b)
Quantification of plaque for-
mation; the percentage of cell
pairs with gap junction plaques
to all cell pairs expressing a
hCx46 variant is given. For
each variant, at least five trans-
fection experiments were per-
formed. The results are given as
the average. The error bars rep-
resent the SEM. The signifi-
cance of the difference was
evaluated by Student’st-test
(** p≤0.01; * p≤0.05)
610 J Bioenerg Biomembr (2012) 44:607–614
plaques (Fig. 1,Fig.2). Gap junction plaques were found in
74.3 % ±1.2 and 75.8 % ±1.7 of all of the transfected cell pairs
expressing hCx46wt and hCx46D3Y, respectively.
To determine whether the gap junction plaques contained
functional gap junction channels, we performed dye transfer
experiments using LY, a negatively charged (-2) 443 Da mol-
ecule, and EthBr, a positively charged (+1) 314 Da molecule,
as tracers (Elfgang et al. 1995;Abbacietal.2008). Both the
wild type and the mutant formed functional gap junction
channels. Of all tested cell pairs expressing hCx46wt,
35.2 % ±3.3 were able to transfer LY, whereas cell pairs
expressing hCx46D3Y showed a significantly increased abil-
ity to transfer LY (62.5 %± 12.5) (Fig. 3). The dye coupling
experiments with EthBr gave comparable results (hCx46wt
29.2 %± 2.4, hCx46D3Y 53 %± 8.3). Quantification of the
plaque formation and size (data not shown) revealed no sig-
nificant differences in the ability of hCx46wt and hCx46D3Y
to form gap junction plaques (Fig. 2). The increase in gap
junction coupling in cells expressing EGFP-labeled
hCx46D3Y was not caused by a higher density of gap junc-
tion channels in the membrane (Fig. 2, Fig. 3).
Connexins of the Cx46-class can form voltage-dependent
gap junction hemichannels when expressed in Xenopus
oocytes (Ebihara et al. 1995; Pal et al. 2000; Paul et al.
1991;Walteretal.2008). We therefore expressed both
hCx46wt and hCx46D3Y in Xenopus oocytes. The hemi-
channels that were formed were analyzed with the two-
electrode-voltage clamp technique. In oocytes expressing
hCx46wt, a voltage-dependent current was found (Fig. 4).
The currents were induced by depolarizing voltages above
-60 mV (Fig. 5). The voltage dependence of the currents is
also shown in the steady-state current-voltage plot (I(V)).
The fitting of the voltage-conductance G(V) plot for
hCx46wt to a Boltzmann equation (Fig. 6) revealed a half
activation voltage (V
1/2
) of -24.02 mV±4.89 and an appar-
ent gating charge zof 1.85± 0.17 (Table 2). In the oocytes
expressing hCx46D3Y hemichannels, a voltage-dependent
current was not observed (Fig. 4,Fig.5). The currents
observed in oocytes expressing the hCx46D3Y mutant are
comparable to those observed in the control oocytes. This
result indicates that, although hCx46D3Y was inserted into
the plasma membrane and formed functional gap junction
channels when expressed in HeLa cells (Fig. 2, Fig. 3), it
was not able to form voltage-sensitive hemichannels when
expressed in Xenopus oocytes (Fig. 4, Fig. 5).
The D3Y mutation replaces a negatively charged residue
with a neutral amino acid residue at the third position
(Fig. 1) and impairs the formation of voltage-dependent
Fig. 3 Dye coupling experiments. A whole-cell patch-clamp configura-
tion was established with a LY- (1 mg/ml) (black bar) or EthBr- (1 mg/ml)
(grey bar) containing pipette-filling solution onto one cell of a HeLa cell
pair expressing EGFP-labeled hCx46 variants. The degree of dye cou-
pling was estimated as the ratio of the sum of the coupled pairs to the sum
of the tested pairs. For each hCx46 variant, at least four transfection
experiments were performed. The results are given as the average. The
error bars represent the SEM. The significance of the difference between
the control (non transfected HeLa cells) and the respectively hCx46
variant was evaluated by Student’st-test ** p≤0.01; * p≤0.05. The
significance of the difference between hCx46wt and the respectively
mutant are indicated by underlined stars
Fig. 4 Representative current evoked in oocytes expressing the differ-
ent hCx46 variants. From a holding potential of -90 mV, the oocytes
were depolarized to 30 mV. The oocytes were injected with the indi-
cated hCx46 variant and AS38. The expression time was 12 h. The
control oocytes were only injected with the AS38
Fig. 5 I(V) plots obtained from oocytes expressing hemichannels com-
posed of hCx46wt (■), hCx46D3Y (○), and hCx46D3E (●), and from
control oocytes (□). The results are given as the average of at least five
different oocytes for each variant. The error bars represent the SEM
J Bioenerg Biomembr (2012) 44:607–614 611
hemichannels (Fig. 4, Fig. 5). Therefore, we speculated that
a reintroduction of a negative charge at the same position
would restore the voltage sensitivity. Using site-directed
mutagenesis, we introduced the negatively charged glutamic
acid (D3E), instead of tyrosine or aspartic acid, to the third
position. Two-electrode voltage-clamp measurements
revealed that like hCx46wt, hCx46D3E expressed in
Xenopus oocytes formed voltage-dependent hemichannels
(Fig. 4, Fig. 5), despite some differences, such as a high
macroscopic conductance (Fig. 5) and a shift of V
1/2
by
approximately 23 mV to -0.53 mV ±0.92 compared with
hCx46wt (Table 2,Fig.6). HeLa cells pairs expressing
hCx46D3E, which was labeled with EGFP showed a reduced
formation of gap junction plaques (63.6 % ±2.6) compared
with cell pairs expressing hCx46wt (74.3 % ±1.2). From the
dye coupling experiments, however, a similar degree of LY
and EthBr transfer between cells pairs forming hCx46D3E
gap junction plaques (LY: 32.2 % ±1.8, EthBr: 30.9 % ±3.6)
and cell pairs forming hCx46wt gap junction plaques (LY:
35.2 % ±3.3, EthBr: 29.2 % ±4.8) was found (Fig. 3).
In summary, the results described above show that the
replacement of aspartic acid, a negatively charged residue,
with tyrosine, a neutral residue, at the third position of the
Cx46 NT did not affect the synthesis and hexamerization of
the connexins or the trafficking and insertion of the connexons
into the plasma membrane. Compared to the wild type, the
D3Y mutation correlated with an increased degree of dye
transfer between the cells and a suppression of the voltage
sensitivity for hemichannels expressed in Xenopus oocytes.
Discussion
The present report examines how the exchange of the third
amino acid residue from a negatively charged aspartic acid
residue to a neutral tyrosine residue (D3Y) in hCx46, which
causes an autosomal dominant zonular pulverulent cataract
(Addison et al. 2006), affects gap junction formation and
functionality. Confocal microscopy analysis of HeLa cells
expressing EGFP-labeled hCx46wt and hCx46D3Y did not
reveal any difference in the expression level and in the forma-
tion of gap junction plaques between the two hCx46 variants
(Fig. 2). It is noteworthy that the transfection efficiency for
both hCx46wt and hCx46D3Yvaried between 50 % and 70 %.
Plaque counting shows that the trafficking of the protein is not
affected by the D3Y mutation (Fig. 2). The mutated
hCx46D3Y protein is synthesized, transported and inserted
into the plasma membrane in a comparable manner and with
a similar intensity to the wild type hCx46. This finding is in
agreement with the observation published by other authors
which shows that mutations in the NT and even deletion of
as much as half of the NT of hCx37 did not affect the transport
of the proteins and their capacity to form gap junction plaques
(Kyle et al. 2008).
We also tested whether the formed gap junction plaques
were functional. LY and EthBr transfer experiments revealed
that the D3Y mutation correlated with an increased degree
of dye coupling (LY: 62.5 % ±12.5, EthBr: 53 % ±8.3)
compared with the wild type (Fig. 3). This increase appears
to be related to the negative charge of the aspartic acid
residue at the third position because a similar ability to
transfer LY and EthBr through gap junction channels was
found in HeLa cells expressing hCx46wt (LY: 35.2 ±3.3,
EthBr: 29.2 % ±2.4) and HeLa cells expressing our gener-
ated mutant hCx46D3E (LY: 32.2 % ±1.8, EthBr: 30.9 %
±3.6). Different scenarios can be proposed to explain the
results observed in the dye transfer experiments for these
hCx46 variants. (a) Within a gap junction plaque composed
of hCx46D3Y, more hemichannels dock to each other and
form open functional gap junction channels between adja-
cent cells. (b) Due to the loss of the negative charge at the
third position, the single-channel permeability to LY and
EthBr of gap junction channels composed of hCx46D3Y is
increased compared with the single-channel permeability of
gap junction channels composed of hCx46wt or hCx46D3E.
Whether mutations in the NT of connexins could affect the
docking of hemichannels in adjacent cells is not yet known.
In contrast, it is known that the N-termini of the six
Fig. 6 G(V) plots obtained from oocytes expressing hemichannels of
hCx46wt (■) and hCx46D3E (●). For each experiment, the conduc-
tance was normalized to the maximum conductance for hCx46wt and
hCx46D3E as appropriate. The data points are the averages of the
normalized values obtained from at least five experiments for each
variant. The error bars are the SEM. The curves represent the fitting of
the data points to the Boltzmann equation
Table 2 Activation parameter as estimated by fitting the G(V) data
points of different experiments with the Boltzmann equation. The data
are given as the averages ±SEM for at least n05 oocytes for each
connexin variant
connexin V
1/2
[mV] z
hCx46wt -24.02 ±4.87 1.85± 0.17
hCx46D3E -0.53 ±0.92 0.43± 0.27
612 J Bioenerg Biomembr (2012) 44:607–614
connexins that form a connexon line the pore entrance and
form a funnel, which might be critical for the opening and
determination of permeability of gap junction channels
(Dong et al. 2006; Kyle et al. 2009; Maeda et al. 2009;
Purnick et al. 2000a). For Cx26, the N-termini of the six
connexins line the pore entrance and form a funnel, which is
stabilized by hydrogen bonds between the aspartic acid
residue at the second position (D2) in one Cx-subunit and
the threonine residue at the fifth position (T5) in the adjacent
Cx-subunit (Beyer et al. 2012; Maeda et al. 2009). In a
homology model of Cx37, an interaction of D3 with F6
and L7 similar to the corresponding interaction in Cx26 is
predicted (Beyer et al. 2012). Despite minor differences in
the amino acid sequences of the Cx37 NT and Cx46 NT, an
organization of the N-termini in a structure comparable to
the funnel proposed for the N-termini of Cx37 can be
assumed. In this structure, the D3 residue plays a key role
in the electrostatic organization of the N-termini. The re-
placement of a charged residue (D3) with a neutral residue
(Y) could affect the electrostatic stability of the funnel and
the permeability of the channels, which would lead to the
results presented in this report (Fig. 3). In addition to form-
ing the funnel, the N-termini line the pore of the channels
and serve as a voltage sensor that contains charged amino
acid residues, which are important for the rapid voltage
gating (Gonzalez et al. 2007; Maeda et al. 2009; Oh et al.
1999,2000,2004; Purnick et al. 2000a,b; Verselis et al.
1994). For Cx46, it was shown that the NT had a crucial role
in voltage-dependent gating (Tong et al. 2004). It was also
found that a replacement of the aspartic acid residue by an
asparagine residue at the third position affected the voltage
gating of Cx46 gap junction channels (Srinivas et al. 2005).
Therefore, it can be hypothesized that the D3Y mutation,
which replaces a negatively charged residue with a neutral
residue, would affect the voltage sensitivity of hCx46 chan-
nels. This hypothesis is verified by the two-electrode voltage-
clamp experiments, which show that the depolarization of
Xenopus oocytes expressing hCx46wt hemichannels activated
a current that was not observed in oocytes expressing
hCx46D3Y hemichannels (Fig. 4, Fig. 5). These results indi-
cate that the D3Y mutation correlated with a loss of voltage
sensitivity. Furthermore, the observation that ooyctes express-
ing hCx46D3E formed voltage-activated hemichannels
(Fig. 4,Fig.5), emphasizes the crucial role of the charged
residue at the third position in the NT for the voltage sensitiv-
ity of hCx46.The mechanism by which this charged residue
controls the voltage gating of hCx46 is not understood so far.
For Cx26, residues 1–10 of the connexins are organized in
helices that can move within a transjunctional voltage field,
leading to opening or closing of the channel vestibule (Maeda
and Tsukihara 2011; Purnick et al. 2000a).For Cx37, a pos-
sibility to form a helical structure between residues 3–16 was
proposed (Kyle et al. 2009). The NT of hCx46 is not
completely similar to the NT of Cx26 and Cx37. However,
as proposed by other authors (Beyer et al. 2012; Maeda and
Tsu kihar a 2011; Srinivas et al. 2005), it could be assumed that
the hCx46 N-termini may adopt helical structures with the
aspartic acid residues facing the cytoplasmic space. Because
of the charged aspartic acid residues, the N-termini could
move in the voltage field, leading to channel opening or
closing (Maeda and Tsukihara 2011). Structure analysis using
NMR spectroscopy or crystal structure analysis, which we
cannot offer, should be performed to clarify the structure of
the hCx46 NT. On the basis of our results, we propose that the
replacement of a negatively charged residue with a neutral
residue at the third position in the NT of hCx46 suppresses a
regulatory mechanism of the resultant channels.
Conclusion
The present report shows that the hCx46D3Y mutation,which
correlates with a cataract, did not change the functional ex-
pression of hCx46. Dye transfer experiments revealed an
increased degree of coupling for cells expressing hCx46D3Y
compared with cell pairs expressing hCx46wt. A replacement
of the aspartic acid residue at third position by a glutamic acid
residue, which is also negatively charged, showed a reduced
formation of gap junction plaques, but the degree of coupling
was similar for cells expressing hCx46D3E and hCx46wt.
Thus, the negatively charged residue at the third position
could be involved in the docking of functional gap junction
channels and/or the modulation of the permeability of hCx46
channels. Additionally, we show that the negatively charged
aspartic acid residue at the third position is essential for correct
voltage sensitivity of hCx46 gap junction hemichannels and
most likely also for the gap junction channels. The loss of
voltage sensitivity is a loss of a regulatory mechanism, which,
in turn, could impair the function of gap junction channels in
the lens, leading to cataract formation.
Acknowledgements This work was supported by Transregio TR37.
We thank Viviane Berthoud for the hCx46 clone.
References
Abbaci M, Barberi-Heyob M, Blondel W, Guillemin F, Didelon J
(2008) Advantages and limitations of commonly used methods
to assay the molecular permeability of gap junctional intercellular
communication. Biotechniques 45:33–62
Addison PK, Berry V, Holden KR, Espinal D, Rivera B, Su H,
Srivastava AK, Bhattacharya SS (2006) A novel mutation in the
connexin 46 gene (GJA3) causes autosomal dominant zonular
pulverulent cataract in a Hispanic family. Mol Vis 12:791–795
Arora A, Minogue PJ, Liu X, Addison PK, Russel-Eggitt I, Webster
AR, Hunt DM, Ebihara L, Beyer EC, Berthoud VM, Moore AT
J Bioenerg Biomembr (2012) 44:607–614 613
(2008) A novel connexin50 mutation associated with congenital
nuclear pulverulent cataracts. J Med Genet 45:155–160
Arora, A., Minogue, P.J., Liu, X., Reddy, M.A., Ainsworth, J.R.,
Bhattacharya, S.S., Webster, A.R., Hunt, D.M., Ebihara, L.,
Moore, A.T., Beyer, E.C. and Berthoud, V.M. (2006) A novel
GJA8 mutation is associated with autosomal dominant lamellar
pulverulent cataract: further evidence for gap junction dysfunction
in human cataract. J Med Genet. 43:e2 (http://www.jmedgenet.-
com/cgi/content/full/43/1/e2)
Berthoud VM, Beyer EC (2009) Oxidative stress, lens gap junctions,
and cataracts. Antioxid Redox Signal 11:339–353
Berthoud VM, Minogue PJ, Guo J, Williamson EK, Xu X, Ebihara L,
Beyer EC (2003) Loss of function and impaired degradation of a
cataract-associated mutant connexin50. Eur J Cell Biol 82:209–221
Beyer EC, Lipkind GM, Kyle JW, Berthoud VM (2012) Structural
organization of intercellular channels II. Amino terminal domain
of the connexins: sequence, functional roles, and structure. Bio-
chim Biophys Acta 1818:1823–1830
Donaldson P, Kistler J, Mathias RT (2001) Molecular solutions to
mammalian lens transparency. News Physiol Sci 16:118–123
Dong L, Liu X, Li H, Vertel BM, Ebihara L (2006) Role of the N-
terminus in permeability of chicken connexin45.6 gap junctional
channels. J Physiol 576:787–799
Ebihara, L., Berthoud, V.M. and Beyer, E.C. (1995) Distinct behavior
of connexin56 and connexin46 gap junctional channels can be
predicted from the behavior of their hemi-gap-junctional chan-
nels. Biophys J.68:1796-1803
Elfgang C, Eckert R, Lichtenberg-Frate H, Butterweck A, Traub O,
Klein RA, Hulser DF, Willecke K (1995) Specific permeability
and selective formation of gap junction channels in connexin-
transfected HeLa cells. J Cell Biol 129:805–817
Falk MM, Kumar NM, Gilula NB (1994) Membrane insertion of gap
junction connexins: polytopic channel forming membrane pro-
teins. J Cell Biol 127:343–355
Gerido DA, White TW (2004) Connexin disorders of the ear, skin, and
lens. Biochim Biophys Acta 1662:159–170
Gonzalez D, Gomez-Hernandez JM, Barrio LC (2007) Molecular basis
of voltage dependence of connexin channels: an integrative ap-
praisal. Prog Biophys Mol Biol 94:66–106
Goodenough DA (1992) The crystalline lens. A system networked by
gap junctional intercellular communication. Semin Cell Biol
3:49–58
Hansen L, Yao W, Eiberg H, Funding M, Riise R, Kjaer KW, Hejt-
mancik JF, Rosenberg T (2006) The congenital “ant-egg”cataract
phenotype is caused by a missense mutation in connexin46. Mol
Vis 12:1033–1039
Krogh A, Larsson B, von Heijne G, Sonnhammer EL (2001)
Predicting transmembrane protein topology with a hidden
Markov model: application to complete genomes. J Mol Biol
305:567–580
Kyle JW, Berthoud VM, Kurutz J, Minogue PJ, Greenspan M, Hanck
DA, Beyer EC (2009) The N terminus of connexin37 contains an
alpha-helix that is required for channel function. J Biol Chem
284:20418–20427
Kyle JW, Minogue PJ, Thomas BC, Domowicz DA, Berthoud VM,
Hanck DA, Beyer EC (2008) An intact connexin N-terminus is
required for function but not gap junction formation. J Cell Sci
121:2744–2750
Lichtenstein A, Gaietta GM, Deerinck TJ, Crum J, Sosinsky GE,
Beyer EC, Berthoud VM (2009) The cytoplasmic accumula-
tions of the cataract-associated mutant, Connexin50P88S, are
long-lived and form in the endoplasmic reticulum. Exp Eye Res
88:600–609
Maeda S, Nakagawa S, Suga M, Yamashita E, Oshima A, Fujiyoshi Y,
Tsukihara T (2009) Structure of the connexin 26 gap junction
channel at 3.5 A resolution. Nature 458:597–602
Maeda S, Tsukihara T (2011) Structure of the gap junction channel and
its implications for its biological functions. Cell Mol Life Sci
68:1115–1129
Mathias RT, Rae JL, Baldo GJ (1997) Physiological properties of the
normal lens. Physiol Rev 77:21–50
Mathias RT, White TW, Gong X (2010) Lens gap junctions in growth,
differentiation, and homeostasis. Physiol Rev 90:179–206
Minogue PJ, Liu X, Ebihara L, Beyer EC, Berthoud VM (2005) An
aberrant sequence in a connexin46 mutant underlies congenital
cataracts. J Biol Chem 280:40788–40795
Minogue PJ, Tong JJ, Arora A, Russell-Eggitt I, Hunt DM, Moore AT,
Ebihara L, Beyer EC, Berthoud VM (2009) A mutant connexin50
with enhanced hemichannel function leads to cell death. Invest
Ophthalmol Vis Sci 50:5837–5845
Oh S, Abrams CK, Verselis VK, Bargiello TA (2000) Stoichiometry of
transjunctional voltage-gating polarity reversal by a negative
charge substitution in the amino terminus of a connexin32 chime-
ra. J Gen Physiol 116:13–31
Oh S, Rivkin S, Tang Q, Verselis VK, Bargiello TA (2004) Determi-
nants of gating polarity of a connexin 32 hemichannel. Biophys J
87:912–928
Oh S, Rubin JB, Bennett MV, Verselis VK, Bargiello TA (1999)
Molecular determinants of electrical rectification of single chan-
nel conductance in gap junctions formed by connexins 26 and 32.
J Gen Physiol 114:339–364
Pal JD, Liu X, Mackay D, Shiels A, Berthoud VM, Beyer EC, Ebihara
L (2000) Connexin46 mutations linked to congenital cataract
show loss of gap junction channel function. Am J Physiol Cell
Physiol 279:C596–C602
Paul DL, Ebihara L, Takemoto LJ, Swenson KI, Goodenough DA
(1991) Connexin46, a novel lens gap junction protein, induces
voltage-gated currents in nonjunctional plasma membrane of
Xenopus oocytes. J Cell Biol 115:1077–1089
Purnick PE, Benjamin DC, Verselis VK, Bargiello TA, Dowd TL
(2000a) Structure of the amino terminus of a gap junction protein.
Arch Biochem Biophys 381:181–190
Purnick PE, Oh S, Abrams CK, Verselis VK, Bargiello TA (2000b)
Reversal of the gating polarity of gap junctions by negative
charge substitutions in the N-terminus of connexin 32. Biophys
J 79:2403–2415
Sohl G, Willecke K (2004) Gap junctions and the connexin protein
family. Cardiovasc Res 62:228–232
Srinivas M, Kronengold J, Bukauskas FF, Bargiello TA, Verselis VK
(2005) Correlative studies of gating in Cx46 and Cx50 hemi-
channels and gap junction channels. Biophys J 88:1725–1739
Thomas BC, Minogue PJ, Valiunas V, Kanaporis G, Brink PR, Berthoud
VM, Beyer EC (2008) Cataracts are caused by alterations of a
critical N-terminal positive charge in connexin50. Invest Ophthal-
mol Vis Sci 49:2549–2556
Tong JJ, Liu X, Dong L, Ebihara L (2004) Exchange of gating properties
between rat cx46 and chicken cx45.6. Biophys J 87:2397–2406
Verselis VK, Ginter CS, Bargiello TA (1994) Opposite voltage gating
polarities of two closely related connexins. Nature 368:348–351
Walter WJ, Zeilinger C, Bintig W, Kolb HA, Ngezahayo A (2008)
Phosphorylation in the C-terminus of the rat connexin46 (rCx46)
and regulation of the conducting activity of the formed connex-
ons. J Bioenerg Biomembr 40:397–405
White TW, Bruzzone R (2000) Intercellular communication in the eye:
clarifying the need for connexin diversity. Brain Res Brain Res
Rev 32:130–137
614 J Bioenerg Biomembr (2012) 44:607–614
