Cardiovascular Research 62 (2004) 276 – 286
www.elsevier.com/locate/cardiores
Review
Biophysical properties of homomeric and heteromultimeric channels
formed by cardiac connexins
Alonso P. Moreno *
Krannert Institute of Cardiology, Indiana University School of Medicine, 1800 N. Capitol Ave. Suite 310, Indianapolis, IN 46202, USA
Received 23 October 2003; received in revised form 27 February 2004; accepted 1 March 2004
Time for primary review 21 days
Abstract
Substantial advances have been made in characterizing the biophysical properties of channels formed exclusively by connexin isoforms
expressed mainly in the heart, e.g., Cx43, Cx45 or Cx40. These properties include transjunctional and transmembrane voltage gating as well as
their perm-selectivity, chemical gating and, at a single channel level, their multiple open states and changes in mode behavior. Nonetheless,
these connexins are rarely expressed individually in a cell and the presence of functional channels constituted by distinct connexin isoforms is
now suspected to be a norm. In fact, combinations of the connexins that form heteromeric channels have been described in some tissues,
increasing the necessity to reinforce the research that leads to understanding the effects of these heteromeric interaction on the gating and
conducting characteristics of the channels. Furthermore, protein – protein interaction studies will help to understand which connexin domains
are involved in these interactions and how they affect the physiology of channels and their interaction with other biological and structural
molecules in the cell.
New information on the biophysical properties of heteromultimeric channels suggests that interaction between connexins and connexons is
not as simple as once thought. Theoretically, changes in the coupling of homomeric connexons (Cx43) in the myocardium may not be
significant enough to change the physiology of the heart or to incite arrhythmias, but when heteromeric channels are present, alteration in
conductance, differential gating sensitivity to bio-gating molecules and changes in voltage sensitivity increase substantially the cell resources
to modulate intercellular coupling, which may participate in the physiology and/or pathology of the cardiovascular tissues.
D 2004 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
Keywords: Heteromeric connexons; Oligomerization; Phosphorylation; Gating; Permeability
1. Introduction
Among the multiple types of channels expressed in
vertebrate cells, only gap junction channels are known to
allow intercellular communication between the cytoplasms of
contiguous cells in tissues, and in some configurations, these
channels allow direct communication between the cytoplasm
and the extracellular media. Gap junction channels can be
formed from one or various proteins named connexins that
constitute a family of highly homologous proteins. A new
family of proteins termed pannexins is emerging, and it will
be necessary to determine their abundance and physiological
relevance in cardiovascular tissue [1].
* Tel.: +1-317-962-0074; fax: +1-317-962-0574.
E-mail address: [email protected] (A.P. Moreno).
Some of these channels are highly selective for small ions
or molecules, and their gating mechanisms involve intracellular messengers; this indicates that their presence is required
to maintain essential functions in many organs. Natural or
induced mutations in these proteins cause the development of
severe diseases [2].
A gap junction channel is formed when two hexameric
connexons dock; one connexon is provided by each of
two contiguous cells (see Fig. 1). Connexons not apposed
by any other connexon are known as hemichannels.
Connexons and hemichannels are predicted to be either
homomeric (all six subunits identical) or heteromeric (two
or more different subunits; see Fig. 1). Full channels are
predicted to exist in various configurations. These channels are homomeric if all subunits are identical, and also
homotypic if both connexons are identical. Channels
formed by different connexins are generally considered
0008-6363/$ - see front matter D 2004 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.cardiores.2004.03.003
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
277
Fig. 1. Combinations of connexins forming connexons, hemichannels, and full channels. Connexin45 and Cx43 are represented in white and grey, respectively.
heteromultimeric. Various subgroups of heteromultimeric
channels are distinguishable. These channels are heterotypic when all six connexins are identical in each connexon, but
different for each connexon; mono-heteromeric when one
homomeric connexon docks with one heteromeric connexon,
and bi-heteromeric when they are formed by two heteromeric
connexons.
The permeability of cardiac gap junction channels has
been studied for more than 35 years [3]. This has been a
great challenge, mainly because the proteins (connexins)
that form these channels display intrinsic isoform diversity
and have the capacity to form heteromeric connexons in
vivo [4]. Since each connexin has unique gating and
perm-selective properties, some combinations of connexins produce new biophysical properties that definitively
change the interconnecting properties between cells. One
clear example is seen in cardiac remodeling, where
changes in connexin expression lead to changes in communication and possibly produce arrhythmias [5]. The
medical ramifications of these changes create an inherent
necessity for understanding interactions between connexon
subunits and their impact on the biophysics of the
channels formed.
Cardiac tissue from most mammalian species expresses
similar types of connexins: Cx40, Cx43, and Cx45. The
abundance of Cx43 in cardiac musculature and of Cx40
and Cx45 in conductive cardiac tissue provides the heart
with communicating elements that allow the organ to
work synchronously [6]. Similarly, during early heart
development, the abundance of Cx45 [7], which is
known to form channels with low permeability, suggests
a strong involvement for this connexin, not only in
conduction of electrical signals but in differentiation of
tissues.
2. Perm-selectivity
Early studies detected diffusion of molecular tracers in
cardiac tissue, indicating that ions and other molecules
could cross between cells [3]; this movement of ions
provided evidence of differences in the conductance of
channels formed by distinct connexins [8]. The use of
fluorescent molecules has helped elucidate some aspects
of the selectivity of these channels; it was determined that
the permeability of gap junction pores depends not only
on the molecular properties of the permeant, but on the
structure and properties of the channel, including its pore
structure [9], the surface charge [10] and, in some
instances, on the interaction between the pore and the
permeant [11]. These differences in perm-selectivity have
been proposed to be relevant in signalling across the
developing heart and represent an important tool for
determining the mechanism of conduction and rectification in particular combinations of connexins.
2.1. Homotypic channels. Unitary conductance
Since the flux of ions across the adult heart is the main
function of gap junctions in that organ, quantification of
conductances of channels formed by distinct connexins
allow correlation of these channels with physiological and
pathological processes. To determine the unitary conductance of gap junction channels, it is necessary to obtain
278
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
electrophysiological measurements through a double
whole-cell voltage-clamp technique using tumour cells as
an exogenous system of expression [2]. Three main states
have been assigned to all cardiac gap junction channels:
(1) open, which is the condition of maximal conductance;
(2) residual, which is the non-voltage dependent state that
is associated with voltage gating, but not with chemical
gating; and (3) the closed or non-conductive state. The
residual state is 10 –30% of maximal conductance and
apparently represents incomplete closure of channels when
the carboxyl terminal (CT) blocks the channel pore[12,13].
This carboxyl terminal does not need to be attached to
channel structures, because residual states can be observed
with independent co-expression of truncated channels and
the carboxyl terminal residues [12].
Among the cardiac gap junction channels, the open
conductance of Cx45 is the smallest at 30 –40 pS [14,15],
that of Cx40 is the largest with values between 175 and 210
pS [16], and Cx43 channels have an intermediate open
conductance with values of 60– 120 pS [17,18]. The differences seen in unitary conductance values among Cx43
channels have been attributed to the phosphorylation state
of the cell [8,18]. Differences in unitary currents during one
single recording are sometimes difficult to explain; the most
plausible rationalization is the formation of heteromeric
connexons (see below), because most tumour cells used in
exogenous systems express endogenous connexins, mainly
Cx45.
2.2. Homotypic channels. Selectivity
Gap junction channels, formed by cardiac connexins,
display great selectivity based on molecular charge and
molecular size. This lends a variety of selective and conductive properties to cardiac regions that express distinct
connexins. Experiments using symmetric ion substitutions
via patch pipettes have been performed, and a new scenario
has developed wherein selectivity is not correlated with the
unitary conductance of channels; this indicates that the
selectivity process is not simple [9].
Charge selectivity varies depending on the connexin isoforms present in the channels. Cx45 has the strongest selectivity for cations with a 10:1 ratio, while Cx40 favours cations
at a ratio of 4:1 and rCx43 at only 1.2:1 [19]. It is important to
note that there is no correlation between unitary conductance
and selectivity. In experiments where selectivity has been
studied using asymmetric solutions, it was reported that
PK+/PCl- ratios for Cx43 and Cx40 were much greater than
those predicted by conductance – mobility ratios of these ions
[20,21]. Apparently, this is due to fixed charges in the pore
region that dominate the conductance ratios.
Studies on selectivity among different cations and anions
are still being processed. For Cx43 channels, their charge
selectivity is small and reports indicate that these channels
are little selective, and there is a linear correlation between
the permeability and the mobility of cations, although this is
not linear for anions [21]. A more complex selectivity
process has been described for connexin40 in which cation
permeability follows the aqueous mobility of entities used,
but anion movement does not match at all. Furthermore, the
electrical conductance of the pore is reduced when anions
permeate, indicating an anomalous interaction between
charged species permeating the pore [20]. Further studies
will be required to understand the interaction of Cx45 in
heterotypic or heteromeric channels.
2.3. Homotypic channels. Permeability to large molecules
The pores of cardiac connexins are permeable to different
large molecules up to f 1 kDa. The quest to find the limiting
diameter, charge selectivity and ability to facilitate passage of
specific cytoplasmic molecules requires the use of distinct
types of molecules, including those with a variety of sizes and
charges.
To determine the limiting diameter of channels, it is
necessary to use either uncharged molecules or molecules
with different sizes but with an identical charge and shape.
Cx43 appears to have a larger limiting diameter than does
Cx40, since it allows mannitol to permeate [21]. Compared to
other connexins like Cx32, Cx37 and Cx43, Cx40 ranks
among the smallest in its limited diameter for molecules to
pass through the pore. This may be significant in the
movement of molecules across tissue boundaries during
differentiation.
Few experiments have been systematically performed to
determine the selectivity of cardiac gap junction channels
for charged molecules. An attempt using tetra-alkylammonium derivatives showed that TMA and TEA (radii of 3.47
and 4 Å, respectively) pass through Cx40 and Cx43
channels, although comparative studies for larger molecules could not be performed. Cx43 allows rapid permeation of 6-carboxyfluorescein and dichlorocarboxyfluores
cein compared to Cx40 and Cx45, and Cx45 (with a smaller
conductance) has greater permeability to these anions than
does Cx40 [9].
Cx43 channels are permeable to glutamate, glutathione,
ADP and ATP [22], and Cx40 is permeable to glutamate.
Although more studies on Cx40 and Cx45 are necessary,
the main conclusion is that differences in permeability to
signalling molecules between cells could cause a temporal
effect or response and could be altered by changing
connexin expression.
An interesting approach to regulating permeability by
transjunctional voltage (Vj) was reported for Cx43, indicating
that the permeability of Cx43 channels at the residual state is
significantly reduced by rectifying properties [23]. Voltage
gating induces narrowing of the pore and greatly reduces dye
transfer. This property could act as a voltage-sensitive gating
mechanism that selectively filters molecules to limit metabolic communication, while still allowing electrical coupling.
How this Vj gating – permeability relationship applies under
physiological conditions remains to be determined.
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
279
In summary, unitary conductance, cation selectivity and
anion selectivity have been approximated for cardiac connexins, although no complete systematic studies have been
performed exclusively for these molecules. Interestingly,
differences in permeability vary during phosphorylation of
connexins [24], indicating that charges close to the mouth of
the pore or in the pore itself alter permeability to large
charged molecules.
All these data forecast that the movement of signalling
molecules according to the connexins expressed causes
differences in physiological responses. Although not yet
studied for cardiac connexins, the combination of Cx26 and
Cx32 in reconstituted bilayers produces channels capable of
selecting between cAMP and cGMP [32].
2.4. Heteromultimeric channels. Unitary conductance
One of the first indications of functional hemichannels
was obtained from Cx46 expressed in Xenopus oocytes [33].
Later, a more physiological preparation from fish retinal cells
showed channels whose current inactivation was strongly
voltage dependent [34]. Among the cardiac connexins, Cx45
has been fully characterized as a functioning hemichannel in
an exogenous expression system [15]. Its unitary conductance of 70 pS does not depend on polarity and its residual
state of 15 pS resembles those found in homotypic full
channels.
It has been proposed that Cx43 hemichannels are
present in some cell systems and that their appearance
seems to be correlated with oxidative stress[35,36]. The
active form of these channels has been recently reported
[37]. The unitary conductance of Cx43 hemichannels is, as
expected, twice the conductance of the full channels that
are found in cell pairs.
Since Cx45 forms abundant hemichannels and Cx43 is
able to form heteromeric connexons with Cx45, it would be
prudent to determine if the few hemichannels observed in
HeLa or other cells are the product of heteromerization,
which allows formation of semi-functional channels when
Cx45 is present.
The three major cardiac connexins have been proven to
form heterotypic channels, although formation of functional
channels between Cx43 and Cx40 appears to be restricted
[16]. For heterotypic Cx43 – Cx45 channels, the unitary
conductance corresponds to the Ohmic sum of the predicted
conductances for each of the homomeric native connexons
( f 50– 60 pS), which indicates a good association between
these two connexins [25]. Cx45 and Cx40 heterotypic
docking produces similar results, with the resulting unitary
conductances very close to 45 pS [26]. For Cx43 – Cx40, the
unitary conductance of channels is dependent on the transjunctional voltage (Vj), with main states of 60 and 100 pS
(Cx43 side made positive or negative, respectively).
Biochemically, Cx45 and Cx43 can be co-immunoprecipitated, indicating a close relationship between these connexins in vivo [27]. In the same way, Cx40 and Cx43 coimmunoprecipitate [28]. Although the results of biochemical
methods suggest formation of heteromeric channels, a complementary method for testing whether heteromeric channels
are actually functional is by finding incorporation of multiple, new unitary conductance levels intermediate between
the conductance levels known to occur in homotypic and
heterotypic channels [29]. For Cx40 and Cx43, most of the
values for unitary conductances are in the lower range
between the unitary conductances of homotypic channels
formed by each connexin; however, an influence from
endogenous Cx45 could not be eliminated.
Co-expression of Cx43 and Cx45 also produces intermediate sizes, but unitary conductances smaller than Cx45
homotypic channels have been found abundantly, indicating
that there are substantial differences between heteromeric
channels formed by different combinations of connexins
[30].
Mono-heteromeric combinations between Cx40 and
Cx43 connexins indicate that their unitary conductances
are smaller than expected from homotypic Cx40 or Cx43,
suggesting that there are probably some connexin combinations that are not functional, although, again, the influence
of endogenous Cx45 cannot be excluded [28]. Interesting
results have been presented in calculating molecular flux
across connexins by using traditional measurements; LY
was found to be five times more permeable though Cx43
channels than through Cx40 channels (f 1600 to 300
molecules/channel per second), and heteromeric channels
yield intermediate flux values [31].
2.5. Hemichannels. Unitary conductance
2.6. Hemichannels. Selectivity
Isolated HeLa cells expressing Cx45 have been shown
to take up propidium iodide (e = + 2) and LY (e = 2) at
relatively similar rates when exposed to low extracellular
calcium solutions, although these data do not clearly
correspond to the normal flux of dyes in Cx45 full
channels from the same cells. Normally, LY diffuses
rapidly, while PI is barely detectable. This change in
diffusion patterns for these dyes indicates that docking
between connexons probably changes the selectivity/permeability ratio for Cx45 channels [15]. Small changes in
the internal shape of the pore might substantially reduce
permeability to large molecules, leaving the unitary
conductance values relatively intact.
The few studies of the selectivity of Cx43 hemichannels
indicate that, as is seen in their homotypic counterparts,
their pore is not selective for charged molecules, since
positively (ethidium bromide [37]) or negatively charged
(LY, 6-carboxyfluorescein) dyes can be taken up by tumour
cells [37,38] in culture or by primary isolated cells [35].
Although interesting, it has been surprising to observe that
the uptake of dye takes place in cells under control
280
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
conditions where hemichannel detection by voltage clamp
has been low [37], indicating that other channels or transporters might be involved.
3. Voltage gating
The sensitivity of gap junction channels to voltage has
been well characterized; the macroscopic conductance
becomes reduced upon transjunctional (Vj) or transmembrane (Vm) voltage. In general, the processes of gating are
complex, and, although some parameters can be used to
compare sensitivity of connexin isoforms, multiple issues
need to be clarified, including the mechanism for instantaneous gating.
3.1. Homotypic channels. Transjunctional voltage dependence (Vj)
Vj gating of connexin channels has been studied in cell
pairs, and these studies show a reduction of the macroscopic
current as voltage of either polarity is applied to either cell
of a pair. This yields a non-linear relationship that can be
fitted for most of connexons with a Boltzmann equation,
stated as G=( Gmax Gmin)/(1 + exp( A(V Vo))) + Gmin.
Here, Gmax and Gmin are the maximal and minimal conductances. ‘‘A’’ equals (nq)/(kt) or the energy term that
expresses the voltage sensitivity where n is the equivalent
of a charge moving through the voltage field, q is the unitary
charge, k is Boltzmann’s constant and t is the temperature.
Vo corresponds to the voltage at which the change in
conductance is half-maximal. In general terms, connexins
vary in their Vo, Gmin and n, indicating that voltage sensitivity and residual conductance for each connexin is different (see Fig. 2).
Voltage gating of cardiac connexins is similar, although
each channel type is regulated by its own gating constants.
The most remarkable property is that at voltages above Vo,
the channels reach a sub-conductance level that is not
voltage dependent [39] and whose size is between 15%
and 30% maximal unitary conductance. This residual conductance has its own gating properties [23], which is likely
to be the result of a partially closed channel. This fast gating
has been related to the carboxyl terminal (CT). As the CT is
reduced in size, the channel loses its fast gating, and slow
gating governs closing of the channel [12,40], apparently
separating the gating mechanism. When the tail of Cx43 is
expressed separately, it interacts with the truncated connexin
to re-establish residual conductances [12]. Furthermore, the
CT of Cx43 is capable of recognizing truncated channels of
Cx40 and inducing gating behaviour that includes similar
residual conductances [13]. Altogether, these data indicate
that residual conductance may represent incomplete closure
of a pore by the CT. It is worth mentioning that this gating
mechanism is similar in other connexins but can be mediated by other molecular regions, as is seen in Cx26 where
the N terminus is responsible for fast gating and residual
conductance [41,42].
Fig. 2. Transjunctional voltage gating. Comparison of macroscopic gating properties between homotypic channels formed of either Cx43, Cx40 and Cx45. Top
left: A representative inactivation of currents when a F 100 mV is applied across Cx43 channels. Bottom left: A cartoon representing a connexin channel with
its two gating domains during the application of a voltage potential of distinct polarity. According to the polarity applied, one of the two domains, here
representing the carboxyl terminal, will gate preferentially.
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
3.2. Heteromultimeric channels. Transjunctional voltage
dependence
In a simple configuration such as heterotypic channels,
the voltage gating properties of connexon channels become
altered. For pairs formed of heterotypic Cx45 –Cx43, it has
been reported that docking of these connexons inhibited
the gating of Cx43 [25]. This was confirmed by modelling
the gating behaviour of these channels, where the voltage
drop across the smaller of the connexons, in this case
Cx45, was much larger than across the opposite connexon
[43]. The gating sensitivity was not inhibited, but the
actual voltage drop was reduced. At a microscopic level,
these channels have an intermediate conductance, but their
residual conductance disappeared by a mechanism still not
determined [25].
Pairing of Cx40 and Cx43 to form heterotypic channels
in RIN or HeLa cells displayed a sum of their properties
instead of an increase in rectification. This sum likely
indicates that these connexins gate to the same polarity
[16]. Using distinct connexin combinations, it is clear that
Cx43 and Cx45 gate to inside negativity, but Cx40 paired
with Cx37 appeared to gate to positivity [44]. This concept
has currently been challenged by experiments with HeLa
cells that express heterotypic Cx45 –Cx40 [26]. The steep
difference in conductance of connexons and lack of
resolution that is caused by series resistance in some
experiments may have hindered finding the real voltage
gating polarity of Cx40. Although conclusive, these results
must be reassessed in other systems in order to verify that
the polarity of Cx40 is not inverted according to the
opposing connexon.
For bi-heteromeric channels, it has been demonstrated
that Cx45, Cx43 and Cx40 can form heteromeric connexons; their voltage gating appears to be a blend of the
properties of the parental homomeric connexons. For
Cx40 and Cx43, and for Cx45 and Cx43, gating at distinct
ratios seems to follow specific values of gating, as if their
combination intermingles, producing novel properties,
where the sensitivity of the connexons seemed changed
[30,45] (but see Ref. [28]). Vj properties for heteromeric
combinations of Cx45/Cx40 connexon combinations are
presently unavailable.
A recent report showed that the formation of monoheteromeric channels, Cx45/Cx43 – Cx43 combinations induced further reduction in Cx43 connexon gating, which
confirmed that these heteromeric Cx45/Cx43 connexons
had a lower unitary conductance and induced a smaller
voltage drop across the opposing homomeric connexon, in
this case, Cx43 [30].
The voltage gating of mono-heteromeric combinations of
Cx40 and Cx43 could be not be fit by the combination of
homotypic and heterotypic properties, indicating that the
heteromeric channels probably have functional significance
with regard to voltage-dependent gating [45] (but see Ref.
[28]).
281
3.3. Homotypic channels. Transmembrane voltage gating
Of all cardiac connexins, Cx45 shows the strongest
sensitivity to transmembrane voltage. This gating occurs
when the membrane potential of both cells changes simultaneously. This type of voltage gating has been called Vi – o
or Vm and is substantialy slower than Vj. In the case of Cx45
channels, the conductance of the junction increases up to
2.5-fold from 100 to 0 mV in the mouse homologue. For
other homologues, the dependence is not that steep, but
there is always an increment to positivity. Cx43 is less
sensitive to Vm and gating occurs in the opposite direction,
where conductance decreases during depolarization [46].
Cx40 appears to be totally insensitive to Vm [47]. The
decrease or increase in conductance in all cases occurs as
a change in open probability, since no changes in unitary
conductance have been reported for different polarities
[25,39,47].
3.4. Heteromultimeric channels. Transmembrane voltage
gating
In heterotypic channels formed by Cx43 – Cx45 or
Cx40– Cx45, the dependence for Vm becomes substantially
reduced (30% and 50%, respectively [26,48]). Since the
sensor for Vm appears to be in the docking domains of
channels [49], it is possible that the sensing and/or gating
mechanisms become altered as docking occurs in heterotypic channels.
3.5. Hemichannels. Voltage gating
The mechanisms of gating are maintained in Cx45 hemichannels, where for hyperpolarizing pulses, the channels
gate rapidly, maintaining most of the properties of the
connexons. Voltage dependence with Vo = 11.1 mV is comparable to that reported for homotypic channels [14,15].
This indicates that the gate and sensor of the hemichannel
are similar to those present after docking in a connexon. For
depolarizing pulses, the Ca2 +-dependent loop gate opens
with slow kinetics, such as those shown for hemichannels
formed by other connexins (Cx46, Cx50).
At a microscopic level, the behaviour of channels resembles gating to Vj voltage at hyperpolarizing potentials, with
changes in conductance to a residual state of 15 pS; this
indicates that these hemichannels follow fast gating where
the CT partially blocks conductance of the channels [15]. In
most of the hemichannels recorded, the presence of loop
gating can be considered the mechanism for keeping the
channel in the completely closed state. This gating is
regulated by extracellular calcium, and gates open under
depolarizing pulses. As for other hemichannels, studies on
cardiac connexins need to be focused on determining
whether this loop gating corresponds to the Vm gating
mechanism in a full channel, and how it contributes to
chemical gating.
282
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
3.6. Loop gating
At the single channel level, the gating of all cardiac
connexin homotypic channels shows slow transitions between open and closed states; this indicates that eventually
these channels can fully close. This slow gating, which was
associated initially with loop gating in Cx46 hemichannels
[50], appears to be present also in full channels. Slow gating
shows gating times similar to those observed during chemical gating, as we will see below, and has been detected in
Cx43 channels where the CT has been truncated [12,51].
4. Chemical gating
As for other connexins, the biophysical properties of
cardiac gap junction channels are known to be biochemically modulated. Some of these compounds are thought to
affect the interaction of channels with their surrounding
membranes, and others are physiological intracellular compounds. The latter effect is understandable, since the pores
of these channels are facing the intracellular millieu. Several
compounds have been reported to reduce junctional coupling, but only a few can be considered to have a direct
effect on gap junction channels.
4.1. N-Alkyl alcohols
Heptanol and octanol, first described as gap junction
channel blockers in invertebrates [52], are known to rapidly
and reversibly inhibit gap junctional conductance. It has
been determined that these alcohols reduce the open channel
probability with little effect on the unitary conductance of
the channels [13,53]. These alcohols are effective in all
cardiac connexins. Their mechanism of action has been
proposed to be mediated by changes in the cholesterol-rich
region that surrounds connexin channels.
4.2. Halothane
Halothane’s rapid action in reducing junctional conductance is reversible and does not affect the unitary conductance of the channels [54]. Although it is not specific for gap
junction channels, since it affects even cellular receptors
[55], it appears that channels from some connexins have
distinct sensitivity to halothane [56]. These differences may
create some interesting structural correlations during investigation of the gating behaviour of heteromultimeric channels using other uncouplers. Halothane is one of the few
uncouplers that has been systematically applied to compare
sensitivity between homomeric and heteromeric connexons
formed by Cx43 and 40. Heteromeric connexons display
more sensitivity to halothane than do homomeric connexons, since uncoupling occurs in heteromeric channels twice
as rapidly. This indicates that the packing of heteromeric
connexons in the membrane offers less stability and renders
a structure that could be affected more readily by lipophilic
agents that affect membrane structure.
4.3. Glycyrrhetinic acids
These represent another group of compounds that reduce coupling in almost all connexins [57,58] and whose
mechanism of action have not yet been determined. These
compounds affect coupling in long-term cell cultures at
low concentrations, although they display toxic effects at
higher concentrations. At low concentrations, glycyrrhetinic acids do not completely block communication among
cells; this probably indicates partial block of the channels
[58]. Reports indicate that these changes are unrelated to
protein phosphorylation and that phosphorylation is not a
requirement, since communication through Cx26 channels
is affected. It appears that the change in conductance is
independent of the expression of channels, but relies
mostly on alteration of channel distribution in plaques
[58].
4.4. Intracellular acidification
The process of ischemia in cardiac cells produces a
strong and significant reduction of intracellular pH that
has been associated with healing after an infarct, or arrhythmias due to electrical uncoupling between cells. This
internal pH reduction has been shown to reduce junctional
conductance. The sensitivity to low pH has been found to be
connexin specific, indicating that the mechanism of action is
related to molecular structure. It is generally accepted that
the CT of connexins participates actively in this process,
since the truncation of the CT in Cx43 [59] abolishes
junctional sensitivity. Nonetheless, not all pH gating
depends on the CT. Truncated Cx45 channels remain
sensitive to pH acidification as do the channels of Cx26, a
natural connexin lacking a CT [60].
The presence of histidine residues in the cytoplasmic
loop is required for gating of Cx43 by pH, although in
Cx46, H95 seems not to be required [61]. Plasmon resonance studies now suggest that specific protein – protein
interactions between the CT region and the CL regions are
required [62]. How direct is this process? Cx46 hemichannels in the excised patch conformation appear to gate with a
high concentration of protons [61] and this has been
hypothesized to be the general mechanism, although for
Cx43 channels some reports suggest the need of an intermediary [63].
For cardiac connexins, Cx43 is most sensitive to low pH,
and Cx45 is the least sensitive. The formation of heteromeric connexons by Cx43 and Cx40 yields channels more
sensitive to acidification [64]. This may have strong consequences in the differential effects of connexins in distinct
regions of the heart, where specific connexins are expressed.
Further comparative studies are required to determine these
differences.
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
pH gating occurs in connexons or hemichannels similarly, as demonstrated for Cx46 and Cx38 where the sensitivity
to pH is maintained in both structures. Cx45-related membrane currents in HeLa cells are sensitive to CO2 perfusion,
although the sensitivity to acidification has not yet been
quantified [15].
283
ylated by TPA, reducing also its permeability to LY, but not
to neurobiotin [27].
Studies of the changes in biophysical properties due to
phosphorylation of Cx40 have been scarce. Activation of
PKA reduces the time Cx40 channels spend in substates,
indicating that the total conductance of the junction probably increases [71].
4.5. Phosphorylation. Homotypic channels
4.6. Heteromultimeric channels
The majority of vertebrate connexins are known to be
posttranslationally modified by phosphorylation. Connexins are targets of numerous protein kinases, including
protein kinase C, mitogen-activated protein kinase, and vSrc tyrosine kinase [65]. Connexin phosphorylation has
been implicated in the regulation of various cellular
processes that include trafficking, assembly, degradation,
gating and selectivity of gap junction channels. In this
section, we will focus mainly on the effects of phosphorylation on the gating and perm-selectivity of cardiac
connexins. Cx43 channels were the first reported to
undergo changes in the distribution of unitary conductance during the activation of C kinases [18]. The serine
residue involved in the shift of unitary conductance
during PKC activation is S368 [66], indicating its direct
involvement in the reduction of total conductance between cells. Rat Cx43 can also react to PKG in a similar
manner but not humanCx43, which does not have a
serine in position 257, showing the relevance of that site
to G kinases [67].
The mechanism by which phosphorylation modifies the
unitary conductance of the channels remains unknown, but
phosphorylation of CT is required. Probably, the alteration
of charge changes the affinity of the main channel, which, in
turn, maintains the channel in a partially open state. A rapid
reduction in junctional communication is also observed
during activation of ERK. This kinase phosphorylates serine
residues 255, 279, and 282. Although there are no single
channel recordings, the rapid change in electrical communication indicates a gating mechanism and possibly a direct
interaction [65]. Further studies would be required to
determine the nature of these changes.
Recent work has established that phosphorylation at
tyrosine Y265 resulted from a cascade of events in which
ERK was a major participant [68]. This work clarified that
the whole CT is required for inhibition of conductance.
Hence, ERK appears to gate Cx43 channels rapidly. The
mechanism associated with this gating requires the presence
of the CT [69], where the presence of CT independently coexpressed with tailless channels recovered inhibition during
phosphorylation [65].
Rat Cx45 responds to PKC activation with the appearance of a new, smaller conductance state, but when mouse
Cx45 is phosphorylated, there is a tendency for the channels
to reside longer in the open state, probably increasing total
conductance [70]. The conductance of the junctions formed
by chicken Cx45 is substantially reduced when phosphor-
The co-expression of Cx45 and Cx43 produces bi-heteromeric channels with reduced sensitivity to activation by
TPA [27]. Homotypic cCx45 and rCx43 have both been
inhibited by as much as 70% after activation with TPA.
When bi-heteromeric channels become phosphorylated during TPA activation, this reduction is less that 50%. Since not
all channels are expected to be identical, the average lower
reduction seems to indicate that some channel combinations
respond less to activation by PKC. The mechanism involved
in this desensitization is unknown and needs to be thoroughly analyzed to determine the interaction between connexins. The effects of phosphorylation on other cardiac
connexin combinations have not yet been studied.
4.7. Quinine and other selective blockers and uncouplers
4.7.1. Quinine
Quinine specifically reduces gap junctional currents in a
concentration-dependent manner for channels formed by
Cx36 and Cx50 with kAs below 100 AM. At 300 AM,
quinine does not substantially block Cx40 or Cx43 but
reduces Cx45 by 50%, apparently through the loop gating
mechanism [72].
4.7.2. Arylaminobenzoates
Arylaminobenzoates block gap junction channels at
concentrations in the 100 AM range. The degree of blocking
by these compounds is not high for all connexins, but can
differentiate between some of them. Although the differences found for arylaminobenzoates are not as striking as
those found for quinine, it is intriguing that both of these
compounds differentially block certain connexin channels
[73].
4.7.3. Tetra-alkylammonium ions
Cx40 is permeable to small tetra-alkylammoniun ions
(tetramethyl- and tetraethyl-), but large ones block the
permeant pore with concentration and voltage dependence.
Apparently, one site of interaction within the pore gates so
quickly that it precludes recording single channels. This
blocking does not interfere with other gating phenomena,
like the Vj gating of Cx40 [74].
4.7.4. Spermine and spermidine
A group of compounds known to block receptors in
neurons is also reported to be capable of selectively block-
284
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
ing Cx40 [75]. Spermine and spermidine block up to 95% of
the current that crosses homotypic junctions. In this case,
spermine reduced the unitary conductance of Cx40 and
significantly reduced the open probability of the channels
in a Vj-dependent manner. Similarly to quinine and other
derivatives, spermine and spermidine are able to discriminate pharmacologically the physiological and pathological
roles of some connexins in tissues that co-express multiple
connexins.
5. Summary
In cardiac tissue, the expression of distinct connexins is
noteworthy in the conduction system of the heart [76]. Some
of the biophysical properties of these connexins increase the
importance of electrical signalling to heart function [77].
This effect may not be seen in the musculature itself, where
an extreme reduction in the number of Cx43 active channels
would be required to jeopardize appropriate communication
between cells [78], but in regions where communication is
less abundant (regions of the SA node, AV node or close to
an infarction site). In these areas, other factors like ischemia,
transmembrane voltage dependence or phosphorylation can
affect overall conductance.
Until now, our knowledge of the biophysical properties
of gap junctional channels has been mainly restricted to
homomeric channels and hemichannels and to a few heteromeric combinations. The temporal and regional co-expression of distinct connexins in a particular region of the heart
suggests formation of heteromultimeric channels. The biophysical properties of these channels could be predicted to
be a mixture of the properties of their components. Nonetheless, important variations have been observed; some
gating properties become augmented or reduced, and single
channel distribution becomes skewed. Phosphorylation, low
intracellular pH, and perm-selectivity also seem to be
modified when these connexin combinations occur.
As we continue to perform controlled experiments and
learn to link subunits to reduce the number of heteromeric
combinations, we will be able to understand more quickly
the way that interactions between these subunits affect the
properties of heteromeric channels. At the molecular level,
interaction between proteins and distinct regions within
these proteins may elucidate some epitopes for interaction
and may help predict how these distinct subunits interact,
not only at the membrane level, but also within their gating
mechanisms.
The demonstration that connexins can form heteromeric
combinations has been performed mostly in exogenous
systems, and few examples have been reported from freshly
isolated cells. Unless we understand how these channels
behave in vivo, in their natural environment, we will be
unable to control their function so that pathological behavior
induced by changes in the expression of these connexins can
be corrected.
References
[1] Bruzzone R, Hormuzdi SG, Barbe MT, Herb A, Monyer H. Pannexins, a family of gap junction proteins expressed in brain. Proc Natl
Acad Sci 2003;100:13644 – 9.
[2] Harris AL. Emerging issues of connexin channels: biophysics fills the
gap. Q Rev Biophys 2001;34:325 – 472.
[3] Weidmann S. The diffusion of radiopotassium across intercalated
discs of mammalian cardiac muscle. J Physiol (Lond) 1966;187:
323 – 42.
[4] Elenes S, Rubart M, Moreno AP. Junctional communication between
isolated pairs of canine atrial cells is mediated by homogeneous and
heterogeneous gap junction channels. J Cardiovasc Electrophys
1999;10:990 – 1004.
[5] Yamada KA, Rogers JG, Sundset R, Steinberg TH, Saffitz JE. Upregulation of Cx45 in heart failure. J Cardiovasc Electrophys 2003;
14:1205 – 12.
[6] Gros DB, Jongsma HJ. Connexins in mammalian heart function.
BioEssays 1996;18:719 – 30.
[7] Delorme B, Dahl E, Jarry-Guichard T, et al. Expression pattern of
connexin gene products at the early developmental stages of the
mouse cardiovascular system. Circ Res 1997;81:423 – 37.
[8] Takens Kwak BR, Jongsma HJ. Cardiac gap junctions: three distict
single channel conductances and their modulation by phosphorylating
treatments. Pflügers Arch 1992;422:198 – 200.
[9] Veenstra RD, Wang HZ, Beblo DA, et al. Selectivity of connexinspecific gap junctions does not correlate with channel conductance.
Circ Res 1995;77:1156 – 65.
[10] Banach K, Weingart R. Voltage gating of Cx43 gap junction channels
involves fast and slow current transitions. Pflügers Arch 2000;439:
248 – 50.
[11] Locke D, Wang L, Zhang Y, Bevans CG, Harris AL. Carbohydrate
modifiers of connexin channel permeability. Biophys J 2003;84:586a.
[12] Moreno AP, Chanson M, Anumonwo J, et al. Role of the carboxyl
terminal of connexin43 in transjunctional fast voltage gating. Circ Res
2002;90:450 – 7.
[13] Anumonwo JM, Taffet S, Gu H, et al. The carboxyl terminal domain
regulates the unitary conductance and voltage dependence of connexin40 gap junction channels. Circ Res 2001;88:666 – 73.
[14] Moreno AP, Laing JG, Beyer EC, Spray DC. Properties of gap
junction channels formed of connexin 45 endogenously expressed
in human hepatoma (SKHep1) cells. Am J Physiol 1995;268:
C356 – 65.
[15] Valiunas V. Biophysical properties of connexin-45 gap junction hemichannels studied in vertebrate cells. J Gen Physiol 2002;119:147 – 64.
[16] Valiunas V, Weingart R, Brink PR. Formation of heterotypic gap
junction channels by connexins 40 and 43. Circ Res 2000;86:E42 – 9.
[17] Burt JM, Spray DC. Single-channel events and gating behavior of the
cardiac gap junction channel. Proc Natl Acad Sci U S A 1988;85:
3431 – 4.
[18] Moreno AP, Fishman GI, Spray DC. Phosphorylation shifts unitary
conductance and modifies voltage dependent kinetics of human connexin43 gap junction channels. Biophys J 1992;62:51 – 3.
[19] Veenstra RD. Size and selectivity of gap junction channels formed
from different connexins. J Bioenerg Biomembr 1996;28:327 – 37.
[20] Beblo DA, Veenstra RD. Monovalent cation permeation through the
connexin40 gap junction channel. Cs, Rb, K, Na, Li, TEA, TMA,
TBA, and effects of anions Br, Cl, F, acetate, aspartate, glutamate, and
NO3. J Gen Physiol 1997;109:509 – 22.
[21] Wang HZ, Veenstra RD. Monovalent ion selectivity sequences of
the rat connexin43 gap junction channel. J Gen Physiol 1997;109:
491 – 507.
[22] Goldberg GS, Lampe PD, Nicholson BJ. Selective transfer of endogenous metabolites through gap junctions composed of different connexins. Nat Cell Biol 1999;1:457 – 9.
[23] Bukauskas FF, Bukauskiene A, Verselis VK. Conductance and per-
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
meability of the residual state of connexin43 gap junction channels.
J Gen Physiol 2002;119:171 – 85.
Kwak BR, van Veen TA, Analbers LJ, Jongsma HJ. TPA increases
conductance but decreases permeability in neonatal rat cardiomyocyte
gap junction channels. Exp Cell Res 1995;220:456 – 63.
Elenes S, Martinez AD, Delmar M, Beyer EC, Moreno AP. Heterotypic docking of Cx43 and Cx45 connexons blocks fast voltage gating
of Cx43. Biophys J 2001;81:1406 – 18.
Hayrapetyan V, Moreno AP. Gating of heterotypic connexin40 and
connexin45 gap junction channels. Biophys J 2003;84:526a.
Martinez AD, Hayrapetyan V, Moreno AP, Beyer EC. Connexin43
and connexin45 form heteromeric gap junction channels in which
individual components determine permeability and regulation. Circ
Res 2002;90:1100 – 7.
Valiunas V, Gemel J, Brink PR, Beyer EC. Gap junction channels
formed by coexpressed connexin40 and connexin43. Am J Physiol
Heart Circ Physiol 2001;281:H1675 – 89.
Valiunas V, Weingart R, Brink PR. Formation of heterotypic gap
junction channels by connexins 40 and 43. Circ Res 2000;86:
E42 – 9.
Zhong G, Hayrapetyan V, Moreno AP. The formation of mono-heteromeric Cx43 – Cx45/Cx45 gap junction channels uncovers gating
and selectivity properties of their channels. Biophys J 2003;82:633a.
Valiunas V, Beyer EC, Brink PR. Cardiac gap junction channels show
quantitative differences in selectivity. Circ Res 2002;91:104 – 11.
Bevans CG, Harris AL. Direct high affinity modulation of connexin channel activity by cyclic nucleotides. J Biol Chem 1999;274:
3720 – 5.
Paul DL, Ebihara L, Takemoto LJ, Swenson KI, Goodenough DA.
Connexin46, a novel lens gap junction protein, induces voltage-gated currents in nonjunctional plasma membrane of Xenopus oocytes.
J Cell Biol 1991;115:1077 – 89.
DeVries SH, Schwartz EA. Hemi-gap-junction channels in solitary
horizontal cells of the catfish retina. J Physiol (Lond) 1992;445:
201 – 30.
Contreras JE, Sanchez HA, Eugenin EA, et al. Metabolic inhibition
induces opening of unapposed connexin 43 gap junction hemichannels and reduces gap junctional communication in cortical astrocytes
in culture. Proc Natl Acad Sci U S A 2002;99:495 – 500.
Kondo RP, Wang SY, John SA, Weiss JN, Goldhaber JI. Metabolic
inhibition activates a non-selective current through connexin hemichannels in isolated ventricular myocytes. J Mol Cell Cardiol
2000;32:1859 – 72.
Contreras JE, Saez JC, Bukauskas FF, Bennett MV. Gating and
regulation of connexin 43 (Cx43) hemichannels. Proc Natl Acad
Sci U S A 2003;100:11388 – 93.
Liu TF, Paulson AF, Li HY, Atkinson MM, Johnson RG. Inhibitory
effects of 12-O-tetradecanoylphorbol-13-acetate on dye leakage from
single Novikoff cells and on dye transfer between reaggregated cell
pairs. Methods Find Exp Clin Pharmacol 1997;19:573 – 7.
Moreno AP, Rook MB, Fishman GI, Spray DC. Gap junction channels: distinct voltage-sensitive and -insensitive conductance states.
Biophys J 1994;67:113 – 9.
Revilla A, Castro C, Barrio LC. Molecular dissection of transjunctional voltage dependence in the connexin-32 and connexin-43 junctions. Biophys J 1999;77:1374 – 83.
Verselis VK, Ginter CS, Bargiello TA. Opposite voltage gating polarities of two closely related connexins. Nature 1994;368:348 – 51.
Oh S, Rubin JB, Bennett MV, Verselis VK, Bargiello TA. Molecular
determinants of electrical rectification of single channel conductance
in gap junctions formed by connexins 26 and 32. J Gen Physiol
1999;114:339 – 64.
Bukauskas FF, Angele AB, Verselis VK, Bennett MV. Coupling
asymmetry of heterotypic connexin 45/connexin 43-EGFP gap junctions: properties of fast and slow gating mechanisms. Proc Natl Acad
Sci U S A 2002;99:7113 – 8.
White TW, Bruzzone R. Multiple connexin proteins in single inter-
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
285
cellular channels: connexin compatibility and functional consequences. J Bioenerg Biomembr 1996;28:339 – 50.
Cottrell GT, Lin R, Warn-Cramer BJ, Lau AF, Burt JM. Mechanism of
v-Src- and mitogen-activated protein kinase-induced reduction of gap
junction communication. Am J Physiol, Cell Physiol 2003;284:
C511 – 20.
Barrio LC, Capel J, Jarillo JA, Castro C, Revilla A. Species-specific
voltage-gating properties of connexin-45 junctions expressed in Xenopus oocytes. Biophys J 1997;73:757 – 69.
Bukauskas FF, Elfgang C, Willecke K, Weingart R. Biophysical properties of gap junction channels formed by mouse connexin40 in induced pairs of transfected human HeLa cells. Biophys J 1995;68:
2289 – 98.
Elenes S, Moreno AP. Murine Cx45. Voltage dependence characterization at physiological conditions. Mol Biol Cell 1998;9:95a.
Bennett MV, Verselis VK. Biophysics of gap junctions. Semin Cell
Biol 1992;3:29 – 47.
Trexler EB, Verselis VK. The study of connexin hemichannels
(connexons) in Xenopus oocytes. Methods Mol Biol 2001;154:
341 – 55.
Revilla A, Bennett MV, Barrio LC. Molecular determinants of membrane potential dependence in vertebrate gap junction channels. Proc
Natl Acad Sci U S A 2000;97:14760 – 5.
Johnston MF, Simon SA, Ramon F. Interaction of anaesthetics with
electrical synapses. Nature 1980;286:498 – 500.
Bastiaanse EM, Jongsma HJ, van der Laarse A, Takens-Kwak BR.
Heptanol-induced decrease in cardiac gap junctional conductance is
mediated by a decrease in the fluidity of membranous cholesterol-rich
domains. J Membr Biol 1993;136:135 – 45.
Burt JM, Spray DC. Volatile anesthetics block intercellular communication between neonatal rat myocardial cells. Circ Res 1989;65:
829 – 37.
Rozental R, Giaume C, Spray DC. Gap junctions in the nervous
system. [Review] [22 refs] Brain Res Brain Res Rev 2000;32:11 – 5.
He DS, Burt JM. Mechanism and selectivity of the effects of halothane on gap junction channel function. Circ Res 2000;86:E104 – 9.
Davidson JS, Baumgarten IM. Glycyrrhetinic acid derivatives: a novel class of inhibitors of gap-junctional intercellular communication.
Structure – activity relationships. J Pharmacol Exp Ther 1988;246:
1104 – 7.
Goldberg GS, Moreno AP, Bechberger JF, et al. Evidence that disruption of connexon particle arrangements in gap junction plaques is
associated with inhibition of gap junctional communication by a glycyrrhetinic acid derivative. Exp Cell Res 1996;222:48 – 53.
Morley GE, Taffet SM, Delmar M. Intramolecular interactions mediate pH regulation of connexin43 channels. Biophys J 1996;70:
1294 – 302.
Stergiopoulos K, Alvarado JL, Mastroianni M, et al. Hetero-domain
interactions as a mechanism for the regulation of connexin channels.
Circ Res 1999;84:1144 – 55.
Trexler EB, Bukauskas FF, Bennett MV, Bargiello TA, Verselis VK. VK.
VK. Rapid and direct effects of pH on connexins revealed by the
connexin46 hemichannel preparation. J Gen Physiol 1999;113:
721 – 42.
Duffy HS, Sorgen PL, Girvin ME, et al. pH-dependent intramolecular
binding and structure involving Cx43 cytoplasmic domains. J Biol
Chem 2002;277:36706 – 14.
Peracchia C, Wang X, Li L, Peracchia LL. Inhibition of calmodulin
expression prevents low-pH-induced gap junction uncoupling in Xenopus oocytes. Pflügers Arch 1996;431:379 – 87.
Gu H, Ek-Vitorin JF, Taffet SM, Delmar M. Coexpression of connexins 40 and 43 enhances the pH sensitivity of gap junctions: a
model for synergistic interactions among connexins. Circ Res
2000;86:E98 – 103.
Lampe PD, Lau AF. Regulation of gap junctions by phosphorylation
of connexins. Arch Biochem Biophys 2000;384:205 – 15.
Lampe PD, TenBroek EM, Burt JM, Kurata WE, Johnson RG.
286
[67]
[68]
[69]
[70]
[71]
A.P. Moreno / Cardiovascular Research 62 (2004) 276–286
Phosphorylation of connexin43 on serine368 by protein kinase c
regulates gap junctional communication. J Cell Biol 2000;149:
1503 – 12.
Kwak BR, Saez JC, Wilders R, et al. Effects of cGMP-dependent
phosphorylation on rat and human connexin43 gap junction channels.
Pflügers Arch 1995;430:770 – 8.
Zhou L, Kasperek E, Nicholson B. Dissection of the molecular
basis of pp60v-src induced gating of connexin 43 gap junction
channels. J Cell Biol 1999;144:1033 – 42.
Homma N, Alvarado JL, Coombs W, et al. A particle-receptor model
for the insulin-induced closure of connexin43 channels. Circ Res
1998;83:27 – 32.
vanVeen TAB, vanRijen HVM, Jongsma HJ. Electrical conductance
of mouse connexin45 gap juntion channels is modulated by phosphorylation. Cardiovasc Res 2000;46:496 – 510.
van Rijen HV, van Veen TA, Hermans MM, Jongsma HJ. Human
connexin40 gap junction channels are modulated by cAMP. Cardiovasc Res 2000;45:941 – 51.
[72] Srinivas M, Hopperstad MG, Spray DC. Quinine blocks specific gap
junction channel subtypes. Proc Natl Acad Sci U S A 2001;98:
10942 – 7.
[73] Srinivas M, Spray DC. Closure of gap junction channels by arylaminobenzoates. Mol Pharmacol 2003;63:1389 – 97.
[74] Musa H, Gough JD, Lees WJ, Veenstra RD. Ionic blockade of the rat
connexin40 gap junction channel by large tetraalkylammonium ions.
Biophys J 2001;81:3253 – 74.
[75] Musa H, Veenstra RD. Voltage-dependent blockade of connexin40
gap junctions by spermine. Biophys J 2003;84:205 – 19.
[76] Verheule S, van Kempen MJ, Postma S, Rook MB, Jongsma HJ. Gap
junctions in the rabbit sinoatrial node. Am J Physiol Heart Circ Physiol 2001;280:H2103 – 15.
[77] van Rijen HV, van Veen TA, van Kempen MJ, et al. Impaired conduction in the bundle branches of mouse hearts lacking the gap junction protein connexin40. Circulation 2001;103:1591 – 8.
[78] Jongsma HJ, Wilders R. Gap junctions in cardiovascular disease. Circ
Res 2000;86:1193 – 7.