Retigabine reduces the excitability of unmyelinated peripheral human axons.

Abstract

Enhancement of membrane K(+) conductance may reduce the abnormal excitability of primary afferent nociceptive neurons in neuropathic pain. It has been shown that retigabine, a novel anticonvulsant, activates Kv7 (KCNQ/M) channels in the axonal/nodal membrane of peripheral myelinated axons. In this study, we have tested the effects of retigabine on excitability parameters of C-type nerve fibers in isolated fascicles of human sural nerve. Application of retigabine (3-10 microM) produced an increase in membrane threshold. This effect was pronounced in depolarized axons and small in hyperpolarized axons. This finding indicates that retigabine produces a membrane hyperpolarization which is limited by the K(+) equilibrium potential. The retigabine-induced reduction in excitability was accompanied by modifications of the post-spike recovery cycle. Most notable is the development of a late subexcitability at 250-400 ms following a short burst of action potentials. All effects of retigabine were blocked in the presence of XE991 (10 microM). The data show that Kv7 channels are present on axons of unmyelinated, including nociceptive, peripheral human nerve fibers. It is likely that activation of these channels by retigabine may reduce the ectopic generation of action potentials in neuropathic pain.

Full text

Page 1

Retigabine reduces the excitability of unmyelinated peripheral human axons
P.M. Langa,b, J. Fleckensteinb, G.M. Passmorec, D.A. Brownc, P. Grafea,*
aDepartment of Physiology, University of Munich, Pettenkoferstrasse 12, D-80336 Munich, Germany
bDepartment of Anesthesiology, University of Munich, 81377 Munich, Germany
cDepartment of Pharmacology, University College London, London WC1E 6BT, United Kingdom
a r t i c l e i n f o
Article history:
Received 15 February 2008
Received in revised form 10 April 2008
Accepted 11 April 2008
Keywords:
Neuronal potassium channel activator
Retigabine
KCNQ
M channel
Kv7
Axonal excitability
Human C-fiber
Electrophysiology
Threshold tracking
Neuropathic pain
Nociception
a b s t r a c t
Enhancement of membrane Kþconductance may reduce the abnormal excitability of primary afferent
nociceptive neurons in neuropathic pain. It has been shown that retigabine, a novel anticonvulsant,
activates Kv7 (KCNQ/M) channels in the axonal/nodal membrane of peripheral myelinated axons. In this
study, we have tested the effects of retigabine on excitability parameters of C-type nerve fibers in isolated
fascicles of human sural nerve. Application of retigabine (3–10 mM) produced an increase in membrane
threshold. This effect was pronounced in depolarized axons and small in hyperpolarized axons. This
finding indicates that retigabine produces a membrane hyperpolarization which is limited by the Kþ
equilibrium potential. The retigabine-induced reduction in excitability was accompanied by modifica-
tions of the post-spike recovery cycle. Most notable is the development of a late subexcitability at
250–400 ms following a short burst of action potentials. All effects of retigabine were blocked in the
presence of XE991 (10 mM). The data show that Kv7 channels are present on axons of unmyelinated,
including nociceptive, peripheral human nerve fibers. It is likely that activation of these channels by
retigabine may reduce the ectopic generation of action potentials in neuropathic pain.
? 2008 Elsevier Ltd. All rights reserved.
1. Introduction
Spontaneous ectopic generation of action potentials in periph-
eral nociceptive neurons is considered to be a key mechanism
underlying the development of neuropathic pain (for review see
Devor, 2006) although this concept may be adequate for specific
forms of nerve lesions only (Eschenfelder et al., 2000). Spontaneous
electrical activity has been observed not only in injured but also in
intact peripheral nociceptive neurons (Ali et al., 1999; Liu et al.,
2000; Michaelis et al., 2000; Wu et al., 2001, 2002; Campbell and
Meyer, 2006; Djouhri et al., 2006). Furthermore, there is a clear
correlation between spontaneous firing in peripheral nociceptive
neurons and the sensation of pain in humans (e.g. Ochoa et al.,
2005) and behavioural signs of pain in animal experiments (e.g.
Djouhri et al., 2006). Based on these findings, suppression of ec-
topic impulse generation is a rational approach for the treatment of
neuropathic pain. In fact, inhibition of action potentials by blockade
of voltage-dependent sodium channels has been successful in some
clinical neuropathic pain conditions (for review see Priestley and
Hunter, 2006).
Potassium channels, too, contribute to the excitability of noci-
ceptive neurons. The importance of Kþconductances for the de-
velopment of spontaneous and ectopic action potential generation
has been documented by experimental findings: (a) blockers of
voltage-sensitive Kþchannels such as tetraethylammonium (TEA)
and 4-aminopyridine (4-AP) strongly excited experimental neuro-
mas that had spontaneous discharge (Devor, 1983) and (b) 4-AP
enhanced spontaneous firing recorded in sensory dorsal root gan-
glion cells (Amir et al., 2005). It is possible, therefore, that the op-
posite effect, i.e. pharmacological activation of the neuronal Kþ
conductance, should be able topreventthe spontaneous generation
of action potentials in nociceptive neurons (Ocana et al., 2004).
One target for pharmacological activation of neuronal Kþcon-
ductance are Kv7 (KCNQ/M) channels which contribute to the ex-
citability of the cell bodies of sensory, including nociceptive,
neurons (Passmore et al., 2003). The International Union of Phar-
macology has assigned the name Kv7 for KCNQ channel proteins
and the effective M current (Gutman et al., 2005). Therefore, Kv7 is
used exclusively in the following text. Kv7-mediated Kþcurrents
also modifythe excitabilityof central (Passmore et al., 2003;Rivera-
Arconada and Lopez-Garcia, 2006) and peripheral (Passmore and
* Corresponding author. Tel.: þ49 89 2180 75221; fax: þ49 89 2180 75216.
E-mail address: p.grafe@lrz.uni-muenchen.de (P. Grafe).
Contents lists available at ScienceDirect
Neuropharmacology
journal homepage: www.elsevier.com/locate/neuropharm
0028-3908/$ – see front matter ? 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.neuropharm.2008.04.006
Neuropharmacology 54 (2008) 1271–1278

Page 2

Brown, 2007; Wladyka et al., 2008) terminals of primary afferent
sensory neurons and have been functionally characterised in
peripheral myelinated axons (Devaux et al., 2004; Schwarz et al.,
2006). In the last few years, several activators and blockers of Kv7
channels have been developed (Munro and Dalby-Brown, 2007;
Xiong et al., 2008). A potent activator is retigabine, a compound
with anti-hyperalgesic activity in animal models of neuropathic
pain (Blackburn-Munro and Jensen, 2003; Dost et al., 2004;
Nielsen et al., 2004; Hirano et al., 2007). Also, it has been observed
in a recent study that retigabine blocks ectopic discharges in
axotomized sensory fibers (Roza and Lopez-Garcia, in press). An
important question is whether retigabine is able to modify the
excitability of peripheral human nociceptive axons, a likely com-
ponent in the development of spontaneous ectopic impulse gen-
eration and neuropathic pain. The excitability of unmyelinated
human C-type nerve fibers can be studied by ‘‘threshold tracking’’
(Bostock et al., 1998) of action potential generation in isolated
segments of human nerve. In this study, we have used this method
to characterise the components of axonal excitability that are
modified by retigabine.
The data presented here have been published previously in
abstract form (Lang et al., 2008).
2. Methods
2.1. Preparations
The experiments on segments of human sural nerve were carried out on 47
isolated fascicles from 10 patients. Approval for this procedure was obtained from
the Ethics Committee of the University of Munich, and the patients gave written
informed consent. The patients (male n¼ 7, female n ¼3) had a median age of
71.5 years (48–89) when they underwent the procedure. Five of them underwent
biopsies of the sural nerve for diagnosis of neurological diseases (like poly-
neuropathy) and five patients needed to undergo amputation due to peripheral
vascular disease. The isolated segments of human sural nerves were cut into
shorter segments of 15–25 mm in length and single nerve fascicles were dissected
free under a microscope. The isolated fascicles were held at each end by suction
electrodes in an organ bath. One suction electrode was used to elicit action po-
tentials, while the other was used as a recording electrode. The distance between
stimulating and recording electrodes was approximately 3 mm. The organ bath
(volume 1 ml) was continuously superfused with solution at a flow rate of 6–8 ml/
min at a temperature of 32?C. The perfusion solution contained (in mM) NaCl
118, KCl 3.0, CaCl21.5, MgCl21.0, D-glucose 5.0, NaHCO325, NaH2PO41.2, and was
bubbled with 95% O2/5% CO2(pH 7.4).
2.2. Electrophysiological methods used for determination of axonal excitability
Axonal excitability parameters were recorded using the QTRAC program (?
Institute of Neurology, London, UK). QTRAC is a flexible, stimulus–response data-
acquisition program, originally written for studies of human nerves in vivo (Bostock
et al., 1998) but also suitable for electrophysiological recordings from isolated pe-
ripheral nerves. In this study, QTRAC was used to record compound actionpotentials
(CAPs) from peripheral C-fibers, to generate stimuli, and to display the results. The
isolated fascicles from segments of human sural nerves were stimulated with
a constant current stimulator (A395, WPI, Sarasota, FL) with a maximal output of
1 mA. The stimulator was controlled by a computer via a data-acquisition board.
Methods used for separating A and C-fiber compound action potentials have been
described previously (Lang et al., 2003). Nerve excitability was tested with 1 ms
current pulses at different frequencies. For threshold tracking experiments, the
stimulus strength was automatically adjusted to maintain the C-fiber CAP at a con-
stant amplitude (40% of the maximum, defined as ‘‘threshold’’). QTRAC was also
used for determination of axonal excitability in the post-spike recovery period. In
a typical stimulation sequence used for this parameter, successive sweeps delivered
every 0.3–1 s contained the following stimulus conditions: (1) a supramaximal
stimulus (to monitor peak response amplitude), (2) a test stimulus alone adjusted to
maintain the C-fiber CAP peak height at 40% of the maximum, and (3) a supra-
maximal conditioning stimulus followed by the test stimulus delivered at variable
delay.
2.3. Chemicals
XE991 (10,10-bis (4-pyridinylmethyl)-9(10H)-anthracenone dihydrochloride)
was purchased from Biotrend Chemikalien GmbH (Cologne, Germany). Retigabine
(N-(2-amino-4-(4-fluorobenzyl-amino)-phenyl)carbamic acid ethyl ester) was
generously provided by NeuroSearch A/S (DK-2750 Ballerup, Denmark).
3. Results
3.1. Retigabine reduces the excitability of unmyelinated axons
The excitability of unmyelinated C-type nerve fibers in isolated
segments of human sural nerve was tested by ‘‘threshold tracking’’.
This technique determines the current necessary to maintain the
peak height of the C-fiber compound action potential (C-CAP) at
40% of the maximum. Addition of retigabine (10 mM) resulted in an
increase of the current strength, i.e. a reduction in axonal excit-
ability. A representative recording is illustrated in Fig. 1. The effect
of retigabine had a fast onset, reached a steady state level after
about 5 min, and could be washed out within about 10 min. Reti-
gabine did not have a clear effect on the peak height of the C-CAP.
However, the increase in threshold current was accompanied by
Fig.1. Retigabine reduces excitability of unmyelinated axons in an isolated segment of human sural nerve. The electrophysiological parameters tested are the maximal peak height
of the compound C-fiber action potential (C-CAP), the current necessary to evoke a C-CAP with 40% of the maximal amplitude, and the latency to 50% of the maximal peak height
(examples illustrated in (B)). The time course of alterations in these parameters produced by retigabine is shown in (A). Note that retigabine produces a reduction in excitability
(increase in ‘‘threshold’’ current), and a slowing in the generation and/or conduction of the action potential (continuous stimulation at 0.33 Hz).
P.M. Lang et al. / Neuropharmacology 54 (2008) 1271–12781272

Page 3

a shift in latency due to a slowing in either or both the generation
and conduction of the action potential.
Quantitatively, the effect of retigabine (10 mM) on threshold
current and latency was rather variable. Therefore, we tested
whether the effect of retigabine depends on axonal membrane
potential. To this end, the effect of retigabine on the excitability of
C-type nerve fibers was compared with a post-spike excitability
parameter determined before the application of retigabine (16
different fascicles). Several observations have been described
which indicate that the post-spike recovery cycle of membrane
threshold can be used as an indirect measure of membrane po-
tential in peripheral myelinated (Kiernan and Bostock, 2000) and
unmyelinated (Weidner et al., 2002; Bostock et al., 2003; Moalem-
Taylor et al., 2007) axons. In this study, post-spike excitability was
determined 40 ms after a supramaximal conditioning stimulus
(Fig. 2B). Post-spike C-fiber excitability in the various nerve prep-
arations differed between ?23% (superexcitability) and 5% (sub-
excitability). We compared this parameter with the percentage
increase in threshold current during application of retigabine
(10 mM). The correlation is illustrated in Fig. 2A. The graph reveals
that retigabine has the strongest inhibitory effect on membrane
excitability in axons with the most prominent post-spike sub-
excitability under resting conditions. In contrast, we find little to no
increase in the threshold of axons with prominent post-spike
superexcitability. This finding indicates a potential-dependent
effect of retigabine. Whilst the excitability of depolarized axons is
strongly reduced by retigabine, membrane hyperpolarization
clearly limits its effect. This observation is in accordance with the
most likely effect of retigabine which is an activation of Kv7
channels.
The effects of retigabine on membrane threshold were accom-
panied byan increase in the latency to 50% of the peak height of the
C-CAP (Fig.1). Quantitatively, an increase in latency of between 1.6
and 10% was observed. These changes in latency, too, were corre-
lated with the post-spike excitability before the application of
retigabine (Fig. 2C). However, the correlation was weaker
(r2¼0.574) than that for retigabine-induced changes in threshold
(r2¼0.857).
3.2. Effects of retigabine are blocked by XE991
XE991 has been characterized as a specific blocker of Kv7.2 and
Kv7.3 potassium channel subunits when used at low micromolar
concentrations (Wang et al., 1998). Therefore, we tested whether
XE991 could block the effects of retigabine on C-fiber excitability.
Representative examples of these experiments are illustrated in
Fig. 3. First, we tested the effect of repeated applications of reti-
gabine (Fig. 3A). We did not observe the development of tachy-
phylaxis. Second, after one control application, retigabine (10 mM)
was applied in the presence of XE991 (10 mM). XE991 strongly
Fig. 2. The decrease in membrane excitability produced by retigabine correlates with post-spike membrane threshold determined before the application of retigabine.
Post-spike excitability was determined using a test and a conditioning stimulus with a 40 ms interspike interval (protocol shown in (B)). The magnitude of post-spike excitability is
given on the x-axis in plot (A). Some preparations had a post-spike superexcitability (reduction in threshold up to 25%); in others, a post-spike subexcitability was observed
(increase in threshold up to 5%). The effect of retigabine on membrane threshold in these preparations is given on the y-axis in plot (A). Note the clear correlation between
post-spike excitability before the application of retigabine and the effect of retigabine on membrane ‘‘threshold’’. Likewise, the effect of retigabine on latency was correlated with
post-spike excitability (C, D).
P.M. Lang et al. / Neuropharmacology 54 (2008) 1271–1278 1273

Page 4

reduced the effect of retigabine on membrane threshold. Quanti-
tatively, a reduction of the retigabine-induced decrease in
membrane excitability by 91.3?10.5% (mean?S.D.; n¼4) was
observed.
3.3. Retigabine modifies the recovery cycle of excitability
In myelinated peripheral axons, Kv7 channels produce a slowKþ
current and reduce membrane excitability in the recovery period
following a short train of action potentials (Schwarz et al., 2006). In
this study, the effects of retigabine and XE991 were tested on the
recovery cycle of unmyelinated nerve fibers. A summary of the data
and the threshold tracking protocol used in these experiments is
illustrated in Fig. 4. Membrane excitability was tested at 40, 160,
250, and 400 ms after a burst of two action potentials (intra-burst
interval 25 ms; stimulus interval 2 s). The data used are mean
values from a selected group of nerve fascicles which showed post-
spike superexcitability between ?15 and ?24% at 40 ms in the
recovery cycle. A prominent, long-lasting membrane super-
excitability was observed under control conditions. Retigabine
reduced this post-spike superexcitability and produced a decrease
in the time constant of the recovery cycle. These effects were
dependent on the concentration used. The time constant was
reduced from 207 to 132 ms (3 mM) and from 226 to 71.5 ms
(10 mM; see Fig. 4B, C), respectively. In contrast, we did not observe
an effect of XE991 (10 mM) on the recovery cycle (Fig. 4D).
3.4. Retigabine enhances late subexcitability
A more detailed analysis of the effect of retigabine on the
recovery cycle of excitability reveals that this compound reverses
the post-spike superexcitability in the late phase of the post-spike
recovery period (250–400 ms) into membrane subexcitability (e.g.
Fig. 4C). This change in excitability would be consistent with acti-
vation of a slow Kþcurrent activated by the burst of action po-
tentials. The time course of this effect was studied in further
experiments. Thereby, three conditioning action potentials were
used because activation of Kv7 channels may be more easily
revealed after a longer lasting membrane depolarization (Baker
et al., 1987; Bostock et al., 2003; Schwarz et al., 2006). A repre-
sentative example is illustrated in Fig. 5. Membrane threshold was
tested 250 ms following a burst of three action potentials (intra-
burst interval 25 ms, stimulus frequency 0.25 Hz; see Fig. 5A). Un-
der control conditions, no difference in membrane threshold
Fig. 3. The effect of retigabine on membrane threshold is blocked by XE991. The excitability parameters tested are identical to those described in Fig.1. (A) This illustrates the effect
of retigabine produced by three consecutive applications. There is no indication of tachyphylaxis. (B) Recording from a different isolated human nerve segment. Following a control
application, retigabine (10 mM) was tested in the presence of XE991 (10 mM). XE991 strongly reduced the effect of retigabine.
P.M. Lang et al. / Neuropharmacology 54 (2008) 1271–1278 1274

Page 5

Fig. 4. Retigabine modifies the recovery cycle of unmyelinated C-type nerve fibers. (A) This illustrates the experimental protocol. The recovery of membrane excitability after a pair
of action potentials (intra-burst interval 25 ms) was tested by a comparison of the ‘‘threshold current’’ necessary to elicit a compound action potential with 40% of the maximum
(parameter 1) either alone (parameter 2) or after the conditioning impulses (parameter 3; stimulus interval 2 s, mean stimulation frequency 0.66 Hz). (B, C) Under control con-
ditions, a period of superexcitability lasting for at least 400 ms can be recorded. This increase in membrane excitability is reduced in the presence of retigabine. The effect of
retigabine depends on the concentration used (3 vs. 10 mM). Note also that a late subexcitability is produced by retigabine. (D) Application of XE991 (10 mM) does not have a clear
effect on the period of superexcitability (mean ?s.e.m.; *p <0.05, **p <0.01, ***p<0.001).
Fig. 5. Retigabine produces late subexcitability in the axonal membrane of unmyelinated peripheral nerve fibers. (A) Membrane excitability was tested by three different stimulus
parameters similar to the protocol used in Fig. 4 (stimulus interval: 4 s). The train of conditioning impulses consisted of three action potentials (intra-burst interval: 25 ms) and was
elicited 250 ms before the test pulse. (B) Continuous recording of the ‘‘threshold’’ current and the peak height of the compound C-fiber action potential determined by using the
different stimulus parameters. Note that retigabine (3 mM) produces late subexcitability 250 ms after a train of action potentials (stimulus parameter 3) whereas the ‘‘resting’’
membrane excitability (stimulus parameter 2) is only slightly reduced. The effect of retigabine is completely blocked by addition of XE991 (10 mM) to the bathing solution.
P.M. Lang et al. / Neuropharmacology 54 (2008) 1271–1278 1275

Page 6

between the post-spike excitability and the ‘‘resting’’ excitability
was observed at this time point (Fig. 5B). However, application of
retigabine (3 mM) produced an increase in membrane threshold.
Addition of XE991 to the retigabine-containing solution completely
blocked this effect.
For the quantitative analysis, membrane excitability tested
250 ms after a burst of three action potentials was compared in the
control solution and in the presence of retigabine. There was only
a small membrane superexcitability in the normal bathing solution
(?0.13?0.43%; n¼6; mean?s.e.m). Application of retigabine
(3 mM) produced a reduction in axonal excitability (subexcitability)
of 4.04 ?0.40% (n¼6; mean?s.e.m.). A more pronounced de-
velopment of membrane subexcitability was observed in the
presence of 10 mM retigabine (4.69?0.29%; n¼7; mean?s.e.m.).
4. Discussion
The observations made in this study indicate that retigabine
produces changes in several electrophysiological parameters of
unmyelinated peripheral human nerve fibers. Previously, effects of
retigabine have been described at the nodal membrane of mye-
linated axons in peripheral nerve (Devaux et al., 2004; Schwarz
et al., 2006), and also at peripheral (Passmore and Brown, 2007;
Wladyka et al., 2008), and at central terminals (Passmore et al.,
2003; Rivera-Arconada and Lopez-Garcia, 2006) of primaryafferent
nerve fibers. These findings, together with the data obtained in the
cell body of sensory neurons (Passmore et al., 2003), indicate that
the functional binding site for retigabine is present on all ana-
tomical compartments of a peripheral sensory neuron. In addition,
after submission of this manuscript, a strong hyperpolarizing effect
of retigabine has been described for axotomized peripheral nerve
endings of afferent sensory neurons in mice (Roza and Lopez-
Garcia, in press). The effects of retigabine on C-type nerve fibers in
isolated human nerve segments were most pronounced in axons
with a depolarized membrane potential. Also, retigabine produced
a late membrane subexcitability in accordance with activation of
a slow Kþconductance, and its effect was blocked by XE991. These
three characteristics are typical for Kv7 channels which are nor-
mally activated by membrane depolarization. The increase in con-
ductance starts at about ?60 mV and the half-maximal voltage is
reached at ?35 mV (Jentsch, 2000; Delmas and Brown, 2005).
Therefore, the findings of this study indicate that retigabine acti-
vates such channels in the membrane of unmyelinated peripheral
human axons.
Retigabine produced a reduction in membrane excitability of
unmyelinated axons (see Fig. 1). This effect was more pronounced
in C-fiber bundles with post-spike subexcitability as compared to
those C-fiber bundleswith
excitability (see Fig. 2). The main source of variability in the
recovery cycle seems to be the nerve tissue itself (e.g. length of
fascicle, disease condition, and/or age of patient). We do not get
enough material to study the importance of these factors system-
atically. Since the recovery cycle in unmyelinated nerve fibers is
correlated with the axonal membrane potential (Weidner et al.,
2002; Bostock et al., 2003; Moalem-Taylor et al., 2007), this implies
that retigabine produces stronger effects on ‘‘resting’’ excitability at
depolarized than at hyperpolarized membrane potentials. This ac-
cords with the known action of retigabine, namely, to activate Kv7
channels and thereby to hyperpolarize the unmyelinated axons.
Several previous studies have shown that retigabine produces
membrane hyperpolarization by a shift of Kþchannel opening to
more negative potentials (Rundfeldt, 1997; Rundfeldt and Netzer,
2000; Wickenden et al., 2000; Tatulian et al., 2001; Passmore et al.,
2003; Schwarz et al., 2006; Vervaeke et al., 2006).
An increase in Kþconductance predicts that the most negative
membrane potential reachable during application of retigabine is
pronounced post-spike super-
the Kþequilibrium potential (EK). In fact, in a pioneering study,
a reversal potential for the effect of retigabine at EKof cultured
cortical neurons has been described (Rundfeldt, 1997). It is con-
ceivable, therefore, that those C-fibers for which retigabine had the
least effect on ‘resting’ excitability were already at a resting po-
tential close to EK. In central unmyelinated axons, too, a potential-
dependent effect of retigabine has been observed (Vervaeke et al.,
2006). An effect of the compound was only seen when the axons
and presynaptic terminals were depolarized using a high extra-
cellular [Kþ] concentration (Vervaeke et al., 2006).
Another effect of retigabine observed in this study is an increase
in latency of the compound action potential. This observation on
unmyelinated peripheral human axons resembles the findings
made using compound action potentials of myelinated and pre-
myelinated axons in rat sciatic and optic nerves (Devaux et al.,
2004). The underlying effect is most likely membrane hyper-
polarisation. However, an increase in membrane Kþconductance
may also contribute to this effect.
Prominent effects of retigabine were observed on the post-spike
recovery cycle (Figs. 4 and 5). The time constant of post-spike
superexcitability (Fig. 4) was reduced from more than 200 ms to
less than 100 ms in the presence of 10 mM retigabine (Fig. 4).
Retigabine produces membrane hyperpolarization and it may be
possible that the change in membrane potential causes the re-
duction in the time constant. To our knowledge, there is only one
previous studyabout the effects of membrane hyperpolarization on
the recovery cycle of human C-type nerve fibers for interspike in-
tervals up to 400 ms (Bostock et al., 2003). In these experiments,
membrane hyperpolarization due to an increase in stimulation
frequency resulted in no change of the recoverycycle time constant
in most of the unmyelinated axons. However, there was a decrease
in the time constant of sympathetic fibers (Bostock et al., 2003). It is
unlikely that only sympathetic fibers contribute to the compound
C-fiber action potentials of isolated human nerve segments.
Therefore, we conclude that the changes in the time constant seen
in the presence of retigabine are mainly due to the opening of
potassium channels (i.e. a reduction in membrane resistance).
Application of retigabine resulted in the development of a late
subexcitability (Fig. 5). This observation indicates the functional
importance of Kv7 channels to the axonal excitability after a burst
of impulses. For this type of experiment, three conditioning action
potentials were used because post-spike subexcitability is en-
hanced with the number of conditioning impulses (Bergmans,
1970; Baker et al., 1987; Bostock et al., 2003; Schwarz et al., 2006;
George et al., 2007). A possible contribution of a slow Kþcurrent
produced by Kv7 channels to a late subexcitability in unmyelinated
peripheral rat axons has been already suggested (George et al.,
2007) but not tested with pharmacological tools. In this study, the
maximum of this effect of retigabine on membrane threshold was
seen at about 250–400 ms following a short burst of action
potentials (Fig. 5). In contrast, in myelinated axons, membrane
subexcitability mediated by Kv7 channels (Devaux et al., 2004;
Schwarz et al., 2006) is produced at about 40–60 ms after a short
burst of action potentials (previously described as H1 by Bergmans,
1970). We did notfind a prominent H1 at this early time point in the
recovery period of isolated unmyelinated C-type nerve fibers in
peripheral human nerve. Also in microneurography recordings of
C-type nerve fibers in rat (George et al., 2007) and human nerve
(Bostock et al., 2003), a late subexcitability has been observed only
at about 200–300 ms. Differences between myelinated and un-
myelinated nerve fibers can be explained by variations in the spa-
tial expression of Kv7 channels in the different types of nerve fibers.
In myelinated fibers, Kv7 channels are localized and highly con-
centrated at nodes. Therefore, the high Kv7 conductance locally
opposes the Naþconductance. In C-fibers, Kv7 is more diffusely
distributed (Wladyka et al., 2008), like Naþchannels. Also, the
P.M. Lang et al. / Neuropharmacology 54 (2008) 1271–1278 1276

Page 7

effect of the slow Kþcurrent on membrane potential may depend
on the membrane time constant. The time constant for the nodal
membrane is much shorter (about 50 ms; Bostock and Rothwell,
1997) than that of the membrane of unmyelinated peripheral axons
(about 110 ms; Bostock et al., 2003). This can explain why Kv7
channels produce subexcitability in the early phase of the recovery
cycle in myelinated and in the late phase in unmyelinated axons.
The effect of retigabine was blocked by low micromolar con-
centrations of XE991. This is a pharmacological indication for
a contribution of Kv7 channels (Wang et al., 1998; Zaczek et al.,
1998). However, we did not find an effect of XE991 itself when
applied at the ‘‘resting’’ potential of the unmyelinated axons. This
situation has been described by others (Rivera-Arconada and
Lopez-Garcia, 2006; Vervaeke et al., 2006). In central axons, the
effects of XE991 on axonal excitability were only revealed during
applicationofhighextracellular
(Vervaeke et al., 2006) or by means of slow depolarising voltage
ramps (Hu et al., 2007). It is most likely, therefore, that the lack of
effect of XE991 given alone is due to the fact that Kv7 channels are
not strongly active at the resting potential of the C-type nerve
fibers, but only become so when their voltage-sensitivity is shifted
by retigabine.
Finally, the findings of this study do support the view that
retigabine may become a possible option in the treatment of neu-
ropathic pain (Blackburn-Munro and Jensen, 2003; Passmore et al.,
2003; Dost et al., 2004; Blackburn-Munro et al., 2005; Surti and Jan,
2005). The reduction in excitability of peripheral C-type nerve
fibers may help not only to suppress the development of sponta-
neous, ectopic action potential generation but also to suppress
stimulus-dependent symptoms such as allodynia and/or hyper-
algesia. Recently, the inhibition of ectopic activity has been dem-
onstrated in axotomized sensory fibers (Roza and Lopez-Garcia, in
press). Reduction of stimulus-dependent symptoms may be caused
by the retigabine-mediated attenuation in the ability to produce
bursts of action potentials.
potassium concentrations
Acknowledgement
We would like to thank the colleagues from the Department of
Hand and Plastic Surgery (Head: Prof. Dr. Stock; University of
Munich) who performed sural nerve biopsies and provided us with
a portion of it. Also, we are very grateful to Dr. Richard Carr for
critical reading of the manuscript, Ms. Christina Mu ¨ller for expert
technical assistance, and Dr. W. Dalby-Brown (NeuroSearch A/S) for
the provision of retigabine under EU FP6 project LSHM-CT-2004-
503038.Thiswork wassupported
schungsgemeinschaft (SFB 391/TPA1) and the UK Medical Research
Council.
by theDeutscheFor-
References
Ali, Z., Ringkamp, M., Hartke, T.V., Chien, H.F., Flavahan, N.A., Campbell, J.N.,
Meyer, R.A., 1999. Uninjured C-fiber nociceptors develop spontaneous activity
and alpha-adrenergic sensitivity following L6 spinal nerve ligation in monkey. J.
Neurophysiol. 81, 455–466.
Amir, R., Kocsis, J.D., Devor, M., 2005. Multiple interacting sites of ectopic spike
electrogenesis in primary sensory neurons. J. Neurosci. 25, 2576–2585.
Baker, M., Bostock, H., Grafe, P., Martius, P., 1987. Function and distribution of three
types of rectifying channels in rat spinal root myelinated axons. J. Physiol.
(Lond.) 383, 45–67.
Bergmans, J., 1970. The Physiology of Single Human Nerve Fibres. Vander, Louvain.
Blackburn-Munro, G., Dalby-Brown, W., Mirza, N.R., Mikkelsen, J.D., Blackburn-
Munro, R.E., 2005. Retigabine: chemical synthesis to clinical application. CNS
Drug Rev. 11, 1–20.
Blackburn-Munro, G., Jensen, B.S., 2003. The anticonvulsant retigabine attenuates
nociceptive behaviours in rat models of persistent and neuropathic pain. Eur. J.
Pharmacol. 460, 109–116.
Bostock, H., Campero, M., Serra, J., Ochoa, J., 2003. Velocity recovery cycles of C
fibres innervating human skin. J. Physiol. (Lond.) 553, 649–663.
Bostock, H., Cikurel, K., Burke, D., 1998. Threshold tracking techniques in the study
of human peripheral nerve. Muscle Nerve 21, 137–158.
Bostock, H., Rothwell, J.C., 1997. Latent addition in motor and sensory fibres of
human peripheral nerve. J. Physiol. (Lond.) 498, 277–294.
Campbell,J.N.,Meyer,R.A.,2006.Mechanismsofneuropathicpain.Neuron52,77–92.
Delmas, P., Brown, D.A., 2005. Pathways modulating neural KCNQ/M (Kv7) potas-
sium channels. Nat. Rev. Neurosci. 6, 850–862.
Devaux, J.J., Kleopa, K.A., Cooper, E.C., Scherer, S.S., 2004. KCNQ2 is a nodal Kþ
channel. J. Neurosci. 24, 1236–1244.
Devor, M., 1983. Potassium channels moderate ectopic excitability of nerve-end
neuromas in rats. Neurosci. Lett. 40, 181–186.
Devor, M., 2006. Response of nerves to injury in relation to neuropathic pain. In:
McMahon, S.B., Koltzenburg, M. (Eds.), Textbook of Pain. Elsevier, pp. 905–927.
Djouhri, L., Koutsikou, S., Fang, X., McMullan, S., Lawson, S.N., 2006. Spontaneous
pain, both neuropathic and inflammatory, is related to frequency of sponta-
neous firing in intact C-fiber nociceptors. J. Neurosci. 26, 1281–1292.
Dost, R., Rostock, A., Rundfeldt, C., 2004. The anti-hyperalgesic activity of retigabine
is mediated by KCNQ potassium channel activation. Naunyn Schmiedeberg’s
Arch. Pharmacol. 369, 382–390.
Eschenfelder, S., Ha ¨bler, H.J., Ja ¨nig, W., 2000. Dorsal root section elicits signs of
neuropathic pain rather than reversing them in rats with L5 spinal nerve injury.
Pain 87, 213–219.
George, A., Serra, J., Navarro, X., Bostock, H., 2007. Velocity recovery cycles of single
C fibres innervating rat skin. J. Physiol. (Lond.) 578, 213–232.
Gutman, G.A., Chandy, K.G., Grissmer, S., Lazdunski, M., McKinnon, D., Pardo, L.A.,
Robertson, G.A., Rudy, B., Sanguinetti, M.C., Stu ¨hmer, W., Wang, X., 2005. In-
ternational Union of Pharmacology. LIII. Nomenclature and molecular re-
lationships of voltage-gated potassium channels. Pharmacol. Rev. 57, 473–508.
Hirano, K., Kuratani, K., Fujiyoshi, M., Tashiro, N., Hayashi, E., Kinoshita, M., 2007.
Kv7.2–7.5 voltage-gated potassium channel (KCNQ2–5) opener, retigabine, re-
duces capsaicin-induced visceral pain in mice. Neurosci. Lett. 413, 159–162.
Hu, H., Vervaeke, K., Storm, J.F., 2007. M-channels (Kv7/KCNQ channels) that reg-
ulate synaptic integration, excitability, and spike pattern of CA1 pyramidal cells
are located in the perisomatic region. J. Neurosci. 27, 1853–1867.
Jentsch, T.J., 2000. Neuronal KCNQ potassium channels: physiology and role in
disease. Nat. Rev. Neurosci. 1, 21–30.
Kiernan, M.C., Bostock, H., 2000. Effects of membrane polarization and ischaemia on
the excitability properties of human motor axons. Brain 123, 2542–2551.
Lang, P.M., Burgstahler, R., Sippel, W., Irnich, D., Schlotter-Weigel, B., Grafe, P., 2003.
Characterization of neuronal nicotinic acetylcholine receptors in the membrane
of unmyelinated human C-fiber axons by in vitro studies. J. Neurophysiol. 90,
3295–3303.
Lang, P.M., Fleckenstein, J., Passmore, G.M., Brown, D.A., Grafe, P., 2008. A neuronal
potassium channel opener reduces the excitability of unmyelinated peripheral
human axons. Acta Physiol. 192 (Suppl. 663) PW03-15.
Liu, X., Eschenfelder, S., Blenk, K.H., Ja ¨nig, W., Ha ¨bler, H., 2000. Spontaneous activity
of axotomized afferent neurons after L5 spinal nerve injury in rats. Pain 84,
309–318.
Michaelis, M., Liu, X., Ja ¨nig, W., 2000. Axotomized and intact muscle afferents but
not skin afferents develop ongoing discharges of dorsal root ganglion origin
after peripheral nerve lesion. J. Neurosci. 20, 2742–2748.
Moalem-Taylor, G., Lang, P.M., Tracey, D.J., Grafe, P., 2007. Post-spike excitability
indicates changes in membrane potential of isolated C-fibers. Muscle Nerve 36,
172–182.
Munro, G., Dalby-Brown, W., 2007. K(v)7 (KCNQ) channel modulators and neuro-
pathic pain. J. Med. Chem. 50, 2576–2582.
Nielsen, A.N., Mathiesen, C., Blackburn-Munro, G., 2004. Pharmacological charac-
terisation of acid-induced muscle allodynia in rats. Eur. J. Pharmacol. 487,
93–103.
Ocana, M., Cendan, C.M., Cobos, E.J., Entrena, J.M., Baeyens, J.M., 2004. Potassium
channels and pain: present realities and future opportunities. Eur. J. Pharmacol.
500, 203–219.
Ochoa, J.L., Campero, M., Serra, J., Bostock, H., 2005. Hyperexcitable polymodal and
insensitive nociceptors in painful human neuropathy. Muscle Nerve 32,
459–472.
Passmore, G.M., Brown, D.A., 2007. Effects of M-channel modulators on peripheral
excitability in rat hairy skin. Soc. Neurosci. Abstr. 681.8.
Passmore, G.M., Selyanko, A.A., Mistry, M., Al Qatari, M., Marsh, S.J., Matthews, E.A.,
Dickenson, A.H., Brown, T.A., Burbidge, S.A., Main, M., Brown, D.A., 2003. KCNQ/
M currents in sensory neurons: significance for pain therapy. J. Neurosci. 23,
7227–7236.
Priestley, T., Hunter, J.C., 2006. Voltage-gated sodium channels as molecular targets
for neuropathic pain. Drug Dev. Res. 67, 360–375.
Rivera-Arconada, I., Lopez-Garcia, J.A., 2006. Retigabine-induced population pri-
mary afferent hyperpolarisation in vitro. Neuropharmacology 51, 756–763.
Roza, C., Lopez-Garcia, J.A. Retigabine, the specific KCNQ channel opener, blocks
ectopic discharges in axotomized sensory fibres. Pain, doi:10.1016/j.pain.2008.
01.031, in press.
Rundfeldt, C., 1997. The new anticonvulsant retigabine (D-23129) acts as an opener
of Kþchannels in neuronal cells. Eur. J. Pharmacol. 336, 243–249.
Rundfeldt, C., Netzer, R., 2000. The novel anticonvulsant retigabine activates
M-currents in Chinese hamster ovary-cells transfected with human KCNQ2/3
subunits. Neurosci. Lett. 282, 73–76.
Schwarz, J.R., Glassmeier, G., Cooper, E.C., Kao, T.C., Nodera, H., Tabuena, D., Kaji, R.,
Bostock, H., 2006. KCNQ channels mediate IKs, a slow Kþcurrent regulating
excitability in the rat node of Ranvier. J. Physiol. (Lond.) 573, 17–34.
P.M. Lang et al. / Neuropharmacology 54 (2008) 1271–12781277

Page 8

Surti, T.S., Jan, L.Y., 2005. A potassium channel, the M-channel, as a therapeutic
target. Curr. Opin. Investig. Drugs 6, 704–711.
Tatulian, L., Delmas, P., Abogadie, F.C., Brown, D.A., 2001. Activation of expressed
KCNQ potassium currents and native neuronal M-type potassium currents by
the anti-convulsant drug retigabine. J. Neurosci. 21, 5535–5545.
Vervaeke, K., Gu, N., Agdestein, C., Hu, H., Storm, J.F., 2006. Kv7/KCNQ/M-channels
in rat glutamatergic hippocampal axons and their role in regulation of excit-
ability and transmitter release. J. Physiol. (Lond.) 576, 235–256.
Wang, H.S., Pan, Z., Shi, W., Brown, B.S., Wymore, R.S., Cohen, I.S., Dixon, J.E.,
McKinnon, D.,1998. KCNQ2 and KCNQ3 potassium channel subunits: molecular
correlates of the M-channel. Science 282, 1890–1893.
Weidner, C., Schmelz, M., Schmidt, R., Hammarberg, B., Orstavik, K., Hilliges, M.,
Torebjork, H.E., Handwerker, H.O., 2002. Neural signal processing: the
underestimated contribution of peripheral human C-fibers. J. Neurosci. 22,
6704–6712.
Wickenden, A.D., Yu, W., Zou, A., Jegla, T., Wagoner, P.K., 2000. Retigabine, a novel
anti-convulsant, enhances activation of KCNQ2/Q3 potassium channels. Mol.
Pharmacol. 58, 591–600.
Wladyka, C.L., Feng, B., Glazebrook, P.A., Schild, J.H., Kunze, D.L., 2008. The KCNQ/M-
current modulates arterial baroreceptor function at the sensory terminal in rats.
J. Physiol. (Lond.), 795–802.
Wu, G., Ringkamp, M., Hartke, T.V., Murinson, B.B., Campbell, J.N., Griffin, J.W.,
Meyer, R.A., 2001. Early onset of spontaneous activity in uninjured C-fiber
nociceptors after injury to neighboring nerve fibers. J. Neurosci. 21, RC140.
Wu, G., Ringkamp, M., Murinson, B.B., Pogatzki, E.M., Hartke, T.V., Weerahandi, H.M.,
Campbell, J.N., Griffin, J.W., Meyer, R.A., 2002. Degeneration of myelinated
efferent fibers induces spontaneous activity in uninjured C-fiber afferents. J.
Neurosci. 22, 7746–7753.
Xiong, Q., Gao, Z., Wang, W., Li, M., 2008. Activation of Kv7 (KCNQ) voltage-gated
potassium channels by synthetic compounds. Trends Pharmacol. Sci. 29,
99–107.
Zaczek, R., Chorvat, R.J., Saye, J.A., Pierdomenico, M.E., Maciag, C.M., Logue, A.R.,
Fisher, B.N., Rominger, D.H., Earl, R.A., 1998. Two new potent neurotransmitter
releaseenhancers,10,10-bis(4-pyridinylmethyl)-9(10H)-anthracenone
10,10-bis(2-fluoro-4-pyridinylmethyl)-9(10H)-anthracenone:
linopirdine. J. Pharmacol. Exp. Ther. 285, 724–730.
and
to comparison
P.M. Lang et al. / Neuropharmacology 54 (2008) 1271–12781278