British Journal of Anaesthesia 96 (1): 7787 (2005)

doi:10.1093/bja/aei271

Advance Access publication November 18, 2005

Local anaesthetics inhibit signalling of human NMDA receptors

recombinantly expressed in Xenopus laevis oocytes: role of
protein kinase C
K. Hahnenkamp1*, M. E. Durieux3, A. Hahnenkamp2, S. K. Schauerte1, C. W. Hoenemann4,
V. Vegh1 4, G. Theilmeier1 and M. W. Hollmann5
1

Department of Anaesthesiology and Intensive Care, and 2Institute of Physiological Chemistry and
Pathobiochemistry, University Hospital, Munster, Germany. 3Department of Anesthesiology,
University of Virginia, Charlottesville, VA, USA. 4Department of Pharmaceutical Chemistry,
Comenius University, Bratislava, Slovakia. 5Department of Anesthesiology, University of Amsterdam,
Amsterdam, The Netherlands
*Corresponding author. E-mail: hahnenkamp@anit.uni-muenster.de
Background. N-methyl-D-aspartate (NMDA)-receptor activation contributes to postoperative
hyperalgesia. Studies in volunteers have shown that intravenous local anaesthetics (LAs) prevent
the development of hyperalgesic pain states. One potential explanation for this beneficial effect is
the inhibition of NMDA receptor activation. Therefore, we studied the effects of LA on NMDA
receptor function.
Methods. The human NR1A/NR2A NMDA receptor was expressed recombinantly in Xenopus
laevis oocytes. Peak currents were measured by voltage clamp in Mg- and Ca2+-free, Ba2+containing Tyrodes solution. Holding potential was 70 mV. Oocytes were stimulated with
glutamate/glycine (at EC50) with or without 10 min prior incubation in bupivacaine, levobupivacaine, S-( )-ropivacaine, or lidocaine (all at 10 910 4 M), procaine (10 4 M), R-(+)-ropivacaine
(10 4 M), QX314 (permanently charged, 510 4 M) extracellularly or intracellularly or benzocaine (permanently uncharged, 510 3 M). We also determined the effect of the protein kinase C
(PKC) inhibitors chelerythrine (510 5 M), calphostin C (310 6 M) and Ro 31-8220 (10 7 M),
and the effect of PKC activation with phorbolester (10 6 M).
Results. Non-injected oocytes were unresponsive to agonist application, but oocytes expressing
NMDA receptors responded with inward currents (1.10.08 mA). All LA concentrationdependently inhibited agonist responses. The inhibition was reversible and stereoselective. Intracellular QX314 reduced responses to 59% of control, but extracellular QX314 was without
effect. Benzocaine reduced responses to 33% of control. PKC inhibitors had no additional
inhibitory effect beyond that of bupivacaine. The effect of PKC activation was abolished in
the presence of bupivacaine.
Conclusion. All LA tested inhibited the activation of human NMDA receptors in a concentration
dependent fashion. This effect may contribute to reduced hyperalgesia and opiate tolerance
observed after systemic administration of LA. The effect is independent of the charge of LA;
site of action is intracellular. The mechanism of action may be mediated by inhibition of PKC.
Br J Anaesth 2006; 96: 7787
Keywords: anaesthetics; local; complications, hyperalgesia; oocytes; Xenopus laevis; protein
kinase C; receptors, NMDA
Accepted for publication: October 3, 2005

Local anaesthetics (LAs) act beyond blocking Na channels.

In addition to modulating processes like inflammation,
coagulation, platelet aggregation and the microcirculation1 2
they also affect hyperalgesic pain states.3 Clinical studies
demonstrate protection against hyperalgesia by systemic

administration of LA.46 These clinical observations support

preclinical studies demonstrating antihyperalgesic effects of
LA in animal models. The mechanisms underlying these
effects are unclear. Koppert et al.7 showed that low-dose
systemic LAs are able to reduce secondary hyperalgesia by a

The Board of Management and Trustees of the British Journal of Anaesthesia 2005. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org

Hahnenkamp et al.

containing Mg2+/Ca2+-free Tyrodes solution containing

Ba2+ (TyrBa, containing 150 mM NaCl, 5 mM KCl,
1.8 mM BaCl2, 10 mM dextrose and 10 mM Hepes,
pH adjusted to 7.4). Microelectrodes were pulled in one
stage from capillary glass on a vertical computer-controlled
electrode puller (Model 773, Campden Instruments Ltd,
Lafayette, IN, USA). Electrode tips were broken to a diameter of 10 mm, providing a resistance of 13 MV, and
filled with 3 M KCl. The oocytes were voltage clamped
using a two-electrode voltage clamp amplifier (OC725C;
Warner Instruments Corp., New Haven, CT) connected to
an IBM-compatible personal computer for data acquisition
and analysis. All measurements were performed at a holding
potential of 70 mV; data were recorded for 120 s.

central mode of action. Since spinal N-methyl-D-aspartate

(NMDA) receptors play a critical role in the development of
secondary hyperalgesia,79 we hypothesized that the beneficial effects of LA on postoperative pain may be explained
in part by the inhibition of NMDA receptor channel activation. To test this hypothesis we studied the influence of
clinically relevant LA on human NMDA receptor function
in an in vitro model. In addition, we investigated the site and
mechanism of action. Our results indicate that all LA studied
inhibit NMDA signalling at clinically relevant concentrations. The mechanism appears to be indirect, via inhibition
of protein kinase C (PKC).

Materials and methods

Study protocols
Oocyte harvesting and preparation

The physiological agonist glutamate (10 5 M), or the

receptor-selective agonist NMDA (10 3 M), in combination
with the obligatory co-agonist glycine (10 5 M), were
diluted in TyrBa solution to the required concentrations
and were delivered into the bath for 20 s. Responses were
quantified by measurement of peak currents and are reported
as mA. Substances used for incubation (LA, PKC activators
and inhibitors) prior to measurements were diluted in MBS
to the required concentrations. Oocytes were incubated in
LA for 10 min. We tested bupivacaine, S-( )-ropivacaine,
lidocaine or levobupivacaine at 10 910 4 M. Procaine and
R-(+)-ropivacaine were tested at 10 4 M. QX314 [N-(2,6)
dimethylphenylcarbamoylmethyl triethylammonium bromide, a quaternary LA, 99.9% permanently charged, which
does not penetrate the cell membrane] was injected into the
oocyte or applied outside the cell to identify an intracellular
or extracellular site of action. For injection QX314 was
diluted in 50 nl of 150 mM KCl to a concentration of 5 mM,
resulting in an intracellular concentration of 510 4 M
QX314. However, the oocyte is heavily compartmentalized,
which will lead to intracellular concentration differences. In
addition it contains a large amount of yolk, which may bind
various compounds with different affinities. Hence the concentration estimate may be prone to considerable variability.
Control cells were injected with 150 mM KCl. Each cell was
voltage clamped 10 min after injection. To differentiate the
effects of charged and uncharged LA, we studied the effect
of benzocaine (510 3 M, permanently uncharged, freely
membrane permeable). To study the role of PKC, cells were
incubated with PKC inhibitors chelerythrine 510 5 M, calphostin C 310 6 M and Ro 31-8220 10 7 M for 50 min
followed by 10 min bupivacaine 10 4 M in the continued
presence of PKC inhibitor. To evaluate an inhibitory effect
of LA on PKC activation in NMDA receptor-expressing
oocytes, we tested two approaches: (i) cells were coincubated in 10 6 M phorbolester (PMA, PKC activator)
and bupivacaine for 5 min, followed by another 5 min in
10 4 M bupivacaine (co-administration); (ii) cells were
incubated in 10 6 M PMA (5 min) followed by 10 min
10 4 M bupivacaine (subsequent application). Experiments

Procedures for Xenopus laevis oocyte isolation, synthesis

of NMDA receptor mRNA and microinjection technique
were published previously.1012 In brief, after approval by
the local Animal Care and Use Committee, female Xenopus
laevis frogs were anesthetized by immersion in cold 0.2%
3-aminobenzoic-methyl-ester until unresponsive to a painful
stimulus (toe pinching). Approximately 200 oocytes were
surgically removed. The oocytes were defolliculated by treatment for 2 h with collagenase type 1A diluted in oocyte Ringers (OR2) solution (containing 82.5 mM NaCl, 2 mM KCl,
1 mM MgCl2 and 5 mM Hepes, pH adjusted to 7.4). Microscopic observation confirmed that the follicle cells had been
removed.

NMDA receptor expression

The NMDA receptor combinations tested consist of NR1 and
NR2 subunits. The human NR1 (3000 bp) and NR2A
(5500 bp) subunits were obtained from Dr P. J. Whiting
(Merck Sharp & Dohme Research Laboratories, Harlow, UK)
as a complementary DNA in pcDNAI/Amp vectors. These
constructs were linearized by either the nuclease XbaI
(NR1A) or EcoRV (NR2A) and transcribed in the presence
of capping analogue by bacteriophage RNA polymerase T7,
using a commercial RNA preparation kit (mMESSAGE
mMACHINE TM T7 Kit, Ambion Inc., Austin, TX). Oocytes
were injected, using an automated microinjector (Nanoject;
Drummond Scientific, Broomall, PA, USA), with 6 ng of
NR1/NR2A subunits in a 1:5 weight ratio in 30 nl RNasefree sterile water. Correct injection was confirmed by a slight
increase in cell size. Oocytes were then stored for 4872 h in
MBS solution (MBS, containing 88 mM NaCl, 2.4 mM
NaHCO3, 0.41 mM CaCl2, 0.82 mM MgSO4, 0.3 mM
Ca2NO3, 10 mg ml 1 gentamycin, 10 mg ml 1 penicillin
and 15 mM Hepes, pH adjusted to 7.4) at 16 C.13 Oocytes
were transferred to fresh dishes with new MBS each day.

Electrophysiology
A single oocyte was positioned in a continuous-flow chamber with a volume of 0.5 ml and superfused (5 ml min 1)

78

Local anaesthetics and NMDA receptors

with PKC inhibitors and PMA were repeated with the

ester-LA procaine 10 4 M to exclude a distinct mechanism
for different types of binding.
For reversibility experiments a different study protocol
with repeated measurements of the same cell was used. First
the response to agonists were measured. The response serves
as control value for the following measurements. After
10 min incubation in LA 10 4 M the response to agonists
was recorded, followed by a measurement after 10 min
wash out.

Westborough, USA. All other chemicals were from Sigma

Aldrich Chemie GmbH (Steinheim, Germany).

Results
Functional expression of NMDA receptors in
Xenopus oocytes
Uninjected oocytes were unresponsive to either glutamate/
glycine (Fig. 1A) or to NMDA/glycine (data not shown). In
contrast, oocytes injected 4872 h before experimentation
with NR1A/2A receptor mRNA responded concentrationdependently with inward currents to glutamate/glycine
(Fig. 1B). Glycine, as expected, evoked inward currents in
the presence of glutamate only (Fig. 1B and D). The selective
agonist NMDA (10 3 M) was applied (in combination with
10 5 M glycine) to demonstrate that the receptors functioned appropriately as NMDA receptors. Responses were
indistinguishable from those induced by glutamate/glycine
(Fig. 1B and C). Therefore, for further experiments the
physiologic agonist glutamate in combination with glycine
was used. EC50 concentrations as determined in a previous
study from our group were employed: glutamate 810 6 M/
glycine 1210 6 M.14

Statistical analysis
Unless stated otherwise, results are reported as meanSEM.
Differences between treatment groups were analysed
using one-way ANOVA, corrected for multiple comparisons
(Dunnets test). Differences between two groups were analysed using Students t-test. Because variability between
batches of oocytes is common, at least 10 oocytes from at
least 3 frogs were studied for each data point and responses
were normalized to control values obtained on the same
day in oocytes from the same batch. Differences between
recordings were analysed using one-way repeated measures
ANOVA, corrected for multiple comparisons (Dunnets test).
P<0.05 was considered significant.

Materials
LAs inhibit glutamate signalling

Levobupivacaine was obtained from Abbott, Eindhoven,

The Netherlands, S-( )-ropivacaine from AstraZeneca,
Germany; R-(+)-ropivacaine from Astra USA Inc.,

We then tested the effects of LA on responses induced

by glutamate/glycine at EC50. S-( )-ropivacaine or

Uninjected, glu+gly (EC50)

Injected, glu+gly (EC50)

Injected, NMDA+gly

D
50 A
10 s

Injected, gly

Fig 1 Pharmacological characterization of NMDA (NR1/2A) receptors, recombinantly expressed in Xenopus laevis oocytes. Example traces of responses to
20 s agonist administration. (A) In oocytes which were not injected with NMDA mRNA. Cells were unresponsive to either glutamate/glycine or NMDA/
glycine. (B) In oocytes injected with NMDA (NR1A/2A) mRNA 2448 h prior to experiments. Glutamate/glycine at EC50 concentration evoked inward
currents. (C) The selective agonist NMDA (10 3 M) in combination with glycine (10 5 M) evoked inward currents in NMDA receptor expressing ooyctes.
Responses were indistinguishable from those stimulated by glutamate/glycine. (D) Glycine (10 3 M) in the absence of glutamate did not evoke currents in
oocytes expressing NMDA (NR1A/2A) receptors.

79

Hahnenkamp et al.

100

104

105

106

107

108

109

Ctrl

104

105

106

107

108

109

Bupivacaine (M)

Lidocaine (M)

105

106

100

104

107

108

109

Ctrl

104

105

106

107

108

109

50

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

50

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

Ropivacaine (M)

Levobupivacaine (M)

Fig 2 (AD) The effect of clinically used LA on NMDA receptor expressing Xenopus laevis oocytes. All tested LA concentrations dependently inhibit
activation by EC50 concentrations of glutamate/glycine. S-( )-ropivacaine or levobupivacaine at concentrations of 10 6 M significantly reduced NMDA
signalling to 727.4% (n=12) (levobupivacaine) and 662.9% (n=11) [S-( )-ropivacaine]. Responses to agonists were reduced to 62.29.4% (n=12) in the
presence of 10 7 M or greater lidocaine and to 726.4% (n=12) following 10 7 M or greater bupivacaine. Because inhibition of currents did not exceed 50%
with the highest LA concentration tested (except for lidocaine and benzocaine) the IC50 was not calculated. Results are plotted as % of control. Black bars
show control responses to EC50 glutamate/glycine, white bars show responses to EC50 glutamate/glycine after 10 min of incubation in different concentrations of local anesthetics. Responses were normalized to control values obtained on the same day in oocytes from the same batch. (*P<0.05 compared
with control responses.). Ctrl, control.

levobupivacaine at concentrations of 10 6 M significantly

reduced NMDA signalling to 727.4% (n=12) (levobupivacaine) and 662.9% (n=11) [S-( )-ropivacaine] (Fig. 2C
and D). Responses to agonists were reduced to 62.25.4
(n=12) in the presence of 10 7 M or greater lidocaine
and to 726.4% (n=12) following 10 7 M or greater bupivacaine (Fig. 2A and B). Because inhibition of currents did
not exceed 50% with the highest LA concentration tested
(except for lidocaine and benzocaine), we did not fit the
data to concentrationinhibition curves or calculate IC50.
To test if the effect of the LA was reversible and to
exclude conclusively the possibility that the inhibitory
effects of LA were due to variability among oocytes, we
exposed oocytes to each of the LA at 10 4 M for 10 min and
subsequently washed out the LA. After a 10-min wash-out in
MBS the responses to agonists were similar to control values
(P>0.05; t-test) (Fig. 3A and B), (n=6). These experiments
yielded results similar to those shown in Figure 2 (Fig. 3A).
Time matched blanks showed no significant differences.

with that of S-( )-ropivacaine. Whereas R-(+)-ropivacaine

was without effect, the clinically used S-( )-ropivacaine
inhibited the responses to glutamate/glycine to 619%
(n=14), indicating a stereoselective effect of LA on
NMDA receptors, making protein interaction of LA with
NMDA receptor signalling most likely (Fig. 4A). Since
this degree of stereoselectivity is much greater than
observed on other receptor systems we repeated these
experiments three times with similar results.

LA inhibit NMDA receptor signalling at an

intracellular site
We then determined if the site of action of the LA was
intracellular or extracellular, using the permanently charged
and therefore virtually non-membrane-permeable lidocaine
analogue QX314. When applied extracellularly, QX314 did
not inhibit agonist responses (responses were 9119% of
control, P>0.05, n=10; Fig. 4B). In contrast, intracellular
microinjection of the compound inhibited the responses to
agonists effectively (by 41%) to 598% (n=10; Fig. 4B). To
exclude an effect of the vehicle, 150 mM KCl was microinjected into the cell and showed no effect (responses
99.611% of control, n=10). This indicates that the site of

LA inhibition of NMDA signalling is stereoselective

To investigate stereoselectivity of LA on NMDA receptors
in oocytes, the effect of R-(+)-ropivacaine was compared

80

Local anaesthetics and NMDA receptors

Recovery

Control

S-()-ropivacaine
0.2 A
200 s

Levobupivacaine
S-()-ropivacaine

Bupivacaine

10 min wash out

10 min
Lidocaine

10 min wash out

*
10 min

10 min

50

Ctrl

10 min wash out

100

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

10 min

50

Ctrl

10 min wash out

100

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

0
Fig 3 (A) Example trace of EC50 glutamate/glycine evoked responses on a single oocyte expressing NMDA (NR1A/2A) receptors. The response
was inhibited to 57% after 10 min incubation S-( )-ropivacaine 10 4 M and restored after 10 min wash out phase in MBS. (B) The observed inhibition
of EC50 glutamate/glycine evoked NMDA (NR1A/2A) receptor responses after 10 min incubation in 10 4 M in all tested local anesthetics (LA) was
reversed after a 10 min wash out phase in MBS solution. Experiments were performed in single oocytes with repeated measurements of the same cell. After
10 min incubation in LA 10 4 M the response to agonists was recorded, followed by a measurement after 10 min wash-out. (*P<0.05 LA vs control and
LA+wash out). Ctrl, control.

action of QX314 on NMDA receptor signalling is located

intracellularly. The fact that the degree of inhibition is
different from that of lidocaine may result from the approximate estimation of intracellular QX314 concentrations
(see Materials and methods section).

when oocytes were exposed to benzocaine 5 mM (n=10).

Therefore, both charged and uncharged LA are able to
inhibit NMDA signalling (Fig. 4B). However, a conclusion
whether the site of action of benzocaine is intra- or extracellular cannot be made.

LA inhibition of NMDA signalling is not

charge-dependent

LA inhibit NMDA receptor signalling by

inhibiting PKC

Our data with QX314 indicate that charged LA can block

NMDA signalling. To clarify whether charge is required for
NMDA signalling block by LA, the effects of the almost
completely uncharged and highly membrane permanent
LA benzocaine on NMDA signalling in oocytes were tested.
The agonist response was inhibited to 334% of control

NMDA receptors are highly regulated by PKC, which

enhances signalling by phosphorylating the C-terminal segments of NMDA receptors.15 16 Since interactions between
LA and PKC have been shown previously,17 and since our
data suggest an intracellular site of action of LA, we hypothesized that LA could affect NMDA signalling indirectly

81

Hahnenkamp et al.

Ctrl

S-()-ropivacaine

R-(+)-ropivacaine

200 nA
10 s

100

S-()ropivacaine

50

R-(+)-ropivacaine

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

0
Intracellular

QX314

QX314 (KCI)

Ctrl (KCI)

50

Extracellular

BENZ

100

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

Fig 4 (A) Example traces of EC50 glutamate/glycine evoked responses in oocytes expressing NMDA receptors. Responses to agonists were inhibited, when
oocytes were incubated with the clinically used S-( )-ropivacaine, but not when incubated with its stereoisomer R-(+)-ropivacaine. Example traces were
from measurements on one day, oocytes were from the same batch. (B) Responses to stimulation of NR1A/2A expressing oocytes with EC50 glutamate/
glycine was inhibited to 61% after 10 min incubation of oocytes in 10 4 M S-( )-ropivacaine (*P<0.05). No effect was noted after incubation in
10 4 M R-(+)-ropivacaine (n=14). (C) Stimulation of NR1A/2A expressing oocytes with EC50 glutamate/glycine after intracellular injection of 510 3 M
QX314 (permanently charged and non membrane permeable) and 10 min of incubation with the resulting intracellular concentration 510 4 M was
inhibited to 59 of control (n=10). QX314 was diluted in 150 mM KCl to maintain the osmolality of the oocyte. Intracellular injection of 150 mM KCl had
no effect on agonist stimulation. Incubation of oocytes with 510 4 M QX314 (extracellular) had no effect on agonist stimulation. Incubation of oocytes
with permanently charged benzocaine (BENZ, 510 3 M) reduced the responses to agonists to 33 of control (n=10; *P<0.05 vs control). Ctrl, control.

by inhibiting PKC. This could result either directly in a

decreased phosphorylation state of the receptor or a
decreased phosphorylation state of a signalling system
downstream of PKC, which in turn would inhibit NMDA
receptor function. To determine whether LA inhibit PKC
and consequently inhibit NMDA signalling, the effect of
PKC inhibition on the inhibitory effect of bupivacaine
was studied. We compared the effects on NMDA signalling
of (i) pre-treatment (incubation for 1 h) with the PKC inhibitors chelerythrine (510 5 M), Ro 31-8220 (10 7 M) or
calphostin C (310 6 M); (ii) bupivacaine (10 4 M); and
(iii) the combination of a PKC inhibitor and bupivacaine.
Ro 31-8220 inhibited responses to agonists to 645%
(n=12), chelerythrine to 667% (n=12) and calphostin C
to 567% (n=11) of control cells. Bupivacaine inhibited
responses to agonists to 579% (n=11) in chelerythrine

experiments, 507% (n=12) in calphostin C experiments and

607% (n=12) in Ro-31-8220 experiments. The combination of the compounds did not further reduce responses to
NMDA receptor agonists as compared with either compound alone [to 6210% in combination with Ro 31-8220
(Fig. 5C), to 629% in combination with chelerythrine
(Fig. 5A) and to 366% in combination with calphostin C
(Fig. 5B)] (P>0.05; n=10). This suggests that LA and the
PKC inhibitors share a common PKC-dependent mechanism, and that LA block NMDA receptor signalling by inhibition of the PKC pathway.
LA should therefore be able to affect PKC activation
of NMDA receptors. To test this hypothesis, we studied
the effect of 10 4 M bupivacaine on phorbolester (PMA)activated NMDA receptors. Bupivacaine alone inhibited
responses to 5510% (n=10) of control in this experiment.

82

Local anaesthetics and NMDA receptors

Ctrl

BUP

CHE

BUP/CHE

100 nA
20 s

80

Ctrl

60

100

BUP/CAL

BUP

20

CAL

40

80
60

BUP/Ro

20

Ro

40
BUP

Glutamate/glycine-induced
peak currents (% of control)

80

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

100

BUP

20

CHE

40

BUP/CHE

60

100

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

Fig 5 The effects of 1 h pretreatment with different PKC inhibitors on NMDA signalling. (A) Example traces of EC50 glutamate/glycine evoked responses in
oocytes expressing NMDA receptors. Responses to agonists are inhibited, when oocytes are incubated with bupivacaine (BUP) or the PKC inhibitor
chelerythrine (CHE). The combination of both (BUP/CHE) agents did not further reduce the evoked responses. Example traces were from measurements on
one day, oocytes were from the same batch. (B) Incubation of cells with chelerythrine (510 5 M) inhibited responses to EC50 glutamate/glycine evoked
NMDA (NR1A/2A) receptor currents to 66% (n=12). Bupivacaine (BUP) inhibited responses to 57% after 10 min incubation (n=11). The combination of
compounds did not further reduce responses to NMDA receptor agonists as compared with either compound alone (62%, n=10). (C) Incubation of cells with
calphostin C (CAL, 310 6 M) inhibited responses to EC50 glutamate/glycine evoked NMDA (NR1A/2A) receptor currents to 56% (n=12). Bupivacaine
(BUP) inhibited responses to 50% (n=12) after 10 min incubation in this experiment. The combination of compounds (BUP/CAL) did not further reduce
responses to NMDA receptor agonists as compared with either compound alone (37%, n=10). (D) Incubation of cells with Ro 31-8220 (Ro, 10 7 M)
inhibited responses to EC50 glutamate/glycine evoked NMDA (NR1A/2A) receptor currents to 64% (n=12). Bupivacaine (BUP) inhibited responses to
60% after 10 min incubation (n=12). The combination of compounds (BUP/Ro) did not further reduce responses to NMDA receptor agonists as compared
with either compound alone (62%, n=10). (*P<0.05 vs control). Ctrl, control.

PMA (10 6 M for 5 min) alone induced a significant

increase to 18810% (n=10) in glutamate/glycine activated
cells expressing NR1A/2A NMDA receptors. Responses to
glutamate/glycine were then elicited in cells (i) that had been

incubated with PMA (10 6 M) for 5 min and subsequently

with bupivacaine (10 4 M) for 10 min, or (ii) that had been
exposed to a combination of PMA (10 6 M) and bupivacaine
(10 4 M) for 5 min and subsequently to bupivacaine

83

Hahnenkamp et al.

(10 4 M) alone for 5 min. Subsequent treatment with bupivacaine did not reverse the stimulatory effect of PMA
(17818%; n=12), suggesting that once PKC activation
(and subsequent receptor phosphorylation) had occurred,
bupivacaine could no longer inhibit NMDA signalling. In
contrast, bupivacaine completely inhibited the stimulatory
effect of the PKC activator PMA on NMDA receptor
currents (10515%; n=10) when the agents were administered simultaneously (Fig. 6A). These findings further substantiate the suggestion that LA inhibit NMDA signalling
indirectly, via inhibition of a PKC dependent pathway.

and an extracellularly located site of action closely related to

those of Mg2+ and ketamine has been suggested for procaine,
an ester-type LA. Thus, to evaluate a potential difference
in effect between amide- and ester-type LA we tested the
effect of procaine on PKC activation by PMA (10 6 M)
and on PKC inhibition by chelerythrine (510 5 M). Procaine (10 4 M, 10 min incubation) inhibited responses to
glutamate/glycine in NR1A/2A NMDA receptor expressing
cells to 606% (n=10) compared with control responses. In
these experiments PKC inhibition by chelerythrine (60 min
incubation) induced a 475% (n=10) decrease in response.
The combination of chelerythrine and procaine did not
further inhibit responses to NMDA receptor agonists as
compared with either compound itself. The combination
of procaine and PKC activator PMA (9312%; n=10) completely abolished the stimulatory effect of PMA on NMDA
receptor currents (16813%; n=10). These findings indicate
that there is no difference in the site of action between
amide- and ester-type LA.

Ester and amide type LA similarly inhibit

PKC signalling
Different sites of action and mechanisms for LA effects on
NMDA signalling have been proposed by Sugimoto et al.18

200

Discussion

150

Our results demonstrate that LA concentration-dependently

and reversibly inhibit glutamate/glycine-induced NMDA
receptor channel (NR1A/2A) activation. This effect is
stereoselective and does not depend on the charge of the
molecules. The site of action is located intracellularly.
Inhibition of PKC seems to be involved. Several groups
have demonstrated that LA are able to suppress hyperalgesic
pain states, particularly secondary hyperalgesia.3 7 19 Our
results might provide a partial explanation for the reduced
hyperalgesia and opiate tolerance observed after administration of LA.6 7 13 Concentrations of LA that significantly
inhibited NMDA receptor channel activation in our study

PMA/BUP
(subseq.)

**
BUP

PMA

50

PMA/BUP
(co-incub.)

100

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

100
80

PRO+CHE

CHE

20

PRO

40

C
150

100

PMA

PRO

50

PRO+PMA

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

Fig 6 (A) The effect of 10 4 M bupivacaine (BUP) on phorbol ester (PMA)

activated NMDA receptors. Bupivacaine alone inhibited responses to 55%
(n=10) of control in this experiment. PMA (10 6 M for 5 min) alone induced
a significant increase of responses to 188% (n=10) in glutamate/glycine
activated cells expressing NR1A/2A NMDA receptors. Responses to EC50
glutamate/glycine were then elicited in cells (i) that had been incubated
with PMA (10 6 M) for 5 min and subsequently to bupivacaine (10 4 M) for
10 min and (ii) that had been exposed to a combination of PMA (10 6 M)
and bupivacaine (10 4 M) for 5 min and subsequently to bupivacaine
(10 4 M) alone for 5 min. Whereas subsequent treatment with
bupivacaine did not reverse the stimulatory effect of PMA (178%,
n=12), bupivacaine completely inhibited the stimulatory effect of the
PKC activator PMA on NMDA receptor currents (105%) (*P<0.05 vs
control, **P<0.05 vs PMA and control). To evaluate a distinct mechanism
between amide- and ester-type LA we tested the effect of procaine (Pro) on
PKC activation by PMA (10 6 M, 5 min) and PKC inhibition by chelerythrine (CHE, 510 5 M, 60 min). Procaine (10 4 M, 10 min incubation)
inhibited responses to glutamate/glycine in NR1A/2A NMDA receptor
expressing cells to 60% (n=10) compared with control responses. (B)
PKC inhibition by chelerythrine induced a 47% (n=10) decrease in response
sizes. Combination with procaine did not further inhibit responses to
NMDA receptor agonists as compared with either compound itself. (C)
The combination of procaine and PKC activator PMA (93%, n=10) completely abolished the stimulatory effect of PMA on NMDA receptor currents
(168%, n=10) (*P<0.05 vs control). Ctrl, control.

60

Ctrl

Glutamate/glycine-induced
peak currents (% of control)

84

Local anaesthetics and NMDA receptors

are within the range as measured in blood during epidural

anaesthesia.14

second measurement of the same cell extending the time

course of 40 min.

NMDA receptor expression

Effects of LA on NMDA signalling in

other models

The NMDA receptor is a protein complex composed of

two classes of subunits, the essential subunit NR1 and the
modulating NR2 subunit (of which four different types
exist: AD). The NR2 subunits alone cannot form functional
channels, but they amplify NR1 activity and induce
functional variability in NMDA receptor signalling. These
subunits co-assemble in various combinations to form
functionally distinct NMDA receptors.2022
In the present study, the NR1 subunit was co-expressed
with the NR2A subunit. This combination is widely distributed in the brain and the dorsal horn of the spinal cord and is
believed to play a relevant physiological role in the development of hyperalgesia and acute opioid tolerance.23 24
Although NMDA receptor expression in Xenopus oocytes
is an established model,11 some limitations of the model
should be noted. First, our experiments were performed at
room temperature, whereas the expressed receptor is derived
from human sources and normally functions at 37 C. This
might theoretically influence its behaviour. However, we
found it more important to maintain the cell membrane
in its normal state of fluidity. Second, NR2 subunits were
co-expressed with NR1 subunits to enhance currents through
expressed receptors and to provide a more physiological
receptor configuration.23 In order to achieve heteromeric
expression, the mRNA was injected into the oocytes in a
weight ratio of 1:5 between NR1A and NR2A.25 We could
not determine expression levels or stoichiometry for the
subunit combinations. Therefore, although we used a defined
mRNA weight ratio between NR1 and NR2 subunits, it is
conceivable that the ratio of NR1 to NR2 varied during
experimentation.

The charge of LA influences its pharmacological properties.

Since charged fraction varies greatly among clinically used
LA it is important to know whether the charge of LA might
influence the inhibition of NMDA receptors. Our results
indicate that both a completely uncharged and also a completely charged LA were able to inhibit NMDA receptor
activation. Interestingly, QX314 only inhibits activation
of the NMDA receptor when applied intracellularly,
but not when applied extracellularly. Therefore the site of
action of charged LA seems to be located intracellularly.
The uncharged benzocaine is freely membrane permeable,
therefore no site of action can be determined. These findings
are in agreement with previous studies. In whole cell patch
clamp experiments in the electrosensory lateral line lobe
of the fish Apteronotus leptorhynchus, the amplitudes of
NMDA-evoked excitatory postsynaptic potentials (EPSPs)
have been shown to be inhibited in the presence of intracellular QX314. The authors suggest an indirect inhibitory
effect of QX314 acting through a block of persistent sodium
channels.26 Our results indicate that QX314 may also
affect NMDA signalling in the absence of sodium channel
signalling.
In contrast to our results Nishizawa and colleagues27
could not find an inhibitory effect on NMDA-induced currents of clinically relevant concentrations of lidocaine or
bupivacaine using the whole cell patch clamp technique
in CA1 mouse pyramidal neurons. Bupivacaine did inhibit
NMDA-induced currents but only at 1 mM concentration.
In addition to model differences and a possible difference in
NMDA receptor subtypes studied (which were not defined
in the study of Nishizawa and colleagues), a potential explanation for these different results might be the choice of
agonists. Whereas in our model the influence of LA on
stimulation with the physiological agonists glutamate and
glycine were tested, the specific but non-physiological
agonist NMDA was used in the study of Nishizawa and
colleagues. A potential inhibition induced by greater concentrations of LA could not be tested in their model because
the neurons became leaky at high LA concentrations.

Study protocol of electrophysiological

measurements
To exclude the possibility that results depend on variability
of the oocytes (which is the main reservation towards this
protocol), we also tested all LA at 10 4 M concentration
with oocytes serving as their own control (protocol 2, shown
in the reversibility experiments).
Results of both protocols yield comparable results, as had
also been shown in preliminary experiments for the design of
our study protocol. Both study protocols have been used
successfully in this model in the past. As our experiments
for the identification of intracellular mechanisms extend to
1 h incubation time, most experiments were performed with
protocol 1. This protocol provides the advantage of longer
incubation times. At least in our hands oocytes do not hold
their membrane potential reliably for that time course. Two
electrodes inside an oocyte for 1 h interfere with the membrane integrity. Therefore protocol 2 cannot be used for a

Effects of ester- and amide-type LAs

Sodium channel blocking potency of LA depends upon the
type of linkage in the intermediate chain. Ester-linked LA
produce more potent Na channel blockade. Sugimoto and
colleagues18 found that ester-type LA inhibit NMDA signalling more potently than did amide-type LA and propose
two distinct extracellularly located mechanisms of action.
Our data cannot confirm these findings. In contrast, we found
that there is no difference in potency concerning the inhibitory effect of LA on NMDA receptor signalling. The site of

85

Hahnenkamp et al.

action is intracellular and no mechanistic difference was

observed between amide- and ester-type LA. In contrast
to our study, Sugimoto and colleagues only observed inhibition of NMDA responses at concentrations of LA greater
than 10 4 M; responses were blocked almost completely
at 10 1 M. At these high concentrations of LA unspecific
membrane effects cannot be excluded and mechanistic
conclusions cannot probably be drawn.

Role of PKC

PKC-dependent phosphorylation of NMDA receptor subunits amplifies NMDA receptor responses, resulting in
enhanced currents through NMDA receptor channels.15
Inhibition of PKC results in reduced responses to NMDA
receptor agonists in trigeminal and spinal cord dorsal horn
neurones28 as well as in oocytes expressing recombinant
NMDA receptors.29 In our study, no additive effect was
observed when oocytes were incubated simultaneously in
LA and one of three structurally distinct PKC inhibitors,
suggesting that LA and PKC antagonists share a common
mechanism. Therefore it appears that the inhibitory effects
of LA on NMDA receptor signalling are indirect and mediated by inhibition of PKC. Despite truncation and point
mutation experiments it remains unclear whether PKCmediated action on NMDA receptors is entirely direct
(via direct phosporylation of NMDA receptors15), indirect
[via a downstream located signalling peptide of the nonreceptor tyrosine kinase (Src) signalling cascade16] or a
mixture of both.30 Our results do not address this issue
but show that inhibition of PKC is most likely the mechanism involved in LA inhibitory action.
In summary, clinically relevant LA concentrationdependently inhibit the activation of human NMDA receptors. The effect is intracellularly located and seems to be
mediated by PKC inhibition.

10

11

12

13

14

15

16

17

18

Acknowledgements
This study was supported by the 2004 Ben Covino Research Award of the
International Anesthesia Research Society (IARS), USA, sponsored by
AstraZeneca, Sweden, and by the Department of Anesthesiology and
Intensive Care, University Hospital Munster.

19

20

References
1 Hahnenkamp K, Herroeder S, Hollmann MW. Regional anaesthesia, local anaesthetics and the surgical stress response. Best
Pract Res Clin Anaesthesiol 2004; 18: 50927
2 Hollmann MW, Durieux ME. Local anesthetics and the inflammatory response: a new therapeutic indication? Anesthesiology
2000; 93: 85875
3 Rowbotham MC, Reisner-Keller LA, Fields HL. Both intravenous
lidocaine and morphine reduce the pain of postherpetic neuralgia.
Neurology 1991; 41: 10248
4 Cassuto J, Wallin G, Hogstrom S, Faxen A, Rimback G. Inhibition
of postoperative pain by continuous low-dose intravenous
infusion of lidocaine. Anesth Analg 1985; 64: 9714
5 Groudine SB, Fisher HA, Kaufman RP Jr, et al. Intravenous lidocaine
speeds the return of bowel function, decreases postoperative

21

22

23

24

86

pain, and shortens hospital stay in patients undergoing radical

retropubic prostatectomy. Anesth Analg 1998; 86: 2359
Koppert W, Zeck S, Sittl R, Likar R, Knoll R, Schmelz M. Low-dose
lidocaine suppresses experimentally induced hyperalgesia in
humans. Anesthesiology 1998; 89: 134553
Koppert W, Ostermeier N, Sittl R, Weidner C, Schmelz M. Low
dose lidocaine reduces secondary hyperalgesia by a central mode
of action. Pain 2000; 85: 21724
Celerier E, Laulin J, Larcher A, Le Moal M, Simonnet G. Evidence
for opiate-activated NMDA processes masking opiate analgesia in
rats. Brain Res 1999; 847: 1825
Sandkuhler J. Learning and memory in pain pathways. Pain 2000;
88: 1138
Le Bourdelles B, Wafford K, Kemp J, et al. Cloning, functional
coexpression, and pharmacological characterisation of human
cDNA encoding NMDA receptor NR1 and NR2A subunits.
J Neurochem 1994; 62: 20918
Yamakura T, Harris R. Effects of gaseous anesthetics nitrous oxide
and xenon on ligand-gated ion channels. Anesthesiology 2000; 93:
1095101
Sullivan LM, Honemann CW, Arledge JA, Durieux ME. Synergistic
inhibition of lysophosphatidic acid signaling by charged and
uncharged local anesthetics. Anesth Analg 1999; 88: 111724
Vaida SJ, Ben David B, Somri M, Croitoru M, Sabo E, Gaitini L. The
influence of preemptive spinal anesthesia on postoperative pain.
J Clin Anesth 2000; 12: 3747
Rygnestad T, Zahlsen K, Bergslien O, Dale O. Focus on mobilisation after lower abdominal surgery. A double-blind randomised
comparison of epidural bupivacaine with morphine vs. lidocaine
with morphine for postoperative analgesia. Acta Anaesthesiol Scand
1999; 43: 3807
Liao GY, Wagner DA, Hsu MH, Leonard JP. Evidence for direct
protein kinase-C mediated modulation of N-methyl-D-aspartate
receptor current. Mol Pharmacol 2001; 59: 9604
Lu WY, Xiong ZG, Lei S, et al. G-protein-coupled receptors
act via protein kinase C and Src to regulate NMDA receptors.
Nat Neurosci 1999; 2: 3318
Do SH, Fang HY, Ham BM, Zuo ZY. The effects of lidocaine on the
activity of glutamate transporter EAAT3: The role of protein
kinase C and phosphatidylinositol 3-kinase. Anesth Analg 2002;
95: 12638
Sugimoto M, Uchida I, Mashimo T. Local anaesthetics have different mechanisms and sites of action at the recombinant
N-methyl-D-aspartate (NMDA) receptors. Br J Pharmacol 2003;
138: 87682
Kastrup J, Petersen P, Dejgard A, Angelo HR, Hilsted J. Intravenous lidocaine infusiona new treatment of chronic painful
diabetic neuropathy? Pain 1987; 28: 6975
Cull-Candy S, Brickley S, Farrant M. NMDA receptor subunits:
diversity, development and disease. Curr Opin Neurobiol 2001; 11:
32735
Kus L, Saxon D, Beitz A. NMDA R1 mRNA distribution in motor
and thalamic-projecting sensory neurons in the rat spinal cord and
brain stem. Neurosci Lett 1995; 196: 2014
Monyer H, Burnashev N, Laurie D, Sakmann B, Seeburg P.
Developmental and regional expression in the rat brain and
functional properties of four NMDA receptors. Neuron 1994;
12: 52940
Tolle TR, Berthele A, Zieglgansberger W, Seeburg PH, Wisden W.
The differential expression of 16 NMDA and non-NMDA receptor subunits in the rat spinal cord and in periaqueductal gray.
J Neurosci 1993; 13: 500928
Watanabe M, Mishina M, Inoue Y. Distinct spatiotemporal distributions of the N-methyl-D-aspartate receptor channel subunit

Local anaesthetics and NMDA receptors

mRNAs in the mouse cervical cord. J Comp Neurol 1994; 345:

3149
25 Liu H, Hollmann M, Liu W, Hoenemann C, Durieux M. Modulation of NMDA receptor function by ketamine and magnesium.
Part I. Anesth Analg 2001; 92: 117381
26 Berman N, Dunn RJ, Maler L. Function of NMDA receptors and persistent sodium channels in a feedback pathway
of the electrosensory system. J Neurophysiol 2001; 86:
161221
27 Nishizawa N, Shirasaki T, Nakao S, Matsuda H, Shingu K. The
inhibition of the N-methyl-D-aspartate receptor channel by local

anesthetics in mouse CA1 pyramidal neurons. Anesth Analg 2002;

94: 32530
28 Chen L, Huang LY. Protein kinase C reduces Mg2+ block of
NMDA-receptor channels as a mechanism of modulation. Nature
1992; 356: 5213
29 Mori H, Yamakura T, Masaki H, Mishina M. Involvement of the
carboxyl-terminal region in modulation by TPA of the NMDA
receptor channel. Neuroreport 1993; 4: 51922
30 Ali D, Salter M. NMDA receptor regulation by Src kinase signalling
in excitatory synaptic transmission and plasticity. Curr Opin Neurobiol 2001; 11: 33642

87