British Journal of Anaesthesia 112 (2): 370–9 (2014)
Advance Access publication 24 September 2013 . doi:10.1093/bja/aet299
Time-dependent modulation of muscarinic m1/m3 receptor
signalling by local anaesthetics
S. Picardi1,2, M. F. Stevens 2,3, K. Hahnenkamp 4, M. E. Durieux 5, P. Lirk 2,3* and M. W. Hollmann 2,3
1
Department of Anaesthesiology, University Hospital Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany
Laboratory of Experimental Intensive Care and Anaesthesiology and 3 Department of Anaesthesiology, Academic Medical Center,
Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
4
Department of Anaesthesiology and Critical Care, University Hospital Münster, Albert-Schweitzer-Campus 1, 48149 Münster, Germany
5
Department of Anaesthesiology, University of Virginia, Charlottesville, VA, USA
2
* Corresponding author. E-mail: [email protected]
Editor’s key points
† Local anaesthetics (LAs)
may have
anti-inflammatory
actions.
† G-protein-coupled
receptors are likely targets
for LAs.
Background. Signalling of several G-protein-coupled receptors of the Gq/11 family is
time-dependently inhibited by local anaesthetics (LAs). Since LA-induced modulation of
muscarinic m1 and m3 receptor function may explain their beneficial effects in clinical
practice, such as decreased postoperative cognitive dysfunction or less bronchoconstriction,
we studied how prolonged exposure affects muscarinic signalling (Wang D, Wu X, Li J, Xiao F,
Liu X, Meng M. The effect of lidocaine on early postoperative cognitive dysfunction after
coronary artery bypass surgery. Anesth Analg 2002; 95: 1134–41; Groeben H, Silvanus MT,
Beste M, Peters J. Combined lidocaine and salbutamol inhalation for airway anesthesia
markedly protects against reflex bronchoconstriction. Chest 2000; 118: 509 –15).
† Xenopus oocytes were
used to invesitgate effects
of lidocaine on muscarinic
m1 and m3 receptors.
Methods. A two-electrode voltage clamp was used to assess the effects of lidocaine or its
permanently charged analogue QX314 on recombinantly expressed m1 and m3 receptors in
Xenopus oocytes. Antisense knock-down of functional Gaq-protein and inhibition of protein
kinase C (PKC) served to define mechanisms and sites of action.
† Lidocine had a biphasic
effect which may be due
to modulation by protein
kinase C.
Results. Lidocaine affected muscarinic signalling in a biphasic way: an initial decrease
in methylcholine bromide-elicited m1 and m3 responses after 30 min, followed by a
significant increase in muscarinic responses after 8 h. Intracellularly injected QX314 timedependently inhibited muscarinic signalling, but had no effect in Gaq-depleted oocytes. PKCantagonism enhanced m1 and m3 signalling, but completely abolished the LA-induced
increase in muscarinic responses, unmasking an underlying time-dependent inhibition of
m1 and m3 responses after 8 h.
† These results may have
future clinical implications
Conclusions. Lidocaine modulates muscarinic m1 and m3 receptors in a time- and Gaqdependent manner, but this is masked by enhanced PKC activity. The biphasic time course
may be due to interactions of LAs with an extracellular receptor domain, modulated by PKC
activity. Prolonged exposure to LAs may not benefit pulmonary function, but may positively
affect postoperative cognitive function.
Keywords: G-protein-coupled receptors; local anaesthetics; muscarinic m1 and m3 receptors
Accepted for publication: 11 June 2013
Block of voltage-gated sodium channels, leading to inhibition
of nerve impulse conduction, is probably the most important
effect of local anaesthetics (LAs) and the major mechanism
underlying their well-known antinociceptive and antiarrhythmic properties. However, recently discovered aspects of LA
function, in particular their anti-inflammatory properties,
seem to provide further significant benefit for patients in the
clinical setting, as they attenuate the surgery-induced stress
response but do not impair physiological host defence.1 Signalling of G-protein-coupled receptors (GPCRs) and in particular
the a-subunit of G-proteins of the Gq/11 family, well known
for their role within the immune response, seem to be one potential target for LAs.2 3
We have previously shown that signalling of different
immunomodulatory Gaq-coupling receptors, such as the
lysophosphatidic acid (LPA) or thromboxane A2 receptor, is
time-dependently inhibited at concentrations of LA similar to
that seen clinically after epidural anaesthesia or i.v. administration.4 An increased inhibitory potency and thus a potentially
increased efficacy on GPCR signalling pathways after prolonged exposure to LA might thus explain the beneficial
effects, such as faster return of gastrointestinal function or
& The Author [2013]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved.
For Permissions, please email: [email protected]
BJA
Local anaesthetics and muscarinic signalling
improved postoperative pain, seen in the clinical setting even
after completion of LA treatment.5
Functional Gaq-protein was critically required for the
observed time-dependent effects of long-term LA exposure,
suggesting that all GPCRs of the Gq/11-family would be timedependently inhibited.4 However, in pilot studies, m1 and m3
muscarinic acetylcholine receptors, both bound to the Gq/11
family of G proteins, behaved differently. m1 and m3 muscarinic
receptors were shown to be more sensitive to LA than the neuronal sodium channel: half-maximal inhibitory concentrations
(IC50) were 18 nM for the m1 receptor and 370 nM for the m3 receptor, respectively, which is 100–1000 times less than
required for sodium channel block (IC50 is 60–200 mM).6 – 8 The
Gaq-protein was defined as the site of action on these receptors
for short-term exposure effects of LA.2 Signalling of muscarinic
receptors plays an essential role in memory and learning, and
is a potential target for the treatment of cognitive deficits.9 Persistent cognitive impairment after surgery is an increasing challenge, particularly in elderly patients, who have higher morbidity
and mortality. I.V. lidocaine was shown to improve early postoperative dysfunction in patients undergoing cardiac surgery.10
Muscarinic m1 and m3 receptors are widely expressed in
airway and lung tissue, mediating alveolar smooth muscle contraction and thus airway hyperresponsiveness.11 Previous
studies demonstrated that nebulized lidocaine improved pulmonary function or mitigated bronchoconstriction after anaesthetic induction and intubation in asthmatic patients
when given i.v.12 13 Since m1 and m3 muscarinic receptors
play an important role clinically, modulation of their signalling
may be useful. We aimed to characterize the effects of longterm exposure to LA on muscarinic signalling, and the underlying mechanisms of action, in Xenopus oocytes.
Methods
All animal experiments were approved by the Animal Research
Committee, University of Maastricht, The Netherlands.
Oocyte experiments
Oocyte harvesting and receptor expression were performed as
described previously.6 Briefly, oocytes were obtained from
anaesthetized female Xenopus laevis frogs, defolliculated
and injected with rat m1 and m3 muscarinic acetylcholine receptor complementary RNA. Agonist-induced calcium-activated
chloride currents [ICl(Ca)] were measured using two-electrode
voltage clamping (sample traces are shown in Fig. 1A and B).
Xenopus laevis frogs were housed in an established frog
colony and fed regular frog brittle twice weekly. Surgery for
oocyte harvesting was performed once every 2 months.
Drug administration
Agonists and LA were used at concentrations used in earlier
studies.6 7 Acetyl-b-methylcholine bromide (MCh) was used
as an agonist for the m1 and m3 muscarinic receptors,
diluted in Tyrode’s solution (150 mM NaCl, 5 mM KCl, 1 mM
MgCl2.6H2O, 2 mM CaCl2.2H2O, 10 mM dextrose, 10 mM
HEPES) to the desired concentrations and superfused over
the oocyte for 10 s. The oocyte was placed next to the inflow
tubing to ensure complete exposure to agonists, treatment,
or both. LAs were diluted in Barth’s solution to concentrations
corresponding to 1/10th of the previously determined IC50.6 7
For extracellular administration of LA, oocytes were incubated
for different durations in plain Barth’s solution with or without
LA. Intracellular administration of QX314 was performed as
described previously.4 Control and treatment responses were
obtained in different oocytes to avoid an alteration of measurements by receptor desensitization.
Oligonucleotide injection
Phosphorothioate oligonucleotides were synthesized by the
University of Virginia Research Facility. The antisense sequence
is complementary to specific 20-base segments with ,50%
homology with other types of X. laevis G-proteins.14 Sense oligonucleotides were used as control. Oocytes were injected
with 50 nl sterile water containing 50 ng per cell antisense or
sense oligonucleotides; 24 h after oligonucleotide injection,
the cells were tested as described previously.7
Experiments using PKC or phosphatase inhibitors
The PKC antagonists chelerythrine (CT, 10 mM, targeted to
the substrate binding site of PKC) or bisindolylmaleimide
(BIM, 10 mM a competitive PKC antagonist for adenosine triphosphate binding to the catalytic domain) were used. As
described previously, the rather high concentrations used
were chosen in order to ensure the inhibition of all PKC isoforms
present in the Xenopus oocyte.4 For the inhibition of receptordephosphorylation, we used the phosphoserine/phosphothreonine phosphatase inhibitor okadaic acid (1 mM). Oocytes
were incubated in PKC or phosphatase inhibitors for 1 h
before administration of LA and/or agonists to assure appropriate inhibition. Responses elicited by stimulation of m1 or m3
receptors by MCh (at EC50) in the presence of LA and the PKC
or phosphatase antagonists were normalized to control responses obtained from oocytes that were incubated only with
the corresponding PKC or phosphatase antagonist.
Binding experiments
Membrane preparation and ligand binding studies were performed as described previously.7 In brief, Chinese hamster
ovary (CHO) cells, stably transfected with the m3 muscarinic receptor, were homogenized. Receptor density and equilibrium
dissociation constants in CHO cell membranes were determined by specific binding of [3H] radiolabelled quinuclydinyl
benzylate ([3H]QNB) (0.1– 16 nM), a muscarinic receptor antagonist, and measured by a scintillation counter. Non-specific
binding was identified by the addition of 5 mM atropine to displace specific binding of [3H]QNB. Five different membrane preparations, assayed with the same batch of radioligand, were
used for each time point.
Materials
Molecular biology reagents were obtained from Promega
(Madison, WI, USA), CHO cells (CRL-1982), stably transfected
371
BJA
Picardi et al.
A
B
MCh 0.42 µM
MCh 0.42 µM+lidocaine 370 nM
500 nA
500 nA
10 s
10 s
*
120
% of control response
D
Lidocaine on m1 signalling
*
*
100
*
80
*
Lidocaine on m3 signalling
*
*
60
40
150
% of control response
C
100
*
*
720
2880
*
*
*
50
20
0
0
5
10
30
60
120 240 480 720 1440 2880
Time (min)
10
30
60
120
Time (min)
Fig 1 Modulation of m1 and m3 signalling by extracellular lidocaine in Xenopus oocytes. (A) Sample traces of ICl(Ca) induced by 10 s administration of
MCh at EC50 in an oocyte expressing the m3 receptor and in the presence of lidocaine at IC50. Peak currents are 1.12 mA (A) and 0.48 mA (B). (C) Effects
of long-term exposure to lidocaine (1.8 nM) on m1 signalling (n¼22). Recombinant expressed m1 receptors were stimulated with MCh (0.57 mM).
(D) Lidocaine (37 nM) on m3 responses, elicited by MCh (0.42 mM, n¼25). Effects were determined at different time points and are shown as mean
(SD), responses, normalized to corresponding control responses. *Significance (P,0.05) compared with the control group.
with the rat muscarinic m3 receptor, were purchased from
ATCC (Manassas, VA, USA) and other chemicals were obtained
from Sigma (St Louis, MO, USA). QX314 was a gift from Astra
Pharmaceuticals, L.P. (Westborough, MA, USA).
Statistical analysis
Results are reported as mean and standard deviation (SD). At
least 22 oocytes were used to determine each data point. As
variability between batches of oocytes is common, responses
were normalized to control response. Statistical comparisons
were made using one-way analysis of variance followed by
the Dunnet correction for multiple comparisons. P,0.05 was
considered significant. SigmaStat 2.0 (Jandel Scientific Corporation, San Rafael, CA, USA) was used for all statistical analyses.
Results
Functional expression of m1 and m3 muscarinic
receptors in Xenopus oocytes
Whereas untreated oocytes did not respond to MCh, a transient
increase in ICl(Ca) was seen after MCh stimulation in oocytes
expressing the m1 and m3 muscarinic receptor. Evidence
372
that signalling of these specific receptors was responsible for
responses was shown in earlier studies since atropine
and pirenzepine attenuated muscarinic m1 signalling and
4-diphenylacetoxy-N-methylpiperidine, an m3 preferring cholinergic receptor antagonist, inhibited m3-mediated signal
transduction.15 16
Lidocaine modulates m1 and m3 muscarinic signalling
in Xenopus oocytes biphasically
Incubation of oocytes which recombinantly expressed m1
muscarinic receptors for various durations with lidocaine
(1.8 nM) induced a biphasic effect on m1 signal transduction.
Whereas lidocaine significantly inhibited MCh (0.57 mM)induced responses of the m1 receptor after 30, 60, and 120
min longer exposure to the LA (.8 h) resulted in a small but significant increase in MCh-elicited responses than compared with
control (Fig. 1C). A biphasic effect was also observed when
oocytes expressing the m3 receptor where incubated with lidocaine (37 nM) for .8 h (Fig. 1D). MCh (0.42 mM) stimulated m3
receptor signalling was significantly inhibited by lidocaine
after 30, 60, and 120 min but as shown for the m1 receptor, significantly increased after prolonged exposure to the LA.
BJA
Local anaesthetics and muscarinic signalling
Functional ligand binding of recombinantly expressed
m3 receptors in CHO cells
To exclude the possibility of muscarinic receptor dysfunction as
an effect of time ([3H]QNB), binding to membranes prepared
from CHO cells, stably transfected with the rat m3 muscarinic
receptor, was studied at different time points. Free drugspecific binding was saturable and reached its maximum at
2.2–4.6 nM (Fig. 2). The saturation curves and Scatchard analyses confirm near steady dissociation constants (Kd) and
receptor densities (Bmax) over time, suggesting muscarinic
m3 receptors to be fully functional even after longer study
times in our model (Table 1).
Intracellularly injected QX314 inhibits m1 and m3
signalling in a time- and G-protein dependent manner
Since time-dependent inhibition of different Gaq-coupling
receptors by LA has been shown to have an intracellular site
of action, we studied the effects of the permanently charged
1h
2000
Specific
NS
0.2
0.1
1000
0.0
1000
0
2000 3000
Bound
0
10
5
0
4000
15
20
4000
3000
Specific
0.3
Bound/free
0.3
Bound [3H]QNB (fmol mg–1 protein)
3000
Bound/free
Bound [3H]QNB (fmol mg–1 protein)
Control
4000
2000
NS
0.2
0.1
1000
0.0
0
1000
2000
Bound
0
5
0.2
0.1
1000
0.0
2000 3000
Bound
0
0
5
10
[3H]QNB
4000
15
20
5000
4000
Specific
0.3
Bound/free
NS
Bound [3H]QNB (fmol mg–1 protein)
Specific
0.3
Bound/free
Bound [3H]QNB (fmol mg–1 protein)
3000
Free
3000
2000
1000
NS
0.2
0.1
0.0
0
0
0
1000
4000
Bound
5
10
Free
(nM)
[3H]QNB
6000
15
20
(nM)
8h
5000
4000
Specific
0.3
3000
Bound/free
Bound [3H]QNB (fmol mg–1 protein)
4h
4000
1000
20
10
15
Free [3H]QNB (nM)
2h
0
4000
0
Free [3H]QNB (nM)
2000
3000
2000
1000
NS
0.2
0.1
0.0
0
0
0
5
4000
2000
Bound
10
6000
15
20
Free [3H]QNB (nM)
Fig 2 Functional ligand binding of recombinantly expressed m3 receptors in CHO cells. [3H]quinuclydinyl benzylate ([3H]QNB) (0.1 –16 nM) binding
to membranes prepared from Chinese ovarian hamster cells stably transfected with the rat muscarinic m3 receptor at different time points (n¼5 at
each time point).
373
BJA
Picardi et al.
lidocaine analogue QX314, administered intracellularly. In
contrast to lidocaine, QX314 (96 mM for m1 and 45 mM for
m3) inhibited MCh (0.57 and 0.42 mM, respectively)-induced
m1 and m3 signal transduction in a time-dependent manner
(Fig. 3A and B, respectively) after 48 h.
Table 1 Radioligand binding of recombinantly expressed m3
receptors in Chinese ovarian hamster cells at different time points.
Data are shown as mean and SD. Kd, dissociation constant; Bmax,
receptor density
Time points
Control
1h
2h
4h
8h
Kd (nM)
0.148
(0.03)
0.152
(0.04)
0.152
(0.03)
0.147
(0.03)
0.154
(0.04)
Bmax (fmol
mg21
protein)
3545
(217)
3479
(225)
3657
(199)
4650
(233)
4760
(231)
Effects of extracellularly applied lidocaine and QX314
on m3 receptors in Gaq-depleted oocytes
The effects of lidocaine and QX314 on m1 and m3 signalling
were clearly different. As lidocaine reaches both the intraand extracellular space, this may be due to an extracellular
QX314i on m1 signalling
A
To ascertain if modulation of Gaq-protein function might be a
potential site of action for this QX314-induced time-dependent
effect on muscarinic signalling,4 we next studied the inhibitory
potency of intracellularly injected QX314 (45 mM) on MChelicited m3 responses in oocytes after depletion of Gaq-protein
using antisense oligonucleotides. m3 receptors were used for
antisense studies as they gave more robust receptor expression
after multiple microinjections.4 According to earlier data on LA
effects on LPA signal transduction, depletion of Gaq-protein
fully abolished the observed QX314 effects on muscarinic
signalling at all time points studied (Fig. 3C), indicating that
time-dependent inhibition by intracellular QX314 is critically
dependent on Gaq-protein function.
120
120
100
*
80
*
*
*
*
60
*
*
*
40
% of control response
% of control response
QX314i on m3 signalling
B
100
*
*
80
*
*
60
*
40
20
20
0
0
5
10
30
5
60 120 240 480 720 14402880
Time (min)
QX314i on m3 signalling Gaq-depleted
C
10
30
60 120 240 480 720 14402880
Time (min)
Lidocaineexon m3 signalling Gaq-depleted
D
175
*
*
150
% of control response
% of control response
125
100
75
50
25
0
*
*
*
*
125
100
75
*
*
*
10
20
30
*
*
50
25
10
20
30
45
60 120 240 480 1440 2880
Time (min)
0
45
60 120 240 480 1440 2880
Time (min)
Fig 3 Sites of action for LA effects on muscarinic signalling. Effects of intracellularly injected QX314 (96 mM for m1 and 45 mM for m3) on (A) m1 and
(B) m3 signalling (n¼25 each). Responses were elicited by stimulation with MCh (0.57 mM for m1 and 0.42 mM for m3 receptors). (C) Effects of intracellularly administered QX314 (45 mM) on m3 signalling after depletion of the Gaq-protein (n¼22). (D) Effects of extracellularly applied lidocaine
(37 nM) on m3 responses, induced by MCh in Gaq-depleted oocytes (n¼24). Data are mean (SD), normalized to corresponding control responses.
*Significance (P,0.05) compared with the control group.
374
BJA
Local anaesthetics and muscarinic signalling
possible mechanism and so we investigated the role of PKC in
lidocaine-induced increased muscarinic responses, using two
different PKC antagonists.
In the absence of LA, inhibition of PKC increased m1 and m3
receptor signalling by 25% of control response after 49 h (1 h
pre-treatment and 48 h study period, data not shown). When
both lidocaine (1.8 nM for m1 and 37 nM for m3) and BIM
(10 mM) were present, MCh-induced muscarinic signalling was
inhibited in a time-dependent manner (Fig. 4A and B). Similar
results were obtained using lidocaine and CT (Fig. 4C and D).
target site, accessible to lidocaine but not to the permanently
charged QX314. To confirm this, we studied the effects of
extracellularly administered lidocaine on m3 responses in
Gaq-protein-depleted oocytes. Extracellular lidocaine (37 nM)
inhibited m3 signalling by 20% after the first hour (Fig. 3D).
This attenuation of signalling results from interaction with an
additional extracellular binding site for uncharged LA at
the m3 receptor.7 However, further incubation of Gaqprotein-depleted oocytes with lidocaine (.4 h) enhanced
m3 signalling even more than in the presence of functional
Gaq-protein.
Thus, the observed biphasic effect of LA on muscarinic signalling seems to consist of two opposing parts: an intracellular
time- and Gaq-protein-dependent inhibition and an even
slower extracellular time-dependent signalling enhancement.
Phosphatase antagonism does not affect the increase
in m1 and m3 responses after long-term exposure to
lidocaine
As receptor phosphorylation seems to play an important role in
the biphasic effect of lidocaine on muscarinic signalling, we
next questioned if inhibition of dephosphorylation using the
phosphoserine/phosphothreonine phosphatase inhibitorokadaic
acid resulted in even further enhancement of signalling.
Pre-treatment of oocytes in okadaic acid (1 mM) for 1 h did not
have any effect on MCh-elicited muscarinic responses (data not
PKC antagonism abolishes the lidocaine-induced
time-dependent increase in m1 and m3 signalling
The slow time course of signalling enhancement suggests
covalent modification of the receptor rather than allosteric
modulation. Receptor phosphorylation might be one
A
B
Lidocaine on m1 signalling BIM
100
Lidocaine on m3 signalling BIM
100
*
80
*
60
*
*
*
40
*
*
*
20
% of control response
% of control response
*
*
*
50
*
720
2880
0
0
5
10
30
60
120 240 480 720 1440 2880
10
30
D
Lidocaine on m1 signalling CT
100
*
*
80
*
60
*
40
120
Lidocaine on m3 signalling CT
100
*
*
*
*
*
*
% of control response
C
60
Time (min)
Time (min)
% of control response
*
*
*
50
*
*
720
2880
20
0
0
5
10
30
60 120 240 480 720 1440 2880
Time (min)
10
30
60
120
Time (min)
Fig 4 PKC antagonism modulates the effect of lidocaine on muscarinic signalling. Effects of extracellularly applied lidocaine on (A) m1 and (B) m3
signalling (1.8 and 37 nM, respectively), stimulated by MCh (0.57 mM for m1 and 0.42 mM for m3 receptors) after pre-treatment with BIM (10 mM,
n¼25) or (C and D) CT (10 mM, n¼25), for 1 h. Data are mean (SD) normalized to corresponding control responses. *Significance (P,0.05) compared
with the control group.
375
BJA
Picardi et al.
shown). However, when cells were additionally treated with
lidocaine (1.8 nM), m1 signalling was initially inhibited in a timedependent manner, followed by a maximal increase in signalling up to 150% of control response after 48 h. A similar biphasic
effect in the presence of okadaic acid and lidocaine (37 nM) was
seen for m3 signal transduction (data not shown).
PKC antagonism abolishes the time-dependent
increase in m1 and m3 signalling by extracellular
QX314
To further characterize the site of action for LA to enhance muscarinic signalling after a certain time, we next studied the
effects of the PKC antagonists BIM or CT and/or extracellularly
applied QX314 (500 mM) on MCh-elicited m3 responses. As
shown in Figure 5A and B, 1 h pre-treatment with each PKC antagonist alone increased m3 signalling by 25% for BIM and
QX314ex on m3 signalling BIM
A
35% for CT. With respect to QX314, m3 responses were not
affected for the first 2 h, but increased after 4 h (effects of
QX314 were determined for oocytes of each series of PKC
antagonists). This increase was completely abolished when
one of the PKC antagonists, either BIM or CT, was already
present, suggesting that LA interact with an extracellular site
at the m3 receptor and, as similar results were obtained for
m1 signalling (data not shown), with an extracellular site at
the m1 receptor, accordingly.
Intracellular lidocaine does not enhance m3 signalling
Direct interaction with PKC could be another potential mechanism of LA to time-dependently enhance muscarinic signalling. To confirm this hypothesis, we injected lidocaine
(500 mM) into oocytes expressing the m3 receptor and prevented diffusion to the extracellular space by vigorously
QX314ex on m3 signalling CT
B
QX314ex 500 µM
QX314ex500 µM
140
120
BIM 10 µM
100
80
BIM 10 µM+QX314ex 500 µM
60
40
20
% of control response
% of control response
140
120
CT 10 µM
100
80
CT 10 µM+QX314ex 500 µM
60
40
20
0
0
5
10
30
60
120 240 480 720 1440 2880
Time (min)
5
10
30
60 120 240 480 720 1440 2880
Time (min)
Lidocainei on m3 signalling Gaq-depleted
C
% of control response
125
100
75
50
25
0
10
20
30
45
60 120 240 480 1440 2880
Time (min)
Fig 5 PKC-dependent modulation of muscarinic m3 signalling by extra- and intracellular LAs. Time course of m3 responses after incubation with
BIM (10 mM) or CT (10 mM), QX314 (500 mM) or the combination of QX314 (500 mM), and BIM or CT. (A) Data obtained in the presence of BIM (n¼25)
and (B) data when CT was used (n¼25). Oocytes were pre-treated with PKC antagonists for 1 h, followed by incubation with QX314 for 48
h. Responses were elicited by MCh (at EC50). (C) Effects of intracellularly applied lidocaine on m3 signalling in Gaq-protein-depleted oocytes
(n¼24). Receptors were stimulated with MCh (at EC50). Data are mean (SD) normalized to corresponding control responses.
376
Local anaesthetics and muscarinic signalling
superfusing the oocyte with Tyrode’s solution. To furthermore
prevent lidocaine-induced modulation of Gaq-protein function, we used oocytes which had been depleted of intact
Gaq-protein by antisense oligonucleotides. As expected, intracellular lidocaine did not have any effects on m3 signalling, as
shown in Figure 5C.
Discussion
This study shows that m1 and m3 muscarinic signalling is timedependently modulated by LA. However, muscarinic responses
are affected by two opposing mechanisms: an initial inhibition
is mediated by an intracellular site of action, requiring functional Gaq-protein, followed by enhanced signalling based
on interference with an extracellular receptor domain modulated by PKC and protein phosphatase activity, which masks
the inhibitory potencies of LA.
Many clinical trials have demonstrated beneficial antiinflammatory effects of LA perioperatively and we previously
showed that LA inhibit signalling of several Gq/11 class GPCRs,
a G protein family that predominantly mediates haemostatic
and inflammatory signalling, in a time- and Gaq-dependent
manner.4
Besides modulation of immune responses, LAs were also
shown to positively affect bronchial hyperreactivity and post
operative cognitive function.10 12 13 Muscarinic m1 receptors,
found in the central nervous system, are involved in processing
memory and learning. Muscarinic m3 receptors, expressed in
smooth muscle cells of the respiratory tract, mediate bronchoconstriction and thus bronchiolar tone. Both may be a potential
target site for LA.9 11 Since both receptors couple to Gaq-protein,
we anticipated their signalling to be inhibited by LA in a similar
manner to the other GPCRs, and thus the observed increase in
signalling was rather unexpected.
Muscarinic receptors are known to be regulated by PKCactivity.
Whereas phosphorylation results in desensitization of receptor
function, inhibition of PKC enhances muscarinic responses.17
In line with Shiga and colleagues,18 we observed an increase
in MCh-elicited signals by 25% in oocytes treated with PKCantagonists, BIM, or CT, when compared with control, suggesting
PKC involvement.
However, pre-treating oocytes with PKC antagonists before incubation with lidocaine prevented the late increase in muscarinic
responses, and unmasked the underlying time-dependent inhibition of muscarinic signalling by LA. These findings suggest that
receptor phosphorylation, dephosphorylation, or both play a
key role in the biphasic effect of lidocaine on muscarinic signalling. Confirmatory data were obtained when okadaic acid was
used. If inhibition of PKC was relevant for restoring timedependent action of LA, block of receptor dephosphorylation
should increase orat least maintain the biphasic LA-effect. Inhibition of protein phosphatases did not affect muscarinic signalling
itself, but resulted in a similar biphasic time course of inhibition by
LA, as expected.19
Thus, up-regulation of receptor function after prolonged exposure to LA seems to be because of PKC activity and receptor
phosphorylation. As PKC is located intracellularly, we assumed
BJA
a direct interaction of LA with the enzyme. However, intracellular injection of QX314 did not have any stimulating effects on
muscarinic signalling, suggesting an extracellular site of
action for the biphasic LA effect, and no direct interactions
with PKC. Since QX314 is a permanently charged lidocaine analogue, the lack of effect did not exclude an intracellular binding
site for uncharged LA, although as intracellular administration
of lidocaine in oocytes depleted of functional Gaq-protein did
not have any effect on muscarinic signalling, this possibility
was dismissed.
As in our earlier studies, we used the Xenopus oocyte model
for our studies. Although the cells have a long survival time
in vitro and are useful to study intracellular actions,2 6 7 20
there are several potential problems with the technique.
Actual PKC activity is often hard to determine in Xenopus
oocytes and so altered calcium-activated chloride currents
were used as a surrogate in the presence of PKC inhibitors.
Although recombinant expression of a single subtype of
muscarinic receptors allows for mechanistic studies due to
easy separation of signalling pathways, it is rather difficult
to make a statement on how lidocaine-induced modulation
of G proteins other than Gaq/11 would affect the effects on
muscarinic m1 and m3 receptors. We have previously shown
that LA enhances signalling of the Gai coupling human adenosine 1 receptor, leading to a decrease in cAMP levels.21
However, even in the presence of muscarinic m2 and m4 receptors, lidocaine will most likely modulate m1 and m3 signalling
as described above, since both receptors primarily couple to
Gaq/11.
Other LAs are likely to affect muscarinic signalling in the
same way. However, as the efficacy of LA to modulate GPCR signalling depends on their physicochemical properties, further
studies are required to determine the size of effect of other
LAs on muscarinic m1 and m3 signal transduction.22
Taken together, our data suggest that prolonged exposure
of muscarinic m1 and m3 receptors to LA may result in a
change of receptor confirmation: new phosphorylation sites,
usually not accessible for PKC, may become available and as
a consequence, these receptors are phosphorylated by basal
PKC activity. Under these circumstances, PKC may induce
up-regulation of receptor function instead of the expected
down-regulation (Fig. 6). Direct interaction between LA and
PKC may not be obligatory to result in the biphasic effect on
muscarinic signalling observed after prolonged exposure.
In conclusion, LAs affect muscarinic signalling via PKCdependent mechanisms, yet without direct interference with
the enzyme. One important finding of the present study is
that LAs do not have to directly modulate PKC in order to
induce PKC-mediated effects. In addition, PKC activity does
not always induce predictable effects under these circumstances. Based on these data, not all GPCRs that signal
through Gaq may be time-dependently inhibited by LA. Prolonged exposure to LA may not result in prolonged inhibition
of muscarinic signalling. For clinical practice, this may mean
that prolonged exposure to LA may not benefit pulmonary
function, but it may benefit cognitive deficits. However, this
remains to be determined in clinical trials.
377
BJA
Picardi et al.
A
LA
SS
T
S
+
P
+
P
PKC
+
PKC
P
Muscarinic signalling ↓↓↓
B
LA
LA
S
+
PKC
T
S
P
P
+
+
PKC
P
Muscarinic signalling
Fig 6 Possible mechanism of action for LA-induced biphasic effect on muscarinic signalling. Simplified scheme of muscarinic m1 and m3 receptors.
LA, binding site for LAs on an extracellular receptor domain; PKC, protein kinase C; T , S , threonine/serine amino acids; and P , phosphate molecule. (A) Muscarinic receptor before long-term exposure to LAs. The receptor is in a balanced state of stimulation and inhibition. Serine/
threonine-rests are available for PKC, leading to attenuation of muscarinic signalling. (B) Muscarinic receptor after long-term exposure to LAs
(.2 h). LAs inhibit muscarinic signalling Gaq-protein dependently for the first 2 h. At the same time, LAs interact with an extracellular receptor
domain, resulting in a change of intracellular receptor conformation. Further serine/threonine amino acids become available for phosphorylation
and after the initial inhibition receptor-function is up-regulated by LAs.
Authors’ contributions
S.P.: conduction of experiments, data analysis, and manuscript
preparation. K.H.: data analysis. M.E.D.: study design and
manuscript preparation. M.F.S.: manuscript preparation. P.L.:
manuscript preparation. M.W.H.: study design, conduction of
experiments, data analysis, and manuscript preparation.
Anaesthesiology, University of Heidelberg, Heidelberg, Germany)
for his support.
Declaration of interest
None declared.
Funding
Acknowledgement
We gratefully acknowledge Prof. Dr Eike Martin, MD, PhD
(Professor of Anaesthesiology and Chair, Department of
378
This work was supported by a M.D.-medical research trainee
grant from the Netherlands Organisation for Health Research
and Development (ZonMW AGIKO to S.P.), by the Medical
Local anaesthetics and muscarinic signalling
Faculty, University of Heidelberg, Germany (F206639) and by
Institutional funds from the Departments of Anaesthesiology,
University of Heidelberg, Germany and Academic Medical
Center, Amsterdam, The Netherlands.
References
1 Hollmann MW, Durieux ME. Local anesthetics and the inflammatory
response: a new therapeutic indication? Anesthesiology 2000; 93:
858– 75
2 Hollmann MW, Wieczorek KS, Berger A, Durieux ME. Local anesthetic
inhibition of G protein-coupled receptor signaling by interference
with Galpha(q) protein function. Mol Pharmacol 2001; 59: 294– 301
3 Wettschureck N, Offermanns S. Mammalian G proteins and their cell
type specific functions. Physiol Rev 2005; 85: 1159–204
4 Hollmann MW, Herroeder S, Kurz KS, et al. Time-dependent inhibition of G protein-coupled receptor signaling by local anesthetics.
Anesthesiology 2004; 100: 852– 60
5 Herroeder S, Pecher S, Schonherr ME, et al. Systemic lidocaine shortens length of hospital stay after colorectal surgery: a doubleblinded, randomized, placebo-controlled trial. Ann Surg 2007;
246: 192–200
6 Hollmann MW, Fischer LG, Byford AM, Durieux ME. Local anesthetic
inhibition of m1 muscarinic acetylcholine signaling. Anesthesiolgy
2000; 93: 497– 509
7 Hollmann MW, Ritter CH, Henle P, de Klaver M, Kamatchi GL,
Durieux ME. Inhibition of m3 muscarinic acetylcholine receptors
by local anesthetics. Br J Pharmacol 2001; 133: 207– 16
8 Scholz A, Kuboyama N, Hempelmann G, Vogel W. Complex blockade
of TTX-resistant Na+ currents by lidocaine and bupivacaine reduce
firing frequency in DRG neurons. J Neurophysiol 1998; 79: 1746– 54
9 Robinson L, Platt B, Riedel G. Involvement of the cholinergic system
in conditioning and perceptual memory. Behav Brain Res 2011; 221:
443– 65
10 Wang D, Wu X, Li J, Xiao F, Liu X, Meng M. The effect of lidocaine on
early postoperative cognitive dysfunction after coronary artery
bypass surgery. Anesth Analg 2002; 95: 1134– 41
BJA
11 Racke K, Juergens UR, Matthiesen S. Control by cholinergic mechanisms. Eur J Pharmacol 2006; 533: 57–68
12 Groeben H, Silvanus MT, Beste M, Peters J. Combined lidocaine and
salbutamol inhalation for airway anesthesia markedly protects
against reflex bronchoconstriction. Chest 2000; 118: 509– 15
13 Hunt LW, Frigas E, Butterfield JH, et al. Treatment of asthma with
nebulized lidocaine: a randomized, placebo-controlled study.
J Allergy Clin Immunol 2004; 113: 853– 9
14 Shapira H, Amit I, Revach M, Oron Y, Battey JF. Ga14 and Gaq mediate
the response to trypsin in Xenopus oocytes. J Biol Chem 1998; 273:
19431– 6
15 Nietgen GW, Hoenemann CW, Chan CK, Kamatchi GL, Durieux ME.
Volatile anaesthetics have differential effects on recombinant m1
and m3 muscarinic acetylcholine receptor function. Br J Anaesth
1998; 81: 569–77
16 Durieux ME. Halothane inhibits signaling through m1 muscarinic
receptors expressed in Xenopus oocytes. Anesthesiology 1995;
82: 174– 82
17 Hosey MM. Diversity of structure, signaling and regulation within
the family of muscarinic cholinergic receptors. FASEB J 1992; 6:
845– 52
18 Shiga Y, Minami K, Shiraishi M, et al. The inhibitory effects of tramadol on muscarinic receptor-induced responses in Xenopus oocytes
expressing cloned M(3) receptors. Anesth Analg 2002; 95: 1269– 73
19 Do SH, Kamatchi GL, Durieux ME. The effects of isoflurane on native
and chimeric muscarinic acetylcholine receptors: the role of protein
kinase C. Anesth Analg 2001; 93: 375–81
20 Hoenemann CW, Nietgen GW, Podranski TA, Chan CK, Durieux ME.
Influence of volatile anesthetics on thromboxane A2 signaling.
Anesthesiology 1998; 88: 440– 51
21 Benkwitz C, Garrison JC, Linden J, Durieux ME, Hollmann MW. Lidocaine enhances Galphai protein function. Anesthesiology 2003; 99:
1093– 101
22 Picardi S, Cartellieri S, Groves D, et al. Local anesthetic-induced inhibition of human neutrophil priming: the influence of structure,
lipophilicity, and charge. Reg Anesth Pain Med 2012; 38: 9–15
Handling editor: H. F. Galley
379