Description
answer this about the article. – the topic of article, journal and author, country, year when published.- the main results, insights and research methods of the publication (that is is understandable). Explain the research in more detail. – Restate the authors’ take-home message followed by your own interpretation of the study. Why do you find this article interesting or not, is it important in clinical practice?one page, understandable
Unformatted Attachment Preview
Cough
BioMed Central
Review
Open Access
Central and peripheral mechanisms of narcotic antitussives:
codeine-sensitive and -resistant coughs
Kazuo Takahama* and Tetsuya Shirasaki
Address: Department of Environmental and Molecular Health Sciences, Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1
Oe-Honmachi, Kumamoto 862-0973, Japan
Email: Kazuo Takahama* – [email protected]; Tetsuya Shirasaki – [email protected]
* Corresponding author
Published: 9 July 2007
Cough 2007, 3:8
doi:10.1186/1745-9974-3-8
Received: 4 December 2005
Accepted: 9 July 2007
This article is available from: http://www.coughjournal.com/content/3/1/8
© 2007 Takahama and Shirasaki; licensee BioMed Central Ltd.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
Narcotic antitussives such as codeine reveal the antitussive effect primarily via the µ-opioid
receptor in the central nervous system (CNS). The κ-opioid receptor also seems to contribute
partly to the production of the antitussive effect of the drugs. There is controversy as to whether
δ-receptors are involved in promoting an antitussive effect. Peripheral opioid receptors seem to
have certain limited roles. Although narcotic antitussives are the most potent antitussives at
present, certain types of coughs, such as chronic cough, are particularly difficult to suppress even
with codeine. In guinea pigs, coughs elicited by mechanical stimulation of the bifurcation of the
trachea were not able to be suppressed by codeine. In gupigs with sub-acute bronchitis caused by
SO2 gas exposure, coughing is difficult to inhibit with centrally acting antitussives such as codeine.
Some studies suggest that neurokinins are involved in the development of codeine-resistant coughs.
However, evidence supporting this claim is still insufficient. It is very important to characterize
opiate-resistant coughs in experimental animals, and to determine which experimentally induced
coughs correspond to which types of cough in humans. In this review, we describe the mechanisms
of antitussive effects of narcotic antitussives, addressing codeine-sensitive and -resistant coughs,
and including our own results.
Introduction
Cough causes via the activation of cough reflex arc consisted of the airway vagal afferent nerves, cough center in
the medulla and the efferent nerves. Inhibiting it at any
site of the arc can be expected to cause antitussive effect.
However, the mechanisms of cough generation, its modulation and antitussive effect of centrally and peripherally
acting antitussives are still largely unclear. Of the many
available narcotic and non-narcotic antitussives, the most
effective are the narcotic antitussives, which are of limited
use due to their inherent undesirable side effects, particularly their narcotic side effects. Even for this codeine, it has
recently been pointed that it is not effective as estimated
from the experimental results in guinea pigs [1]. Also,
chronic coughs are often resistant to treatment with
codeine. Thus, there is a need for new types of antitussives
that can suppress chronic coughs. It is unclear why some
coughs, such as chronic cough, are resistant even to treatment with potent antitussives such as codeine, although it
is known that coughing is a neural reflex. In this review,
we discuss the mechanisms of the effects of narcotic antitussives on coughing using experimental animals, and further, the resistance of coughs to narcotic antitussives,
describing our recent findings regarding codeine-sensitive
and -insensitive coughs in guinea pigs.
Page 1 of 8
(page number not for citation purposes)
Cough 2007, 3:8
Opioid receptor subtypes and antitussive effects
The antitussive mechanisms of narcotic antitussives are
not fully understood. The available evidence clearly indicates that narcotic antitussives act on opioid receptors [24]. Binding studies concerning guinea-pig and human
opioid receptors demonstrated that codeine and dihydrocodeine, gold standard narcotic antitussives, were
more selective to the µ-opioid receptor than other κ- or δopioid receptors [3,5]. Ki value of [3H]codeine (3.7 × 10-7
M) for replacement of [3H]-D-Ala2, MePhe4, Gly-ol5]
enkephalin (DAMGO), a µ-selective ligand, in guinea pigs
[3] was close to the KD value of 5.6 × 10-7 M for the saturable binding of [3H]codeine in the lower brain stem of
guinea-pigs [6]. µ-Selective morphine has much more
potent antitussive activity in cat [7]. κ-Agonists also have
antitussive activity. Therefore, both µ- and κ-opioid receptors have been considered as candidates for being the
receptors which contribute to antitussive activity.
Further, pharmacological studies carried out by using rats,
mice and µ1-opioid receptor deficient mice suggested that
µ2- rather than µ1-subtype of the µ-opioid receptor contributes to the antitussive activity of opioids [8,9]. Unfortunately, there is argument against these results in mice
and rats, because it has been unable to reliably obtain a
cough-like behavior in mice and rats. In addition, Ohi et
al. [10] recently found that the motor patterns of rats and
guinea pigs during cough-producing stimuli were significantly different. In rats, two different types of behavior
were observed and one of them did not conform to the
classic definition of a cough. Codeine suppressed both
behaviors. For these reasons, it has been addressed that
rats and mice are not viable as models of cough.
This issue seems to cast its shadow over the conflicting
results about the role of δ-opioid receptors in producing
the cough inhibiting effect of narcotic antitussives. Kamei
et al [11] demonstrated that [D-Pen2,5]enkephalin
(DPDPE), a selective δ-agonist, did not have an antitussive effect, but rather inhibited the antitussive effects of
DAMGO and K-50488H, a selective κ opioid receptor agonist found in rats. But, δ-antagonists such as naltrindole
and naltriben reduced the number of capsaicin-induced
coughs in mice and rats [12,13]. Mu- and κ-opioid receptor antagonists did not antagonize the δ-antagonistinduced antitussive activities. Conversely, Kotzer et al. [5]
showed that the highly selective δ-agonist SB 227122
inhibited the cough-reflex induced by citric acid in
guinea-pigs. The antitussive effect of SB227122 was antagonized by the δ-antagonist SB 244525. This δ-antagonist
itself did not have an antitussive effect. Kotzer et al. have
also reported that naltrindole binds to human µ- and κopioid receptors at significant levels. Further studies are
required to confirm whether the controversy presented
above comes from differences in the species of experimen-
http://www.coughjournal.com/content/3/1/8
tal animals used and/or differences in the pharmacological properties of each δ receptor agonist and antagonist
used.
Apart from the above, we have recently found evidence for
another possible mechanism of the antitussive effects of δantagonists. In a patch clamp study using single brain
neurons, naltrindole and naltriben both inhibited the currents caused by activation of G-protein-coupled inwardly
rectifying K+ (GIRK) channel [14]. GIRK channels couple
to the 5-HT1A receptor, and contribute to a negative feedback mechanism of 5-HT release. Dextromethorphan,
which is a representative non-narcotic antitussive and has
an inhibitory effect on GIRK channel activated currents
[15], antagonized the 5-HT-induced hyperpolarization
and depolarized the membrane potential, generating
action potentials in dorsal raphe neurons. Thus, inhibition of this channel may increase the 5-HT level in the
CNS. In human volunteers, infusion of 5-HT or its precursor reduced cough responses to a chloride-deficient solution [16]. In contrast, reduction of 5-HT levels has been
found to inhibit the antitussive effects of narcotic and
non-narcotic antitussives [17]. Stimulation of raphe
nuclei depresses discharges in inspiratory motoneurons
[18,19]. The 5-HT1A receptor agonist inhibited cough
responses, although it stimulated cough response at high
doses [20]. 5-HT2/5-HT1 receptor antagonists inhibited
any morphine-induced antitussive effect in humans [21].
In addition, DMGO increased 5-HT efflux in dorsal raphe
nucleus [22]. Taken together, the above findings suggest
that antitussive effects of δ-antagonists are at least partly
due to the inhibition of GIRK channel currents [23].
Next, we will discuss the site of antitussive action of opioids in the CNS. Results of in vivo experiments suggest that
centrally acting antitussives primarily act on the brainstem cough center. Recently, Gestreau et al. [24] reported
that fictive cough selectively increased Fos-like immunoreactivity (FLI) in the interstitial and ventrolateral subdivision of the nucleus tractus solitarius (NTS), the reticular
formation (the medial part of the lateral tegmental field,
and the internal division of the lateral reticular nucleus),
the ambigual complex (the nucleus retroambiguus, the
para-ambigual region, and the retrofacial nucleus), and
the medial parabrachial nucleus in cat. In all the nuclei,
codeine significantly reduced the increase in FLI. Further,
laryngeal afferent stimulation enhanced FLI in periaqueductal gray matter (PAG) and dorsal raphe nucleus in cat
[25].
µ-Opioid receptors are expressed intensely or moderately
in the ambiguus nucleus, NTS, dorsal vagal nerve nucleus,
medial parabrachial nucleus, PAG and raphe nuclei [2629]. In these regions, κ-opioid receptors are also expressed
with similar or less potent density. δ-Opioid receptors are
Page 2 of 8
(page number not for citation purposes)
Cough 2007, 3:8
generally less abundant in the brainstem, but the pneumotaxic center, including the nucleus parabrachialis, contains a very high density of δ-binding site. In the NTS and
ambiguus nucleus, it is expressed weakly. Here, caudal
NTS and its neighboring ventromedial region has been
considered as a strong candidate for being the cough
center, because this region primarily receives sensory
input from the lower airway [30,31] and its stimulation
causes cough-like response [32,33]. The NTS is more
heavily labeled by the µ-ligand than by the κ-ligand in
guinea pigs and cats [29,34]. Further, the µ2 sites have
been found to be associated with respiratory depressant
effects of opioids, whereas the µ1 sites have been found to
be associated with the analgesic effects of opioids in
mouse brain [35]. Microinjection of codeine into the NTS
inhibited a fictive cough reflex in guinea pigs [36]. µ-Opioid receptor agonist presynaptically inhibited excitatory
postsynaptic currents in the NTS [37]. Kappa- and δ-opioid receptor agonists also inhibited excitatory postsynaptic potentials in the NTS but they are less effective than µopioid receptor agonist [38].
Given these together with the reported affinity of narcotic
antitussives for opioid receptors, narcotic antitussives
might have a primary site of antitussive effect on µ-opioid
receptors in the NTS, although there is a report that the
antitussive effects of codeine are not blocked by naloxone
in cats [38]. In addition to the NTS, the raphe nuclei may
be a candidate for being the site of action of narcotic antitussives, since stimulation of the raphe nuclei depresses
the reflex activity caused by stimulation of the superior
laryngeal or vagal nerve in respiratory interneurons of the
NTS, without affecting respiratory rhythm [39]. This characteristic seems to be in accordance with properties that
antitussives are presumed to have.
Peripheral opioid receptors and antitussive
effects
Mu-opioid receptors are located in both the central and
peripheral nervous systems. Adcock [40] has written a
nice review about the sensory opioid receptor and antitussive activity of narcotic antitussives. Inhalations of nebulized codeine, morphine and a peripherally acting
specific µ-opioid receptor agonist produced antitussive
effects in guinea pigs [41,42]. Therefore, it is plausible that
inhaled opioid antitussives exert their effect by inhibiting
tachykinergic transmission of excitatory non-adrenergic
non-cholinergic (eNANC) nerves via a blockade of µ-opioid receptors in the airway, although it is unknown
whether opioids affect peripheral opioid receptors when
administered via conventional routes.
In addition to the effect on sensory fibers, opioid agonists
also appear to inhibit airway cholinergic transmission
[43-45]. Opioid-induced inhibition of the cholinergic
http://www.coughjournal.com/content/3/1/8
bronchoconstriction induced by electric field stimulation
(EFS) in guinea-pig is caused partly by an inhibitory
action on the eNANC nerve, and partly by a direct effect
on cholinergic transmission [44]. The inhibitory effect of
µ-opioid ligands on EFS-induced cholinergic contraction
of the airway’s smooth muscle was also found in human
preparation. This effect is presumably caused by inhibiting the acetylcholine release from the postganglionic parasympathetic nerve fibers [45]. Here, controversial
opinions exist as to whether the airway contraction
induces a cough response or not. However, it has been
known that coughing in patients with cough variant
asthma [46] is inhibited by bronchodilators such as
adrenergic β2 stimulants [47]. This fact seems to indicate
that the kind of cough such as that found in cough variant
asthma may be caused by smooth muscle contraction in
the airway.
Postganglionic parasympathetic nerve fibers in the airway
arise from the paratracheal ganglia (PTG). Their excitability is controlled by the preganglionic neurons via central
vagal reflex. In addition, they can be modulated by a
peripheral reflex mechanism because the collateral
branches of neurokinin-containing C-fibers project to the
PTG neurons [48] and enhance cholinergic transmission
in the PTG, probably via neurokinin releases [49]. Further,
we have recently found that bradykinin inhibits the Mtype K+ current in the acutely dissociated PTG neurons of
rats, causing depolarization and action potential generation [50]. In addition, bradykinin potentiated nicotinic
ACh currents in PTG neurons [51]. Thus, the PTG are
thought to be not only a relay neuron of the parasympathetic nerve, but also integrative sites for the neuromodulation of normal airway function and important for
pathogenesis in airway inflammation. Interestingly, ophiopogonin-D, an active constituent of bakumondo-to, a
Chinese herbal medicine, hyperpolarized the membrane
potential via activation of the K+ current, reducing the cell
excitability of PTG neurons [52]. Bakumondo-to is found
to be effective for treating clinically chronic coughs
[53,54], and to inhibit codeine-registrant coughs as that
expressed in the experimental model described above
[55,56]. Therefore, we speculate that the excitability of
PTG neurons may contribute to pathological condition,
including some kinds of chronic cough. In this context,
we examined the effects of codeine in dissociated PTG
neurons. However, codeine did not induce any currents in
PTG neurons, and had no effect on high-voltage-activated
(HVA) Ca2+, bradykinin- induced or nicotine-induced
currents in the neurons. Bradykinin-induced potentiation
of nicotinic currents in the neurons was also not affected
by codeine (unpublished data).
To summarize this section, µ-opioid receptors locate in
the airway vagal sensory neurons and, at least inhaled opi-
Page 3 of 8
(page number not for citation purposes)
Cough 2007, 3:8
oids, inhibit both eNANC nerve activity and cholinergic
contraction of smooth muscles through acting on µ-opioid receptors. The PTG neurons seem to be a possible target for peripherally acting antitussives. However, opioids
have no effect on the PTG neurons.
Codeine-sensitive and -resistant coughs
The larynx is the most sensitive site for elicitation of the
cough reflex by mechanical stimulation, followed by tracheal bifurcation and the lower half of the trachea, in that
order [57]. We have recently found that coughs elicited by
mechanical stimulation of the tracheal bifurcation were
relatively resistant to suppression by codeine in guinea
pigs, whereas mechanically induced coughs in the trachea
close to the larynx were effectively inhibited by codeine
[58].
Sensory receptors in airway vagal afferents have been classified into 5 groups, which include rapidly adapting receptors (RARs), Aδ-nociceptors and bronchial C-fiber
receptors. These 3 receptors listed above appear to contribute to cough responses. RARs are myelinated Aδ fibers
and have a low threshold for mechanical stimuli, but are
resistant to chemical stimuli. Conversely, Aδ-nociceptors
and unmyelinated C-fiber receptors have a high threshold
for mechanical stimuli, but a low threshold for chemical
stimuli such as bradykinin and capsaicin. Recently, a 6th
receptor group called “cough receptor” has been identified [59]. Its properties are similar to those of RARs, but
they have a slower conduction velocity and did not
respond to stretching. In the larynx and upper trachea,
“cough receptors” appear to play a primary role in regulation of the cough response [59]. Research by Widdicombe
[57] indicated that the larynx and the tracheal bifurcation
are abundantly innervated by RARs which presumably
include “cough receptors”. Conversely, chemoreceptors
involved in cough responses are mainly distributed in the
lower trachea, particularly around the tracheal bifurcation. Thus, differences in codeine resistivity between areas
of the lower airway may be due to differences in the distribution of these various types of sensory fibers.
In guinea pigs, the effects of codeine on mechanically elicited coughs at each lower airway site were strengthened by
repeated treatment with large doses of capsaicin [58]. This
capsaicin treatment caused degeneration and dysfunction
of the C- and Aδ-nociceptors [60-63] and consequently
reduced cough generation caused by citric acid and capsaicin, but not coughs caused by nicotine or mechanical
stimulation [64]. In addition, angiotensin-converting
enzyme inhibitors (ACEIs), which sensitize nociceptive
fibers, induced codeine-resistant chronic cough in conscious guinea pigs [65]. Consequently, it has been
hypothesized that coughs mediated by nociceptive fibers
were resistant to codeine. In our own study, capsaicin
http://www.coughjournal.com/content/3/1/8
administered topically to the tracheal bifurcation caused a
cough response that was resistant to codeine, whereas topical application to the larynx side of the trachea did not
cause a cough response [66]. In a preliminary histochemical study using guinea pigs, we found that substance P
(SP)-like immunoreactivity is lower in the larynx side of
the trachea than in the tracheal bifurcation. In addition,
the density of SP-immunoreactive nerves has been found
to be significantly higher in patients with cough-variant
asthma than in normal subjects and patients with classic
asthma [67]. The above findings support the hypothesis
that coughs mediated by nociceptive fibers may be resistant to codeine treatment.
In a chronic bronchitis model of rats produced by SO2 gas
exposure, SP content in the trachea was elevated [68]. In a
similar model using guinea pigs, codeine did not inhibit
the cough responses elicited by mechanical stimulation of
the larynx side of the trachea or the tracheal bifurcation.
Epithelial shedding was not observed, but neutral
endopeptidase (NEP) levels and NEP activity in the trachea and bronchus were significantly lower than those of
normal guinea pigs. NEP degrades a variety of peptides,
including bradykinin, SP and other tachykinins [69]. At
high doses, a NEP inhibitor elicits a cough response in
normal guinea pigs [70]. Based on findings that bradykinin and tachykinins are potent inflammatory mediators,
and that neurokinins such as SP are released from C-fiber
(eNANC nerve) terminals, it has been suggested that
coughing induced by inflammatory peptides is resistant to
codeine. However, codeine has been found to significantly suppress the cough response induced or enhanced
by NEP inhibitors [70]. In addition, opioids peripherally
inhibit tachykinergic transmission in the guinea pig bronchus [71-73]. Thus, it appears that NEP inhibition or tachykinin release from peripheral C-fiber terminals is not
sufficient to explain mechanisms of induction of codeineresistant cough. However, in a preliminary study, we
found that inhaled neurokinin A caused codeine-resistant
cough in guinea pigs. Also, in that study, co-administration of codeine and an antagonist for the neurokinin 2
(NK2) receptor almost abolished citric acid-induced
coughing in conscious guinea pigs, in spite of the fact that
citric acid-induced coughs were hard to completely inhibit
even with high doses of codeine, when codeine alone was
given (Fig. 1).
In summary, evidence suggests that RAR or “cough receptor”-mediated coughs are sensitive to codeine but coughs
triggered by neurokinin-containing nociceptive nerves are
resistant to it. In support of this suggestion, there is a finding that the expression of transient receptor potential
vanilloid-1 (TRPV-1) is increased in the airway nerves of
patients with chronic cough [74]. In addition to TRPV-1,
it has recently been reported that acid sensing ion chan-
Page 4 of 8
(page number not for citation purposes)
Cough 2007, 3:8
http://www.coughjournal.com/content/3/1/8
Codeine
Codeine + SR 48,968
Number of cough
(% of pre-administration)
120
100
80
60
40
P
Purchase answer to see full
attachment
