The cold-sensitive cation channel TRPM8 is a target for menthol, which is used routinely as a cough suppressant and as an additive to tobacco and food products. Given that cold temperatures and menthol activate neurons through gating of TRPM8, it is unclear how menthol actively suppresses cough. In this study we describe the antitussive effects of (−)-menthol in conscious and anesthetized guinea pigs. In anesthetized guinea pigs, cough evoked by citric acid applied topically to the tracheal mucosa was suppressed by menthol only when it was selectively administered as vapors to the upper airways. Menthol applied topically to the tracheal mucosa prior to and during citric acid application or administered continuously as vapors or as an aerosol to the lower airways was without effect on cough. These actions of upper airway menthol treatment were mimicked by cold air delivered to the upper airways but not by (+)-menthol, the inactive isomer of menthol, or by the TRPM8/TRPA1 agonist icilin administered directly to the trachea. Subsequent molecular analyses confirmed the expression of TRPM8 in a subset of nasal trigeminal afferent neurons that do not coincidently express TRPA1 or TRPV1. We conclude that menthol suppresses cough evoked in the lower airways primarily through a reflex initiated from the nose.
cold temperatures are sensed by subsets of cold-sensitive mammalian primary afferent neurons through activation of the cation channel TRPM8 (1, 20, 28). TRPM8-expressing neurons constitute a functionally distinct population of neurons specifically dedicated to innocuous cold sensing. Menthol also activates TRPM8, an effect that likely accounts for the cooling sensations associated with topical menthol application. Due in large part to its cooling and soothing effects on mucosal surfaces, menthol has been employed for centuries in medical applications and in consumer products. One of the most common uses of menthol is for the symptomatic treatment of upper and lower airway diseases with cough as a chief complaint (11, 48).
In addition to its therapeutic applications, menthol is also used to enhance the flavor of tobacco. The soothing sensation of menthol in cigarettes has been scrutinized in recent years, because it may disguise the irritant effects of the tobacco smoke, facilitating and rewarding the behavior and thus rendering smoking cessation efforts more difficult (14, 49, 50, 52). Indeed, Wise and colleagues described a counterirritant effect of menthol in the human nose (52), whereas Willis and co-workers (49) recently observed that menthol acts as a counterirritant in mice through activation of TRPM8, providing objective evidence that the pharmacological action of inhaled menthol vapor may reduce the perception of cigarette smoke–induced irritation.
Cough is triggered by airway irritation due to cigarette smoke but may also be suppressed by menthol. The mechanisms of menthol antitussive action are unknown but may include effects on respiratory tract secretion, increased mucosal blood flow, relief of bronchoconstriction, and/or direct or indirect effects on cough-initiating sensory neuronal pathways (29, 49, 51, 52). It should be noted, however, that antitussive effects of menthol are not a universal finding. Cough suppression has been reported in two studies performed by Morice and colleagues (22, 30), and there was perceived benefit in children with nocturnal cough receiving menthol containing vapor rub treatment (37), but Kenia et al. (18) documented no effect of inhaled (−)-menthol on the cough reflex in children, whereas Haidl and colleagues (13) observed no antitussive effects of a 1% (∼30 mM) menthol aerosol prior to bronchoscopy. Perhaps route of administration is critical to the antitussive actions of menthol.
The ability of menthol to activate sensory nerves is difficult to reconcile with cough suppression, at least with a mechanism directly targeting the sensory nerves implicated in cough. Rather, the data suggest an inhibitory effect resulting from the activation of a separate subset of sensory nerves. Excitatory and inhibitory effects of pulmonary and extrapulmonary sensory nerves on the cough reflex have been described, including excitatory effects of activating TRPA1- and/or TRPV1-expressing nasal afferent nerves in animals and in healthy human volunteers (5, 38, 39), and inhibitory effects induced by activation of pulmonary C-fibers (44, 45). In the present study we addressed the hypothesis that menthol suppresses cough secondary to activating subsets of afferent neurons expressing the cation channel TRPM8. We present molecular, physiological, and pharmacological evidence that the antitussive effects of menthol occur secondary to the activation of TRPM8+/TRPV1− nasal trigeminal afferent neurons.
All experiments were performed on male Dunkin Hartley guinea pigs (250–350 g, pathogen free; Harlan, Boxmeer, the Netherlands), and all study protocols were first approved by local ethical and animal care and use committees, and conformed to animal welfare guidelines.
Citric acid–induced cough in awake guinea pigs.
Awake animals were placed in a double-chambered body plethysmograph (type 855; Hugo Sachs Electronik, Germany) and were exposed to 0.4 M citric acid aerosol (Fisher, Slovakia) for 10 min. This design allows for simultaneous recordings of cough and lung mechanics. The aerosol was generated via a jet nebulizer (Pariprovocation test I; Pari Starneberg, Germany) and delivered to the head part of the body plethysmograph. Aerosols (median particle size diameter 1.2 μm) were evacuated from the chamber with the same rate of aerosol delivery (5 liters/min) by vacuum.
Analog recordings (Multiscriptor Hellige 21; Germany) of respiratory reflexes and breathing pattern were monitored using a pneumotachograph (Godart, Germany) with a Fleish head connected to the head chamber. Cough was detected using a microphone placed in the roof of the head chamber and connected to a tape recorder. Cough was defined as a sudden enhancement of expiratory airflow accompanied by typical cough sound, with responses expressed as the total number of coughs induced during challenge. The number of elicited cough efforts was concurrently counted and finally analyzed by two independent and experienced researchers, with one of the evaluators blinded to the procedures performed. Cough responses were evaluated in each animal after exposure to room air or menthol vapors, with challenges performed 1 wk apart [using methods comparable to those described by Laude et al. (22)]. In separate experiments coughing evoked by citric acid (0.01–0.3 M) was studied before and 30 min after oral administration (gavage) of dissolved menthol (30–100 mg/kg) or an equivalent volume of the vehicle (10% ethanol in water) in an unpaired experimental design.
Cough challenge in anesthetized guinea pigs.
Male Dunkin Hartley guinea pigs (250–350 g, pathogen free; Harlan, Indianapolis, IN) were anesthetized with urethane (1.5 g/kg ip). The adequacy of the anesthesia was assessed by monitoring withdrawal responses to a sharp pinch of a hind limb or responses during surgery. When the experiments were completed, animals were euthanized by asphyxiation in a vessel filled with carbon dioxide.
Anesthetized guinea pigs were secured supine on a warming pad. A midline incision in the neck exposed the trachea, which was cannulated at its caudal-most end with a bent luer stub adaptor. The cannula, placed approximately equidistant from the larynx and carina (1–2 cm), was attached to a length of tubing that terminated inside a water-jacketed organ bath continuously filled with humidified and warmed air (this design serves as an artificial warming and humidifying nose). The innervation and vasculature of the trachea were carefully preserved throughout the dissection. A pressure transducer attached to a side port in the tracheal cannula monitored respiratory efforts that were recorded digitally (Biopac, Santa Barbara, California). Once the trachea was cannulated, the remaining rostral segment of the extrathoracic trachea was opened lengthwise. Polyethylene (PE) tubing was threaded through the upper airways via the larynx. Warmed, oxygenated Krebs buffer comprising (in mM) 118 NaCl, 5.4 KCl, 1 NaHPO4, 1.2 MgSO4, 1.9 CaCl2, 25 NaHCO3, and 11.1 dextrose pH 7.4, was superfused (3 ml/min) over the tracheal mucosa. The cyclooxygenase inhibitor indomethacin (3 μM) was added to the perfusate to limit formation of neuromodulatory prostanoids, although we have shown previously that cyclooxygenase inhibition does not modulate cough responses to citric acid in anesthetized animals (6) or coughing evoked by bradykinin in conscious animals (unpublished observations). The buffer was introduced into the tracheal lumen from the caudal-most exposed segment of the trachea and removed at the rostral end of the trachea by attaching the PE tubing threaded through the upper airways to a vacuum.
After a 5-min equilibration period, coughing was evoked by applying citric acid topically to the tracheal mucosa. Citric acid (0.001–2 M) was dissolved in water and applied in 100-μl aliquots directly into the Krebs buffer perfusing the trachea. Concentration response curves were constructed in an ascending fashion, with doses administered at 1-min intervals. Cough was defined on the basis of visual confirmation of a cough-like respiratory effort and a change in tracheal pressure that produced a >500% increase in peak expiratory pressure preceded by an enhanced inspiratory effort all in <1 s.
Coughing was evoked in control animals and in animals coincidentally challenged with menthol or cold air. Menthol vapors were delivered selectively to the upper airways through a cannula placed in the nostril and connected to an air pump (30°C air source) or directly to the lower airways via the tracheal cannula. A 130-ml chamber containing 150 mg of menthol crystals was connected in series with the air pump and nasal cannula (36, 42). Menthol vapors were administered to the lower airways using the same chamber connected to the air pump supplying fresh air to the artificial nose. Menthol (0.1 mM) was also delivered selectively to the trachea by addition to the tracheal perfusate. We also studied the effects of delivering frigid air (generated through 1 meter of tubing placed in an ice bath and connected to the air pump) to the upper or lower airways on breathing pattern and citric acid–evoked coughing. Icilin (10 μM) was administered by addition to the tracheal perfusate or by bolus administration to the nose. Changes in breathing pattern in response to these stimuli and the number of coughs evoked by citric acid during their administration was recorded using an unpaired experimental design.
Retrograde neuronal tracing and single-cell RT-PCR analysis.
In brief anesthesia, the fluorescent tracer DiI (dissolved in DMSO to 2% and subsequently diluted in saline to 0.05%; Invitrogen, Carlsbad, California) was injected into the nasal turbinate unilaterally at two sites according the methods described by Taylor-Clark et al. (46). In separate animals, neurons innervating the lower airways were labeled by DiI injection into the trachea or intrapulmonary airways. Seven to 14 days later the animals were euthanized by overdose and exsanguination, the trigeminal ganglia or jugular and nodose ganglia were harvested, and the ganglia were then incubated in enzyme buffer (2 mg/ml collagenase type 1A and 2 mg/ml dispase II in Ca2+-, Mg2+-free Hanks' balanced salt solution) for 30 min at 37°C. Neurons were then dissociated by trituration with three glass Pasteur pipettes of decreasing tip pore size, washed, suspended in l-15 medium containing 10% fetal bovine serum (l-15/FBS) and transferred onto Poly-d-Lysine/Laminin-coated coverslips. After the suspended neurons had adhered to the coverslips for 2 h, DiI-labeled neurons identified via fluorescent microscopy were individually harvested into a glass pipette (tip 50–150 μm) pulled with a micropipette puller (P-87, Sutter) by applying negative pressure. Single-cell RT-PCR studies were performed on labeled neurons as described previously (21, 33, 34).
Results are presented as a mean ± SE of n experiments, where n refers to a single animal. The majority of the experiments were performed using an unpaired experimental design. Differences in group means were compared by t-test ANOVA and compared post hoc using Scheffé's test for unplanned comparisons. P < 0.05 was considered statistically significant.
Citric acid and urethane were purchased from Sigma (St. Louis, MO). Icilin was purchased from Tocris (Minneapolis, MN). Menthol (+) and (−) were purchased from MP Biochemicals (Aurora, OH).
Menthol inhibits cough in awake guinea pigs.
Orally administered menthol (100 mg/kg) significantly inhibited cough evoked by citric acid. Coughing evoked by low doses of citric acid (0.01–0.1 M) and the cumulative coughs evoked by all doses of citric acid (0.01–0.3 M) were inhibited. Overall, citric acid evoked 19 ± 2 coughs in control animals and 10 ± 2 coughs following orally administered menthol (n = 10–18; Fig. 1; P < 0.05). A lower dose of menthol (30 mg/kg) administered orally was without effect on cough (17 ± 4 coughs; n = 4; P > 0.1).
We also studied the effects of menthol vapor on 0.4 M citric acid–evoked coughing using our two-chamber plethysmograph design. Under these conditions, responses were variable and thus the menthol vapor fell just short of producing a statistically significant inhibition of citric acid–evoked coughing (13 ± 1 vs. 8 ± 2 coughs at baseline and during menthol challenge, respectively; n = 10; P = 0.05).
Effect of cold air, menthol, and icilin on breathing pattern and cough.
Cold (∼10°C) or warm (30°C) air was delivered to the nasal cavity of anesthetized guinea pigs through a nasal cannula. Flows were generated by an air pump at a rate of ∼30 ml/min, which is within the physiological range for guinea pigs. Control animals had the nasal cannula (without airflow) introduced into one nostril. Cold air significantly, acutely, and persistently decreased breathing rate (∼20% drop). Warm air (30°C) produced only modest (<10%), variable, and thus statistically insignificant effects on respiratory rate (Figs. 2 and 3). Adding (−)-menthol vapors to the warm air challenges produced a marked slowing of respiratory rate, comparable to the effects of cold air. Vapors of the inactive isomer, (+)-menthol, had little or no effect on respiratory rate (Figs. 2 and 4).
Cold air flow or (−)-menthol vapors delivered selectively to the nose nearly abolished coughing evoked by citric acid applied topically to the tracheal mucosa (Figs. 3 and 4). No significant effects of warm air or the less active isomer (+)-menthol were apparent, with citric acid evoking 8 ± 1, 7 ± 1, and 6 ± 1 coughs in control, warm air, and nasal (+)-menthol treated animals, respectively (P > 0.1).
The route of menthol delivery profoundly influenced its effects on respiratory rate and cough. Thus, in contrast to the inhibitory effects of nasal cold air and nasal (−)-menthol vapors on breathing rate, menthol delivered directly to the tracheal mucosa produced no consistent changes in breathing pattern (Fig. 2). If anything, menthol delivered to the intrapulmonary airways as vapors or as an aerosol increased respiratory rate while producing audible sounds of wheezing and evidence of increased airway secretions in some animals. Neither of these routes of menthol delivery inhibited citric acid–evoked coughing (Fig. 5).
Icilin, which like menthol activates TRPM8 (but also TRPA1), failed to alter breathing pattern when delivered selectively to the tracheal mucosa and did not inhibit coughing induced by citric acid. On the contrary, tracheal icilin administration enhanced citric acid–evoked coughing (9 ± 2 and 16 ± 1 coughs following vehicle or 10 μM icilin administration to the tracheal perfusate, respectively; n = 5–7; P < 0.05). This effect of icilin may be attributable to TRPA1-dependent activation of the jugular C-fibers innervating the trachea, which we have previously shown to sensitize cough responses evoked by citric acid challenge in anesthetized guinea pigs (26). Nasal challenge with a 100-μl bolus of 30 μM icilin evoked coughing immediately upon administration in two of three guinea pigs studied followed by respiratory slowing comparable to that evoked by menthol (Fig. 2). Due to its nonselective actions on TRPA1 and TRPM8, the effects of icilin on citric acid–evoked coughing were not extensively studied.
TRPM8 gene expression in trigeminal and vagal afferent neurons.
Using the single-cell RT-PCR technique we found that 60% (12/20) of labeled trigeminal nasal afferent neurons express mRNA for TRPM8. The TRPM8 gene was co-expressed with TRPV1 and TRPA1 mRNA in 7 of these 12 neurons, and they likely represent nasal nociceptors (35% of the overall labeled population). We have shown previously that either TRPV1 or TRPA1 activation in the nose enhances cough responsiveness in animals and in patients (5, 38, 39), thus it seems unlikely that these TRPM8-expressing nasal afferent neurons account for the inhibitory effects of nasal cold air or menthol challenges. The remaining five TRPM8+ neurons (25% of the overall population) did not express TRPV1 or TRPA1, and these neurons are likely the cold and menthol-sensitive nasal trigeminal afferents that initiate the inhibitory effects of menthol on cough (Fig. 6).
Consistent with most previous studies (33, 54, 57) and consistent with the lack of effects of menthol on citric acid–evoked coughing when delivered to the lower airways, none of the retrogradely labeled TRPV1-nodose ganglia neurons (0/6) and few of the retrogradely labeled jugular ganglia neurons (2/5) innervating the trachea or intrapulmonary airways expressed TRPM8. Interestingly, a majority of the retrogradely labeled TRPV1+ nodose ganglia neurons (6/8) innervating the airways and lungs were found to express TRPM8.
Inhaled menthol exerts complex olfactory, sensory, and respiratory effects in patients and in animals (9, 35, 41). Menthol has also been reported to inhibit coughing. The goal of the present study was to determine the mechanism by which menthol prevents coughing in a relevant animal model.
We confirmed the results of Laude et al. (22), documenting the antitussive effects of orally administered (−)-menthol. Likely due to the variability of the citric acid challenge model, menthol vapor treatments of conscious animals failed to achieve statistical significance (P = 0.05, not < 0.05), but overall, our results confirm previous results by revealing an inhibitory effect of menthol on cough.
Guinea pigs are obligatory nasal breathers and so when they are exposed to vapors or aerosols, a considerable proportion of aerosolized substances are deposited onto the nasal mucosa (23). The nose is thus a likely target for menthol action when administered as an aerosol or vapor, and also perhaps when administered orally. Thus, nasal effects after ingestion of substances has been documented in gustatory rhinitis, which is characterized by watery rhinorrhea after eating pungent foods (e.g., wasabi, which contains the TRPA1 activator allylisothiocyanate). These responses most likely occur following stimulation of trigeminal sensory nerve endings in the upper aerodigestive tract (16).
Several lines of experimental evidence suggest that menthol prevents cough through activation of TRPM8. Thus, in addition to menthol, we observed that cold air inhalation also inhibited coughing. The less active isomer of menthol, (+)-menthol (10), was without effect on citric acid–evoked coughing when administered to the nose. Menthol could conceivably be antitussive through effects on TRPA1, but we found that menthol modulated cough in our anesthetized animals only in airways in which nerves consistently express TRPM8 (the upper airways). In these upper airways, the neurons expressing TRPA1 also express TRPV1, and we have shown previously that either TRPA1 or TRPV1 activation in the nose enhances subsequently evoked cough responses (5, 38, 39). It is difficult to envision a scenario in which activation of the same nerve subtypes could both enhance and inhibit subsequently evoked coughing. Thus, although we like other investigators (36, 42) are hampered in our ability to study multiple doses of menthol administered as vapors and by solubility limitations of menthol in our oral dosing studies, we believe the data argue in favor of a TRPM8-dependent antitussive effect of menthol on cough. Perhaps with the recent reports of potent and selective blockers of TRPM8 (2, 19, 25), which are at present not available commercially, this hypothesis may soon be addressed using pharmacological approaches.
Our results generated in anesthetized animals suggest that the antitussive effects of menthol occur through a reflex initiated from the nose. Neurophysiological, immunohistochemical, and molecular analyses have documented TRPM8 expression in nasal trigeminal afferents but minimal expression of TRPM8 by bronchopulmonary vagal afferents (11, 33, 54, 55). Consistent with these previous studies, we found that menthol and cold air delivered selectively to the nose inhibited coughing evoked by citric acid applied topically to the tracheal mucosa, whereas administering menthol directly to the trachea or delivering menthol to the intrathoracic airways had no effect on citric acid–evoked cough. Icilin applied topically to the tracheal mucosa also failed to inhibit cough.
As discussed above, menthol and icilin may also activate the polymodal irritant receptor and ion channel TRPA1, which may account for the burning sensations occasionally reported in studies of menthol challenge (16, 32, 52). Our molecular analyses identified at least two populations of TRPM8-expressing nasal trigeminal afferent neurons, with one population co-expressing TRPA1 and TRPV1. Such patterns of TRP channel expression have been noted previously (1, 3, 20, 43, 53, 56). We speculate that TRPM8-dependent inhibitory actions of these stimuli on cough depend upon the activation of the nasal trigeminal afferent neurons that do not co-express TRPA1 or TRPV1. We have shown previously in guinea pigs, cats, and human subjects that TRPA1 or TRPV1 stimulation in the nose enhances cough responsiveness (4, 5, 38–40); this argues against an antitussive effect of menthol transduced through TRPV1 and TRPM8-expressing neurons. In fact, we observed coughing in two of three animals challenged with nasal icilin.
Our studies do not rule out modulatory effects of menthol and cold air transduced through activation of vagal afferent nerves. If anything, icilin applied topically to the tracheal mucosa and menthol delivered selectively to intrathoracic airways tended to enhance cough responsiveness. The sensitizing effects of icilin on cough might be attributed to a TRPA1-dependent activation of the TRPV1/TRPA1 expressing jugular C-fibers (26, 31), which we show here may also express TRPM8. Adverse effects of menthol have been reported (8, 24, 49, 52).
We conclude that menthol suppresses cough by a reflex transduced through TRPM8-dependent activation of nasal trigeminal afferent neurons. These effects of menthol were apparent in conscious animals challenged with citric acid aerosols, a cough that likely depends on TRPV1-dependent activation of tachykinin-containing jugular C-fibers (7, 12, 31, 47), and in anesthetized guinea pigs, when coughing was evoked by citric acid applied in bolus challenges directly to the tracheal mucosa, which depends on TRPV1-independent activation of the acid-sensitive, mechanically sensitive cough receptors (6, 27). Menthol may thus be a broadly effective antitussive, preventing cough responses evoked by multiple mechanisms and pathways. These observations support the validity of upper airway menthol vapors in controlling cough, an approach that may be clinically effective in children (37). It is unclear whether comparable therapeutic benefits can be achieved in adults. Our results do not rule out additional actions of menthol and cold air transduced through effects on other airway or extrapulmonary afferent nerves, or through TRPM8-independent mechanisms. Such studies await the availability of more selective activators and blockers of TRPM8. The central inhibitory pathways connecting the upper airways to cough pattern generation also await further analysis (40, 41, 54, 55).
J. Plevkova is supported by VEGA Grant 1/0031/11 and by the Center of Experimental and Clinical Respirology (CEKR), which is funded by the European Union. This work was also supported by National Heart, Lung, and Blood Institute Grant HL083192 to B. J. Canning.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: J.P., M.K., I.P., M.T., and B.J.C. conception and design of research; J.P., M.B., L.S., and N.M. performed experiments; J.P., M.K., I.P., M.B., L.S., M.T., N.M., and B.J.C. analyzed data; J.P., M.K., I.P., M.B., L.S., M.T., and B.J.C. interpreted results of experiments; J.P., M.K., L.S., N.M., and B.J.C. prepared figures; J.P. and B.J.C. drafted manuscript; J.P. and B.J.C. edited and revised manuscript; J.P., M.K., I.P., M.B., L.S., M.T., N.M., and B.J.C. approved final version of manuscript.
- Copyright © 2013 the American Physiological Society