Introduction

Neurogenic inflammation is the inflammation that results from the release of neuropeptides from primary sensory nerve terminals. It plays an important role in regulating cough reflex and airway responsiveness. C-fibers, along with rapidly adapting stretch receptors (RARs) and slowly adapting stretch receptors (SARs), are the main kinds of sensory nerve fibers in the cough reflex. The nerve terminals of C-fibers are also the main source of neuropeptides like substance P (SP) and calcitonin gene-related peptide in airway. Various chemical stimuli such as capsaicin, bradykinin, citric acid, hypertonic saline and sulfur dioxide can activate C-fibers, resulting in releasing of neurokinins from the nerve terminals [1]. Via specific receptors, these neuropeptides can cause vascular leakage, smooth muscle spasm, mucus secretion, tissue edema and modulate immune cells functions [2, 3], which can further stimulate RARs and cause cough and airway hyperresponsiveness (AHR) [1]. Neutral endopeptidase (NEP) is the main degrading enzyme of neuropeptides in airway. Inhibition of NEP activity by toxic inhalation or its specific inhibitor can enhance the cough responsiveness [2, 4]. High level of neuropeptides and their receptors and NEP activity defect in lung have been reported in asthma, chronic obstructive pulmonary disease (COPD) and respiratory viral infection. Neurogenic inflammation is believed to contribute to the enhanced cough, AHR, mucus secretion and inflammatory cell influx in these diseases [3].

Cigarette smoke (CS) exposure is the main cause of chronic bronchitis, in which the CS-enhanced neurogenic inflammation plays an important role in chronic cough [5, 6]. Nicotine in CS can evoke cough by activating neuronal nicotinic acetylcholine receptors (nAChR), and sensitize the TRPV1 response to capsaicin with or without activation of nAChR [7, 8]. α,β-unsaturated aldehydes, which are abundant in CS, are also found to mediate the cigarette smoke-induced neurogenic inflammation through TRPA1 receptor [9]. Besides, various inflammatory mediators, such as tumor necrotic factor α (TNF-α), reactive oxygen species (ROS), prostanoids and leukotrienes, induced by CS-exposure can further enhance the neuropeptides synthesis in vagal bronchopulmonary sensory nerves and sensitize these nerves [10, 11]. Furthermore, the free radicals in CS can depress the activity of NEP in airways, delaying the clearance of neuropeptides and augmenting the airway neurogenic inflammation [12]. Thus the animal models with acute or chronic CS exposure are considered as useful tools for researching neurogenic inflammation relative pathological cough [13].

In recent decades, antagonists of the neurokinin receptors, bradykinin receptors and transient receptor potential cation channel have been developed as antitussives [14]. The effects of some existing antitussives, such as codeine, moguisteine, levodropropizine (LVDP), on the neurogenic inflammation also have been studied in some physical or acute pathological animal models [15–18]. However, the effects of these antitussives on the neurogenic inflammation in chronic airway disease model were barely discussed yet. In present study, effects of four antitussives, including codeine phosphate (CP), moguisteine, LVDP and a recently reported antitussive naringin [19], on airway neurogenic inflammation as well as enhanced cough and breathing pattern to capsaicin challenge were examined in a guinea pig model of chronic cough induced by chronic CS exposure.

Methods

Animal

Male and female Hartley strain guinea pigs (300–350 g) were purchased from Guangdong Medical Laboratory Animal Center (Guangdong, China). The animals were maintained in ordinary animal cages, with food and water available ad libitum and randomly divided into two groups in time course study and six groups in testing study. Some animal deaths happened during the 8 weeks experiment when many measures had been made to avoid possible death (like vitamin C addiction in drinking water, cleaning the cage and bedding more often, adding fresh vegetable besides the standard feed). Most of the death had happened before the 7th week and the death also happened in normal group. Thus the numbers of animals in different groups were not all the same at the end of the study (n = 6–8 in time course study and n = 5–9 in testing study). The study was performed according to the Helsinki declaration and all experimental procedures were approved by the Animal Care and Use Committee of the School of Life Sciences, Sun Yat-sen University, PR China. Adequate measures were taken to minimize pain of experimental animals.

Chemicals

Capsaicin (30.5 mg, Hubei Artec Carbohydrate Chemistry Co., Ltd., China) was dissolved in ethanol (1 ml) and Tween 80 (1 ml), then further dissolved with 0.9 % saline solution (8 ml) to yield a 0.01 M stock solution [20]. The stock solution was then diluted with saline to give a 50 μM working solution. CP was purchased from Qinghai Pharmaceutical Co., Ltd, (Xining, China). Moguisteine was purchased from Beijing HuaFeng United Technology Co., Ltd (Beijing, China). LVDP was purchased from Hunan Jiudian Pharmaceutical Co., Ltd, (Changsha, China). Naringin was extracted by our laboratory (extracted from Citrus grandis ‘Tomentosa’ by water, deposited in ethanol, with concentrated filtrate obtained after one to ten times of recrystallization. HPLC purity is >98.3 % (batch no. 20080203) determined by peak area normalization. HPLC chromatogram of naringin is shown in Fig. 1). All these antitussives were dissolved in saline and orally administered at a gavage volume of 5 ml/kg body weight.

Fig. 1
figure 1

HPLC chromatogram of naringin. The purity analysis was performed on a Shimadzu (HPLC) LC-6A instrument (Shimadzu Corp., Kyoto, Japan) with a Dionex C18 column (5 μm, 4.6 mm × 250 mm, USA) and a TL9000 Chromatographic Station. The mobile phase was prepared by a 45/55 (v/v) mixture of methanol/water and the pH was adjusted to 3.0 with acetic acid. The injection volume was 20 μL. The UV detector was set at a wavelength of 283 nm

Full size image

CS exposure and drug treatments

CS exposures were performed as described by Luo et al. [21]. Commercially available filter cigarettes (Cocopalm brand cigarettes; China Tobacco Guangdong Industrial. Co., Ltd., Guangdong, China) were used for the CS exposure. Each cigarette contained 1.2 mg of nicotine and 13 mg of tar according to the manufacturer’s specifications. Animals were consciously and unrestrainedly exposed to CS in smoke chamber (80 cm × 80 cm × 100 cm). They were repeatedly exposed to the smoke of 10 cigarettes/day, 1 h/day, 6 consecutive days/week, for up to 8 weeks. During the period of CS exposure, the temperature in the chamber was maintained at about 26 °C.

For time course study, the normal group and CSE group animals were not treated with vehicle or antitussives. For testing study, antitussives and vehicle were administered to animals 1 h before each CS exposure session during the 7th and 8th week according the result of time course study. Vehicle, CP (4.8 mg/kg), moguisteine (24 mg/kg), LVDP (14 mg/kg) and naringin (18.4 mg/kg) were orally administered respectively to CSE group, CP group, moguisteine group, LVDP group and naringin group. Doses of CP, LVDP and moguisteine were equal to their clinically used doses (for CP and LVDP) or the clinical trial dose (for moguisteine) [22] as calculated by body surface area. The dose of CP (4.8 mg/kg) was much lower than the previously reported dose (60 mg/kg) with CNS effect in guinea pigs [23], and it did not produce phanerous sedation in our preliminary experiment either. The dose of naringin was chosen according our previous study [21]. Age-matched non smoke-exposed and vehicle-administered animals were used as normal group.

Measurements of cough and enhanced pause (Penh)

The measurements of cough and Penh were performed as previously described [21]. In order to find out the formation time point of the CS-exposure induced chronic cough, the cough responsiveness of guinea pigs in normal group and CSE group were measured before and 24 h after 2, 4, 6 and 8 weeks of CS exposure in time course study. In testing study, the measurements were carried out 24 h after 8 weeks of CS exposure. The drugs and vehicle were administered 1 h before cough challenge. Conscious guinea pigs were individually placed in an unrestrained whole body flow plethysmograph (FWBP) (Buxco, NY, USA). The rate of bias flow was 2.5 L/min in the plethysmograph. After 2 min of adjustment, animals were challenged with nebulized (Aerogen (Ireland) Ltd, Galway Ireland) capsaicin solution (50 μM) for 2 min at a nebulization rate of 0.5 ml/min. Coughs were recorded by both Biosystem XA software (Buxco Electronics, Sharon, CT) and a blinded observer during the capsaicin challenge and for a further 8 min after challenge. The trained blinded observer manually accessed cough according the cough like posture of guinea pig (splaying of the front feet and forward stretching of the neck, and opening of the mouth) in the transparent plethysmography and the box flow change recorded by the Buxco system (Fig. 2). The cough analyzer software of Buxco system automatically recognized the cough without audio monitoring but basing on the analysis of the box flow during one cough like action (Compared to sneeze, cough has higher value of the area under the box flow curve during compression and longer time between 50 % max of the compressive spike and 50 % min of the expulsive spike) [24].

Fig. 2
figure 2

The waveform of box flow during the cough of guinea pig in plethysmography

Full size image

Penh is a non-dimensional parameter based on a characteristic change in the expiratory waveshape of the unrestrained flow plethysmography, which can reflect the breathing pattern. Penh is the product of (Te/Rt-1)×PEF/PIF, where PEF = peak expiratory height in the FWBP waveform; PIF = peak inspiratory height; Te is the expiratory time and Rt is the time to expire 65 % of the volume [25]. Penh was recorded by the Buxco system simultaneously with cough recording in testing study. Raw Penh was recorded during 2 min of adjustment period, 2 min of capsaicin inhalation and 8 min further afterward. However, only the last 10 min of the time-Penh area under the curve (Penh-AUC) was calculated to reflect the changes in breathing pattern induced by capsaicin challenge (Fig. 3) [6]. All animals were challenged by capsaicin only once.

Fig. 3
figure 3

The representative time-Penh curves of two guinea pigs challenged by 50 μM capsaicin. Raw Penh was recorded during 2 min of adjustment period, 2 min of capsaicin inhalation and 8 min further afterward. Only the last 10 min of the time-Penh area under the curve (Penh-AUC) was calculated to reflect the changes in breathing pattern induced by capsaicin challenge. CSE cigarette smoke exposed, Normal non-smoke exposed

Full size image

Preparation of lung tissue sample

Guinea-pigs terminally anaesthetised with pentobarbital sodium (30 mg/kg) were humanely killed by exsanguination 4 h after the capsaicin challenge. Lung tissue samples were collected and stored at −80 °C until used. The thawed lung tissue samples were placed in 10× volume of cold PBS (0.1 M, pH 7.4) and homogenized for 90 s by automatic homogenizer (T10B, IKA, Staufen, German) on ice. The homogenate was then centrifuged in 4 °C at 3,500 rpm for 15 min. The supernatant was collected and stored at −80 °C until used.

Determination of SP content

The content of SP in the supernatant of lung tissue homogenate was measured with enzyme-linked immunosorbent assay (ELISA) kits (Bluegene, Shanghai, China) and presented as pg/mg protein. The total protein of lung tissue homogenate supernatant was measured with BCA assay kit (Beyotime, Shanghai, China).

Assay of NEP activity

NEP enzymatic activity was determined by fluorometric method in flat bottom 96-well plate. 20 μL lung tissue homogenate supernatant was added to the PBS (0.1 M, pH 7.4) containing 100 μM N-dansyl-d-Ala–Gly-p-(nitro)-Phe–Gly (DAGNPG, Sigma, MO, USA), which is internally quenched substrate for NEP, and with or without 40 μM (DL)-thiorphan (Santa Cruz, California, USA), which is NEP specific inhibitor. The reaction system was incubated in a final volume of 200 μL at 37 °C for 30 min, and the dansyl-d-Ala–Gly (DAG) production was fluorometrically detected (excitation 342 nm, emission 562 nm) by Infinite M200 microplate reader (Tecan, Austria). The fluorescence of 0–100 μM DAG (Sigma, MO) was also measured and the standard curve was computer-fitted by linear regression to extrapolate the values of the samples. The total protein of lung tissue homogenate supernatant was measured with BCA assay kit (Beyotime, Shanghai, China). 1U NEP activity corresponds to 1 μM DAG production per minute per milligram protein.

Western blotting of NK-1 receptor

Lung tissue homogenate supernatant was separated by 5–8 % SDS-PAGE, transferred to a nitrocellulose membrane, and blotted with primary (overnight, 4 °C), then secondary (1 h) antibodies. The primary antibodies included rabbit polyclonal anti-NK-1 antibody (1:1000; catalog no. NB300-101) (Novus Biologicals, Littleton, USA) and mouse monoclonal anti-β-actin antibody (1:1000; catalog no. AA128) (Beyotime, Shanghai, China). The secondary antibodies included horseradish peroxidase-labeled goat anti-rabbit IgG (H+L) for NK-1 (1:1000; catalog no. A0208) and horseradish peroxidase-labeled goat anti-mouse IgG (H+L) for β-actin (1:1000; catalog no. A0216) (Beyotime, Shanghai, China). The antigen/antibody complexes were visualized by an ECL method according to the manufacture’s instruction (Beyotime, Shanghai, China). The fluorescent images were captured by FluorChem Q system (ProteinSimple, Santa Clara, CA).

Statistical analysis

Datas are expressed as mean ± SD. Statistical analysis was performed using student’s t test for the time course study and one-way analysis of variance (ANOVA) followed by Dunnett’s t test for the testing study. Kruskal–Wallis H test and Mann–Whitney U test were used for the data that did not pass Levene’s homogeneity of variance test. A P value <0.05 was taken as significant. Statistical analysis was performed using SPSS for windows 16.0 (SPSS Inc. Chicago, USA).

Results

Time course of enhanced cough to capsaicin following chronic CS-exposure

Chronic CS-exposure induced a bi-phase time course of enhanced cough to capsaicin in present study. The cough responsiveness to capsaicin in guinea pigs was significantly enhanced after 2 weeks of CS exposure, following a decrease to baseline at the 4th week. With CS-exposure continuing, the cough responsiveness was reincreased significantly at the 6th week and stayed at high level for at least 2 weeks. These results suggested that the chronic enhanced cough was established in guinea pigs after 6 weeks of CS exposure (Fig. 4).

Fig. 4
figure 4

Time course of cough responsiveness to capsaicin following chronic CS exposure. Eight weeks of CS exposure (10 cigarettes/day, 1 h/day, 6 consecutive days/week) caused bi-phase time course of cough responsiveness to capsaicin in guinea pigs. Data presents the number of coughs in 10 min induced by capsaicin challenge. Each bar represents the mean ± SD. ## P < 0.01, # P < 0.05, significantly different from normal group; n = 6–8 animals per treatment group

Full size image

Effect of antitussives against the CS-enhanced cough and breathing pattern change

CS exposure for 8 weeks significantly increased the cough responsiveness and induced excessive breathing pattern change in guinea pigs when challenged by capsaicin (reflected by increased Penh-AUC). All the antitussive drugs tested effectively depressed the enhanced cough responsiveness to capsaicin (Fig. 5a). For Penh-AUC, CP and LVDP showed somewhat but not significant inhibitory effect (P values are 0.070 and 0.097 respectively). Moguisteine did not exert marked effect on Penh-AUC. However, naringin was significantly effective on attenuating the increasing of Penh-AUC induced by chronic CS exposure (Fig. 5b).

Fig. 5
figure 5

Effect of antitussives on the enhanced cough (a) and breathing pattern change (b) induced by chronic CS-exposure in guinea pigs. Age-matched non-smoke-exposed animals were used as normal group. Vehicle, CP (4.8 mg/kg), moguisteine (24 mg/kg), LVDP (14 mg/kg) and naringin (18.4 mg/kg) were orally administered for 2 weeks respectively. Each bar represents the mean ± SD. ## P < 0.01, # P < 0.05, significantly different from normal group; **P < 0.01, *P < 0.05, significantly different from model group; n = 5–9 animals per treatment group

Full size image

Effect of antitussives against the CS-induced lung SP content

The SP content in lung was markedly increased by 8 weeks of CS exposure in CSE group guinea pigs. The treatment of CP, moguisteine and LVDP for 2 weeks did not significantly reduce the SP level in lung. However, naringin showed slight but significant effect on reducing SP content (Fig. 6).

Fig. 6
figure 6

Effect of antitussives against the SP content in lung tissue. Age-matched non-smoke-exposed animals were used as normal group. Vehicle, CP (4.8 mg/kg), moguisteine (24 mg/kg), LVDP (14 mg/kg) and naringin (18.4 mg/kg) were orally administered respectively. Each bar represents the mean ± SD. ## P < 0.01, significantly different from normal group; *P < 0.05, significantly different from model group; n = 5–9 animals per treatment group

Full size image

Effect of antitussives against the CS-inhibited NEP activity

Eight weeks of CS exposure caused a slight but significant decrease in NEP activity in lung, which was effectively inhibited by 2 weeks of treatment of naringin. Administration of CP, moguisteine and LVDP for 2 weeks did not show significant effect on the NEP activity (Fig. 7).

Fig. 7
figure 7

Effect of antitussives against the NEP activity in lung tissue. Age-matched non-smoke-exposed animals were used as normal group. Vehicle, CP (4.8 mg/kg), moguisteine (24 mg/kg), LVDP (14 mg/kg) and naringin (18.4 mg/kg) were orally administered respectively. Each bar represents the mean ± SD. ## P < 0.01, significantly different from normal group; *P < 0.05, significantly different from model group; n = 5–9 animals per treatment group

Full size image

Effect of antitussives against the CS-induced NK-1 expression

As shown in Fig. 8, NK-1 receptor expression in lung could be significantly increased by 8 weeks of CS exposure in guinea pigs. Naringin treatment for 2 weeks significantly decreased the expression of the NK-1 receptor, while CP, moguisteine and LVDP did not showed significant effect.

Fig. 8
figure 8

Effect of antitussives against the NK-1 receptor expression in lung tissue. Age-matched non-smoke-exposed animals were used as normal group. Vehicle, CP (4.8 mg/kg), moguisteine (24 mg/kg), LVDP (14 mg/kg) and naringin (18.4 mg/kg) were orally administered respectively. a Representative immunoblot of NK-1 receptor and β-actin in whole lung tissue. b Densitometric measurements of NK-1 receptor expression in whole lung tissue. Each bar represents the mean ± SD (n = 4). # P < 0.05, significantly different from normal group; *P < 0.05, significantly different from model group

Full size image

Discussion

Neurogenic inflammation system plays an important role in the regulating cough responsiveness and airway reactivity [3]. SP released from the terminals bronchopulmonary sensory nerves such as C-fibers is one of the most important neurokinins relative with cough hyperreactivity [4]. Via NK-1 receptor and NK-2 receptor, SP can induce the bronchoconstriction, mucus secretion, edema, vascular leakage, which can further initiate the cough reflex via RARs [11, 26]. In another hand, neuropeptides in lung can be cleaved and inactivated by the NEP that locates at the surface of various types of cells. Thus NEP in lung plays a critical negative controlling role in airway neurogenic inflammation system [27]. In present research, CS-exposed guinea pigs were treated with CP, moguisteine, LVDP and naringin respectively, and the therapeutic effects of these antitussives on the airway SP content, NK1 receptor expression, NEP activity as well as enhanced cough were examined.

Cigarette smoking is strongly correlative with the chronic cough symptom in chronic bronchitis [28, 29]. Smoke-exposed guinea pig models are considered to be useful animal models of pathological cough [21, 30]. Via up-regulating the expression and releasing of neurokinins, inactivating airway NEP and enhancing the responsiveness of RARs to SP, acute and chronic smoke exposure can both accelerate the sensitivity of cough reflexes [6, 12, 31, 32]. In our previous research, cough hyperresponsiveness and chronic airway inflammation were induced in guinea pigs by 8 weeks of CS exposure [21]. In present study, a bi-phase time course of cough responsiveness to capsaicin following chronic CS exposure was further confirmed in guinea pigs. A transient increase was observed after 2 weeks of CS exposure, which was in accordance with the reports of Lewis et al. [6]. And the chronicity of enhanced cough responsiveness was found to be established after 6 weeks of CS exposure. This bi-phase time course of cough responsiveness is somewhat similar to the inflammation response of pulmonary to chronic CS exposure, which was comprised of acute phase in about 1–2 weeks and chronic phase in about 2–8 weeks [33–36]. Besides, as antitussives are usually recommended to be used temporally when necessary to avoid impeding the sputum clearance [29, 37], we chose to expose the animals to CS for 8 weeks and carried out the antitussive intervention in the last 2 weeks in testing study to examine their therapeutic effect on this chronic model of enhanced cough. As shown in the results, the cough responsiveness and Penh-AUC to capsaicin challenge were markedly increased in CSE group animals after 8 weeks of CS exposure, which were in consistent with our previous work [21]. At the same time, SP content and NK-1 receptor expression were also significantly increased, while NEP activity was markedly depressed in CSE group. These results suggested that chronic CS exposure had induced airway neurogenic inflammation along with enhanced cough and excessive breathing pattern change to capsaicin challenge in guinea pigs, which was suitable for studying chronic pathological cough. But something should be noted is that, as SP-immunoreactive nerves are more abundant in guinea pig airways than in human [2], the extrapolation from guinea-pig to man basing on this guinea pig model should be careful.

Codeine is a central acting antitussive which is usually recommended for treating chronic cough in bronchitis [29]. But some researches had reported that codeine was ineffective on the cough in COPD patients and in allergic challenged, SO2 exposed or enalapril treated animals. The inflammation, infection or CS facilitated C-fibers mediated neurogenic inflammation in airway was presumed to be the main reason for these codeine-resistent coughs [15, 38]. In present study, CP still showed significant effect on the enhanced cough, which was in accordance with previous reports of Lewis et al. [6]. But it did not significantly inhibit the CS induced hyper Penh-AUC and airway neurogenic inflammation level. These results indicated that the inhibitory effect of codeine on chronic CS exposure-enhanced cough in guinea pig was not necessarily dependent on inhibiting airway neurogenic inflammation.

Moguisteine and LVDP are peripheral antitussives which are chemically unrelated to codeine. Moguisteine can directly reduce the excitatory response of RARs to capsaicin instead of acting primarily on C-fiber receptor [16, 17], which might be the reason for moguisteine to reduce cough without affecting the airway neurogenic inflammation level. The peripheral antitussive effect of LVDP is thought to be mediated by decreasing the response of C-fibres to chemical stimuli [16]. But there are also some reports showing LVDP could reduce capsaicin induced bronchoconstriction and plasma extravasation with SP or NK1 receptor independent pathway [39, 40]. In present study, 2 weeks of treatment of LVDP was effective on the chronic CS exposure-enhanced cough responsiveness to capsaicin but ineffective on the SP content, NK1 expression and NEP activity. These results suggested that there should be some other neurokinin mediators involved in the antitussive effect of LVDP in the CS-exposure induced chronic cough.

Naringin is a recently reported peripheral antitussive. Its antitussive effect on non-pathological cough is C-fibers independent and is presumed to be exerted via modulating RARs activity [19]. In present study, therapeutic use of naringin showed significant antitussive effect on CS induced chronic cough. Besides, the excessive breathing pattern change to capsaicin, SP content, NK-1 receptor level and declined NEP activity were also significantly inhibited by naringin. As neuropeptides and receptors synthesis can be enhanced by airway inflammatory mediators like TNF-α, ROS, prostanoids and leukotrienes during chronic CS exposure [10, 11]. The inhibitory effect of naringin on CS exposure induced chronic airway inflammation may be relative with its effects on the airway SP content and NK1 receptor expression in lung [21]. However, the exact causal relationship between its effects on the airway inflammation and neurogenic inflammation needs further investigation. The airway NEP activity can be inactivated by free radicals in the CS and SOD aerosol inhalation can prevent this inactivation [12]. Naringin itself is a flavanone with strong scavenging activity on free radicals [41], and our previous study also showed that naringin could prevent the decrease of SOD activity in lung in CS-exposed guinea pigs [21]. Thus the antioxidant characteristic of naringin may have contributed to its protecting effect on the NEP activity in present study.

Conclusions

Neurogenic inflammation plays an important role in the enhanced cough induced by chronic CS-exposure. In present study, a bi-phase time course of cough responsiveness to capsaicin following 8 weeks of CS exposure was confirmed in a guinea pig model. Two weeks of therapeutic treatments of CP, moguisteine, LVDP and naringin were all effective on the chronic CS-exposure enhanced cough to capsaicin in guinea pig. Besides, naringin had further significantly depressed the CS-exposure facilitated airway neurogenic inflammation, which may be relative with its inhibitory effect on the chronic airway inflammation.