Title page
Title:Methyl eugenol inhibits phasic inspiratory synaptic inputsto tracheobronchial airway vagal preganglionic neurons
Authors:LiliHou1, LeiZhu2, DongyingQiu3, ZhenweiLiu1, MinZhang1,Guoqing Zhang1, LiqingZhang4, Qiang Li1**, XinZhou1*
Affiliations:1, Department of Respiratory Medicine, Shanghai General Hospital, Shanghai Jiao Tong University; 2,Department of Respiratory Medicine, Zhongshan hospital, Fudan University;3, Department of Gerontology, Zhongshan Hospital, Fudan University;4, Taian Disabled Soliders’ Hospital of Shandong Province.
Corresponding author:
Name: Li Qiang1, Zhou Xin1
Address:100 Haining Road,Shanghai, 200080,China
Fax: 021-37798863
E-mail: , ,
Figures: 5
Abstract
Objective:Methyl eugenol (ME), a natural compound existing in lots of Chinese herbs, could cause respiratory inhibition, but the underlying mechanism remains elusive. The aim of the current study was to explore the mechanism involved by investigating the effect of ME ontracheobronchial airway vagal preganglionic neurons (TPVNs) at neural and synaptic levels.
Methods:TVPNs in the external formation of the nucleus ambiguous (eNA) were labeled by injecting rhodamine into the tracheal wall of anesthetized Sprague-Dawley rats. Inspiratory-activated TVPNs (IA-TVPNs) and inspiratory-inhibited TVPNs (II-TVPNs) were identified in thebrainstem sliceswith rhythmically firing, and then the effect of ME was investigated using patch clamp techniques.
Results:ME at 10μmol L-1 didn’t affect the frequency and intensity of the hypoglossal bursts.ME at 100μmol L-1initially didn’t change or temporarily increased, but inhibited the frequency, amplitude, area and duration of the hypoglossal bursts after prolonged exposure. Coincidentally, the amplitude, area and duration of phasic inspiratory inward current in IA-TVPNs during inspiratorybursts wereinhibited. In addition, the frequency of bursting sEPSCs decreased; while the amplitude increased and finally both werecompletely blocked. However, the tonic glutamatergic sEPSCs, tonic GABAergic and glycinergic sIPSCs in IA-TVPNs remained unaltered. Application of 100μmol L-1 ME significantly inhibited phasic inspiratory outward currents and bursting sIPSCs of II-TVPNs, but didn’t affect tonic glutamatergic sEPSCs and glycinergic sIPSCs.
Conclusion:The current study demonstrated that therespiratory inhibition caused by ME was due to the inhibition of ME on phasic inspiratory synaptic inputs to TVPNs.
Key words: Respiration, Synaptic transmission,Neurons,Methyl eugenol, Patch Clamp Techniques.
Count of abstract words: 252
Word count of manuscript：3342
Abbreviations list:
ME,Methyl eugenol; ACSF, artificial cerebral spinal fluid;CNQX, 6-Cyano-7-nitroquinoxaline-2, 3-dione; DMSO, dimethyl sulfoxide; eNA, the external formation of the nucleus ambiguous; sEPSCs, spontaneous excitatory postsynaptic currents; sIPSCs: spontaneous inhibitory postsynaptic currents; TVPNs,tracheobronchialvagal preganglionic neurones;IA-TVPNs, inspiratory-activated tracheobronchialvagal preganglionic neurones; II-TVPNs, inspiratory-inhibited tracheobronchial airway vagal preganglionic neurons.
- Introduction
Methyl eugenol (3, 4-dimethoxyallybenzene; ME) is a bioactive component extracted using the modern techniques from various kinds of Chineseherb 1-3such as Asarum, rhizomes of Acorustatarinowii, Agastacherugosa, etc. ME has been well recognized for its biological actions, i.e. anaesthetic,4, 5anticonvulsant,6, 7 antinociceptive,8antioxidative and anti-inflammatory9effects.It was recently reported that inhibition of ME on peripheral Nav1.7 channels was demonstrated to contribute to its anesthetic and antinociceptive effects.10ME at higher concentrations (≥100 μmol/L) exerted inhibitory effects on neural activation by evoking large GABAergic currents in hippocampal neurons.11 However, the further mechanism by which ME affects respiratory diseases remain to be elucidated. Previous in vitro studies exhibited that ME relaxed the airway smooth muscles, indicating it might target at the peripheral sites for the therapy of asthma. But over-dosageof intravenously administered ME caused slow breath, even led to apnea12 in rodents, suggesting that ME might cause side effects at the medullary respiratory centre. In our previous study, ME was proved to inhibit central respiratory drive to inspiratory-activated airway vagal preganglionic neurons.13However, there has been a dearth of further evidence to proveit.
Respiratory rhythm originates from thePre-Bötzingercomplex (PBC) located in the ventral medulla.14, 15Pacemaker neurones, considered as the central respiratory rhythm generator and connected by bidirectional electrical coupling and unidirectional excitatory synaptic transmission,16-18 provide projections to cranial motoneurones17including hypoglossal neurones and TVPNs. It has been previously demonstrated that a regular central inspiratory rhythm caused by inspiratory neurones in the PBC can activate hypoglossal neurones by providing various excitatory inputs, 19which exert coordinated control over swallowing, respiration, and suckling, etc.17
TVPNs located in the external formation of the nucleus ambiguus (eNA)20-22 provide cholinergic preganglionicterminals to the intrinsic tracheobronchial ganglia.Excitation of TVPNs induced pronounced contraction of airway smooth muscles,23-25increases of airway secretion and blood flow,26, 27which coordinated respiration physiologically and pathophysiologically. Our previous studies demonstrated that central inspiratory activity excited some neurons in eNA by enhancing bursting excitatory inputs and inhibited some neurons by enhancing bursting inhibitory inputs.20, 21, 28, 29As a result, TVPNs have been identified as inspiratory-activated tracheobronchial airway vagal preganglionic neurones (IA-TVPNs) and inspiratory-inhibited tracheobronchialairway vagal preganglionic neurones (II-TVPNs). In the tracheobronchial gangliathe postganglionic neurons also include ‘phasic’ and ‘tonic’ ones.The former fire in phase with inspiration andinnervate the tracheobronchial smooth muscles, whereas the latterfire tonically during inspiration and innervate the intercartilaginous spaces.30Postganglionic neurons firing similarly with preganglionic neuronshave also been identified in vitro; bothin humans and in animals.31-33It is thus reasonable to hypothesize that the IA-TVPNs innervatedpreferentially phasic postganglionic neurons and II-TVPNs,the tonicpostganglionic ones.
In the current study, we investigated the effects of ME on the synaptic inputs to both IA-TVPNs and II-TVPNs. The results showed that bath-applied 100 μmol L-1 ME induced dramaticallyinhibition in respiratoryrhythm and the intensity of hypoglossal bursts, and in the phasic inspiratory synaptic inputsto TVPNs, but didn’t affect the tonic synaptic inputs, both in IA-TVPNs and II-TVPNs.
- Materials and Methods
All animal care and procedures were approved by the Medical Experimental Animal Administration Committee of Shanghai and by the Laboratory Animal Ethics Committee of Shanghai General Hospital.
Retrograde fluorescent labelling of tracheobronchial airway vagal preganglionic neurons (TVPNs)
The 3 to 5-day-old Sprague-Dawley rats (Shanghai Institute for Family Planning and Shanghai General Hospital) wereanesthetized by inhalation of halothane, just as descriptions in our previous studies21, 22, 28, 29, 34. When the rat lost response to the limbspinched, the body was wrapped in ice-water-filled bags to decrease the body temperature and to slow the heart rate. When the automatic breathing disappeared (within two minutes), the animal was fixed on an ice-water-filled bag in a supine posture. A ventral midline incision was made in the neck to expose the extrathoracic trachea, followed by an injection of rhodamine (XRITC, Molecular Probes, 1% solution, 0.5μl) into the trachea wall between the fourth and eighth tracheal cartilage with a glass pipette, whose tip diameter was 30μmol L-1, before it was attached to a syringe through polyethylene tubing. Rinsed with saline containing 5mg/ml streptomycin sulfate and 50,000 U/mlpenicillin, the incision wassutured. The animals were heated with a thermo-pad to help recovery.
During the whole surgical period of approximately five minutes, the body temperature was below 10 ºC;the animal hasno spontaneous breathing or struggling. Within 3 min after the surgery,the animals started to breathe automatically,and withinanother 5 min, they started free moving. A single dose of morphine (10 mg kg-1) was intraperitoneallyinjected to relieve the postoperative pain. It was taken 48 h for the animals to recover.
2.1.Slice preparation
The animal, 48-52 h after the surgery, was deeply anesthetized with halothane again as described above, before decapitated. The brain was taken out of the skull within one min to besubmerged in cold (4 °C) artificial cerebral spinal fluid (ACSF) of the following composition in mmol L-1: NaCl 124, KCl 3.0, KH2PO4 1.2, CaCl2 2.4, MgSO4 1.3, NaHCO3 26, d-glucose 10. The solution was constantly bubbled with 95% O2-5% CO2, withapH of 7.4. The cerebellum was removed, and the hindbrain was isolated under a dissection microscope. The brainstem was then secured in theslicing chamber of a vibratome (Leica VT 1000S) filled with the same ACSF. The rostral end of the brainstem was set upwards;the dorsal surface wasglued to an agar block facing the razor.
The brainstem was serially sectioned in the transverse planein variable thickness. Once the nucleus ambiguus as a landmark was visible,a single medulla slice of 500-800 μm thickness with one to two hypoglossal rootlets in each side was made for experimentation.The slice preparation produces rhythmic discharge during inspiration in hypoglossal rootlets.35The slice was transferred into the recording chamber to be superfused with ACSF at a flow rate of 8-11 ml/min. The rostral cutting plane of the slice was set upwards to be identified fluorescently and recorded synaptic activities of the TVPNs by patch clamp. Bath temperature was maintained at 23 ± 0.5 °C, and the concentration of KCl in ACSF was increased to 10 mmol L-1 to keepa steady recording of the respiratory rhythm.
2.2.Electrophysiological recording
The patch pipettes were filled with a K+gluconate-dominated solution.28When the TVPNswere normallyclamped at -80 mV, neurons showing bursting sEPSCs during inspiratory bursts were identified as IA-TVPNs.And under cell-attached configuration,the neurons which exhibited trains of inspiratory dischargeswere identifiedas IA-TVPNs.The II-TVPNs were identified as those that showed phasic inspiratory bursts of the inhibitory (outward) synaptic activity under a holding voltage of -50 mV. In each slice was testedonlyoneTVPN.28
The patch-clamp signal was amplified with an Axopatch 700B amplifier (10 kHz sampling frequency; 1 kHz filter frequency), digitized with 1322A digidata, and collected with Clampex 9.2 software (Axon Instruments, USA). The raw and integrated hypoglossal signals were fed into the computer simultaneously with the patch-clamp signal as described before.20-22, 28
2.3.Drug application
ME was dissolved in dimethylsulfoxide(DMSO) to make a stock solution of 100 mmol L-1, and diluted to 10, 100 μmol L-1 with ACSF. DMSO was bath-applied for 20 min prior to the presence of ME in the control. The slices were submerged in the perfusate of ME for 1h unless the respiratory bursts were blocked. ME was used only once in each slice.6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 50 μmol L-1), a mixture of picrotoxin (10 μmol L-1)and strychnine (1 μmol L-1) were used to block Non-NMDA glutamate receptors,GABA receptors and glycine receptors, respectively. All drugs were purchased from Sigma–Aldrich (St. Louis, MO, USA).
2.4.Data analysis
The tonic glutamatergic sEPSCs, tonic GABAergic and glycinergic sIPSCs, the bursting sEPSCs and sIPSCs were analyzed using MiniAnalysis (Synaptosoft, version 4.3.1) with minimal acceptable amplitude of 10 pA. The intensity (duration, amplitude, area) of the phasic inspiratory inward/outward currents and the hypoglossal bursts were analyzed using Clamfit 9.2 (Axon instruments, USA), the changes of which were expressed as the percentages of the control values. Before analysing the phasic inspiratory inward currents and the hypoglossal bursts, selected segments of them were low-pass filtered at 5-HZ with the eight-pole Bessel filter. The data from 10 consecutive inspiratory phases prior to application of MEwere analyzed as the control. Thedata obtained from the last 5 inspiratorydischarges prior to the rhythm disappearance were consideredas the maximal effectswhen hypoglossal bursts were completely blocked during drug application.Respiratory frequency was manually calculated. The data from5-min recording prior to application of ME were analyzed as the control, and those from the last 5 inspiratory discharge prior to the rhythm disappearance, as the maximal effects prior to the blockade of the inspiratory discharge. The results are presented as mean±SD, and compared as paired-Student’s t-test. Significant difference was set at p<0.05.
3．Results
3.1. Changes in frequency, amplitude, duration and area of hypoglossal bursts
Initially, ME at 10 μmol L-1 increased slightly the frequency, but didn’t affect the frequency (4.30±0.59 vs. 4.05±0.45 bursts min-1,P=0.10), amplitude (7.07±15.28%, P=0.30) duration (2.19±8.64%,P>0.22) or area (2.52±4.42%, P=0.22) of the hypoglossal bursts with prolonged exposure of ME (6 slices,Figure.1. A).
When applied to 20 brainstem slices, ME at 100 μmol L-1 increasedslightlythe frequency (16 of 20) and temporarily the amplitude (12 of 20) of the hypoglossal bursts. The changes started within 4 to 12 min (Fig.1 B). Initially, the frequency (4 of 20) and the amplitude (8 of 20) were unchanged (data not shown). However, both the frequency and amplitude were progressively inhibited over time, and finally abolished following exposure of 16-50 min to ME. Before blockade of respiratory-likerhythm, ME induced a significant inhibition of the frequency (5.62±1.67 vs 2.25±0.80 bursts min-1,P＜0.01; n=20) and the amplitude (45.30±16.58%,P＜0.01; n=12) of the hypoglossal bursts. In 16 of 20 slices, ME caused slight increases in the duration and area of the hypoglossal bursts, which started at the time points of 2-5 min after ME application (Fig .1.C and D). No significant effect was observed on the other four slices regarding the duration and area of the hypoglossal bursts (data not shown). The duration (59.10±18.89%,P＜0.01; n=12) and area (46.14±13.40%,P＜0.01; n=12) were also attenuated with a prolonged application of ME before a complete blockade at last. The hypoglossal bursts appeared spike-like just before the blockade of the respiratory-like rhythm. Under the control conditions a representative hypoglossal burst was compared with the last one prior to the abolishment of the respiratory rhythm (Fig 1D). A summary was made of the data showing the maximal changes in the frequency, amplitude, duration and area (n=12, Figure 1F, G).
All of the responses to ME were reversible (Fig. 1B). It took 3-10 min for the hypoglossal bursts to reappear after washing. In our previous study, wedemonstrated that there were no obvious alterations in the frequency, amplitude, area and duration of hypoglossal bursts during 2-hour recording.
3.2. Effect of ME onthe synaptic inputs to IA-TVPNs
3.2.1. Inhibited duration, amplitude and area of the phasic inspiratory inward currents in IA-TVPNs
IA-TVPNs showed the phasic inspiratory inward currents and bursting sEPSCs under a holding voltage of -80 mV (6 IA-TVPNs) or -50 mV (2 IA-TVPNs) (Fig.2). Bath application of 100μmol L-1, ME inhibitedprogressively the intensity of phasic inspiratory inward currents in 10 IA-TVPNs whatever happened to the hypoglossal bursts initially. The amplitude was decreased by 34.24±12.24% (P<0.01, n=8); the area, by 56.24±12.94% (P<0.01, n=8); and the duration, by 42.10±8.58% (P<0.01, n=8), respectively. A representative phasic inspiratory inward current during the control recording was compared with the last inward current prior to blockade of the respiratory-like rhythm following an application of 100μmol L-1ME (Fig.2D). A summary was made of the maximal alterations in amplitude, area and duration (Fig. 2E).
3.2.2. Decreased frequency and increased amplitude of bursting sEPSCs in IA-TVPNs
The effect of ME on bursting sEPSCs during the inspiratory bursts and on tonic glutamatergic sEPSCs during inspiratory intervals were observed with the neurons clamped at -80 mV or -50 mV. Ineight IA-TVPNs, initially with instantaneous augmentation of the hypoglossal bursts, the frequency or the amplitude of the bursting sEPSCs increased simultaneously. Interestingly, with a decline of the hypoglossal bursts over time, the frequency was progressively inhibited, but the amplitude was temporarily augmented, especially prior to the blockade of the hypoglossal bursts (Fig 2A, control; Fig 2B, prior to blockade of respiratory rhythmic activity). The frequency was decreased by 27.29% and the amplitude was increased by 68.81%, respectively (Fig.3.). In two IA-TVPNs, the bursting sEPSCs declined with the attenuation of the hypoglossal bursts, both in frequency and in amplitude (data not shown). Additionally, the bursting sEPSCs disappeared with a complete blockade of the hypoglossal bursts, whatever happened initially to the frequency and amplitude.
It was found that 100μmol L-1ME caused a slight increase (n=3), a slight decrease (n=3), and no changes (n=2) of the frequency or amplitude of the tonic glutamatergic sEPSCs in individual IA-TVPNs, while the overall changes were not obvious (n=8, P>0.05) (Fig. 2B). The data were summarized (Fig 2F and G). Furthermore, the tonic glutamatergic sEPSCs, as the bursting sEPSCs were completely blocked, still remained unaltered (Data was not shown).
3.2.3. No effect on tonic GABAergic and glycinergic sIPSCs in frequency and in amplitude
Under a holding voltage of -50 mV (4 IA-TVPNs), an application of 100μmol L-1ME induced a slight decrease in frequency initially, but didn’t cause any significant changes in the frequency or amplitude with a prolonged application (n=4, P>0.05) (Fig. 4). The effects of ME on both excitatory and inhibitory synaptic currents were reversible. At the end of the experiments, the bath application of CNQX abolished all the bursting and tonic glutamatergic sEPSCs, and the combined application of picrotoxin and strychnine blocked all tonic GABAergic and glycinergic sIPSCs.
3.3. Attenuated the phasic inspiratory outward currents in amplitude and area
Under the same holding voltage, the II-TVPNs exhibited phasic inspiratory outward currents and bursting sIPSCs (Fig.5.). 100 μmol L-1ME caused limited changes in the tonic glutamatergic sEPSCs (n=4, P>0.05,Fig.5. D, E) and glycinergic sIPSCs (n=4, P>0.05, Fig.5. F, G), whereas the phasic inspiratory outward currents were significantly inhibited. The amplitude was decreased by 49.58±10.37% (P<0.01, n=6) and the area by 78.56±16.44% (P<0.01, n=6), respectively (Fig.5.). The bursting sIPSCs were also almost inhibited (Fig.5.). At the end of the experiments, all the spontaneous sEPSCs and sIPSCs were abolished by CNQX (50μmol L-1) and a mixture of strychnine (1μmol L-1) and picrotoxin (10μmol L-1), respectively.Surprisingly, focal application of CNQX (50μmol L-1) abolished almost all the phasic inspiratory outward currents and bursting sIPSCs.
Discussion
The current study presented three major findings: ME inhibited significantly the frequency and intensity of the hypoglossal bursts in a progressive manner and blocked completely the respiratory rhythm; ME decreased conspicuously the phasic synaptic inputs with a reduction of respiratory rhythm, in both IA-TVPNs and II-TVPNs; and ME had no effect on tonic synaptic inputs during the inspiratory intervals, in both IA-TVPNs and II-TVPNs.
ME andβ-asarone are believed as important constituentsof Asarum volatile oil. In our previous study,β-asarone had been found to inhibit the hypoglossal bursts in brainstem slices with respiratory rhythm in a dose-dependent manner,20suggesting that it was one of the components which caused respiratory disturbance during Asarum application. Generally, ME is thought to play a crucial role in treating diseases as a component of Asarum in Traditional Chinese Medicine (TCM). In the current study, ME was also found to induce a complete blockade of both the respiratory rhythm and the frequency and the intensity of hypoglossal bursts, demonstrating at least its inhibitoryeffect on the central respiratory drive from neuronsproducing respiratory rhythm in the PBC to hypoglossal neurons. This indicates that apnea caused by ME in rodents12 may be due to the suppression of the central excitatory drive to hypoglossal neurons. However, the time course of ME-induced blockade of the hypoglossal discharge was shorter than that ofβ-asarone,20so wasthe time course of ME-induced recovery of the hypoglossal respiratory-like rhythm20.