Math & Engineering

Glucose effects on gastric motility and tone evoked from the rat dorsal vagal complex

Description
Glucose effects on gastric motility and tone evoked from the rat dorsal vagal complex
Published
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Diabetes mellitus has been reported to be associated witha significant incidence of gastrointestinal (GI) tractsymptoms, including constipation, nausea, abdominalpain and diarrhoea (Enck et al. 1994). Studies carried outin tertiary care centres have shown GI symptoms in20–60% of diabetic patients (Feldman & Schiller, 1983;Clouse & Lustman, 1989), though the prevalence of GIsymptoms in diabetic patients selected from the generalpopulation is much less common (Janatuinen et al. 1993;Malecki et al. 2000). However, data from several studiesshow a weak but significant correlation between GI tractsymptoms and delayed gastric emptying in diabeticpatients (Horowitz et al. 1986, 1991; Wegener et al. 1990).Although the cause of delayed gastric emptying is notcompletely understood, Yamano & colleagues (1997)surmised that this may be caused by changes in bloodglucose levels, in addition to the autonomic neuropathicchanges known to occur with long-standing diabetesmellitus (Sampson et al. 1990; Vinik & Suwanwalaikorn,1997). Indeed, there is evidence from both human andanimal studies indicating that blood glucose level can actas a modulator of gastric motility (Sakaguchi & Shimojo,1984; Barnett & Owyang, 1988; Bjornsson et al. 1994).Animal studies to date have implicated vago-vagalreflexes in the action of glucose to alter gastric function(Sakaguchi & Shimojo, 1984; Sakaguchi et al. 1994). Thesereflexes consist of three components, the first of which isa sensory limb comprising chemosensory and mechano-sensory elements linked to vagal afferent nerves (Rogers et al. 1995). Data received by these sensory elements arefunnelled via a glutamatergic synapse into the brainstemat the level of the nucleus of the tractus solitarius (NTS)(Rogers et al. 1995; Sykes et al. 1997). Many of theperipheral vagal afferents that synapse in the NTS do soat the level of the medial subnucleus of the tractussolitarius (mNTS) (Altschuler et al. 1989). The NTS, inturn, sends projections to the efferent vagal neurones inthe dorsal motor nucleus of the vagus (DMV), which Glucose effects on gastric motility and tone evoked fromthe rat dorsal vagal complex Manuel Ferreira Jr*, Kirsteen N. Browning‡, Niaz Sahibzada*,JosephG.Verbalis†, Richard A. Gillis*and R. Alberto Travagli‡ *Departments of Pharmacology and †Medicine, Georgetown University Medical Center,Washington, DC and ‡Division of Gastroenterology and Department of Physiology,University of Michigan, Ann Arbor, MI, USA (Received 2 February 2001; accepted after revision 7 June 2001) 1.To examine the effects of glucose on the central components of the vago-vagal reflex control of gastric function, we performed both in vivo and in vitro experiments on neurones in the medialnucleus of the tractus solitarius (mNTS) and in the dorsal motor nucleus of the vagus (DMV). 2.In the in vivo anaesthetized rat preparation, unilateral microinjection of D -glucose (10 or 50m M (60nl) _1 ) in mNTS produced inhibition of gastric motility and an increase in intragastricpressure. D -glucose had no effect in the DMV. 3.In the in vitro  rat brainstem slice preparation, whole-cell recordings of DMV neurones showedthat increasing the glucose concentration of the perfusion solution from 5m M to 15 or 30m M produced outward currents of 35±5pA (n= 7) and 51±10pA (n= 11), respectively. Thesewere blocked by tetrodotoxin and picrotoxin, indicating that glucose was acting indirectly tocause the release of GABA. Decreasing the glucose concentration of the perfusing solution byone-half produced an inward current of 36±5pA (n= 7). 4.Stimulation of the NTS evoked inhibitory postsynaptic currents (IPSCs) in DMV neurones. Theamplitude of the evoked IPSCs was positively correlated with glucose concentration. Perfusionwith the ATP-sensitive K + (K ATP ) channel opener diazoxide mimicked the effect of reducedglucose, while perfusion with the K ATP channel blocker glibenclamide mimicked the effects of increased glucose. 5.Our data indicate that glucose had no direct excitatory effect on DMV neurones, but DMVneurones appear to be affected by an action of glucose on cell bodies of mNTS neurones viaeffects on an ATP-sensitive potassium channel. Journal of Physiology  (2001), 536.1 , pp  .141–152 12276 141  project to the parasympathetic ganglia and the entericganglia innervating the digestive tract (Rogers et al. 1995). Most of the projections from the NTS to the DMVappear to be inhibitory (McCann & Rogers, 1994) and,although the neurotransmitter released is unknown,indirect evidence suggests that it is GABA (Feng et al. 1990; Travagli et al. 1991; Washaban et al. 1995; Sivarao et al. 1998; Browning & Travagli, 1999). Glucose exerts pronounced effects both on vagal sensorynerves and on central components of the reflexes. Thehepatic portal area appears to have glucose sensors linkedto hepatic vagal afferent nerves (Sakaguchi & Shimojo,1984; Sakaguchi et al. 1994). In fact, glucose administeredinto the hepatic portal vein has been reported to decreasehepatic vagal afferent discharge rate (Niijima, 1969;Niijima & Mequid, 1994). Neurones in both NTS and DMVhave also been shown to be affected by glucose. Glucoseinjected into the DMV of anaesthetized rats has beenshown to decrease gastric motility and intragastricpressure (Sakaguchi et al. 1985, 1994). Conversely, gastricmotility or pressure did not seem to be affected whenglucose was injected into the NTS, although additionalstudies indicated that glucose injected into the NTS couldreduce gastric acid secretion (Sakaguchi & Sato, 1987). Electrophysiological studies of these two brainstem nucleiindicate the presence of both glucoresponsive (i.e. gluco-excitatory) and glucosensitive (i.e. gluco-inhibitory)neurones (Mizuno & Oomura, 1984; Adachi et al. 1995),and, in the case of the NTS, the majority of the gluco-responsive neurones were shown to be linked to the K ATP channel (Dallaporta et al. 2000).The purpose of the present study was to evaluate thepotential role of each of these brainstem nuclei in theeffects of glucose to modulate vagal control of gastricmotility and tone. This was accomplished using twocomplementary approaches, namely, microinjection of glucose into the mNTS and the DMV of the anaesthetizedrat while monitoring intragastric pressure and motility,and patch-clamp analysis of electrical activity of DMVneurones exposed to different concentrations of glucose.A preliminary account of this paper has been presentedpreviously (29th Annual Meeting of the Society forNeuroscience, 1999). METHODS In vivo  rat studies Animal preparation. Experiments were performed on maleSprague-Dawley rats (n= 59) weighing 250–350g (Taconic,Germantown, NY, USA). Prior to each experiment, food waswithheld overnight but water was provided ad libitum  . Animals wereanaesthetized with an intraperitoneal injection of a cocktail(3mlkg _1 ) containing urethane (800mgkg _1 ) and a -chloralose(60mgkg _1 ) dissolved in 3ml of 0.9% saline. Body temperature wasmonitored by a rectal thermometer and maintained at 37±1°C withan infrared heating lamp. Before surgery, all animals werepretreated with dexamethasone (0.8mg, subcutaneously) tominimize brain swelling. Animal care and experimental procedureswere performed in accordance with the NIH guidelines and with theapproval of the Animal Care and Utilization Committee of Georgetown University, Washington, DC, USA. Surgery. Rats were intubated to maintain an open airway and forinstituting artificial respiration when necessary. The carotid arterywas cannulated with polyethylene tubing (PE 50) to monitor bloodpressure. Blood pressure was recorded using a bridge amplifierconnected to a MacLab (ADInstruments, Milford, MA, USA) dataacquisition system. Data were stored on computer (Apple MacintoshG3 connected to MacLab) for analysis at a later time. In someanimals, ligatures were placed around the vagus nerves and weretied, to be avulsed at a later time. For each experiment an intra-gastric balloon, made from the little finger of a small latex glove, wastied around polyethylene tubing (PE 160) and was inserted into thestomach via the fundus. The tubing was connected to a pressuretransducer, which was connected to a bridge amplifier (MacLab).Data were saved for analysis at a later time. The stomach wasinflated by introducing warm saline (2–3ml) into the balloon, toachieve a baseline pressure of 6–15mmHg. The animals were thenpositioned in a stereotaxic apparatus (David Kopf, Tujunga, CA,USA). A partial dorsal craniotomy was performed to expose thefourth ventricle. The cerebellum was retracted slightly while using a26gauge needle to cut the subarachnoid covering. Calamusscriptorius (cs) was viewed from the dorsal aspect and used as a pointof reference (see below). Microinjection technique. D -Glucose was dissolved in 0.9% saline. Ahistological marker (Fast Green dye) for locating injection sites wasadded to drug solutions in a 1–2mgml _1 concentration. The pH of alldrug solutions was brought to 7.0–7.2. Double-barrelled pipetteswith a tip diameter of between 30 and 60µm were used. All micro-injections were given unilaterally. Injections were given in volumesof 60nl and administered by hand-controlled pressure. Micro-injections were given within 5s. Calamus scriptorius was used as azero reference point. Stereotaxic co-ordinates were srcinally chosenbased on histology in Paxinos & Watson (1986). Co-ordinates for theDMV ranged from 0.3–0.5mm rostral to cs, medial–lateral0.3–0.5mm lateral from the midline, and dorsal–ventral0.5–0.7mm from the dorsal surface of the medulla. Co-ordinates forthe mNTS ranged from 0.3–0.5mm rostral to cs, 0.5–0.7mm lateralfrom the midline, and from 0.4–0.6mm from the dorsal surface of the medulla. These co-ordinates were chosen based on co-ordinatesdeveloped in our earlier published study of nicotine (Ferreira et al. 2000). Histologic verification. At the end of the experiment all rats werekilled with an overdose of pentobarbital. Brains were removed andfixed in a mixture of 4% paraformaldehyde and 20% sucrose for atleast 24h. The brain was cut into 50µm thick coronal sections andstained with Neutral Red. The location of nuclear groups was studiedin relation to microinjection sites using the atlas of Paxinos & Watson(1986). Data analysis. Data were analysed using the Chart Software fordata analysis made for MacLab (ADInstruments). Before micro-injections were performed, the lowest points of the intragastricpressure (IGP) trace were obtained over a 5min control period, and asingle value was calculated as the mean of all of these points and usedas an index of gastric tone. After microinjections into the mNTS, themaximum value in the trace was taken as the largest increase ingastric tone. The percentage change from baseline in IGP was thencalculated. Data for IGP are reported as percentage change frombaseline since baseline IGP varied between animals. It should benoted that all data that are shown to be statistically significant aresignificant when analysed both as raw data and as percentage changefrom baseline. Data appear as means (percentage change frombaseline for IGP)± S . E . M . To calculate the area under the curve for the M. Ferreira Jr and others  142 J. Physiol. 536.1  phasic contractions appearing in the IGP traces, the MacLab dataanalysis package was utilized. The program calculates the area underthe curve as an integral that is represented by: ∑ ( y_y  Baseline( t)  )  w ∆ t  , where y  _ y  Baseline denotes the first and last data point in theselection and ∆ t  is the sample interval (5min in our study). Student’spaired t  tests were performed in all cases as animals served as theirown controls. Differences were considered significant at P< 0.05. In vitro  brain slice studies Gastric-projecting DMV neurones were labelled as describedpreviously (Browning et al. 1999). Briefly, 12-day-old rat pups of either sex were anaesthetized deeply (indicated by abolition of thefoot-pinch withdrawal reflex) with a 6% solution of 2-bromo-2-chloro-1,1,1-trifluoroethane (Halothane) with air (400–600mlmin _1 ) before an abdominal laparotomy was performed. Duringsurgery anaesthesia was maintained by placing the head of the rat ina custom-made anaesthetic chamber through which the halothanemixture was perfused. Crystals of the retrograde tracer Dil wereapplied to the serosal surface of the gastric fundus, corpus or theantrum/pylorus. The application site was embedded in a fast-hardening epoxy resin that was allowed to dry for several minutesbefore the entire surgical area was washed with warm saline. Thewound was closed with 5/0 silk suture and the animal allowed torecover for 10–15 days.The brainstems were removed as described previously (Travagli et al. 1991; Browning et al. 1999). Briefly, the rats were placed in ananaesthetic chamber and anaesthetized with halothane before beingkilled by severing the major blood vessels in the chest. The brainstemwas removed and placed in oxygenated Krebs solution at 4°C (seebelow for composition). Using a vibratome, six to eight coronalsections (200µm thick) containing the dorsal vagal complex were cutand stored in oxygenated Krebs solution at 30°C for at least 1h priorto use. A single slice was transferred to a custom-made perfusionchamber (volume 500µl) and kept in place using a nylon mesh. Thechamber was maintained at 35°C by perfusion with warmed,oxygenated Krebs solution at a rate of 2.5–3.0mlmin _1 .Prior to electrophysiological recording, gastric-projecting DMVneurones were identified using a Nikon E600-FS microscopeequipped with epifluorescence filters suitable for visualizing Dil.Once the identity of a labelled neurone was confirmed, whole-cellrecordings were made under brightfield illumination. Electrophysiological recording. Whole-cell recordings were madewith patch pipettes (3–8MΩ resistance) filled with a potassiumgluconate solution (see below for composition) using an Axopatch IDsingle electrode voltage-clamp amplifier (Axon Instruments, FosterCity, CA, USA). Perforated-patch recordings were made usingpipettes (3–8MΩ resistance) filled with a potassium gluconatesolution containing gramicidin (see below for composition andSpruston & Johnston (1992) for detailed technique). Recordings were made from neurones unequivocally labelled withDil. Data were filtered at 2kHz, digitized via a Digidata 1200Cinterface (Axon Instruments), acquired, stored and analysed on anIBM PC utilizing pCLAMP8 software. Recordings were acceptedonly if the series resistance (i.e. pipette+access resistance) was<15MΩ. In addition, the neuronal membrane had to be stable at theholding potential, the action potential evoked following injection of depolarizing current had to have an amplitude of at least 60mV andthe membrane potential had to return to the baseline value followingthe action potential after-hyperpolarization. Electrical stimulation. Bipolar tungsten electrodes were used toelectrically stimulate the centralis and medialis subnuclei of the NTS.Paired stimuli (0.1–1.0ms, 10–500µA; 50–300ms interval) wereapplied every 20s to evoke submaximal excitatory (EPSCs) orinhibitory (IPSCs) postsynaptic currents. All experiments involving evoked IPSCs were carried out in thepresence of 1m M kynurenic acid (to eliminate spontaneous andevoked EPSCs) and with 0.5m M QX314 in the internal pipettesolution (to eliminate antidromically evoked action potentials).QX314, 0.5m M , was also used in the internal pipette in experimentsinvolving evoked EPSCs. A minimum of six control EPSCs or IPSCswere obtained prior to each drug application. Drugs were applied tothe bath via a series of manually operated valves. To assess theeffects of drugs, each neurone served as its own control (i.e. theresults obtained after administration of a drug were compared tothose before administration using Student’s paired t  test). Results areexpressed as means± S . E . M .Significance was set at P< 0.05. Drugs and solutions. Materials used for the in vivo rat studies wereas follows: D -glucose, L -glucose, urethane, a -chloralose, Fast Greendye and dexamethasone. All were purchased from Sigma ChemicalCo. (St Louis, MO, USA) except dexamethasone which was purchasedfrom Elkins-Sinn (Cherry Hill, NJ, USA). Materials used for the in vitro  studies were as follows: extracellularsolution materials (Krebs solution) (m M ): NaCl, 126; NaHCO 3 , 25;KCl, 2.5; MgCl 2 , 1.2; CaCl 2 , 2.4; NaH 2 PO 4 , 1.2; and glucose, 5;maintained at pH 7.4 by bubbling with 95%O 2 –5%CO 2 . Equi-molarKrebs solutions were prepared by either substituting sodium chloridefor glucose (as in the case of 2.5m M solution) or vice versa(as in thecase of 11, 15 and 30 m M solutions).Intracellular solution materials (m M ): potassium gluconate, 128; KCl,10; CaCl 2 , 0.3; MgCl 2 , 1; Hepes, 1; EGTA, 1; ATP, 2; and GTP, 0.25;adjusted to pH 7.35 with KOH.Intracellular solution materials for perforated patch (m M ): potassiumgluconate, 128; KCl, 10; CaCl 2 , 0.3; MgCl 2 , 1; Hepes, 1; and EGTA, 1;with 100µgml _1 gramicidin, adjusted to pH 7.35 with KOH.1,1fi-Dioctadecyl-3,3,3fi,3fi-tetramethylindocarbocyanine perchlorate(DilC 18 (3); Dil) was purchased from Molecular Probes (Eugene, OR,USA). 2-Bromo-2-chloro-1,1,1-trifluoroethane (Halothane) and allother materials listed for the composition of the solutions werepurchased from Sigma Chemical Co. RESULTS In vivo studies Gastric motility was decreased and intragastricpressure was increased by microinjection of D -glucosein the mNTS but not in the DMV Unilateral microinjections of D -glucose (60nl of a 10m M solution) into the DMV were performed in eightanaesthetized rats while monitoring intragastric pressureand motility. We did not observe any effect on eitherintragastric pressure or gastric motility in any of theexperiments (Figs 1 and 2). In three experiments, D -glucose (60nl of a 50m M solution) was tested byunilateral microinjection into the DMV. Again, no effectof D -glucose on gastric function was observed (Fig.2).Finally, to provide a more sensitive endpoint (i.e. a higherlevel of baseline gastric motility) for detecting aninhibitory effect of D -glucose microinjected into theDMV, we administered insulin (5units) subcutaneously inthree animals prior to unilateral microinjection of 60nl of  Glucose effects in dorsal vagal complex  J. Physiol. 536.1 143  a 10m M solution of D -glucose into the DMV. Despite theincrease in gastric motility produced by insulin-inducedhypoglycaemia, unilateral microinjection of D -glucoseinto the DMV had no effect on gastric motility (data notshown). The sites in the dorsomedial medulla where D -glucose was microinjected in the experiments describedabove are depicted in Fig.3. As can be noted, all sites werelocated within the confines of the DMV. Contrary to the lack of effect of microinjections of  D -glucose into the DMV, unilateral microinjection of 60nlof a 10m M solution of D -glucose into the mNTS producedinhibition of gastric motility with an immediate onset,but a relatively slow developing increase in intragastricpressure (n= 5; Figs 2 and 4). These effects dissipatedover a period of 15–30min but could be reproduced in thesame experiment by repeated microinjection of 60nl of a10m MD -glucose solution (data not shown). When 60nl of a 50m M solution of D -glucose was microinjectedunilaterally into the mNTS (n= 5), the effect on motilityand intragastric pressure appeared to be intensified(Fig.2). Three types of control experiments wereperformed. First, D -glucose (60nl of a 10m M solution)was microinjected into sites outside the mNTS (Fig.3).These microinjections did not produce an effect onmotility or on intragastric pressure (data not shown). Theother two types of controls were microinjection of either60nl of 0.9% saline solution or 60nl of a 10m ML -glucosesolution into the mNTS or DMV (Fig.3). When 10m ML -glucose was microinjected into the mNTS, a 2.7±1.5%change in intragastric pressure (n= 5; P> 0.05) and a_52±49mmHgs _1 change in the area under the curve (n= 4; P> 0.05) occurred. When 10m ML -glucose wasmicroinjected into the DMV, a 1.2±0.7% change inintragastric pressure (n= 3; P> 0.05) and a _36±24mmHgs _1 change in the area under the curve (n= 3; P> 0.05) occurred. There was no significant differencebetween the responses elicited by saline or L -glucosemicroinjected into these nuclei, ruling out the possible roleof injectate osmolarity in the response of D -glucose. In nocase did these control microinjections affect motility orintragastric pressure (Fig.2). M. Ferreira Jr and others  144 J. Physiol. 536.1 Figure2 . Effects of D -glucose microinjected intothe DMV and mNTS on intragastric pressure(IGP) and the area under the curve (i.e. area of phasic antral contractions) A , percentage change of IGP due to microinjection of saline (designated as 0 glucose), 10 and 50m M (in60nl) D -glucose into the DMV ( 5 ), and the responsedue to microinjection of saline (designated as0glucose), 10 and 50m M (in 60nl) D -glucose into themNTS ( 4 ). B  , change of the area under the curve(mmHgs _1 ) in response to microinjection of saline(designated as 0 glucose), 10 and 50m M (in 60nl) D -glucose into the DMV ( 5 ), and microinjection of saline (designated as 0 glucose), 10and 50m M (in60nl) D -glucose into the mNTS ( 4 ). IGP data arerepresented as the means± S . E . M .of 3–6microinjection sequences. Change in the area underthe curve data are represented as the means± S . E . M .of 3 or 4 microinjection sequences. * P< 0.05 usingStudent’s group t  test. Figure1 Effect of unilateral microinjection of D -glucose (60nlof 10m M solution) into the DMV on intragastricpressure and gastric motility (see text for details).  Ipsilateral vagotomy had no effect on either theinhibition of gastric motility or the increase in intragastricpressure induced by D -glucose (n= 4; Table1). Bilateralcervical vagotomy, however, prevented D -glucose fromproducing an increase in IGP (n= 4; Table 1). We werenot able to evaluate bilateral cervical vagotomy on phasiccontractions because sectioning both cervical vagusnerves per se  abolished all phasic contractions (Table 1).The lack of effect of ipsilateral vagotomy and therequirement for bilateral vagotomy for abolishingresponses evoked from the mNTS was noted before(Ferreira et al. 2000), and is due to the fact that neuronesof the mNTS connect to both the right and left DMVs(Blessing et al. 1991). Finally, the sites in the dorsomedialmedulla where D -glucose was microinjected in theexperiments described above are depicted in Fig.3. Ascan be noted, all sites were located within the confines of the mNTS. In vitro studies Postsynaptic effects: increasing the extracellularglucose concentration induced an outward current inDMV neurones Whole-cell recordings. Whole-cell patch clamp recordingswere made from 77 identified gastric-projecting neurones(35 fundus, 22 corpus and 20 antrum/pylorus-projectingneurones). In 75 of these neurones, the glucoseconcentration was raised from 5to 11m M (n= 8), 15m M (n= 28) or 30m M (n= 49) (note: some neurones wereexposed to more than one concentration of glucose).Twenty-nine neurones (38%) responded with an outwardcurrent, 1neurone (1%) responded with an inwardcurrent with the remaining 47 neurones (61%) showingno effect. The magnitude of the outward current was 12±1.5pA (n= 4), 36±4.6pA (n= 7) and 51±9.7pA (n= 11) Glucose effects in dorsal vagal complex  J. Physiol. 536.1 145 Figure 3. Microinjection sites for D -glucose andcontrol studie sTwo camera lucida drawings of coronal sections of themedulla are depicted to illustrate the location of thepipette tips indicating the microinjection sites of  D -glucose ( w ), D -glucose controls ( 0 ) and L -glucose orsaline ( 1 ). Microinjection sites (tip of micropipette)were located in these two rostral–caudal areas of themedulla, namely at obex, and slightly caudal to obex(i.e. 0.2mm caudal to obex). For clarity, all mNTSinjection sites are shown on the left and DMV sitesare shown on the right. mNTS, medial nucleus of thetractus solitarius; TS, tractus solitarius; CC, centralcanal; DMV, dorsal motor nucleus of the vagus; XII,hypoglossal; AP , area postrema. Figure4 Effect of unilateral microinjection of D -glucose (60nlof 10m M solution) into the mNTS on intragastricpressure and gastric motility (see text for details). Figure5 . Superfusion with 30m M glucoseinduced an outward current that was sensitive totetrodotoxin and picrotoxin Representative traces of gastric-projecting DMVneurones voltage clamped at _50mV showing thatelevating the glucose concentration of thesuperfusing Krebs solution from 5 to 30m M inducedan outward current (current allowed to plateau, thenwash was performed so that the current could returnto baseline levels, typically after 10min) that wasinhibited by the synaptic blocker tetrodotoxin(incubation of 10min; 1µ M , A ) as well as by theGABA A receptor channel-selective antagonistpicrotoxin (50µ M , B)  .
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks