Neurophysiology of Ganglia of Auerbach's Plexus

Neurophysiology of Ganglia of Auerbach's Plexus

J. D. WOOD

Department of Physiology, University of Kansas Medical Center, Kansas City,Kansas 66103

SYNOPSIS. The enteric plexuses of the automatic nervous system may be considered, onthe basis of both function and morphology, to be a simple integrative nervous system ofvertebrate animals. Microelectrcde studies of single unit activity within enteric gangliareveal four distinct types of ganglion cells distinguished on the basis of pattern of spikedischarge. These are (i) burst-type units which spontaneously discharge bursts of spikesat periodic intervals; (ii) fast- and slowly-adapting mechanoreceptors; (iii) tonic-typeunits which respond to mechanical stimulation with prolonged, all-or-nothing trains ofspikes; (iv) single-spike units which spontaneously discharge single action potentials atvariable intervals. The enteric plexuses are adapted for control of the intestinal muscu-lature which behaves as an electrical syncytium activated by myogenic pacemakerpotentials. The mechanism of neural control is integration of continuous neurogenic in-hibition of the inherently excitable musculature.

Formidable technical problems arise inthe study of integrative action of and in-formation processing by groups of neuronswithin the mammalian central nervoussystem, primarily because of the large num-bers of neurons and complexity of connec-tions within the system. Because of thesedifficulties, many comparative neurophysi-ologists, as evidenced by the titles on theprogram of this symposium, study therelatively simple nervous systems foundamong invertebrate species. The gangliawhich constitute the central nervous systemof invertebrate phyla are advantageous forneurophysiological investigation becausethey contain small numbers of identifiableneurons and do not present the difficultiesin identification of connections and inter-actions that are inherent in the mammaliancentral nervous system. Since the propertiesof the constituent neurons are generallythe same for botli groups ol animals, it isassumed that discoveries and concepts de-rived from the less complex systems ofinvertebrates can be extrapolated to func-tional models of the central nervous systemof vertebrates, including man. Studies oninvertebrate systems are justified becauseno integrative system of comparable sim-

Soine of the work described here was supportedby National Science Foundation Grant GB-31292 andKansas University Medical Center General ResearchSupport Grant RR-05373.

plicity and convenience for study has beendemonstrated previously in chordates.

In the present paper, I shall present ex-perimental findings which suggest that theganglia of the enteric plexuses of the au-tonomic nervous system of vertebrates maybe considered to be simple integrative sys-tems analogous to ganglia of the nervoussystems of invertebrates in terms of easeof study and relative simplicity. The addedusefulness of the enteric plexuses for neuro-physiological research is that they representa "simple nervous system" which is alreadypresent in vertebrates and in particular inhomeothermic mammals.

STRUCTURE OF AUERBACH'S PLEXUS

Auerbach's plexus lies between the circu-lar and longitudinal muscle coats of themuscularis externa of the intestine and con-sists of ganglia, interconnected by bundlesof nerve fibers (Fig. 1). This forms theprimary plexus, from which are derived asecondary and a tertiary plexus of smallernerves that supply the smooth muscle. Theliterature describing the morphology ofAuerbach's plexus as well as the other con-stituents of the enteric plexuses is vast andhas been reviewed in depth by Schofield(1968). My discussion of morphology willbe confined to pointing out that the struc-tural characteristics of ganglia of the enteric

973

Dow

nloaded from https://academ

ic.oup.com/icb/article/14/3/973/2008541 by guest on 20 August 2022

974 J. D. WOOD

FIG. 1. Photomicrographs o£ Auerbach's plexus. A,Auerbach's plexus of cat jejunum as viewed with adissecting microscope after vital staining with meth-ylene blue. A strip of longitudinal muscle was re-moved to expose the plexus. C, circular musclelayer beneath plexus; G, ganglion; F, bundle ofnerve fibers connecting two ganglia; L, underside offolded back strip of longitudinal muscle; S, serosalsurface of intact longitudinal muscle. B, Livingneurons of a ganglion of Auerbach's plexus ofguinea-pig jejunum as seen with Normarski optics.C, capillary blood vessel; P, periphery of ganglion;E, recording electrode; G, ganglion cell soma; N,nucleus of ganglion cell; A, axon of ganglion cell.

plexuses more closely resemble the neuraltissue of the central nervous system thanother autonomic ganglia. The entericganglia share with ensembles of neurons ofthe central nervous systems of both verte-brates and invertebrates morphological or-ganization which is common to systems withintrinsic integrative capability, but uncom-mon to sympathetic ganglia which functionchiefly as relay-distribution centers.

Ultrastructural studies of enteric gangliashow the following structural similaritieswith central nervous tissue. The neural andglial elements are densely packed within the

ganglia, and there is virtual absence of ex-tracellular space. Blood vessels, in somespecies, do not enter the ganglia and thebasal lamina surrounding each ganglionmay constitute a physical barrier to entranceof substances from the extraganglionicspace. Extensive synaptic connections withdifferent terminals involving at least fourdifferent kinds of synaptic vesicles occurboth on perikarya and within a dense neu-ropile. A given enteric ganglion cell mayreceive extensive synaptic input from up tothree different kinds of neurons as deter-mined by the features of the synapticvesicles in the presynaptic endings (Gabella,1972).

METHODS

Most of the results that will be reviewedin this paper were obtained by extracellularrecording with metal microelectrodes withinganglia viewed at 35X magnification witha dissecting stereomicroscope after vitalstaining with methylene blue (Fig. \A). Inthese experiments, individual neuronswithin a ganglion were not resolved op-tically; instead, the single units were identi-fied by characteristic patterns of dischargeof action potentials. A second method wasused for visual resolution of and applica-tion of recording electrodes to individual,unstained ganglion cells of Auerbach'splexus. This method utilized a flat, rela-tively transparent preparation of the mus-cularis externa that was obtained byopening a segment of small bowel alongthe mesenteric border and removing themucosal layer. A Normarski differential in-terference contrast optical system provideda clear image of Auerbach's plexus locatedbetween the two muscle layers of the intactmuscularis externa of this preparation (Fig.IB). Details of methodology appear inWood (1970) and Wood and Mayer (1974).

PATTERNS OF ELECTRICAL DISCHARGE

Both spontaneous and stimulus-evokedpatterns of action potentials have beenrecorded from neurons within myentericand submucosal ganglia of segments of

Dow



NEUROPHYSIOLOGY OF GANGLIA OF AUERBACH'S PLEXUS 975

FIG. 2. Discharge of a burst-type unit of Auerbach'splexus of jejunum of cat. A, Burst pattern recordedwith a slow time base. B, Single burst of spikes fromsame neuron recorded with expanded time base. C,Superimposed traces of ten consecutive bursts ofspikes from the same neuron. Each trace of theoscilloscope was triggered by first spike of eachburst. Note that variance of interspike interval isleast for first interspike interval and increases forsubsequent intervals.

intestine in vitro (Yokoyama, 1966; Wood,1970, 1973b; Ohkawa and Prosser, 1972a;Wood and Mayer, 1973a, 1974). These neu-rons can be classified into three main cate-gories on the basis of the pattern ofdischarge of spikes (Wood, 1970). The firstcategory consists of burst-type neuronswhich discharge spontaneously, periodicbursts of spikes with silent interburst in-tervals (Fig. 2). Mechanosensitive neurons,some of which are silent and some of whichshow ongoing discharge of single spikes,can be induced by mechanical distortion ofthe ganglionic structure to discharge spikesat increased frequency and form the secondcategory (Figs. 6, 7). The third categoryconsists of single-spike neurons which show

continuous discharge of action potentialscharacterized by irregular interspike inter-vals and lack of response to mechanicalstimulation (Fig. 6C). Each type of neuronhas been detected in the myenteric plexusof the cat, dog, and guinea pig (Wood andMayer, 1973b).

Burst-type neurons

The steady, systematic discharge of burst-type neurons (Fig. 2) continues for longperiods of time (more than 2 hr in isolatedsegments of intestine) and continues also inpreparations which have been stored in arefrigerator at 5 C for periods up to 48 hrand then rewarmed to 37 C (Wood andMayer, 1973a).

These units usually discharge at highestfrequency at the onset of each burst and theinterspike intervals progressively increasein duration toward the end of the burst(Fig. 2B). Some burst-type neurons, how-ever, show interspike intervals which both atthe onset and termination of the bursts areof long duration relative to the midregionof the burst. For most burst units, thevariance of the duration of each consecutiveinterspike interval is least for the intervalbetween the first and second spikes of eachburst, and the variance of succeeding inter-spike intervals is progressively increased(Fig. 2C). Nonsequential interspike-intervalhistograms for burst units are usually bell-shaped and skewed to the right. Statisticaldata obtained from spike-interval histo-grams appear in Wood (1973ft) and Woodand Mayer (1974).

Measurements of time intervals betweenthe first spike of consecutive bursts of spikes(i.e., the interburst intervals) show consid-erable variance both within an individualspecies and among different species (Woodand Mayer, 1974). Frequency distributionhistograms of interburst intervals usuallyshow a characteristic mode in a given spe-cies; in the cat for example, the modeusually approximates 6 sec. Some burstunits show relatively small variance of in-terburst interval over prolonged spans oftime (Fig. 3ZJ). Other burst units may yielda relatively large standard deviation of in-

Dow



976 J. D. WOOD

5 0 .

zUJ

o

oN -X :SD

2419 55:13 50

Z

O

4 0 . 1 N :

1 * :1 so

3 3 77 03:0.94

1 6 12 18 24 30INTERVAL (SEC)

b 12INTERVAL (SEC)

Fig. 3. Histograms of interval distributions of con-secutive bursts of spikes of two different ganglioncells of Auerbach's plexus of cat jejunum. The oi-dinate lepresents the pioportion of the total num-ber of intenals and the abscissa represents theduration of the interburst intervals. Histograms areconstructed with 1-sec interval classes. Given foreach histogram are number of intervals (N), meaninterval (X) and standard deviation (SD) of thedistribution. The unit of histogram A dischargedat intervals of either 6 sec or at multiple of thisinterval. The unit of histogram B discharged withonly small variation around a mean interburst' inter-vel of 6 sec.

terburst intervals, but, as in Figure 3A, theinterburst-interval histograms show onlysmall variation of the individual peakswhich are multiple and clearly defined at6, 12, 18, and 24 sec. Bullock (personalcommunication) remarked that these neu-rons seem to behave as "clocks" with atiming mechanism that runs continuouslybut with an "alarm" that sometimes fails.The discharge patterns of some burst-typeganglion cells are dominated by cyclicchanges in the spike activity, with eachcycle of activity showing the following se-quential changes in spike discharge: (i) asilent period of relatively long duration;(ii) a regular series of bursts of spikes; (iii)less regular bursts; (iv) continuous dischargeof spikes (Figs. 4, 5).

Several different pharmacological agentshave been applied to the burst-type neuronsin vitro (Wood, 1970; Ohkawa and Prosser,19726). In general, only agents which areactive at cholinergic synapses seem to alterthe ongoing discharge of these units. Thecatecholamines, serotonin, and the ami noacids glycine, glutamate, and gamma-amino-butyric acid produce little or no effects.Acetylcholine stimulated some, but not all,

of the tested burst units (Wood, 1970;Ohkawa and Prosser, 19726), and bothmuscarinic and nicotinic receptor blockingdrugs stop the discharge of some, but notall, of the tested burst units. Excitatoryeffects of applied acetylcholine are antago-nized by atropine. However, concentrationsof atropine two to three orders of magnitudegreater than concentrations required toblock muscarinic receptors of the longi-tudinal muscle of the intestine are requiredfor blockade of burst-type cells. This mayresult from diffusion barriers which sur-round the ganglia (Gabella, 1972) and op-pose entrance of certain pharmacologicalagents in much the same manner as thesheath surrounding the ventral nerve cordof arthropods (Bullock and Horridge, 1965,p. 1133-1134).

The ease with which it is possible to

H I I I D I

mm*

4-H-H- f-4

HG. 4. A record from Auerbach's plexus of guinea-pig jejunum showing conversion from a burstpattern of discharge to a pattern of continuous dis-charge of spikes and reversion back to a burstpattern. Record is continuous from top to bottomtrace. (From Wood, 1973ft.)

Dow




20

10

mlklSEQUENTIAL INTERVALS

FIG. 5. Computer plot (Calcomp-IBM 1130) of cy-clic variation of sequential interburst intervals fora burst unit of Auerbach's plexus of guinea-pigjejunum. Same unit as Figure 4. Ordinate is timeinterval between first spike of sequential bursts.Abscissa represents 575 consecuthe interburst inter-

vals which occurred over a time span of 23 min.Ordinate values of zero indicate periodic conversionsfiom burst pattern to continous discharge of spikes.Note long-duration interburst interval followingeach episode of continuous firing. (From Wood,1973b.)

record electrical activity from single burstunits for prolonged periods of time permitssequential testing of several pharmacologi-cal agents on the same neuron. In suchexperiments, a population of burst neuronsemerged in which the firing patterns ofthe cells were unaltered by a number ofpharmacological compounds which are im-plicated as being either neurotransmittersubstances or antagonists of transmission atspecific chemical synapses in the centraland autonomic nervous systems (Wood,1970). This suggested to me that these burst-type neurons which do not respond to anyof the suspected transmitter substances maybe spontaneous oscillators which do notreceive synaptic input from other neurons,and that the second population of burst-type neurons, which are activated by acetyl-choline and blocked by atropine, may bedriven by input from the spontaneous oscil-lators. Interactions between different burst-type cells have been recorded (vide infra)and lend support to this idea.

Insufficient knowledge of the mechanismsof genesis of the burst-type discharge pre-

cludes adequate understanding, at the pres-ent time, of the patterned activity of theseneurons. The formation of burst activitycould be dependent either upon interneu-ronal interactions within an ensemble ofneurons, such as in the thoracic gangliaduring control of flight in insects (Wilson,1970), or upon single neurons with endoge-nous burst generating capability, an ex-ample of which would be the burstgenerating neurons of Aplysia (Strum-wasser, 1967). If the implications of thepharmacological studies are correct, and ifsome of the myenteric burst-type neuronsdo not receive synaptic input mediated bya chemical substance, then the latter mech-anisms of genesis of burst discharge wouldbe most probable.

Some reports of results obtained by intra-cellular recording from myenteric neuronshave appeared recently (Hirst et al., 1972;Holman et al., 1972; Nishi and North,1973), but none has reported any continu-ous patterned activity. Many of the neuronsthat were impailed during these intracellu-lar studies responded to electrical stimula-

Dow



978 J. D. WOOD

tion with a single spike followed byprolonged hyperpolarization lasting for pe-riods up to 20 sec. This kind of behaviorwould prevent repetitive firing in thesecells.

The cyclic alteration of discharge patterndisplayed by certain burst-type neurons(Figs. 4, 5) could be the result of inter-neuronal interactions. The unstable dis-charge patterns of these units are reminis-cent of results obtained by Perkel (1964)with computer models. Perkel's findings in-dicate that when a neuron is receiving inputfrom two pacemaker neurons, the regularityof burst discharge of the follower cell de-pends on the coefficient of variation ofinterval between both sets of inputs, andthe burst pattern rapidly deteriorates torandom spiking with increased variance atthe two inputs.

Calvin (1972) has analyzed, for mam-malian cortical neurons, mechanisms of pre-synaptic production of burst discharge inpostsynaptic neurons. When he assumed alinear relationship between firing frequencyand amplitude of summed EPSP's in thepostsynaptic neuron, his model predictedthat burst activity in at least 20 differentinputs to the postsynaptic cell must developin time in order to generate burst dischargein the postsynaptic neuron. A maximum ofonly six different burst-type units have beenfound within a single enteric ganglion(Wood and Mayer, 1973a).

Mechanosensitive neurons

Three kinds of units within the myen-teric ganglia respond with an increase inrate of firing to mechanical distortion ofthe ganglion. One of the mechanosensitiveunits behaves like a typical slowly-adaptingmechanoreceptor and another like a fast-adapting mechanoreceptor (Fig. 6). Thethird kind of mechanosensitive cell is acti-vated by mechanical stimulation to dis-charge relatively long duration trains ofspikes (Fig. 7). The discharge frequency ofthe third kind of unit is independent ofthe intensity of stimulation, and the dis-charge continues in a set pattern for rela-tively long periods of time after the original

Q

2sec

» [aOiN

4sec

FIG. 6. Examples of discharge of two kinds ofmcchanoreceptors and a neuron insensitive to me-chanical stimulation within Auerbach's plexus of catjejunum. A, Slowly-adapting mechanoreceptor. B,Fast-adapting mechanoreceptor which dischargedonly at onset of stimulation. C, A single-spike typeneuron which continuously discharged spikes anddid not respond to mechanical stimulation. Me-chanical stimulation was a transient forward-reversemovement of glass platinum electrode as indicatedby horizontal bars on the records.

mechanical stimulus is removed. The dis-charge patterns of these cells behave as all-or-nothing events. Once the cell is activatedby the mechanical stimulus, it discharges astereotyped train of spikes independent ofthe original stimulus. When first discovered,this kind of myenteric neuron was arbi-trarily termed a tonic-type mechanoreceptor(Wood, 1970). This terminology is no longersuitable, because it now appears that theyare not first-order mechanoreceptor neu-rons. These cells may be higher order neu-rons which are activated by input frommechanoreceptors (vide infra), and the termtonic-type enteric neuron has been sug-gested as a more appropriate name (Woodand Mayer, 1974; Mayer and Wood, 1974).

Enteric mechanoreceptors (Fig. 6) maybe activated by either a glass probe pressedon the ganglion (Wood, 1970) or by anelectrode with a large diameter tip whichdoes not penetrate the ganglionic sheathand which serves as both recording elec-trode and stimulus probe (Wood andMayer, 1974; Mayer and Wood, 1974).

Slowly-adapting mechanoreceptors (Fig.

Dow



NEUROPHYSIOLOGY OF GANGLIA OF AUERBACH'S PLEXUS

FIG. 7. Discharge of a tonic-type enteric neuron inAuerbach's plexus of cat jejunum. A, Discharge ofof a train of spikes in response to mechanical stim-ulation (horizontal bar) followed by conversion toa burst pattern of discharge. Stimulation was tran-sient forward-reverse movement of a glass-platinumelectrode with tip diameter of 25/jin. Upper three

6A) show sustained discharge without signsof adaptation during a stimulus of constantintensity, and the frequency of discharge isdirectly related to the intensity of stimula-tion (Wood, 19736; Wood and Mayer,1974). These cells may be activated re-peatedly over periods up to 3 hr and may ormay not show ongoing discharge prior tostimulation. Fast-adapting mechanorecep-tors (Fig. 6B) give an intensity-dependentdischarge at the onset of the stimulus andquickly fatigue during a sustained stimulus(Mayer and Wood, 1974). Rarely, with ourtechniques, do these cells show an "off-discharge."

Mechanosensitivity is specific to mechano-

traces are continuous records. Fourth trace is con-tinuous with third trace and shows consecutivebursts of spikes with time between consecutivebursts given in sec. B, Waveform of an action po-tential of the train-like discharge pattern. C, Wave-form of an action potential of the burst-likedischarge pattern. (From Mayer and Wood, 1974.)

receptors and is not a general property ofall myenteric ganglion cells. Other neuronswithin the ganglia show ongoing dischargewhich may be similar to that of the me-chanoreceptors, but these neurons are notexcited by the same mechanical stimulationwhich increases firing rate of the mechano-receptors (Fig. 6C).

The behavior of these enteric mechano-receptors is very much like that of gastro-intestinal mechanoreceptors which dis-charge in afferent fibers of the vagus andsplanchnic nerves (Gernandt and Zotter-man, 1946; Paintal, 1954; Iggo, 1957; Bes-sou and Perl, 1966; Mei, 1970; Sharma etal., 1972; Ranieri et al., 1973). This suggests

Dow



980 J. D. WOOD

2 4.

2 2 .

2 0 .

1 8 .

1 6 .

1 4 .

1 2 .

1 0 .

S .

6 .

4 .

2 -

7A A>7

O 0«> «7TO

^ • •A o 7 7 »

A AO 7 • «

O •

*AA " 70 A »

OA 7O A 7

0O •

A 7 7 7O

O A

O A A » •O 0 A • «

i 4 i s 10 to f iTSi iT I M E ( S E C )

* 26 * 3b 32

FIG. 8. Frequency of discharge of tonic-type neuronsof Auerbach's plexus of cat jejunum plotted at 1-sec time intervals following the initiation of dis-charge. Four different units are represented on thegraph. (From Mayer and Wood, 1974.)

that the same mechanoreceptors may pro-vide information to both the central nerv-ous system and the integrative network ofthe enteric ganglia.

The characteristics of the mechanosensi-tivity of tonic-type enteric neurons are un-like intestinal mechanoreceptors andmechanoreceptors elsewhere in the body.These neurons appear not to encode di-rectly either the time derivative or staticintensity of the stimulus. Instead, theyseem to signal, with a stereotyped trainof spikes, only the occurrence of me-chanical distortion of the ganglion. Oncethe tonic-type units are activated (Fig.7/4), they generate the characteristic dis-charge pattern independent of the originalstimulus, and when the discharge stops, thecells appear to be refractory to reactivationfor a period of 0.5 to 2 min (Wood andMayer, 1974; Mayer and Wood, 1974). Fol-lowing the refractory period, the statisticalparameters of a subsequent spike train re-semble the parameters of the initial re-sponse. Interspike-interval analysis of thespike trains of tonic-type cells (Fig. 8) showsthat the impulse frequency is usually rela-tively constant at or shortly after the onsetof the spike train; then the discharge fre-quency progressively decreases as a linearfunction of time until firing terminates(Mayer and Wood, 1974).

Close association of discharge of slowly-

adapting mechanoreceptors with dischargeof tonic-type neurons has been observed(Wood, 1970; Wood and Mayer, 1974;Mayer and Wood, 1974). The nature of theclose association of the discharge of the twotypes of cells suggests that tonic-type neu-rons may be triggered by input derivedfrom the mechanoreceptors.

The tonic-type neurons may fire also inburst-type patterns. The neuron illustratedby Figure 7 displayed a capacity to dis-charge in the characteristic pattern of bothtonic-type neurons and burst-type neurons.Observations also have been made in whichthe burst-type discharge of such cells arecoupled temporally to discharge of slowly-adapting mechanoreceptors (Wood, 1973&).

The overall behavior of the tonic-typeneurons is not consistent with these cellsbeing first-order receptor neurons; it seemsmore probable that they are higher-orderneurons. No counterpart of the tonic-typetrains of spikes have been reported in cen-trally directed sensory fibers from the gut,suggesting that these cells may be involvedonly in intrinsic integrative mechanisms ofthe gut.

Single-spike neurons

Single-spike units show ongoing dischargeof action potentials at relatively low fre-quencies with no consistent pattern to theactivity (Fig. 6C). There is large fluctuationin duration of interspike intervals pro-ducing broad, flat-topped interspike-inter-val histograms with relatively large valuesfor the coefficient of variation (Wood,1973b; Wood and Mayer, 1974).

Single-spike units have noteworthypharmacological properties. Their rate ofdischarge is increased by exogenous acetyl-choline and other nicotinic agonists (Wood,1970), and discharge rate is reduced bynorepinephrine at alpha adrenergic recep-tors (Sato et al., 1973).

Neuronal interactions

An advantage of multiunit recordingwith relatively large electrode tips in Auer-bach's plexus is that it is possible to record

Dow



NEUROPHYSIOLOGY OF GANGLIA OF AUERBACH'S PLEXUS §81

10 30 50LATENCY (MSEC)

FIG. 9. Temporal coupling between burst-type unitsof Auerbach's plexus of cat jejunum. A, Recordingof coupled discharge of two different units by thesame electrode. B, Frequency distribution histogramof the latent periods between the terminal spike ofthe first unit to discharge and the initial spike ofthe "follower" unit. Given are number of latencyintervals (N), mean latency (X) and standard devi-ation (SD) of the distribution.

simultaneously, with a single electrode, theactivity of two or more units within oneganglion and to investigate relationshipsbetween the discharge patterns of differentneurons (Wood and Mayer, 1973a).

One kind of interaction is sequential tem-poral coupling between the discharge ofpairs of different burst-type neurons (Fig.9A). The second neuron always dischargeda burst of spikes with a fixed latency, greaterthan 20 msec, following the burst dischargeof the first neuron (Fig. 9B). Analysis of theinterspike intervals of each of these coupledburst units reveals distinct differences be-tween the discharge characteristics of thefirst and second neurons of the pair (Wood,1970; Ohkawa and Prosser, 1972a; Woodand Mayer, 1974). The first member of apair usually discharges fewer spikes perburst and at higher frequency than thesecond member. The coefficient of variationcomputed from interspike-interval histo-grams is smaller for the first cell, andinterspike-interval histograms of the secondcell of a coupled pair show a greateramount of positive skewness. Interspike-interval histograms for coupled burst-typeunits usually show two distinct peakscorresponding to differences in dischargefrequency of the two different cells (Fig.10). The time intervals between eachpaired discharge of these coupled neuronsare similar to the interburst intervals ofsingle burst units that on the electrical re-

FIG. 10. Interspike-interval histogram for temporallycoupled burst units of Auerbach's plexus of catjejunum is bimodal, corresponding to differentfrequency of spike discharge of the two neurons.

Same neurons as in Figure 9. Ordinate of histogramis number of intervals and abscissa is interval dura-tion in msec. Computer bin width 0.5 msec.

Dow



982 J. D. WOOD

FIG. 11. Synchronous discharge of two different neu-rons in Auerbach's plexus of the cat jejunum. A,Discharge pattern recorded with a slow time base.B, Single burst recorded on expanded time baseshows that each burst consists of two spikes fromone and a single spike from a second cell. C, Super-imposed traces of 12 consecutive groups of spikesfrom the same two neurons. Each trace of the oscil-loscope was triggered by first spike of each group.Note relatively constant latent period between firstlarge spike and smaller amplitude, triphasic spike.

cordings do not show temporal coupling toanother cell.

The neuronal connections which accountfor temporal coupling of the pairs of burstunits are not understood. The time delaybetween the first and second bursts of apair seems quite long to be attributable to

synaptic delay between a "driver" anddirectly-driven "follower" cell.

Burst-type activity in which each burst iscomprised of spikes from two different unitsalso may be recorded within myentericganglia (Fig. 11). In these situations, it ispossible that presynaptic fibers common toboth neurons produce the synchronous ac-tivity of the different units (Wood andMayer, 1974).

Another kind of neuronal interaction in-volves inhibition of burst-type discharge byactivity of another neuron (Fig. 12). InFigure 12, 35% of the bursts of large ampli-tude spikes were accompanied by actionpotentials of a second neuron, the dischargeof which appeared to inhibit firing of theburst-type cell. The discharge of the in-hibitory units in these cases are not tightlycoupled to a particular phase of the burst,but may occur at various times during aburst (Fig. 12C-12F). The inhibitory dis-charge may also precede the burst-discharge,in which case the frequency of spikes of theburst is reduced (Fig. 12B).

The waveform of the spikes, as recordedextracellularly, of both tonic-type neuronsand mechanoreceptors sometimes displays anotched appearance which could resultfrom either synchronous activation of twounits and super-position of the spikes orfrom composite action potentials of a singleneuron (Wood and Mayer, 1974).

FUNCTIONS OF ENTERIC NEURONS

Some of the enteric neurons receive inputfrom the central nervous system. One groupof enteric nerve cells have nicotinic-cholinergic receptors which are activatedby preganglionic fibers of the parasympa-thetic division of the autonomic nervoussystem, according to classical concepts(Campbell, 1970). These neurons are motorin that they release acetylcholine and excitethe gastrointestinal muscle. Another groupof enteric neurons are activated by fibers ofthe vagus nerves and produce inhibition ofthe muscle of the lower esophagus andstomach (Campbell, 1970). These neuronsrelease at the neuroeffector junction a non-adrenergic inhibitory transmitter substance

Dow



NEUROPHYSIOLOGY OF GANGLIA OF AUERBACH'S PLEXUS

I I I I M I I I I I I i

UM11U^^^^^^^^^^^^^^B ^^^^^^^^^V^^^H ^^^^^^^^^^^^^^P

MM1

nifnffffTfFIG. 12. Inhibition of burst-type discharge by activi-ty of a second neuron in Auerbach's plexus of catjejunum. A, Single burst of spikes showing no in-hibitory discharge of second neuron. B, Discharge ofsecond cell preceding burst discharge reduced fre-quency of discharge of burst unit. C, Discharge of

which may be a purine nucleotide (Burn-stock, 1972). Neurohistochemical observa-tions (Norberg, 1964; Jacobowitz, 1965;Norberg and Sjoqvist, 1966) suggest that avast majority of the post-ganglionic sympa-thetic fibers to the gut synapse directly withenteric ganglion cells.

None of the enteric neurons which re-ceive input from the central nervous systemhas as yet been identified by direct electro-physiological methods. The pharmacologi-cal behavior of the single-spike type neuronsis suggestive of terminal ganglion cells ofclassical parasympathetic pathways. How-ever, it is not certain whether neuronswhich are classified electrophysiologically assingle-spike units do constitute a function-ally homogenous group of cells.

Although the histochemical observationsof numerous noradrenergic synapses onenteric ganglion cells suggest an extensiveinfluence of extrinsic sympathetic nerves on

0.2 SEC

second unit occurred following first spike of burstand inhibited discharge of spikes at onset of burstdischarge. D-F, Same burst unit, occurrence of in-hibitory discharge followed second spike of burst,fourth spike of burst and sixth spike of burstrespectively. (From Wood and Mayer, 1974.)

these cells, only the electrical discharge pat-terns of the single-spike neurons are alteredby exogenous noradrenaline (Sato et al.,1973). Furthermore, in my laboratory wefind (unpublished) that noradrenaline re-duces the rate of discharge of less than 20%of the single-spike cells to which it is ap-plied during in vitro experiments. Ex-ogenous noradrenaline- does reduce theamplitude of stimulus-evoked, cholinergicEPSP's with no effect on the resting po-tential in intracellular recordings from en-teric ganglia (Holman et al., 1972), andthis inhibition may be a presynaptic effectof noradrenaline on cholinergic terminals(Paton and Vizi, 1969).

There is little doubt that much of theprogramming and integration of the mo-tility patterns of the intestine are performedby intrinsic neuronal mechanisms (cf. Alva-rez, 1946), and it seems that the physio-logical properties of the intestinal muscle

Dow



984 J. D. WOOD

and the intrinsic nervous system should bemutually complimentary. Spontaneous con-tractile activity is an inherent property ofintestinal smooth muscle in the absence ofary neuronal influence. Two basic mech-anisms account for these properties. Firstly,the intestinal musculature displays proper-ties of an electrical syncytium (i.e., a three-dimensional core-conductor). There is noprotoplasmic continuity between the mus-cle cells; however, there are regions offusions of plasma membranes between con-tiguous cells, and these "tight junctions"function as pathways of low electrical re-sistance for propagation of excitation be-tween adjacent muscle fibers (Bozler, 1948;Nagai and Prosser, 1963; Barr et al., 1968;Tomita, 1970). In the absence of any neu-ronal influence, the electrical connectionsbetween cells account for the three-dimensional spread of excitation through-out a sheet of smooth muscle tissue, whichoccurs when a localized stimulus of thresh-old intensity is applied at any point on thetissue (Wood, 1972). Secondly, organizedexcitation of the network of electrically-coupled muscle cells is initiated by myo-genic pacemaker mechanisms which areomnipresent in the intestinal musculature(Prosser and Bortoff, 1968; Bass, 1968).

The structural and functional organiza-tion of transmission from nerve to musclein this system also is specialized and com-patible with the physiological properties ofthe musculature. Adaptations which appearto be of functional significance in neuronalcontrol of the contractile activity of themuscle are as follows: (i) No specializedneuromuscular junctions occur and non-localized release of transmitter substancefrom a relatively small number of neuronsproduces a diffuse action on many differentmuscle fibers (Richardson, 1958; Taxi, 1965;Bennett and Rogers, 1967; Paton and Zar,1968). (ii) The transmitter substances per-sist and exert their effects at the musclereceptors for relatively prolonged periodsof time (Kuriyama et al., 1967; Bennett andBurnstock, 1968; Hidaka and Kuriyama,1969). (iii) The space constant of cable-likeintestinal muscle is equivalent to 10 to 20muscle cell lengths, and a synaptic junction

I'IG. 13. Release of neuronal inhibition from thecircular muscle layer of a segment of cat jejunumby tetrodotoxin. Tetrodotoxin (1.2 X 10-6 g/ml)was added to arrow. Upper trace, mechanical activi-ty; lower trace, electrical activity. After tetrodotoxin,muscle action potentials appear at falling phase ofelectrical slow waves, and the amplitude of eachslow wave-associated contraction is increased. (FromWood. 1972.)

potential thereby encompasses relativelylarge areas of the conducting syncytium(Tomita, 1970). (iv) Since the muscle fibersare electrically coupled, transmitter-inducedelectrical current of an innervated cell canspread into and affect adjacent muscle fiberswhich receive no direct nervous influence(cf. Johansson and Ljung, 1968).

One function of the steady-patterned dis-charge of the burst-type neurons seems to becontinuous inhibition of the spontaneousmyogenic activity of the circular musclelayer of the intestine. In isolated segmentsof intestine with ongoing discharge of burstunits, muscle action potentials and the asso-ciated contractile activity are absent (Fig.13). Myogenic pacemaker potentials are al-ways present and when the neuronal ac-tivity is blocked experimentally withtelrodotoxin (Fig. 13), each cycle of themyogenic pacemaker triggers intense dis-charge of action potentials and large ampli-tude muscle contractions (Wood 1970, 1972;Ohkawa and Prosser, 1972b; Wood andHarris, 1972; Wood and Marsh, 1973; Biberanil Fara, 1973). Many different situationssuch as application of local anesthetic drugs,long periods of cold storage, and congenitalabsence of enteric ganglion cells, all ofwhich involve functional ablation of entericneurons, are associated with conversionfrom a hypoirritable condition of the circu-lar muscle to a hyperirritable state (Wood,1972, 1973rt). When the burst neurons areactive, neither mechanical stimulation

Dow




(Wood, 1972) nor electrical stimulation(Kobayashi et al., 1966; Wood and Perkins,1970) effectively elicit contractile responsesof the circular muscle. Following neuronalblockade, both electrical shock and me-chanical stimulation readily trigger muscleaction potentials and waves of contractileactivity which may be propagated for dis-tances of several centimeters in either direc-tion along the longitudinal axis of anisolated segment of intestine (Wood andPerkins, 1970; Wood, 1972). These observa-tions suggest that blockade of activity ofenteric neurons releases the circular musclefrom an inhibitory influence and permitsexcitation and conduction, mediated bymyogenic mechanisms.

The transmitter substance which is re-leased from the tonically-active inhibitoryneurons does not appear to be a catechola-mine (Wood, 1972; Wood and Marsh, 1973).It is probably the same as the nonadrenergicinhibitory transmitter substance which isreleased from intramural ganglion cells dur-ing transmural electrical stimulation andwhich produces intense hyperpolarizationand inhibition of the intestinal muscle(Bennett and Burnstock, 1968). The trans-mitter substance may be the purine nucleo-tide, ATP (Burnstock, 1972).

Drugs which block muscarinic-type cho-linergic receptors also remove inhibitionfrom the circular muscle and release myo-genic action potentials and contractions.The mechanism of action of atropine andother muscarinic agents appears to be block-ade of cholinergic synapses within the en-teric ganglia, because electrical stimulationof the inhibitory neurons, in the presenceof atropine, again produces inhibition ofthe myogenic activity, and the inhibitoryresponse is antagonized by letrodotoxin(Wood, 1972; Wood and Harris, 1972;Wood and Marsh, 1973).

Since blockade of cholinergic synapsesappears to release the muscle from ongoinginhibition, I have suggested that cholinergictransmission between at least two differentneurons may be involved in the tonically-active inhibitory pathways and have con-structed the model shown in Figure 14 toillustrate possible neural connections and

FIG. 14. Diagram of hypothetical circuitry for in-hibitory nervous control of the circular musclelayer of the small intestine. A, Spontaneously activeoscillator which provides excitatory input to Neuron/i via a cholinergic synapse. B, Non-spontaneouslyactive "follower" neuron which releases a non-adrenergic inhibitory transmitter substance withina bundle of circular muscle fibers. R, Intestinalmechanoieceptor actuates tonic-type neuron. T.Tonic-type neuron when activated discharges pro-longed train of spikes which constitutes citherexcitatory or inhibitory input to Neuron B. M,Bundle of circular muscle fibers. C, Low-resistanceelectrical connections between individual musclelibers.

functional implications of the inhibitorynetwork (Wood, 1972; Wood and Marsh,1973). The evidence for this model is largelyinductive and based on the observationsthat have been reviewed on the foregoingpages.

In Figure 14, Neuron A is a spontaneousoscillator (burst-type unit) which receivesno synaptic input. Neuron B is a nonspon-taneously active neuron which is driven byNeuron A through a cholinergic synapse.Neuron B releases at its terminals a non-adrenergic inhibitory transmitter substancewhich produces hyperpolarization of themuscle and inhibits genesis of muscularaction potentials. Any treatment, such asapplication of tetrodotoxin, which blocksaction potential production in neuronswould halt ongoing release of the inhibitorytransmitter from Neuron B, and the musclewould be released from inhibition. In theabsence of neuronal inhibition, each cycle

Dow



986 J. D. WOOD

of the myogenic pacemaker system triggersmaximal activity in all of the muscle fibers.Blockade of the synapse between Neuron Aand Neuron B likewise would halt ongoingrelease of the inhibitory transmitter sub-stance from Neuron B and release the mus-cle from inhibition. In this situation, directexcitation of the inhibitory terminals ofNeuron B by experimental electrical stimu-lation should again produce inhibition ofthe activity of the muscle. This has beenobserved experimentally (Wood, 1972;Wood and Marsh, 1973).

In this model, Neurons A and B are con-sidered to be burst-type units, because it isthe steady-ongoing discharge of this kind ofganglion cell that constitutes the prevalentneuronal electrical activity in preparationsin which myogenic activation of the circularmuscle layer is continuously suppressed.Neuron A represents the burst-type neuronswith discharge patterns that are not in-fluenced by experimental application of abattery of potential neurotransmitter sub-stances. These cells are likely also to be thesame as the burst neurons which are ob-served to discharge at precisely timed inter-vals (Fig. 3/4). In the experimentalobservations of temporal coupling of dis-charge of two different burst-type cells (Figs.9, 10), Neuron A might correspond to thecell that discharges first in the sequenceand Neuron B could be the coupled "fol-lower." However, this supposition is tenu-ous if it is assumed that the two cells arelinked "in series" by a single synapse, be-cause the observed latency of the couplingseems to be too long to result from only onesynaptic delay (Fig. 9B). Neuron B of thismodel corresponds to the burst-type unitson which acetylcholine is an agonist andatropine, an antagonist. Neuron B ismodeled as a ganglion cell which receivesmore than one different kind of synapticinput. This neuron might correspond to theburst type units which show erratic dis-charge patterns that in some cases appearto result from synaptic input from otherneurons (Figs. 4, 7, 12).

If tonically active inhibitory neurons con-tinuously suppress myogenic activation ofthe circular muscle of the intestine, then it

is implicit that some patterns of motoractivity are dependent upon integrated re-lease of the muscle from inhibition. Mostlikely, the neuronal circuitry of the entericplexuses is connected such that sensoryinformation derived from either chemo- ormechanoreceptors ultimately leads to in-hibitory synaptic effects on the dischargeof the inhibitory neurons (Neuron B) to themuscle, with the consequent effect of releaseof inhibition from the muscle itself. Somemotility patterns of the intestine include aphase of active inhibition of the muscle;these patterns probably involve neuronaltransformation of sensory information intoincreased excitatory input to the intrinsicinhibitory neurons. The increased synapticactivation of the inhibitory neurons wouldproduce greater intensity of inhibition atthe muscle.

Some of the several kinds of interactionswhich have been observed to occur betweenenteric ganglion cells may represent corre-lates of neural connections involved in mod-ulation of the inhibitory control of the mus-cle. An hypothetical arrangement of con-nections between the enteric mechanorecep-tors, tonic-type neurons, and burst-type cellsis included in Figure 14. In this scheme, ac-tion potentials from mechanoreceptors(Neuron R) provide synaptic input whichtriggers discharge of prolonged trains ofspikes in tonic-type neurons (Neuron T).Tonic-type neurons, in turn, provide synap-tic input that may be either excitatory orinhibitory to the secondary burst-type neu-rons (Neuron B). Depending upon thenature of the connections, the prolongedactivity of the tonic-type cells would pro-duce either sustained excitation or inhibi-tion of the inhibitory burst neuron; in theformer circumstance, more intense inhibi-tion of the muscle would ensue and in thelatter, inhibition would be released from allmuscle fibers that are influenced by theparticular inhibitory neuron.

The functional significance of neuronalinhibitory control is evident when consid-ered together with the inherent excitabilityand electrical syncytial properties of theintestinal musculature. The question arisesas to how myoid conduction of excitation is

Dow




regulated in the intact bowel. Peristalticwaves have been observed to start at thepharynx and continue uninterrupted to theanus (Alvarez, 1946). Certainly it is notadvantageous for propagation over the en-tire extent of the gastrointestinal tract tooccur each time a local region of the musclebecomes excited. In fact, a predominantmotor pattern of the intestine is segmen-tation movements involving contractionslimited to narrow bands of circular muscle.

With the assumption of appropriate pro-gramming of the inhibitory nervous com-ponents, it is possible to account forregulation of the distance over which ex-citation spreads within the layer of muscle,the direction of propagation of the excita-tion, and the strength of the contractileresponse. For example, if the ongoing in-hibition were removed from all of the mus-cle fibers confined to the circumference ofa short segment of bowel, then all of themuscle fibers of that segment would beexcited simultaneously by the ever-presentmyogenic pacemaker to contract synchro-nously, and the result would appear as alocalized segmental contraction of largeamplitude. If the neuronal programmingwere such that inhibition was selectivelywithdrawn from only a small proportion ofthe total number of muscle fibers aroundthe same segment, then a smaller number offibers would be depolarized to action po-tential threshold by the pacemaker, and theensuing contraction would be of relativelysmaller amplitude. If the inhibitory inner-vation were integrated in such a way thatinhibition was removed sequentially fromadjacent segments, propagation of myogenicexcitation over greater distances along thelongitudinal axis of the gut would occur,and the contractile response would havethe appearance of a peristaltic wave. If theneural connections provided for unidirec-tional removal of inhibition, then direc-tionality of myoid propagation would beimposed upon the muscular system. Bothforward and reverse peristalsis occur in theintestine, and this implies existence in theenteric nervous system of neuronal circuitryfor selective bidirectional control of propa-gation of excitation within the musculature.

These suggestions arise from experimentalresults which indicate that with progressiveblockade of inhibitory neuronal activity,the amplitude of pacemaker-induced con-tractions progressively increases, and follow-ing complete blockade of the inhibitorysystem, spread of excitation is bidirectionaland extensive (Wood.. 1972: Wood andMarsh, 1973).

Nervous control of the longitudinal mus-cle layer of the intestine appears to differfrom control of the circular muscle. Inin vitro preparations, when the enteric neu-rons are active, the longitudinal musclelayer shows rhythmic contractions, whereasthe circular layer is quiescent (Wood, 1970).Paton and Zar (1968) have suggested thatcontractile responses of the longitudinalmuscle layer may be mediated exclusivelyby acetylcholine released from enteric neu-rons. It is well known that there is a tonicoutput of acetylcholine from the myentericplexus (Paton and Zar, 1968); this acetyl-choline is probably released by the steadyactivity of some of the enteric neurons.

Finally, it is interesting that the digestivetract of the squid also contains spontane-ously active intrinsic neurons which appearto produce tonic inhibition of a myogenicmusculature (Wood, 1969). In most mol-luscs, movement of ingested materialthrough the alimentary system is achievedby ciliary activity. However, in the courseof evolution from continuously feeding,microherbivorous molluscs to macrocarniv-orous predators, the cephalopods acquirednonstriated muscle and an "enteric nervoussystem" within the wall of the gut. Motorfunction in the squid digestive systemclosely resembles that of the mammaliangut and may reflect similar selective pres-sures and adaptations to similar feedinghabits in parallel evolution of digestive mo-tor systems in the chordates and cephalo-pods.

REFERENCES

Alvarez, W. C. 1046. An introduction to gastroenter-ology. Paul B. Hoeber, Inc., Medical Book Dept.of Harper and Bros., Xcw York.

Bair, L., W. Berger, and M. M. Dewey. 1968. Elec-

Dow



988 J. D. WOOD

trical transmission at the nexus between smoothmuscle cells. J. Gen. Physiol. 51:347-368.

Bass, P. 1968. In vivo electrical activity of the smallbowel, p. 2051-2074. In Handbook of physiology,Sect. 6, Alimentary canal. Vol. IV. Amer. Physiol.Soc, Washington, D.C.

Bennett, M. R., and G. Burnstock. 1968. Electro-physiology of the innervation of intestinal smoothmuscle, p. 1709-1732. In Handbook of physiology,Sect. 6, Alimentary canal. Vol. IV. Amer. Physiol.Soc, Washington, D.C.

Bennett, M. R., and D. Rogers. 1967. A study of theinnervation of the taenia coli. J. Cell. Biol. 33:573-596.

Bessou, P., and E. R. Perl. 1966. A movement detec-tor of the small intestine. J. Physiol. (London)182:404-426.

Biber, B., and J. Fara. 1973. Intestinal motility in-creased by tetrodotoxin, lidocaine and procaine.Experientia 29:551-552.

Bozler, E. 1948. Conduction, automaticity and tonusof visceral muscle Experientia 4:312-318.

Bullock, T. H., and G. A. Horridge. 1965. Structureand function in the nervous system of inverte-brates. W. H. Freeman Co., San Francisco, Calif.

Burnstock, G. 1972. Purinergic neurons. Pharmacol.Rev. 34:509-581.

Calvin, W. H. 1972. Synapdc potential summationand repetitive firing mechanisms: input-outputtheory for the recruitment of neurons into epilep-tic bursting firing patterns. Brain Res. 39:71-94.

Campbell, G. 1970. Automatic nervous supply toeffector tissues, p. 451-494. In E. Bulbing, A. F.Brading, A. W. Jones and T. Tomita [ed.], Smoothmuscle. Williams and Wilkins Co., Baltimore, Md.

Gabella, G. 1972. Fine structure of the myentericplexus in the guinea-pig ileum. J. Anat. 111:69-97.

Gernandt, B., and Y. Zotterman. 1946. Intestinalpain: an electrophysiological investigation on mes-enteric nerves. Acta Physiol. Scand. 12:56-72.

Hidaka, T. and H. Kuriyama. 1969. Responses of thesmooth muscle membrane of guinea-pig jejunumelicited by field stimulation. J. Gen. Physiol. 53:471-486.

Hirst, G. D. S., M. E. Holman, C. L. Prosscr, and I.Spence. 1972. Some properties of the neurons ofAuerbach's plexus. J. Physiol. (London) 225:60 P.

Holman, M. E., G. D. S. Hirst, and I. Spence. 1972.Preliminary studies of the neurons of Auerbach'splexus using intracellular microelcctrodcs. Aust.J. Exp. Biol. Med. 50:795-801.

Iggo, A. 1957. Gastrointestinal tension receptors withunmyelinated afferent fibers in the vagus of thecat. Quart. J. Exp. Physiol. 42:130-141.

Jacobowitz, D. 1965. Histochemical studies of theautonomic innervation of the gut. J. Pharmacol.Exp. Ther. 149:356-364.

Johansson, B., and B. Ljung. 1968. Role of myogenicpropagation in vascular smooth muscle responseto vasomotor nerve stimulation. Acta. Physiol.Scand. 73:501-510.

Kobayoshi, M., C. L. Prosser, and T. Nagai. 1966.

Electrical interaction between muscle layers of catintestine. Amer. J. Physiol. 211:1281-1291.

Kuriyama, H., T. Osa, and N. Toida. 1967. Nervousfactors influencing the membrane activity of in-testinal smooth muscle. J. Physiol. (London) 191:251-270.

Mayer, C. J., and J. D. Wood. 1974. Properties ofmechanosensitive neurons within Auerbach's plex-us of the small intestine of the cat. Arch. Ges.Physiol. (Submitted)

Mei. N. 1970. Mechanorecepteurs vagaux digestifschez le chat. Exp. Brain Res. 11:502-514.

Nagai, T., and C. L. Prosser. 1963. Electrical para-meters of smooth muscle cells. Amer. J. Physiol.204:915-924.

Nishi, S., and R. A. North. 1973. Intracellular re-cording from the myenteric plexus of the guinea-pig ileum. J. Physiol. (London) 321:471-491.

Norberg, R. A. 1964. Adrenergic innervation of theintestinal wall studied by fluorescence microscopy.Int. J. Neuropharmacol. 3:379-382.

Norberg, K. A. and F. Sjoqvist. 1966. New possibili-ties for adrenergic modulation of synaptic trans-mission. Pharmacol. Rev. 18:743-751.

Ohkawa, H., and C. L. Prosser, 1972a. Electrical ac-tivity in myenteric and submucous plexuses of catintestine. Amer, J. Physiol. 222:1412-1419.

Ohkawa, H., and C. L. Prosser, 19726. Functions ofneurons in enteric plexuses of cat intestine. Amer.J. Physiol. 222:1420-1426.

Paintal, A. S. 1954. A study of gastric stretch recep-tors. Their role in peripheral mechanisms ofsatiation of hunger and thirst. J. Physiol. (Lon-don) 126:225-270.

Paton, W. D. M., and E. Z. Vizi. 1966. The inhibitoryaction of noradrenaline on acetylcholine outputby guinea-pig ileum longitudinal muscle strip.Brit. J Pharmacol. 35:10-28.

Paton, W. D. M., and M. A. Zar. 1968. The originof acetylcholine released from guinea-pig intestineand longitudinal muscle strips. J. Physiol. (Lon-don) 194:13-33.

Perkel, D. H. 1964. Neurophysiological models:methods and applications, Rand Corp. Mem.RM-4247-NIH.

Prosscr, C. L., and A. Bortoff. 1968. Electrical activi-ty of intestinal muscle under in vitro conditions,p. 2025-2050. In Handbook of Physiology, Sect. 6,Alimentary canal. Vol. IV. Amer. Physiol. Soc,Washington, D.C.

Ranieri, F. N. Mei, and J. Cousillat. 1973. Les affer-enecs splanchniques provenant des mechanore-cepteurs gastrointestinaux et peritoneaux.. Exp.Brain Res. 16:276-290.

Richardson, K. C. 1958. Electron microscopic obser-vations on Auerbach's plexus in the rabbit withspecial reference to the problem of smooth muscleinnervation. Amer. J. Anat. 103:99-136.

Sato, T., I. Takayanagi, and K. Takagi. 1973. Phar-macological properties of electrical activities ob-tained from neurons in Auerbach's plexus. Jap. J.Pharmacol. 23:665-671.

Schofield, G. C. 1968. Anatomy of muscular and neu-

Dow




ral tissues in the alimentary canal, p. 1579-1627.In Handbook of Physiology, Sect. 6, Alimentarycanal. Vol. IV. Amer. Physiol. Soc, Washington,D.C.

Sharma, K. N., H. L. Jacobs, V. Gopal, and S. Dua-Sharma. 1972. Vagosympathetic modulation of gas-tric mechanoreceptors: effects of distension andnutritional state J. Neural. Trans. 33:113-154.

Strumwasser. F. 1967. Types of information storedin single neurons, p. 291-316. In C. A. G. Wiersma[ed.], Invertebrate nervous systems. Univ. ChicagoPress, Chicago.

Taxi. J. 1965. Contribution a l'etude des connectionsdes neurones moteur du systeme nerveux auto-nome. Ann. Sci. Natur. Zool., Paris, Ser. 12, VII:413-674.

Tomita, T. 1970. Electrical properties of mammaliansmooth muscle, p. 197-243. In E. Bulbring, A. F.Brading, A. W. Jones, and T. Tomita [ed.],Smooth muscle. Williams and Wilkins Co., Balti-more, Md.

Wilson, D. M. 1970. Neural operations in arthropodganglia, p. 397-409. In F. O. Schmitt [ed.], Theneurosciences: second study program. The Rocke-feller Univ. Press, N.Y.

Wood, J. D. 1969. Electrophysiological and pharma-cological properties of the stomach of the squid,Loligo pealii. Comp. Biochem. Physiol. 30:813-824.

Wood, J. D. 1970. Electrical activity from singleneurons in Auerbach's plexus. Amer. J. Physiol.219:159169.

Wood, J. D. 1972. Excitation of intestinal muscle by

atropine, tetrodotoxin and xylocaine. Amer. J.Physiol. 222:118-125.

Wood, J. D. 1973a. Electrical activity of the intestineof mice with hereditary megacolon and absenceof enteric ganglion cells. Amer. J. Dig. Dis. 18:447-488.

Wood, J. D. 1973fc. Electrical discharge of singleenteric neurons of guinea-pig small intestine.Amer. J. Physiol. 225:1107-1113.

Wood, J. D., and B. R. Harris. 1972. Phase relationships of the intestinal muscularis: effects of atro-pine and xylocaine. J. Appl. Physiol. 32:734-737.

Wood, J. D., and D. R. Marsh. 1973. Effects ofatropine, tetrodotoxin and lidocaine on reboundexcitation of guinea-pig small intestine. J. Phar-macol. Exp. Ther. 184:590-602.

Wood, J. D., and C. J. Mayer. 1973a. Patterned dis-charge of six different neurons in a single entericganglion. Arch. Ges. Physiol. 338:247-256.

Wood, J. D., and C. J. Mayer. 1973&. Electrical dis-charge of single enteric neurons of the intestineof the cat, dog and guinea-pig. Fed. Proc. 32:377.

Wood, J. D., and C. J. Mayer. 1974. Discharge pat-terns of single enteric neurons of the small intes-tine of the cat, dog and guinea-pig, p. 387-408.IV International Symposium GastrointestinalMotility, Banff, Canada. Mitchell Press, Van-couver, Canada.

Wood, J. D., and W. E. Perkins. 1970. Mechanicalinteraction between longitudinal and circular axesof the small intestine. Amer. J. Physiol. 218:762-768.

Yokoyama, S. 1966. Aktionspotentiale der Ganglien-zelle des Auerbachschen Plexus in Kaninchen-dunndarm. Arch. Ges. Physiol. 288:95-102.

Dow



Dow



Neurophysiology of Ganglia of Auerbach's Plexus

Documents