Myogenic and Neural Control Systems for Esophageal Motility

Vol. 73, No.6 Printed in U.S A .

73:1345-1352, 1977 Copyright © 1977 by the American Gastroenterological Association

GASTROENTEROLOGY

MYOGENIC AND NEURAL CONTROL SYSTEMS FOR ESOPHAGEAL MOTILITY s.

K.

SARNA, PH.D.,

E. E.

DANIEL, PH.D., AND

w.

E.

WATERFALL, M.D.

Departments of Electrical Engineering, Surgery and Neurosciences, McMaster University, Hamilton, Ontario, Canada

The role of myogenic and neural control systems in esophageal motility was studied in anesthetized opossums by applying electrical pulses at 20 to 40 v, and 100 to 400msec pulse width directly to muscle layers, by cervical vagal stimulation (10 to 25 pulses per sec, 0.5 to 5-msec pulse width, 10 to 40 v), and by balloon distention. Direct muscle stimulation resulted in a propagated contraction in 13 of 22 opossums (proximal propagation from a distal stimulation site and vice versa). The velocity of propagation was of the same order of magnitude as that of a spontaneous swallow (<5 em per sec). The propagated contractions were not blocked by intravenous or close intraarterial atropine and hexamethonium or by intraarterial tetrodotoxin. Smooth muscle depolarization by intraarterial KCl or tetraethylammonium brought about propagated contractions in those opossums that did not show this in the first place. Generally, these propagated contractions could also be initiated in these opossums by applying 2 to 3 stimulating pulses 1 to 1.5 sec apart. The propagated contractions in response to direct muscle stimulation were observed in all opossums 2 to 20 min after death. Cessation of vagal stimulation and balloon relaxation produced "off-responses" which were blocked by tetrodotoxin. The propagation of off-responses was much faster than the swallowinduced peristaltic contractions. In conclusion, the myogenic control system in the esophagus is capable of producing propagated contractions independently which resemble normal esophageal peristalsis in propagation velocity, and may be the underlying system responsible for it. This system may, however, be modulated by the extrinsic and intrinsic nerves. and reanastomosis, 22 and muscle bath studies. 23 Recently, records have been made from single efferent or afferent vagal fibers during peristalsis. 7• 10 The studies to date indicate that motility in the striated muscle portion of the esophagus is controlled almost entirely by activity programmed in the central nervous system in the sense that extrinsic nerve denervation abolished peristalsis, and bolus deviation or section had little effect on the propagation of deglut itive contraction. 19-21 A well organized and complex system of central nervous system control has been elucidated. 17- 18 This system has feedback from the esophageal bolus but can operate without it. In the smooth muscle portion of the esophagus, however, vagotomy has little effect on peristalsis and both bolus deviation and transection with reanastomosis affect its performance. 22 Such findings show that control of peristalsis in this part of the esophagus depends upon local input from the descending bolus, but Roman et al. 17• 18 have shown that in an intact esophagus, the neurogenic control may also be present. Because myogen ic control operates in the rest of the Received January 21, 1977. Accepted July 13, 1977. gastrointestinal tract, we have evaluated the hypotheAddress requests for reprints to: Dr. S.K. Sarna, Department of sis that the myogenic control has also a role to play in Surgery, McMaster University Medical Center, 4W8, 1200 Main esophageal motility, and the deterioration in performStreet West, Hamilton, Ontario, Canada LSS 4J9. This study was supported by the Medical Research Council of ance after transection or bolus deviation in smooth Canada. muscle esophagus may be attributable to the impair-

The control of esophageal motility has been assumed to be primarily a neurogenic phenomenon. This assumption has seemed reasonable because in most species the esophagus and the fundus are the only two regions in the gastrointestinal tract which apparently do not have the omnipresent electrical control activity, also known as slow wave activity or pacesetter activity . 1 The electrical events of electrical control activity have come to be recognized as the basic elements of the myogenic control of motility. Also, in most mammals, the smooth muscle forms only a part of the esophagus; indeed, in some instances like the dog, only the lower sphincter is smooth muscle. An important myogenic control is not typical of striated muscle. A considerable amount of work has been done to elucidate the nature and role of extrinsic and intrinsic nerves and the afferent-efferent vagal loop in the control of esophageal motility, using techniques such a neural stimulation,z-12 vagal sectioning, 13-18 bolus transmission with and without deviation,19-21 transection

1345

1346

Vol. 73 , No.6

SARNA ET AL.

ment, altered input, or propagation of the myogenic control activity. In view of this hypothesis, this study was undertaken to determine whether an independent myogenic control system does exist in the smooth muscle portion of the esophagus, and how it interacts with the extrinsic and intrinsic neuronal controls to control esophageal motility. Methods Healthy North American opossums (2.5 to 4 kg) anesthetized with 30 mg of ketamine per kg (given intramuscularly) and a 1.5-ml per kg solution of chloralose (2%) and urethane (10%) given intraperitoneally were used in all experiments. The abdominal cavity was opened by a midline incision extending 10 em distally from the sternum and the thoracic cavity was opened by the section and removal of 2 to 3 ribs. Up to four strain gauges, similar to those of Bass and Wiley24 were sutured on the smooth muscle portion of the esophagus, two in the thoracic portion, and two in the abdominal portion (fig. 1). Artificial respiration was started just before opening the chest. Up to four bipolar silver wire electrodes (0.015 em diameter and 0.7 em length, 0.5 to 1.0 em apart) were also implanted intramuscularly to record from and stimulate smooth muscle. In 16 of 22 experiments, only the abdominal cavity was opened to minimize alteration of control systems and to reduce the mortality. In this case, only two strain gauges (SG3 and SG4) and two or three electrodes (E2, E3, and E4, or E2 and E4) were used. The mechanical and electrical activities were recorded on a six-channel curvilinear Beckman recorder (Beckman Instruments, Inc., Fullerton, Calif.) with lower and upper cut-off frequencies set at 0.16 and 22Hz, respectively, for electrical recordings. The total of 22 opossums were divided into three groups. In group A (12 opossums), a set of bipolar stimulating electrodes was inserted below the right or th e left vagus in the neck,

~-BALLOON =EI SG2

0 D

SG3

0

SGI

DIAPHRAGM

= E2 =E3

SG4

0

=E4

FIG. 1. Diagram showing the arrangement of strain gauges and electrodes in opossum esophagus. SGI to SG4 represent the strain gauges andEI toE4, the bipolar electrodes.

SG3 ~ SG" E3

A

B

SG 3

~

SG"

-~~

E3

0 .05 mV

l

-ss.c-

~'\.r-~-

f

i c

FIG. 2. Responses to vagal stimulation and balloon distention. A , vagal stimulation at 25 Hz, 5 msec, and 10 v applied for 5 sec resulted in an initial sustained contraction. There was an "off response" when stimulation was stopped. B , atropine (100 Jl.g per kg given intravenously) abolished the initial sustained contraction, but had no effect on off response. C, balloon distention (15 cc) for 5 sec and its relaxation also produced an off respon se. E3 shows electrical activity associated with these responses. SG3, SG4 , and E3 were 1.5, 3.7, and 2.6 em from lower esophageal sphincter. Heavy black line represents stimulus artifact. Arrows in C show balloon inflation.

and the vagus proximal to it, as well as the other vagus, was tied to prevent afferent stimulation. Vagal stimulation parameters were 10 to 25 pulses per sec, 0.5 to 5-msec pulse width, 3 to 5-sec train duration, and 10 to 40 v amplitude. Direct smooth muscle stimulation was achieved with 1, 2, or 3 pulses of 100 to 400-msec pulse width a nd 20 to 40-v pulse amplitude applied 1 sec apart. In group B (6 opossums), the vagus was left undisturbed. In group C (4 opossums), the vagus was left undisturbed at the beginning, but it was cut later in the experiment. In 6 opossums of group A and all of group B, the left gastric artery supplying blood to the distal 3 to 5 em of esophagus was cannulated for the perfusion of drugs (fig. 1). Branches going to the fundus or duodenum were ligated. A balloon made of thin rubber (1.5 to 2.0 em long) was constructed and inserted into the esophagus through the mouth. It was positioned 2 to 4 em proximal to the diaphragm. It was inflated rapidly with 10 or 15 cc of air. The balloon pressures were recorded from aT-tube in the airline.

Results

Group A opossums (both vagi tied in the neck). The electrical stimulation of the cervical vagus at 10 to 25 pulses per sec, 0.5 to 5-msec pulse width, and 10 to 30 v in 12 opossums resulted in a gradual and sustained contraction simultaneously at all strain gauges (fig. 2A), which correlated with visually observed shortening of the esophagus. When the stimulation was turned off, a distally propagating contraction ensued as shown in figure 2A. The velocity of propagation varied considerably. Only approximate estimates of propagation velocity are given here because of the small distance between strain gauges (<2 em center to center when stretched to allow measurement; the strain gauges themselves were 1 em wide) and because of variations in the

December 1977

1347

MYOGENIC AND NEURAL CONTROL SYSTEMS

amount of stretch required to measure the distance in different opossums. In the distal esophagus the propagation velocity after a stimulus of 5-sec duration varied from 5 em per sec up to contractions appearing to be simultaneous at both strain gauges (>40 observations in 22 opossums). Balloon distention (10 to 15 cc) usually produced no response, but its relaxation produced a distally propagating contraction (fig. 2C). The propagation velocities for 5-sec distention were similar in values and variability (>60 observations in 22 opossums) to those produced by vagal stimulation. Both of these responses have been referred to as "off responses" in the literature. 4 • 25 Direct muscle stimulation at E1 to E4 at 30 to 40 v and 200 to 400-msec pulse width usually produced a contraction at the nearest strain gauge only (fig. 3, A and B). In 5 of 12 opossums, this contraction propagated to the most proximal strain gauge when stimulation was applied at a distal site and to the most distal strain when it was applied at a proximal site (fig. 4, A and B). The velocity of propagation was less than 5 em per sec in both directions. The velocity of propagation during spontaneous swallows (fig. 4D), observed occasionally in some opossums, was also less than 5 em per sec (30 observations). When stimulation was applied at E2 or E3, the contractions propagated in both directions. Atropine (100 to 500 )1-g per kg given intravenously) and hexamethonium (5 to 10 mg per kg given intravenously) did not block the off-responses to balloon stimu-

SG3 ~

- - - - ,,r----- - A

SG3

~

0 .05 mV

SG 4

1

-:ss.cc

0

FIG. 4. Propagated responses to direct muscle stimulat ion in alive opossum and spontaneous swallow. A , distally propagated response to direct muscle stimulation at E2. B, proximally propagated r esponse to direct muscle stimulation a t E4. C, proximally and distally propagated response to direct muscle stimulation at E3, SG3, SG4, E2, E3, and E4 were 3.9, 1.8, 5.3, 2.7, and 0.7 em from lower esophageal sphincter, respectively. D, contractions during a spontaneous swallow.

lus in 12 opossums. Further intraarterial doses of atropine (100 to 400 )1-g) and hexamethonium (10 mg) in 6 opossums did not block this response either. The effect on responses to vagal stimulation was variable. Intravenous atropine (100 )1-g per kg) did not block off-respones in response to vagal stimulation in any of 12 opossums in this series, but blocked or SOl ~--]\_considerably reduced the initial sustained contraction associated with longitudinal muscle3 in all cases (fig. 2B). Intravenous atropine (100 )1-g per kg) and hexameSG 2 - - - - - - - - - - - - - - o.oo 1 mV thonium (5 mg per kg) together blocked these vagal off503 ~ responses in 4 of 12 opossums~ Atropine (100 J.Lg) and -ss.chexamethonium (10 mg) given intraarterially after in904 - ---travenous drugs, blocked the vagal off-response com- .A .. pletely in 1 of 6 opossums and had no effect in the other opossums. SO I In all of the above cases, the effectiveness and distribution of intraarterial perfusions was checked with a 5so 2 ____/"'-. _ __ J.Lg perfusion of carbachol given before atropine. Initially, it caused a large sustained contraction and then for 2 to 4 min, irregularly repeated contractions (fig. 5); each was accompanied by what appeared to be a slow S04 _ _ depolarization wave or control potential and a burst of spikes or response potentials. These effects were blocked by atropine perfusions. £ The above protocol of drugs had no effect on direct FIG. 3. Local responses to direct muscle stimulation in alive muscle stimulation responses whether they were local opossum and propagated response after death. A , local response at or propagating before the administration of drugs. GuaSGI to direct muscle stimulation at electrode E1 in alive opposum. nethidine (6 mg per kg given intravenously) in 4 oposB, local response at SG4 to direct muscle stimulation at E4 in alive sums, which showed only local responses, had no effect opossum. In both cases, propagation did not occur. C, distally on their propagation . propagated response to direct muscle stimulation at E1, 10 min After the above perfusions of atropine and hexameafter death. D, proximally propagated response to direct muscle thonion, tetrodotoxin (TTX; 20 )1-g) was perfused instimulation at E4, 11 min after death. E, proximally and distally traarterially in 6 opossums. It completely blocked both propagated responses to direct muscle stimulation at E2, 12 min after death. SG1, SG2, SG3, SG4, E1, E2, and E4 were 9.3, 7.7, 3.8, the balloon off-responses, and the vagal off-responses 1.4, 10.0, 4.6, and 0.5 em from lower esophageal sphincter, respec- (if they were still present). TTX perfusion had no effect on responses to direct muscle stimulation. tively. I~

~

1348

Vol . 73, No.6

SARNA ET AL.

SG4N E3~,(~~~~~ 1

\

'

°,;,~

l -IOSEC

5.0 U9

FIG. 5. Repeated contractions caused by intraarterial carbachol (5.0 !Lg). Larger doses produced a strong initial contraction followed by these repeated contractions lasting for 2 to 4 min. E3 shows activity-like control potentials superimposed by response potentials. Each of these complexes is associated with a contraction. SG3, SG4, and E3 were 4.3, 1.6, and 2.8 em from lower esophageal sphincter, respectively.

The above opossums, after perfusions of atropine, hexamethonium and TTX, were killed by asphyxiation. Two to 8 min after death, the direct muscle stimulation always initiated propagating contractions, even in those cases in which propagation did not occur during life (fig. 3, C, D, and E) . The propagation was proximal if distal electrode site was stimulated (fig. 3D), distal if proximal electrode site was stimulated (fig. 3C), and originated in the middle and spread in both directions if the middle electrode was stimulated (fig. 3E). The velocity of propagation in both directions was nearly the same and < 5 em per sec. This propagating response lasted for 8 to 10 min and then both propagating and local responses gradually failed - in that order about 20 min after death. In 4 opossums, in which there was no propagating response to direct muscle stimulation, 1 ml of KCl (46 mM) or 10 mg of tetraethylammonium (TEA) was infused intraarterially. This bolus was flushed into the esophagus with 1 to 3 ml of Krebs solution. Initially, it caused a sustained contraction at the two strain gauges. The contraction at the distal strain gauge was stronger and lasted longer. Soon after the sustained contraction caused by TEA ended, direct muscle stimulation resulted in propagation in either direction (fig. 6). This lasted for 4 to 8 min and then propagation stopped. Local responses were still present. Soon after the sustained contraction in response to KCl ended, the local contractions in response to direct muscle stimulation were larger in amplitude, but there was no propagation. As the effect of KCl was disappearing, the contractions were propagating for 1 to 2 min and then propagation stopped again. The period of propagation with KCl was short and could not always be found, but with TEA it was always present and lasted for 4 to 8 min. In these same opossums, a gauze soaked with KCl was placed on the esophagus. This again resulted in propagating responses to direct muscle stimulation. When the gauze was removed, only local responses were observed. Group B opossums (both vagi undisturbed) . In 6 opossums the vagi were left undisturbed. In 5 of these opossums, direct muscle stimulation resulted in propagating responses - proximal propagation from distal stimulation site and vice versa. Intravenous atropine (up to 500 !Lg per kg) and hexamethonium (5 mg per kg) had no effect on responses to direct muscle stimulation responses or on off-responses to balloon distention. Intraarterial doses of atropine (100 fLg) and hexametho-

SG4

~

A

SG3 ~~

SG4 ~ ~ ----..~,,--

0.05 mV

--r;-1

-s Sec-

FIG. 6. Effect of intraarterial tetraethylammonium (TEA) on response to direct muscle stimulation. A and B , direct muscle stimulation at E2 and E4 resulted in local responses only at SG3 and SG4, respectively. C and D , following 10 mg of intraarterial TEA the responses to direct muscle stimulation at E2 and E4 propagated distally and proximally, respectively . The local contractions were also increased in size . This state lasted for 4 to 6 min and then propagation stopped again. Local responses were present as before. SG3, SG4, E2, and E4 were 4.3 , 2.2, 5.1, and 1.1 em from lower esophageal sphincter, respectively.

nium (10 mg) in addition to the above doses had no effect on these two responses. Intraarterial TTX (20 JJ.-g) abolished the balloon responses but did not block the propagation of muscular responses. When these opossums were killed by asphyxiation, the propagation in response to direct muscle stimulation was present and improved at about the time it appeared in group A opossums. Improvement consisted of larger local and propagating contractions. In the 1 opossum that did not show propagating contractions, a 10-mg dose of TEA (given intraarterially) resulted in propagat ing contractions in response to muscle stimulation as before . Group C opossums (effect of vagotomy and repeated direct muscle stimulation). The vagi were initially left undisturbed in these opossums. In 2 of 4 opossums, single pulse direct muscle stimulation did not r esult in a propagating contraction, but when the stimulus was repeated two to three times at 1-sec intervals, propagation occurred (fig. 7). Figure 7, A to D show proximal and distal propagations in response to double stimulus, whereas figure and 7, E and F show that in another opossum a weak propagation was present in response to double stimulus, but it improved to a normal propa-

1349

MYOGENIC AND NEURAL CONTROL SYSTEMS

December 1977

SG3~

vJ~

SG 4

_____!\___ ----;---

~

~

~

~

____!\_

_j

__jv.\ c

B

A

D

SG3 ~

~

~

SG 4

~

~

~

-ss.c0.05 mV

~

1

F

E

G

FIG. 7. Effect of repeated stimulation on response to direct muscle stimulation. A, a single stimulus at E2 resulted in a local response at SG3 only . B, a double stimulus at the same electrode resulted in a propagated response. The local contraction at SG3 was larger than before. C, a single stimulus at E4 resulted in a local contraction at SG4 and a feeble propagated response at SG3. D, double stimulus at the same electrode increased the amplitude of local contraction at SG4 and resulted in a stronger propagated contraction at SG3. E, a single stimulus at E2 resulted in a local contraction only at SG3. F, double stimulus at the same electrode increased the amplitude of local contraction and resulted in a feeble propagated response at SG4. G, triple stimulus at the same electrode resulted in a yet stronger local contraction at SG3 and a stronger propagated contraction at SG4. SG3, SG4, E2, and E4 were 4.2, 2.0, 5.4, and 0.8 em from lower esophageal sphincter (LES) , respectively, in the first opossum (A to D), and 4.0, 1.9, 5.0, and 0.6 em from LES for the second opossum (E to G).

SG3 ~~ SG4

-r ~/L B

A

SG3 ~ 0.05 mV

1

SG4 ~

---tt"--c

D

FIG. 8. Effect of vagotomy on response to direct muscle stimulation. A and B, proximally and distally propagated responses to direct muscle stimulation at E2 and E4, respectively, before vagotomy. C and D, both vagi in the neck were first ligated and then cut. None of these procedures had any effect on proximal or distal propagation except perhaps for velocity of propagation. SG3, SG4, E2, and E4 were 4.0, 2.1, 5.0, and 1.0 em from lower esophageal sphincter, respectively.

gation when triple stimulus was applied. In all of these 4 opossums, both vagi in the neck were first ligated and then cut. None of these procedures had any effect on propagation (fig. 8). Discussion Three mechanisms act in conjunction to control motility of the stomach, the small intestine, and the colon: (1) the myogenic control mechanism; (2) the extrinsic

neural control mechanism; and (3) the intrinsic neural control mechanism. Together they control the occurrence of contractions and their sequence in time and space. In the esophagus, control of peristalsis by extrinsic nerves firing in a centrally determined pattern has been suggested in studies by Roman and Tieffenbach, 17• 18 and at least for the smooth muscle portion of the gullet, control by patterns of activity from intrinsic neuronal mechanisms has also been suggested. 2 • 4 • 13 • 18 Recently Weisbrodt and Christensen23 suggested that a nervous control mechanism may operate because of the distally increasing latency of the off-responses of smooth muscle to nonadrenergic inhibitory nerves. Upon simultaneous nonadrenergic nerve-induced hyperpolarization, more distal smooth muscle in the esophagus responds later, on cessation of stimulation, and hence, the contraction appears to propagate distally. Technically speaking, this is not a propagating contraction because each group of muscles is contracting independently with a built-in timing system. No interaction with neighboring groups of muscle cells is involved, but the time sequence is similar to that in a propagating contraction. The velocity of propagation with this mechanism is, however, faster than in normal swallows. Another mechanism of propagation which has not been seriously considered involves active muscle to muscle propagation of electrical activity and hence contractions. The data from this study support the operation of such a mechanism. This study shows that direct stimulation of the smooth muscle of the opossum esophagus produces contractions which can propagate in a peristaltic or an antiperistaltic fashion at a velocity similar to that of peristaltic contractions produced by swallows. In contrast, contractions produced by cessa-

1350

SARNA ET AL.

tion of vagal stimulation or balloon deflation (off-responses) appeared to propagate considerably faster and sometimes were simultaneous. The off-responses to balloon distention or vagal stimulation were present even when both vagi had been tied proximally (which inhibited central effects), indicating that their origin and propagation did not require central integration or an afferent- efferent vagal loop through the central nervous system. Organization of these responses to balloon distention and vagal stimulation appeared to depend only upon structures within the esophageal wall. They were neurally mediated as has been shown before, 2 • 26• 27 because they were blocked by intraarterial TTX. They were noncholinergic, as even very large doses of intravenous and intraarterial atropine and hexamethonium did not block them in most cases. The neuronal network may involve primarily nonadrenergic inhibitory neurones and the contraction may occur because of hyperpolarization of smooth muscle membranes and a subsequent rebound phenomenon when the stimulus is removed, as suggested by Christensen.26 Vagal stimulation caused firing of the excitatory as well as inhibitory fibers. Sustained "on-contractions" were present and were blocked by atropine in contrast to the "off-contractions." They may have resulted from contraction of longitudinal muscle; if so, the only contractions of circular muscle are off-responses. Vagal stimulation of nonadrenergic inhibitory neurones may overcome any cholinergic excitation of circular muscle resulting in hyPerpolarization, and a rebound depolarization on cessation of vagal stimulation. This outcome was observed when isolated strips of circular muscle from the Australian brush-tailed opossum were studied by Daniel et al, 27 during field stimulation of all intrinsic neurones. The blocking of these responses by intraarterial injections in some cases may be attributable to the local anesthetizing effects of high concentrations of atropine. 28 Cholinergic synapses may not occur in the pathway from vagus to nonadrenergic neurones. Even when the vagal off-responses were blocked, the balloon off-responses still persisted. This suggests that the network of intrinsic nonadrenergic inhibitory neurones has an input from extrinsic neurones, but can be stimulated by distention and initiate contractions independent of the network of extrinsic neurones. Also, the input to the intrinsic network from the vagus may be more susceptible to local anesthesia by atropine than are the elements within this network. In some opossums, there were no propagating responses to direct muscle stimulation. Inasmuch as intraarterial TTX blocked all off-responses but did not restore propagation in these opossums, a tonic inhibitory effect of nerves could not account for the lack of propagation. Furthermore, TTX did not block propagation of contractions from direct muscle stimulation when it was present, implying that the myogenic system is capable of generating and propagating contractions independently of any neural input. Failure of TTX to promote propagating contractions in some opossums excludes the hypothesis that failure

Vol . 73, No.6

of propagation resulted from hyperpolarization after vagal handling. Such hyperpolarization could prevent sufficient depolarization in response to oscillations in neighborning cells to trigger an oscillation and stop propagation. This hypothesis is consistent with the findings that depolarization by TEA or KCl initiated propagation when it was absent and that after death (which should lead to depolarization as electrogenic pumps fail and ions run downhill) propagation always occurred for a period of time. The more likely alternative explanation is that damage or depletion of another mediator inhibited vagal-depolarizing activity. TTX could not restore such a mediator. Another possibility is that vagal handling or other procedures released circulating catecholamine which hyperpolarized the smooth muscle cells. Failure of propagation from excessive depolarization of smooth muscle after vagal handling is inconsistent with the fact that KCl, TEA, and death, which should depolarize the muscle, restored propagating responses. Repeated direct muscle stimulation produced larger local contraction and initiated propagating contractions when single stimulation failed to do so. If a stronger contraction implies a larger depolarization, then the group of cells that depolarized in direct response to stimulation provided a larger input to the neighboring group of cells and hence, overcame the effect of hyperpolarization. The implicit hypothesis is that the esophageal smooth muscle behaves like a chain of bidirectionally coupled "one-shot" oscillators. When depolarized up to the threshold level by a stimulus, be it external or from the neighborning cells, such an oscillator undergoes a cycle of oscillation (control potential) and several rapid oscillations (response potentials). In several instances, we recorded control potentials with superimposed response potentials associated (fig. 5) with contractions. At this time, it is not established that these oscillations correspond to the control and response potentials in the rest of the gastrointestinal tract. However, our recordings and those of Christensen and DanieF9 and Diamant30 show the possibility of an esophageal control potential associated with electrical response activity and contractions. Others have also recorded response potentials associated with contractions. 3 1-34 The chain of events during esophageal motor activity consistent with the hypothesis of a chain of bidirectionally coupled one-shot oscillators can be summed up as follows. A contraction initiated by a swallow is propagated in the striated muscle portion of the esophagus under the control of higher centers acting through extrinsic nerves, 17, 18 and activates smooth muscle possibly by electrical current spread. From there onward, the propagation in smooth muscle is effected mainly by the propagation of electrical activity from cell to cell. This propagation may, however, be modulated by intrinsic and extrinsic nerves. Roman and Tieffenbach 17 • 18 have shown that the higher centers have a weaker control over smooth muscle portion of esophagus as compared to the striated muscle portion. A mechanism involving bilaterally coupled muscle

December 1977

MYOGENIC AND NEURAL CONTROL SYSTEMS

oscillators would not depend upon any 'ITX-sensitive neural input but could be blocked by hyperpolarizing the cells or after vagal handling. Consistent with such a mechanism, our study shows that the contractions can propagate in both directions at about the same velocity. This velocity is similar to that initiated by a swallow, but swallows normally propagate only distally. Directionality of swallow-induced peristalsis could be conferred by refractoriness; i.e., by the fact that the refractory period of proximal oscillators outlasts the period of excitatory input from distal active oscillators. However, during secondary peristalsis, refractoriness would not prevent peristalsis in both directions; thus, this is not a satisfactory basis for directionality. On the other hand, directionality could be conferred by nervous influences (distal inhibition followed by off-contractions) when associated with bolus transit. Janssens et al. 22 have shown that the bolus plays an important role in peristalsis in this region; presumably, it does so by modifying the inherent ability of the muscle to propagate a contraction, a property demonstrated in this study. However, the possibility exists that the operative procedure used for bolus deviation interfered with propagation of muscular activity. Obviously, there is a need to determine whether secondary peristalsis, primary peristalsis, or propagated responses to muscle stimulation utilize similar mechanisms in whole or in part. Also, there is a need to study awake, untraumatized animals to compare these mechanisms operating in their normal state. We conclude that in the smooth muscle esophagus, the myogenic and the neural control systems can interact to control motility, but the role and the mode of action of both of these control systems is different in the esophagus than in the rest of the gastrointestinal tract. In the stomach, the small intestine, and the colon, the spontaneous electrical control activity depolarizes the muscle and, if acetylcholine is also present, a contraction occurs. For the esophagus, the myogenic control system is independently capable of producing and propagating contractions. This system, can, however, be affected by neural input to block prO!"!'igation or to initiate a propagated contraction. Teleologically, a chain of one-shot oscillators, rather than spontaneously active oscillators, may be better suited to serve esophageal function, which is primarily to transfer a bolus rapidly, as compared with slow propulsion and mixing for the rest of the gut. A chain of one-shot oscillators can be more readily coordinated with a voluntary swallow rather than can a system of spontaneously active oscillators oscillating at their own frequency. REFERENCES 1. Sarna SK: Gastrointestinal electrical activity: terminology. Gas-

troenter ology 68:1631- 1635, 1975 2. Christensen J : Patterns and origin of some esophageal responses to stretch and electrical stimulation. Gastroenterology 59:909916, 1970 3. Lund GF, Christensen J: Electrical stimulation of esophageal smooth muscle and effects of antagonists. Am J Physiol

1351

217:1369-1374, 1969 4. Christensen J, Lund GF: Esophageal responses to distension and electrical stimulation. J Clin Invest 48:408-419, 1969 5. Mukhopadhyay AK, Weisbrodt NW: Neural organization of esophageal peristalsis·: role of vagus nerve. Gastroenterology 68:444-447, 1975 6. Car A, Jean A, Roman C: A pontine primary relay for ascending projections of the superior laryngeal nerve. Exp Brain Res 22:197- 210, 1975 7. Jean A, Car A, Roman C: Comparison of activity in pointine versus medullary neurones during swallowing. Exp Brain Res 22:211-220, 1975 8. Crousillat J, Ranieri F, Mei N, et al: Effets de la stimulation des fibres et des mecanorecepteurs splanchniques sur l'activite unitaire du cortex somesthesique. Pfluegers Arch 348:127-137, 1974 9. Jean A: Localisation et activite des neurones deglutiteurs bulbaires. J Physiol (Paris) 64:227-268, 1972 10. Jean A: Localization et activite des neurones deglutiteurs bulbaires. J Physiol (Paris) 64:227-268, 1972 11. Car A: La commande corticale de Ia deglutition. These, illustration (see Reference 12), 1975 12. Car A: La commande corticale de Ia deglutition. These, Faculte des sciences et techniques, Universite de droit, d'economie et des sciences d'Aix-Marseille, 1975 13. Roman C, Tieffenbach L: Motricite de l'oesophage a musculeuse lisse apres bivagotomie. J Physiol (Paris) 68:733-762, 1971 14. Ueda M, Schlegel JF, Code CF: Electric and motor activity of innervated a nd vagally denervated feline esophagus. Am J Dig Dis 17:1075-1088, 1972 15. Carveth SW, Schlegel JF, Code CF, eta!: Esophageal motility after vagotomy, phrenicotomy, myotomy and myomectomy in dogs. Surg Gynecol Obstet 114:31-42, 1962 16. Hwang K: Mechanism of tr ansportation of the content of the esophagus. J Appl Physiol6:781-796, 1954 17. Roman C, Tieffenbach 1 : Enregistrement de l'activite unitaire des fibres motrices vagales destinees a l'oesophage du babouin. J Physiol (Paris) 64:479-506, 1972 18. Tieffenbach L, Roma n C: Role de !'innervation extrinsique vagale dans Ia motricite de l'oesophage a musculeuse lisse: etude electromyographique chez le chat et le babouin. J Physiol (Paris) 64:193-226, 1972 19. Janssens J , Valembois P, Vantrappen G, et al: Is the primary peristaltic contraction of the canine esophagus bolus-dependent? Gastroenterology 65:750-756, 1973 20. Janssens J, Valembois P, Hellemans J, et al: Studies of the necessity of a bolus for the progression of secondary peristalsis in the canine esophagus. Gastroenterology 67:245-251, 1974 21. Longhi EH, Jordan PH: Necessity of a bolus for propagation of primary peristalsis in the canine esophagus. Am J Physiol 220:609-612, 1971 22. Janssens J , Vantrappen G , DeWever Y, et a!: Peristalsis in smooth muscle esophagus after transection a nd bolus deviation. Gastroenterology 71:1004- 1009, 1976 23. Weisbrodt NW, Christensen J: Gradients of contractions in the opossum esophagus. Gastroenterology 62:1159-1166, 1972 24. Bass P , Wiley JN: Contractile force transducer for recording muscle activity in unanesthetized animals. J Appl Physiol 32:567-569, 1972 25. Dodds WJ, Wood JD, Arndorfer RC, et al: Esophageal contractions induced by vagal stimulation (VS) in the opossum (abstr.) Gastroenterology 70:879, 1976 26. Christensen J: Neural control mechanisms in the esophagus: a cryptic motor innervation. In Proceedings of IV International Symposium on Gastrointestinal Motility. Vancouver, Mitchell Press, 1973, p 607-611 27. Daniel EE, Taylor GS, Holman ME: The myogenic basis of

1352

SARNA ET AL.

active tension in the lower esophageal sphincter (abstr). Gastroenterology 70:874, 1976 28. Goodman LS, Gilman A: The Pharmacological Basis of Therapeutics. New York, Macmillan, 1970, p 526-528 29. Christensen J, Daniel EE: Electric and motor effects of autonomic drugs on longitudinal esophageal smooth muscle. Am J Physiol 211:387-394, 1966 30. Diamant NE: Electrical activity of the cat smooth muscle esophagus: a study of hyperpolarizing responses. Proceedings of IV International Symposium on Gastrointestinal Motility . Vancouver, Mitchell Press, 1973, p 593-605

Vol. 73, No . 6

31. Hellemans J, Vantrappen G, Valembois P , et al: Electrical activity of striated and smooth muscle of the esophagus. Am J Dig Dis 13:320-334, 1968 32. Arimori M, Code CF, Schlegel JF, et al: Electrical activity of the canine esophagus and gastroesophageal sphincter: its relations to intraluminal pressure and movement of material. Am J Dig Dis 15:191-208, 1970 33. Hellemans J , Vantrappen G: Electromyographic studies on canine esophageal motility. Am J Dig Dis 12:1240-1255, 1967 34. Inouye T: Electromyographic investigation of the esophagus in animals. Laryngoscope 76:1502-1519, 1966