Electrical stimulation of esophageal smooth muscle and effects of ...

AMERICAN

JOURNAL

OF PHYSIOLOGY

Vol. 217, No. 5, November, 1969. Printed in U.S.A.

Electrical

stimulation

of esophageal

smooth muscle and effects of antagonists GORDON F. LUND AND JAMES CHRISTENSEN Gastroenterology Research Laboratory, Department of Internal LUND,

GORDON F., AND JAMES CHRISTENSEN. Electrical stimulation smooth muscle and effects of antagonists. Am. J. Physiol. 217(5): 1369-1374. 1969.-Smooth muscle strips from the esophageal body of the opossum were arranged to show, separately, responses of longitudinal, circular, and mucosal muscle layers to electrical stimulation by l.O-msec square waves at 30 v. Longitudinal and mucosal muscle contracted throughout the stimulus. In both layers, responses were abolished by tetrodotoxin; hexamethonium, nicotine, propranolol, tolazoline, d-tubocurarine, and methysergide had no effect; atropine only partially antagonized the responses. Strips of circular muscle contracted briefly at the beginning or the end of the stimulus, the “on response” and the “off response.” The on response was best produced with d-c stimulation and was unaffected by drugs. The off response was abolished by tetrodotoxin but not by hexamethonium, nicotine, propranolol, tolazoline, d-tubocurarine, methysergide, and atropine. It is concluded that the responses of longitudinal and mucosal muscle contain both atropine-sensitive and atropineinsensitive components. The on response of the circular muscle is a direct response of the muscle to stimulation. The off response either may be mediated by noncholinergic nerves or could be a rebound contraction following excitation of nonadrenergic inhibitory nerves.

of esophageal

autonomic esophageal

innervation; electrophysiology of smooth motility; gastroenterology; motility; opossum

muscle;

MOVEMENTS OF the esophagus differ from the movements of the stomach and intestine. Also, certain diseases selectively disturb esophageal motor function in man ( 1416, 20). The human esophagus contains mostly smooth muscle ( 1). These facts indicate that esophageal smooth muscle differs from other gastrointestinal smooth muscle either intrinsically or in the pattern of its autonomic innervation, ‘or both. We have used the opossum (DideZ’his virginiana) to study esophageal smooth muscle function because the distal twothirds of that organ contains only smooth muscle; in most laboratory animals striated muscle predominates throughout the esophagus (3, 23). In several preparations of the opossum esophagus, we have shown that the muscle layers of the esophagus differ functionally, and that their movements are integrated in the whole esophagus isolated from its central innervation (8). These functional differences of the muscle layers were most apparent in responses to electrical excitation of isolated strips of esophageal wall. Longitudinal muscle of the muscularis propria and the muscularis mucosae always contracted throughout the stimulus period. We called this the “duration response.” Circular THE

Medicine,

University

of Iowa, Iowa City, Iowa 5.2240

muscle of the muscularis propria never showed a duration response, but usually showed an CCoff response,” a brief contraction following stimulation. Rarely it contracted briefly in an “on response” just after the beginning of the stimulus. The work reported here defines the characteristics of the optimal stimulus for these responses and describes the modification of the responses by drugs to indicate some characteristics of the integrative pathways. METHODS

Adult opossums were anesthetized with intraperitoneal sodium pentobarbital (50 mg/kg) and the esophagus was excised. Muscle strips, 1 X 2 cm, were cut from 4-8 cm above the gastroesophageal junction. Longitudinal strips, with the mucosa removed, displayed activity chiefly of the longitudinal layer of the muscularis propria. Transverse strips displayed activity of the circular layer of the muscularis propria. Separate strips of mucosa were cut longitudinally to study actions of the muscularis mucosae whose cells are oriented mostly longitudinally (8). The strips were vertically mounted in separate 50-ml baths of Krebs solution bubbled with 95 % 02-5 % CO2 at 36-37 C, and were attached by fine silver chains to force-displacement transducers (Grass Instrument Co., model FT.O3C), at a basal tension of 2 g. Tension changes were displayed on an inkwriting polygraph (Beckman type R Dynograph). The Krebs solution contained (MM): Na+ 138.5, Kf 4.6, Ca2f 2 5 Mg2+ 1.2, Cl125, HC0321.9, POd31.2, sop 1.2, glucose 49.3. The tissue was stimulated with two fine stainless steel electrodes, 3 mm apart, which pierced each strip near the fixed end. The anode was nearest the fixed end of the strip except where the reverse position is stated in the RESULTS. A square-wave stimulator (Grass Stimulator model S8, with stimulus isolation unit SIU 4678) delivered IO-set trains of monophasic square waves at intervals of about 1 min. Square-wave frequency, voltage, and duration were varied as described in the RESULTS. Voltage calibration represents direct-current voltage outputs of the stimulator measured before each experiment. The following drugs, dissolved in distilled water, were added to each bath as necessary for the experiments: acetylcholine bromide, atropine sulfate, carbamylcholine chloride, physostigmine sulfate, hexamethonium bromide, methysergide maleate, nicotine sulfate, propranolol hydrochloride, tolazoline hydrochloride, tetrodotoxin, and d-tubocurarine chloride. All drugs were added in volumes of less than 1 ml to achieve bath concentrations expressed as grams per milliliter of the base. Before a drug was added

G. F. LUND

1370

AND

J. CHRISTENSEN

to the bath, three or more control responses were recorded to stimuli delivered at I-min intervals. Test responses were the first three or more responses to identical stimuli delivered at least 5 min after adding the drug. Responses were graded by measuring the maximal amplitude of contractions. RESULTS

Longitudinal Laytr of Muscularis and Muscularis Mucosa

Propria

Longitudinal strips of muscularis propria, with mucosa removed, and longitudinal strips of muscularis mucosa were seldom spontaneously active. Effective electrical stimulation of these strips always produced the duration response, a sustained contraction throughout the stimulus period. Square-wave frequency, voltage, and pulse duration were varied independently to obtain optimal stimulus characteristics for the response. Pulse durations of less than 0.1 msec were ineffective. The amplitude of response was nearly maximal with pulses of 1 .O msec; we used l.O-msec pulses because longer pulses are likely to damage tissues. Responses of near maximal amplitude occurred at 20-30 v. As frequency increased from 1 cycle/set, the response amplitude increased to become maximal at 40-50 cycles/set, and gradually diminished at higher frequencies. Single shocks produced no response. The amplitude at any given frequency was slightly greater if the frequency was increased rather than decreased. Also, amplitudes slightly increased as the strip aged. These factors account, in part, for the variance of the frequency response curve shown in Fig. 1. To examine the effects of drugs on this response, we used 1 0-set trains of square waves of 30 v, 1 .O msec, 45 cycles/set, delivered at 1-min intervals. Tetrodotoxin, 10m7 g/ml, 12Or

T

CONCENTRATION

OF

(grams per milliliter)

TETRODOTOXIN

FIG. 2. Effect

of tetrodotoxin on responses of 3 muscle layers of opossum esophagus to electrical stimulation. Each curve represents combined data from four strips. Longitudinal and mucosal mean duration responses of those muscle layers. Circular means off response of that layer. Stimulus characteristics were those described in text and with optimal frequencies of 45 cycles/set for longitudinal and mucosal muscle and 10 cycles/set for circular muscle; lo-30 min of exposure to tetrodotoxin was allowed before testing. Vertical bars indicate 1 SD from mean. 172f43.6

T

I

I

I

C

10-g

I

I

5x10-9 10-e

I

I

5x10-8 lo-7

CONCENTRATION

1

I

5x to-7 10-6

OF ATROPINE

1

5x10-6

I

10-s

I

5x10-5

J

10-4

(grams per milliliter)

FIG. 3. Effect of atropine on responses of 3 muscle layers of opossum esophageal smooth muscle to electrical stimulation. Curves represent combined data from 4 strips each of mucosal, longitudinal, and circular muscle layers. Stimulus characteristics were 30 v, 1 msec, and 10 cycles/set for circular muscle, 45 cycles/set for other layers. In duration responses of longitudinal and mucosal muscle, amplitude was measured only during lo-set stimulus period. At each drug concentration there was at least 10 min of exposure to drug before test.

FREOUENCY

OF STIMULATION

(cycles

per second)

FIG. 1. Frequency-response curves of 3 smooth muscle layers of opossum esophagus. Stimuli were lo-set trains of 1.0~msec square waves, at 30 v, delivered at 60-set intervals. Curves were obtained by increasing and then decreasing frequency 4 times and stimulating only once at each frequency. Each curve represents combined data from 4 strips. Longitudinal means duration response of the outer longitudinal muscle layer. Mucosa means duration response of mucosal muscle. Circular means off response of circular muscle as described in text. Ordinate shows response amplitude as percent of largest contraction observed for each strip. Vertical bars show 1 SD from mean response at each frequency.

abolished the duration response; the effect was dose related (Fig. 2). Hexamethonium, low4 g/ml, and nicotine, 10V4 g/ml, had no effect. In the outer longitudinal muscle strips, atropine delayed the onset of contractions, reduced their amplitude, and retarded relaxation; the effect was dose related (Figs. 3, 4A and 40). Atropine antagonism to contractions was greatest and often complete during the first 8-10 set of the stimulus period. If a train was prolonged more than 10 set, a contraction developed or progressed to a height which was usually less than controls. Relaxation following the end of a prolonged stimulus was delayed in onset and

ELECTRICAL

STIMULATION

1371

OF ESOPHAGUS

AfROPINE W8

ATRVPiNE‘ k6

ATRVPINE W4 FIG. 4. Effect of an-opine on opossum esophageal smooth muscle responses to electrical stimulation. Traces show contractions (indicated in grams of tension) in response to stimulation for periods shown as black bars under each contraction. Time scale for A, B, and C are shown by upper bracket in the lowerright corner; other bracket is for D. In A, duration response in 4 strips of longitudinal muscle is antagonized by atropine. Note in A that off response of circular muscle is translated into shortening of long axis of strips and thus appears as an artifact which is superimposed on contraction of longitudinal muscle (see also D). If stimulus occurs during delayed relaxation, temporary inhibition occurs (channels c and d in panel III). In B, duration response of 4 strips of mucosal muscle is not completely antagonized by atropine. In C, off response of 4 strips of circular muscle is potentiated by atropine. In D, prolongation of stimulus period of 4 strips of longitudinal muscle in presence of large concentrations of atropine reveals a delay in onset of contraction and a delayed relaxation; off response of circular muscle appears as an artifact. Spontaneous activity of strip recorded by channel b is never seen in fresh strips but is not uncommon in presence of these large concentrations of atropine.

gradual. Another stimulus train delivered during the period of prolonged relaxation facilitated the relaxation (Fig. 4A). The effects of au-opine on the mucosal muscle were similar except that au-opine tended to retard the rate of contraction of mucosal muscle rather than to delay the onset of contraction as observed for outer longitudinal muscle (Fig. 4B). d-Tubocurarine, 10m4 g/ml, had no consistent effect on either duration response. Propranolol, 5 X lo+ g/ml, had no effect, but concentrations of lo4 g/ml depressed the duration response. Tolazoline, 5 X lo+ g/ml, either had no effect on the duration response or slightly potentiated it; higher concentrations caused spontaneous contractions not opposed by au-opine, 10m5 g/ml, or tetrodotoxin, lo-’ g/ml. Methysergide, lop5 g/ml, had no effect. Circular Layer of Muscularis Propria Transverse strips showed no spontaneous activity. Electrical stimulation could elicit two responses in this muscle layer. A brief contraction, the off response, usually occurred after the end of the stimulus. A similar response occasionally occurred at initiation of the stimulus, the on response. They were examined separately. Off response. The electrical stimulus parameters were varied independently to obtain optimal stimulating condi-

tions. The response to varying pulse duration was similar to that of the longitudinal strips; a l.O-msec duration was used. A maximally effective voltage was usually between 15-20 v, but a 30-v stimulus was usually used. The amplitude of the response varied with the frequency. Single shocks produced no response. As frequency increased from 1 cycle/set, the amplitude increased sharply to reach a maximum at 10 cycles/set, diminished sharply at higher frequencies, and was often absent above 40 cycles/set (Fig. 1). As in longitudinal muscle, the response amplitude varied with the frequency to which the strip had been previously exposed. The variation of amplitude, however, was greater for the off response than for the duration response. To measure the effects of drugs on this response, we used 10-set trains of square waves of 30 v, 1.O msec, 10 cycles/set, delivered at 1-min intervals. Tetrodotoxin, lo-’ g/ml, abolished the off response and the effect was dose related (Fig. 2). Neither hexamethonium, lo4 g/ml, nor nicotine, lo4 g/ml, affected the off response. Atropine did not block this response (Figs. 3, 4C, and 5), though it was sometimes depressed at lo+ g/ml (Fig. 5). Higher concentrations of au-opine potentiated the off response (Figs. 3 and 4C). Atropine, low6 g/ml, abolished contractions produced by carbachol, lo-’ g/ml, and acetylcholine, lo+ g/m, in the

1372

G. F. LUND TABLE

260

smooth

1. E$ectS muscle

Of stribs

antagonists

on

to

stimulation

electrical

AND

TeSpOnSeS

J. CHRISTENSEN Of

OpOSSUm

esophageal

240

Kl Electrical stimulation

q Ach stimulation Antagonist

Concentration

Duration Duration Off Response oi Rg:eso”3 o:f Response oi Longitudinal Circular Muscle Muscle Muscle Ef- No. 01i Ef- No. oj fect strips fect strips

Tetrodotoxin Hexamethonium Nicotine Atropine d-Tubocurarine Propranolol Tolazoline Methysergide

8 100 P &

5 80

Effect

No. o Efstrips fect

8 4

6 4

A 0

14 4

0 0

1o-4 1o-6 10-d

4 8

22

4 15 8

0 0 0

4 16 4

0 0 0

4 16 4

5X10-6 5X10-6 10-S

7 6 4

4 4 4

0 0 0

5 8 4

0 0 0

5 8 4

-

20

lo’7 IO”

of

strips

Stimuli were lo-set trains of square waves, 1.0 msec, 30 v. Frequency was 45 cycles/set for duration response of longitudinal and mucosal muscle, and 10 cycles/set for off response of circular muscle. The on response of circular muscle was elicited with a IO-set d-c pulse at 30 v. A indicates antagonism, 0 means no effect.

.:..::..:. ..... ::....::. :::....:. :::....:.

c 5x1@ IO” IO-’ 10-7 Eserine ro-7 Atropine 2x10-9 lo+ Acetylcholine 10-6 loo6 loo6 lo+ lo+

No.

10-T 10-J

60 40

On Response of Circular Muscle

IO” lo* IO-510-4 lo-6

DRUG CONCENTRATIONS (grams per milliliter) 5. Effect of atropine on responses of opossum circular esophageal smooth muscle to electrical stimulation and to stimulation by acetylcholine. Each bar represents mean responses of 4 strips. Stimuli were 10-set trains of square waves of 30 v, 1 msec, 10 cycles/set delivered at 60-set intervals. Vertical lines represent 1 SD from mean. Note that atropine abolished excitation by acetylcholine whereas atropine only depresses off response to electrical stimulation; SOY0 depression of off response by atropine, 10m6 g/ml, was greatest ever observed in these experiments.

A summary of the effects the muscle strips to electrical Table I.

of drugs on the responses stimulation is presented

of in

FIG.

presence of physostigmine, 10B7 g/ml (Fig. 5). d-Tubocurarine, 10m5 g/ml, had no effect on the off response. Propranolol, 5 X 10B6 g/ml, had no effect but the response was depressed at concentrations of low4 g/ml. Tolazoline, 5 X 10m5 g/ml, either had no effect or potentiated the off response; higher concentrations caused spontaneous contractions not antagonized by atropine or tetrodotoxin. Methysergide, 10B5 g/ml, had no effect. On response. The on response was difficult to study because it did not occur consistently. It was rarely observed at 10 cycles/set, 1 msec, 30 v, particularly after the first few stimulus periods. An on response, however, always occurred with direct-current stimulation or square waves at very high frequencies if the cathode was nearer the lower fixed end of the preparation. If the anode was the lower electrode, as was the usual arrangement, direct-current stimulation resulted primarily in an offlike response. This offlike response could be easily distinguished from the off response observed at low frequencies because the latter response was not so altered by reversal of electrode polarity. We could not study drug effects systematically because direct-current stimulation rapidly damages the tissue. The response, however, was not opposed by tetrodotoxin, 10B7 g/ml, hexamethonium, 10m4 g/ml, nicotine, 1 Oa4 g/ml, atropine, 1 Om6 g/ml, d-tubocurarine, low5 g/ml, propranolol, 5 X lo-+ g/ml, tolazoline, lOA g/ml, or methysergide, 10m5 g/ml-

DISCUSSION

Study of the autonomic innervation of esophageal smooth muscle requires techniques to selectively stimulate the various motor nerves. In some organs the nerves may be selected by dissection as in the Finkleman preparation of the duodenum (9) and the preparations of colon described by Garry and Gillespie ( 10, 11). Postganglionic sympathetic motor nerves generally reach these viscera separate from the motor nerves of craniosacral origin. In the esophagus,, however, the sympathetic and parasympathetic nerves mingle in the esophageal plexus before they enter the muscle wall and the sympathetic nerves reach the plexus in numerous delicate strands (22). Thus, it would be difficult to achieve discriminate stimulation of extrinsic nerves. Indiscriminate stimulation of intramural nerve elements is produced by transmural stimulation as described by Paton (24), and the techniques used by Bennett (2) and Campbell (4). W e f ound, however, that the paired intramural electrodes described in METHODS produced satisfactory results. Field stimulation did not work. Isolated strips permitted a clearer separation of the functions of the muscle layers than did such tubular organ preparations as the Trendelenberg preparation and the whole esophagus preparation we have previously described (8). The electrodes implanted in the fixed end of the strips did not greatly hinder contractions, and the contractions involved the entire strip. Electrical stimulation of this preparation may excite both local inhibitory and excitatory nerves as well as act directly on the muscle cells. The responses observed under the various conditions must be viewed as the net result. Tetrodotoxin is considered to block nerve conduction without affecting smooth muscle directly ( 18). The ability of tetrodotoxin to eliminate the off response of circular muscle and

ELECTRICAL

STIMULATION

OF ESOPHAGUS

the duration responses of both longitudinal and mucosal muscle indicates that these responses are nerve mediated. The on response of circular muscle produced by direct current stimulation even in the presence of tetrodotoxin appears, therefore, to result from direct muscle stimulation. Some autonomic nerves, however, may be resistant to the effects of tetrodotoxin ( 12) Hexamethonium and nicotine did not affect the nervemediated off and duration responses. These responses, therefore, are due to excitation of postganglionic elements of the neural plexuses. The absence of antagonism to these responses by tolazoline and propranolol indicate that adrenergic nerves acting upon adrenergic alpha and beta receptors are not involved. The depression of responses by large concentrations of propranolol can be attributed to the local anesthetic properties of this drug (21). The potentiation of these responses and the excitation produced by large concentrations of tolazoline suggest that there may be excitatory adrenergic alpha receptors in this tissue. Excitatory alpha receptors have been identified in esophageal smooth muscle in the opossum (unpublished data) and in other species (5, 6). The excitation produced by tolazoline is not nerve mediated since tetrodotoxin did not alter this action. Since methysergide did not alter the responses to electrical stimulation, serotonin apparently is not directly involved. The duration responses of both longitudinal and mucosal muscle are antagonized by atropine, indicating that these responses involve cholinergic mechanisms acting on musProlonged electrical stimulation, howcarinic receptors. ever, could overcome the effect of atropine, and atropinized strips excited by prolonged stimulation were slow to relax. A delayed a tropine-resistant response to vagal stimulation in longitudinal esophageal smooth muscle of the chicken was reported recently by Hassan ( 13). It was surprising that atropine did not obliterate the off response of circular muscle. It is likely that atropine could penetrate to the circular layer in sufficient concentration to effect blockade; similar concentrations of atropine effectively blocked contractions induced by acetycholine and carbachol. Occasionally, atropine did depress the off response slightly so that some of the response is attributable large doses to cholinergic nerves. In man, by comparison, of atropine given intravenously weaken but do not abolish

1373 the esophageal responses to swallowing ( 17). Two different mechanisms m ight be postulated to explain the atropineresistant off response of the circular muscle layer. A noncholinergic motor innervation The existmay be present. ence of noncholinergic motor nerves in the circular muscle of the guinea pig ileum has been suggested by Kottegoda ( 19). Alternatively, excitation of inhibitory nerves may predominate during stimulation and result in a hyperpolarized state of the muscle; the return to the resting potential after stimulation may be sufficiently rapid to cause activation. Thus the off responses could be a rebound initiated contraction. The inhibition, however, would be nonadrenergic since the adrenergic alpha and beta blocking agents, tolazoline and propranolol, did not abolish the off contraction. Bennett (2) and Campbell (4) observed Ccafter-contrac tions” following stimulation in the taenia coli, and they attribute these responses to rebound excitation; these responses were potentiated by atropine. Atropine at high concentrations strongly excites opossum esophageal smooth muscle (7) and the potentiation of the off response by atropine may be due to this effect. In conclusion, the duration responses of the mucosal and longitudinal muscle appear to be mediated by similar means. They exhibit similar pharmacological and- optimal stimulus characteristics. Both have an atropine-sensitive and an atropine-resistant component that are nerve mediated. The off response of the circular muscle is mediated motor nerves by nerves which mav be either noncholinergic or nonadrenergic inhibitory nerves. The on response observed in circular muscle, produced by direct-current stimulation, appears to result from a direct effect on the muscle; the response was not blocked by any of the drugs used. A different and nerve-mediated on response, however, may also be present since early in experiments an on response was occasionally of 10 seen at the low frequencies cycles/set irrespective of electrode polarity. This study was supported by Research Grant AM 11242 and Training Grant AM 5390 from the National Institute of Arthritis and Metabolic Diseases. Present address of G. F. Lund : Dept. of Zoology, University of Iowa, Iowa City, Iowa 52240. J. Christensen is a Markle Scholar in Academic Medicine. Received

for publication

29 November

1968.

REFERENCES The muscle content of the lower 1. AREY, L. B., AND M. J. TREMAINE. oesophagus in man. Anat. Record 56 : 3 15-320, 1933. M. R. Rebound excitation of the smooth muscle cells of 2. BENNETT, the guinea pig taenia coli after stimulation of intramural inhibitory nerves. J. Physiol., London 185 : 124-l 3 1, 1966. 3. BOTHA, G. S. M. The Gastroesophageal Junction. Boston: Little, Brown, 1962, p. 65-81. 4. CAMPBELL, G. Nerve-mediated excitation of the taenia of the guinea pig caecum. J. Physiol., London 185 : 148-159, 1966. Effects of some autonomic J., AND E. E. DANIEL. 5. CHRISTENSEN, Exptl. drugs on circular esophageal smooth muscle. J. Pharmacol. Therap. 159 : 243-249, 1968. J., AND E. E. DANIEL. Electric and motor effects of 6. CHRISTENSEN, autonomic drugs on longitudinal esophageal smooth muscle. Am. J. Physiol. 211: 387-394, 1966. 7. CHRISTENSEN, J., AND G. F. LUND. Atropine excitation of esophageal smooth muscle. J. Pharmacol. Exfltl. Therap. 163 : 287-289, 1968.

8. CHRISTENSEN, J., AND G. F. LUND. Esophageal responses to distension and electrical stimulation. J. Clin. Invest. 48 : 408-419, 1969. B. On the nature of inhibition in the intestine. J. 9. FINKLEMAN, Physiol., London 70 : 145-157, 1930. R. C., AND J. S. GILLESPIE. An In vitro preparation of the 10. GARRY, distal colon of the rabbit with orthosympathetic and parasympathetic innervation. J. Physiol., London 60-6 1P, 1954. The responses of the muscula11. GARRY, R. C., AND J. S. GILLESPIE. ture of the colon of the rabbit to stimulation, in vitro, of the parasympathetic and of the sympathetic outflows. J. Physiol., London 128 : 557-576, 1955. 12. HASHIMOTO, Y., M. E. HOLMAN, AND A. J. MCLEAN. Effect of tetrodotoxin on the electrical activity of the smooth muscle of the vas deferens. Nature 2 15 : 430-432, 1967. contraction of the chicken isolated 13. HASSAN, T. A hyoscine-resistant oesophagus to stimulation of the vagus and descending oesopha34: 205-206P, 1968. geal nerves. B-it. J. Pharmacol.

1374 14. INGELFINGER, F. J. Esophageal motility. Physiol. Rev. 38: 533-584, 1958. 15. INGELFINGER, F. J. The esophagus. Gastroenterolopy 41 : 264-276, 1961. 16. INGELFINGER, F. J. The esophagus, March, 1961 to February, 1963. Gastroenterology 45 : 24 1-264, 1963. 17. KANTROWITZ, P. A.,C.L. SIEGEL,AND T.R. HENDRIX. Differences in motility of the upper and lower esophagus in man and its alteration by atropine. Bull. Johns Hopkins Hosfi. 118 : 476-491, 1966. 18. KAO, C. Y. Tetrodotoxin, saxitoxin and their significance in the study of excitation phenomena. Pharmacol. Rev. 18 : 997-1049, 1966. 19. KOTTEGODA, S. R. Are the excitatory nerves to the circular muscle

G. F. LUND

20. 21.

22. 23. 24.

AND

J. CHRISTENSEN

of the guinea pig ileum cholinergic? J. Physiol., London 197 : 1718P, 1968. KRAMER, P. The esophagus. Gastroenterology 49 : 439-463, 1965. LEVY, J. V. Myocardial and local anesthetic actions of betaadrenergic receptor blocking drugs : relationship to physicochemical properties. European J. Pharmacol. 2 : 250-257, 1968. MITCHELL, G. A. G. The nerve supply of the gastroesophageal junction. &it. J. Surg. 26: 333-345, 1938-1939. OPPEL, A. Lehrbuch der vergleichenden mikroscopischen Anatomie der Wirbeltiere. Jena : G. Fischer, 1897, vol. 2, p. 117-148. PATON, W. D. M. The response of the guinea pig ileum to electrical stimulation by coaxial electrodes. J. Physiol., London 127 : 4041P, 1955.