Origin of Slow Waves in the Canine Colon

Transcription

1 GASTROENTEROLOGY 1983;8: Origin of Slow Waves in the Canine Colon N. G. DURDLE, Y. J. KINGMA, K. 1. BOWES, and M. M. CHAMBERS Surgical Medical Research Institute, Departments of Electrical Engineering and Surgery, University of Alberta, Edmonton, Alberta, Canada The objectives of this work were to determine the origin of slow wave activity in the canine colon, to examine the slow wave characteristics in the circular and longitudinal muscle layers, and to examine the roles played by each of these layers in the generation of this activity. Extracellular electrical activity was recorded in vitro from strips of intact muscle wall and from isolated circular and longitudinal muscle using either multiple electrodes applied to one side of the specimen or two electrodes applied simultaneously to opposite sides of the tissue. Intracellular electrical activity was also studied in intact muscle, in isolated circular muscle, and in isolated longitudinal muscle. Slow were recorded extracellularly from circular and longitudinal muscle when the two layers formed the intact muscle wall; they were also recorded from isolated circular muscle, but not from isolated longitudinal muscle. Removal of the submucosa from the circular muscle surface abolished slow recorded from both isolated circular muscle and the intact specimen. Exposure of the mucosal surface of isolated circular muscle to a hypertonic solution of KCl abolished slow, while exposure of the serosal surface to the same concentrations had no effect. Slow are not generated in longitudinal muscle. Slow in circular muscle are dependent on the integrity of the junction between the submucosa and the innermost circular layer. Low frequency cyclic variations of intracellular potential (slow, basic electrical rhythm, electrical control activity, or pacesetter potentials) occur in the smooth muscle cells of the stomach, small intes- Received August, Accepted September 10, Address requests for reprints to: N. G. Durdle, Department of Electrical Engineering, University of Alberta, Edmonton, Alberta. Canada T6G 2G7. This research was supported in part by the Medical Research Council of Canada by the American Gastroenterological Association /83/ $03.00 tine, and colon of most animals, including humans (1). There is general agreement that the slow wave originates in the longitudinal layer and propagates into the circular muscle layer of the stomach and small intestine (2-). There is controversy, however, about the origin, frequency, and incidence of the slow in the colon. Sarna et al. (5) described multiple frequencies present in the human colon. Bowes et al. (6) and Chambers et al. (7) have, however, presented evidence indicating that the canine and human colons have a single omnipresent dominant frequency. There is general agreement that electrical coupling between regions in the colon is poorer than that in the stomach and the small intestine (5,6,8). The only evidence concerning the origin of the slow wave activity in the colon comes from a study on isolated cat colon by Christensen et al. (9) in which the circular muscle was found to be the site of origin. The objectives of the present study were to determine the site of origin of slow wave activity in the canine colon, to examine slow wave characteristics in the circular and longitudinal muscle layers, and to examine the roles played by each of these layers in the generation of this activity. Methods Sections of colon were surgically removed from healthy mongrel dogs anesthetized with sodium pentobarbitol. Immediately upon removal, the segments were washed with oxygenated Krebs-Ringer solution and cut open along the mesenteric border. These sections were trimmed to form either longitudinally or circularly oriented strips with dimensions 1.5 x 2.0 cm for extracellular experiments or 0.2 x 1.0 cm for intracellular experiments. Six types of tissue strips were prepared: Type 1: Intact muscle wall; mucosa and muscularis mucosa were removed using sharp dissection. Type 2: Intact muscle wall; mucosa and muscularis mucosa were removed using sharp dissection.

2 375 DURDLE ET AL. GASTROENTEROLOGY Vol. 8, No.2 In addition, the submucosal tissue remaining was removed by blunt dissection. Type 3: Isolated longitudinal muscle; muscle layers separated by blunt dissection. Type :, submucosa attached. Type 5:, submucosa removed. Type 6: Isolated submucosal tissue. Histological sections were taken from each experiment to confirm the accuracy of dissection. Extracellular Electrode Measurements Two different types of preparations were used: (a) "single-sided" preparations in which pressure electrodes were placed at multiple sites on one side of the tissue specimen and (b) "double-sided" preparations in which two pressure electrodes were applied simultaneously to both sides of the tissue specimen. Single-sided experiments. These were conducted in a horizontal tissue chamber, filled and perfused with oxygenated Krebs-Ringer solution (poz == 220 mmhg) and maintained at 37.5 C. Applied to the upper surface of the tissue with an average force of 6 g were 1 to low noise, low drift silver-silver chloride electrodes (10). The reference electrode was a large silver-silver chloride electrode placed in the bath fluid (Figure 1). Twenty-six preparations were used to compare slow wave characteristics of the various tissue preparations. These experiments are tabulated in Table 1. Double-sided experiments. These experiments were conducted using the equipment shown in Figure 2. The tissue specimen was mounted vertically between two electrically isolated chambers, each containing 20 ml of oxygenated Krebs-Ringer solution. Two extracellular pres- Table 1. Single-Sided Extracellular Experiments Prep. Recording type Preparation site n Result 1 Intact muscle Mucosal 25 Slow Intact muscle Serosal 25 Slow 2 Intact muscle with sub- Mucosal 12 No slow mucosa removed Intact muscle with sub- Serosal 12 No slow mucosa removed 3 Isolated longitudinal Inner 10 No slow muscle Isolated longitudinal Serosal 10 No slow muscle Submucosal 10 Slow Outer 10 Slow 5 Inner No slow with submucosa re- moved Outer No slow with submucosa re- moved 5 Isolated submucosa 2 No slow sure electrodes were applied to opposite sides of the tissue specimen, and a silver-silver chloride reference electrode was placed in each bath. Double-sided recordings were obtained from segments of intact muscle wall and from isolated circular and longitudinal layers in a total of 12 experiments (Table 2). To investigate the effect of depolarization of intestinal muscle caused by excess potassium chloride (11), the double-sided recordings were repeated with an increased KCI concentration on each side of the tissue specimen. In other experiments, 10-5 M tetrodotoxin (TTX) was applied to each side of the specimen. In a final series of experiments, recordings were made of the potential difference between the two reference electrodes, in the absence of pressure electrodes, when the two chambers were separated by type 1 and type tissue preparations. Data from both single- and double-sided experiments were recorded on an fm magnetic tape recorder, digitized, and processed on a Hewlett-Packard 21 MX/E computer (Hewlett-Packard Co., Palo Alto, Calif.) using fast Fourier transform, zero crossing, and cross-correlation programs. ELECTRODES ----:;.,,-Lr:.,,-L--o~III!!1!-'., ~~~~~~~~~R ---,.'~~"'-.. SOLUTION!=r=::::;?"L,L-- REfERENCE ELECTRODE ~,...,,<--+-- COLON MUSCLE STRIP Figure 1. Single-sided experimental system. The specimen is perfused with oxygenated Krebs-Ringer solution maintained at 37.5 C. Four extracellular pressure electrodes are applied to the tissue by means of a combined electrode holder and force transducer. Intracellular Electrode Measurements Intracellular electrical activity was studied using a microelectrode manipulator and holder, a hydraulic manipulation system, a battery-powered electrometer-input amplifier, and the constant temperature tissue bath originally used in the single-sided extracellular studies. The hydraulic system permitted precise control of electrode movement while mechanically isolating the operator from his vibration sensitive equipment. The amplified intracellular potential was simultaneously recorded on a storage

3 February 1983 ORIGIN OF CANINE COLONIC SLOW WAVES 377 Electrodes To Recordino Channel Electrode Holder on FDrce Transducer To Recording Channel 1 Reference Electrode y~-+} To Recording Channel 2 Reference Electrode Chamber No.1 Constant Figure 2. Cross section of double-sided experimental system. The specimen is mounted vertically between two electrically isolated chambers, each with oxygenated Krebs-Ringer solution. Two pressure electrodes are applied to opposite sides of the specimen. Table 2. Double-Sided Extracellular Experiments Prep. type Preparation Recording site n Result Intact muscle Intact muscle Intact muscle Isolated longitudinal muscle Isolated longitudinal muscle Intact muscle t [KCI] on serosal side t [KCI] on mucosal side t [KCI] on longitudinal side t [KCI] on mucosal side 10-6 M TTX 10-6 M TTX Ref. electrode Ref. electrode (no pressure electrode) Ref. electrode Ref. electrode (no pressure electrode) Slow on both sides Larger amplitude on mucosal side Slow of normal amplitude No slow Slow on both sides Slow of normal amplitude No slow No effect on slow No slow Sometimes burst activity No effect on burst activity Cyclic signal Same frequency as slow wave No slow

4 378 DURDLE ET AL. GASTROENTEROLOGY Vo!' 8, No.2 Table 3. Intracellular Experiments Prep. type Preparation Recording site n Result 1 Intact muscle Throughout cir- 37 Slow cular muscle Intact muscle Throughout lon- 8 No slow gitudinal mus- c1e 3 Isolated longi- Throughout layer 5 No slow tudinal mus- c1e Isolated circular Throughout layer 5 Slow muscle with submucosa attached oscilloscope, an fm tape recorder, and a polygraph recorder. Tissue specimens were mounted with the cross-section uppermost. In this way, cells of the circular and longitudinal muscle layers were penetrated from the side of the cut specimen. Intracellular experiments were conducted on 37 specimens from canine proximal and midcolon. Measurements were obtained from cells across the different muscle layers in the intact muscle wall as well as across isolated circular and longitudinal muscle layers. Results The results of single-sided extracellular, double-sided extracellular, and intracellular electrode studies are summarized in Tables 1, 2, and 3, respectively. Regular, continuous slow were seen in all single-sided and double-sided recordings from both the mucosal and serosal sides of the intact muscle wall and the isolated circular muscle layer in which the submucosal layer was intact. Single-Sided Experiments A typical record of slow wave activity recorded from four extracellular electrodes placed on the mucosal side of the intact muscle wall is shown in Figure 3. The mean frequency of slow in the proximal canine colon was 5.9 ± 1.1 cpm. At no time were slow recorded extracellularly from isolated longitudinal muscle. A typical result from the serosal side of isolated longitudinal muscle is shown in Figure. Similar results were obtained from the circular side of longitudinal muscle. Removal of the submucosal layer from the intact muscle wall and from isolated circular muscle resulted in complete abolition of slow. Figure 5 shows recordings from four electrodes on the mucosal side of intact muscle wall after removal of the submucosa. No slow were recorded extracellularly from isolated submucosal tissue. Double-Sided Experiments Figure 6 shows a typical record obtained from a double-sided experiment on an intact muscle specimen. The slow recorded from the serosal side were always of smaller amplitude than those from the mucosal side. No phase difference could be observed between the slow obtained from the serosal and mucosal sides of the specimen, nor from the two sides of isolated circular muscle. In the latter preparation, the signals recorded from the two sides were always of approximately equal amplitude. No slow wave activity was recorded on either side of isolated longitudinal muscle, but a higher frequency burst activity was sometimes seen (Figure 7). This signal has a mean burst frequency of 23.5 cpm and an average of cycles/burst. The bursts occurred an average of 1.82 times/min. This activity was not present in all specimens, and was unaffected by 10-6 M tetrodotoxin The concentration of KCI on each side of intact muscle wall and isolated circular muscle specimens (submucosa attached) was increased to 30 mm. Within 15 s of increasing the concentration of KCI on the mucosal side of both tissue preparations, slow wave activity on both sides of the specimen was abolished. This is shown for an intact muscle specimen in Figure 8. The effect could be completely reversed by washing with normal Krebs-Ringer solution. An increased concentration of KCl applied to the longitudinal side of the intact muscle wall or to the isolated circular muscle produced no effect. The potential of the large silver-silver chloride reference electrode on the mucosal side of an intact muscle segment when measured with respect to a No.1 No. Figure 3. Slow wave recordings obtained from four pressure electrodes applied to the mucosal side of an intact muscle specimen. The electrodes are oriented in the circular direction.

5 February 1983 ORIGIN OF CANINE COLONIC SLOW WAVES 379 r60secj ttbl"".i!!i!!!ui"'hiiiiimiii.iiiiii".!!!willmiii.iiiimitiliiiiimiiiii.'liil1t!'"tll!l!i!i!iiiii'!i""iiii"""'"ll11tut""""i1i11"'''iii!ltll llfiilt ll lll,,'m'''''it'wh'''''''!l1ii!!""'ut!!t!!.ttmiim""" fie j No 1 No 2 I 06mV --.l Figure. Recordings obtained from two pressure electrodes applied to the serosal side of isolated longitudinal muscle in a single-sided experiment. No regular activity was observed. similar reference electrode on the serosal side (in the absence of pressure electrodes) varied cyclically at the same frequency as the slow wave. No such cyclic variation was obtained when isolated circular muscle separated the two reference electrodes. Intracellular Experiments Intracellular measurements were obtained from cells on both sides of intact muscle wall and both sides of isolated circular and longitudinal muscle. Two hundred thirty-three successful penetrations were made into circular muscle cells of intact muscle preparations from 37 specimens. A typical recording from such a cell is shown in Figure 9A. The mean resting potential and slow wave amplitude were ± 6.8 mv and 28.9 ± 5.5 my, respectively. Twenty-eight penetrations were made into the longitudinal muscle cells of eight intact muscle preparations. Figure 9B shows such a recording with no slow wave activity. The mean resting 160seel potential for longitudinal muscle cells was ± 7.0 my. Intracellular recordings confirmed that removal of the submucosal layer completely abolished slow wave activity in the circular muscle layer. Attempts to measure intracellular activity in the isolated submucosal tissue were unsuccessful. Histological specimens obtained from the tissues of isolated circular and longitudinal muscle layers confirmed that the tissue dissection occurred in the plane between the circular and longitudinal muscle layers. The isolated submucosal layer was mainly connective tissue. A thin layer of smooth muscle cells was seen in this layer, but further investigation is needed to show if this was in fact part of the circular muscle layer. Discussion Our results show that slow are not generated in longitudinal muscle; slow were never recorded either extracellularly or intracellular- I"1WtI1I!!!!!!I!ltt!!l!!!!Uttttl1!!tttllb1!HIUMt!!!!!I!!!tttU!n!!tt!!!!!!I!t!J!!1!U!UM'!Uttt!Jt!!1It!!rtttUH!!tn!!!!!t!U!!!!!t!!!!t!!!1!lIJt!!lltUUIIIU!!!1!!1In!!JH!tt!!tt!!U.,!t!'!!I!IItI!t!IIHHU!I!ItIUUUltt!!!Ulltt!'nlUH!I!!!!!!!I!t!!,Mi Figure 5. Single-sided pressure electrode recordings obtained from the mucosal side of an intact muscle specimen after removal of the submucosa. The regular slow wave activity, seen before removal of the submucosa, was completely abolished. No. J No. 2 ~ OAmV ---.l No.

6 380 DURDLE ET AL. GASTROENTEROLOGY Vol. 8, No. 2 j10seci,mlmmttmtmtitll!t!llil!!!!ii!!!!i!i!!itnll!!i!uu!l!!i!!!!!i!""tlutlt """'II1!I!I!IIII!!ltll!!""ItI""!!I!I!""tt"""M!1!9'1I1III!!M11!"!H III!I!""!!'.. rr' *!!!S'IMU"'HHH'MIIMWlttWttM"b 7 #5 No r 2.0mV ----.l. No.2 Figure 6. Double-sided recording from an intact muscle wall specimen. Channell was recorded from the serosal side; channel 2 from the mucosal side. ly from isolated longitudinal muscle cells. It could be argued that the absence of slow in this tissue was caused by the trauma of dissection. This is very unlikely, however, since slow were recorded on both sides of isolated circular muscle obtained by the same dissection techniques. In the intact muscle wall, no phase lag was recorded between the longitudinal muscle and circular muscle slow wave signals. This indicates either that there is very tight coupling between the two layers, or that one of the layers is passive. Since slow were never measured either extracellularly or intracellularly from either side of isolated longitudinal muscle, it must be concluded that this is the passive layer (12). Slow wave amplitudes recorded on the serosal side of longitudinal muscle were always smaller than those on the circular muscle side of the intact muscle wall, in the double-sided preparation. Upon removal of the longitudinal layer, slow of approximately equal amplitude were recorded from both sides of the remaining circular layer. This indicates that the slow recorded on the longitudinal side are conducted passively from the circular layer. The longitudinal layer attenuates the slow wave signal, causing it to have a lower amplitude at the serosa. Further evidence to support the hypothesis that slow are not generated in longitudinal muscle is found in the cyclic variations recorded from one reference electrode with respect to the other reference electrode across the intact muscle wall. These variations were measured in the double-sided preparation, across intact muscle tissue, in the absence of pressure electrodes. The same measurement made across a specimen of isolated circular muscle produced no cyclic variation. We concluded that the cyclic variations were potentials developed across the longitudinal layer as a result of currents flowing from the circular layer to the serosal side of the specimen. The second finding of this study is that the appearance of slow in the circular muscle is dependent on the integrity of the junction between the submucosa and the innermost circular layer. In the presence of the submucosal layer, slow wave activity was recorded both extracellularly and intracellularly on both sides of intact and isolated circular muscle. Removal of the submucosal layer resulted in complete abolition of slow wave activity in both isolated circular muscle and in the intact muscle wall. This was confirmed using both extracellular and intracellular measurements. The effects of an increased concentration of KCI on No.1 No.2 Figuff 7. Double-sided recording of high frequency burst activity from extracellular electrodes on both sides of isolated longitudinal muscle. Channell was recorded from the serosal side; channel 2 from the mucosal side.

7 February 1983 ORIGIN OF CANINE COLONIC SLOW WAVES 381 r sec---1,w,!lii'!mmitiimuiii.iuii!llilib"ii.!liihu!!!i11i1ii11himii!!!!hiiiii!!t!!!!!iu,i"w"mllti!!i!!ii11t1'",!!!!!iiiiiitill"!liii!!!lii.!llthmi!!liiimiiiihii'!'mli!itmmitt!mtmmlh!mihmmii!lii!!'!m'hmitmm,,_,... ti No I No. 2 I I,J"-"\ /'~ _,"-, --I-./'-, r-\ ""'--.J- ~'-../"''\.\_/\...f''-v'_r-:'',,--.aj''\.../~,.r~\./-''v''j- ~'/V\_ 1 mv.,~ ~ ~ v v - v' --.-l \ \ L KC1 concentration changed to 30 mm Figure 8. Effect of increasing the concentration of KCl to 30 mm on the mucosal side of a double-sided recording. Within 15 s, the regular normal slow wave activity was abolished. the double-sided preparation showed that the submucosal layer, or the inner circular muscle cells, playa major role in slow wave generation. Kuriyama (11) showed that hypertonic solutions of KCI cause a depolarization of smooth muscle and a reduction of slow wave amplitude. An increased concentration, however, of KCl applied to the outer side of both the intact muscle wall and the isolated circular muscle had no effect on the slow wave. This indicates that the longitudinal layer and outer circular muscle cells are not the source of slow wave activity. In contrast to this, an increased concentration of KCI applied to the mucosal side of both the intact wall and isolated circular layer leaves no doubt that the submucosal layer, the inner circular layer, or the integrity of the junction between the two is responsible for the generation of this activity. Two conclusions can be drawn from this study: (0) 1. Slow are not generated in longitudinal muscle. 2. The appearance of slow in circular muscle is dependent on the integrity of the o Volts -100 mv - r- 20 sec----, III ti'lll "'11111t t tlllllllllll'ii'1i t "'1111 lit 130 sec, 11111!11t1tlltlt!lI1!!I!t1tIllI!1!11!ttlttt!!ltI!1!I!ttlllt7pBl\!jjj!It!t!!!!!!I!!! o Volts -I (b) V;~ mv Figure 9. a. Intracellular recording from a circular muscle cell of the intact muscle wall. b. Intracellular recording from a longitudinal muscle cell of the intact muscle wall. junction between the submucosa and the innermost circular layer. Two questions arise as a result of this work. The first relates to the origin of the slow wave activity. What component of the submucosal tissue or inner circular layer is responsible for slow wave generation? Is it the thin layer of smooth muscle cells seen in the submucosal tissue layer, or is this layer actually debris from the circular muscle layer? Additional study using light and electron microscopy is required to examine the morphology of this layer. The second question relates to the control of longitudinal contractions. There is general agreement (13,1) that slow control contractile activity in the stomach and small intestine. Does the absence of slow wave activity in longitudinal muscle indicate that slow play no role in the control of longitudinal contractions? Slow wave contractions are known to occur primarily during the plateau phase of the slow wave (13,1). During this phase of each slow wave, circular muscle cells are close to the intracellular threshold potential required for the cells to contract. We have found that the potential of the plateau phase of the slow wave is -9 my. The mean resting potential of longitudinal muscle cells (-52.0 my), however, is very close to the plateau phase in circular muscle cells. If the threshold potential required for contraction is the same in both layers, longitudinal cells are always ready to contract. Any stimulus, such as a contraction in the circular muscle layer, could very easily result in contraction of longitudinal muscle layer. In addition to this, the higher frequency burst activity observed in isolated longitudinal muscle will very likely produce associated contractions. In fact, the repetition frequency of this burst activity is close to the 0.5/min frequency of peristaltic contractions observed by Kocylowski et al. (15) in an ex vivo canine

8 382 DURDLE ET AL. GASTROENTEROLOGY Vol. 8, No.2 colon preparation. Thus, the burst activity, and not the slow wave, may be the major factor involved in the control of longitudinal contractions. References 1. Alvarez WC, Mahoney LJ. Action currents in stomach and intestine. Am J PhysioI1921;58: Papasova MP, Nagai T, Prosser CL. Two component slow in smooth muscle of cat stomach. Am J Physiol 1968;21(): EI-Sharkawy TY, Morgan KG, Szurszewski JH. Intracellular electrical activity of canine and human gastric smooth muscle. J Physiol (London) 1978;279: Bortoff A. Electrical transmission of slow from longitudinal to circular intestinal muscle. Am J Physiol 1980; 209(b): Sarna SK, Bardakjian BL, Waterfall WE, Lind JF, Daniel EE. The organization of human colonic electrical control activity. In: Gastrointestinal motility. Christensen J, ed. New York: Raven Press; 1980: Bowes KL, Shearin NL, Kingma YJ, Koles ZJ. Frequency analysis of electrical activity in dog colon. In: Gastrointestinal motility in health and disease. Duthie HL, ed. London: MTP, 1978: Chambers MM, Bowes KL, Kingma YJ, Bannister C, Cote KR. In vitro electrical activity in human colon. Gastroenterology 1981;81: Christensen J. Myoelectric control of the colon. Gastroenterology 1975;68: Christensen J, Caprilli R, Lund G. Electric slow in circular muscle of cat colon. Am J Physiol 1969;217(3): Kingma YJ, Lenhard JM, Durdle NG, Bowes KL. A small silver-silver chloride electrode for measurement of low frequency biological signals. In: Proceedings of the 9th Conference of the Canadian Medical and Biological Engineering Society, Hamilton, Ontario, August Kuriyama H. The influence of sodium, potassium and chloride on the membrane potential of the smooth muscle of taenia coli. J Physiol (London) 1963;166: Bortoff A, Michaels D, Mistretta P. Dominance of longitudinal muscle in propagation of intestinal slow. Am J Physiol (Cell Physiol 9) 1981;20:C Sarna SK, Daniel EE, Kingma YJ, Simulation of slow wave electrical activity of small intestine. Am J Physiol 1971; 221(1): Sarna SK, Daniel EE, Kingma YJ. Simulation of electric control activity of the stomach by an array of oscillators. Dig Dis Sci 1972;17(): Kocylowski M, Bowes KL, Kingma YJ. Electrical and mechanical activity in ex vivo perfused total canine colon. Gastroenterology 1979;77: