Comparative Anatomy of the Myenteric Plexus of the Distal Colon in Eight Mammals

Transcription

1 GASTROENTEROLOGY 1984;86: Comparative Anatomy of the Myenteric Plexus of the Distal Colon in Eight Mammals JAMES CHRISTENSEN, MICHAEL J. STILES, GARY A. RICK, and JEAN SUTHERLAND Research Laboratories, Division of Gastroenterology-Hepatology, Department of Internal Medicine, University of Iowa College of Medicine, Iowa City, Iowa The behavior of the most distal part of the colon in a variety of species suggests that the innervation of this part may differ from that of more proximal parts. Silver impregnation was used to demonstrate the arrangement of the myenteric plexus of the distal colon in eight species (rat, guinea pig, rabbit, Australian possum, American opossum, cat, dog, and monkey). A distal zone, -5%-20% of the total length of the colon above the anal verge in the nonrodents, was characterized by a plexus of very irregularly disposed intersecting nerve bundles of highly variable size with few and small ganglia; this zone was absent in the three rodent species. A next most distal zone, -10%-65% of the total colonic length, contained a stellate plexus of large, regularly disposed ganglia interconnected by small nerve fiber bundles upon which were superimposed large dark-staining nerve bundles; these bundles began to be seen at the level of the irregular rectal plexus and ran cephalad, bypassing some ganglia but giving off branches to others. These, called shunt fascicles, contained many myelinated nerve fibers. Above this zone, the plexus was a stellate plexus throughout the remainder of the colon. The motor function of the distal part of the colon differs from that of the rest of the organ. Elliott and Barclay-Smith (1) noted in several carnivores and rodents that the dominant pattern of contraction of the distal colon was a powerful peristaltic wave that could be excited by electrical stimulation of the sacral visceral nerves. This effect did not seem to extend to proximal levels of the colon. They did not specify actual distances. Common experience also suggests that the distal colon functions differently from more proximal regions: normally, in defecation, it appears to be only the rectum and, at most, a variable length of the sigmoid and descending colon that empty. The anatomy of the nerves in this segment seems to have been generally assumed to be the same as it is in the rest of the colon. The anatomy of the myenteric plexus of the distal colon and rectum has been examined mainly in rodents (2,3). We investigated the idea that the myenteric plexus may display some unique anatomy to account for the special function of the distal part of the colon, the part that can be excited to contract by the sacral visceral nerves. Received August 18, Accepted November 18, Address requests for reprints to: James Christensen, M.D., Division of Gastroenterology-Hepatology, Department of Internal Medicine, Room C31-P, General Hospital, University of Iowa, Iowa City, Iowa This work was supported in part by Research Grant AM from the National Institutes of Health, and in part by Grant HL from the National Institutes of Health to the University of Iowa College of Medicine in support of summer fellowships for medical students. The authors acknowledge the kindness of Dr. John Dent and Dr. John B. Furness of the Flinders University Medical Center, Adelaide, South Australia, where part of the work was done by the American Gastroenterological Association /84/$3.00 Methods The gross arrangement of the myenteric plexus of the distal colon was studied in several different classes of animals: carnivores were represented by the cat and dog; rodents were represented by the rabbit, rat, and guinea pig; marsupials were represented by the American opossum and the Australian brush-tailed possum; primates were represented by the cynomolgus monkey. In all cases, only healthy, mature animals were used, having body weights characteristic of adults of each species. The animals were killed by intraperitoneal injection of sodium pentobarbital (or by a blow to the head in the case of the rodents). The distal half of the colon (or, in some cases, the whole colon) was immediately removed, emp-

2 April 1984 MYENTERIC PLEXUS OF DISTAL COLON 707 0, - 0> 0> C 0,,- ~E ~ E. o~.0 CO ~~ ~ Species '".,,-0 '0 0 {l ". ()::; «CL () a: a: COlonic Length -36cm " 17em 30cm SOem 35cm 145cm 40cm 25cm 100 0:.'" ",...J ~ ' 2.0.l8!!ti 0-.cO <"".0 ~c.~ ~,,~ Figure 1. Key: ~:; Rectal Plexus [ill = Stellate Plexus with 0 = Stellate Plexus Shunt,Fasc icles The averaged proportionate lengths of the colon occupied by the three different patterns of the plexus. Colonic length refers to the average distance from the end of the rectum to the ileocecal junction in situ. The three different patterns are described in the text. tied, and washed in Krebs' solution. Particular care was taken to remove the colon all the way to the rectal mucocutaneous junction. The specimen, kept in Krebs' solution aerated with 95% O 2-5% CO 2 at 37 C, was dis- sected free of surrounding tissues, opened lengthwise, and pinned flat to a sheet of balsa wood so as to stretch it in both dimensions to a point just short of tearing of the tissue. The dimensions of the organ were increased to about 150% of control by this degree of stretch. The tissues were fixed and stained by the silver-impregnation technique of Richardson (4). After at least 1 wk of immersion in methanol-free formalin, the tissues were cut into segments to fit microscope slides, and the circular muscle layer was removed to expose the myenteric plexus attached to the longitudinal muscle layer. The segments were then mounted, dehydrated, cleared, and exposed to a solution of silver and copper nitrates at 37 C for min. After reduction in 2,4-dinitrophenylhydrazine, they were put in a solution of ammoniacal silver nitrate containing an excess of ammonia and a small quantity of formalin. These quantities were varied slightly to slow the rate of color development so that the depth of staining could be controlled. Color development was arrested by immersion of the tissues in a solutiqn of sodium sulfite. The tissues were washed, dehydrated, cleared, and mounted. Considerable damage to the plexus often occurred in the removal of the circular muscle layer. With practice, the degree of such damage could be reduced. From each species, at least five complete sets of slides were obtained Figure 2. The rectal myenteric plexus of the Australian possum. The end of the rectum is to the left. The pattern seen here is the same a s that found in the distal rectum in all nonrodents. The calibration bar represents 0.5 mm.

3 708 CHRISTENSEN ET AL. GASTROENTEROLOGY Vol. 86, 1\'0. 4 that were judged to be as free of damage as possible. Each set represented contiguous segments of the colon from the anal verge to the proximal level of amputation. In some cases, samples were also taken from the proximal colon. Ganglion populations were determined at intervals above the distal end of the internal anal sphincter. For counting, a square was marked on each slide at a point where ganglia were completely preserved. The square was 1 cm z in the large animals (cat, dog, monkey, American opossum, and Australian possum) and 0.5 cm z in the small animals (rat, guinea pig, and rabbit). Ganglion population densities were averaged among the five sets of slides from each species. Results Nonrodent Species In the nonrodent species, the distal colon could be divided into three parts on the basis of the gross features of the myenteric plexus (Figure 1). In the three rodent species, the most distal of these parts, called part 1, was absent, as explained below. In part 1, in the most distal part of the rectum, the plexus was a network of nerve bundles intersecting at a variety of angles and intervals, so that the pattern had no regularity (Figure 2). No system of a major plexus with superimposed secondary and tertiary plexuses (as is commonly seen in the small intestine) could be discerned. The nerve fiber bundle intersections were usually devoid of ganglia. Nerve cell bodies lay mostly in very small ganglia, some ganglia lying near nerve bundle intersections, others being inserted in fiber bundles remote from intersections, and still others being pedunculated to form what have previously been called parafascicular ganglia in the plexus of the esophagus (5,6). The fiber bundles sometimes formed knots of tangled bundles that have been previously called labyrinthine nodes (6). These were of no consistent pattern. In the cat and opossum, these knots tended to be large and complex, while in the other species they were small and simple. This exceedingly irregular plexus involved a space of the terminal rectum that varied among species (Figure 1). Ganglia were seen at the level of the internal anal sphincter. The sphincter, of course, could not be identified physiologically in these studies, but it was assumed to be the thickening of circular muscle at the distal end of the rectum. Figure 3. The myenteric plexus from 6 cm above the end of the rectum in the Australian possum. Several large shunt fascicles, as described in the text, lie upon a plexus of stellate ganglia. The calibration bar represents 0.5 mm.

4 April 1984 MYENTERIC PLEXUS OF DISTAL COLON 709 In part 2, the next most proximal part of the distal colon, the plexus was characterized by two features. The first feature was the presence of a stellate plexus in which large ganglia, fairly regularly distributed, were connected to one another by nerve fiber bundles-the classical pattern of the plexus. Upon this stellate plexus there was superimposed an array of very thick and very densely staining fiber bundles that ran in the long axis of the colon (Figure 3). These have been previously called shunt fascicles (6). These shunt fascicles were first apparent distally in the irregular plexus of the first most distal segment, part 1. They were uniformly arranged in the circumference and extended cephalad, with some branching (Figure 4). They tapered proximally as they branched so that they merged gradually into the stellate plexus. The shunt fascicles were loosely attached to the longitudinal muscle layer, so that they were easily removed with the circular muscle layer. In the cat and dog rectum, where the shunt fascicles are very thick, they often lay in a groove indenting the longitudinal muscle layer (Figure 5). The shunt fascicles bypassed many ganglia but gave off branches to others along their course. Rarely, a shunt fascicle was interrupted by passing through a ganglion, but when one did, it passed alongside the ganglion, its fibers clearly bypassing the mass of the ganglion. Shunt fascicles were usually compact, but occasionally they broadened, sometimes splitting into two to five parallel fiber bundles that later converged again to form a compact bundle (Figure 6). In such regions of broadening or splitting, the number of branches to adjacent ganglia was greater than it was in the regions where the fascicle was compact. Broadening and splitting of the shunt fascicles were generally more frequent at the more proximal levels of this part of the colon, and were more prominent in the dog than in the other nonrodent species. These shunt fascicles contained myelinated nerve fibers (Figure 7).. In part 3, the third region of the distal colon, the plexus was arranged in the stellate pattern alone, Figure 4. The shunt fascicles of the rectum of the cat. The distal rectum li es at the top of the photograph. Nearly the whole breadth of the proximal rectum is represented. In this lightly stained preparation. the ganglia cannot be seen. The calibration bar represents 1.0 mm.

5 710 CHRISTENSEN ET AL. GASTROENTEROLOGY Vo!' 86, No. 4 Figure 5. A single shunt fascicle in an unstained preparation from the distal cat colon. Some remaining circular muscle bundles can be seen at the left, overlying the shunt fascicle. To the right, it has been lifted from the groove (arrow) in the longitudinal muscle layer in which it rests. The calibration bar represents 0.5 mm. with large ganglia, regularly distributed and connected together by small nerve fiber bundles. The pattern continued throughout the rest of the colon. less in the area of the rectal plexus than it is in the areas containing the stellate plexus with shunt fascicles and in the area of the stellate plexus alone. Rodent Species In the rat, guinea pig, and rabbit, the stellate plexus extended all the way to the internal anal sphincter. The sparse and irregular plexus of the distal rectum described above in the nonrodent species was not found in any of these species. A few shunt fascicles were seen in all three species but they were very thin. The zone containing such thin shunt fascicles extended some distance. The lengths of the colon over which this pattern extended are shown in Figure 1. The mean ganglionic population density in each of these three regions in each species is shown in Table 1. Ganglion cells themselves could not be counted because the supporting cells, which are stained by the silver, obscured the outlines of the nerve cell bodies. The data suggest that ganglionic density is Table 1. Ganglion Density in the Three Parts of the Distal Colon and Rectum in Eight Species Part 2 (Stellate plexus Part 1 with Part 3 (Rectal shunt (Stellate plexus) fascicles) plexus) Cynomolgus monkey 55 :±: :±: :±: 95 Australian possum 47 :±: :±: :±: 33 American opossum 30 :±: 5 45 :±: 5 58 :±: 7 Dog 37 :±: 6 40 :±: 5 72 :±: 9 Cat 17 :±: 4 42 :±: 2 39 :±: 2 Rabbit 334 :±: :±: 27 Guinea pig 563 :±: :±: 35 Rat 794 :±: :±: 74 Ganglion density is measured as gangjia/cm2. All values given are mean:±: SE; n = 5.

6 April 1984 MYENTERIC PLEXUS OF DISTAL COLON 711 Figure 6. Shunt fascicles from the distal colon of the dog. A compact fascicle is seen in the center of the photograph. Broadened or split shunt fascicles appear both to the left and right. The calibration bar represents 0.5 mm. Discussion This study shows a distinction between rodents and omnivores-carnivores in the anatomy of the myenteric plexus of the most distal part of the colon. Rodents have mainly been examined in this area (2,3), and the shunt fascicles are sparse and easily torn away with the circular muscle. Stohr (7) observed in his long review of over 30 yr ago that the myenteric plexus of the distal colon has had little study. This deficit has been rectified only partly since then. The most notable contribution is that of Stach (8) who examined the innervation of the colon in dog, cat, rabbit, and humans. His paper, which has not received much attention from the English-speaking world, clearly describes the structures that are here called shunt fascicles. He called them the "ascending nerves of the pelvic plexus." Using nerve degeneration techniques, he established that these structures convey fibers from the dorsocranial part of the pelvic plexus, and he was able to trace such fibers all the way to the level of the ileocecal junction, though the shunt fascicles themselves could not be traced grossly above the transverse colon. Stach did not describe the features of the distal rectal myenteric plexus that we observed; however, he did show that the shunt fascicles contain both myelinated and unmyelinated fibers, and proposed that the former are parasympathetic afferents and sympathetic efferents, and that the latter are fibers arising from intrinsic nerves of the plexus itself. Stach cites an earlier description by Iljina and Lawrentjew (9) of the myenteric plexus of the distal dog colon, but that description is very cursory. Lee (10), using planar sections of the colon wall in dogs, sought to determine the relative densities of myelinated fibers at various levels of ascending, transverse, and descending colon. He found myelinated fibers to be most abundant in the myenteric plexus, much less abundant in the submucous plexus, and very sparse in the mucosa. In the myenteric plexus, the proportion of myelinated fibers increased along the colon, the relative ratios being 1: 3: 9 in the ascending, transverse, and descending

7 712 CHR1STENSEN ET AL. GASTROENTEROLOGY Vol. 86, No.4 Figure 7. A 25-lLm planar section through a shunt fascicle of the distal colon of the cat, stained with Sudan B. Myelin sheaths can be seen as dark structures with interruptions at intervals, the nodes of Ranvier. The calibration bar represents 10 /-Lm. colon, respectively. Lee did not observe the shunt fascicles in his preparations. Holmes (11), despite the title of his paper, did not, in fact, examine the rectum but only the anal canal and most distal rectum. Although he studied monkeys, cats, rabbits, guinea pigs, and rats, he commented only on the structure of the plexus in the most distal rectum in rat and guinea pig, noting the hexagonal or rectangular pattern of the nerve bundles. He may have missed the unusual arrangement of the rectal plexus in the nonrodents because his attention was directed to the anal canal. Holmes (11) did note that there is a profuse nerve plexus in the areolar tissue around the lower rectum, containing no ganglia, but he did not describe it further. In the absence of a complete physiological description of the behavior of the distal colon, it is difficult to state the significance of this anatomy. The anatomy of the plexus of the distal colon of omnivores-carnivores could be related to the fact that such species store the fecal volume and deliver a large mass at infrequent intervals. In contrast, rodents defecate small masses at frequent intervals, seeming not to store feces for very long. Even such a simple concept, however, rests upon common observation rather than upon objectively gathered evidence. The temporal patterns of defecation that characterize different classes or species of animals have apparently not been described to any extent. In the Australian possum, the American opossum, dogs, and cats when killed, the distal colon often contains a large soft fecal mass. In contrast, the rodent colon usually contains hard fecal pellets spaced quite uniformly all the way to the anal verge. The shunt fascicles may be specialized pathways for communication between the distal rectum and the more proximal parts of the colon to which they extend. This communication could occur in both directions. Distention of a balloon in the distal colon in humans excites relaxation of the internal anal sphincter (12), and the shunt fascicles could well be the pathways involved. In defecation, a coordinated peristaltic contraction involves a long segment of the

8 April 1984 MYENTERIC PLEXUS OF DISTAL COLON 713 distal colon. The shunt fascicles could be involved in this coordination. The myelinated nerve fibers in the shunt fascicles are of particular interest, for myelinated fibers are infrequent in the myenteric plexus. They could be preganglionic fibers, and the shunt fascicles seem to be intraplexal extensions of the sacral visceral nerves, as was indicated by the nerve degeneration studies done by Stach (8). Thus, these shunt fascicles may be specialized pathways by which centers in the central nervous system can selectively influence the function of different parts of the colon. The work reported here was half-completed before we were aware of the German literature describing the shunt fascicles. Thus, this paper represents a rediscovery of that neglected anatomy. The irregularity of the rectal plexus in nonrodent species seems not to have been previously noted. References 1. Elliott TR, Barclay-Smith E. Antiperistalsis and other muscular activities of the colon. J Physiol (Land) 1904;31: Irwin DA. The anatomy of Auerbach's plexus. Am J Anat 1931;49: Matsuo H. A contribution on the anatomy of Auerbach's plexus. Jap J Med Sci Anat 1934;4: Richardson KC. Studies on the structure of autonomic nerves in the small intestine, correlating the silver-impregnated image in light microscopy with the permanganate-fixed ultrastructure in electron-microscopy. J Anat 1960;94: Christensen J, Robison BA. Anatomy of the myenteric plexus of the opossum esophagus. Gastroenterology 1982;83: Christensen J, Rick GA, Robison BA, Stiles MJ, Wix MA. Arrangement of the myenteric plexus throughout the gastrointestinal tract of the opossum. Gastroenterology 1983; 85: Stohr P. Zussammen Fassende Ergebnisse tiber die mikroscopische Innervation des Magen-Darmkanals. Ergeb Anat Entwicklungsgesch 1952;34: Stach W. Uber die in der Dickdarmwand azendieren Nerven des Plexus pelvinus und die Grenze der Vagalen und sakralparasympathetischen Innervation. Z Mikrosk Anat Forsch 1971;84: iljina WI. Lawrentjew BJ. Zur Lehre von der Cytoarchitektonik des peripherischen autonomen Nervensystems. III. Ganglien des Rektums. Z Mikrosk Anat Forsch 1932;30: Lee 1M. The distribution of the myelinated nerves in the colon of the dog. Arch Jap Chir 1956;25: Holmes AM. Observations on the intrinsic innervation of the rectum and anal canal. J Anat 1961;95: Schuster MM, Hendrix TR. Mendeloff AI. The internal anal sphincter response: manometric studies on its normal physiology, neural pathways, and alterations in bowel disorders. r Clin Invest 1963;42: