- Email: [email protected]
Histochemical localization of nitric oxide-synthesizing neurons and vascular sites in the guinea-pig intestine
0306-4522/92 $5.00 + 0.00 Pergamon Press Ltd IBRO
Neuroscience Vol. 51, No. 4, pp. 791 799, 1992
Printed in Great Britain
H I S T O C H E M I C A L L O C A L I Z A T I O N OF NITRIC OXIDE-SYNTHESIZING NEURONS AND VASCULAR SITES IN THE G U I N E A - P I G INTESTINE K. NICHOLS,*A. KRANTIS*t and W. STAINES~ *Digestive Diseases Research Group, Department of Physiology and :~Department of Anatomy, University of Ottawa, Ottawa, Canada K1H 8M5 A~tract--Laminar preparations of fixed segments of the guinea-pig intestine were examined for nitric oxide synthase activity using reduced nicotinamide adenine dinucleotide phosphate and nitroblue tetrazolium salt as substrates. Under conditions specific for detecting nitric oxide synthase-related diaphorase activity, a subpopulation of neural elements in the myenteric plexus, deep muscular plexus and submucosa were intensely stained. Intensely stained nerve fibres were distributed throughout the meshworks of the myenteric plexus and its innervation of the circular muscle, and in the submucosa within Henle's plexus. Intensely stained nerve cells and their processes were evident in most myenteric ganglia but were rare in ganglia of Henle's plexus. Stained ganglion cells comprised types I, II and VI of the morphologically defined enteric nerve cells. Stained neural elements were increasingly prevalent within successively more caudal segments of the intestine. In addition to neuronal staining, arterioles of the submucosal vascular network displayed distinct, punctate patches of staining distributed over their surface. Perivascular nerve fibre staining was absent. These results show nitric oxide synthase activity to be present within neurons and fibres of the major enteric nerve layers and within submucosal blood vessels throughout the guinea-pig small and large intestine.
Nitric oxide (NO) is a free radical molecule produced from the amino acid L-arginine in neural and nonneural tissue by the action of a family of N O synthases. 32'34 N O is a neural messenger in the brain where its synthesis in some cells is stimulated by Ca 2+ entry subsequent to excitatory amino acid (N-methylD-aspartate) receptor stimulation. U p o n synthesis, N O diffuses to neighbouring cells where it interacts with the ferrous haeme of soluble guanylate cyclase and stimulates formation of cGMP. This process has been seen to carry signals between neurons, and between neurons and glial cells. 35 Proposed roles for N O in C N S function include synaptic plasticity and long-term potentiation, and the activity-dependent determinant of neural development. 22'23 In addition, it is proposed that vasodilator nerves release N O and therefore are involved in the neural control of cerebral blood flow. 45 N O is also produced by central and peripheral blood vessels 26'34'37'39'49 and is believed to be responsible for the activity of the endotheliumderived relaxant factor ( E D R F ) . 28'34 There is now pharmacological evidence for applied N O to specifically relax the muscularis of the
?To whom correspondence should be addressed. Abbreviations: EDRF, endothelium-derived relaxant factor; NADPH, nicotinamide adenine dinucleotide phosphate, reduced form; NANC, non-adrenergic, non-cholinergic; NBT, nitroblue tetrazolium; PBS, phosphate-buffered saline; PB, phosphate buffer; VIP vasoactive intestinal polypeptide.
rat gastric fundus and ileocaecal junction, canine small and large i n t e s t i n e , 2~'9"~3"~4 guinea-pig stomach and ileum ~7'z7 and opossum oesophageal circular muscle. ~s'46 This has led to the proposal that N O mediates non-adrenergic, non-cholinergic ( N A N C ) inhibitory motor innervation of the mammalian gastrointestinal tract. Little is known, however, about the disposition of N O in the gut wall; if N O is a transmitter of N A N C inhibitory nerves it must be present in myenteric neurons innervating the muscularis as is the case for the rat. 6 This raises questions as to the distribution of these neurons in the gut wall. What proportion of myenteric nerves are NOproducing? What is the pattern of innervation of these neurons. Although a vasodilator role for N O at cerebral and peripheral sites has been proposed for various species, the distribution of N O with respect to the intestinal vasculature is unknown. These questions are addressed in this study where we have modified a histochemical technique for localizing N O synthase-related diaphorase activity in rat brain 24"2s'4° and applied it to stretch preparations of laminae dissected from the wall of the guinea-pig intestine. The guinea-pig is the most investigated and best understood species with regards to enteric nervous innervation and function. This is partly due to the ease with which the various nerve and muscle layers can be separated by fine dissection. The histochemical reaction employed in this investigation is based on the presence of an enzyme, known as a diaphorase, which can reduce a tetrazolium dye, 791
792
K. NICHOLSet al.
nitroblue tetrazolium (NBT), in the presence of the co-factor, reduced nicotinamide adenine dinucleotide phosphate (NADPH), with the resultant product being a visible formazan precipitate. There are now several lines of evidence which suggest that N O synthase activity fully accounts for the NADPHdependent diaphorase activity identified by this histochemical reaction in fixed tissue. NADPH diaphorase activity has been observed in human kidney cells transfected with cDNA for NO synthase. 7~5'~6'43 In addition, purified NO synthase exhibits NADPH diaphorase activity while its own activity is competitively inhibited by NBT. 25 Finally, NADPH diaphorase activity co-localizes not only with NO synthase immunoreactivity, but also with the mRNA for NO synthase both centrally and in the periphery; the latter including myenteric neurons.8'~6Using this modified histochemical technique we have now localized NO synthase activity in the guinea-pig intestine, and here present evidence for NO synthase activity at specific sites in various enteric, neural and vascular networks. EXPERIMENTAL PROCEDURES
Freshly decapitated male guinea-pigs weighing400-500 g were given an abdominal midline incision and all intestinal segments were removed, carefully cleared of contents and immediately washed in pre-chilled 0.1 M sodium phosphatebuffered saline (PBS), pH 7.2. Individual intestinal segments 2.5-4.0 cm in length were cut along the mesenteric border, opened and pinned (mucosa up) to styrofoam. Gut segments were then subjected to a 2-h incubation period at 4°C in 4% paraformaldehyd¢-0.4% picric acid in 0.16 M sodium phosphate buffer (PB), pH 7.0. The pinned out segments were then rinsed three times in pre-chilled0.1 M PBS, pH 7.2, and stored at 4°C in fresh PBS. Laminar preparations Dissection and isolation of the different layers (laminae; Refs 28, 29) of the intestinal wall was carried out using a modification of the method described by Furness and Costa) 9 Individual segments were removed from the styrofoam blocks, stretched and pinned to the base of a 0.1 M PBS-filled container (pH 7.2) with ser0sal side facing up. To obtain laminar preparations of the myenteric plexus, a superficial incision was made along one end of the tissue segment, sul~cient to allow the serosa, longitudinal muscle and myenteric plexus to be peeled away together using fine forceps. This laminar preparation was then positioned with the myenteric plexus uppermost and any adherent circular muscle stripped away. The remaining tissue segment was then further dissected to expose the submucosal layer, by first stripping away the circular muscle, then repositioning the tissue with the mucosal layer uppermost for blunt dissection of the mucosa. These laminar preparations of the (i) myenteric plexus and (ii) submucosal layer (mucosal side down) were then stretched on to electrostatic glass slides, allowing the outer edges of the preparations to air dry and adhere. Enzyme histoehemistry The air-dried preparations were permeabilized by overnight incubation at 4°C in 3% Triton X-100 in a 1: 1 mixture of 0.1 M PBS, pH 7.2, and 10% sucrose in 0.1 M PB, pH 7.2. After washing in several changes of sodium PB (10 mM, pH 8.0), laminar preparations were incubated in a darkened,
moist chamber for 1 h at 37 C with reaction medium for NADPH diaphorase activity consisting of 1 mM NADPH. 0.5 mM NBT, and 0.3% Triton X-100 in 10 mM sodium PB, pH 8.0. Control preparations were treated in the same way but without NADPH in the reaction medium. To ensure that the diaphorase activity observed was not due to DT diaphorase, a second control procedure was carried out by testing the resistance to 0.1 mM dicumarol placed in the reaction medium. Tissues were given a final wash with PB (10raM, pH 8.0) air dried, coverslipped under Permount and analysed by light microscopy. Chemicals NBT and /~-NADPH were purchased from Sigma. The remaining reagents were purchased from BDH RESULTS Nerve and vascular networks, and smooth muscle layers of the gut wall could be easily identified in laminar stretch preparations of the guinea-pig small and large intestine treated for histochemical demonstration of NADPH diaphorase activity under conditions specific for NO synthase-related activity. No staining of gut tissue occurred in the absence of NADPH and all the staining features described below were resistant to inhibition with dicumarol. DT diaphorase, also capable of reducing tetrazolium dyes to formazan precipitates in the presence of NADPH or NADH, is selectively inhibited by dicumarol. '-~ Therefore, DT diaphorase does not account for the diaphorase activity exhibited here. Myenterie plexus In laminar preparations of the myenteric plexus, intensely stained cells were present within or localized to ganglia (Figs 1-6) and occasionally found in the interconnecting fibre bundles (fasciculi) of the primary meshwork of the plexus (Fig. 7). The relative numbers of stained cells/ganglia are shown in Table 1. The density distribution of stained ganglion cells was not constant along the length of the intestine. Rather, the highest frequency of cells/ganglion occurred in the ileum and colon, and lowest in the proximal duodenum. The cytoplasm of the NADPH diaphorase-positive cells was completely labelled except for the nucleus. In addition, reaction product was visible within very long, fine processes enabling description of both the morphology of labelled cells and their projections, Stained cells encompassed a number of morphological types according to the classification by Stach, 42 including types I and II (Figs 2-6) and type VI (Fig. 5). Of these, diaphorase-positive neurons corresponding to type II were most common. In addition, the longest emergent processes were always associated with type II cells. The perikarya of these cells were usually located at the poles of the ganglia with the large emergent process projecting out into the interconnecting fasciculi. These processes could often be traced considerable distances coursing through up to four ganglia of the primary meshwork on into the fine tertiary meshwork that innervates the circular
NO-synthesizing neurons and vascular sites muscle and submucosa (Figs 1, 7). There was no pattern to the projection of labelled fibres in the myenteric meshworks, projecting as they did, circumferentially, radially, aborally and orally. In many instances these labelled fibres were lost from view before any termination could be seen, suggesting that many of the labelled fibres projected considerably further than could be discerned here. Most of the labelled fibres within a ganglion appeared to originate from outside the ganglion, some of which bore varicosities en passant, Others ramified within the ganglion around clearly unlabelled cells. Intensely stained extraganglionic cells41 were rarely encountered (Fig. 7) and when found were small and round (approximately 20/~m in diameter). The primary meshwork fasciculi contained many stained fibres and therefore it was difficult to distinguish whether these extraganglionic cells gave off positive processes. In parts of the tissue where the circular muscle remained attached to the myenteric plexus, intensely stained fibres could be seen coursing amongst the muscle cells (Fig. 13). This pattern is characteristic of the meshwork of nerve fibres that ramify in the deep muscular nerve plexus within the muscularis.2°'-'~4~In regions of the tissue preparation where the circular muscle and myenteric plexus were stripped away, leaving just the longitudinal muscle layer, only a diffuse nonspecific staining was evident. Submucous plexus
Henle's or Schabadash's plexus displayed a profuse network of intensely stained varicose fibres within individual ganglia and throughout the interconnecting fasciculi. This fibre network appeared to be more profuse than in the myenteric plexus and singlestained fibres were distinctly varicose in appearance (Figs 9 11). These fibres could often be traced long distances (up to 800/tin). Within the ganglia, stained varicose fibres could be seen to ramify around clearly unlabelled cells. Few ganglia contained labelled cells. The maximum number of labelled cells/ganglia was found to be two. These cells could be easily identified as type II cells (Figs 10, 11). In preparations where the circular muscle was not completed stripped away, labelled elements of the deep muscular plexus could be seen (Fig. 13) as well. B l o o d t,essels
Underlying the Henle's or Schabadash's plexus an extensive vascular network is visible (Figs 8, 12). The blood vessels and associated paravascular nerve bundles were in close contact with and often connected to fibres of Henle's plexus. Arterioles, easily distinguished by their smaller diameter and characteristic striated appearance displayed an extensive labelling over their surface (Figs 8, 9, 12, 13). The punctate labelling of the arterioles was comprised of uniformly large puncta distributed relatively evenly over the visible surface of the blood vessels (Figs 8, 9). These
793
puncta appeared to represent patches of NO synthase activity within the endothelial cells rather than NO synthase activity on approximated neurites. The venules were not labelled and there was no obvious perivascular labelling. DISCUSSION
As currently understood, there are two forms of NO synthase, one (the constitutive form) which is found in neurons and vascular endothelial cells producing NO for intercellular signalling and a second (inducible) form of NO synthase which is found in macrophages and is used to synthesize NO for use as a toxin in cell-mediated immune responses. The activity of the constitutive form of NO synthase generally accounts for the NADPH diaphorase reactivity of the histochemical procedure employed in this i n v e s t i g a t i o n . 7"8"~-~'1624 NADPH diaphorase activity has been reported in murine enteric neurons at certain stages of development23 and in rat myenteric neurons co-localized with NO synthase immunoreactivity.~6 However, the results of this study show for the first time, the localization and distribution of NO synthase activity in different laminae of the guinea-pig intestine that are similar to our findings for the rat intestine,z* Moreover, the discrete staining of enteric sites by this method allowed for morphological identification of elements staining positive for NO synthase activity. NO synthase-related diaphorase activity was found in both myenteric and submucous neurons and their emergent processes. The frequency of labelled cells seen in this study indicates that less than 10% of myenteric nerve cells have the capacity to produce NO. It is uncertain whether the lightly stained myenteric ganglion cells represent a subcategory of cells that are characterized by low NO synthase activity. Three of the six morphologically classified myenteric nerve cells were, however, found to label intensely under the treatment conditions employed, suggesting that NO synthase is widely distributed in the myenteric plexus. Pharmacological studies in different species show NO to be a powerful relaxant of gastrointestinal smooth muscle2 4,9,13.14A7.18.27,30.38.46and stimulation of intestinal nerves liberate NO. 3 These electrical stimulations induced relaxations of the muscularis that could be blocked by inhibitors of NO synthesis. Taken together, these data are strong evidence for NO to be a mediator of NANC inhibitory motor transmission. The identity of the neurotransmitter(s) mediating the inhibitory motor innervation of the gut wall, however, is controversial. 334~ Moreover, Costa et al. 12 provide evidence for at least two different NANC enteric inhibitory motor nerves. Vasoactive intestinal polypeptide (VIP) and ATP have been proposed as candidate NANC inhibitory transmitters~O,~l,~9,20and these substances have also been localized to types I-III cells of the rodent intestine.2° Whether the NO-positive neurons identified in this
794
K. NI(:HoLs et
study represent N A N C inhibitory neurons already described, or another subset of N A N C inhibitory neurons remains to be determined. To this end we are currently exploring whether ATP, or peptides including VIP are co-localized with NO, and whether the pharmacological actions of these inhibitory agents are indeed mutually exclusive. Only a small number of submucosal nerve cells localized in ganglia of Henle's or Schabadash's plexus displayed N O synthase activity. A conspicuous feature of the submucosal preparations was, however, the extensive labelling of varicose fibres within Henle's or Schabadash's plexus. The ratio of labelled fibres to ganglion cells was high compared to the myenteric plexus. The situation is not unique since we recently identified G A B A e r g i c nerve fibre innervation in the rat and guinea-pig intestinal submucosa to be extensive, while G A B A e r g i c cells were rare. 29 The submucosal nerves are integrated with the myenteric plexus and modulated by innervations from extrinsic autonomic ganglia. 2~'4~ As such, the submucosa normally contain an extensive and complex array of nerve fibres of diverse origins, it is possible that a proportion of the nerve fibre ramifications that were shown to display N O synthase-related diaphorase activity in this investigation are visceral afferents. In addition, a large population of diaphorase-positive neurons and fibres have been found in the autonomic ganglia of the rat. ~
al.
A prominent feature of the actions o l NO m mammals is its effect on vascular smooth muscle (for review, see Ref. 31). No is a powerful vasodilator of central and peripheral blood vessels in a variety of species with similar actions and properties to E D R F . 3~'47 We found extensive N O synthase activity in the submucosal arterioles. The labelling of blood vessels was similar in all regions of the small and large intestine indicative of a homogeneous distribution to the vasculature. This is also a feature of submucosal blood vessels in preparations we have examined from the rat. 36 The NO-related labelling of the enteric vasculature bore none of the characteristics of a perivascular distribution of NO-positive nerve fibres. This suggests an absence of NO vasodilator nerves and therefore no direct NO-related neural modulation of vasomotor activity in the guinea-pig submucosa. Whether this NO-related labelling is associated with the endothelium or vascular smooth muscle cannot be resolved here. However, the pre~ncc of NO-synthesizing sites on blood vessels may represent some intrinsic role for N O as a mediator of neurohumoral control of blood flow, similar to its E D R F like actions at cerebral blood vessels. The ability of specific enteric neurons to produce and release N O together with the ability of this molecule to relax gastrointestinal smooth muscle is highly suggestive of a transmitter/modulator role for enteric NO. Our results show N O synthase-related
Figs 1-13. Light microscopic micrographs of laminae from the guinea-pig intestine treated for histochemical localization of NO synthase-related diaphorase activity. Scale bars = 100 # m. Fig. 1. Laminae of the myenteric plexus from the proximal colon. Intensely stained nerve cells and their processes are present in their ganglia (G). Intensely stained fibres (small arrowheads) can be seen within the ganglia and interconnecting primary (Pp) and secondary (Ps) meshworks of this nerve network. Figs 2 and 3. Myenteric ganglia (G) from the proximal colon showing intensely stained nerve cells and their processes typical of type I and II cells. Fig. 4. High magnification micrograph of a myenteric ganglion from the proximal colon, showing an intensely stained type I cell. Fig. 5. High magnification micrograph of a myenteric ganglion from the proximal colon, showing intensely stained type II and VI cells. Fig. 6. High magnification micrograph of a myenteric ganglion from the distal ileum, showing two type II cells. Fig. 7. Micrograph of the interconnected meshworks of the myenteric plexus from the distal ileum, showing an intensely stained extraganglionic cell in the primary (Pp) meshwork. Fig. 8. Submucosal laminae from the proximal jejunum. Intensely stained varicose fibres (small arrowheads) can be seen coursing throughout the nerve bundles of Henle's plexus. Fibres are evident in a ganglion (G), ramifying around clearly unlabelled cells. Arterioles (Va), but not the venules (Vv), of the underlying vascular network contain intensely stained punctate deposits over their surface. Fig. 9. High magnification of the vascular network and Henle's plexus from the distal ileum. The distinct pattern of labelling of nerve fibres (small arrowheads) vs blood vessels is evident. Fig. 10. Henle's plexus from the distal ileum, showing two labelled ganglion cells and their processes. Intensely stained fibres can be seen coursing within the ganglion (G) and in the interconnecting nerve bundles (small arrowheads) of this plexus. Fig. I 1. A ganglion (G) from Henle's plexus of the distal ileum, showing an intensely stained type 1I cell. Fig. 12. High magnification micrograph of the nerve and vascular networks of the proximal duodenum submucosa, Intensely stained varicose fibres can be seen coursing through a ganglion (G) and in the nerve bundles of Henle's plexus. The underlying arterioles (Va) display intense punctate labelling. Fig. 13. A lamina from the submucosa of the proximal jejunum. Intensely stained fibres (small arrowheads) can be seen coursing in the plane of the circular muscle layer (cm) still adherent to the underlying submucosal laminae. Submucosal nerve and vascular networks can be seen beneath the muscle layer.
I
m I
jff) vvvvv.
-" o
z
~.,
Z
NO-synthesizing neurons and vascular sites
797
Va B
Figs 10 13.
798
K. NICHOLSeta/.
Table 1. Frequency of nicotinamide adenine dinucleotide phosphate (reduced form)-diaphorase reactive cells in the myenteric plexus in different regions of the guinea-pig intestine Region
Duodenum
Cells per ganglion
5+ 1
Jejunum Ileum Colon 8+ 2
10 + 2
9+ 1
Numbers represent the mean number of cells per ganglion (+ standard deviation) from manual counts obtained in 10 preparations/region.
diaphorase labelling of a spectrum of enteric nerve types along the length of the intestine. This indicates that N O may be released at multiple gastrointestinal
sites similar to a variety of putative enteric transmitters. Although this myenteric and submucosal nerverelated N O synthase labelling displayed similar histochemical sensitivity to that associated with the submucosal arterioles, the pattern of labelling suggests that they are not functionally linked. These findings show the distribution of NOrelated elements to neural and vascular networks in the guinea-pig intestine and raise a number of questions as to the functional significance of enteric NO. We are now using transmitter identification techniques to examine the involvement of N O with pre- and postsynaptic elements to determine the role of N O in enteric neural function.
REFERENCES
1. Aimi Y., Fujimura M., Vincent S. R. and Kimura H. (1991) Localization of NADPH-diaphorase-containing neurons in sensory ganglia of the rat. J. comp. Neurol. 306, 382-392. 2. Boeckxstaens G. E., Pelekmans P. A., Rampart M., Ruytjens I. F., Verbeuren T. J., Herman A. G. and Van Maerke Y. M. (1990) GABAA receptor-mediated stimulation of non-adrenergic non-cholinergic neurons in the dog ileocolonic junction. Br. J. Pharmac. 101, 460-464. 3. Boeckxstaens G. E., Pelckmans P. A., Bogers J. J., Bult H., De Man J. G., Oosterbosch L,, Herman A. G. and Van Maereke Y. M. (1991) Release of nitric oxide upon stimulation of non-adrenergic non-cholinergic nerves in the rat gastric fundus. J'. Pharmac. exp. Ther. 256, 441-447. 4. Boeekxstaens G. E., Pelckmans P. A., Bult J., De Man K. G., Herman A. G. and Van Maercke Y. M. (1991) Evidence for nitric oxide as mediator of non-adrenergic, non-cholinergic relaxations induced by ATP and GABA in the canine gut. Br. J. Pharmac. 102, 434-438. 5. Branchek T. A. and Gershon M. D. (1989) Time course of expression of neuropeptide Y, calcitonin gene-related peptide, and NADPH diaphorase activity in neurons of the developing murine bowel and the appearance of 5-hydroxytryptamine in mucosal enterochromaflin cells. J. comp. Neurol. 285, 262-273. 6. Bredt D. S., Hwang P. M. and Snyder S. H. (1990) Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347, 768-770. 7. Bredt D. S., Hwang P. M., Glatt C. S., Lowenstein C,, Reed R. R. and Snyder S. H. (199t) Cloned and expressed nitric oxide synthase structurally resembles cytochrome P-450 reductase. Nature 351, 714-718. 8. Bredt D. S., Glatt C. E., Hwang P: M., Fotuhi J., Dawson T, M. and Snyder S. H. (1991) Nitric oxide synthase protein and mRNA are discretely localized in neuronal populations of the mammalian CNS together with NADPH diaphorase. Neuron 7, 615-624. 9. Bult H. (1990) Nitric oxide as an inhibitory non-adrenergic non-cholinergic neurotransmitter. Nature 345, 346-347. 10. Burnstock G. (1972) Purinergic nerves. Pharmac. Rev. 24, 509-563. 11. Costa M. and Furness J. B. (1983) The origins, pathways and terminations of neurons with VIP-like immunoreactivity in the guinea-pig small intestine. Neuroscience 8, 665-676. 12. Costa M., Furness J. B. and Humphreys C. M. S. (1986) Apamin distinguishes two types of relaxation mediated by enteric nerves in the guinea-pig gastrointestinal tract. Arch. Pharmac. 332, 79-88. 13. Dalziel H. H., Thornbury K. D., Ward S. M. and Sanders K. M. (1991) Involvement of nitric oxide synthetic pathway in inhibitory junction potentials in canine proximal colon. Am. J. Physiol. 260, G789-G792. 14. D'Amato D., Curro D. and Montuschi P. (1991) L-N-nitro arginine partially inhibits non-adrenergic non-cholinergic relaxation in the rat gastric fundus. Gastroenterology 100, A434. 15. Dawson T. W., Dawson V. I., Bredt D. S., Fotuhi M., Hwang P. M., Uhl G. R. and Snyder S. H. (1991) Nitric Oxide Synthase/NADPH Diaphorase Neurons: Role in Neurotoxicity, 21st Annual Meeting, New Orleans, Nov. 10-15, Vol. 17, 314.7. 16. Dawson T. M., Bredt D. S., Fotuhi J., Hwang P. M. and Snyder S. H. (1991) Nitric oxide synthase and neuronal NADPH diaphorase are identical in brain and peripheral tissues. Proc. natn. Acad. Sci. U.S.AI 88, 7797-7801. 17. Desai K. M., Sessa W. C. and Vane J. R. (1991) Involvement of nitric oxide in the reflex relaxation of the stomach to accommodate food or fluid. Nature 351, 477-479. 18. Du C., Murray J., Conlin J. L. and Bates J. N. (1991) Nitric oxide: mediator of non-adrenergic non-cholinergic hyperpolarization of opossum esophageal circular smooth muscle. Gastroenterology. 100, A439. 19. Furness J. B. and Costa M. (1980) Types of nerves in the enteric nervous system. Neuroscience 5, I--20. 20. Furness J. B. and Costa M. (1987) The Enteric Nervous System. Churchill Livingstone, New York. 21. Gabella G. (1979) Innervation of the gastrointestinal tract. Int. Rev. Cytol. 59, 129-193. 22. Gaily J. A., Montague P. R., Reeke G. N. and Edelman G. M. (1990) The NO hypothesis: possible effects of a short-lived, rapidly diffusible signal in the development and function of the nervous system. Proc. natn. Acad. Sci. 87, 3547-3551. 23. Garthwaite J., Charles S. L. and Chess-Williams R. (1988) Endothelium-derived relaxing factor release on activation of NMDA receptors suggests role as intercellular messenger in the brain. Nature, Lond. 336, 385-388. 24. Hope B. T. and Vincent S. R. (1989) Histochemical characterization of neuronal NADPH-diaphorase. J. Histochem. Cytochem. 37, 653-661. 25. Hope B. T., Michael G. J., Knigge K. M. and Vincent S. R. (1991) Neuronal NADPH diaphorase is a nitric oxide synthase. Proc. natn. Acad. Sci. U.S.A. 88, 2811-2814.
NO-synthesizing neurons and vascular sites
799
26. Ignarro L. J., Buga G. M., Wood K. S., Byrns R. H. and Chaudhuri G. (1987) Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc. natn. Aead. Sci. U.S.A. 84, 9265 9269. 27. Knudsen M. A., Tottrup A. and Aarhus L. (1991) Evidence for two mediators of NANC inhibition in circular muscle of the guinea-pig ileum. Gastroenterology 100, A459. 28. Krantis A. and Clark D. (1991) Localization of [3H]GABA-labelled nerve fibre networks in the rat intestinal mucosa. J. auton, nerv. Syst. 34, 195-200. 29. Krantis A. K. and Clark D. (1991) High-affinity uptake of [3H]GABA by submucous ganglion cells, nerve fibres and peri- and para-vascular fibres in guinea-pig and rat intestine. J. auton, nerv. Syst. 32, 251 258. 30. Li C. G. and Rand M. J. (1990) Nitric oxide and vasoactive intestinal polypeptide mediate non-adrenergic, non-cholinergic inhibitory transmission to smooth muscle of the rat gastric fundus. Eur. J. Pharmac. 191, 303 309. 31. Long C. J. and Berkowitz B. A. (1989) Minireview--What is the relationship between the endothelium derived relaxant factor and nitric oxide? Life Sci. 45, 1 14. 32. Marietta M. A. (1989) Nitric oxide: biosynthesis and biological significance. Trends" biol. Sci. 14, 488 492. 33. Matusak O. and Bauer V. (1986) Effect of desensitization induced by adenosine 5'-triphosphate, substance P, bradykinin, serotonin, gamma-aminobutyric acid and endogenous non-cholinergic, non-adrenergic transmitter in the guinea-pig ileum. Eur. J. Pharmac. 126, 199 209, 34. Moncada S., Palmer R. M. J. and Higgs E. A. (1989) The biological significance of nitric oxide formation from L-arginine. Bioehem. Soc. Trans. 17, 642-4544. 35. Moncada S., Palmer M. J. and Higgs E. A. (1991~ Nitric oxide: physiology, pathophysiology and pharmacology. Pharmac. Rev. 43, 109 142. 36. Nichols K., K rantis A. and Staines W. (1991) NADPH-diaphorase activity, a marker of nitric oxide (NO)-synthesizing neurons, is localized to enteric neurons and vasculature. Physiol. Can. 21, 249. 37. Pahner R. M. J., Ferridge A. C. and Moncada S. (1987) Nitric oxide release accounts for the biological activity of endothelium-derived releasing factor. Nature 327, 524 526. 38. Pelckmans P. A., Boeckxstaens J. G., Deman, H. B., Herman A. G. and Vanmaercke Y. M. (1991) Nitric oxide as an inhibitory non-adrenergic non-cholinergic neurotransrnitter in the gastrointestinal tract. Gastroenterology 100, A480. 39. Pique J. M., Whittle B. J. R. and Esplugues J. V. (1989) the vasodilator role of endogenous nitric oxide in the rat gastric microcirculation. Eur. J. Pharmac. 174, 293 296. 40. Scherer-Singler U., Vincent S. R., Kimura H. and McGeer E. G. (1983) Demonstration of a unique population of neurons with NADPH-diaphorase histochemistry. J. Neurosci. Meth. 9, 229 234. 41. Schofield G. C. (1989) Anatomy of muscular and neural tissues in the alimentary canal. In Handbook (~l~Physiology, Section6, Vol. 1, pp. 1579 1627. Am. Physiol. Soc. 42. Stach W. (1989) A revised morphological classification of neurons in the enteric nervous system. In Nerves and the Gastrointestinal Tract (eds Singer M. V. and Goebell H.), pp. 29 44. Klewer Academic Publishers, MA, U.S.A. 43. Snyder S. H. and Bredt D. S. (1991) Nitric oxide as a neuronal messenger. Trends pharmae. Sci. 12, 125 128. 44. Toda N., Baba H. and Okamura T. (1990) Role of nitric oxide in non-adrenergic, non-cholinergic nerve-mediated relaxation in dog duodenal longitudinal muscle strips. Jap. J. Pharmac. 53, 281 284. 45. Toda N. and Okamura T. (1990) Modification by L-N-monomethylarginine (L-NMMA) of the response of nerve stimulation in isolated dog mesenteric and cerebral arteries. Jap. J. Pharmae. 52, 170 173. 46. Tottrup A., Suane D. and Forman R. (1991) Nitric oxide mediates NANC inhibition in opossum lower esophageal sphincter. Am. J. Physiol. 260, G385-G389. 47. Vallance P., Collier J. and Moncada S. (1989) Nitric oxide synthesized from L-arginine mediates endothelium-dependent dilatation in human veins. Cardiovasc. Res. 23, 1053 1057. 48. Westfall D. P., Hagoboom G. K., Colby J., O'Donnel J. P. and Fedan J. S. (1982) Direct evidence against a role of ATP as the non-adrenergic, non-cholinergic inhibitory neurotransmitter in guinea-pig taenia coli. Proc. natn. Acad. Sci. U.S.A. 79, 7041 7045. 49. Whittle B. J. R., Lopez-Belmonte J. and Rees D. D. (1989) Modulation of the vasodepressor actions of acetylcholine, bradykinin, substance P and endothelin in the rat by a specific inhibitor of nitric oxide formation. Br. J. Pharmac. 98, 646 652. (Accepted 18 June 1992)
NSC 5t/4
("
Neuroscience Vol. 51, No. 4, pp. 791 799, 1992
Printed in Great Britain
H I S T O C H E M I C A L L O C A L I Z A T I O N OF NITRIC OXIDE-SYNTHESIZING NEURONS AND VASCULAR SITES IN THE G U I N E A - P I G INTESTINE K. NICHOLS,*A. KRANTIS*t and W. STAINES~ *Digestive Diseases Research Group, Department of Physiology and :~Department of Anatomy, University of Ottawa, Ottawa, Canada K1H 8M5 A~tract--Laminar preparations of fixed segments of the guinea-pig intestine were examined for nitric oxide synthase activity using reduced nicotinamide adenine dinucleotide phosphate and nitroblue tetrazolium salt as substrates. Under conditions specific for detecting nitric oxide synthase-related diaphorase activity, a subpopulation of neural elements in the myenteric plexus, deep muscular plexus and submucosa were intensely stained. Intensely stained nerve fibres were distributed throughout the meshworks of the myenteric plexus and its innervation of the circular muscle, and in the submucosa within Henle's plexus. Intensely stained nerve cells and their processes were evident in most myenteric ganglia but were rare in ganglia of Henle's plexus. Stained ganglion cells comprised types I, II and VI of the morphologically defined enteric nerve cells. Stained neural elements were increasingly prevalent within successively more caudal segments of the intestine. In addition to neuronal staining, arterioles of the submucosal vascular network displayed distinct, punctate patches of staining distributed over their surface. Perivascular nerve fibre staining was absent. These results show nitric oxide synthase activity to be present within neurons and fibres of the major enteric nerve layers and within submucosal blood vessels throughout the guinea-pig small and large intestine.
Nitric oxide (NO) is a free radical molecule produced from the amino acid L-arginine in neural and nonneural tissue by the action of a family of N O synthases. 32'34 N O is a neural messenger in the brain where its synthesis in some cells is stimulated by Ca 2+ entry subsequent to excitatory amino acid (N-methylD-aspartate) receptor stimulation. U p o n synthesis, N O diffuses to neighbouring cells where it interacts with the ferrous haeme of soluble guanylate cyclase and stimulates formation of cGMP. This process has been seen to carry signals between neurons, and between neurons and glial cells. 35 Proposed roles for N O in C N S function include synaptic plasticity and long-term potentiation, and the activity-dependent determinant of neural development. 22'23 In addition, it is proposed that vasodilator nerves release N O and therefore are involved in the neural control of cerebral blood flow. 45 N O is also produced by central and peripheral blood vessels 26'34'37'39'49 and is believed to be responsible for the activity of the endotheliumderived relaxant factor ( E D R F ) . 28'34 There is now pharmacological evidence for applied N O to specifically relax the muscularis of the
?To whom correspondence should be addressed. Abbreviations: EDRF, endothelium-derived relaxant factor; NADPH, nicotinamide adenine dinucleotide phosphate, reduced form; NANC, non-adrenergic, non-cholinergic; NBT, nitroblue tetrazolium; PBS, phosphate-buffered saline; PB, phosphate buffer; VIP vasoactive intestinal polypeptide.
rat gastric fundus and ileocaecal junction, canine small and large i n t e s t i n e , 2~'9"~3"~4 guinea-pig stomach and ileum ~7'z7 and opossum oesophageal circular muscle. ~s'46 This has led to the proposal that N O mediates non-adrenergic, non-cholinergic ( N A N C ) inhibitory motor innervation of the mammalian gastrointestinal tract. Little is known, however, about the disposition of N O in the gut wall; if N O is a transmitter of N A N C inhibitory nerves it must be present in myenteric neurons innervating the muscularis as is the case for the rat. 6 This raises questions as to the distribution of these neurons in the gut wall. What proportion of myenteric nerves are NOproducing? What is the pattern of innervation of these neurons. Although a vasodilator role for N O at cerebral and peripheral sites has been proposed for various species, the distribution of N O with respect to the intestinal vasculature is unknown. These questions are addressed in this study where we have modified a histochemical technique for localizing N O synthase-related diaphorase activity in rat brain 24"2s'4° and applied it to stretch preparations of laminae dissected from the wall of the guinea-pig intestine. The guinea-pig is the most investigated and best understood species with regards to enteric nervous innervation and function. This is partly due to the ease with which the various nerve and muscle layers can be separated by fine dissection. The histochemical reaction employed in this investigation is based on the presence of an enzyme, known as a diaphorase, which can reduce a tetrazolium dye, 791
792
K. NICHOLSet al.
nitroblue tetrazolium (NBT), in the presence of the co-factor, reduced nicotinamide adenine dinucleotide phosphate (NADPH), with the resultant product being a visible formazan precipitate. There are now several lines of evidence which suggest that N O synthase activity fully accounts for the NADPHdependent diaphorase activity identified by this histochemical reaction in fixed tissue. NADPH diaphorase activity has been observed in human kidney cells transfected with cDNA for NO synthase. 7~5'~6'43 In addition, purified NO synthase exhibits NADPH diaphorase activity while its own activity is competitively inhibited by NBT. 25 Finally, NADPH diaphorase activity co-localizes not only with NO synthase immunoreactivity, but also with the mRNA for NO synthase both centrally and in the periphery; the latter including myenteric neurons.8'~6Using this modified histochemical technique we have now localized NO synthase activity in the guinea-pig intestine, and here present evidence for NO synthase activity at specific sites in various enteric, neural and vascular networks. EXPERIMENTAL PROCEDURES
Freshly decapitated male guinea-pigs weighing400-500 g were given an abdominal midline incision and all intestinal segments were removed, carefully cleared of contents and immediately washed in pre-chilled 0.1 M sodium phosphatebuffered saline (PBS), pH 7.2. Individual intestinal segments 2.5-4.0 cm in length were cut along the mesenteric border, opened and pinned (mucosa up) to styrofoam. Gut segments were then subjected to a 2-h incubation period at 4°C in 4% paraformaldehyd¢-0.4% picric acid in 0.16 M sodium phosphate buffer (PB), pH 7.0. The pinned out segments were then rinsed three times in pre-chilled0.1 M PBS, pH 7.2, and stored at 4°C in fresh PBS. Laminar preparations Dissection and isolation of the different layers (laminae; Refs 28, 29) of the intestinal wall was carried out using a modification of the method described by Furness and Costa) 9 Individual segments were removed from the styrofoam blocks, stretched and pinned to the base of a 0.1 M PBS-filled container (pH 7.2) with ser0sal side facing up. To obtain laminar preparations of the myenteric plexus, a superficial incision was made along one end of the tissue segment, sul~cient to allow the serosa, longitudinal muscle and myenteric plexus to be peeled away together using fine forceps. This laminar preparation was then positioned with the myenteric plexus uppermost and any adherent circular muscle stripped away. The remaining tissue segment was then further dissected to expose the submucosal layer, by first stripping away the circular muscle, then repositioning the tissue with the mucosal layer uppermost for blunt dissection of the mucosa. These laminar preparations of the (i) myenteric plexus and (ii) submucosal layer (mucosal side down) were then stretched on to electrostatic glass slides, allowing the outer edges of the preparations to air dry and adhere. Enzyme histoehemistry The air-dried preparations were permeabilized by overnight incubation at 4°C in 3% Triton X-100 in a 1: 1 mixture of 0.1 M PBS, pH 7.2, and 10% sucrose in 0.1 M PB, pH 7.2. After washing in several changes of sodium PB (10 mM, pH 8.0), laminar preparations were incubated in a darkened,
moist chamber for 1 h at 37 C with reaction medium for NADPH diaphorase activity consisting of 1 mM NADPH. 0.5 mM NBT, and 0.3% Triton X-100 in 10 mM sodium PB, pH 8.0. Control preparations were treated in the same way but without NADPH in the reaction medium. To ensure that the diaphorase activity observed was not due to DT diaphorase, a second control procedure was carried out by testing the resistance to 0.1 mM dicumarol placed in the reaction medium. Tissues were given a final wash with PB (10raM, pH 8.0) air dried, coverslipped under Permount and analysed by light microscopy. Chemicals NBT and /~-NADPH were purchased from Sigma. The remaining reagents were purchased from BDH RESULTS Nerve and vascular networks, and smooth muscle layers of the gut wall could be easily identified in laminar stretch preparations of the guinea-pig small and large intestine treated for histochemical demonstration of NADPH diaphorase activity under conditions specific for NO synthase-related activity. No staining of gut tissue occurred in the absence of NADPH and all the staining features described below were resistant to inhibition with dicumarol. DT diaphorase, also capable of reducing tetrazolium dyes to formazan precipitates in the presence of NADPH or NADH, is selectively inhibited by dicumarol. '-~ Therefore, DT diaphorase does not account for the diaphorase activity exhibited here. Myenterie plexus In laminar preparations of the myenteric plexus, intensely stained cells were present within or localized to ganglia (Figs 1-6) and occasionally found in the interconnecting fibre bundles (fasciculi) of the primary meshwork of the plexus (Fig. 7). The relative numbers of stained cells/ganglia are shown in Table 1. The density distribution of stained ganglion cells was not constant along the length of the intestine. Rather, the highest frequency of cells/ganglion occurred in the ileum and colon, and lowest in the proximal duodenum. The cytoplasm of the NADPH diaphorase-positive cells was completely labelled except for the nucleus. In addition, reaction product was visible within very long, fine processes enabling description of both the morphology of labelled cells and their projections, Stained cells encompassed a number of morphological types according to the classification by Stach, 42 including types I and II (Figs 2-6) and type VI (Fig. 5). Of these, diaphorase-positive neurons corresponding to type II were most common. In addition, the longest emergent processes were always associated with type II cells. The perikarya of these cells were usually located at the poles of the ganglia with the large emergent process projecting out into the interconnecting fasciculi. These processes could often be traced considerable distances coursing through up to four ganglia of the primary meshwork on into the fine tertiary meshwork that innervates the circular
NO-synthesizing neurons and vascular sites muscle and submucosa (Figs 1, 7). There was no pattern to the projection of labelled fibres in the myenteric meshworks, projecting as they did, circumferentially, radially, aborally and orally. In many instances these labelled fibres were lost from view before any termination could be seen, suggesting that many of the labelled fibres projected considerably further than could be discerned here. Most of the labelled fibres within a ganglion appeared to originate from outside the ganglion, some of which bore varicosities en passant, Others ramified within the ganglion around clearly unlabelled cells. Intensely stained extraganglionic cells41 were rarely encountered (Fig. 7) and when found were small and round (approximately 20/~m in diameter). The primary meshwork fasciculi contained many stained fibres and therefore it was difficult to distinguish whether these extraganglionic cells gave off positive processes. In parts of the tissue where the circular muscle remained attached to the myenteric plexus, intensely stained fibres could be seen coursing amongst the muscle cells (Fig. 13). This pattern is characteristic of the meshwork of nerve fibres that ramify in the deep muscular nerve plexus within the muscularis.2°'-'~4~In regions of the tissue preparation where the circular muscle and myenteric plexus were stripped away, leaving just the longitudinal muscle layer, only a diffuse nonspecific staining was evident. Submucous plexus
Henle's or Schabadash's plexus displayed a profuse network of intensely stained varicose fibres within individual ganglia and throughout the interconnecting fasciculi. This fibre network appeared to be more profuse than in the myenteric plexus and singlestained fibres were distinctly varicose in appearance (Figs 9 11). These fibres could often be traced long distances (up to 800/tin). Within the ganglia, stained varicose fibres could be seen to ramify around clearly unlabelled cells. Few ganglia contained labelled cells. The maximum number of labelled cells/ganglia was found to be two. These cells could be easily identified as type II cells (Figs 10, 11). In preparations where the circular muscle was not completed stripped away, labelled elements of the deep muscular plexus could be seen (Fig. 13) as well. B l o o d t,essels
Underlying the Henle's or Schabadash's plexus an extensive vascular network is visible (Figs 8, 12). The blood vessels and associated paravascular nerve bundles were in close contact with and often connected to fibres of Henle's plexus. Arterioles, easily distinguished by their smaller diameter and characteristic striated appearance displayed an extensive labelling over their surface (Figs 8, 9, 12, 13). The punctate labelling of the arterioles was comprised of uniformly large puncta distributed relatively evenly over the visible surface of the blood vessels (Figs 8, 9). These
793
puncta appeared to represent patches of NO synthase activity within the endothelial cells rather than NO synthase activity on approximated neurites. The venules were not labelled and there was no obvious perivascular labelling. DISCUSSION
As currently understood, there are two forms of NO synthase, one (the constitutive form) which is found in neurons and vascular endothelial cells producing NO for intercellular signalling and a second (inducible) form of NO synthase which is found in macrophages and is used to synthesize NO for use as a toxin in cell-mediated immune responses. The activity of the constitutive form of NO synthase generally accounts for the NADPH diaphorase reactivity of the histochemical procedure employed in this i n v e s t i g a t i o n . 7"8"~-~'1624 NADPH diaphorase activity has been reported in murine enteric neurons at certain stages of development23 and in rat myenteric neurons co-localized with NO synthase immunoreactivity.~6 However, the results of this study show for the first time, the localization and distribution of NO synthase activity in different laminae of the guinea-pig intestine that are similar to our findings for the rat intestine,z* Moreover, the discrete staining of enteric sites by this method allowed for morphological identification of elements staining positive for NO synthase activity. NO synthase-related diaphorase activity was found in both myenteric and submucous neurons and their emergent processes. The frequency of labelled cells seen in this study indicates that less than 10% of myenteric nerve cells have the capacity to produce NO. It is uncertain whether the lightly stained myenteric ganglion cells represent a subcategory of cells that are characterized by low NO synthase activity. Three of the six morphologically classified myenteric nerve cells were, however, found to label intensely under the treatment conditions employed, suggesting that NO synthase is widely distributed in the myenteric plexus. Pharmacological studies in different species show NO to be a powerful relaxant of gastrointestinal smooth muscle2 4,9,13.14A7.18.27,30.38.46and stimulation of intestinal nerves liberate NO. 3 These electrical stimulations induced relaxations of the muscularis that could be blocked by inhibitors of NO synthesis. Taken together, these data are strong evidence for NO to be a mediator of NANC inhibitory motor transmission. The identity of the neurotransmitter(s) mediating the inhibitory motor innervation of the gut wall, however, is controversial. 334~ Moreover, Costa et al. 12 provide evidence for at least two different NANC enteric inhibitory motor nerves. Vasoactive intestinal polypeptide (VIP) and ATP have been proposed as candidate NANC inhibitory transmitters~O,~l,~9,20and these substances have also been localized to types I-III cells of the rodent intestine.2° Whether the NO-positive neurons identified in this
794
K. NI(:HoLs et
study represent N A N C inhibitory neurons already described, or another subset of N A N C inhibitory neurons remains to be determined. To this end we are currently exploring whether ATP, or peptides including VIP are co-localized with NO, and whether the pharmacological actions of these inhibitory agents are indeed mutually exclusive. Only a small number of submucosal nerve cells localized in ganglia of Henle's or Schabadash's plexus displayed N O synthase activity. A conspicuous feature of the submucosal preparations was, however, the extensive labelling of varicose fibres within Henle's or Schabadash's plexus. The ratio of labelled fibres to ganglion cells was high compared to the myenteric plexus. The situation is not unique since we recently identified G A B A e r g i c nerve fibre innervation in the rat and guinea-pig intestinal submucosa to be extensive, while G A B A e r g i c cells were rare. 29 The submucosal nerves are integrated with the myenteric plexus and modulated by innervations from extrinsic autonomic ganglia. 2~'4~ As such, the submucosa normally contain an extensive and complex array of nerve fibres of diverse origins, it is possible that a proportion of the nerve fibre ramifications that were shown to display N O synthase-related diaphorase activity in this investigation are visceral afferents. In addition, a large population of diaphorase-positive neurons and fibres have been found in the autonomic ganglia of the rat. ~
al.
A prominent feature of the actions o l NO m mammals is its effect on vascular smooth muscle (for review, see Ref. 31). No is a powerful vasodilator of central and peripheral blood vessels in a variety of species with similar actions and properties to E D R F . 3~'47 We found extensive N O synthase activity in the submucosal arterioles. The labelling of blood vessels was similar in all regions of the small and large intestine indicative of a homogeneous distribution to the vasculature. This is also a feature of submucosal blood vessels in preparations we have examined from the rat. 36 The NO-related labelling of the enteric vasculature bore none of the characteristics of a perivascular distribution of NO-positive nerve fibres. This suggests an absence of NO vasodilator nerves and therefore no direct NO-related neural modulation of vasomotor activity in the guinea-pig submucosa. Whether this NO-related labelling is associated with the endothelium or vascular smooth muscle cannot be resolved here. However, the pre~ncc of NO-synthesizing sites on blood vessels may represent some intrinsic role for N O as a mediator of neurohumoral control of blood flow, similar to its E D R F like actions at cerebral blood vessels. The ability of specific enteric neurons to produce and release N O together with the ability of this molecule to relax gastrointestinal smooth muscle is highly suggestive of a transmitter/modulator role for enteric NO. Our results show N O synthase-related
Figs 1-13. Light microscopic micrographs of laminae from the guinea-pig intestine treated for histochemical localization of NO synthase-related diaphorase activity. Scale bars = 100 # m. Fig. 1. Laminae of the myenteric plexus from the proximal colon. Intensely stained nerve cells and their processes are present in their ganglia (G). Intensely stained fibres (small arrowheads) can be seen within the ganglia and interconnecting primary (Pp) and secondary (Ps) meshworks of this nerve network. Figs 2 and 3. Myenteric ganglia (G) from the proximal colon showing intensely stained nerve cells and their processes typical of type I and II cells. Fig. 4. High magnification micrograph of a myenteric ganglion from the proximal colon, showing an intensely stained type I cell. Fig. 5. High magnification micrograph of a myenteric ganglion from the proximal colon, showing intensely stained type II and VI cells. Fig. 6. High magnification micrograph of a myenteric ganglion from the distal ileum, showing two type II cells. Fig. 7. Micrograph of the interconnected meshworks of the myenteric plexus from the distal ileum, showing an intensely stained extraganglionic cell in the primary (Pp) meshwork. Fig. 8. Submucosal laminae from the proximal jejunum. Intensely stained varicose fibres (small arrowheads) can be seen coursing throughout the nerve bundles of Henle's plexus. Fibres are evident in a ganglion (G), ramifying around clearly unlabelled cells. Arterioles (Va), but not the venules (Vv), of the underlying vascular network contain intensely stained punctate deposits over their surface. Fig. 9. High magnification of the vascular network and Henle's plexus from the distal ileum. The distinct pattern of labelling of nerve fibres (small arrowheads) vs blood vessels is evident. Fig. 10. Henle's plexus from the distal ileum, showing two labelled ganglion cells and their processes. Intensely stained fibres can be seen coursing within the ganglion (G) and in the interconnecting nerve bundles (small arrowheads) of this plexus. Fig. I 1. A ganglion (G) from Henle's plexus of the distal ileum, showing an intensely stained type 1I cell. Fig. 12. High magnification micrograph of the nerve and vascular networks of the proximal duodenum submucosa, Intensely stained varicose fibres can be seen coursing through a ganglion (G) and in the nerve bundles of Henle's plexus. The underlying arterioles (Va) display intense punctate labelling. Fig. 13. A lamina from the submucosa of the proximal jejunum. Intensely stained fibres (small arrowheads) can be seen coursing in the plane of the circular muscle layer (cm) still adherent to the underlying submucosal laminae. Submucosal nerve and vascular networks can be seen beneath the muscle layer.
I
m I
jff) vvvvv.
-" o
z
~.,
Z
NO-synthesizing neurons and vascular sites
797
Va B
Figs 10 13.
798
K. NICHOLSeta/.
Table 1. Frequency of nicotinamide adenine dinucleotide phosphate (reduced form)-diaphorase reactive cells in the myenteric plexus in different regions of the guinea-pig intestine Region
Duodenum
Cells per ganglion
5+ 1
Jejunum Ileum Colon 8+ 2
10 + 2
9+ 1
Numbers represent the mean number of cells per ganglion (+ standard deviation) from manual counts obtained in 10 preparations/region.
diaphorase labelling of a spectrum of enteric nerve types along the length of the intestine. This indicates that N O may be released at multiple gastrointestinal
sites similar to a variety of putative enteric transmitters. Although this myenteric and submucosal nerverelated N O synthase labelling displayed similar histochemical sensitivity to that associated with the submucosal arterioles, the pattern of labelling suggests that they are not functionally linked. These findings show the distribution of NOrelated elements to neural and vascular networks in the guinea-pig intestine and raise a number of questions as to the functional significance of enteric NO. We are now using transmitter identification techniques to examine the involvement of N O with pre- and postsynaptic elements to determine the role of N O in enteric neural function.
REFERENCES
1. Aimi Y., Fujimura M., Vincent S. R. and Kimura H. (1991) Localization of NADPH-diaphorase-containing neurons in sensory ganglia of the rat. J. comp. Neurol. 306, 382-392. 2. Boeckxstaens G. E., Pelekmans P. A., Rampart M., Ruytjens I. F., Verbeuren T. J., Herman A. G. and Van Maerke Y. M. (1990) GABAA receptor-mediated stimulation of non-adrenergic non-cholinergic neurons in the dog ileocolonic junction. Br. J. Pharmac. 101, 460-464. 3. Boeckxstaens G. E., Pelckmans P. A., Bogers J. J., Bult H., De Man J. G., Oosterbosch L,, Herman A. G. and Van Maereke Y. M. (1991) Release of nitric oxide upon stimulation of non-adrenergic non-cholinergic nerves in the rat gastric fundus. J'. Pharmac. exp. Ther. 256, 441-447. 4. Boeekxstaens G. E., Pelckmans P. A., Bult J., De Man K. G., Herman A. G. and Van Maercke Y. M. (1991) Evidence for nitric oxide as mediator of non-adrenergic, non-cholinergic relaxations induced by ATP and GABA in the canine gut. Br. J. Pharmac. 102, 434-438. 5. Branchek T. A. and Gershon M. D. (1989) Time course of expression of neuropeptide Y, calcitonin gene-related peptide, and NADPH diaphorase activity in neurons of the developing murine bowel and the appearance of 5-hydroxytryptamine in mucosal enterochromaflin cells. J. comp. Neurol. 285, 262-273. 6. Bredt D. S., Hwang P. M. and Snyder S. H. (1990) Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347, 768-770. 7. Bredt D. S., Hwang P. M., Glatt C. S., Lowenstein C,, Reed R. R. and Snyder S. H. (199t) Cloned and expressed nitric oxide synthase structurally resembles cytochrome P-450 reductase. Nature 351, 714-718. 8. Bredt D. S., Glatt C. E., Hwang P: M., Fotuhi J., Dawson T, M. and Snyder S. H. (1991) Nitric oxide synthase protein and mRNA are discretely localized in neuronal populations of the mammalian CNS together with NADPH diaphorase. Neuron 7, 615-624. 9. Bult H. (1990) Nitric oxide as an inhibitory non-adrenergic non-cholinergic neurotransmitter. Nature 345, 346-347. 10. Burnstock G. (1972) Purinergic nerves. Pharmac. Rev. 24, 509-563. 11. Costa M. and Furness J. B. (1983) The origins, pathways and terminations of neurons with VIP-like immunoreactivity in the guinea-pig small intestine. Neuroscience 8, 665-676. 12. Costa M., Furness J. B. and Humphreys C. M. S. (1986) Apamin distinguishes two types of relaxation mediated by enteric nerves in the guinea-pig gastrointestinal tract. Arch. Pharmac. 332, 79-88. 13. Dalziel H. H., Thornbury K. D., Ward S. M. and Sanders K. M. (1991) Involvement of nitric oxide synthetic pathway in inhibitory junction potentials in canine proximal colon. Am. J. Physiol. 260, G789-G792. 14. D'Amato D., Curro D. and Montuschi P. (1991) L-N-nitro arginine partially inhibits non-adrenergic non-cholinergic relaxation in the rat gastric fundus. Gastroenterology 100, A434. 15. Dawson T. W., Dawson V. I., Bredt D. S., Fotuhi M., Hwang P. M., Uhl G. R. and Snyder S. H. (1991) Nitric Oxide Synthase/NADPH Diaphorase Neurons: Role in Neurotoxicity, 21st Annual Meeting, New Orleans, Nov. 10-15, Vol. 17, 314.7. 16. Dawson T. M., Bredt D. S., Fotuhi J., Hwang P. M. and Snyder S. H. (1991) Nitric oxide synthase and neuronal NADPH diaphorase are identical in brain and peripheral tissues. Proc. natn. Acad. Sci. U.S.AI 88, 7797-7801. 17. Desai K. M., Sessa W. C. and Vane J. R. (1991) Involvement of nitric oxide in the reflex relaxation of the stomach to accommodate food or fluid. Nature 351, 477-479. 18. Du C., Murray J., Conlin J. L. and Bates J. N. (1991) Nitric oxide: mediator of non-adrenergic non-cholinergic hyperpolarization of opossum esophageal circular smooth muscle. Gastroenterology. 100, A439. 19. Furness J. B. and Costa M. (1980) Types of nerves in the enteric nervous system. Neuroscience 5, I--20. 20. Furness J. B. and Costa M. (1987) The Enteric Nervous System. Churchill Livingstone, New York. 21. Gabella G. (1979) Innervation of the gastrointestinal tract. Int. Rev. Cytol. 59, 129-193. 22. Gaily J. A., Montague P. R., Reeke G. N. and Edelman G. M. (1990) The NO hypothesis: possible effects of a short-lived, rapidly diffusible signal in the development and function of the nervous system. Proc. natn. Acad. Sci. 87, 3547-3551. 23. Garthwaite J., Charles S. L. and Chess-Williams R. (1988) Endothelium-derived relaxing factor release on activation of NMDA receptors suggests role as intercellular messenger in the brain. Nature, Lond. 336, 385-388. 24. Hope B. T. and Vincent S. R. (1989) Histochemical characterization of neuronal NADPH-diaphorase. J. Histochem. Cytochem. 37, 653-661. 25. Hope B. T., Michael G. J., Knigge K. M. and Vincent S. R. (1991) Neuronal NADPH diaphorase is a nitric oxide synthase. Proc. natn. Acad. Sci. U.S.A. 88, 2811-2814.
NO-synthesizing neurons and vascular sites
799
26. Ignarro L. J., Buga G. M., Wood K. S., Byrns R. H. and Chaudhuri G. (1987) Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc. natn. Aead. Sci. U.S.A. 84, 9265 9269. 27. Knudsen M. A., Tottrup A. and Aarhus L. (1991) Evidence for two mediators of NANC inhibition in circular muscle of the guinea-pig ileum. Gastroenterology 100, A459. 28. Krantis A. and Clark D. (1991) Localization of [3H]GABA-labelled nerve fibre networks in the rat intestinal mucosa. J. auton, nerv. Syst. 34, 195-200. 29. Krantis A. K. and Clark D. (1991) High-affinity uptake of [3H]GABA by submucous ganglion cells, nerve fibres and peri- and para-vascular fibres in guinea-pig and rat intestine. J. auton, nerv. Syst. 32, 251 258. 30. Li C. G. and Rand M. J. (1990) Nitric oxide and vasoactive intestinal polypeptide mediate non-adrenergic, non-cholinergic inhibitory transmission to smooth muscle of the rat gastric fundus. Eur. J. Pharmac. 191, 303 309. 31. Long C. J. and Berkowitz B. A. (1989) Minireview--What is the relationship between the endothelium derived relaxant factor and nitric oxide? Life Sci. 45, 1 14. 32. Marietta M. A. (1989) Nitric oxide: biosynthesis and biological significance. Trends" biol. Sci. 14, 488 492. 33. Matusak O. and Bauer V. (1986) Effect of desensitization induced by adenosine 5'-triphosphate, substance P, bradykinin, serotonin, gamma-aminobutyric acid and endogenous non-cholinergic, non-adrenergic transmitter in the guinea-pig ileum. Eur. J. Pharmac. 126, 199 209, 34. Moncada S., Palmer R. M. J. and Higgs E. A. (1989) The biological significance of nitric oxide formation from L-arginine. Bioehem. Soc. Trans. 17, 642-4544. 35. Moncada S., Palmer M. J. and Higgs E. A. (1991~ Nitric oxide: physiology, pathophysiology and pharmacology. Pharmac. Rev. 43, 109 142. 36. Nichols K., K rantis A. and Staines W. (1991) NADPH-diaphorase activity, a marker of nitric oxide (NO)-synthesizing neurons, is localized to enteric neurons and vasculature. Physiol. Can. 21, 249. 37. Pahner R. M. J., Ferridge A. C. and Moncada S. (1987) Nitric oxide release accounts for the biological activity of endothelium-derived releasing factor. Nature 327, 524 526. 38. Pelckmans P. A., Boeckxstaens J. G., Deman, H. B., Herman A. G. and Vanmaercke Y. M. (1991) Nitric oxide as an inhibitory non-adrenergic non-cholinergic neurotransrnitter in the gastrointestinal tract. Gastroenterology 100, A480. 39. Pique J. M., Whittle B. J. R. and Esplugues J. V. (1989) the vasodilator role of endogenous nitric oxide in the rat gastric microcirculation. Eur. J. Pharmac. 174, 293 296. 40. Scherer-Singler U., Vincent S. R., Kimura H. and McGeer E. G. (1983) Demonstration of a unique population of neurons with NADPH-diaphorase histochemistry. J. Neurosci. Meth. 9, 229 234. 41. Schofield G. C. (1989) Anatomy of muscular and neural tissues in the alimentary canal. In Handbook (~l~Physiology, Section6, Vol. 1, pp. 1579 1627. Am. Physiol. Soc. 42. Stach W. (1989) A revised morphological classification of neurons in the enteric nervous system. In Nerves and the Gastrointestinal Tract (eds Singer M. V. and Goebell H.), pp. 29 44. Klewer Academic Publishers, MA, U.S.A. 43. Snyder S. H. and Bredt D. S. (1991) Nitric oxide as a neuronal messenger. Trends pharmae. Sci. 12, 125 128. 44. Toda N., Baba H. and Okamura T. (1990) Role of nitric oxide in non-adrenergic, non-cholinergic nerve-mediated relaxation in dog duodenal longitudinal muscle strips. Jap. J. Pharmac. 53, 281 284. 45. Toda N. and Okamura T. (1990) Modification by L-N-monomethylarginine (L-NMMA) of the response of nerve stimulation in isolated dog mesenteric and cerebral arteries. Jap. J. Pharmae. 52, 170 173. 46. Tottrup A., Suane D. and Forman R. (1991) Nitric oxide mediates NANC inhibition in opossum lower esophageal sphincter. Am. J. Physiol. 260, G385-G389. 47. Vallance P., Collier J. and Moncada S. (1989) Nitric oxide synthesized from L-arginine mediates endothelium-dependent dilatation in human veins. Cardiovasc. Res. 23, 1053 1057. 48. Westfall D. P., Hagoboom G. K., Colby J., O'Donnel J. P. and Fedan J. S. (1982) Direct evidence against a role of ATP as the non-adrenergic, non-cholinergic inhibitory neurotransmitter in guinea-pig taenia coli. Proc. natn. Acad. Sci. U.S.A. 79, 7041 7045. 49. Whittle B. J. R., Lopez-Belmonte J. and Rees D. D. (1989) Modulation of the vasodepressor actions of acetylcholine, bradykinin, substance P and endothelin in the rat by a specific inhibitor of nitric oxide formation. Br. J. Pharmac. 98, 646 652. (Accepted 18 June 1992)
NSC 5t/4
("