| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Article |
Toyama Institute for Cardiovascular Pharmacology Research, Azuchi-machi, Chuo-ku, Osaka, Japan (N.T.); and Division of Pharmacology, University of Antwerp, Wilrijk, Belgium (A.G.H.)
Abstract
Abstract I. Introduction II. Nitric Oxide as a Neurotransmitter A. Discovery of Nitrergic Nerves B. Mechanism of Nitric Oxide Actions C. Is the Nitrergic Neurotransmitter Free Radical Nitric Oxide or a Stable Analog of Nitric Oxide? D. Cotransmitters Responsible for Nonadrenergic Noncholinergic Inhibitory Responses E. In Vivo Studies III. Nitrergic Innervation in Various Regions A. Esophagus B. Stomach C. Duodenum, Jejunum, Ileum, and Ileocolonic Junction D. Colon, Rectum, and Internal Anal Sphincter E. Sphincter of Oddi and Gall Bladder F. Liver and Pancreas 1. Liver. 2. Pancreas. IV. Secretion and Neurogenic Nitric Oxide V. Species and Regional Differences in the Nitrergic Regulation VI. Development and Aging VII. Prejunctional Regulation VIII. Gastrointestinal Blood Flow Regulation by Nitrergic Nerves IX. Pathological Implications X. Summary
Gastrointestinal (GI) smooth muscle responses to stimulation of the nonadrenergic noncholinergic inhibitory nerves have been suggested to be mediated by polypeptides, ATP, or another unidentified neurotransmitter. The discovery of nitric-oxide (NO) synthase inhibitors greatly contributed to our understanding of mechanisms involved in these responses, leading to the novel hypothesis that NO, an inorganic, gaseous molecule, acts as an inhibitory neurotransmitter. The nerves whose transmitter function depends on the NO release are called "nitrergic", and such nerves are recognized to play major roles in the control of smooth muscle tone and motility and of fluid secretion in the GI tract. Endothelium-derived relaxing factor, discovered by Furchgott and Zawadzki, has been identified to be NO that is biosynthesized from L-arginine by the constitutive NO synthase in endothelial cells and neurons. NO as a mediator or transmitter activates soluble guanylyl cyclase and produces cyclic GMP in smooth muscle cells, resulting in relaxation of the vasculature. On the other hand, NO-induced GI smooth muscle relaxation is mediated, not only by cyclic GMP directly or indirectly via hyperpolarization, but also by cyclic GMP-independent mechanisms. Numerous cotransmitters and cross talk of autonomic efferent nerves make the neural control of GI functions complicated. However, the findingsrelated to the nitrergic innervation may provide us a new way of understanding GI tract physiology and pathophysiology and might result in the development of new therapies of GI diseases. This review article covers the discovery of nitrergic nerves, their functional roles, and pathological implications in the GI tract.
Nonadrenergic noncholinergic (NANC1) inhibitory neurotransmission has long been recognized in the gastrointestinal (GI) tract (Abrahamsson, 1986
). Although part of the induced responses to efferent nerve stimulation, such as relaxation and inhibitory junction potentials (IJP), are apparently mediated by peptides (Fahrenkrug, 1993; Shuttleworth and Keef, 1995) and ATP (Burnstock, 1972), the major mechanisms remained unclarified until 1990. In the late 1980s, three groups independently reported that endothelium-derived relaxing factor (EDRF) first discovered by Furchgott and Zawadzki (1980) is nitric oxide (NO) (Ignarro et al., 1987, 1988; Palmer et al., 1987; Furchgott, 1988). Murad et al. (1978) provided evidence that nitroglycerin exerts vasorelaxation due to the release of NO which activates soluble guanylyl cyclase to yield cyclic GMP. Moncada et al. (1989) reported that NO is formed by an enzyme, called NO synthase (NOS) that transforms L-arginine to L-citrulline in the presence of cofactors such as tetrahydrobiopterin, reduced nicotinamide-adenine dinucleotide phosphate (NADPH), and FAD/FMN. In contrast to the inducible form of the NOS (iNOS), the two constitutive NO synthases, i.e., the neuronal (nNOS) and endothelial (eNOS), are activated by Ca2+ in the presence of calmodulin (Bredt and Snyder, 1990). Palmer et al. (1988) first introduced the compound NG-monomethyl-L-arginine (L-NMMA), which inhibits the enzyme, and helped us to elucidate the physiologically important roles of NO in the whole body. This led to the discovery that the major inhibitory neurotransmitter of the efferent autonomic nervous system at the level of the anococcygeus muscle (Li and Rand, 1989), retractor penis (Gillespie and Sheng, 1990), artery (Toda and Okamura, 1990), gut (Bult et al., 1990; Toda et al., 1990a), and trachea (Tucker et al., 1990) is NO. In the field of GI tract research, detailed analyses established the role of neurons in which NO acts as an inhibitory neurotransmitter. These nerves, called "nitrergic" (Rand, 1992a) or "nitroxidergic" (Toda and Okamura, 1992), receive electrical information from the central nervous system via parasympathetic preganglionic fibers and ganglia. This review article describes the history of the discovery of nitrergic nerves in the GI tract, the nitrergic innervation in various regions of the GI tract, and the interaction with other autonomic nerves, the blood flow regulation of the GI tract by these nerves, as well as their pathophysiological implications.
II. Nitric Oxide as a Neurotransmitter
A. Discovery of Nitrergic Nerves
Bult et al. (1990), Toda et al. (1990a), and Li and Rand (1990) for the first time provided evidence that NO acts as an inhibitory NANC neurotransmitter in the isolated canine ileocolonic junction, canine duodenal longitudinal muscle, and rat gastric fundus, respectively. According to Bult et al. (1990), superfused ileocolonic preparations liberate vasorelaxant substance(s) in response to transmural electrical stimulation or dimethylphenylpiperazinium (DMPP). The induced relaxation was abolished by treatment of the tissue with the NOS inhibitor NG-nitro-L-arginine (L-NA), superoxide anions, or oxyhemoglobin (oxyHb), and the effect of L-NA was reversed by L-arginine. Toda et al. (1990a) and Li and Rand (1990) noted that relaxations induced by electrical field stimulation of duodenal and gastric fundus strips were abolished by L-NMMA, but not by D-NMMA, and the addition of L-arginine, but not D-arginine, restored the responses. Release of the neurotransmitter NO from the myenteric plexus by electrical stimulation has been visualized with NO/luminal-induced chemiluminescence in the guinea pig ileum (Wiklund et al., 1996, 1997). In addition, evidence supporting the idea that nNOS activation and NO formation from L-arginine are involved in the relaxation induced by electrical nerve stimulation in the rat gastric fundus has been obtained by measuring L-citrulline in the incubation medium (Curro et al., 1996). Direct evidence for the release of NO from neurons has been obtained by detecting the major metabolites of the NO-citrulline pathway with capillary electrophoresis at the single cell level (Moroz et al., 1999).
NOS immunoreactive and NADPH diaphorase-positive nerve fibers have histologically been shown to be present in the myenteric plexus of the rat intestine (Bredt et al., 1990) and the myenteric and submucous plexuses of the GI tract from other mammals (Barbiers et al., 1993; Nichols et al., 1995; Rodrigo et al., 1998). Vittoria et al. (2000) studied the distribution of nitrergic neurons in the bovine gut from the esophagus to the rectum and noted the NOS immunoreactivity and NADPH diaphorase to be constantly colocalized in the same neuron.
Burns et al. (1996), Ward et al. (1998), and Ward and Sanders (2001) revealed that nitrergic inhibitory responses are abolished in mutant rats that lack c-Kit-positive interstitial cells of Cajal in the esophagus, stomach, and pyloric sphincter; thus, the interstitial cells are expected to play quite an important role in nitrergic neurotransmission, possibly by transducing electrical and mechanical NO signals. Immunohistochemical distribution of c-Kit-positive cells and NOS-positive neurons has also been demonstrated in the human gut (Nemeth and Puri, 2001) and esophageal sphincter (Watanabe et al., 2002).
Review articles by Sanders and Ward (1992), Rand (1992b), Lefebvre (1993, 2000), Boeckxstaens and Pelckmans (1997), and Curro and Preziosi (1998) summarize the evidence supporting the hypothesis that NO or its analog acts as an inhibitory neurotransmitter in the GI tract. However, Correia et al. (2000) noted that among the products that can be formed during NOS catalyzed L-arginine NG-oxidation, only NH2OH caused relaxations that exhibited a pharmacological profile similar to that induced by nitrergic nerve stimulation in the rat duodenum, and these observations suggested that this molecule may be an actual neurotransmitter acting directly or by a release of NO.
NO is synthesized from L-arginine to yield citrulline by the constitutive nNOS (Forstermann et al., 1994). Neuronal NOS in the brain (Bredt and Snyder, 1990) and iNOS (Hevel et al., 1991; Stuehr et al., 1991) are mostly soluble enzymes, whereas eNOS is more than 90% particulate (Forstermann et al., 1991). The nNOS-containing subcellular structure has not morphologically been confirmed. A recent study by Van Geldre et al. (2004) demonstrated that using a protocol involving strong homogenization of the rat small intestine, about 50% of the immune reactive nNOS, was particle-bound in a subcellular structure.
Ca2+ together with calmodulin is required for nNOS activation (Bredt and Snyder, 1990). In the canine ileocolonic junction, 
-conotoxin GVIA, an N-type Ca2+ channel inhibitor (McCleskey et al., 1987; Miller, 1987), or exposure of the tissue to Ca2+-free media reduced the inhibitory response to nitrergic nerve stimulation, whereas L-type Ca2+ channel inhibitors nifedipine and verapamil were without effect (Boeckxstaens et al., 1993a). NO induced relaxation of rectal circular muscle with a concomitant decrease in intracellular Ca2+ level (Takeuchi et al., 1998a). Therefore, electrically generated nerve action potentials trigger the influx of Ca2+ through N-type channels that appear to be prerequisite for the nNOS activation. Evidence for the capacity of recycling citrulline to arginine has been reported in the proximal colon (Shuttleworth et al., 1995) and internal anal sphincter smooth muscle (Rattan and Chakder, 1997). The presence of enzymes, such as argininosuccinate synthetase and argininosuccinate lyase (Hecker et al., 1990), capable of transforming citrulline to arginine, has immunohistochemically been established (Shuttleworth et al., 1995). This seems to be one of the mechanisms of maintaining the function of nitrergic neurons in the GI tract.
Nicotine (Toda et al., 1992; Kojima et al., 1993; McLaren et al., 1993; Mozhorkova et al., 1994), DMPP (Boeckxstaens et al., 1993a; Takeuchi et al., 1996), K+ (Toda et al., 1992), 5-HT (Boeckxstaens et al., 1990b; Bogers et al., 1991), ATP, and GABA (Boeckxstaens et al., 1991b) reportedly stimulate nitrergic nerves and release NO or a NO analog in the GI tract. Nerve action potentials appear to contribute to the inhibitory response to these chemical stimuli because the induced relaxation is susceptible to tetrodotoxin. The actions of nicotine and DMPP were diminished by hexamethonium. The Ca2+ channels involved in the Ca2+ influx for nNOS activation in response to DMPP were suggested to be of a non-L, non-N-type, which was in contrast to the N-type channel involved in the response to electrical nerve stimulation (Boeckxstaens et al., 1993a). A similar finding as to the Ca2+ channel types responsible for the action of nicotine was also obtained in the cerebral artery, although in the arteries, the action of nicotine was resistant to tetrodotoxin, which abolished the response mediated by nerve action potentials (Toda et al., 1995). The M1-muscarinic receptor activation by McN-A-343 also produced NO-mediated relaxation in the intestine (Olgart and Iversen, 1999). The response was abolished by L-NA, L-NMMA, ODQ (1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one), a soluble guanylyl cyclase inhibitor (Garthwaite et al., 1995), and tetrodotoxin, whereas guanethidine or hexamethonium did not affect the response. Susceptibility to tetrodotoxin is consistent with the view that ganglionic M1 receptors are stimulated by the agonist to generate nitrergic nerve action potentials.
B. Mechanism of Nitric Oxide Actions
EDRF is identical to NO in blood vessels (Ignarro et al., 1987; Palmer et al., 1987; Furchgott, 1988), and NO exerts its actions via activation of soluble guanylyl cyclase in smooth muscle cells that produce cyclic GMP as an intracellular messenger (Waldman and Murad, 1987). These hypotheses have widely been accepted in vascular smooth muscle. However, in GI smooth muscles, the NO-induced relaxation is mediated by the guanylyl cyclase-cyclic GMP pathway and also by membrane hyperpolarization. In GI smooth muscles from animals (Desai et al., 1991; McLaren et al., 1993; Olgart and Iversen, 1999) or humans (Bartho et al., 2002), involvement of cyclic GMP in NO-induced relaxation has been derived from studies using the soluble guanylyl cyclase inhibitors methylene blue and ODQ or the cyclic GMP phosphodiesterase inhibitor zaprinast (Ward et al., 1992) or measuring directly the cyclic GMP contents (Toda et al., 1992). Ward et al. (1992) suggested that inhibitory junction potentials mediated by neurogenic NO are presumably coupled with the enhanced production of cyclic GMP. Ny et al. (2000) noted that electrical field stimulation caused a biphasic relaxation, an initial rapid one followed by a slowly developing phase, in stomach fundus strips from wild-type mice, whereas only a slow relaxation was observed in the strips from mice lacking cyclic GMP-dependent protein kinase I. These relaxations were mediated by NO and a non-NO inhibitory molecule. The responses to 3-morpholino-sydnoninime, a NO donor, and 8-bromo-cyclic GMP were markedly impaired in the strips from the knockout mice. It was suggested that cyclic GMP-dependent protein kinase I plays a central role in the NO/cyclic GMP signaling cascade. In contrast, the relaxation induced by neurogenic NO was not inhibited by treatment with methylene blue (Allescher et al., 1993; Takakura and Muramatsu, 1999). Cyclic GMP-independent, apamin-sensitive relaxation induced by NO was also reported by Martins et al. (1995) and Borjesson et al. (1997).
NO and NO donors enhance the open state of Ca2+-activated K+ channels and mimic hyperpolarization by NANC inhibitory nerve stimulation (Thornbury et al., 1991). The apamin-sensitive and -insensitive IJP is abolished by L-NAME, suggesting that nerve-derived NO opens Ca2+-activated K+ channels to induce hyperpolarization (Christinck et al., 1991). Transmural field stimulation at high frequencies elicited fast and slow phases of relaxation and hyperpolarization, and L-NA depressed the slow phase and partially reduced the fast phase, suggesting that the NO-induced hyperpolarization is involved in the relaxation (Shimamura et al., 1993; Suzuki et al., 2003). Cl- channels may also be involved in the NO-evoked IJP (Suzuki et al., 2003). From studies on wild mice and mutant mice that lacked the inositol triphosphate type 1 receptor, Suzuki et al. (2000) noted that the lack of this receptor attenuated IJP evoked by nitrergic nerve stimulation but did not alter those associated with norepinephrine. It was suggested that basal release of NO caused an oscillatory pattern of electrical and mechanical activity. Suthamnatpong et al. (1994) suggested that IJP is not involved in relaxation in the rat proximal colon. On the other hand, there are a number of reports showing that NO and other transmitters are independently involved in the responses to nitrergic nerve stimulation. In the lower esophageal sphincter (LES), relaxations induced by NANC inhibitory nerve stimulation were reduced by NOS inhibitors and oxyHb, and the remaining response was abolished by apamin (Boyle et al., 1991). In the human colon, NO- and ATP-induced relaxations mimicked relaxations evoked by NANC neurogenic stimulation, and the L-NA-resistant response was abolished by apamin, suggesting that NO and possibly ATP are involved in the inhibitory neurotransmission (Boeckxstaens et al., 1993c). Zagorodnyuk and Maggi (1994) have suggested that apamin-sensitive and apamin-resistant components of the evoked IJP utilize different NANC transmitters, i.e., ATP and NO, respectively. However, according to Imaeda et al. (1998), both the nitrergic and adrenergic (ATP-mediated) inhibitory nerves contribute equally to the IJP generation via opening of Ca2+-activated K+ channels, and the relaxation is mainly produced by NO in a membrane potential-independent manner. Okamura et al. (1998b) have reported that L-NA-resistant relaxation induced by NANC inhibitory nerve stimulation in the canine colon is suppressed by ouabain and lowering of external K+, suggesting the involvement of the electrogenic Na+ pump. In this preparation, K+ channel inhibitors, such as apamin, charybdotoxin, tetraethylammonium, and glibenclamide, are without effect. The relaxing factor responsible for the Na+ pump activation has not yet been identified.
Possible mechanisms involved in NO-induced relaxation and hyperpolarization are 1) cyclic GMP-dependent reduction of cellular free Ca2+ without changing the membrane potential, 2) cyclic GMP-dependent opening of apamin-sensitive K+ channels or other types of ion channels to produce hyperpolarization and relaxation, and 3) cyclic GMP-independent mechanisms, such as actions of NO on ion channels involved in muscle contractility, either directly or via membrane hyperpolarization. Figure 1 summarizes the mechanisms of action of neurogenic NO in the GI tract. Proposed mechanisms of action of nitrergic nerve stimulation differ in animal species and regions of the GI tract studied. Despite the evidence for a cyclic GMP-independent mechanism in some rodents, cyclic GMP seems to be a key substance for nitrergic inhibitory responses in most mammalian GI tracts, including that of humans.
|
This question has been raised soon after EDRF was identified to be NO (Ignarro et al., 1987; Palmer et al., 1987; Furchgott, 1988). The reason for the question was the observation that responses to EDRF or nitrergic nerve stimulation in some smooth muscle preparations, including the GI tract, were not reduced by superoxide anion generators or a NO scavenger (La et al., 1996) in concentrations sufficient to suppress the effect of exogenous NO or NO donors. In addition, the neurogenic response was not augmented by superoxide dismutase. Therefore, stable analogs of NO like S-nitrosothiols or unidentified substance(s) were postulated to be the NANC neurotransmitters in the canine colon (Thornbury et al., 1991), rat gastric fundus (Jenkinson et al., 1995), and rat duodenum (Correia et al., 1999). Using the bioassay method, it was reported that NOS products, considered to be NO or S-nitrosothiols liberated from neurons in the donor tissue, relaxed the assay tissue and that this response was potentiated by superoxide dismutase (Boeckxstaens et al., 1991a,c), suggesting that the released substance was susceptible to superoxide anions. If the substance had been released in a stable form, superoxide anions would not have destroyed its activity. Furthermore, the fluorescent method using DAF-2 (4,5-diaminofluorescein/diacetate) and 4-amino-5-methylamino-2',7'-difluorofluorescein for detection of NO demonstrated the release of NO in rat brain slices (Kojima et al., 1998, 2001). There is no visible fluorescence of these indicators interacting with S-nitrosothiols (T. Nagano, personal communication), suggesting that DAF only binds to NO. If the thiol compounds were stable enough to be protected from degradation in extracellular fluids, the data obtained from the bioassay method and with DAF would not support the hypothesis that the neurotransmitter is a stable analog of NO.
Evidence against the idea that S-nitrosothiols act as intermediate compounds has been presented elsewhere (De Man et al., 1995b, 1998, 1999), and many other investigators, who sought a way to solve this problem, were also led to the conclusion that free radical NO most likely acts as the nitrergic neurotransmitter by excluding the possible involvement of S-nitrosothiols (Boeckxstaens et al., 1994; De Man et al., 1996a,b; Lefebvre, 1996; Colpaert and Lefebvre, 2000). Immunohistochemical analysis of porcine gastric fundus revealed a 100% colocalization of Cu2+/Zn2+ superoxide dismutase and nNOS in the intrinsic neurons of the myenteric plexus (Colpaert et al., 2002). Therefore, it is hypothesized that there are endogenous antioxidant defense mechanisms, enzymatic and nonenzymatic, for the neurotransmission process to be mediated by free NO. Similar conclusions were also obtained from studies using an irreversible inhibitor of superoxide dismutase (Martin et al., 1994; Lilley and Gibson, 1995; Paisley and Martin, 1996; Okamura et al., 1998a; Tanaka et al., 1999).
Goyal and He (1998) noted that nitrergic slow IJP and hyperpolarization evoked by the NO· redox donor, diethylenetriamine/NO adduct, were suppressed by guanylyl cyclase inhibitors but not by apamin in the guinea pig ileal circular muscle, whereas hyperpolarization caused by sodium nitroprusside and solubilized NO gas was antagonized by apamin but not by guanylyl cyclase inhibitors. Thus, they speculated that NO· redox is the nitrergic neurotransmitter.
D. Cotransmitters Responsible for Nonadrenergic Noncholinergic Inhibitory Responses
Coexistence of choline acetyltransferase/cholinesterase with nNOS, VIP, and neuropeptide Y (NPY) in the neurons of the myenteric plexus (Fang et al., 1993; Ekblad et al., 1994; Barbiers et al., 1995) would indicate that nerve fibers containing these molecules are of vagal origin. It has been demonstrated that nitrergic as well as cholinergic nerves innervating the cerebral and ophthalmic arteries arise from the superior salivatory nucleus in the brain stem sending off parasympathetic efferent neurons (Spencer et al., 1990) via the greater petrosal nerve and pterygopalatine ganglion (Ayajiki et al., 2000; Toda et al., 2000a,b; Toda and Okamura, 2003). A histological study has demonstrated that vagal preganglionic efferents terminate on NOS-containing neurons in the rat gastric fundus (Berthoud, 1995).
Li and Rand (1990) have reported that NANC-mediated neurogenic relaxation of the rat gastric fundus is reduced and VIP-induced relaxation is abolished by treatment with a VIP antibody. The residual relaxation in the presence of VIP antibody is further reduced by L-NMMA. Therefore, these authors suggested that NO as well as VIP is involved in the NANC-mediated relaxation. There is also evidence for a differential release of NO and VIP by NANC nerves in the rat gastric fundus (Boeckxstaens et al., 1992; D'Amato et al., 1992). In studies on isolated feline and porcine gastric fundus, Barbier and Lefebvre (1993) and Lefebvre et al. (1995) have concluded that short-lasting and sustained relaxations induced by electrical nerve stimulation are mediated by NO and that a peptide, possibly VIP, is involved during sustained, low-frequency stimulation. NO-independent inhibitory responses to NANC nerve stimulation were reduced by 
-chymotrypsin in the canine proximal colon (Keef et al., 1994). Jin et al. (1996) and Murthy et al. (1996) suggested the dual origin of NO from nerves and muscle and its interplay with VIP in regulating relaxation of the rabbit and rat stomach. Recent studies by Dick and Lefebvre (2000) and Dick et al. (2001, 2002) on pig, guinea pig, and mouse gastric fundus smooth muscle cells have provided evidence that inducible NOS probably induced by the isolation procedure may be involved in the relaxant effect of VIP, thereby underscoring the importance of the experimental method used. From findings that nicotine-induced NANC relaxation in the gastric fundus was reduced by a specific anti-VIP serum and that nicotine increased the outflow of VIP- and histidine isoleucine-like immunoreactivity from the strips, involvement of VIP and possibly histidine isoleucine in the mechanical response was suggested (Curro and Preziosi, 1997). According to Bayguinov et al. (1999), the NANC inhibitory neurotransmitters in the gastric fundus are NO and VIP mediating responses via cyclic GMP and cyclic AMP, and there is no evidence supporting a serial cascade in which VIP is coupled to NO-dependent responses. In the proximal colon, the VIP-induced inhibitory response was reportedly induced by stimulation at the high frequency of 20 Hz only when nitrergic and purinergic NANC responses were depressed (Matsuyama et al., 2003). Histochemical studies have demonstrated the colocalization of NOS immunoreactivity or NADPH-diaphorase and VIP-like immunoreactivity in myenteric plexus and enteric neurons (Keef et al., 1994; Barbiers et al., 1995; Lefebvre et al., 1995; Guo et al., 1997).
Contribution of NO and ATP to NANC inhibitory neurotransmission has been noted in the human colon (Boeckxstaens et al., 1993c), rat pyloric sphincter (Soediono and Burnstock, 1994), rat ileum (Smits and Lefebvre, 1996a), and mouse jejunum (De Man et al., 2003). Soediono and Burnstock (1994) have concluded that ATP seems to act through P2Y-purinoceptors in the rat pyloric sphincter. Studies on the rat pylorus by Ishiguchi et al. (2000) have also shown that NO and ATP act in concert to mediate NANC relaxation. However, these authors suggested that P2X purinoceptors are involved in the ATP-induced relaxation. Recently, De Man et al. (2003) suggested the major involvement of P2Y receptors of the P2Y1 subtype located postjunctionally and P2X receptors located pre- and/or postjunctionally in the mouse jejunum. Coexistence of ATP and NO was reported in myenteric neurons in the rat ileum (Belai and Burnstock, 1994). NOS-related activity and NPY immunoreactivity are extensively colocalized in layers of the human (Nichols et al., 1994) and porcine gut wall (Wu et al., 2003). NOS-positive neurons of the myenteric plexus display Y1 NPY receptors in the human colon (Peaire et al., 1997), suggesting possible modulation of nitrergic nerve function by NPY receptors. It has been reported that NANC inhibitory responses to nerve stimulation are mediated by NO and by neurotensin via apamin-sensitive K+ channels in the rat ileum (Yamaji et al., 2002).
Pituitary adenylyl cyclase-activating peptide (PACAP) is postulated to be an inhibitory neurotransmitter in the rat colon (Grider et al., 1994; Kishi et al., 1996) and guinea pig taenia coli (Jin et al., 1994). McConalogue et al. (1995) noted that in the guinea pig taenia caecum, nitrergic IJP were reduced in tissues desensitized to PACAP and that the IJP and hyperpolarization induced by exogenous PACAP were blocked by apamin. Cocaine- and amphetamine-regulated transcript (CART) peptide, originally isolated from the rat striatum (Douglass et al., 1995), is also expressed in the peripheral nervous system (Dun et al., 2000) and enteric neurons (Couceyro et al., 1998). Ekblad et al. (2003) found that CART peptide is present in numerous myenteric neurons throughout the rat GI tract, and the addition of CART attenuated the NO donor-induced relaxations in rat colonic strips, suggesting CART may exert a modulatory action on NO neurotransmission. Kadowaki et al. (1999) demonstrated that about 30% of the total NOS-immunoreactive neurons had contact with 5-HT-positive nerve fibers in the guinea pig distal colon and suggested that NO may participate in the 5-HT-evoked Cl- secretion.
Nitrergic inhibitory responses of the GI tract were also observed in in vivo studies. Electrical stimulation of the peripherally cut end of the right vagus nerve resulted in relaxations of the opossum LES and caused peristaltic and nonperistaltic contractions in the esophageal body (Tottrup et al., 1991b). L-NA abolished the sphincter relaxation evoked by nerve stimulation, this effect being fully reversed by L-arginine infusion. The L-arginine-NO pathway seems to play an important role in the relaxation, but not for esophageal body peristalsis. In conscious dogs, L-NAME delayed gastric emptying of solid meals, and L-arginine given with L-NAME shortened the delay of gastric emptying. This delay was due to stimulation of pyloric and proximal duodenal contractions (Orihata and Sarna, 1994). They speculated that NO facilitated gastric emptying by partially inhibiting pyloric and proximal duodenal contractions. Pyloric sphincter relaxations may also be involved. De Winter et al. (2002) showed that L-NAME delayed the gastric emptying of a semiliquid meal in mice without any effect on the intestinal transit. In conscious dogs with enterically isolated ileocolonic loops, L-NA increased the phasic pressure in ileum and ileocolonic sphincter and the sphincter tone, but did not abolish colonic relaxation during ileal distension, suggesting that non-nitrergic nerve pathways mediate the reflex colonic relaxation (Leelakusolvong et al., 2002). In the anesthetized rat, L-NAME increased jejunal intraluminal pressure and initiated phasic intestinal contractions. These responses were inhibited by concurrent administration of L-arginine, but not D-arginine (Calignano et al., 1992). Endogenous NO is suggested to play a role in the modulation of intestinal motility. From findings obtained in anesthetized rats that an initial increase in intragastric pressure by vagal nerve stimulation was enhanced while the induced decrease in intragastric pressure was abolished by L-NAME, Lefebvre et al. (1991) suggested that NO is involved in gastric relaxation. In anesthetized dogs, pyloric activity induced by duodenal field stimulation was inhibited by antral field stimulation and electrical vagal stimulation, and intra-arterial L-NAME reduced the inhibition. It has been reported that the functional significance of NO release from NANC nerves in the isolated guinea pig stomach is to bring about adaptive relaxation through a reflex response to increases in gastric pressure (Desai et al., 1991). In conscious rats exposed to cold at 4°C for 2 to 3 h, nitrergic and cholinergic neurons were activated in the gastric myenteric ganglia through vagal nicotinic pathways (Yuan et al., 2001).
III. Nitrergic Innervation in Various Regions
It has been reported that NANC inhibitory responses are mainly mediated by nitrergic nerves in esophagus from the opossum (Christinck et al., 1991; Murray et al., 1991), cat (Ny et al., 1995), cow (Barahona et al., 1998), and human (Richards et al., 1995) and also in the LES from the dog (De Man et al., 1991; Daniel et al., 2002), opossum (Tottrup et al., 1991a,b; Conklin et al., 1993; Uc et al., 1999), guinea pig (Imaeda et al., 1998), cat (Kortezova et al., 1996), mouse (Kim et al., 1999; Sivarao et al., 2001), and human (McKirdy et al., 1992; Preiksaitis et al., 1994; Tomita et al., 1997). Biphasic, rapid and sustained, relaxations of the feline LES are suggested to be mediated by NO and VIP or VIP-like peptides (Kortezova et al., 1996), and those of the opossum sphincter appear to be mediated by NO and an unidentified substance (Uc et al., 1999). In LES strips from humans, inhibitory motorneurons were efficiently stimulated both by electrical field stimulation and stimulation of nicotinic receptors located in nerve terminals, releasing NO and an apamin-sensitive neurotransmitter (Gonzalez et al., 2004). In studies on the LES from wild-type and genetically engineered eNOS and nNOS mice, Kim et al. (1999) noted that relaxations to electrical nerve stimulation were abolished by L-NA in wild-type mice, and the responses were not affected by a lack of the eNOS gene, but were absent in nNOS knockout mice. There was no difference in the sensitivity to a NO donor in these three groups. Thus, the authors concluded that nNOS rather than eNOS is the enzymatic source of NO. Cyclic GMP is involved in the NO-mediated relaxation in canine (Daniel et al., 2002) and bovine LES (Barahona et al., 1999). Histological studies have demonstrated the presence of neurons containing nNOS or NADPH diaphorase and costaining for VIP and galanin in the human (Singaram et al., 1994), cat (Ny et al., 1995), monkey (Rodrigo et al., 1998), guinea pig (Morikawa and Komuro, 1998), bovine (Vittoria et al., 2000), and porcine esophagus (Wu et al., 2003). Nitrergic innervation was also observed in striated muscles of the guinea pig (Morikawa and Komuro, 1998) and rat esophagus (Neuhuber et al., 1994; Kuramoto et al., 1999).
In the anesthetized opossum and cat, L-NAME or L-NA inhibited LES relaxation induced by vagal stimulation, swallowing, and balloon distension (Paterson et al., 1992) and decreased the amplitude of peristaltic contraction in the very distal esophagus (Xue et al., 1996), suggesting that NO is an important mediator for the timing of peristalsis in the distal esophagus for LES relaxation and for swallowing. In cats under ketamine sedation, it was suggested that tonic contractile activity in the esophageal body is mainly caused by a continuous cholinergic excitatory input, and a NO inhibitory mechanism may have a complementary role in the tonic regulation (Zhang et al., 2004). According to Beyak et al. (2000), intravenous injections of L-NA reduced the number of oropharyngeal swallows and the induction of primary peristalsis in the smooth muscle portion but not in the striated muscle portion, whereas administration of L-NA into the fourth ventricle in the brain reduced the swallows and peristalsis in both smooth and striated muscles. Therefore, NO released in the central nervous system appears to be involved in esophageal peristalsis and swallowing, and the neural substrates mediating striated and smooth muscle peristalsis may be anatomically and neurochemically distinct. In humans, intravenous infusions of L-NMMA resulted in a reduction of the latency period between swallows and the onset of contractions in the distal esophagus (Konturek et al., 1997b) or increased the amplitude of peristaltic pressure waves in the distal esophagus and inhibited the increase in LES relaxation during gastric distension (Hirsch et al., 1998). Endogenous NO is likely to be involved in the timing of esophageal peristalsis and sphincter relaxation in humans.
On the basis of histological studies, diminished nitrergic inhibitory neurotransmission to the LES (Mearin et al., 1993) or to interstitial cells of Cajal in the LES was speculated to be the pathophysiological mechanism of esophageal achalasia (Watanabe et al., 2002). Sivarao et al. (2001) noted that mice with nNOS gene disruption had LES hypertension with impaired relaxation resembling achalasia. On the other hand, wild-type mice lacking intramuscular interstitial cells of Cajal had a hypotensive sphincter with unimpaired relaxation, suggesting that the interstitial cells of Cajal do not play a role in nitrergic neurotransmission. NADPH diaphorase histochemistry showed a reduction of myenteric nitrergic neurons and fibers in the circular muscle of congenital esophageal stenosis, but the peptidergic neurons (VIP, substance P, and galanin) were unaltered (Singaram et al., 1995). The specific lack of NO inhibitory innervation may be an important mechanism for the pathogenesis of stenosis and aperistalsis of the esophagus in this disorder. In contrast, nitrergic relaxation of LES obtained from patients with reflux esophagitis was increased compared with that in normal sphincter specimens (Tomita et al., 2003). Portal hypertension activates NOS genes in the esophageal mucosa in rats, and this phenomenon may facilitate the development and rupture of esophageal varices (Tanoue et al., 1996).
Cardiac antiarrythmic therapy in patients with mexiletine, a class Ib antiarrythmic drug, is accompanied by a high incidence of gastrointestinal side effects (Campbell, 1987). Mexiletine attenuated NANC relaxation of the rabbit LES, possibly by inhibiting myenteric nitrergic neurotransmission (Kohjitani et al., 2003b). The intravenous anesthetics, ketamine and midazolam, inhibited the K+-induced neurogenic relaxation of the rabbit LES (Kohjitani et al., 2001). The authors suggest that ketamine inhibits the response by the extracellular production of superoxide anions and that midazolam inhibits it by inhibiting NOS activity (Kohjitani et al., 2003a).
Intragastric infusion of sildenafil, a phosphodiesterase-5 inhibitor, induced a decrease in LES tone and resting pressure and in the amplitude and propagation velocity of esophageal contractions in normal subjects (Bortolotti et al., 2001a; Rhee et al., 2001). Reduced degradation of intracellular messenger cyclic GMP may participate in the enhanced nitrergic inhibition. Intravenous administration of sildenafil also reduced the amplitude of esophageal contractions and LES pressure in anesthetized cats (Zhang et al., 2001). The authors suggested that the effect may benefit patients with achalasia. In patients with idiopathic achalasia (Bortolotti et al., 2000) and symptomatic hypertensive LES (Bortolotti et al., 2002), sildenafil infused into the stomach in a double-blind manner decreased the LES tone, residual pressure, and wave amplitude. In an open study on 11 patients with nutcracker esophagus, hypertensive LES, or achalasia and in a randomized double-blind study on six volunteers, sildenafil lowered LES pressure and propulsive forces in the esophageal body in both the healthy subjects and patients (Eherer et al., 2002). A subset of patients with hypertensive LES or nutcracker esophagus may benefit from sildenafil, but the side effects might be a limiting factor.
|
Neurogenic relaxation and IJP were reportedly mediated by the neurotransmitter NO in the gastric fundus and pyloric sphincters from the rat (Boeckxstaens et al., 1991a; McLaren et al., 1993; Jenkinson et al., 1995; Curro et al., 1996; Curro and Preziosi, 1998; De Man et al., 1998; Lefebvre, 1998), guinea pig (Kojima et al., 1993), dog (Bayguinov et al., 1999), pig (Colpaert and Lefebvre, 2000), mouse (Selemidis and Cocks, 2000; Suzuki et al., 2000; Ergun and Ogulener, 2001), Japanese monkey (N. Toda, K. Ayajiki, and T. Okamura, unpublished data), and human (Tomita et al., 1999; Tonini et al., 2000). Typical responses to nitrergic nerve stimulation of the isolated monkey pyloric sphincter are shown in Fig. 2. L-NA abolished the relaxation that was reversed by L-arginine, but not D-arginine, suggesting that the response is associated exclusively with nerve-derived NO. It was reported that the ability of the nitrergic neurotransmitter to induce relaxation of the rat gastric fundus was influenced by the mechanism used to induce tone, and sarcoplasmic/endoplasmic reticulum Ca2+ ATPase appeared to play a role in nitrergic relaxation (Van Geldre and Lefebvre, 2004). From studies using wild-type and M3 muscarinic receptor knockout mice, Stengel and Cohen (2003) obtained evidence supporting the presence of M1 receptor-mediated relaxation in the stomach and suggested that when the M3 receptor responsible for contraction was eliminated or blocked, M1 receptor-mediated nitrergic relaxation was enhanced, possibly leading to alterations in gastric emptying. Melatonin had no effect on the basal tonus of the isolated rat gastric fundus but reduced the inhibitory response to nitrergic nerve stimulation (Storr et al., 2002b). Together with findings obtained from the electrophysiological study on mouse colonic preparations and studies on NOS activity in enteric synaptosomes and melatonin receptor transcripts in rat GI tracts, these authors concluded that a direct inhibition of NOS activity might be involved in the reduction by melatonin of inhibitory neurogenic responses.
Histochemical studies of the human stomach (Belai and Burnstock, 2000; Smith et al., 2001) and electron-microscopic analyses of the cat pylorus (Feher et al., 2001) showed the localization of neurons containing nNOS or NADPH diaphorase. Site-specific gene expression of nNOS variants, including nNOS-a,-b,-c and nNOS
, was revealed in distinct functional regions (e.g., gastric fundus versus pylorus) of human (Saur et al., 2000) and rat gastrointestinal tract (Saur et al., 2002). They suggested that the different expression of these proteins is implicated in different biological functions. Miller et al. (2001) demonstrated colocalization of NOS-immunoreactive cell bodies of myenteric ganglia and those containing heme oxygenase-2 in the human antrum and jejunum, and they postulated a possible role for CO as a neurotransmitter and a possible interaction between heme oxygenase and NOS pathways in inhibitory neurotransmission.
A cyclic GMP-dependent mechanism appears to be involved in the nitrergic relaxation of the rat and mouse gastric fundus (Williams and Parsons, 1995; Chang et al., 1998; Bayguinov et al., 1999; Selemidis and Cocks, 2000). Cyclic GMP-dependent protein kinase I apparently plays a key role in the NO/cyclic GMP signaling pathway in the mouse gastric fundus (Ny et al., 2000). However, okadaic acid, a protein phosphatase inhibitor, also interfered with the NO-mediated inhibitory neurotransmission (Storr et al., 2002a), suggesting that attenuated dephosphorylation of proteins impairs the inhibitory process. Whether this is due to antagonism of the nitrergic relaxation or enhanced contraction by phosphorylated myosin light chain in association with phosphatase inhibition remains to be determined.
NO released from NANC nerves in the guinea pig stomach appears to be involved in adaptive relaxation through a reflex response to increases in intragastric pressure (Desai et al., 1991). It has been reported that metabolites of L-arginine mediate neuronal inhibition of canine pyloric motor activity in vivo (Allescher et al., 1992) and facilitate gastric emptying by inhibiting pyloric contractions (Orihata and Sarna, 1994). Gastric distension-induced pyloric relaxation appears to be mediated via a vago-vagal reflex and NO release in the rat (Ishiguchi et al., 2001). The primary site of action of nitrergic mechanisms on the gastric fundic tone in conscious dogs is likely to be at a presynaptic level on the vagal cholinergic efferent nerves (Paterson et al., 2000). In healthy volunteers, it is suggested that NO is involved in maintaining the basal fundic tone and in the meal-induced fundic relaxation (Kuiken et al., 2002b), and the gastric accommodation reflex involves activation of nitrergic neurons (Tack et al., 2002).
In the stomach of ovariectomized female rats, NOS activity and mucus secretion were increased, and the severity of ulcer lesions was decreased compared with controls, suggesting that estrogen deficiency protects the gastric mucosa by NO-mediated mucus hypersecretion (Morschl et al., 2000). In contrast, Shah et al. (2000, 2001) noted that nitrergic relaxation of isolated gastric fundus and nNOS protein were increased in rats treated with estrogen, but not by progesterone, or in pregnant rats. The nNOS protein was also increased in ovariectomized rats treated with estrogen when compared with ovariectomized rats receiving vehicle. Results from experiments with L-NAME indicated the involvement of neurogenic NO in the estrogen-induced and pregnancy-related changes. Therefore, the possibility is suggested that estrogen, rather than progesterone, may be responsible for the delay in gastric emptying and increase in colonic transit time observed during pregnancy that is mediated by nitrergic inhibitory neurotransmission.
nNOS-deficient knockout mice showed gastric dilation or enlargement (Huang et al., 1993) and delayed gastric emptying of solids and liquids compared with wild-type and eNOS-deficient mice (Mashimo et al., 2000). Wild-type mice treated with L-NA showed delay in emptying of solids but not liquids. Thus, these authors concluded that chronic depletion of NO from nNOS resulted in delayed gastric emptying. Similar defects in gastric emptying were also observed in genetic and streptozotocin-elicited models of diabetes in mice; they also showed impaired nitrergic inhibitory responses of the pyloric muscle and reduced nNOS protein and mRNA in the pylorus (Watkins et al., 2000). nNOS expression and pyloric function were restored to normal levels by insulin treatment, and delayed gastric emptying was reversed with sildenafil. These findings implicate novel therapeutic approaches to diabetic gastropathy. Cellek et al. (2003) demonstrated that NOS-containing neurons innervating the gastric pylorus of streptozotocin-induced diabetic rats lost some of nNOS content and function in the first phase and led to complete loss of nitrergic function in the second phase; the changes in the first phase were reversible with insulin therapy, but the neurodegeneration in the second phase was irreversible. In normal subjects, antral and duodenal wave frequency and amplitude were lower during and after sildenafil administration in the gut than during and after placebo (Bortolotti et al., 2001b). NO-mediated brisk relaxant responses were impaired in the gastric fundus of mdx dystrophic mice, an animal model of Duchenne muscular dystrophy compared with normal mice (Baccari et al., 2000). Quintana et al. (2004) demonstrated synthesis of NO in postganglionic myenteric neurons during early endotoxemia that seemed to mediate gastric hypocontractility in rats.
C. Duodenum, Jejunum, Ileum, and Ileocolonic Junction
There are numerous data supporting the hypothesis that NO plays a pivotal role in NANC relaxation or IJP associated with electrical field stimulation or nicotinic agonists in the duodenum, jejunum, and ileum from a variety of mammals, including the dog (Toda et al., 1990a, 1991; Bogers et al., 1991), human (Maggi et al., 1991; Zyromski et al., 2001), rat (Berrino et al., 1995; Sotirov et al., 1998; Donat et al., 1999; Correia et al., 2000; Okishio et al., 2000b; Tanovic et al., 2001; Yamaji et al., 2002), guinea pig (Toma et al., 1999), hamster (Matsuyama et al., 1999, 2002), and mouse (Ogulener et al., 2001). The ileocolonic junction also has intense nitrergic innervation in the dog (Boeckxstaens et al., 1990a, 1993b; Bogers et al., 1991; Leelakusolvong et al., 2002), cat (Mozhorkova et al., 1994), opossum (Matsuda et al., 1998), and pig (Kajimoto et al., 2000) that would contribute to the regulation of the ileocolonic transit of intraluminal contents. Muscarinic M1 receptors also mediate nitrergic inhibitory responses in the rat intestine (Olgart and Iversen, 1999) and cat duodenum (Shikova and Kortezova, 2000). NO is likely an inhibitory neurotransmitter in the human jejunal longitudinal smooth muscle, acting via a mechanism mediated by guanylyl cyclase; however, other mechanisms are also involved (Zyromski et al., 2001).
The presence of neurons containing NADPH diaphorase or nNOS has histologically been established in the small intestine from humans (Fang et al., 1993; Timmermans et al., 1994) and animals (Wilhelm et al., 1998; Hens et al., 2002). There is an anatomical heterogeneity of the nitrergic system in the ileum of guinea pig, rat, rabbit, and cat (Wilhelm et al., 1998).
Involvement of endogenous NO in the regulation of intestinal motility was noted in the rat (Calignano et al., 1992) and guinea pig (Holzer, 1997). Kuiken et al. (2002a) have shown that inhibition of NO biosynthesis triggers the onset of a rapidly propagating phase III-like activity and shortens the postprandial period, and they have indicated that NO is involved in the modulation of fasting and postprandial small intestinal motility in humans. In jejunal longitudinal muscle strips obtained from rats subjected to chronic extrinsic denervation for 1 to 8 weeks by small bowel transplantation, contractile responsiveness was inhibited possibly by increased functioning of NANC inhibitory nerves, in which NO was partially involved (Balsiger et al., 2001). In anesthetized rats, laparotomy and manipulation of the small intestine inhibited the intestinal transit of Evans blue, and this inhibition was partially reversed by pretreating the rats with reserpine and completely reversed by additional treatment with L-NA, suggesting the involvement of adrenergic and nitrergic pathways in the pathogenesis of experimental ileus (De Winter et al., 1997). Duodenal strips isolated from rats with experimental diabetes induced by streptozotocin administration responded to nicotine, DMPP, or electrical stimulation with L-NA-sensitive relaxations that were substantially smaller than those in control rats (Martinez-Cuesta et al., 1995; Kaputlu et al., 1999). Therefore, impairment of nitrergic innervation may contribute to the abnormalities of intestinal motility associated with diabetes.
D. Colon, Rectum, and Internal Anal Sphincter
A functional role of nitrergic neurons in the colon has been determined in humans (Bartho et al., 2002) and in dogs (Dalziel et al., 1991; Okamura et al., 1998b), rabbits (Smith and Muir, 1991), rats (Middleton et al., 1993; Borjesson et al., 1997), guinea pigs (Zagorodnyuk and Maggi, 1994), and hamsters (Matsuyama et al., 2003). In healthy subjects and patients with irritable bowel syndrome, stimulation of the NO-cyclic GMP pathway by administration of sildenafil decreased the rectal tone but did not influence rectal distensibility, and relaxation of the rectum was accompanied by an increase in rectal volumes to reach perception thresholds (Fritz et al., 2003). Giant migrating contractions of the rat colon, possibly mediated by neuronal release of acetylcholine, appear to be partially suppressed by constitutive release of NO (Li et al., 2002). Nitrergic innervation also contributes to the regulation of the smooth muscle tone in the rat rectum (Takeuchi et al., 1998a) and in the guinea pig (Stebbing et al., 1996) and opossum internal anal sphincter (Rattan and Chakder, 1997). It was reported that NANC nerve stimulation, CO, and a CO-releasing molecule produced internal anal sphincter relaxation via guanylyl cyclase/cyclic GMP-dependent protein kinase activation (Rattan et al., 2004). Sun et al. (2004) noted that bile acids increased colonic permeability in rats via a mechanism that was inhibited by NO in the distal colon and suggested a possible role of NO in maintaining the epithelial barrier function. Stebbing et al. (1996) provided direct anatomical evidence for a descending nitrergic rectoanal neuronal pathway. Histochemical studies demonstrated nitrergic innervation in the colon of humans (Nichols et al., 1994, 1995; Peaire et al., 1997), pigs (Barbiers et al., 1993), and guinea pigs (Wang et al., 1996).
In colonic strips isolated from rabbits with trinitrobenzene sulfonic acid-induced colitis, relaxations elicited by electrical field stimulation were abolished, and nNOS expression was decreased (Depoortere et al., 2002). Similar findings were also observed in the severely inflamed colon and caecum from pigs infected with Schistosoma japonicum (Balemba et al., 2002). These alterations may contribute to impaired intestinal motility and absorption. Nitrergic innervation disappeared in the colon of patients with Hirschsprung's disease (Guo et al., 1997) and in the internal anal sphincter of patients with achalasia of this sphincter (Hirakawa et al., 1995).
E. Sphincter of Oddi and Gall Bladder
NO is a major transmitter of NANC neurogenic relaxation of the sphincter of Oddi in the guinea pig (Mourelle et al., 1993; Pauletzki et al., 1993), opossum (Allescher et al., 1993), Australian possum (Baker et al., 1993), rabbit (Lonovics et al., 1994), dog (Tanobe et al., 1995), pig, and human (Sand et al., 1997). Nitrergic nerve stimulation induced relaxation in association with an increase in the tissue cyclic GMP content in the rabbit sphincter of Oddi (Szilvassy et al., 1998). Recombinant human hemoglobin may alter the motor function of the sphincter of Oddi in vitro and in vivo by binding endogenous NO (Cullen et al., 1996). Neurons containing nNOS immunoreactivity or NADPH diaphorase were histochemically detected in the sphincter of Oddi of the human, monkey (De Giorgio et al., 1994), dog (Tanobe et al., 1995), guinea pig (Wells et al., 1995), and possum (Baker et al., 1993; Simula et al., 2001). The presence of NOS has also been demonstrated in the gall bladder and biliary pathways of the guinea pig (Grozdanovic et al., 1994), monkey, and human (De Giorgio et al., 1994; Uemura et al., 1997). In humans, NOS-containing nerve fibers are more abundant in the sphincter of Oddi than in the gall bladder (De Giorgio et al., 1994). In anesthetized cats, L-NA increased the flow resistance exerted by the sphincter of Oddi, but did not alter gall bladder motility, suggesting that nitrergic tonic inhibition is, if any, minimal in the gall bladder (Thune et al., 1995). Cholinergic nerve stimulation or agonists like cholecystokinin and acetylcholine caused contraction of gall bladder smooth muscle, and parasympathetic, nitrergic/VIPergic nerve activation evoked relaxation of the sphincter of Oddi. These events are likely coordinated with motility and secretory events in the upper GI tract, delivering bile into the duodenum (Shaffer, 2000).
In anesthetized dogs, systemic infusion of L-NAME produced hypertension and increased the sphincter of Oddi and duodenal motilities (Kaufman et al., 1993). L-Arginine, but not D-arginine, blocked these increases, suggesting that regulation of the baseline sphincter of Oddi and duodenal motility involves the generation of NO from L-arginine. In the anesthetized Australian possum, high doses of a highly selective iNOS inhibitor influenced the sphincter's motility by inhibiting nNOS activity; thus, the effects need to be considered in relation to therapeutic doses of this agent (Sandstrom et al., 2004). Local somatothermal stimulation applied to anesthetized cats and rabbits relaxed the sphincter of Oddi. Pretreatment with L-NAME blocked the relaxation, which was reversed by L-arginine (Chiu et al., 1998). Local heat-induced sphincter of Oddi relaxation was also seen in patients undergoing endoscopic retrograde cholangiopancreatography. The authors concluded that local thermal stimulation inhibits the sphincter Oddi motility by activating the heat-sensitive neuronal release of NO.
It was noted that nitrergic relaxation of the sphincter of Oddi was impaired in hyperlipidemic rabbits (Szilvassy et al., 1996). Neurogenic relaxation and an increase in the tissue cyclic GMP content were reversed in preparations from hyperlipidemic rabbits that underwent a treatment with farnesol (Szilvassy et al., 1998). In the sphincter of Oddi isolated from rabbits pretreated with lovastatin, nitrergic relaxation was impaired, and the relaxation was recovered by combined treatment with lovastatin and farnesol (Sari et al., 2001). Topical applications of the NO donor S-nitroso-N-acetylcysteine (Slivka et al., 1994) or nitroglycerin (Luman et al., 1997) to patients undergoing routine sphincter Oddi manometry reduced or abolished the phasic and tonic contractions of the sphincter of Oddi. A novel clinical approach using local NO donors may be utilized to control GI motility and regulate sphincteric function.
1. Liver.
At the hepatic hilus in the cat, a rich plexus of nNOS-immunoreactive nerve fibers and ganglia was detected around the interlobular branches of the bile duct, and the fibers were observed running along the length of a few vessels and in the ducts of the deep parenchyma (Esteban et al., 1998b). It is suggested that nNOS is involved in the autonomic control of hepatic blood flow and the regulation of hepatobiliary excretory activity and hepatocellular metabolic function. Similar results were also observed in the rat, rainbow trout,
, superoxide anion; GTP, guanosine triphosphate; MB, methylene blue; ODQ, 1H-[1,2,4]oxadiazolo[4,3a]quinoxalin-1-one; cGMP, cyclic GMP.