We discuss three pathologies of the intestine below, inflammation, diabetes and ileus. Although we have considered these separately, it should be noted that practically any irritation or insult to the GI tract causes inflammation. Thus, even handling of the intestine that causes ileus results in inflammation,71,72 as do damage to the mucosa and ischaemia.
Effects of inflammation on sympathetic neurons
Inflammation of the intestine causes changes in functions that are strongly influenced by sympathetic nerve activity, notably changed motility and secretion.73 Moreover, inflammation in one region causes changes in other regions,74–76 and, as detailed above, sympathetic reflex pathways are prominent in the communication between regions. However, few investigations have looked specifically at changes in sympathetic neurons.
Release of noradrenaline from sympathetic endings in the gut, when they are experimentally stimulated, is markedly reduced during colitis.77–79 Reduction is seen both in the area of intestine that is inflamed and in remote regions.77–79 N-type Ca2+ channels at the terminals of sympathetic postganglionic neurons open to admit Ca2+ for transmitter release and inhibition of these channels appears to contribute to the diminished release. Colitis reduced voltage-gated Ca2+ currents (ICa) in sympathetic postganglionic neurons by inhibiting ω-conotoxin GVIA-sensitive N-type Ca2+ channels, and the ω-conotoxin GVIA-sensitive component of noradrenaline release was significantly smaller in the colon during colitis. The inhibition of ICa was accompanied by a decrease in mRNA encoding the N-type Ca2+ channel alpha subunit and a depolarizing shift in voltage-dependent activation of the current.80 Over-expression of presynaptic α2-adrenoceptors on sympathetic nerve terminals may also contribute to reduced noradrenaline release during colitis.78
Despite release of noradrenaline being reduced, sympathetic cell bodies are activated by GI inflammation, as shown by the induction of c-Fos in sympathetic postganglionic neurons of the inferior mesenteric ganglia following inflammation of the colon81 and hyperexcitability of celiac ganglion neurons after inflammation of the distal ileum.82 A recent study, that examined the impact of stress on reactivation of TNBS-colitis in rats, reported that the susceptibility to reactivation of colitis was associated with cardiovascular indicators of increased sympathetic nervous system activity. However, inhibition of transmitter release by bretylium administration did not prevent the reactivation of colitis by stress.83
It is feasible that the inhibition of release described above is a negative feedback reaction to the increased activity of the neurons. In fact, ulcerative colitis patients do not have increased circulating noradrenaline levels,84 which may suggest that increased excitability is off-set by lowered release. It will be important for future studies to resolve the apparently discordant findings of increased sympathetic neuronal activity and decreased sympathetic neurotransmitter release in animal models of IBD. This might be achieved by measuring transmitter release in vivo, or in response to activation of reflexes, rather than in response to electrical stimulation or artificial depolarization. It is also important to note that release studies have not distinguished release from different functional sub-classes of sympathetic neurons, e.g., motility inhibiting vs vasoconstrictor neurons, whereas functional studies suggest that there are specific effects of inflammation on subgroups of neurons.82 One potential solution will be to utilize local amperometry techniques to quantify catecholamine release from individual populations of sympathetic postganglionic neurons.85
The extracellular catabolism of purines, that are co-transmitters released from sympathetic neurons, appears to be modulated by inflammation. Sympathetic purinergic vasoconstriction of submucosal arterioles of mouse colon was dramatically reduced in TNBS and DSS models of IBD, due to enhanced catabolism of ATP by the ectonucleoside 5′-triphosphate diphosphohydrolase, CD39.47,86 Consistent with this, a recent study reported a diminished purinergic vasoconstriction of mesenteric arteries from IBD patients.87 Interestingly, investigation of synaptic transmission in the ENS during TNBS-induced colitis in guinea-pigs revealed an increase in purinergic transmission during colitis,88 suggesting that the extracellular catabolism of purines may be differentially regulated within ganglia and in the vicinity of the vasculature.
To date, there is a lack of agreement in the literature on whether there are structural changes in the sympathetic innervation of the GI tract in IBD. A recent study of mice with chronic DSS colitis reported a loss of TH immunoreactivity in the colon,65 whereas studies of acute DSS and TNBS-induced colitis were unable to detect changes in TH immunoreactivity.47,79 Studies of ileum from Crohn’s disease patients reported an increase in TH immunoreactivity in the myenteric plexus, including within enteric perikarya, and in the mesenteric vasculature compared to uninflamed controls.87,89 Further studies will be required to examine whether ulcerative colitis and Crohn’s disease, with their differing aetiologies, have distinct effects on sympathetic neuroanatomy.
Diabetes causes profound alterations to the structure and function of the GI tract and considerable changes to the chemical coding and properties of enteric neurons.90 Diabetic sympathetic neuropathy is characterized by the selective development of dystrophic and degenerative changes in the sympathetic nerve terminals innervating the GI tract.91,92 Different populations of sympathetic neurons differ in their susceptibility to diabetes, possibly due to differences in vulnerability to oxidative stress; prevertebral ganglia are greatly affected, whereas paravertebral neurons are relatively spared.93 Rats with streptozotocin-induced diabetes display markedly swollen dystrophic axons in the ileum and an altered pattern of innervation.91,92 By contrast, the sympathetic innervation of the distal colon was comparatively normal in this model of experimental diabetes.92 The alterations in structure are accompanied by functional changes. For example, in humans, diabetics with evidence of sympathetic involvement have more rapid transit through the distal small bowel.94 Despite these important observations, there is a paucity of mechanistic information on the impact of the altered sympathetic innervation in diabetes on the ENS or on GI function. How the ENS adapts to the reduced functional inhibitory innervation is an important question that remains to be answered.
When the handling of the intestine is extensive, or peritonitis is induced, inhibition of motility following the operative procedure can be long-lasting. This condition is known as postoperative or adynamic ileus. It has a sympathetic reflex component, but is also contributed to by other mechanisms, particularly in persistent ileus. In its acute phase, postoperative ileus is relieved by sympathectomy or splanchnic anaesthesia95,96 or by inhibition of transmission from the postganglionic neurons.97,98 Spinal anaesthesia or sympathetic blockade is less effective when ileus results from severe and widespread peritonitis.99,100 In particular, prolonged ileus involves an inflammatory infiltration of the external musculature and myenteric plexus region, is not relieved by blocking noradrenergic transmission, but is reduced by opiates.100