CrossTalk proposal: 5-HT is necessary for peristalsis

The overwhelming preponderance of the 5-HT of every known mammal is in the gut (Erspamer, 1966; Gershon & Tack, 2007; Gershon, 2013). Enteric 5-HT must be an important signalling molecule to be so conserved. Most enteric 5-HT is in enterochromaffin (EC) cells, but smaller amounts are present in myenteric neurons. Despite their small numbers, serotonergic neurons project widely throughout the enteric nervous system (ENS) and also innervate interstitial cells of Cajal (ICC) (Okamoto et al. 2014). The abundance and variety of enteric 5-HT receptors (5-HT_1-7 and subtypes) also suggest that 5-HT plays a significant role in GI physiology (Smith et al. 2014). The efficacy of therapies targeting enteric 5-HT or its receptors against GI motility disorders supports this idea (Gershon & Tack, 2007; Gershon, 2013). Beyond motility and secretion, putative roles that 5-HT plays include metabolism, osteogenesis, immunity, neurogenesis and neuroprotection (Gershon, 2013). The multiplicity of enteric 5-HT targets and receptors complicates ascertaining the physiological roles of 5-HT. Controversy is thus to be expected and has appeared in recent papers, which question whether EC or neuronal 5-HT has anything to do with peristalsis (Keating & Spencer, 2010; Spencer et al. 2011; Sia et al. 2013; Spencer et al. 2013). These papers are important not because enteric 5-HT is vestigial (it is not) or that its roles in normal and abnormal GI motility can be ignored (they cannot). Instead, they highlight common misunderstandings about peristaltic reflexes and 5-HT cellular biology. The papers focus on colonic migrating motor complexes (CMMCs), which are aborally propagating propulsive contractile complexes, essentially peristaltic reflexes. The authors assert that they can evoke CMMCs after mucosal removal or depleting 5-HT with reserpine; therefore, they conclude that neither EC cells, nor neuronal 5-HT is necessary for CMMCs. To comprehend what is misunderstood, it is necessary to discuss basic information about 5-HT and GI motility.

Two tryptophan hydroxylase isoforms, TPH1 and TPH2, are rate limiting in 5-HT biosynthesis, TPH1 in EC cells, and TPH2 in serotonergic neurons (Gershon, 2013). 5-HT is synthesized in the cytosol but stored in vesicles. Reuptake terminates actions of 5-HT. Because 5-HT is charged, two transporters are required for transmembrane transport, a vesicular monoamine transporter (VMAT1 in EC cells and VMAT2 in neurons) (Henry et al. 1998) and a plasmalemmal serotonin reuptake transporter (SERT) (Blakely, 2001). Reserpine inhibits VMAT (Henry et al. 1998). Intracellular 5-HT is thus reduced due to enhanced catabolism; but reserpine does not prevent 5-HT biosynthesis or constitutive release. That requires deletion or inhibition of TPH, which when isoform-selective, distinguishes mucosal from neuronal 5-HT (Li et al. 2011; Gershon, 2013). SERT deletion amplifies 5-HT effects. Because GI motility is abnormal after deletion of SERT (Chen et al. 2001), either isoform of TPH (Li et al. 2011; Gershon, 2013), or exposure to 5-HT antagonists/agonists (Monro et al. 2002; Smith et al. 2014), 5-HT clearly influences GI motility.

Peristalsis is a general term applied to enteric motile behaviour that should not be conflated with the ENS-mediated peristaltic reflex that can drive propulsion (Gershon & Tack, 2007; Gershon, 2013; Furness et al. 2014). That reflex, first called the ‘law of the intestine’ (Bayliss & Starling, 1899), is an oral contraction and anal relaxation; it is evoked by increased intraluminal pressure and involves polarized neural pathways within the ENS (Furness et al. 2014; Smith et al. 2014). Many enteric cells, not just neurons and muscle, participate in peristaltic reflexes (Smith et al. 2014). The peristaltic reflex is only one of many activity patterns encoded within the ENS (Furness et al. 2014).

Mucosal pressure/distortion or chemical stimuli release 5-HT from EC cells and evoke peristaltic reflexes (Bülbring & Lin, 1958; Bertrand et al. 2008). Luminally applied 5-HT mimics pressure (Bülbring & Crema, 1958; Bülbring & Lin, 1958). Fecal pellets apply pressure to the mucosa and thereby release 5-HT, which entrains CMMCs (Heredia et al. 2009). Mucosal removal, anaesthesia, or asphyxiation all abolish mucosally evoked peristaltic reflexes (Bülbring & Crema, 1958; Bayguinov et al. 2010; Dickson et al. 2010). Mucosally released 5-HT acts on 5-HT₃ and/or 5-HT_1P (or 5-HT₇) receptors to stimulate intrinsic primary afferent neurons (IPANs) (Pan & Gershon, 2000; Bertrand et al. 2008; Dickson et al. 2010), which engage the ENS (Kirchgessner et al. 1992; Bayguinov et al. 2010; Okamoto et al. 2014). IPANs are found in both plexuses (Kirchgessner et al. 1992; Bayguinov et al. 2010; Okamoto et al. 2014), and appear to link mucosal and neuronal 5-HT pools together (Okamoto et al. 2014; Smith et al. 2014). 5-HT₃ antagonists can block CMMCs when applied around fecal pellets, as do intraluminal 5-HT₃ and 5-HT₇ antagonists, suggesting that these antagonists act locally on the EC cell-to-IPAN junction (Heredia et al. 2009; Smith et al. 2014).

Evidence suggests that neuronal, as well as mucosal, 5-HT is critical for peristaltic reflexes. Serosally applied 5-HT desensitizes ENS receptors, thereby inhibiting peristaltic reflexes (Bülbring & Crema, 1958; Smith et al. 2014). 5-HT antagonism interferes with transmission in ENS pathways, CMMCs, and tonic inhibition in the colon (Monro et al. 2002; Dickson et al. 2010). Neuronal 5-HT can mediate intestinal slow excitatory postsynaptic potentials (sEPSPs): 5-HT₇ antagonists inhibit sEPSPs in IPANs (Monro et al. 2005), as well as CMMCs (Dickson et al. 2010). Tryptamine, which first releases and then depletes endogenous 5-HT, initially induces but then abolishes sEPSPs without affecting similar responses to exogenous 5-HT (Takaki et al. 1985). Anti-idiotypic antibodies, which bind selectively to all 5-HT receptors, also mimic sEPSPs before blocking them irreversibly (Wade et al. 1994).

Radial stretch of the bowel wall activates high threshold mechanosensitive interneurons that activate CMMCs (Heredia et al. 2009). Mucosal stimuli and radial stretch evoke similar reflex responses because nerve pathways from each converge on final common neurons (Smith et al. 1992, 2007). Mucosal reflexes alone propel small fecal pellets that do not produce radial stretch down the colon (Heredia et al. 2013); fluid or larger pellets that stretch the gut can be propelled in the absence of the mucosa or mucosal 5-HT (Spencer et al. 2011; Heredia et al. 2012, 2013). Because stimuli that short circuit mucosal activation evoke CMMC-like responses does not mean the mucosa and its 5-HT are not physiologically critical. Stimuli restricted to the mucosa cannot evoke CMMCs in the TPH1KO colon and thus are 5-HT-dependent (Heredia et al. 2013). If the TPH1KO bowel is stretched, CMMC-like responses are evoked; however, they do not propagate and thus are not CMMCs. In an analogy, the lower leg can be made to move involuntarily through the patellar reflex or voluntarily. Voluntary leg movement does not obviate the need for quadriceps muscle spindles to evoke patellar reflexes. Circuits in the myenteric plexus can be engaged in the absence of mucosal 5-HT to give rise to contractile activity; however, under physiological circumstances the mucosa is present and, when pressed, secretes 5-HT. When the gut is intact, therefore, 5-HT will do what it does when the mucosa releases it, initiate peristaltic reflexes.

The argument (Spencer et al. 2013) that because reserpine-induced 5-HT depletion fails to prevent CMMCs, 5-HT is not needed is invalid. Because reserpine only inhibits VMAT, it cannot drive tissue 5-HT to zero. Reserpine lowers intracellular 5-HT to levels that may be difficult to detect (Bülbring & Crema, 1959; Spencer et al. 2013); however, the 5-HT that remains activates receptors. In fact, constitutive 5-HT release in reserpine-treated animals enhances intestinal motility (Bülbring & Crema, 1959). The continued secretion of 5-HT thus explains the ability of 5-HT₃ and 5-HT₄ antagonists to block responses in reserpinized preparations (Sia et al. 2013; Spencer et al. 2013). Importantly, 5-HT₃ antagonists do not affect the CMMC-like activity in TPH1KO mice (Heredia et al. 2013). The abnormality of CMMCs in TPH1KO mice establishes that physiologically meaningful peristaltic reflexes are 5-HT dependent (Heredia et al. 2013).

Total GI transit and colonic motility are slowed in TPH2KO mice but gastric emptying is accelerated (Li et al. 2011; Gershon, 2013). Normal GI motility thus requires neuronal 5-HT; however, because neuronal 5-HT is a growth factor, the TPH2KO ENS is severely hypoplastic (Li et al. 2011), which could be responsible for defective GI motility. Accelerated gastric emptying probably occurs because serotonergic activation of gastric inhibitory motor neurons is impaired in mice lacking neuronal 5-HT (Li et al. 2011). Patterns of GI motility other than peristaltic reflexes evidently compensate for the defective peristaltic reflex/CMMC of TPH1KO mice (Li et al. 2011; Gershon, 2013). Compensatory mechanisms, such as enhanced prostaglandin synthesis may also be mobilized to generate CMMC-like activity in the absence of mucosal 5-HT (Heredia et al. 2012; Smith et al. 2014).

In conclusion, 5-HT has satisfied all of the criteria needed to identify it a mediator of peristaltic reflexes. EC cells and enteric neurons synthesize 5-HT. Exogenous 5-HT and stimuli that release endogenous 5-HT evoke peristaltic reflexes, which are lost or impaired when mucosal and/or neuronal 5-HT is depleted or 5-HT receptors are antagonized. Mucosal and neuronal 5-HT are thus essential for physiological manifestation of peristaltic reflexes.

Call for comments

Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘Last Word’. Please email your comment, including a title and a declaration of interest to jphysiol@physoc.org. Comments will be moderated and accepted comments will be published online only as ‘supporting information’ to the original debate articles once discussion has closed.

Biographies

Terence Smith (right) is Professor in the Department of Physiology and Cell Physiology at the University of Nevada-Reno, USA, where he is the Director of the Dynamic Imaging Core. After working for several years in solid state physics, he received his PhD in neuropharmacology/electrophysiology from Monash University, Victoria, Australia under Professors Mollie Holman and David Hirst. His interests have focused on the circuitry in the enteric nervous system and how this affects gut pacemakers.

Michael Gershon (left) is Professor of Pathology and Cell Biology at Columbia University, College of Physicians and Surgeons. He received his MD degree from Cornell University, did post-doctoral research with Professor Edith Bülbring in the Department of Pharmacology of Oxford University, and chaired the Department of Anatomy and Cell Biology at Columbia until 2006. His interests are in enteric neuronal development, cell biology, and function as well as the roles serotonin plays in the bowel.

Additional information

Competing interests

None declared.

Funding

This work was funded by grants from the National Institutes of Health: RO1 DK45713 (T.K.S.) and NS 12969, NS 15547 (M.D.G.).