Serotonin (5-HT) is released from the enterochromaffin cells and plays an important role in regulating intestinal function. Although the release of 5-HT is well documented, the contribution of the serotonin reuptake transporter (SERT) to the levels and actions of 5-HT in the intestine is unclear. This study aimed to demonstrate real-time SERT activity in ileal mucosa and to assess the effects of SERT inhibition using fluoxetine. Electrochemical recordings were made from the mucosa in full-thickness preparations of rat ileum using a carbon fiber electrode to measure 5-HT oxidation current and a force transducer to record circular muscle (CM) tension. Compression of the mucosa stimulated a peak 5-HT release of 12 ± 6 μM, which decayed to 7 ± 4 μM. Blockade of SERT with fluoxetine (1 μM) increased the peak compression-evoked release to 19 ± 9 μM, and the background levels of 5-HT increased to 11 ± 7 μM (P < 0.05, n = 7). When 5-HT was exogenously applied to the mucosa, fluoxetine caused a significant increase in the time to 50% and 80% decay of the oxidation current. Fluoxetine also increased the spontaneous CM motility (P < 0.05; n = 7) but did not increase the CM contraction-evoked 5-HT release (P > 0.05, n = 5). In conclusion, this is the first characterization of the real-time uptake of 5-HT into the rat intestine. These data suggest that SERT plays an important role in the modulation of 5-HT concentrations that reach intestinal 5-HT receptors.
- enterochromaffin cell
- gastrointestinal tract
- serotonin reuptake transporter
serotonin (5-ht) is produced and stored in the enterochromaffin (EC) cells of the intestinal epithelium (19, 34). Released 5-HT acts as a sensory mediator and plays an important role in visceral sensation and in regulating gut function by activating 5-HT receptors on the afferent nerve terminals (1, 22, 26). Activation of the intrinsic sensory neurons of the enteric nervous system by 5-HT is responsible for initiating or enhancing local intestinal motor reflexes (11, 42–44) and for increasing secretory function (e.g., 39). EC-derived 5-HT also activates extrinsic afferent fibers and thus contributes to the relay of sensory information to the central nervous system (24, 25, 41).
The actions of 5-HT are terminated by uptake via the serotonin reuptake transporter (SERT, a Na+/Cl−-dependent transporter) (31). SERT is the target for many important therapeutic drugs such as Prozac (fluoxetine), a member of the serotonin-selective reuptake inhibitors. SERT has been localized to many intestinal epithelial cells using immunohistochemistry to detect protein (44) and Northern blots to detect mRNA (13). The SERT gene is subject to a number of naturally occurring polymorphisms in the coding and promoter regions, some of which have been linked to the symptoms of irritable bowel syndrome (IBS) (e.g., 46) or to the pharmacogenomics of IBS treatment (12, 37). A novel splice variant of SERT has been identified that is specific for the intestinal epithelium (30) but has not yet been linked to disease. SERT is therefore a potential target for drug therapy of functional bowel disorders such as IBS (27, 41). Previous studies have shown that there are decreased levels of SERT expression in patients with ulcerative colitis or IBS (e.g., 14) and in some animal models of colitis (28, 29).
The aim of this project was to characterize the real-time activity of SERT in the rat ileal mucosa and to assess the effects of SERT inhibition on 5-HT release kinetics and motility by using the serotonin-selective reuptake inhibitor fluoxetine. We used electrochemical methods that have a high temporal and spatial resolution and that measure 5-HT release directly from a small group of intact intestinal EC cells (2, 3). Thus we can apply a mechanical stimulus and measure 5-HT release and quantify the activity of SERT directly at the site of action.
MATERIALS AND METHODS
All experiments were performed using Sprague Dawley rats of either sex, fed with a standard lab diet (n = 54, 300 to 500 g) obtained from the University of New South Wales. Rats were anesthetized with intraperitoneal injection of pentobarbital sodium, and the carotid arteries were severed. All procedures were approved by the University of New South Wales Animal Experimentation Ethics Committee. A 3-cm long (full width, ∼1.5 cm) segment of distal ileum was removed ∼10 cm from the ileocecal junction. The segment was placed in warmed (35°C) oxygenated (95% O2-5% CO2) physiological saline (composition in mM: 117 NaCl, 1.2 NaH2PO4, 1.2 MgCl2, 2.5 CaCl2, 4.7 KCl, 25 NaHCO3, and 11 glucose) and was pinned flat, mucosa side up, in a small organ bath (Fig. 1A). The flow of physiological saline was ∼5 ml/min, and the bath volume was ∼3 ml. In later experiments, the segment was superfused with a physiological saline containing papaverine (100 μM) to paralyze the smooth muscle. Unless noted, all drugs were purchased from Sigma-Aldrich, St. Louis, MO.
Electrochemical recordings of 5-HT release were made using carbon electrodes, a technique recently validated in gastrointestinal tissues by our laboratory and others (2, 3, 9, 32). In the present study, we used similar methods to our previous study (2). We have estimated that the majority of 5-HT release recorded using a small carbon fiber electrode comes from a group of 5–7 EC cells with a smaller contribution from the surrounding EC cells. We used a carbon fiber electrode (CFE), which consisted of a single carbon fiber (7 μm in diameter, 200 μm exposed at the tip) insulated with a glass micropipette (1.5-mm outer diameter, 0.86-mm inner diameter; Clark Electromedical Instruments, Kent, UK) and a pellet of woods metal plus copper wire to establish an electrical connection. Current recordings were made using a VA-10 amplifier (NPI Electronics, Tamm, Germany), digitized at 1–5 kHz (Digidata 1200B; Axon Instruments, Union City, CA), and recorded for analysis on a personal computer using PClamp 9.0 (Axon Instruments, 0.5 kHz filtering with a 50-Hz notch filter). Preparations were visualized at 20× magnification using an upright dissecting microscope.
The CFE was pretreated with Nafion (an anionic exchange resin; Sigma-Aldrich)(see Ref. 21) and was voltage clamped at +300 to +400 mV vs. a Ag/AgCl ground. At this potential, 5-HT and the catecholamines will oxidize but not other biological chemicals; the 5-HT oxidation current appears as a positive current deflection (2). Furthermore, the identity of the oxidation currents were investigated using cyclic voltammetry techniques (ramp: 0 V to +1 V; rate 0.2 V/s or −0.5 V to +1 V and back, ∼0.5 V/s). The oxidation peak from a control solution of 5-HT was compared with that observed from the overflow mucosa. The voltage at which peak oxidation occurred for the 5-HT control solution and for the overflow were very close (+315 and 345 mV, respectively; Fig. 2A). The late peak seen in the mucosa overflow trace in Fig. 2A is most likely attributable to the oxidation of melatonin (2, 10, 33). We did not record this during our amperometry experiments because the potential used to oxidize 5-HT (+400 mV) was too low to oxidize melatonin (> +650 mV).
To calibrate electrodes, 5-HT from a stock solution (10 mM in distilled H2O, 5-hydroxytryptamine creatinine sulfate; Sigma-Aldrich) was dissolved in physiological saline (10–20 μM) and pressure ejected (10 psi, Picospritzer III; Parker Instrumentation, Sydney, Australia) from a micropipette (10–20 μm tip diameter) onto the exposed carbon fiber electrode. Precision micromanipulators (MP-1 and M-3333; Narishige, Tokyo, Japan) were used to position the CFE and the 5-HT-containing micropipette within 20 μm of each other (and away from the preparation). Pressure ejection of 5-HT (pulse duration increasing from 0.05 to 2.5 s) evoked a peak oxidation current that was proportional to the concentration of 5-HT ejected and was used to calculate levels of 5-HT recorded in the mucosa (Fig. 2B).
A similar arrangement of the CFE and micropipette was used to measure SERT-dependent uptake of 5-HT into the mucosal epithelium (Fig. 1B) with the use of electrochemical techniques developed in the central nervous system (CNS) (8, 16, 17). In this instance, both the electrode and the micropipette were placed on the mucosal surface such that SERT in the epithelium contributed to the decay kinetics of the exogenous 5-HT oxidation current. SERT-dependent uptake of 5-HT was then assessed by comparing decay under control conditions and during blockade of SERT. Preliminary experiments showed that pressure ejection of physiology saline only at the same pressures and durations (pulse duration: 0.05 to 0.25 s; 10 psi) did not evoke significant endogenous 5-HT release. Once the arrangements of the electrode and micropipette were established in control, they remained fixed for the reminder of the SERT-dependent uptake experiments.
Measurements of 5-HT oxidation current commenced after 30 min or more of tissue equilibration. To measure mechanically stimulated 5-HT release and/or background levels of 5-HT, the CFE was used to compress the mucosa at three or more points from each preparation. Readings were taken of the peak and background (steady state) levels of 5-HT at each of these points and the results for each averaged. To measure SERT function, the time course of the exogenous 5-HT oxidation current was recorded from a single point on the mucosa under control conditions (minimum 3 repetitions) and during superfusion of fluoxetine (1 μM, Sigma-Aldrich; minimum 3 repetitions). Fluoxetine is not itself electroactive at positive potentials (35). All electrodes showed a gradual and linear decrease in sensitivity over time. This effect was minimized by reducing the number of exposures of the electrode to 5-HT and by holding the electrode at 0 mV in between applications of 5-HT. If a sudden decrease in electrode sensitivity was detected, then a fresh electrode was calibrated and used instead. In rodents, mast cells also contain some 5-HT (45). No attempt was made to differentiate 5-HT release from EC cells and possible release of 5-HT from mast cells; however, compound 48/80 (2–20 μg/m, a mast-cell degranulator) was used in two preparations and was found not to cause detectable release of 5-HT. This observation is in line with previous results from noninflamed control tissues where compound 48/80 did not cause a 5-HT-mediated activation of extrinsic afferents (e.g., Ref. 15).
Measurement of circular muscle tension.
A force transducer (FT-03C; Grass, Quincy, MA) was used to record circular muscle (CM) tension in some experiments. The preparations were pinned loosely with one side attached to the force transducer (see Fig. 1A) and were about 15 mm wide by 25 mm in length. The CM was placed under 0.1 to 0.5 g of resting tension. The CFE was used to detect 5-HT oxidation current on the side of the preparation opposite from the force transducer (Fig. 1A). The frequency and amplitude of spontaneous contractions were measured under control conditions and in the presence of drugs.
Measurements were calculated as means ± SE, with n representing the number of animals used. Student's t-test was used to compare data for significant differences; P values < 0.05 were considered significant. Unless noted, all tests were one tailed and paired. ANOVA with a post hoc Tukey-Kramer test was used to compare multiple measures. The concentration of 5-HT released from the mucosa was calculated from the oxidation current generated at the exposed tip of the CFE, and the current produced by a known concentration of 5-HT was applied exogenously. The effect of fluoxetine on 5-HT dynamics was analyzed by comparing the amplitude, time to peak, and the 50% (t50) and 80% (t80) decay times of the 5-HT signal during the control and fluoxetine periods. The effect of fluoxetine on CM contractions was analyzed by comparing the frequency and amplitude of contractions between control and fluoxetine periods from the force transducer measurements. Motility index was calculated as: log(SUM × amplitude) × (number of waves + 1) (18). Area under the curve (AUC) for CM contraction and 5-HT oxidation currents was calculated using PClamp 9.0 with the baseline taken as the 80% return to the starting level; 80% return was chosen on the basis of our previous experience (8).
Compression-evoked 5-HT release and background levels of 5-HT.
Previous data from guinea pig and mouse have suggested that compression of the epithelium is a reliable stimulus for evoking mucosal 5-HT release (2, 3). We tested this finding in the rat by using the CFE to evoke 5-HT release by compressing the mucosa (n = 16). In all preparations, quick compression of the mucosa caused a transient increase in 5-HT (the peak) followed by a decay toward background (steady state) 5-HT levels (Fig. 3A). Compression of the mucosa stimulated a peak 5-HT release of 5.17 ± 0.73 nA (time to peak: 0.50 ± 0.14 s; n = 8), which decayed to 1.47 ± 0.66 nA (measured at 10–25 s postcompression). After taking into account the calibration for each electrode, peak 5-HT release was calculated to be 16 ± 7 μM, which decayed to a background level of 6 ± 3 μM (38% of peak, Fig. 3B).
Effects of SERT inhibition on 5-HT levels.
We determined whether SERT-dependent uptake limited the peak release or steady state levels of 5-HT, or both, in a further 13 preparations. In control, compression of the mucosa stimulated a peak 5-HT release of 12 ± 6 μM, which decayed to 7 ± 4 μM (Fig. 4), similar to that seen above. Superfusion with fluoxetine (1 μM) caused the peak compression-evoked release to increase to 19 ± 9 μM (158% of control) and the background levels of 5-HT to rise to 11 ± 7 μM (160% of control; n = 7; P < 0.05). To determine whether the background levels of 5-HT were altered by the large levels of 5-HT released by compression, background levels were assessed by bringing the electrode into very close contact with the mucosa without compression. In control, background oxidation current levels were 0.53 ± 0.26 nA, whereas in the presence of fluoxetine (1 μM) this was increased to 1.19 ± 0.59 nA (225% of control) (P < 0.05; n = 6). The increase in the background 5-HT levels in the presence of fluoxetine, as measured after a compression-evoked release (n = 7) or a gentle touch (n = 6), was not significantly different when compared as a percent of control (P > 0.05; unpaired t-test).
Exogenous 5-HT uptake kinetics in paralyzed intestine.
These experiments aimed to determine the extent to which SERT-dependent uptake changed the kinetics of 5-HT signaling. The gut was paralyzed by using the phosphodiesterase inhibitor papaverine (100 μM) to eliminate interference from spontaneous smooth muscle contractions, which can evoke endogenous 5-HT release (see below and Ref. 3). Under these conditions, endogenous release of 5-HT contributes very little to the exogenous 5-HT induced oxidation current (see materials and methods). When the electrode was calibrated by pressure ejection of exogenous 5-HT far from the tissue, the average time to peak was 0.08 ± 0.04 s with average amplitude at 5.42 ± 0.84 nA. The time for 50% decay was 0.28 ± 0.08 s, and the time to 80% decay was 0.62 ± 0.14 s. The time to peak and decay were measured without tissue and with only passive diffusion of the 5-HT into the continuously exchanging solution (5 ml/min; 3-ml bath volume).
This profile changed when the electrode was tested while touching the mucosal epithelium. Oxidation currents induced by pressure application of 5-HT (10 μM) still had a fast rising phase, with an average time to peak of 0.14 ± 0.04 s and a similar average amplitude of 5.78 ± 0.87 nA (Fig. 5A). However, the decay phase was much slower and was composed of two components: a fast decay followed a slower decay. The time to 50% decay (t50: 0.43 ± 0.14 s) and time to 80% decay (t80: 1.67 ± 0.52 s; P < 0.05; n = 10) were both significantly increased. When fluoxetine (1 μM) was added to the paralyzed preparation, there was a significant increase in the background levels of 5-HT from 8 ± 3 μM to 18 ± 7 μM (244% of control; P < 0.05; n = 8). Fluoxetine caused a significant increase in both t50, which increased to 0.82 ± 0.21 s (191% of control; P < 0.05; n = 10), and in t80, which increased to 2.58 ± 0.63 s (156% of control; P < 0.05; Fig. 5B). There was no change in the time to peak, but there was a small but significant decrease in the peak amplitude (control: 5.78 ± 0.92 nA; fluoxetine: 4.12 ± 0.86 nA, 71% of control). This decrease in peak was probably caused by a desensitization of the CFE attributable to the prolonged contact with the mucosa that was required for these experiments (see materials and methods).
Effect of SERT inhibition on endogenous 5-HT release.
From our previous study, it was clear that muscle activity can drive 5-HT release (3). In the present study, we used the spontaneous CM contraction-evoked 5-HT release to test the idea that SERT-dependent uptake contributes to the kinetics of endogenously released 5-HT. We first investigated whether fluoxetine had any effect on the spontaneous CM contractions present in many of our rat ileum preparations. Fluoxetine (1 μM) did not significantly increase the frequency of spontaneous contractions (control: 11/min; fluoxetine: 13/min; P > 0.05, n = 7) or the average amplitude of CM contractions (control: 0.3 ± g; fluoxetine at 0.4 ± g; P > 0.05, n = 7). Interestingly, when both frequency and amplitude were taken into account using a motility index (see materials and methods), there was a small but significant increase in the motor activity (control: 1.29 ± 0.53; fluoxetine: 1.58 ± 0.65; P < 0.05; n = 7).
We next used these spontaneous CM contractions to assess the effects of SERT blockade on endogenous release of 5-HT. 5-HT oxidation current was measured simultaneously during force generated by CM contraction (spontaneous contractions at 0.1 g and 0.5 g of resting tension were binned together; n = 5 each). In control, there was a clear correlation between the magnitude of the contraction and the magnitude of the 5-HT oxidation current (Fig. 6); that is, a larger contraction was associated with detection of a larger 5-HT peak. In the presence of fluoxetine, there was no significant change in CM contraction (control: 0.47 ± 0.25 g; fluoxetine: 0.43 ± 0.22 g; 6 samples each; n = 5) or 5-HT peak concentration (control: 1.5 ± 1.2 μM; fluoxetine: 0.8 ± 0.7 μM; 6 samples each; n = 5; P > 0.05; ANOVA).
To determine whether there were more subtle effects of fluoxetine on the 5-HT signal, the AUC was calculated for three of the previous preparations (7 or more samples for CM contraction and 5-HT oxidation current). In control, AUC for CM contraction was 510 ± 360 g/ms and while in the presence of fluoxetine it was 770 ± 410 g/ms (n = 3; P > 0.05). In control, 5-HT oxidation current was 8,920 ± 1,440 nA/ms, whereas with fluoxetine it was 5,510 ± 1,680 nA/ms (n = 3; P > 0.05). In addition, when individual events were plotted against each other (i.e., contraction vs. 5-HT), we found that the slope did not change or was reduced in presence of fluoxetine (Fig. 6B).
These data show that fluoxetine at 1 μM causes accumulation of high concentrations of 5-HT near the EC cells. These high concentrations could contribute to a reduction in further 5-HT release and explain the apparent lack of effect of fluoxetine on CM contraction-induced 5-HT release. To test this idea, we used a lower concentration of fluoxetine that should block SERT by ∼50%. In a further three preparations, CM contraction-evoked 5-HT was measured in control and during superfusion with fluoxetine (10 nM). Fluoxetine did not cause a significant change in CM contraction amplitude (control: 0.14 ± 0.01 g; fluoxetine: 0.11 ± 0.01 g; 6 or more samples each; n = 3) or 5-HT peak concentration (control: 2.3 ± 0.3 μM; fluoxetine: 1.7 ± 0.2 μM; 6 or more samples each; n = 3; P > 0.05; ANOVA). In addition, we reanalyzed our previous data using fluoxetine (1 μM) in an effort to see whether there were subtle changes in 5-HT release early in the superfusion before fluoxetine had reached equilibrium. Unfortunately, we could detect no differences in the 5-HT peak concentration or AUC measured during the first ∼10 min of superfusion vs. the second 10 min. Together, these data suggest that CM contraction-induced 5-HT release is not strongly regulated by SERT.
The main finding of this study was that SERT-dependent uptake of 5-HT contributes to maintaining the background levels of mucosal 5-HT and in limiting the peak of evoked 5-HT release. This is the first real-time demonstration of the uptake of 5-HT in the rat intestine and provides an important contribution to our growing knowledge of the moment-to-moment regulation of 5-HT levels within the intestinal epithelium.
SERT blockade and background levels of 5-HT.
In the present study, experiments measuring 5-HT oxidation currents in unparalyzed rat intestine showed increased background levels of 5-HT during fluoxetine superfusion. We would predict that the high levels of 5-HT were attributable to blockade of SERT and accumulation of free 5-HT in and around the mucosa. However, fluoxetine was associated with a small increase in motility and, as contractions can release mucosal 5-HT (3), this might have contributed to the higher 5-HT levels. We tested this by comparing background levels of 5-HT in unparalyzed vs. paralyzed tissue and found a similar increase (unparalyzed: 225% of control; paralyzed: 244%), suggesting that the blockade of SERT is the main factor in the increase. This effect of fluoxetine to increase background 5-HT levels is supported by studies in guinea pig ileum though any confounding effects of motility were not examined (9).
5-HT uptake kinetics in rat ileal mucosa.
Electrochemical methods were employed to detect changes in 5-HT oxidation current in real time. This allowed us to calculate 5-HT uptake by mucosal SERT using pressure ejection of a known concentration of 5-HT onto the intestinal mucosa and measuring the 5-HT oxidation signal decay times as has been done in the CNS (8, 16, 17). The immediate sharp decay of the 5-HT signal was most likely due to rapid diffusion into the surrounding solution, whereas the prolonged “foot” of the signal was probably due to 5-HT uptake via SERT located on the intestinal epithelial cells (e.g., 13). Since pressure ejection of saline alone does not show detectable 5-HT release, any effect of fluoxetine should be on the kinetics of the exogenous 5-HT signal. A measure of SERT activity can then be seen by comparing the 50% and 80% decay times in control vs. those measured in the presence of fluoxetine. Fluoxetine binds to SERT and blocks the uptake of 5-HT into the epithelial cells; hence fluoxetine should increase the levels of free 5-HT that are present in the intestine and/or should lengthen the time that 5-HT levels are elevated. Similar to the CNS, these data can help to characterize 5-HT uptake, which can then be used as a baseline for future studies.
The effects of fluoxetine on 5-HT release and motility.
Inhibition of SERT by fluoxetine enhances neurotransmission and the spread of excitation in the bowel, leading to increased peristaltic and secretory reflexes (22). For example, Wade et al. (44) have shown that fluoxetine increased the peristaltic reflex in the guinea pig large intestine at low concentrations (10 nM); however, at high sustained concentrations (>100 nM), the rate of propulsion was reduced. The authors hypothesized that this was caused by desensitization of 5-HT receptors to high and/or sustained concentrations of 5-HT; an idea that has gained acceptance in the literature (e.g., 28). In this study, the concentration of fluoxetine used was 1 μM, which should block all SERT activity. In keeping with Wade et al. (44), our fluoxetine-treated tissues showed variable but nonsignificant changes in the frequency of contractions and in the amplitude of individual contractions. However, when a motility index was calculated for these same data, it was found that fluoxetine did cause a small but significant increase in the overall motility of the preparation.
We used these spontaneous contractions of the CM to assess the effects of SERT blockade on physiologically evoked, endogenous release of 5-HT. We would predict that, in the presence of fluoxetine, each contraction of the CM would be associated with a larger endogenous 5-HT signal (Fig. 6). Our data did not, however, support this prediction. We found that the average peak and AUC calculations for 5-HT oxidation current and CM contraction revealed no significant change in the presence of fluoxetine. One possibility is that the higher background levels of 5-HT acted to reduce the amount of 5-HT released per contraction. Perhaps, as noted above, this occurred by activation of inhibitory receptors or desensitization of excitatory receptors on the EC cells (e.g., 20, 38); blockade of these receptors could help to answer this question. However, since the process was driven by the complete block of SERT and subsequent increase in 5-HT levels, we tested the idea that partial blockade of SERT (which can enhance the peristaltic reflex, 44), would cause a detectable increase in the CM contraction-evoked 5-HT release. As with our initial experiments, we found that partial blockade yielded no enhancement in 5-HT released per contraction. Taken together these data suggest that, for small release events, the peak concentrations of 5-HT are not tightly controlled by SERT.
Where is SERT located?
An interesting question is where SERT is located in relation to the EC cells and their targets. An original report by Wade et al. (44) showed that most SERT was located in the crypts. In contrast, more recent studies in mouse, guinea pig, and human have shown SERT immunoreactivity in many epithelial cells throughout the crypt-villus axis (13, 23, 28, 29) though not in the 5-HT-containing EC cells. How then do the actions of SERT affect the 5-HT signal? In this study, we have shown that background levels of 5-HT increase with blockade of SERT, in keeping with work in guinea pig (9). We also have shown an increase in the peak of the compression-evoked 5-HT release and in the time to decay of exogenous 5-HT. However, when we looked at contraction-evoked 5-HT release, we found that, contrary to our expectations, the levels of 5-HT were unchanged. Taken together, our data show that SERT-dependent uptake contributes to the shaping of large 5-HT release events and to the background levels of 5-HT but may contribute less to the shaping of smaller release events. Intriguingly, the role of SERT in shaping 5-HT release may not be fixed. SERT can be up- or downregulated in both the short and long term (40). Furthermore, SERT levels are known to change during pathological states (e.g., 14). Thus the role of SERT in shaping 5-HT release is a dynamic process that warrants further investigation.
To explain our present findings, we have developed a conceptual model shown schematically in Fig. 7. In this model, SERT acts mainly after contraction-released 5-HT has reached its target (neighboring epithelial cells, immunocytes, nerve terminals, etc). However, when large amounts of 5-HT are released, such as following mucosal compression with the CFE or exogenous application of 5-HT, SERT may act to reduce the peak concentrations. Similarly, 5-HT that has acted on its target can now move on and contribute to the background levels of 5-HT, which are also controlled by SERT.
Visceral hypersensitivity is clearly a problem that, in part, depends on release of 5-HT from the EC cells (36) and its subsequent reuptake and catabolism. It would be predicted that increased 5-HT availability would worsen symptoms of hypersensitivity. Paradoxically, clinical use of serotonin-selective reuptake inhibitors does not seem to worsen symptoms (27, 41), suggesting that either receptors for 5-HT have been desensitized (as discussed above), or that blockade of SERT does not increase the peak levels of 5-HT seen by receptors on the sensory nerves. Further experiments looking at 5-HT release simultaneous with nerve recordings of extrinsic afferents (15) or intrinsic afferents (7) should resolve this issue.
Is 5-HT signaling the same across species?
The real-time dynamics of 5-HT release have been characterized with electrochemical techniques in guinea pig and mouse; here we report on 5-HT release for the rat ileum. Broadly speaking, we found that in the rat ileum the levels of compression-evoked 5-HT release (COMP: ∼14 μM) and of background 5-HT (BG: ∼6 μM) were similar to those recorded previously from mouse ileum (COMP: ∼10 μM; BG: ∼6 μM) (4), mouse colon (COMP: ∼7 μM; BG: ∼2 μM) (5), and guinea pig ileum (COMP: ∼12 μM; BG: ∼3 μM) (2); other studies did not routinely convert oxidation current to 5-HT concentration (9, 32). This indicates that the regulation of 5-HT signaling is highly conserved across these species. Furthermore, it suggests that electrochemical techniques for recording 5-HT release are now well established in gastrointestinal research. Indeed, we have recently used these methods to examine 5-HT release and SERT activity in inflamed mouse colon (5) and in surgical samples from human colon (6).
The main finding of this study was that a SERT inhibitor, fluoxetine, reduced 5-HT uptake in the rat intestine. Blockade of SERT enhanced compression-evoked 5-HT release, increased background levels of 5-HT in the mucosa, and prolonged the decay times of exogenous 5-HT pressure ejected onto the surface of the mucosa. These data suggest that SERT plays an important role in the modulation of 5-HT concentrations that reach intestinal 5-HT receptors. Future studies assessing serotonergic signaling in the mucosa could provide insight into novel therapeutics for functional and inflammatory bowel disorders.
This work was supported by the National Health and Medical Research Council (Australia) Grants #510202 (to P. Bertrand) and #455243 and the School of Medical Sciences, University of New South Wales.
The authors thank Dr. Alan Lomax for helpful comments on drafts of this manuscript and Drs. Shaun Sandow, Tim Murphy, and Rebecca Haddock and Ms. Lauren Howitt for the generous gift of rat tissue.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2008 the American Physiological Society