| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
MUCOSAL BIOLOGY
Department of Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine, University of California, Davis, California
Submitted 21 February 2006 ; accepted in final form 25 April 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-methyl glucose (nonmetabolizable substrate of SGLT-1 and -3) but not 2-deoxy-D-glucose (substrate for GLUT-2) or isoosmotic mannitol. Gastric motility and intestinal fluid secretion, measured in anesthetised rats, were significantly inhibited and stimulated, respectively, by duodenal glucose but not galactose, which has a much lower affinity for SGLT-3 than glucose. Duodenal glucose but not galactose stimulated the release of 5-HT into mesenteric lymph and stimulated the discharge of duodenal vagal afferent fibers. mRNA for SGLT-3 was identified in the duodenal mucosa. Together these data suggest that detection of glucose in the intestine may involve SGLT-3, possibly expressed by enterochromaffin cells in the intestinal mucosa, and release of 5-HT.
gastric emptying; intestinal fluid secretion; vagal afferents; serotonin; intestinal feedback
There is evidence that release of GIP and GLP-1 in response to glucose involves a mechanism similar to insulin release from the pancreas, via generation of ATP and opening of ATP-sensitive K+ (KATP) channels in the membrane (25). However, a different mechanism for glucose-induced release of 5-HT from enterochromaffin (EC) cells has been suggested. We have shown that in an EC cell line, BON, release of 5-HT stimulated by glucose is mimicked by -methyl glucose, a nonmetabolizable analog of glucose, and inhibited in the presence of the competitive blocker of SGLT, phloridzin (14). This suggests a mechanism involving transport of glucose via SGLT in glucose-mediated release of 5-HT in the gut. Furthermore, glucose-induced inhibition of gastric motility in awake rats is attenuated by coinfusion of phloridzin (24), suggesting that transport of glucose into enterocytes via SGLT is important in the ability of the gut to detect the presence of luminal glucose and subsequent 5-HT-mediated activation of the vagal afferents. The SGLT family of transporters consists of SGLT-1, SGLT-2, and SGLT-3. SGLT-1 and SGLT-3 are found in the gut. In the intestine, the predominant form is SGLT-1, which is responsible for glucose and galactose absorption from the lumen of the gut. SGLT-1 is a high-affinity transporter for both D-glucose and galactose; SGLT-3 is a low-affinity transporter for D-glucose. We noted that detection of extracellular glucose by BON cells occurred at high concentrations (25–50 mM), suggesting involvement of SGLT-3 rather than SGLT-1. Recently, further evidence has been obtained to support the idea that SGLT-3 may serve as a glucose sensor in smooth and skeletal muscle (4). Human SGLT-3 (hSGLT-3), which has 70% homology with hSGLT-1, is expressed in smooth and skeletal muscle and is possibly localized to neurons in the enteric nervous system. In hSGLT-3-transfected oocytes, glucose was not transported across the membrane but caused depolarization of the membrane potential with a Km of 20 mM; this depolarization was specific for D-glucose (galactose had no effect) and was blocked by phloridzin.
The specific aim of the present study was to determine whether SGLT-3 is involved in glucose detection in the intestinal lumen. The sensory transduction mechanism may lie at or beyond the apical membrane and may reside within epithelial cells, possible the EC. To test our hypothesis, we measured changes in a number of physiological parameters in response to different analogs of glucose and to galactose. The first specific aim was to determine the role of a member of the SGLT family of transporters by measuring the effects of different glucose analogs on inhibition of gastric emptying. The second specific aim was to determine whether the sensing of glucose has the characteristics of SGLT-1 or SGLT-3; we measured inhibition of gastric motility, intestinal fluid flux, vagal afferent activity, and release of 5-HT in response to glucose and compared the response with that obtained with galactose. To determine whether SGLT-3 is expressed in the rat intestinal mucosa, we measured mRNA by real-time quantitative PCR.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Experiments were performed with Sprague-Dawley rats (Harlan, San Diego, CA) of initial weight 200–280 g, maintained on regular laboratory chow. Rats were fasted overnight but allowed water ad libitum before all surgical and experimental procedures. Animal experiments were reviewed and approved by the UC Davis Institutional Animal Care and Use Committee.
Measurement of Gastric Emptying
Surgical procedures. The procedure to place gastric and duodenal cannulas has been described in detail elsewhere (12). Briefly, rats (n = 6) were anesthetised with pentobarbital sodium (50 mg/kg ip, Nembutal; Abbott Laboratories, North Chicago, IL). A small stainless steel Thomas cannula was inserted into the forestomach, exteriorized through the abdominal wall, and capped. The duodenal cannula [polyethylene tubing (PE-90), Intramedic; Clay Adams, Parsippany, NJ] was inserted into the duodenum 1–2 cm distal to the pylorus. Rats were allowed to recover for 2 wk before being used in experiments and were used for a period of up to 3 mo after surgery. In the period during which rats were recovering from surgery, they were accustomed to light restraint in Bollman cages. The duodenal cannula was flushed daily with 0.9% saline and plugged with petroleum jelly.
On experimental days, fasted rats were placed in Bollman cages, the gastric cannula was opened, and any residual gastric contents flushed out with warm 0.9% saline. The stomach was then allowed to drain freely for 30–45 min. Three milliliters of 0.9% saline containing the nonabsorbable marker phenol red (60 mg/l) were instilled into the stomach, and the cannula was closed. After 5 min, the contents of the stomach were collected, the volume was measured, and the contents were centrifuged. The concentration of phenol red was determined in the instilled and recovered fluid, and the rate of gastric emptying was calculated using the method of Hunt and Knox (13), which calculates the volume emptied from the stomach including any gastric secretions. Data for gastric emptying are expressed as the percentage of the liquid emptied after 5 min. Values are means ± SE.
Experimental protocols. Gastric emptying was measured in two consecutive control periods with no perfusion of the duodenum, with 10 min between procedures. Perfusion of the duodenum with glucose analogs or mannitol (330, 660, or 990 mM; total amounts 30, 60, or 90 mg) at a rate of 0.05 ml/min was started and continued for 10 min; gastric emptying was measured in the 5- to 10-min period; in some experiments, gastric emptying was measured at 20–25 and 35–40 min after duodenal perfusion of glucose to determine the time course of the inhibitory response to glucose. Rats were randomized for treatment using a Latin square design; only one test perfusion was performed on each day, and animals were used no more than every third day. Not all rats received each treatment because the full protocol could not be completed as one rat was euthanized because of a leaking gastric fistula.
Measurements of Gastric Motility in Anesthetised Rats
This method has been published previously (22). Briefly, rats (200–250 g) were anesthetised with urethane (1.25 g/kg ip; Sigma, St. Louis, MO), and a catheter was placed into the trachea to ensure a clear airway [PE-240, 1.67 mm inside diameter (ID), 2.42 mm outside diameter (OD)]. The abdomen was opened, the pylorus was ligated, gastric contents were gently flushed with warm 0.9% saline through an incision in the forestomach, and a catheter (Silastic, 2 mm ID, 3.2 mm OD) was placed through the incision to measure intraluminal gastric pressure (IGP). A catheter (PE-90) was placed in the proximal duodenum for intestinal infusion and in the distal duodenum (5–6 cm distal to the pylorus) to drain intestinal perfusates. After a recovery period of about 45–60 min, the stomach was filled with warm saline (0.9%, 1 ml) and kept under continuous pressure of 5–6 cmH2O for 60 min to normalize baseline IGP. IGP was displayed and collected online for the duration of the experiment. Changes of IGP were measured and analyzed as the maximal change (decrease or increase) of intraluminal pressure (cmH2O).
Experimental protocols. Inhibition of gastric motility was measured during duodenal perfusion with saline and compared with intraduodenal perfusion with glucose or galactose (10 min at 0.05 ml/min; 660 mM, 60 mg and 990 mM, 90 mg, respectively); the order of perfusion was randomized.
Measurement of Intestinal Fluid Flux in anesthetised Rats
Net ileal fluid absorption or secretion was measured in anesthetised rats (pentobarbital sodium, 60 mg/kg ip) using modifications of a previously described method (20). After a midline abdominal incision, a small incision was made into the terminal ileum 
2 cm rostral to the ileocecal junction. A preweighed 2.5% agarose cylinder (type II-A: 140 mM NaCl/5 mM KCl; diameter 4 mm, length 3 cm) was inserted into the lumen, and the intestine was sutured closed. After 90 min, the agarose cylinder was removed and reweighed; the difference between the initial and final weight of the agarose cylinder was taken as a representative measure of the net absorption/secretion in that segment of intestine. Data are expressed as microliters of fluid per centimeter squared per hour.
Experimental protocols. Before anesthesia, rats were randomized into three groups: water, 1 ml; glucose (660 mM, 120 mg and 990 mM, 180 mg); or galactose (990 mM, 180 mg) was administered by oral gavage, and rats were anesthetised after 15 min.
Collection of Mesenteric Lymph
Surgical preparation and lymph collection was performed as previously described (9). Briefly, rats (300–350 g) were anesthetised with methohexital sodium (60 mg/kg ip, Brevital; Jones Pharma, St. Louis, MO). The mesenteric lymph duct was cannulated with a polyvinyl chloride tube (Medical Grade, 0.50 mm ID, 0.80 mm OD; Dural Plastics and Engineering, Dural, Australia), fixed in place with a drop of ethyl cyanoacrylate glue (Krazy Glue; Elmers Products, Columbus, OH), and externalized through a stab wound in the right flank. A second cannula (Silastic, 1 mm ID, 2.15 mm OD) was placed in the duodenum, secured in place with a silk suture and additionally fixed with a drop of ethyl cyanoacrylate glue, and externalized. After surgery, rats were placed in Bollman cages and were allowed to recover, and a maintenance solution (0.2 M glucose, 145 mM NaCl, and 4 mM KCL) was perfused continuously through the duodenal cannula at a rate of 3 ml/h for 24 h during the recovery period to ensure hydration and electrolyte and nutritional maintenance.
Control lymph was collected for 60 min during perfusion of the intestine with physiological saline, followed by collection for a further 60 min during intestinal perfusion with 660 mM glucose or galactose (0.05 ml/min, 90 mg). Lymph was collected in ice-chilled tubes, centrifuged, and stored at –80°C until they were assayed for 5-HT using a commercial ELISA kit (Beckman Coulter).
Recording of Duodenal Vagal Afferent Fiber Discharge
The technique has been published previously (10). Briefly, a segment of the thoracic esophagus, stomach, and proximal duodenum was removed from anesthetised rats and immersed in oxygenated modified Ringer solution. The pancreas and stomach were removed except for the pylorus and adjacent antrum, and the subdiaphragmatic dorsal vagus nerve was identified. A catheter was placed into the gastroduodenal artery, and the hepatic, left, and right gastric arteries were tied. The segment was placed into an organ bath and perfused continuously with oxygenated Ringer solution at 2.0–2.5 ml/min flow rate, and the temperature of the organ bath was maintained at 33 ± 1°C.
Action potentials were recorded from the dorsal gastric vagal nerve trunk; action potentials were sent to a preamplifier (DAM-6 X100, 100–10 kHz band-pass filter; World Precision Instruments, Sarasota, FL), displayed on a digital storage oscilloscope (model 2211, Tektronix), and recorded on-line using a digital tape recorder (Sony high-density linear A/D D/A optical digital audio tape deck, DTC-700). In addition, unit potentials were simultaneously sent to a PC computer equipped with an A/D board (DT2831; Data Translation, Marlboro, MA). In addition, unit potentials were simultaneously sent to a A/D module (Micro 1401 MK2; CED, Cambridge, UK). Electrophysiological recording from duodenal vagal afferents was started 30 min after the preparation during stable recording conditions. Units were selected by the presence of spontaneous activity in the nerve strand. Each nerve strand containing spontaneously active units was tested for the response to an intra-arterial injection of 2-methyl 5-HT (10 pmol). In seven preparations, nerve strands containing 5-HT-responsive units were tested to determine their response to intraduodenal infusion of glucose and galactose (0.05 ml, 10 min, 50 mM).
With the use of the acquisition module of SPIKE2 impulse analysis software (SPIKE2 version 5, CED), units within upper and lower threshold settings of the amplifier were acquired on-line onto the hard drive of a computer (PC). Single units were discriminated off-line from the multi-unit recordings based on their shape. The response pattern of different units was analyzed and displayed separately. Response magnitudes were normalized by a response quotient (RQ), calculated as the ratio of the spike count after treatment to that of 5 min of spike counts before. An RQ value of >1.20 was taken to indicate an excitatory response.
Real-Time Quantitative PCR
These studies were done in collaboration with the Lucy Whittier Molecular Core Facility at the University of California, Davis, School of Veterinary Medicine. Tissue was taken from anesthetised rats. The ileum was opened along the antimesenteric border and placed mucosal side up on an ice-cold glass plate. The mucosa was scraped away from the muscle. The mucosal and muscle samples were lysed in 1x ABI lysis buffer (Applied Biosystems) and stored at –20°C until processing. Total RNA was extracted from the lysates using an automated nucleic acid workstation (model 6700, Applied Biosystems) according to the manufacturer's instructions. Complementary DNA was synthesized using 100 U SuperScript II (Invitrogen, Carlsbad, CA), 600-ng random hexadeoxyribonucleotide [pd(N)6] primers, 10 U RNaseOut, and 1 mM dNTPs (Life Technologies) in a final volume of 40 µl. The reverse transcription reaction proceeded for 50 min at 42°C. After addition of 60 µl of water, the reaction was terminated by heating for 5 min to 95°C and cooling on ice.
Each PCR reaction contained 400 nM of each primer, 80 nM of the TaqMan probe and commercially available PCR mastermix (TaqMan universal PCR mastermix, Applied Biosystems) containing 10 mM Tris·HCl (pH 8.3), 50 mM KCl, 5 mM MgCl2, 2.5 mM deoxynucleotide triphosphates, 0.625 U AmpliTaq Gold DNA polymerase per reaction, 0.25 U AmpErase UNG per reaction, and 5 µl of the diluted cDNA sample in a final volume of 25 µl. The samples were placed in 96-well plates and amplified in an automated fluorometer (ABI PRISM 7700 sequence detection system, Applied Biosystems). Amplification conditions were 2 min at 50°C, 10 min at 95°C, 40 cycles of 15 s at 95°C, and 60 s at 60°C.
For each target gene, two primers and an internal, fluorescent-labeled TaqMan probe [5' end, reporter dye FAM (6-carboxyflourescein); 3' end, quencher dye TAMRA (6-carboxytetramethylrhodamine)] were designed using Primer Express software (Applied Biosystems). SGLT-3: forward primer (SGLT-3–38f), 5'-CCCCAGAGCCACCTCCAT-3'; reverse primer (SGLT-3–112r), 5'-CCAGAAAATAGATGACAATGACTGAGAT-3'; TaqMan probe (SGLT-3–58p), 5'-FAM-TCTGACCACATCCGAAATGCTGCTG-TAMRA-3'. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH): forward primer (GAPDHf), 5'-GCACCACCAACTGCTTAGCAC-3'; reverse primer (GAPDH-557r), 5'-TCTTCTGGGTGGCAGTGATG-3'; TaqMan probe (GAPDH-509p) 5'-FAM-TCGTGGAAGGACTCATGACCACAGTCC-TAMRA-3'. SGLT-1: Assay-on-Demand, no. Hs00165793_m1, Applied Biosystems. The length of the PCR products was held very short (between 73 and 170 bp) to enable high-amplification efficiencies.
Final quantitation was done using the comparative CT method (User Bulletin no. 2, Applied Biosystems) and is reported as relative transcription or the n-fold difference relative to a calibrator cDNA (control group). In brief, the endogenous control feline GAPDH was used to normalize the cytokine gene signals. The normalized values (
CT) were calibrated against the normalized values of the control group for each target gene (CT). The relative linear amount of target molecules relative to the calibrator was calculated as 2–CT. Therefore, all gene transcription data are expressed as an n-fold difference relative to the calibrator.
Data analysis. Values were compared using one-way ANOVA followed by a Tukey test and considered significantly different if P < 0.05. Differences in mesenteric lymph concentrations were determined using a two-tailed unpaired Student's t-test.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Under control conditions during intraduodenal perfusion of saline (0.05 ml/min for 10 min), 76 ± 3% of the contents of the stomach emptied in 10 min (n = 6). Glucose produced a dose-dependent decrease in gastric emptying (Fig. 1A); intraduodenal perfusion of glucose produced a significant inhibition of gastric emptying at doses of 60 and 90 mg but not 30 mg (P < 0.001 vs. saline, n = 5). Perfusion of the intestine with glucose (90 mg) produced a short-lasting decrease in gastric emptying; gastric emptying returned to control levels by 25 min after the start of infusion of glucose [glucose, 5–10 min vs. 20–25 min; 47 ± 2 vs. 72 ± 2, not significant (NS), n = 5]. Intraduodenal perfusion with mannitol (90 mg) had no significant effect on gastric emptying.
|
-methyl glucose (90 mg; glucose analog that is a substrate for SGLT and is not metabolized) produced an inhibition of gastric emptying that was not significantly different from that produced by glucose (90 mg; NS, n = 5, Fig. 1B). In contrast, 2-deoxy-D-glucose (90 mg; a substrate for GLUT-2, the basolateral glucose transporter, but not SGLT) had no significant effect on gastric emptying. Effect of Glucose and Galactose on Gastric Motility
As previously reported (22), perfusion of the duodenum with glucose (60 or 90 mg) produced a significant decrease in gastric motility in anesthetised rats compared with perfusion of saline (Fig. 2). Tonic intragastric pressure decreased by 0.65 ± 0.11 cmH2O during perfusion with 60 mg of glucose (n = 6; P < 0.01). In contrast, intraduodenal perfusion with galactose (60 and 90 mg) had no significant effect in inhibiting gastric motility; rather, at the 90-mg dose, galactose increased gastric motility (change in IGP; 0.6 ± 0.09 cmH2O; P < 0.05).
|
In anesthetised rats treated with vehicle (1 ml water by gavage 15 min before induction of anesthesia), the net fluid movement across the ileum was absorptive (n = 16, Fig. 3). Treatment with glucose (120 and 180 mg) or -methyl glucose (180 mg) produced a significant decrease in intestinal fluid absorption (P < 0.01, n = 5–8 in each group). However, treatment with galactose (180 mg) had no significant effect on intestinal fluid flux (NS, n = 5).
|
In awake rats, basal release of 5-HT was measured in mesenteric lymph (Fig. 4). The release of 5-HT was significantly increased by perfusion of the intestine with glucose (15-min output: 1.77 ± 0.46 vs. 4.08 ± 1.12 pmol saline vs. 660 mM glucose, P < 0.05; 0.05 ml/min for 60 min; n = 7). The peak increase in the output of 5-HT occurred in the first 15 min (equivalent to perfusion of 90 mg glucose) and then decreased toward the baseline despite continued glucose perfusion. In contrast, galactose had no significant effect on 5-HT concentration in mesenteric lymph (15-min output: 2.01 ± 0.39 vs. 3.00 ± 0.25 pmol, saline vs. galactose 660 mM, NS, n = 6). The peak 5-HT release was higher in rats in which the intestine was perfused with glucose compared with galactose perfusion (change from basal, glucose vs. galactose, 2.64 ± 0.75 vs. 0.66 ± 0.46 pmol); however, this did not reach statistical significance (P < 0.055; n = 7 and 6, glucose vs. galactose).
|
Data were obtained from seven fibers of the vagus nerve trunk, and single-fiber activity was obtained from 23 units. Of these 23 units, 18 responded with an increase in firing in response to intra-arterial injection of the 5-HT3 receptor (5-HT3R) agonist, 2-methyl 5-HT (mean RQ = 1.54 ± 0.5). Of these 18 units, 7 responded to intraduodenal perfusion of glucose (mean RQ = 1.32 ± 0.2); however, no glucose-sensitive units responded to intraduodenal galactose (mean RQ = 1.05 ± 0.4, n = 6 glucose-sensitive fibers).
Real-Time PCR for SGLT-1 and SGLT-3
The relative gene transcription for both SGLT-1 and SGLT-3 was determined by real-time PCR in the rat intestinal mucosa. As expected, SGLT-1 expression was high in the intestinal mucosa of the duodenum and the ileum but not in the colon. We also detected SGLT-3 mRNA in these mucosal samples (Fig. 5).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-methyl glucose) that are substrates for the SGLT family of transporters. Metabolism of monosaccharide within the epithelial cell layer does not appear to be required, because 3-O-methyl glucose, a nonmetabolizable substrate of SGLT, was as effective to inhibit gastric motility as glucose and -methyl glucose. In addition, it is unlikely that the mechanism involves a postepithelial mechanism, because 2-deoxy-D-glucose, a substrate for GLUT-2 responsible for the transport of glucose across the basolateral membrane of enterocytes, had no effect on the rate of gastric emptying. This supports a role for a member of the SGLT family of transporters in mediating the response to glucose in the intestine. These data are consistent with other published observations; we previously demonstrated that intestinal perfusion of glucose inhibits gastric motility via a phloridzin-sensitive pathway in anesthetised rats, suggesting that SGLT may play a role in the initiation of the response to glucose (24). Recent evidence has suggested that hSGLT-3, a member of the SLC5 family of transporters, might be a glucose sensor, because it binds but does not efficiently transport glucose (4). In addition, glucose induces membrane depolarization in oocytes transfected with SGLT-3, suggesting that the transporter may be associated with intracellular signaling events. Human and pig SGLT-3 are the only species to be fully characterized, and each has a much lower affinity for galactose than glucose. We used this observation as the basis to discriminate between SGLT-1 and SGLT-3; we determined the effect of glucose vs. galactose on inhibition of gastric motility, stimulation of vagal afferent activity, release of 5-HT, and alteration of intestinal fluid flux. Inhibition of proximal gastric motility, shown to be important in the regulation of gastric emptying, was observed in response to glucose but not galactose. Glucose-mediated inhibition of gastric motility is mediated via a vago-vagal reflex pathway involving 5-HT3Rs located on vagal afferent nerve terminals in the intestinal mucosa (2, 23). Glucose activates vagal afferent fiber discharge via a release of 5-HT and activation of the 5-HT3R (30). Galactose was ineffective in stimulating vagal afferent fiber activity. Because galactose has a much lower affinity for both pig and human SGLT-3 than glucose, yet the same affinity for SGLT-1, these results suggest that the mechanism involved in detection of glucose in the intestinal mucosa involves SGLT-3 rather than SGLT-1. It should be noted, however, that the characteristics of rat SGLT-3 are not known and caution should be exercised in interpretation of structure-function relationships of proteins across species.
Instillation of glucose solutions into the intestine will release 5-HT from EC in the wall of the gut. Release of 5-HT from EC results in activation of vagal afferents and vagal reflex pathways in response to intestinal glucose that is mediated via 5-HT3Rs, but the exact mechanism between the lumen and the nerve terminals remains to be elucidated (21). Using the EC-like cell line BON, we demonstrated that release of 5-HT in response to glucose is also dependent on a SGLT protein (14). We concluded that this was SGLT-1; BON cells express mRNA and the protein for SGLT-1, and release of 5-HT from BON cells was phloridzin sensitive. However, in light of the new findings suggesting that the glucose sensor may be SGLT-3, two observations suggest that release of 5-HT may be dependent on SGLT-3 rather than SGLT-1. Application of phloridzin alone stimulated release of 5-HT from BON cells, suggesting that binding of glucose or phloridzin to SGLT-3 may be the sufficient stimulus to elicit 5-HT release. Second, the concentrations of glucose required to elicit 5-HT release from BON cells were in excess of the Km for glucose binding to SGLT-1. Recent evidence suggests that, at least for human SGLT-3, the affinity for binding of glucose is considerably lower than hSGLT-1 (4). Therefore, in BON cells it is possible that SGLT-3 mediates 5-HT release in response to glucose. We have preliminary evidence that BON cells express mRNA for SGLT-3 (Freeman and Raybould, unpublished observations). In the present study, we determined whether release of 5-HT in response to intestinal glucose perfusion in vivo has the characteristics consistent with a role for SGLT-3 in EC. We measured concentrations of 5-HT in mesenteric lymph; determining 5-HT concentrations in mesenteric lymph has several advantages over plasma; mesenteric lymph represents the interstitial fluid in the gut wall, and the lack of platelets allows accurate determination of 5-HT concentrations. Glucose was more effective than galactose in releasing 5-HT, consistent with the known properties of SGLT-3. mRNA for SGLT-3 was detected in the intestinal mucosa, supporting a role for SGLT-3 in the detection of glucose and the release of 5-HT (Fig. 5). However, it remains to be determined whether this involves an enteric reflex, consistent with the localization of SGLT-3 to enteric neurons in human tissue (4) or whether enterochromaffin cells express SGLT-3. The observation that SGLT-3 is found in the intestinal mucosa provided some evidence that the latter may be the case, and EC may be an important link in this signaling mechanism. Experiments with the EC line BON would suggest that glucose may have a direct effect to stimulate release of 5-HT, although the intracellular signaling pathways are not known but may involve membrane depolarization as shown in tranfected oocystes (4).
In immunochemical studies of the human intestine, expression of SGLT-3 was localized to the neurons in the myenteric plexus; preliminary observations suggest that these were cholinergic neurons (4). Activation of a cholinergic reflex pathway in the intestine by 5-HT and 5-HT3Rs has been shown to stimulate fluid secretion (20). Therefore, we determined the effect of glucose, given by oral gavage, on fluid flux in the distal ileum. Under basal conditions in anesthetised rats, the intestine undergoes net absorption of fluid. The direct effect of glucose on net luminal fluid balance is to promote absorption, either by passive movement or by water transport along with glucose and sodium (7). The effect of glucose in the proximal gut on fluid flux in the distal gut is to increase fluid secretion, presumably to bring intestinal contents to isotonicity in the first part of the postprandial period. It is well-established that intestinal fluid absorption and secretion can be regulated by the enteric nervous system and that 5-HT acting at 5-HT3Rs plays a role (8, 28). Activation of 5-HT3Rs on nerve terminals depolarizes myenteric neurons, some of which have been identified as intrinsic primary afferent neurons (1, 29). In the present study we measured effects of luminal glucose on fluid secretion in the terminal ileum. In this model, intestinal fluid flux is stimulated by agonists of 5-HT3R (Raybould and Freeman, unpublished observations), and secretion stimulated by mechanical stimulation of the mucosa or by abdominal irradiation is reduced by 5-HT3R antagonist (20). In the present study, the terminal ileum was absorptive under basal conditions and glucose produced a decrease in intestinal fluid absorption, the magnitude of which was dependent on the amount of glucose administered. In contrast, galactose had no effect on intestinal fluid absorption. These data further support the role of SGLT-3 rather than SGLT-1 in the detection of glucose in the proximal gastrointestinal tract.
There are other glucose-sensitive cells within the body that play a role in glucose homeostasis, including pancreatic 
-cells, hypothalamic neurons, and enteroendocrine cells. The release of GIP and GLP-1 in response to glucose is mediated by the regulation of KATP channels (27). However, more recent evidence suggests that other mechanism may be involved, including SGLT-1 and G protein-coupled sweet-tasting receptors (6, 11). There is evidence that at least the glucose-induced release of GLP-1 may involve SGLT-1, although similar to -cells of the pancreas, it is likely to also involve KATP channels. This mechanism is unlikely to be the case in release of 5-HT, because nonmetabolizable analogs of glucose are able to induce release of 5-HT and to activate vago-vagal reflexes.
In summary, we have obtained evidence to support the hypothesis that SGLT-3 is involved in glucose-sensing in the wall of the gut. Sensing of glucose at the brush-border membrane of the intestine is important to mount appropriate reflex control of gastrointestinal secretion and motility in the postprandial period.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
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.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-cells: a model for the study of other glucose-regulated cells in gut, pancreas, and hypothalamus. Diabetes 50: 1–11, 2001.This article has been cited by other articles:
|
|
 |
|
  M. Kidd, I. M. Modlin, B. I. Gustafsson, I. Drozdov, O. Hauso, and R. Pfragner Luminal regulation of normal and neoplastic human EC cell serotonin release is mediated by bile salts, amines, tastants, and olfactants Am J Physiol Gastrointest Liver Physiol, August 1, 2008; 295(2): G260 - G272. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Zheng and H.-R. Berthoud Neural Systems Controlling the Drive to Eat: Mind Versus Metabolism Physiology, April 1, 2008; 23(2): 75 - 83. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. T. Stearns, A. Balakrishnan, J. Rounds, D. B. Rhoads, S. W. Ashley, and A. Tavakkolizadeh Capsaicin-sensitive vagal afferents modulate posttranscriptional regulation of the rat Na+/glucose cotransporter SGLT1 Am J Physiol Gastrointest Liver Physiol, April 1, 2008; 294(4): G1078 - G1083. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Wan and K. N. Browning D-Glucose modulates synaptic transmission from the central terminals of vagal afferent fibers Am J Physiol Gastrointest Liver Physiol, March 1, 2008; 294(3): G757 - G763. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Tome From gut nutrient sensing to nutrient perception: a cooperative role involving CCK and 5-HT? Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2007; 292(3): R1061 - R1062. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Bezencon, J. le Coutre, and S. Damak Taste-Signaling Proteins Are Coexpressed in Solitary Intestinal Epithelial Cells Chem Senses, January 1, 2007; 32(1): 41 - 49. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
