Abstract
Effects of the nonbenzamide 5-hydroxytryptamine4 agonist SDZ HTF 919 on gastrointestinal motility are unclear. Our aim was to assess the in vivo effects on gastrointestinal and colonic transit of radiolabeled residue and on colonic phasic contractility. In six female dogs, transit was measured over a period of 2 days by radioscintigraphy and colonic motility was measured by pneumohydraulic perfusion manometry of the proximal and distal colon. SDZ HTF 919 was administered initially by bolus i.v. infusion, followed by s.c. injection 8 and 16 hr later. Doses tested were 0.03, 0.1 and 0.3 mg/kg, and isotonic saline and vehicle served as controls in each dog. Stomach and small bowel transit was not significantly altered by SDZ HTF 919. Overall, i.v. SDZ HTF 919 accelerated colonic transit during the first 1 hr, compared with controls. These effects were significant even with the lowest dose of SDZ HTF 919. Responses to higher infusion doses were more variable. SDZ HTF 919 did not cause significant changes in quantitative pressure indices, such as amplitude or motor index, in the small bowel or colon. Prolonged postprandial colonic contractions, each lasting >30 sec, were noted after each i.v. agent and were significantly more frequent with the 0.03 mg/kg dose than with control (vehicle) treatment. Thus, SDZ HTF 919 accelerates canine colonic transit in vivo during the first 1 hr after i.v. administration. SDZ HTF 919 appears to be a promising agent for stimulation of mammalian colonic transit.
Motor activity of the colon has several major functions, i.e., mixing of contents to facilitate exchange and absorption of water and electrolytes, transportation of colonic contents caudally, maintenance of colonic bacterial population and defecation (Sarna, 1991). Because of its relative inaccessibility, the colon remains one of the least understood organs of the gastrointestinal tract. However, recent advances in intraluminal recording techniques as well as simultaneous scintigraphic measurement of contents within regions of the colon have allowed a better understanding of colonic function (Karaus and Wienbeck, 1991). Migrating and nonmigrating complexes of electrical and contractile activity during fasting were recognized in the canine colon by some authors (Sarna et al., 1984). Others showed that the colonic interdigestive motor complexes in dogs did not “migrate” and were not related to small bowel migrating motor complexes (Schuurkes and Tukker, 1980). Postprandially, contractile activity is more irregular, and the interval between cyclic motor complexes is prolonged, relative to fasting (Sarna and Lang, 1989). Colonic contractions in dogs show two dominant frequencies, i.e., 0.5 to 2 contractions/min and 4 to 6 contractions/min (Schuurkes and Tukker, 1980; Sarna, 1991) or 7 to 9 contractions/min (Fioramonti et al., 1980).
Several neurotransmitters, including acetylcholine, peptides, amines and other chemicals such as fatty acid-derived substances, modulate colonic contractile activity. Serotonin is thought to play a role in several intestinal functions, including secretion (Beubler et al., 1989), modulation of peristalsis (Buchheit and Buhl, 1993;Briejer et al., 1995) and mediation of visceral afferent signaling, as in the emesis associated with chemotherapy (Cubeddu, 1992). 5-HT-containing neurons are found in the colon of several species, such as rodents and mammals, and four distinct receptor subtypes as well as two less well-characterized categories are recognized, based on operational (agonist and antagonist rank order), transductional (second messenger coupling) and structural (gene and amino acid sequence) criteria (Schmidt and Peroutka, 1989; Saxena, 1994). 5-HT3 and 5-HT4 mechanisms have been more thoroughly characterized in recent years (Schmidt and Peroutka 1989), and their potential applications to the treatment of human disease have been reviewed elsewhere (Camilleri and von der Ohe, 1994). The substituted benzamide cisapride is predominantly a 5-HT4 agonist, with some 5-HT3 antagonist activity; this drug appears to have relatively weak effects on motor function in the human colon (Krevsky et al., 1987), but it has proven efficacious in the treatment of disorders of upper gastrointestinal motility (Wiseman and Fauld, 1994).
SDZ HTF 919 (fig. 1) is a nonbenzamide, potent, partial agonist at 5-HT4 receptors (Pfannkuche et al., 1995). The compound has negligible affinity for 5-HT3receptors. SDZ HTF 919 potently stimulated the peristaltic reflex in isolated guinea pig ileum and colon, through 5-HT4receptors (Grider and Foxx-Orenstein, 1996). The time to achieve maximum drug concentration after i.v. or oral administration of SDZ HTF 919 in dogs is 1 to 2 hr; the absolute bioavailability in dogs is 22 to 27%. The mean terminal half-life of SDZ HTF 919 in dogs is 7 hr, but there is extensive first-pass hepatic metabolism (41–63% in dogs) (de Bruijn, 1993).
Chemical structure and formula of SDZ HTF 919.
Preliminary comparative studies with the benzamide 5-HT4agonist cisapride suggested that the latter was less potent than SDZ HTF 919 (Pfannkuche et al., 1995). Animal studies have revealed stimulatory effects of SDZ HTF 919 on motor functions throughout the gastrointestinal tract. Gastric emptying of solids was stimulated in rats. In the canine small bowel, migrating motor complexes were replaced by intense irregular activity consistent with phase II contractions; in mice, impaired colonic transit was restored to normal (Pfannkuche et al., 1995). However, dose-related effects of SDZ HTF 919 on in vivo measurement of gastrointestinal and, in particular, colonic transit, as well as the phasic contractile responses in the postprandial period, are unclear. The aim of our study was to evaluate the dose-related in vivo effects of SDZ HTF 919 on gastrointestinal and colonic transit and colonic motility in dogs, after their ingestion of a standard meal.
Materials and Methods
Canine model.
Experiments were performed on six healthy, conscious, female dogs, each weighing 18 to 22 kg. Access to the abdominal cavity was achieved by a mini-laparotomy under general anesthesia with Brevital (Eli Lilly, Indianapolis, IN). Atropine (0.6–0.9 mg) was given i.m. preoperatively to minimize shortening of the colon due to handling during the celiotomy. Six manometric catheters were secured in the small bowel and colon (one 1 meter distal to the angle of Treitz, one 1 meter proximal to the ileocecal valve and the rest evenly distributed in the colon). A cecal cannula was placed in the dependent portion of the cecum. During surgery, 500 ml of lactate Ringer solution with 5% dextrose was given i.v. The dogs were allowed 7 to 10 days to recover after surgery.
Gastrointestinal transit.
The procedures for radiolabeling the meal and colon residue have been previously described in detail (Camilleri et al., 1989). One millicurie of [99mTc]pertechnetate was used to radiolabel 10 g of Amberlite 410 pellets (Sigma Chemical Co., St. Louis, MO), which were subsequently mixed with the standard dog meal (2 cans of I/D; Hill’s Pet Nutrition, Inc., Topeka, KS). To label colonic residue, 10 g of Amberlite 120 pellets (Sigma) were radiolabeled with 0.05 mCi of111InCl3 and suspended in 10 ml of saline. This slurry was infused into the fasting dog’s cecum over 5 min, by means of a syringe that was fitted into the cecal cannula. An anteriorly positioned, large-field-of-view, gamma camera with a medium-energy collimator was used to localize pellets in the digestive tract and thereby allow assessment of gastric, small bowel and colonic transit [as previously performed in humans (von der Ohe et al., 1993) and dogs (Wen et al., 1995)]. Windows of 147 keV and 240 keV (±10%) were used to detect and quantitate 99mTc and 111In, respectively. A standardized protocol for scanning times was followed in all experiments (fig. 2). Corrections were applied for isotope decay and time-activity curves generated for gastric, small bowel and colonic regions.
Experimental design.
Manometry of small bowel and colon.
A manometric perfusion system under air pressure control (i.e., pneumohydraulic infusion) (Arndorfer et al., 1977) was used to measure phasic contractility of the small bowel and colon. During these studies, dogs were placed in a Pavlov sling and were allowed to rest on a gamma camera, to facilitate simultaneous measurements of both intestinal phasic pressure activity and transit through the gastrointestinal tract.
Experimental design (fig. 2).
A Latin-square design was used to randomize the delivery of saline, vehicle and three doses of the drug SDZ HTF 919 (0.03, 0.1 and 0.3 mg/kg). SDZ HTF 919 was kindly supplied by Sandoz Pharma Ltd. (Basel, Switzerland); the vehicle for SDZ HTF 919 contained 1-methyl-2-pyrrolidinone. The first dose of the drug was given i.v. and was followed by s.c. injections at 8-hr intervals until all isotope was shown by scintigraphy to have emptied from the dog’s colon.
Each experiment started after a 24-hr period of fasting. Before the ingestion of the radiolabeled meal, the 111In-labeled pellets were infused via the cecal cannula over a period of 5 min. Thirty minutes later the medication was infused i.v., and after an additional 30 min the dog was allowed to ingest the99mTc-containing standardized dog meal. Scans were taken immediately after infusion of the 111In-labeled pellets into the cecum, as well as at the time of medication infusion and ingestion of the meal. Thereafter, a standardized protocol of scanning every 15 min for the first 2 hr and every 30 min for the next 4 hr facilitated measurements of gastrointestinal transit. Additional scans were taken at 7, 8, 25, 35 and 49 hr after the infusion of medication or until the isotope was completely emptied from the colon. The study was approved by Mayo’s Institutional Animal Care and Use Committee.
Data analysis of gastrointestinal and colonic transit.
The percentage of isotope remaining in the stomach was plotted against time, and the lag time (time for 10% of the isotope to be emptied from the stomach) and the slope (or fractional emptying rate) of the gastric emptying curve were calculated as in previous studies, using a linear model (Camilleri et al., 1989). The slope was estimated by linear regression analysis of all data points beyond the lag time. The gastric emptying data over the first 8 hr were also analyzed using a power exponential model (Elashoff et al., 1982; Camilleriet al., 1989) and summarized as k (or the slope from power exponential analysis), β (or the initial shape of the curve) and time for 50% emptying (T1/2). The formula used was as follows: prop t = exp[−(k·t)β]. Small bowel transit was estimated by subtracting the time for 10% of the isotope to empty from the stomach from the time taken for 10% of the99mTc to reach the colon (Greydanus et al., 1990).
Colonic transit was summarized as the geometric center (or weighted average) of counts at each hour during the first 6 hr of the experiment and at 24 hr. The geometric center was calculated as recently suggested by Wen et al. (1995), by dividing the colon into three segments, i.e., proximal colon, left colon and rectosigmoid colon. The stool radioactivity content constituted the fourth region of interest for the purpose of calculating the geometric center. The rectosigmoid colon was identified as that portion of the left colon located below the level of the most proximal point of the colon on the right side of the abdomen. The geometric center was calculated by multiplying the proportion of isotope in each of the four colonic regions by factors of 1, 2, 3 and 4, respectively.
Manometric recordings were analyzed using a peak-finding program that has been adapted in our laboratory from a commercially available program (Vaxlab; Digital Equipment, Boston, MA). Recordings were filtered for respiratory and motion artifacts. After visual analysis of the colonic manometric tracings, two types of phasic pressure activities were analyzed, 1) all phasic contractile activity that had a frequency of >1 contraction/min and 2) contractions of >30-sec duration (i.e., up to 0.5 contraction/min). The average amplitude of contraction and the motility index per hour for each of these two types of contractile activity were calculated, and the mean data for the colonic recordings in each dog were estimated for the 2-hr postprandial period in each experiment. The motility index was calculated by using the following formula: motility index = log(sum of amplitudes × number of contractions + 1).
Statistical analysis.
For each dog under each condition (saline, vehicle and dose of SDZ HTF 919), a summary of the measured responses was computed (e.g., lag duration and T1/2 for gastric emptying and geometric center at 1, 2, 3, 4, 5 and 6 hr for colonic transit). These summary values were then analyzed using a univariate, repeated-measures, analysis of variance to assess the overall effects of vehicle and drug. Pairwise comparisons among treatments were based on the paired t test; corresponding two-sided P values are reported in the text. Significant values were declared for P < .05; values that were <.1 were included in the text or tables. Because SDZ HTF 919 was administered as a bolus i.v. injection and its pharmacokinetics show a relatively rapid time to achieve maximum drug concentration and metabolism, we were particularly interested in the initial effects of the agent and we analyzed the colonic transit data during the first 2 hr as a primary efficacy parameter. Geometric centers of colonic isotope were also compared during 2 to 6 hr and at 24 hr, to determine whether there were delayed effects of the agent. Data in this manuscript are expressed as mean ± S.E.M.
Results
Effects of SDZ HTF 919 on gastrointestinal transit.
There were no differences between transit profiles during saline and vehicle treatment and no overall effects of the medication on gastric emptying (fig. 3, which shows median data derived fromk and β of power exponential analysis for each group, and tables 1 and 2) or small bowel transit (table1). The gastric emptying data appeared to show acceleration with the 0.03 and 0.3 mg/kg doses, but the large variability in the 0.1 mg/kg group reduced the significance of the overall drug effects. The similarity of gastric emptying with 0.03 and 0.3 mg/kg suggested that there was no dose-related effect of the agent.
Summary plots of gastric emptying data in five treatment groups; the plots were derived from the group’s mediank and β estimated by power exponential analysis. There was no significant drug effect, compared with saline or vehicle.
Effect of SDZ HTF 919 on gastric emptying (by linear analysis) and small bowel transit time (SBTT)
Effect of SDZ HTF 919 on gastric emptying (power exponential analysis)
Effects of SDZ HTF 919 on colonic transit.
There were no differences in colonic transit between saline and vehicle treatments. When we compared colonic transit over the first 1 hr (i.e.,from time of instillation of isotope into the colon at 0 hr through the time of the i.v. bolus medication at 30 min and the time of meal ingestion at 1 hr) during treatment with vehicle and each of the three doses of SDZ HTF 919 individually, we were unable to show significant effects of the drug. However, because the effects on colonic transit profiles of the three drug doses were not different by analysis of linear or quadratic trends, we grouped all three doses of the SDZ HTF 919 to assess whether there was an overall drug effect, compared with saline control. This analysis showed a significant drug-time interaction (P = .03) over the first 1 hr after administration of SDZ HTF 919, which was associated with some acceleration (P = .07) of transit during the first 0.5 hr and a significant effect (P = .006) for the whole 1 hr.
The distribution of isotope in the different regions of the colon during the earlier part of the study, 30 min after i.v. infusion of SDZ HTF 919, is shown graphically in figure 4. To evaluate the colonic profiles at subsequent time points, we assessed the paired differences in colonic geometric center at each of the first 6 hr, compared with base line (0 hr), for the experiments performed with vehicle treatment and the three individual doses of SDZ HTF 919. Profiles of colonic transit for one dog and the mean time-activity curves for the six dogs are shown in figure 5. Table3 indicates the P values of the pairwise comparisons of each drug dose and vehicle at different time points. The two higher doses of SDZ HTF 919 appeared to have effects qualitatively similar to those of the lowest dose, 0.03 mg/kg. However, there was a high degree of variation in the early effects, particularly that of 0.3 mg/kg in two of the six dogs. Indeed, from 3 hr onward, the transit profile was significantly faster with 0.3 mg/kg, compared with vehicle.
Distribution of radioactivity in proximal, distal and rectosigmoid colon at 30 min after i.v. SDZ HTF 919. Note the trend toward more distal distribution of isotope with medication but no significant trends related to the dose of medication used.
Mean colonic transit time-activity curves in six dogs. Note that the effect of medication is seen almost immediately and precedes the ingestion of the meal. There is a trend toward a second increment in the aborad movement of chyme at about 7 to 8 hr, when the second s.c. injection of SDZ HTF 919 was administered. GC, geometric center.
Results (P values) of paired t tests for Δ values (geometric center at specified time minus geometric center at base line) comparing vehicle vs. dose of SDZ HTF 919
The second acceleration of colonic transit noted in figure6 coincides with images taken immediately after the second s.c. administration of the agent at 8 hr. However, there were no significant differences in the colonic geometric center at 8 hr in the different groups.
Examples of jejunal, ileal and colonic phasic pressure activity recorded by manometry. Note the contractions of relatively long duration observed at all levels in the colon, particularly after the meal was administered. There appear to be no immediate effects of the drug on overall phasic pressure activity before the ingestion of the meal.
Effect of SDZ HTF 919 on small intestinal phasic pressure activity.
Phasic pressure responses of the small intestine during the first 2 hr postprandially were characterized by irregular contractile activity, typical of the “fed” response, that replaced the fasting migrating motor complexes observed before meal ingestion. Quantitation of phasic contractility showed no drug effects on contraction amplitude (data not shown) or motility indices (table4).
Effect of SDZ HTF 919 on small bowel phasic pressure activity (mean ± S.E.M.)
Effect of SDZ HTF 919 on colonic phasic pressure activity.
Table 5 shows a summary of the effects of saline, vehicle and SDZ HTF 919 on average proximal and distal colonic motility indices. There was an overall increase in the phasic motility index induced by meal ingestion (P = .0006, by paired t test for all groups); however, the increase in the colonic motility index postprandially was no different between the groups, suggesting that the meal rather than the drug led to an increase in the motility index and the drug did not increase this index further.
Effect of SDZ HTF 919 on colonic phasic pressure activity (mean ± S.E.M.)
There were phasic contractions lasting >30 sec (fig. 6) after drug and in the postprandial period. The postprandial increase in the number of waves lasting >30 sec over the fasting period was significant (P < .05) for treatment with SDZ HTF 919 at 0.03 mg/kg, compared with vehicle (table 6).
Effect of SDZ HTF 919 on colonic contractions of >30-sec duration (mean ± S.E.M.)
Discussion
This study demonstrates that the nonbenzamide 5-HT4agonist SDZ HTF 919 significantly stimulates transit of solid residue in the colon. Our study suggests that the higher doses (0.1 and 0.3 mg/kg) are not more efficacious than the lower dose (0.03 mg/kg) of SDZ HTF 919. Our data are generally consistent with many of the previous observations on the effects of this agent on the motor function of the gut in animals. The reason for the lack of a demonstrable dose-related effect of the drug on colonic motility is unclear. Several possibilities should be considered. First, the dose range used may have been at the upper end of the dose-response curve. Second, the drug may act at sites other than 5-HT4 receptors at higher doses, thereby inhibiting its potential to stimulate colonic motility or obscuring its pro-transit effects. There is evidence that SDZ HTF 919 is an agonist at 5-HT1D receptors (Buchheit and Pfannkuche, 1992), an action that has been shown to reduce tone in the mammalian gastrointestinal tract (Coulie, 1996), including feline stomach and human colon (B. Coulie and J. Tak, University of Leuven, Belgium, personal communication). Third, the drug may not act via a receptor-mediated mechanism to stimulate transit. Finally, the drug may not affect colonic transit consistently; this was particularly evident in the large variance of the transit measurements with the 0.1 mg/kg dose in this study.
SDZ HTF 919 stimulated gastric emptying of solids in conscious rats, but it exhibited a bell-shaped dose-response curve when given orally or i.p. in doses similar to those used in our study (de Bruijn, 1993). The ED50 for SDZ HTF 919 in acceleration of liquid emptying from the stomach in conscious guinea pigs was 0.1 μg/kg i.p. (de Bruijn, 1993). In conscious dogs, we were unable to replicate the bell-shaped dose-response curve shown previously.
SDZ HTF 919 stimulated small intestinal motility in guinea pigs, and it induced a phase II-like motility pattern in the upper and lower part of the small intestine of conscious dogs. The functional significance of such contractile activity is unclear; however, we did not observe a significant acceleration in small bowel transit with SDZ HTF 919.
In a pharmacological model of impaired large intestinal motility in mice treated with the alpha-2 adrenergic agonist lidamidine, i.p. administration of 0.01 mg/kg SDZ HTF 919 restored colonic motility to control levels. Similarly, after the i.p. administration of SDZ HTF 919 at doses of 1.0 μg/kg to 1.0 mg/kg, mice developed soft stools (K. H. Buchheit, personal communication). Our data suggest that this effect on stool consistency may result at least partly from acceleration of colonic transit.
Our data also confirmed the importance of 5-HT4 receptors in the control of canine colonic motor function. Fink and Friedman (1960) previously demonstrated the differential effects of serotonin on proximal and distal colon. Sleisenger et al. (1959)demonstrated that intra-arterial administration of serotonin stimulated the colon of intact anesthetized dogs. Subsequently, identification of receptor subtypes for serotonin suggested the existence of a type 4 receptor subtype in the digestive tract, chiefly on cholinergic nerves (Elswood et al., 1991). The in vitro work of several groups using circular muscle strips from guinea pig ascending colon (Elswood et al., 1991; Briejer et al., 1993; Kojima and Shimo, 1995) or human colon (Tam et al., 1994) and in vivo studies in the dog colon (Kadowakiet al., 1993) confirm the stimulation of colonic contractility with 5-HT4 agonists. Recently, Grider and Foxx-Orenstein (1996) showed that SDZ HTF 919 can trigger the peristaltic reflex in rat and guinea pig colon via5-HT4 receptor activation. In humans, the substituted benzamide cisapride appears to predominantly stimulate proximal colonic emptying in health (Krevsky et al., 1987) and in the disease colonic inertia (Krevsky et al., 1989), which is characterized most frequently by delayed proximal colonic transit (Stivland et al., 1991). We have confirmed the stimulation of canine colonic transit, in a chronic in vivo dog model, by the nonbenzamide 5-HT4 agonist SDZ HTF 919.
The motor mechanisms responsible for colonic propulsion are unclear; high-amplitude propagated phasic contractions are thought to be responsible for mass movements, but they are relatively infrequent and were not observed during the 2 hr after SDZ HTF 919 infusions. Contraction, amplitude and motility indices were not different postprandially among treatment groups. Hence, the motor parameters, including possible effects of SDZ HTF 919 on colonic tone, which may influence colonic propulsion, require further evaluation. Thus, colonic tone is clearly increased in human models of accelerated transit, such as carcinoid diarrhea (von der Ohe et al., 1993), and decreased in chronic colonic megacolon (von der Ohe et al., 1994) and in some patients with slow-transit constipation (O’Brienet al., 1996).
In summary, SDZ HTF 919 appears to be a promising agent to stimulate motility and transit in the colon and may also stimulate gastric emptying, although the latter effects are not as consistent as are effects on colonic transit. Future studies are needed to further elucidate the motor mechanisms resulting in the accelerated transit. Ultimately, the therapeutic use of this agent will require similar dose-response pharmacodynamic studies in humans and appropriate placebo-controlled trials in carefully selected patients with objectively delayed transit, particularly those with delays in proximal colonic emptying, which are quantifiable using radioscintigraphy and a delayed-release capsule that delivers isotopically labeled solid residue to the ileocolonic junction (Proano et al., 1990;Stivland et al., 1991).
Acknowledgments
We thank Cindy Stanislav for secretarial assistance.
Footnotes
-
Send reprint requests to: Michael Camilleri, M.D., Gastroenterology Research Unit, Alfred 2–435, Mayo Clinic, 200 First St. S.W., Rochester, MN 55905.
-
↵1 This study was supported in part by a research grant from Sandoz Pharma Ltd.
- Abbreviation:
- 5-HT
- 5-hydroxytryptamine
- Received June 7, 1996.
- Accepted November 29, 1996.
- The American Society for Pharmacology and Experimental Therapeutics
