Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Gastric emptying (GE) is delayed in diseases such as diabetes mellitus (Horowitz et al., 1986, 1987; Brogna et al., 1989; Janssens et al., 1990; Schmid et al., 1991; Peeters et al., 1992) and functional dyspepsia (Corinaldesi et al., 1987; Davis et al., 1988). Gastric emptying also is delayed in as many as 50% of patients after truncal vagotomy, antrectomy, and Roux-Y gastrojejunostomy (Hocking et al., 1981; Vogel et al., 1983; Carlson et al., 1991). The common factor in these conditions is upper gastrointestinal motor dysfunction. Therefore, a prokinetic drug with a primary site of action in the upper gut (stomach, and first half of the small intestine) might provide benefit in treatment of these disorders.

Motilin, a 22-amino acid peptide, has been shown to accelerate GE in normal subjects (Christofides et al., 1979, 1981) and in patients with diabetic gastroparesis (Schmid et al., 1991; Peeters et al., 1992) when given as an i.v. infusion. Motilin is thought to accelerate GE by increasing the force of postprandial antral contractions and by promoting coordination between antral and duodenal motor activity (Annese et al., 1992). Erythromycin (ERY), a 14-membered macrolide with antibiotic activity, binds to motilin receptors in the antrum and duodenum (Peeters et al., 1989). It also stimulates contractile activity similar to motilin (Annese et al., 1992). Additionally, ERY accelerated GE in normal subjects and in diabetic gastroperetic patients to the same extent as motilin (Janssens et al., 1990).

ABT-229 (8,9-anhydro-4"-deoxy-3'-N-desmethyl-3'-N-ethylerythromycin B-6,9-hemiacetal) is a more potent synthetic derivative of ERY with no antibiotic activity (Lartey et al., 1995, Faghih et al., 1998). As with ERY, ABT-229 is thought to stimulate contractile activity through activation of motilin receptors (Clark et al., 1999). Previously, ABT-229 has been reported to stimulate contractile activity of the antrum and small intestine in fasted conscious dogs (Faghih et al., 1998). The objective of this study was to determine the effect of ABT-229 on GE of a solid meal as well as postprandial motor activity of the antrum and duodenum, and to compare them with the effects of cisapride (CIS) and ERY.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Experiments were conducted on six conscious beagle dogs, weighing 9.5 to 11.3 kg and trained to stand in a sling. The procedures used in this study were approved by the Institutional Animal Care and Use Committee of Abbott Laboratories, Abbott Park, IL.

Surgical Preparation. After an overnight fast, dogs were initially anesthetized with thiopental, 20 mg/kg i.v., and prepared for surgery in accordance with standard procedures. Isoflurane (1-1.5%) delivered via a semiclosed system was used during the surgical procedure as a general anesthetic. Through a midventral laparotomy, a silicon catheter (3.2-mm o.d. × 1.6-mm i.d.) was placed intraluminally with its tip 2 cm distal to the pylorus. A stainless steel collection cannula was placed 20 cm distal to the pylorus. Additionally, two strain gage force transducers (RB Products, Stillwater, MN) were sutured to the serosal surface of the antrum 2 and 5 cm proximal to the pylorus, and four transducers were placed on the duodenum 2, 6, 10, and 14 cm distal to the pylorus. The catheter was tunneled s.c. to the midscapular region and connected to an s.c. access port (Access Technologies, Skokie, IL). The abdominal incision was closed in two layers. An access port catheter also was inserted into the external jugular vein. At least 2 weeks were allowed for recovery from surgery. Experiments were initiated only after the animals were consuming a normal diet.

Recording and Analysis of Contractile Activity. Contractile activity was recorded with a Grass polygraph (model 7) equipped with 7P1 low-level d.c. preamplifiers and 7DA driver amplifiers. The signals were simultaneously digitized at 10 Hz into computer files for identification of individual contractions and determination of the area under each contraction. Each record was analyzed from time of feeding until 90% of the meal had emptied; the data are expressed as the average motility index (area/minute) during that time. The area of each contraction at each site was standardized to the mean area of the 10 largest contractions during phase III activity at that site. This was done to account for differences between sensitivities of transducers at different sites. Additionally, the records were inspected visually for phenomena the computer program might not recognize, such as gastroduodenal coordination. Gastroduodenal coordination was defined as a contraction or group of contractions that originated in the antrum, while the duodenum was quiescence, and then propagated aborad into the duodenum within 10 s, migrating through the duodenum at a constant velocity.

Experimental Protocol. After an overnight fast, dogs were placed in a sling for GE studies. Beginning 30 min before the animals were fed, a solution containing a nonabsorbable marker [polyethylene glycol 4000 (PEG)] was perfused at 0.5 ml/min through the duodenal catheter and continued for the remainder of the experiment. Simultaneously, an i.v. infusion of either vehicle, ABT-229 (0.17, 0.83, 2.5, or 5.0 µg/kg/min), CIS (10 µg/kg/min), or ERY (33.3 µg/kg/min) was initiated and continued for 30 min at a volume rate of 0.24 ml/min. At the end of the drug infusion, the dogs were fed 175 g of commercial dog food (Alpo Prime Cuts) mixed thoroughly with 125 mg of chromium oxide (Cr₂O₃), a solid-phase marker. After feeding, chyme samples were collected every 5 min for the first 30 min to identify the lag phase of the GE curve. After this point, samples were collected at 40 and 50 min and then at 20-min intervals until solid food particles were no longer present.

Sample Analysis and Calculation of GE. The samples were centrifuged at 2000 rpm for 20 min, and the volumes of the supernatant (V_ln) and the solid pellet (V_sn) were measured. One milliliter of the supernatant (PEG_l) and perfusion solution (PEG_p) were used to determine the PEG concentrations by the method of Malawer and Powell (1967). The concentration of Cr₂O₃ was measured from the sediment of each sample by the method of Bolin et al. (1952).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

GE. ABT-229 dose dependently accelerated GE compared with vehicle (Fig. 1). CIS and ERY also accelerated GE at the doses tested (Fig. 1). ABT-229 at the two highest doses significantly decreased the lag phase, as did CIS compared with vehicle (Fig. 2). Additionally, there was a significant difference in t_lag between the 0.17 and 0.83 µg/kg/min doses of ABT-229, as well as ERY and the two highest doses of ABT-229 (Fig. 2). ABT-229 at the three highest doses, in addition to ERY and CIS, significantly decrease t_1/2 compared with vehicle (Fig. 2). Compared with vehicle, t_full was significantly decreased by all doses of ABT-229, as well as by ERY and CIS (Fig. 2). The GE rate during the linear phase was significantly increased by all doses of ABT-229, as well as by ERY and CIS compared with vehicle (Table 1); however, there was no difference in GE rate between doses of ABT-229 and/or between CIS and ERY. In all experiments, the regression coefficient for the linear phase was >0.9.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Cumulative mean GE curves of a solid meal in response to ABT-229, ERY, and CIS. All compounds significantly increase the slope compared with vehicle.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Effect of ABT-229, ERY, and CIS on the duration of the GE lag phase (tlag), t1/2, and tfull. n = 6; *P < .05 compared with control; P < .05 compared with 2.5; §P < .05 compared with 5.0.

Postprandial Contractile Activity. Two types of coordinated gastroduodenal contractile activity were observed (Figs. 3 and 4). At the two highest doses of ABT-229 and with CIS, a contractile pattern was induced that was characterized by a high amplitude (equal to the maximum amplitude observed during phase III activity) propagated antral contraction with quiescence in the duodenum. A migrating cluster of contractions in the duodenum then followed the antral contraction. This type of coordinated activity occurred during the first 60 min after the meal (Fig. 3). The frequency of this type of coordinated contractile pattern was significantly increased compared with vehicle and with the two lowest doses of ABT-229 and ERY (Table 2).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Recording illustrating high-amplitude, coordinated gastroduodenal contractile activity. This type of activity was seen in the first 60 min postprandially at the two highest doses of ABT-229 and with CIS. Note the quiescence in the duodenum during the antral contraction. Arrows indicate coordinated contractile activity. A, antrum and D, duodenum and the number after A and D indicates the distances of the transducers from the pylorus.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Recording illustrating low-amplitude, coordinated gastroduodenal contractile activity. This type of activity was recorded throughout the postprandial period with all doses of ABT-229, as well as with CIS and ERY. Note the quiescence in the duodenum during the antral contraction. Arrows indicate coordinated contractile activity. A, antrum and D, duodenum and the number after A and D indicates the distances of the transducers from the pylorus.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Effect of ABT-229, ERY, and CIS on the motility index of the antrum and duodenum. *P < .05 compared with vehicle.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The rate of GE is a function of the difference in pressures between the stomach and duodenum and of resistance to flow across the gastroduodenal junction (Kelly, 1981). Therefore, a compound that stimulates antral contractions, while the duodenum is quiescence, would be expected to increase flow across the gastroduodenal junction. Additionally, if propagated duodenal contractions occur after the antral contraction, the chyme would be carried away, reducing resistance to flow across the gastroduodenal junction when the next antral contraction occurs. This type of contractile activity is classified as gastroduodenal coordination (Kelly, 1981). In contrast, if duodenal contractions occur at the same time as the antral contraction, the pressure gradient across the gastroduodenal junction would be decreased and there would be less flow. All three compounds examined in this study induced gastroduodenal coordination.

ERY has previously been shown to be a gastrokinetic agent in both animals (Lin et al., 1994) and humans (Annese et al., 1992; Tack et al., 1992). ERY is thought to exercise its gastrokinetic effects by increasing the motility of the antrum (Annese et al., 1992; Tack et al., 1992), increasing proximal gastric tone (Bruley DesVarannes et al., 1995), and increasing the coordination between antral and duodenal contractions (Annese et al., 1992; Tack et al., 1992). At the dose of ERY used in this study, we also observed an increase in antral motility and gastroduodenal coordination.

This study shows that ABT-229, a synthetic derivative of ERY without antibiotic activity (Lartey et al., 1995; Faghih et al., 1998), dose dependently accelerated gastric emptying of a solid meal by decreasing the lag phase and increasing the rate of GE during the linear phase. Additionally, ABT-229 increased postprandial contractile activity and gastroduodenal coordination in the dog. Depending on the parameter examined, ABT-229 appears to be ~7- to 40-fold more potent than ERY in the conscious dog.

GE may be accelerated by at least two mechanisms with ABT-229. First, at the two highest doses of ABT-229, the lag phase was significantly decreased, probably as a result of high-amplitude, coordinated gastroduodenal contractions observed in the early postprandial period. It is also likely that ERY would have stimulated high-amplitude, coordinated activity in the dog if given at a higher dose than in this study. In humans, ERY has been reported to stimulate high-amplitude, coordinated gastroduodenal contractile activity (Annese et al., 1992) and in dogs to decrease the lag phase at higher doses than were used in this study (Lin et al., 1994). CIS also had been reported to decrease the lag phase, which is probably a result of the induction of high-amplitude, coordinated gastroduodenal contractile activity (Schuurkes, 1990). Second, GE may be accelerated through both the stimulation of low-amplitude, coordinated gastroduodenal contractile activity and an increase in motility index observed throughout the linear phase of GE. These factors are probably the cause of increased GE rate and appear to be shared by ABT-229, ERY, and CIS.

There is a third possible mechanism by which ABT-229 may enhance GE rate. Previously, ERY has been reported to increase postprandial proximal gastric tone and pressure in humans (Bruley DesVarannes et al., 1995). We did not, however, measure gastric tone in this study, so it remains to be determined if increased GE rate in dog is triggered by a rise in proximal gastric tone.

In conclusion, we have shown that ABT-229 dose dependently accelerates GE by increasing postprandial gastroduodenal coordination and by increasing the motility index. Furthermore, ABT-229 is ~7- to 40-fold more potent than ERY in this regard.