Cyclic interdigestive pancreatic exocrine secretion: Is it mediated by neural or hormonal mechanisms?

GASTROENTEROLOGY

1992;102:1378-1384

Cyclic Interdigestive Pancreatic Exocrine Secretion: Is It Mediated by Neural or Hormonal Mechanisms? DANIEL W. ZIMMERMAN, MICHAEL G. SARR, C. DANIEL SMITH, C. PHIFER NICHOLSON, RORY R. DALTON, DARLENE BARR, JAMES

D. PERKINS,

and EUGENE

Gastroenterology Research Unit and Department Rochester, Minnesota

P. DIMAGNO of Surgery, Mayo Clinic and Mayo Foundation,

Cyclic interdigestive exocrine pancreatic secretion and duodenal motility are closely linked. However, the mechanisms controlling this association are not well understood. The aim of this study was to determine whether a neural or hormonal mechanism controls the temporal association of interdigestive secretion and duodenal motility. In five dogs, the pancreas was autotransplanted to the pelvis with anastomosis of the pancreatic duct orifice to the bladder. Electrodes were positioned to monitor motility patterns of the in situ duodenum. After 10 days, dogs were studied on four occasions during fasting. Pancreatic output of amylase activity continued to cycle, but the periodicity of enzyme peaks (mean t SE) was different from the period of the duodenal migrating motor complex (MMC) (60+ 3 vs.125 f 7 minutes; P < 0.05). When grouped according to phase of duodenal MMC, amylase output per 10 minutes during phase I was significantly less than the outputs during phase II or III (135 + 52,214k 76,and 228f 73 X 103U; PC 0.05). However, there was no temporal relationship of the cyclic output of amylase to duodenal phase III. No differences were found when amylase output was analyzed for the 30 minutes before phase III compared with the 30minutes after phase III (687 If:253 vs. 376 + 110 X lo3U; P > 0.05). Plasma motilin concentrations varied with duodenal MMC, but no relationship existed between plasma motilin or plasma pancreatic polypeptide and peaks in amylase output. This study suggests that the close temporal coordination of interdigestive pancreatic exocrine secretion and duodenal motility is controlled primarily by a neural mechanism.

orethan

75 years ago, Boldyreff described a cyclic contractile activity of the upper gastrointestinal tract that occurred during fasting and was associated with a cyclic secretion of gastric, biliary, and pancreatic juices.’ In 1969,Szurszewski’ charac-

M

terized a cyclic, interdigestive myoelectric activity of the small intestine that began in the duodenum and migrated along the entire small intestine. This event later was named the migrating motor complex (MMC).3p4We’ recently rediscovered the findings of Boldyreff; at the same time, Vantrappen et a1.5 published similar observations. Since then, we and others have affirmed the closely linked temporal association of several interdigestive motor and secretory activities of the upper gut. Interdigestive secretion of gastric acid and pepsin occurs cyclically and peaks just before the onset of phase III of the gastric MMC5se Similarly, duodenal bicarbonate secretion’ and the delivery of pancreaticobiliary secretions into the duodenum,“’ related to contractile activity of the gallbladder,g~‘O also occur cyclically and peak just before the onset of phase III in the duodenum. The MMC has been described as an “intestinal housekeeper”” that sweeps the gut free of luminal debris and bacterial2 to ready the gut for the next meal. Association of this motor activity with the admixture of bile and secretions from the stomach, duodenum, and pancreas would aid this function. The mechanisms controlling this closely linked, temporal coordination between motor and secretory activity remain unknown but probably involve hormones, nerves, or both. Support for a hormonal control comes from studies investigating the initiation and control of the MMC. Plasma concentrations of motilin, a putative regulatory peptide hormone, are known to cycle in temporal association with the gastric MMC,13~14and when given exogenously will initiate both a premature MMC14*15and an increase in pancreatic secretion into the duodenum.” These data support the hypothesis that cyclic secretion of motilin (or some other hormone) coordinates motor and secretory activity. In contrast, others have pro0 1992by

the American Gastroenterological 0018-5085/92/$3.00

Association

EXOCRINE

April 1992

SECRETION

FROM AUTOTRANSPLANTED

PANCREAS

1379

posed that control of upper gut motility is under primary neural contro1,17~‘* suggesting that central neural mechanisms may also coordinate temporally both motor and secretory activity. Another possible neural control mechanism may involve local short reflexes over gastropancreatic or duodenopancreatic neural connections’g~20 that could coordinate gastric and duodenal motor activity with pancreatic exocrine secretion. The aim of our study was to determine whether exocrine secretion from the autotransplanted, neurally isolated pancreas continues to cycle in coordination with cycles of duodenal motility, motilin, and pancreatic polypeptide. In addition, we compared interdigestive exocrine secretion from the autotransplanted pancreas with responses duringintravenous infusion of motilin and cholecystokinin octapeptide (CCK-8). Materials and Methods Preparation

of Dogs

Surgical procedures, postoperative care, and experimental procedures were approved by the Institutional Animal Care and Use Committee of the Mayo Foundation. Five healthy female mongrel dogs (19-22 kg) were anesthetized with pentothal sodium (25 mg/kg) and maintained on inhaled halothane. Using a midline celiotomy, the entire pancreas was dissected from the duodenum, carefully preserving the pancreatic duct and superior pancreatoduodenal artery and vein. Typically, the distal one third of the uncinate process (ventral lobe) and distal one third of the body/tail (dorsal lobe) of the pancreas became ischemit and were excised after ligating the pancreatic parenchyma proximally with a heavy silk ligature. The pancreatic duct was isolated on a l-2-cm button of surrounding duodenum, and the superior pancreatoduodenal vessels were transected. The entire pancreas was removed, quickly placed in a basin of iced saline, and perfused intraarterially with an albumin preservative solution (4°C)” until the venous effluent was clear of blood. Simultaneously, the duodenal defect was closed transversely to avoid narrowing the lumen, and the proximal gastroduodenal artery and vein were ligated. The superior pancreatoduodenal vessels were then anastomosed end-to-side with the right common iliac vessels. The dome of the bladder was incised, and the duodenal button containing the pancreatic duct was sewn to the bladder in two layers, thereby allowing drainage of all pancreatic exocrine secretion directly into the bladder (Figure 1). Nine monopolar, Ag-AgCl serosal electrodes were sewn to the small bowel with the first electrode positioned near the site of the removed pancreatic duct and the remainder spaced at IO-cm intervals. Dogs were given 2 weeks to recover, during which plasma glucose was monitored daily to ascertain viability of the pancreatic autograft. Diets were supplemented with pancreatic enzyme replacement (3 tablespoons Viokase; A. H. Robbins, Philadelphia, PA) to assure luminal digestion of ingested nutrients.

Bladder

Figure 1. Autotransplantation of canine pancreas. The entire pancreas is mobilized based on the superior pancreatoduodenal vessels with a button of duodenum containing the major pancreatic ductal orifice. The vascular anastomoses are to the iliac vessels. The duodenal button is implanted into the bladder.

Conduct of Experiments After an overnight fast, each of the five dogs was studied on four separate occasions while resting in a Pavlov sling. A double-lumen, l6F Foley-type catheter was inserted into the bladder; an episiotomy of the introitus performed at the time of pancreatic autotransplantation facilitated placement of this catheter. A Harvard infusion pump delivered a 150 mmol/L NaCl solution at 2 mL/min through the proximal port of the urinary catheter; this constant perfusion helped to assure complete collection of lo-minute intervals of pancreatic exocrine secretions. Bladder effluent (infusate, urine, and pancreatic secretion) was collected continuously through the second lumen by gravity flow. Bladder effluents were collected in IO-minute samples, stored on ice, and frozen at the end of the experiment. Small bowel myoelectric activity was monitored continuously with an 8-channel Grass model 7D recorder (Grass Instruments, Quincy, MA) using alternatingcurrent amplifiers with a time constant of 1 second. Each day’s experiment included two or three complete interdigestive myoelectric cycles. Also, an intravenous catheter was placed at the beginning of each experiment, and plasma samples were collected on ice at SO-minute intervals or during each phase of the duodenal interdigestive myoelectric cycles and frozen for later radioimmunoassay. At the end of each experiment, exogenous motilin (0.1 pg/ kg) (Peninsula Laboratories, Belmont, CA) was given intra-

1380 ZIMMERMAN ET AL.

venously for 30 seconds beginning 30 minutes after the last duodenal phase III as described previously.14 Bladder effluent was collected for an additional 20 minutes after motilin administration. Each dog was studied a fifth time during infusion of CCK-8 (Peninsula Laboratories). The dogs were prepared for this study, and specimens of urine were collected in the same manner as described above. A Harvard pump delivered a constant intravenous 150 mmol/L NaCl infusion containing CCK-8 and plasma protein (1.0 mL plasma/50 mL solution to minimize adherence of peptide to surfaces of the syringe and tubing) in stepwise increasing doses of 20,40,80,160, and 320 ng - kg-’ 0h-‘. CCK-8 was infused at each rate for 1 hour.

Chemical Analyses Bladder effluents were thawed and assayed for amylase activity (Perkin-Elmer amylase reagent C; Coleman Instruments, Inc., Oak Brook, IL) as described previ0us1y.‘~ The serum samples were assayed for motilin and pancreatic polypeptide as a single batch using well-characterized radioimmunoassays.23

Data Analysis Myoelectric activity. Recordings of myoelectrical activity were analyzed by visual inspection and classified according to the criteria for phases of the MMC as described by Code and Marlett3 and by Tanaka and Sarr.” The mean length of each phase and of the entire MMC cycle was measured from the proximal duodenal electrode. Interdigestive pancreatic secretion. Amylase outputs per 10 minutes were analyzed in several ways. First, each lo-minute amylase output was classified according to the phase of the MMC measured in the proximal duodenal electrode. The mean lo-minute output per day was then calculated for each phase of the MMC in each dog, and grand means for all dogs were calculated. Second, the amylase outputs centered around duodenal phase III were calculated by determining the mean amylase output during each duodenal phase III and during each lo-minute period in the 30 minutes before and the 30 minutes after the phase III. Third, for each experiment we determined whether pancreatic secretion cycled without reference to duodenal motility by using previously described criteria of peaks of pancreatic enzyme secretion.23 Fourth, to examine whether phases of the duodenal MMC were related to peaks of amylase output, myoelectric activity was examined for the 30 minutes before and the 30 minutes after a peak in amylase output. For every 10 minutes, a value of 1, 2, or 3 was assigned if phase I, II, or III of the MMC, respectively, was present in the duodenum, and mean values were calculated. Previously validated23 peaks of pancreatic output were defined as either (a) a series of four or more lo-minute outputs of amylase in which at least two consecutive outputs of increasing value were followed by a lower output, or (b) a series of four or more amylase outputs in which one

GASTROENTEROLOGY Vol. 102,No. 4

lo-minute output was higher than the previous value and was followed by at least two successively lower outputs. In addition, the peak value had to be more than two times the lowest value in each group comprising a peak. The mean time between pancreatic secretory peaks (cycle duration) was then calculated. Plasma peptides. Plasma concentrations of motilin and pancreatic polypeptide were grouped according to the phase of the duodenal MMC, and mean values were obtained. To determine whether these peptide concentrations were related either to duodenal phase III or to peaks in amylase output, mean values of plasma peptide concentrations were calculated, when available, 20-30 minutes before, at the time of, and 20-30 minutes after each duodenal phase III. Similar calculations were made of peptide concentrations centered around each peak of amylase output. Response to exogenous motilin. The amylase output after motilin was analyzed in two ways. First, the total 20-minute amylase output was compared with the amylase output during the same 20 minutes of a spontaneous MMC cycle (i.e., because motilin was given 30 minutes after a spontaneous phase III, the interval used was 30-50 minutes after phase III). Second, to compare the response to exogenous motilin (a premature phase III) with spontaneous phase III, the total 20-minute amylase output after induction of a premature phase III by motilin (interval o-20 minutes after motilin) was compared with the amylase output o-20 minutes after a spontaneous phase III. Response to exogenous CCK-8. The mean lo-minute amylase output was calculated for each successive dose of CCK-8. Also, the lo-minute amylase outputs in each experiment during the 5-hour CCK-8 infusions were examined for peaks of amylase output as described above, and the time between peaks was calculated when cycles were present.

Statistical

Analysis

A repeated-measures analysis of variance was used to compare differences of amylase output among phases of the MMC and plasma motilin concentrations among phases of the MMC. Student’s t test for paired data was used to compare amylase outputs centered around peaks of motility (phase III). In this analyses, the sum of the amylase output for 30-o minutes before phase III was compared with the sum of the output for O-30 minutes after phase III. Similarly, Student’s t test for paired data was used to compare the cycle duration of duodenal motility (period of MMC) with the duration of cycles of amylase output and the sum of the amylase output after exogenous motilin with the sums of the output during the two intervals of spontaneous motility. When multiple comparisons were done, an adjusted a value was calculated using Bonferroni’s correction. Mean amylase output per 10 minutes during each dose of CCK-8 was compared using analysis of variance. P values of x0.05 were considered significant. Summary values in the text are presented as the mean + SEM.

April 1992

EXOCRINE

Results Health of Dogs All dogs remained healthy and active throughout the study. Each dog lost initially 3-5 kg of weight despite normal appetites and formed stools, but thereafter weights remained stable. Blood glucose levels in all dogs remained
Figure 2. Interdigestive myoelectric activity after autotransplantation of pancreas. Note spontaneous phase III of the MMC and a “premature” phase III induced by motilin. El-F., are serosal electrodes.

SECRETION

FROM AUTOTRANSPLANTED

PANCREAS

1381

(P < 0.01). During phase I, amylase output was less than during phase II or phase III (135 + 52, 214 f 78, and 228 -t 73 X lo3 U, respectively; P < 0.05). However, there was no significant difference between the total amylase outputs for the 30 minutes before and after phase III (687 + 253 vs. 378 -+ 110 X lo3 U; P > 0.05). However, cycles of amylase output were identified, and peaks occurred every 60 + 3 minutes (Figure 3). The time between duodenal phase 111s(125 f 7 minutes) was different (P < 0.05) from the time between peaks in amylase output. The phases of the duodenal MMC were different 30 minutes before and at the time of the amylase peak, but the difference was small (1.5 f 0.07 vs 1.7 + 0.07; P < 0.02). Plasma Peptides, Motility, and Amylase Output Plasma motilin concentrations varied cyclically with the phases of duodenal MMC. Motilin concentrations were greater during phase III than during phases I and II (283 + 27 vs. 213 + 32 or 221+ 23 pg/mL; P < 0.05). In addition, plasma motilin levels during phase III were significantly greater than concentrations 30 minutes before and after phase III (288 + 22, 203 + 22, and 220 t- 35 pg/mL, respectively; P < 0.025). In contrast, no differences in motilin concentrations were present around peaks of amylase output. Fasting plasma pancreatic polypeptide concentrations were low in all dogs after pancreatic autotransplantation. Normal fasting canine plasma concentrations in our laboratory vary from 40 to 100 pg/mL.23 In two dogs, plasma concentrations were always at or below the lowest limit of the assay (20 pg/mL). In the other three dogs, mean values during the different

1382

ZIMMERMAN ET AL.

GASTROENTEROLOGYVol.102, No.4

1.200,000

time interval immediately after spontaneous III (264f 76 X lo3U; P = 0.06).

1,000,000

phase

CCK-8 Infusion F

600,000

.z 2. $

600,000

z 2

400,000

200,000

0 $0

rio

li0

2io

360

8

360

Time (Minutes)

Figure 3. Representative daily experiment on one dog after autotransplantation of the pancreas. Amylase output cycles at a rate faster than the cycle length of the duodenum without apparent temporal association with duodenal phase III (W)activity. A peak in amylase output (arrow) appears to follow the infusion of motilin (0.1 &kg, intravenous). See text for details of response to motilin.

phases of the MMC ranged from 20 to 55 pg/mL. There was no relationship between plasma pancreatic polypeptide concentrations and phases of the MMC, peaks in amylase output, or cycles of plasma motilin concentration. Exogenous Motilin Motilin (0.1 pg/kg, intravenous) induced a premature phase III in the duodenum within 10 minutes in 14 of 20 experiments. The total amylase output for the 20 minutes after motilin administration was greater than the total amylase output for the comparable interval during a spontaneous MMC cycle 30-50minutes after phase III (509 f 185 vs.274 f 141 X lo3U; P < 0.03)and the comparable 20-minute

All concentrations of CCK-8 (40-320 ng. kg-’ * h-‘) significantly increased amylase output over the mean output during fasting (Figure 4). However, there was no further significant increase in output of amylase at CCK-8 doses of >4O ng +kg-‘. h-l. The mean amylase output in each dog during administration of each dose of CCK-8 in general was greater than the peaks in amylase output during fasting. During the stepwise increases in CCK-8 administration (Figure 5), cyclic output of amylase continued in all dogs as occurred during fasting. The duration of these cycles (80 + 18 minutes) was similar to the intervals of peaks in amylase output during fasting (60 + 3 minutes). Discussion Our data show that interdigestive pancreatic enzyme secretion of the extrinsically denervated pancreas and upper gastrointestinal motility cycle at independent rates. Under these circumstances, enzyme output cycles two times faster than the MMC in the upper gut. We also found that motilin cycles in concert with the duodenal MMC and that exogenous motilin causes a peak of pancreatic enzyme secretion simultaneous with the onset of a premature phase III. It is unlikely that motilin coordinates the relationship between interdigestive pancreatic secretion and gut motility without extrinsic innervation of the pancreas because there was no association between naturally occurring plasma motilin concentrations and peaks of enzyme output. These data show that extrinsic innervation of the pancreas is necessary to

(DoseofCCK)

1 ,ooo,ooo ,

I si ,F 2. 8 -m

600,000

-

400,000

-

200,000

-

2’ooo~oww

0

20

40 Dose CCK-OP

80

160

&

Amylase Output I 10 Minutes

320

(ngkgihr)

Figure 4. Amylase output from the autotransplanted canine pancreas during intravenous CCK-8 infusion. CCK-8 doses of 40-180 ng *kg-’ *h-l increased amylase output over the mean output during fasting. ?? , Amylase output per 10 minutes (mean + SEM; n = 3 dogs); *values >O ng . kg-l - h-l (P < 0.03).

0

, , . . .1x1 ...

60

I

r

,

I

160

r

I

r,

I

240

.

.

I

,*

300

Time (Minutes)

Figure 5. Representative response of one dog with an autotransplanted pancreas to intravenous infusion of CCK-8. Note ongoing cycles of amylase output.

April 1992

EXOCRINE SECRETION FROM AUTOTRANSPLANTED

maintain the normal, closely linked temporal coordination between interdigestive pancreatic secretion and upper gut motility. Previous studies addressing neural vs. hormonal control of the temporal coordination of interdigestive upper gut motility and pancreatic exocrine secretion have been inconclusive. A primary hormonal role is suggested by studies showing that plasma motilin cycles in temporal coordination with the upper gut MMC13*14and that exogenous motilin induces both a premature MMC and a peak in pancreatic secretion, 14*15 In addition, antimotilin serumz5 but not anti-pancreatic polypeptide (anti-PP) serumz6 abolishes cycles of both motility and pancreatic secretion. However, many other studies support a primary neural regulation. Magee and Narusez7**’ used muscarinic and nicotinic cholinergic antagonists to abolish cycles of pancreatic secretion (and motility) as well as the pancreatic secretory response to motilin. Moreover, intraduodenal lidocaine abolished cycles of motility and plasma PP, but pancreatic secretion continued to cycle.” In previous experiments, we showed that total duodenectomy abolished cycling of plasma motilin (and PP), but pancreatic secretion continued to cycle at a normal rate23; moreover, the coordination between interdigestive cycles of gut motility and pancreatic secretion was disrupted. Together, these reports and our study, which shows a lack of coordination between duodenal motility and cycles of amylase secretion from the autotransplanted (neurally isolated) pancreas, show conclusively that extrinsic neural input controls the temporal coordination of interdigestive upper gut motility and pancreatic secretion. Whether the coordinating neural input is mediated by vagal,30 sympathetic, or short gastropancreatic or duodenopancreatic pathways from the gut enteric nervous systemlg remains unknown. Our experiments do not exclude the possibility that plasma motilin plays a role in regulating upper gut motility and pancreatic secretion. In the neurally intact gut, cycles of plasma motilin may help to coordinate the onset of interdigestive cycles of motility and secretion by inducing mechanisms mediated by extrinsic innervation.3* It is also possible that motilin increases peaks of pancreatic secretion. This hypothesis would explain why amylase output was greater during phases II and III (when plasma motilin values were greater) than during phase I (when plasma motilin was decreased) in our study, and why a pharmacological dose of motilin induced a peak in pancreatic secretion, The mechanism of maintaining cyclic secretion from the autotransplanted (neurally isolated) pancreas or after disruption of duodenopancreatic neural connections after duodenectomy remains un-

PANCREAS

1383

known. Others have shown cyclic exocrine pancreatic secretion in humans after pancreatic allotransplantation,32 although no attempt was made to measure motility. It is unlikely, although possible, that an unrecognized hormone controls this cyclic secretion, especially because cycles of secretion and motility are no longer coordinated after either duodenectomy23 or pancreatic autotransplantation. Previous pharmacological studies have shown that nicotinic cholinergic antagonists abolish cycles of interdigestive pancreatic secretion.27~28 These findings, with the current study, suggest that cyclic pancreatic secretion during the interdigestive state is mediated by intrinsic neural innervation of the pancreas.*’ This cyclic secretion appears to continue, but with greater amounts of amylase secreted, during CCK infusion. This finding agrees with previous work by Magee and Naruse33 in the intact dog. In summary, interdigestive pancreatic secretion continued to cycle in vivo in the neurally isolated, autotransplanted canine pancreas. However, there was no temporal coordination between pancreatic secretory cycles and upper gut motility. These findings suggest that interdigestive cyclic pancreatic secretion is controlled by an intrapancreatic mechanism, probably the intrinsic neural innervation of the pancreas. In contrast, temporal coordination between duodenal motility and pancreatic secretion during fasting is controlled by extrinsic neural mechanisms References 1. DiMagno EP, Hendricks

2. 3.

4. 5.

6.

JC, Go VLW, Dozois RR. Relationships among canine fasting pancreatic and biliary secretions, pancreatic duct pressure, and duodenal phase III motor activity-Boldyreff revisited. Dig Dis Sci 1979;24:689-693. Szurszewski JH. A migrating electric complex of the canine small intestine. Am J Physiol 1969;217:1757-1763. Code CF, Marlett JA. The interdigestive myoelectric complex of the stomach and small bowel of dogs. J Physiol 1975;246:289-309. Sarna SK. Cyclic motor activity: migrating motor complex: 1985. Gastroenterology 1985;89:894-913. Vantrappen GR, Peeters TL, Janssens J. The secretory component of the interdigestive migrating motor complex in man. Stand J Gastroenterol 1979;14:663-667. Keane FB, DiMagno EP, Malagelada J-R. Duodenogastric reflux in humans: its relationship to fasting, antroduodenal motility and gastric, pancreatic, and biliary secretion. Gastroenterology 1981;81:726-731. Konturek SJ, Thor P. Relation between duodenal alkaline secretion and motility in fasted and sham-fed dogs. Am J Physiol 1986;251:G591-G596. Keane FB, DiMagno EP, Dozois RR, Go VLW. Relationships among canine interdigestive exocrine pancreatic and biliary flow, duodenal motor activity, plasma pancreatic polypeptide, and motilin. Gastroenterology 1980:78:310-316. Traynor OJ, Dozois RR, DiMagno EP. Canine interdigestive and postprandial gallbladder motility and emptying. Am J Physiol 1984;246:G426-G432.

1384 ZIMMERMAN

ET AL.

GASTROENTEROLOGY Vol. 102, No. 4

10. Matsumoto

T, Sarna SK, Condon RE, Dodds WJ, Mochinaga N. Canine gallbladder cyclic motor activity. Am J Physiol 1988;255:G409-G416. 11. Code CF, Schlegel JF. The gastrointestinal interdigestive housekeeper: motor correlates of the interdigestive myoelectric complex of the dog. In: Daniel EE, ed. Proceedings of the IV International Symposium on Gastrointestinal Motility. Vancouver: Mitchell Press, 1974:631-634. 12. VanTrappen G, Janssens J, Hellemans J, Ghoos Y. The interdigestive motor complex of normal subjects and patients with bacterial overgrowth of the small intestine. J Clin Invest 1977;59:1158-1166. 13. Itoh Z, Takeuchi S, Aizawa I, Mori K, Taminato T, Seino Y, Imura H, Yanaihara N. Changes in plasma motilin concentration and gastrointestinal motor activity in conscious dogs. Am J Dig Dis 1978;23:929-935. 14. Sarr MG, Duenes JA. Site of action of morphine sulfate and motilin in the induction of “premature” phase III-like activity in the canine gastrointestinal tract. Surgery 1988;103:653-

23. Malfertheiner

24.

25.

26. 27.

447. 28. Magee DF, Naruse

29.

661. 15. Wingate

DL, Ruppin A, Thompson HH, Green WER, Domschke W, Wttnsch E, Demling L, Ritche HD. The gastrointestinal myoelectric response to 13-Nle-motilin infusion during interdigestive and digestive states in the conscious dog. Acta Hepatogastroenterol 1977;24:278-287. 16. Lee KY, Shiratori K, Chen YF, Chang T-M, Chey WY. A hormonal mechanism for the interdigestive pancreatic secretion in dogs. Am J Physiol 1986;251:G759-G764. 17. Hall KE, El-Sharkawy TY, Diamant NE. Vagal control of migrating motor complex in the dog. Am J Physiol 1982;243: G276-G284. 18. Magee DF, Naruse S. Neural control of periodic

secretion of in fasting dogs. Am J Physiol

19. 20.

21.

22.

the pancreas and the stomach 1983;344:153-160. Kirchgessner AL, Gershon MD. Innervation of the pancreas by neurons in the gut. J Neurosci 1990;10:1626-1642. Huchtebrock H-J, Niebel W, Singer MV, Forssmann WG. Intrinsic pancreatic nerves after mechanical denervation of the extrinsic pancreatic nerves in dogs. Pancreas 1991;6:1-8. Perkins JD, Fromme GA, Narr BJ, Southhorn PA, Marsh CL, Munn SR, Engen DE, Sterioff S. Pancreas transplantation at Mayo: II. Operative and perioperative management. Mayo Clin Proc 1990;65:483-485. Layer P, Zinsmeister AR, DiMagno EP. Effects of decreasing intraluminal amylase activity on starch digestion and postprandial gastrointestinal function in humans. Gastroenterology 1986;91:41-48.

P, Sarr MG, Spencer MP, DiMagno EP. Effect of duodenectomy on interdigestive pancreatic secretion, gastrointestinal motility, and hormones in dogs. Am J Physiol 1989;257:G415-G422. Tanaka M, Sarr MG. Role of the duodenum in the control of canine gastrointestinal motility. Gastroenterology 1988;94: 622-629. Lee KY, Chang TM, Chey WY. Effect of rabbit antimotilin serum on myoelectric activity and plasma motilin concentration in fasting dog. Am J Physiol 1983;245:G547-G553. Thor PJ, Konturek JW, Konturek SJ. Pancreatic polypeptide and intestinal motility in dogs. Dig Dis Sci 1987;32:513-519. Magee DF, Naruse S. The role of motilin in periodic interdigestive pancreatic secretion in dogs. J Physiol 1984;355:441-

30.

31.

32.

33.

S. Neural control of periodic secretion of the pancreas and the stomach in fasting dogs. J Physiol 1983;344:153-160. Chen MH, Joffe SN, Magee DF, Murphy RF, Naruse S. Cyclic changes of plasma pancreatic polypeptide and pancreatic secretion in fasting dogs. J Physiol 1983;341:453-461. Poulsen J, Delikaris P, Lovgreen NA, Schwartz TW. Impaired pancreatic innervation after pyloric transection in dogs. Stand J Gastroenterol 1983;18:17-22. Hakim NS, Soper NJ, Spencer MP, Sarr MG. Role of extrinsic and intrinsic nerves in hormonal induction of the migrating motor complex in the jejunum. J Invest Surg 1989;2:437-446. Tibell A, Linder R, Ostman R, Hellstrom PM, Tyden G, Johansson C. Cyclic exocrine secretion of the transplanted human pancreas in the fasting state. Transplantation Proc 1990;22:1596-1597. Magee DF, Naruse S. The effect of cholecystokinin-related peptides on periodic pancreatic secretion in fasting dogs. J Physiol 1988;403:15-25.

Received May 31, 1991. Accepted October 22, 1991. Address requests for reprints to: Michael G. Sarr, M.D., Gastroenterology Research Unit, Mayo Clinic, 200 First Street SW., Rochester, Minnesota 55905. Supported in part by Research Grants DK39337 and DK34988 from the National Institutes of Health, U.S. Surgical Corporation, and Mayo Foundation. Presented at the American Gastroenterological Association on May 21, 1991, in New Orleans, Louisiana, and published in abstract form (Gastroenterology 1991;100:A849). The authors thank J. A. Duenes for technical assistance and D. I. Frank and L. Bakken for preparation of the manuscript.