DIRECT AND INDIRECT EFFECTS OF DEXTROSE AND AMINO ACIDS ON GUT MASS

Transcription

1 GASTROENTEROLOGY 72:76-71, 1977 Copyright 1977 by the American Gastroenterological Association Vol. 72, No.4, Part 1 Printed in U.S.A. DRECT AND NDRECT EFFECTS OF DEXTROSE AND AMNO ACDS ON GUT MASS MONROE H. SPECTOR, M.D., GARY M. LEVNE, M.D., AND JULUS J. DEREN, M.D. Department of Medicine, Philadelphia Veterans Administration Hospital and the University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania Oral intake of an elemental diet maintains small intestinal mucosal mass compared to the atrophy seen after intravenous infusion of the same diet. The greatest difference in intestinal mass occurs in the proximal bowel and is thought to occur because of rapid absorption in the proximal small bowel. This study was designed to determine the effects of the individual components of the elemental diet and their site of administration within the small bowel on small intestinal mass. Rats were maintained on intravenous alimentation and the proximal gut (by intragastric infusion) or the ileum was continuously infused with equal volumes of3% dextrose, 5% dextrose, 5% amino acids, saline, or 3% mannitol. Mter 1 week of combined intravenous alimentation and gut infusion, the rats were killed and parameters of small intestinal epithelial mass were determined sequentially for the entire bowel. Although saline- and mannitol-infused controls did not differ from uninfused intravenously fed rats, proximal infusion of 3% dextrose reproduced the effects of a complete elemental diet. Proximal infusion of amino acids but not 5% dextrose had a limited effect on the duodenum and jejunum. leal infusion of 3% dextrose led to local hyperplasia of the site of infusion and in addition produced hyperplasia of the proximal gut. leal amino acid infusion, but not 5% dextrose infusion, led to local ileal hyperplasia. We conclude that: (1) intraluminal dextrose and amino acids have direct effects in maintaining gut mass (2) the gut is more responsive to amino acids compared to 5% dextrose, and (3) ileal 3% dextrose infusion leads to remote effects in the proximal gut, perhaps mediated by hormonal or neurovascular factors. Oral intake maintains small intestnal mass whereas intravenous alimentation leads to atrophy similar to that seen with fasting.1-3 n addition, intraluminal nutrition maintains the normal proximal to distal gradient of mass as a result of rapid absorption of nutrients from the proximal small bowel. n a previous study, we demonstrated that a complete elemental diet consisting of 3% dextrose, 5% amino acids, vitamins, and minerals, when given orally, prevented small intestinal atrophy seen with intravenous alimentation. 3 This study was designed to examine the effects of the individual constituents of the elemental diet on the small bowel in order to: (1) assess their relative contribution to the maintenance of small intestinal mass and (2) determine the regional effects of infusion into either the proximal or distal bowel. Received May 19, Accepted September 28, Address requests for reprints to: Gary M. Levine, M.D. ll1h3, Department of Medicine, Veterans Administration Hospital, University and Woodland Avenues, Philadelphia, Pennsylvania This study was supported by Veterans Administration MRS funds (Project ) and by grants-in-aid from J. Aaron and Company, New York, New York. The authors wish to express their gratitude to Elizabeth Yezdimir for her excellent technical support and to Marianne Adams for her secretarial assistance. Methods Preparation of animals and tissues. Sprague-Dawley male rats (weight 21 to 25 g) were prepared by placing catheters in the right jugular vein (for intravenous alimentation) and either in the stomach or ileum for enteral infusion. After catheter placement, rats were maintained unrestrained in individual cages and were infused with an elemental diet consisting of 3% dextrose, 5% amino acids, minerals, and vitamins by constant infusion pumps as previously reported. 3 nitially, during the first 2 days of study, animals were infused at 3 to 4 ml per day to allow for adaptation to intravenous and intraluminal load. During the last 5 days ofthe study, 45 to 6 ml per day were infused. ntravenous alimentation was used to maintain growth and adequate caloric intake regardless of enteral infusion. The rate of intravenous alimentation was adjusted to allow approximately equal caloric intake among the different groups studied. Protocol : gastric infusion study. The effects of intra gastric infusion of glucose and amino acids were studied by placing a silastic catheter in the rumen portion of the stomach with the proximal end brought through the same harness mechanism used for intravenous alimentation. Groups of rats were infused with equal volumes of one of the following five solutions: saline (6 animals), 3% mannitol (4 animals), 5% mixture of essential and nonessential amino acids (Freamine, McGaw Laboratories, Glendale, Calif.) (7 animals), 5% dextrose (6 animals), and 3% dextrose (6 animals). Protocol : distal small intestinal infusion study. The effects of infusing dextrose and amino acids directly into the 76

2 April 1977 EFFECTS OF DEXTROSE AND AMNO ACDS 77 distal bowel were studied by placing a silastic catheter approximately 35 cm from the ileocecal valve and exteriorizing as described above. The following groups were studied: isotonic saline and catheter implantation without infusion (5 animals), 3% mannitol (4 animals), 5% amino acids (6 animals), 5% dextrose (6 animals), and 3% dextrose (6 animals). After 7 days of combined enteral and intravenous infusion, animals were stunned, bled, and decapitated. The small bowel, from the pylorus to the ileocecal valve, was rinsed thoroughly with chilled saline, excised, and detached from its mesentery. The gut was then blown dry with 1 cc of air, excessive saline was removed by stripping, and the gut was blotted dry. Total gut weight and total length (under 15 g of tension) were determined. The intestines of proximally infused animals were divided into eight approximately equal segments. Segment 1 represented the duodenum and segment 8 represented terminal ileum. The intestines of ileal infused animals were divided by first isolating a 1cm "catheter segment" approximately 5 cm proximal and distal to the site of the catheter and then dividing the remaining gut into eight additional segments. Each segment was weighed, its length was determined, and the mucosa was removed by scraping. Mucosal scrapings were homogenized with saline, 5 times w/v. Analytical methods. Mucosal protein was determined according to the method of Lowry et al.,4 and mucosal DNA was determined colorimetrically with a modified diphenylamine reagent,s. 6 using highly polymerized calf thymus DNA as standard. Serum gastrin concentrations were determined by radioimmunoassay according to the method of Yalow and Berson 7 using the Squibb mmunotope kit (E. R. Squibb & Sons, New Brunswick, N.J.). The assay system was sensitive to concentrations of less than 5 pg per ml, and intrasample variation was less than 1%. Statistical analysis was carried out by unpaired Student's t-tests and data were expressed as the mean ± SEM. Results Body Weight and Gut Length All animals gained weight during the period of the study. There were no significant differences between groups in initial and final weights as shown in table 1. Total gut length was similar in all groups. Protocol : Gastric nfusion Study Total gut weight. Gut weights were significantly greater in animals infused with amino acids (6.15 ±.37 g, P <.5) and 3% dextrose (7.16 ±.4 g, P <.1), than in saline-infused animals (5.33 ±.23 g). Gut weight of 5% dextrose-infused animals (5.67 ±.22 g) did not differ significantly from that in the saline control groups. 3% mannitol infusion (5.66 ±.24 g) had no effect on small intestinal mass. nfusion TABLE 1. Comparison of body weight a ntragastric leal nitial Final nitial Final 3% dextrose 236 :!: 4 26 ± :!: :!: 14 5% dextrose 233 :!: ± ± ± 7 5% amino acids 243 :!: :!: :!: ± 13 3% mannitol 236 :!: :!: ± ± 5 Saline 246 :!: ± ± ± 5 a Values given are weight in grams ± SEM., ".. '" r!:!! 3 2 i J. A Amino Acid.... Soline Co""ol 5 11 l c: Segment FG. 1. Effect of dextrose and amino acids on mucosal weight (milligrams per centimeter) after intragastric infusion. nfusion of 3% dextrose led to greater mucosal weight from segment 1 to 7 (P <.1), whereas amino acids had an effect only on segments 1 to 3 (P <.1). nfusion of 5% dextrose, and of saline did not differ. Segmental gut determinations (gut and mucosal weight, mucosal protein, and DNA). The data for gut weight, mucosal protein, and DNA per segment paralleled mucosal weight in all cases and hence will not be shown for the sake of brevity. As shown in figure 1, intragastric infusion of 3% dextrose led to greater mucosal weight along the entire length of the small intestine compared to controls. These values for 3% dextrose infusion were almost identical to data obtained previously with the oral administration of the complete elemental diet. 3 A considerable proximal-distal gradient (segment 1 to 8) was present after 3% dextrose infusion as compared to the saline control group. All measured parameters (gut weight, mucosal weight, DNA, and protein) ranged from greater than 8% higher in segment 1 to greater than 3% higher in segment 7 versus controls (P <.1 inclusive). The effect of 5% amino acids, shown in figure 1, was restricted to the proximal intestine (segments 1 to 3). A less marked response was elicited by 5% amino acid, which produced 2 to 5% increment in the above parameters versus control (P <.1) and was not as great as that reported for the complete elemental diet. 3 nfusion of 5% dextrose, as shown in figure 1, and 3% mannitol (data not shown) had no effect on small intestinal mass. The values for control animals infused with saline and 3% mannitol, as well as those infused with 5% dextrose, were similar to those previously reported for intravenously fed animals whose intestines were unperfused. 3 Protocol : Distal Small ntestinal nfusion Study Total gut weight. Gut weight was significantly greater in animals infused with 3% dextrose (7.85 ±.13 g, P <.1) than saline control (5.6 ±.37 g). The total gut weight of 5% amino acid-infused rats (5.74 ±.28 g) and 5% dextrose-infused rats (5.67 ±.4 g) did not differ from controls. n addition, 3% mannitol (4.757 ±.14 g) had no effect on small intestinal mass. H

3 78 SPECTOR ET AL. Vol. 72, No. 4, Part 1 Segmental gut determinations (gut and mucosal weight, mucosal protein, and DNA). For the sake of brevity, with the exception of the results of 3% dextrose infusion, data for gut weight, mucosal protein, and DNA are not shown. All data are compared to salineinfused controls. There was no effect on the distal intestine after simple catheter insertion, after ileal saline infusion, or by 3% mannitol infusion compared with control data from the proximally infty'ed groups. However, 3% dextrose evoked a considerable increase in mucosal weight, mucosal protein, and DNA in segments adjacent to and beyond the catheter site in all animals studied (at least 1% versus control, P <.1). This is shown in figures 2, 3, and 4, with each panel representing 1 animal. Surprisingly, the most proximal segments of the small intestine clearly remote from the infusion area by at least 7 cm also showed more than 5% greater mucosal weight in 5 of6 animals (P <.1, for the group as a whole compared to controls). The proximal effect was maximal in the duodenal segment and decreased distally. A proximal to distal gradient was also noted beginning at the infusion site and decreasing caudally. nfusion of 5% amino acids into the ileum (fig. 5) caused a moderate increase locally in mucosal weight at the catheter site in all animals and beyond the catheter site in 3 of 6 animals studied. All parameters (gut and mucosal weight, mucosal protein, and DNA) were at least 5% greater than in controls (P <.1). No effect was noted proximally. The effects of 5% dextrose were minimal and not statistically significant (less than 15% higher than controls). Serum gastrin determinations. No significant differences in serum gastrin levels were noted between groups after either proximal or distal dextrose infusion, despite marked differences in small intestinal mass (table 2). However, gastrin levels were higher than in saline controls, significantly so after proximal 3% dextrose infusion and ileal 5% dextrose infusion (P <.5). Samples were not available from amino acid-infused animals for comparison. Discussion These studies demonstrate the effects of dietary constitutents on the epithelial mass of the small intestine. First, both dextrose and amino acids appear capable of maintaining gut mass, although amino acids appear more potent on a molar and weight basis. Second, we have shown that the distal small intestine is capable of '" 6 ' 8 o ' O i i B B B (g '" B B o odic \' 4 ',, 4 o i 4 o _ B FG. 3. Effect of 3% dextrose on mucosal protein content after ileal infusion. Results as in figure o " "...: = SAL NE CONTROL = 3 \=CATHETER \. u "- (f) (!) 8 \ 4 o \ FG. 2. Effect of3% dextrose on mucosal weight (milligrams per centimeter) after ileal infusion. Eachpanel represents 1 animal and the arrow indicates the site of infusion. n addition to significantly greater mucosal weight at and distal to the site of infusion (P <.1), there also was a remote effect on the proximal bowel in 5 of 6 animals studied (P <.1), compared to controls.

4 April 1977 EFFECTS OF DEXTROSE AND AMNO ACDS 79 responding as vigorously as the proximal intestine when adequate contact of nutrients is ensured by direct infusion. Third, our study suggests that both intraluminal nutrition and indirect effects playa role in regulating small intestinal mass. Both 3% dextrose and 5% amino acids infused into the proximal intestine through an intragastric catheter led to greater epithelial mass than in controls. nfusion of 3% dextrose increased gut mass to the same extent as a complete elemental diet of 3% dextrose plus 5% amino acids as previously reported, and promoted the proximal-distal gradient. 3 Amino acids led to a lesser, but significant increment in gut mass, whereas infusion of 5% dextrose had no effect. We could not test the effect of amino acid concentrations comparable to 3% dextrose because of poor solubility. The lack of response to 3% mannitol infusion of both the proximal and distal intestine eliminates the possibility that the response :J: 2466 :J: 1..8 ' o a.2 o 1.?.8 o 1..8 o.8 o.4..2 o 1..8 ' o o FG. 4. Effect of3% dextrose on mucosal DNA content after ileal infusion. Results as in figure 2. seen with hypertonic dextrose was caused by hypertonicity alone, or by the movement of other nutrients from the plasma into the lumen. n addition, intraluminal distention (produced by infusion of normal saline and mannitol) or other mechanical effects from the placement of the catheter were not sufficient to elicit any response in the control animals studied. nfusion of 3% dextrose into the proximal intestine led to significant increases in gut mass in the distal bowel as well, despite the fact that sugars are rapidly and actively absorbed in the upper small bowel. 8 The distal effects of intragastric 3% dextrose infusion may have been elicited by hormonal, neural, or vascular responses, which secondarily promoted an increment in gut mass. The infusion of 3% dextrose into the ileum led to greater mucosal mass both at and caudad to the infusion site and surprisingly had a striking remote effect in the duodenum and proximal jejunum, similar to that seen in the proximal gut after intragastric 3% dextrose infusion. The significant local response of the ileum to the hypertonic infusion is evidence that intraluminal contact of intraluminal nutrients with the bowel is of major importance in maintaining and stimulating small intestinal growth. n addition, the ileal infusion studies demonstrate that part of the response to intraluminal nutrition is mediated indirectly, as shown by the remote effects in the proximal bowel after ileal 3% dextrose infusion. These remote effects were not caused by reflux TABLE 2. Comparison of serum gastrin after gut infusion ntragastric leal nfusion No. pg/ml ± SEM No. pg/ml ± SEM 3% dextrose ± 4.8 a ± 9.6 5% dextrose ± ± 4.3 a Controlsb 8 6. ± 5.8 a P <.5 versus control. b Saline-infused and catheter-noninfused ' i.. o e= SAL NE CONTROL = AMNO ACDS "'= CATHETER ::E (J) (.!) ::E o L-+--t--+--+_ OL ' " i. o 4 i "- 4 :. o '. i e. o FG. 5. Effect of ileal infusion of amino acids on mucosal weight. There was a significant increase at and distal to the site of infusion (P <.1) compared to controls.

5 71 SPECTOR ET AL. Vol. 72, No.4, Part 1 of nutrient material into the proximal small intestine, because the remote effect was always most marked in the proximal segments, tapering off distally. Caudad to the infusion site there was also a decreasing effect as nutrients were absorbed. Had reflux occurred, the proximal-distal gradient would have been bell-shaped, with a maximum effect near the catheter, decreasing toward both proximal and distal bowel. These studies are consistent with the observation of Dworkin et al.,9 who demonstrated that infusing nutrients into the gut in continuity of Thiry-Vella fistula rats elicited an increase in gut mass in the excluded loops, and the observations of Espinoza et al., 1 who demonstrated that ileal glucose perfusion led to an increase in intestinal glycolytic enzymes in the proximal gut of the rat. The infusion of a 5% mixture of essential and nonessential amino acids also appears to be a potent stimulator of small intestinal epithelial growth in both the proximal and distal bowel. However, distal infusion of amino acids, although demonstrating the importance of local administration and intraluminal nutrition, did not evoke a remote effect in the proximal bowel, as did hypertonic dextrose. Whether one, several, or a group (i.e., essential versus nonessential) of amino acids is necessary for local hyperplasia requires further investigation. Several mechanisms may be responsible for the remote changes observed. Altmann ll and Heller and colleagues 12 have both provided evidence that pancreatic and/or biliary secretions are important in mediating changes in small intestinal mass. These secretions probably play only a limited role in modulating the effects of 3% dextrose infusion, because hypertonic glucose does not stimulate pancreatic exocrine flow and in fact has been shown to inhibit pancreatic secretion in man. 13 n the past several years, Johnson and his co-workers 14 have provided a large body of evidence suggesting that gastrin is a gastrointestinal trophic hormone. For this reason we measured serum gastrin concentrations at the time the animals were killed. Sufficient samples were available to compare the effects of proximal and distal dextrose infusion with the controls used in this study. Although animals receiving dextrose infusion had higher gastrin levels than controls, there were no significant differences in gastrin levels among proximally or distally infused dextrose groups, despite marked differences in small intestinal mass between these groups. These data mitigate against the responsibility of gastrin for the remote effects observed after ileal 3% dextrose infusion. Other factors, not examined in this study, which may playa role include changes in small intestinal mucosal blood flow after hypertonic glucose infusion 15 and the release of other gastrointestinal hormones such as glucagon 16 or gastric inhibitory peptide. 17 n conclusion, our studies provide preliminary data as to the effects of individual classes of dietary components on small intestinal mass. Either glucose or amino acids appear capable of eliciting this response, although amino acids appear more potent. The ileum is as capable of responding as the jejunum when adequate contact of nutrients is ensured by direct infusion. The data also provide additional evidence that indirect effects, perhaps mediated by hormonal or neurovascular factors, playa role in the maintenance of small intestinal mass. REFERENCES 1. Steiner M, Bourges HR, Freedman LS, et al: Effect of starvation on tissue composition of the small intestine of the rat. Am J Physiol 215:75-77, McManus JPA, sselbacher KJ: Effect of fasting versus feeding on the rat small intestine. Gastroenterology 59: , Levine GM, Deren JJ, Steiger E, et al: Role of oral intake in maintenance of gut mass and disaccharidase activity. Gastroenterology 67: , Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the folin phenol reagent. J Bio Chern 193: , Schmidt G, Thannhauser SJ: A method for the determination of desoxyribonucleic acid, ribonucleic acid, and phosphoproteins in animal tissues. J Bio Chern 161:83-89, Croft DN, Lubran N: The estimation of deoxyribonucleic acid in the presence of sialic acid: application to analysis of human gastric washings. Biochem J 95:612-62, Yalow RS, Berson SA: Radioimmunoassay of gastrin. Gastroenterology 58:1-14, Fisher RB, Parsons DS: Glucose absorption from surviving rat small intestine. J Physiol (Lond) 11: , Dworkin LD, Levine GM, Farber NJ, et al: Small intestinal mass of the rat is partially determined by indirect effects of intraluminal nutrition. Gastroenterology 71:626-63, Espinoza J, Bennett Clark S, Hritz A, et al: Regulation of rat proximal small intestinal enzyme activity by ileal glucose perfusion. Gastroenterology 71: , Altmann GG: nfluence of bile and pancreatic secretions on the size of the intestinal villi in the rat. Am J Anat 132: , Heller R, Tawil T, Weser E: Effect of bile and pancreatic secretions on mucosal hyperplasia after small bowel resection (abstr). Clin Res 22:21A, Sum PT, Preshaw RM: ntra-duodenal glucose infusion and pancreatic secretion in man. Lancet 2:34-341, Johnson LR, Lichtenberger LM, Copeland EM, et al: Action of gastrin on gastrointestinal structure and function. Gastroenterology 68: , Yu YM, Yu LCC, Chou CC: Distribution of blood flow in the intestine with hypertonic glucose in the lumen. Surgery 78:52-525, Creutzfeldt W, Feurle G, Ketterer H: Effect of gastrointestinal hormones on insulin and glucagon secretion. N Eng J Med 282: , Cataland S, Crockett SE, Brown JC, et al: Gastric inhibitory polypeptide stimulation by oral glucose in man. J Clin Endocrinol Metab 39: , 1974