THE SMALL intestinal mucosa exhibits an extremely

Transcription

1 /98/$03.00/0 Vol. 139, No. 7 Endocrinology Printed in U.S.A. Copyright 1998 by The Endocrine Society Intestinal Adaptation after Extensive Small Bowel Resection: Differential Changes in Growth and Insulin- Like Growth Factor System Messenger Ribonucleic Acids in Jejunum and Ileum* THOMAS R. ZIEGLER. MARK P. MANTELL, JESSIE C. CHOW, JOHN L. ROMBEAU, AND ROBERT J. SMITH Department of Medicine, Division of Endocrinology and Metabolism, Emory University School of Medicine (T.R.Z.), Atlanta, Georgia 30322; the Department of Surgery, Hospital of the University of Pennsylvania (M.P.M., J.L.R.), Philadelphia, Pennsylvania 19104; and Joslin Research Laboratory, Harvard Medical School (J.C.C., R.J.S.), Boston, Massachusetts ABSTRACT The distal small bowel exhibits greater adaptive growth than proximal segments after partial small intestine resection. To explore this process, we evaluated adaptive cellularity, intestinal insulin-like growth factor (IGF) system messenger RNA (mrna) transcripts, and effects of recombinant IGF-I treatment in jejunum and ileum of adult rats. Gastrostomy-fed animals underwent 80% jejuno-ileal resection or intestinal transection and reanastomosis without resection, followed by infusion of human recombinant IGF-I (2.4 mg/kg day) or vehicle. After 7 days, resected rats demonstrated modest adaptive growth in jejunum and marked cell proliferation in ileum. Resection increased IGF-I mrna in both jejunum (183%) and ileum (249%) and THE SMALL intestinal mucosa exhibits an extremely rapid rate of cell turnover, which exceeds that of most body tissues (1). After partial small bowel resection, residual intestinal cells undergo an accelerated growth and renewal process, termed intestinal adaptation. The adaptive response is characterized by increased villus height and crypt depth as a result of accelerated mucosal cell proliferation, smooth muscle cell hypertrophy, and dilation and lengthening of the remnant intestine (2, 3). The extent to which the residual intestine adapts after resection, and thus increases nutrient and fluid absorptive capacity, is the ultimate determinant of clinical outcome in patients with short bowel syndrome (4). Although intestinal adaptation has not been well characterized in humans, it has been studied in animal models of extensive small bowel and/or colonic resection (5 9). In rats, jejuno-ileal resection alone induces adaptive growth responses in the small intestine and the colon, with the most Received February 5, Address all correspondence and requests for reprints to: Robert J. Smith, M.D., Metabolism Section, Joslin Diabetes Center, One Joslin Place, Boston, Massachusetts smithr@joslab.harvard.edu. * This work was supported in part by NIH Clinical Associate Physician Award 3-MO1-RR S1 and an Emory University Research Committee Grant (to T.R.Z.); NIH Grants DK-43038, DK-48503, and DK and Trauma Center Grant GM (to R.J.S.); Diabetes and Endocrinology Research Center Grant DK (to the Joslin Diabetes Center); a V.A. Merit Review Grant (to J.L.R.); and NIH Training Grant DK and a Mary K. Iacocca Foundation Postdoctoral Fellowship (to J.C.C.). up-regulated IGFBP-4 mrna levels in both tissues. IGFBP-3 mrna fell significantly in ileum after resection. IGF-I infusion modestly increased ileal cellularity after resection, but had no effect in jejunum. IGF-I markedly increased IGFBP-3 mrna levels in jejunum after both transection and resection. These data confirm that bowel resection induces greater adaptive growth in ileum than jejunum. IGF-I administration modestly increases ileal, but not jejunal, growth after resection. Increased levels of intestinal IGF-I and IGFBP-4 mrna suggest roles for IGF-I and IGFBP-4 in mediating small bowel adaptation. Higher levels of jejunal IGFBP-3 mrna may be related to limited jejunal vs. ileal growth after extensive jejuno-ileal resection. (Endocrinology 139: , 1998) vigorous proliferative response occurring in the residual small bowel (2). The ileum exhibits a more robust adaptive response than the duodenum or jejunum (5, 8). The ileum also undergoes significant adaptive changes after partial colonic resection. These findings suggest that the residual ileum plays a particularly important role in adaptive nutrient absorption after intestinal resection (2). The underlying cellular mechanisms that lead to greater adaptive growth in distal vs. proximal small intestine in short bowel syndrome remain uncertain. The extent of augmented intestinal cell proliferation and function after bowel resection is affected by route of feeding (enteral greater than parenteral) (4), specific gut-trophic nutrients such as glutamine (9), the content of gastric and pancreatic-biliary luminal secretions (2, 3), and endogenous gut hormones (1). The messenger RNAs (mrnas) for several peptide growth factors are up-regulated in intestinal cells after partial small bowel resection. These include enteroglucagon (10 12), cholecystokinin (12), and insulin-like growth factor-i (IGF-I) (6 9). Exogenous systemic administration of recombinant hormones, including epidermal growth factor, GH, and IGF-I, increase intestinal cell proliferation and improve intestinal absorptive function after bowel resection (4, 6 7, 9, 11, 13, 14). IGF-I and IGF-binding proteins (IGFBPs) appear to be localized at specific cellular sites throughout the intestine, suggesting that the IGF system mediates important cellular functions in gut tissue (13 17). Recently, several lines of 3119

2 3120 IGF-I SYSTEM IN SMALL INTESTINAL ADAPTATION Endo 1998 Vol 139 No 7 evidence in nonintestinal tissues suggest that IGFBP-3 is inhibitory to cell growth (18). Studies in the HT-29 human colon cell line suggest that IGFBP-4 also has growth inhibitory effects (19). These findings raise the possibility that local expression of IGFBPs may influence small intestinal adaptation after gut resection. To further evaluate the potential role of the IGF system in adaptive growth of proximal and distal small intestine, we studied jejunal and ileal cellularity and concomitant changes in endogenous IGF-I, IGFBP-3, and IGFBP-4 mrnas in a rat model of massive small bowel resection in the presence or absence of recombinant IGF-I treatment. Animals Materials and Methods The animal protocols and procedures were approved by the institutional animal care and use committees of the University of Pennsylvania and the Joslin Diabetes Center. Male Sprague-Dawley rats, weighing g (Charles River Laboratories, Portage, MI), were acclimated to laboratory conditions for several days and fed standard rat food (Ralston Purina, St. Louis, MO) and water ad libitum. Operative procedures Rats underwent laparotomy for 80% jejuno-ileal resection or transection and gastrostomy placement procedures as previously described (9). Briefly, two groups of rats underwent jejunal transection and reanastomosis without bowel resection at a point 10 cm distal to the ligament of Treitz. Two additional groups underwent an 80% jejuno-ileal resection extending from 10 cm distal to the ligament of Treitz to 10 cm proximal to the ileo-cecal valve. Concurrent with the intestinal operation, a SILASTIC brand gastrostomy tube (0.03 in. id; Dow Corning, Midland, MI) was placed and secured (9). Next, an osmotic minipump (model 2001, Alzet, Palo Alto, CA), containing either IGF-I and vehicle or vehicle alone, was inserted into a dorsal sc pocket. Treatment regimens Gastrostomy feeding with a polymeric liquid formula diet (Enercal Plus, Wyeth, Radnor, PA) was used to control dietary intake. The diet at full strength provided 1.5 Cal/ml and g protein/ml, with a caloric distribution of 55% carbohydrate, 15% protein, and 30% fat. Gastrostomy feedings were administered in all animals as continuous 24-h infusions via a Harvard pump syringe system (Harvard Apparatus, Natick, MA) on the following schedule: normal saline at 1.25 ml/h for 18 h after surgery, half-strength diet at 1.25 ml/h on postoperative day 1, and full-strength diet at 2.5 ml/h from postoperative day 2 until the end of the study period on day 7. The diet formula as administered from days 2 7 provided approximately 360 Cal/kg day and 14 g protein/ kg day (2.24 g nitrogen/kg day), which is adequate for growing adult rats (6, 9). Water was provided ad libitum. The transected and 80% bowel-resected rats were randomly assigned to receive 7-day continuous infusions of either recombinant human IGF-I (2.4 mg/kg day in 0.1 m acetic acid) or an equal volume of 0.1 m acetic acid vehicle. Thus, the four groups of animals studied were: 1) transected/vehicle (n 8), 2) transected/igf-i (n 9), 3) resected/ vehicle (n 8), and 4) resected/igf-i (n 7). Body weight and tissue isolation The body weights of the rats were determined preoperatively after the acclimation period and at the end of the study, before tissue removal. After the 7-day treatment period, animals were anesthetized with ip sodium pentobarbital (50 mg/kg), and a laparotomy was performed. Plasma was obtained from the inferior vena cava and stored at 20 C for later analysis. The small intestine was rapidly removed from peritoneal and vascular connections from a point just proximal to the ligament to Treitz to the ileal-cecal junction, and the lumen was gently flushed with ice-cold saline to remove residual liquid diet and intestinal secretions. The intestine was then suspended from a ring stand with a distal 5-g weight to enable consistent identification of appropriate jejunal and ileal segments, and the surgical anastomosis was located. A jejunal segment (9 cm) was removed from the ligament of Treitz to a point 1 cm proximal to the anastomosis, and an ileal segment (9 cm) was removed from the ileal-cecal junction to a point 1 cm distal to the anastomosis. The proximal 2 cm of the jejunal segment and the distal 2 cm of the ileal segment were isolated, weighed, and frozen in liquid N 2 for subsequent protein and DNA analysis. The remaining 7 cm of the jejunal and ileal segments were frozen in liquid N 2 for subsequent mrna studies. The tissue segments used for mrna studies were stored at 80 C, and the segments to be used for cellularity indexes were stored at 20 C. After blood and tissue removal, the animals were killed by exsanguination. Plasma IGF-I assay Total plasma IGF-I concentration was measured after acid-ethanol extraction with an RIA kit (Nichols Institute, San Juan Capistrano, CA). Previous studies in rats have shown that the acid-ethanol extraction method almost completely removes the major circulating IGF-I carrier protein, IGFBP-3, and yields values comparable to those obtained by HPLC (20). Intestinal weight, and protein and DNA contents The total combined weight per cm of the 2- and 7-cm jejunal and ileal segments was calculated as an index of the mass of these segments. Jejunal and ileal protein and DNA content per cm were determined as indexes of gut cellularity, as previously described (2, 5, 9). The jejunal and ileal protein/dna ratio was calculated as an index of relative changes in cellular hypertrophy and hyperplasia during gut adaptation. Jejunal and ileal mrna analysis Frozen intestine was pulverized in liquid N 2, and total RNA was extracted according to the method of Chomczynski and Sacchi (21) after homogenization with a Polytron (Brinkmann Instruments, Westbury, NY) in denaturing buffer containing guanidine thiocyanate, N-lauroylsarcosine, and -mercaptoethanol. For Northern blotting, 30- g aliquots of jejunal and ileal RNA from individual animals were resolved in formaldehyde-agarose gels, and the integrity of the RNA was confirmed by ethidium bromide staining. The RNA then was transferred to nylon filters (GeneScreen Plus, New England Nuclear, Boston, MA) by overnight blotting, fixed by UV cross-linking, and hybridized as described previously with specific rat complementary DNA (cdna) probes (6, 9). IGF-I mrna was determined using an 800-bp IGF-I fragment (from Dr. P. Rotwein, Washington University School of Medicine, St. Louis, MO). This particular IGF-I cdna probe was previously shown to be effective in demonstrating steady state IGF-I mrna transcripts at 7.8, , and kb in rat intestine using similar experimental methods (6, 9). IGF-I receptor mrna levels were assessed using a 781-bp fragment cloned in our laboratory, which spans portions of the - and -subunits of the rat IGF-I receptor (22). IGFBP-3 and IGFBP-4 mrnas were determined using specific 699- and 444-bp fragments, respectively (from Dr. S. Shimisaki, Whittier Institute, La Jolla, CA). After washing successively in 2 SSC (standard saline citrate)-0.1% SDS for 20 min at 22 C, in 1 SSC-0.1% SDS for 30 min at 22 C, and finally in 0.1 SSC-0.1% SDS at 42 C for min, specific mrna bands were identified and quantitated using a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). To verify equal RNA loading, the filters were stored until the 32 P label decayed to background and then rehybridized with a 2.0-kb mouse 18S ribosomal cdna probe (from P. Bachierre, Centre de Recherche de Biochimie et Génétique Cellulaires, Toulouse, France). Statistical analysis The study was arranged in a 2 2 factorial design, with resection and IGF-I treatment as main effects. Treatment groups were compared using two-factor factorial ANOVA, which analyzed the main effects of resection and IGF-I administration and their interaction. P values for the two-factor ANOVA analyses are presented in the text. One-factor ANOVA was used to detect significant intergroup differences. In this case, specific study groups (transected/vehicle, transected/igf-i, re-

3 IGF-I SYSTEM IN SMALL INTESTINAL ADAPTATION 3121 sected/vehicle, and resected/igf-i) were compared post-hoc using the Fisher s protected least significant difference (PLSD) test. P values for the one-factor ANOVA analyses are shown in the tables and figures. All data are presented as the mean se. P 0.05 was considered statistically significant. Results Postoperative body weight change The same volume of liquid diet was infused daily via gastrostomy in the four study groups. Resected animals developed mild diarrhea after initiation of gastrostomy feeding. Stool volumes decreased over the subsequent few days and were not quantitated. Transected/vehicle-infused rats lost a small amount of body weight postoperatively (not significant), demonstrating a failure of weight gain as a consequence of the catabolic effects of the operative procedures (Table 1). Small intestinal resection significantly decreased body weight (P 0.01), reflecting loss of gut tissue, catabolic effects of operation, and probable malabsorption due to short bowel syndrome. IGF-I administration significantly attenuated body weight loss in both transected and resected groups (P 0.05). Plasma IGF-I TABLE 1. Effects of small bowel resection and IGF-I treatment on postoperative body weight change and plasma IGF-I concentrations Study group BW change Plasma IGF-I Transected/vehicle Transected/IGF-I a,b Resected/vehicle a,c Resected/IGF-I d a,b See text for details of operations and IGF-I and vehicle regimens. Data are the mean SE. Body weight change represents the difference between the immediate preoperative weight and the body weight at death on day 7. The sample sizes were: transected/vehicle, n 8; transected/igf-i, n 9; resected/vehicle, n 8; and resected/igf-i, n 7. Significance was determined by one-factor ANOVA and Fisher s PLSD test. a P 0.05 vs. transected/vehicle. b P 0.05 vs. resected/vehicle. c P 0.05 vs. transected/igf-i. d P 0.05 vs. all other groups. TABLE 2. Effects of small bowel resection and IGF-I treatment on jejunal and ileal growth indexes Study group Wet wt (mg/2 cm) Jejunal segment Protein (mg/2 cm) DNA ( g/2 cm) Bowel resection resulted in a small decrease in the mean plasma IGF-I concentration, which was not statistically significant (22% lower in transected/vehicle vs. resected/vehicle groups; Table 1). Circulating IGF-I levels increased approximately 2-fold with IGF-I administration in both transected and resected groups (P 0.01). Intestinal wet weight, and protein and DNA contents Small intestinal growth indexes are shown in Table 2. The jejunum and ileum exhibited distinct adaptive growth responses after partial small intestinal resection and IGF-I administration. With resection, jejunal wet weight increased significantly (38% greater in resected/vehicle vs. transected/ vehicle groups), but there was a more marked increase in ileum wet weight (77% greater in resected/vehicle vs. transected/vehicle groups). Protein content per cm increased to a similar extent in jejunum and ileum (53% and 59% greater with resection in the same groups, respectively), whereas there was much more of a postresection increase in DNA content per cm in ileum than jejunum (204% increase in ileum and 21% in jejunum in resected/vehicle vs. transected/vehicle groups). There was a small, but statistically significant (27%), increase in the protein/dna ratio in jejunum postresection, suggesting cellular hypertrophy. This contrasted with a 55% decrease in the protein/dna ratio in ileum, consistent with hyperplasia and a decrease in average cell size. These differences between jejunum and ileum also were evident in the resected/igf-i vs. transected/igf-i experimental groups. IGF-I treatment did not significantly alter any of the jejunal cellularity indexes. By contrast, IGF-I increased ileal wet weight and protein content per cm in both resected and transected control animals. There was no significant interaction between resection and IGF-I effects in the ileum by ANOVA, supporting the conclusion that the IGF-I-stimulated increase in wet weight and protein per cm was additive to the effect of resection. IGF system mrna content IGF-I mrna transcripts were evident at 7.8, , and kb on Northern blots of total RNA preparations from jejunum and ileum (Fig. 1, A and B). Quantitative PhosphorImager data on the predominant 7.8-kb species showed that small intestinal resection significantly increased IGF-I steady state mrna levels in both jejunum and ileum (83% and 149% increases, respectively). Administration of IGF-I to transected rats significantly increased jejunal IGF-I mrna (64%). There was a similar 58% increase in mean ileal IGF-I mrna with IGF-I treatment, but this was not statistically Protein/DNA ratio Wet wt (mg/2 cm) Protein (mg/2 cm) Ileal segment DNA ( g/2 cm) Protein/DNA ratio Transected/vehicle Transected/IGF-I Resected/vehicle a,b a,b d a a,b a,b Resected/IGF-I a,b a,b a,b d d d a,b See text for details of operations and IGF-I and vehicle regimens. Data are the mean SE. The sample sizes were: transected/vehicle, n 8; transected/igf-i, n 9; resected/vehicle, n 8; and resected/igf-i, n 7. Significance was determined by one-factor ANOVA and Fisher s PLSD test. a P 0.05 vs. transected/vehicle. b P 0.05 vs. transected/igf-i. c P 0.05 vs. all other groups.

4 3122 IGF-I SYSTEM IN SMALL INTESTINAL ADAPTATION Endo 1998 Vol 139 No 7 FIG. 1. A, IGF-I mrna in rat jejunum 7 days after bowel transection or resection and vehicle or IGF-I infusion. B, IGF-I mrna in rat ileum under the same experimental conditions. Northern analysis of steady state mrna levels was performed as described in Materials and Methods. The left panels show representative Northern blots from the four experimental conditions in jejunum and ileum (A and B, respectively). Quantitation of the major 7.8-kb IGF-I mrna transcript is shown on the right. Data are expressed in arbitrary units as a percentage of the control (transected/vehicle) values, corrected for corresponding 18S ribosomal mrna determined on the same blots after the IGF-I probe had decayed to background. The sample sizes were: transected/vehicle, n 8; transected/igf-i, n 9; resected/vehicle, n 8; and resected/igf-i, n 7. *, P 0.05 vs. transected/vehicle;, P 0.05 vs. transected/igf-i (by onefactor ANOVA and Fisher s PLSD test). significant. No effect of IGF-I administration on IGF-I mrna was evident in the jejunum of resected rats, and there was a modest, but significant (22%), decrease in IGF-I mrna in the ileum of resected/igf-i vs. resected/vehicle rats. IGF-I receptor mrna also was identified and quantified by Northern blotting. The levels of the 11-kb IGF-I receptor mrna transcript exhibited considerable variability between individual animals, without significant resection or IGF-I treatment main effects by ANOVA (Table 3). With combined resection and IGF-I administration, the jejunal IGF-I receptor mrna content was significantly lower than that in the transected/igf-i group (22%). There was a somewhat greater decrease in the mean ileum IGF-I receptor mrna in the same groups (49%), but this was not significant. Intestinal resection did not have a significant main effect on IGFBP-3 mrna in the jejunum (Fig. 2A), but reduced IGFBP-3 mrna levels in the ileum (P 0.01; Fig. 2B). IGF-I administration markedly increased jejunal IGFBP-3 mrna levels (203% greater in transected/igf-i vs. transected/vehicle and 90% greater in resected/igf-i vs. resected/vehicle). By contrast, ileal IGFBP-3 mrna was unaltered by IGF-I administration in the absence or presence of resection. IGFBP-4 mrna levels in jejunum were significantly increased by both intestinal resection and IGF-I treatment (P 0.03; Fig. 3A) without interaction by ANOVA. In the ileum,

5 IGF-I SYSTEM IN SMALL INTESTINAL ADAPTATION 3123 TABLE 3. Effects of small bowel resection and IGF-I treatment on jejunal and ileal IGF-I receptor mrna levels Study group Jejunal segment Ileal segment Transected/vehicle Transected/IGF-I Resected/vehicle Resected/IGF-I 75 3 a See text for details of operations and IGF-I and vehicle regimens. Data are the mean SE, expressed as a percentage of transected/ vehicle (control) values. Content of mrna in arbitrary units is normalized to a percentage of the control values on each Northern blot. Data were corrected for RNA loading by quantifying 18S ribosomal mrna. The sample sizes were: transected/vehicle, n 8; transected/ IGF-I, n 9; resected/vehicle, n 8; and resected/igf-i, n 7. Significance was determined by one-factor ANOVA and Fisher s PLSD test. a P 0.05 vs. transected/igf-i. resection significantly increased IGFBP-4 mrna (P 0.04; Fig. 3B), but there was no effect of IGF-I. Discussion The current study is the first to evaluate concomitant changes in growth responses and the levels of IGF system mrnas in the proximal and distal small intestine after bowel resection and IGF-I treatment. We used an established model of 80% small bowel resection to study the potential role of the IGF system in small intestinal adaptation (7, 23, 24). Our experimental design incorporated vehicle- and IGF-I-treated animals and controls subjected to bowel transection and reanastomosis without removal of intestinal tissue. The transected controls allowed an evaluation of IGF-I treatment effects on gut growth and IGF-I system steady state mrnas after intestinal operation without consequent short bowel syndrome. Nutritional status and enteral nutrient intake markedly affect intestinal growth, plasma IGF-I concentrations, and tissue content of IGF-I system mrnas (16, 25). Therefore, we used continuous gastrostomy feeding to control enteral nutrient intake in our study animals. The catabolic effect of the laparotomy and gastrostomy procedures was demonstrated by slight postoperative body weight loss rather than the expected weight gain in the transected control rats despite the administration of an adequate tube feeding diet. The much greater weight loss in the animals undergoing intestinal resection reflected both the increased initial removal of intestinal tissue and the subsequent consequences of short bowel syndrome with its associated malabsorption. IGF-I treatment decreased weight loss in the resected and transected animals. Systemic IGF-I administration in rats has previously been shown to enhance intestinal growth (23, 26) and improve nutrient absorption after partial intestinal resection (6, 27). In pigs, oral administration of IGF-I has been shown to affect intestinal cell differentiation, as evidenced by increased small intestinal disaccharidase activity (28). Therefore, body weight gain with IGF-I in this study may be due to increased nutrient absorption as well direct IGF-I effects on tissue growth. Circulating IGF-I levels are often decreased in catabolic and postoperative states (29), and the mean plasma IGF-I concentration was lower in the resected than in the transected rats in this study. Although this difference was not statistically significant, it is consistent with a greater degree of catabolic stress and malabsorption in the resected animals. The increase in plasma IGF-I levels with IGF-I infusion in both the transected and resected rats suggests a possible beneficial role for IGF-I administration in maintaining plasma levels after major intestinal operation. We measured wet weight and the protein and DNA contents of residual bowel segments as indexes of gut adaptation in response to intestinal resection (2, 5, 30). Wet weight and DNA content increased after resection to a greater extent in ileum than in jejunum, suggesting more marked gut adaptation in the distal residual small intestine. Further, the postresection protein/dna ratio was significantly increased in jejunum, but was markedly decreased in ileum. These differential changes are consistent with a greater degree of adaptive hyperplasia in the ileum than the jejunum, even though the proximal and distal small bowel segments were anastomosed in continuity. Different jejunal and ileal responses also were evident in IGF-I-treated animals. IGF-I did not alter jejunal weight, but significantly increased ileal wet weight in both transected and resected rats. In resected rats treated with IGF-I, DNA content per cm increased to a much greater extent in residual ileum than in jejunum (304% vs. 21%). IGF-I administration also modestly increased ileal protein content postresection, whereas residual jejunum was unresponsive to IGF-I. Although the residual ileum was clearly responsive to exogenous IGF-I, the trophic effects of IGF-I were of much smaller magnitude than the adaptive growth response in the ileum induced by resection. Of interest, anabolic effects of IGF-I in the ileum were greater in resected than transected animals. Thus, endogenous factors induced by resection (1, 10 12) appear to interact with exogenous IGF-I to facilitate ileal growth after massive small bowel resection. The differences that we observed in proximal vs. distal small bowel growth are consistent with the results of a previous study on gut adaptation in the mucosa of younger rats (7). However, the findings in the previous study differ from ours in that the duodenal-jejunal mucosa exhibited a greater growth response to IGF-I treatment than the ileum (7). In intact unoperated rats, IGF-I also was reported to be more trophic to proximal than distal small bowel segments (31). The distinct tissue responses to IGF-I in our study compared with previous reports may result from a number of factors, including characteristics of the specific experimental model, animal age, diet, and the IGF-I dose. Differential responses to resection and growth factor administration in mucosal vs. full thickness gut tissue also may occur, as different cell types populate the mucosal epithelium and submucosal and muscular layers (25). A major consequence of intestinal adaptation is an increase in mucosal surface area and intestinal absorptive capacity; therefore, studies that quantitate nutrient absorption in specific gut segments after IGF-I therapy will be of interest (6, 26). The underlying mechanisms of intestinal adaptation after resection remain uncertain, but are probably multifactorial and interactive (1, 9). Small bowel resection resulted in marked increases in both jejunal and ileal IGF-I mrna abundance (to 183% and 249% of the control value, respectively). Increased tissue IGF-I mrna levels were maintained even

6 3124 IGF-I SYSTEM IN SMALL INTESTINAL ADAPTATION Endo 1998 Vol 139 No 7 FIG. 2. A, IGFBP-3 mrna in rat jejunum 7 days after bowel transection or resection and vehicle or IGF-I infusion. B, IGFBP-3 in rat ileum under the same experimental conditions. Northern analysis of steady state mrna levels was performed as described in Materials and Methods. The left panel shows representative Northern blots from the four experimental conditions. Quantitation of the 2.4-kb IGFBP-3 mrna transcript is shown on the right. Data are expressed in arbitrary units as a percentage of the control (transected/ vehicle) values, corrected for corresponding 18S ribosomal mrna determined on the same blots after the IGFBP-3 probe had decayed to background. The sample sizes were: transected/vehicle, n 8; transected/ IGF-I, n 9; resected/vehicle, n 8; and resected/igf-i, n 7. *, P 0.05 vs. transected/vehicle,, P 0.05 vs. resected/vehicle (by one-factor ANOVA and Fisher s PLSD test). after 7 days of IGF-I administration. A greater increase in IGF-I mrna in ileum vs. jejunum correlated with the different cell proliferation responses in these tissues as assessed by DNA content. A recent study has shown that IGF-I mrna is localized mainly in the small intestinal proliferative crypt cell zone in adult rats (32). Our study together with previous reports (6, 9, 17, 27, 32) are consistent with a role for endogenous IGF-I in the hypertrophic and hyperplastic processes in jejunum and ileum that are induced by massive small intestinal resection. Previous studies have shown that gut IGF-I receptors are especially abundant in the small intestinal crypt region (11), suggesting that crypt cells may be primary targets for IGF-I action. A consistent pattern of IGF-I receptor mrna regulation was not evident in our experimental animals. There was a significant decrease in the jejunum and a decrease in the mean IGF-I receptor mrna in the ileum that was not significant in the IGF-I-treated resected rats, which may reflect hormone-induced receptor down-regulation. However, receptor mrna was not decreased in IGF-Itreated transected animals. Recent data suggest that circulating or local tissue IGFBPs modulate IGF-I action and may affect tissue growth independent of an interaction with IGF-I (18). In the current study, steady state IGFBP-4 mrna levels rose markedly in both jejunum and ileum after bowel resection. This is consistent with other recent studies, in which we have demonstrated increased adaptive ileal and colonic growth in association with increased IGFBP-4 mrna levels (6, 9). In situ hybridization studies in fetal intestine have shown that IGFBP-4 mrna is localized to the proliferative crypt epithelium (15). Although IGFBP-4 was shown to inhibit growth in a human colonic cell line (19), the association between cell proliferation and increased local IGFBP-4 mrna during adaptive hyperplasia suggests a relationship between intestinal growth responses and IGFBP-4 or processes that lead to increased IGFBP-4 mrna. Resection and IGF-I treatment had markedly different effects on intestinal IGFBP-3 and IGFBP-4 mrnas. In contrast to the postresection increase in IGFBP-4 mrna, IGFBP-3 mrna levels were unchanged in the jejunum and significantly decreased in the ileum after resection. Recent evidence strongly suggests that intact IGFBP-3 or its pro-

7 IGF-I SYSTEM IN SMALL INTESTINAL ADAPTATION 3125 FIG. 3. A, IGFBP-4 mrna in rat jejunum 7 days after bowel transection or resection and vehicle or IGF-I infusion. B, IGFBP-4 mrna in rat ileum under the same experimental conditions. Northern analysis of steady state mrna levels was performed as described in Materials and Methods. The left panel shows representative Northern blots from the four experimental conditions in jejunum and ileum (A and B, respectively). Quantitation of the 2.2-kb IGFBP-4 mrna transcript is shown on the right. Data are expressed in arbitrary units as a percentage of the control (transected/ vehicle) values, corrected for corresponding 18S ribosomal mrna determined on the same blots after the IGFBP-4 probe had decayed to background. The sample sizes were: transected/vehicle, n 8; transected/igf-i, n 9; resected/vehicle, n 8; and resected/igf-i, n 7. *, P 0.05 vs. transected/vehicle;, P 0.05 vs. transected/igf-i (by one-factor ANOVA and Fisher s PLSD test). teolytic fragments may inhibit IGF-I-dependent and independent cell growth and proliferation in nonintestinal tissues (18, 33 34). Thus, diminished postresection intestinal IGFBP-3 mrna in the ileum, as observed in this study, and a consequent decrease in local levels of the binding protein may contribute to accelerated intestinal growth. This could occur through increased local free IGF-I concentrations (35) or decreased direct growth inhibitory effects of IGFBP-3 (18). A striking difference between jejunum and ileum was a very marked ( 3-fold) increase in jejunal, but not ileal, IGFBP-3 mrna in both transected and resected animals after IGF-I treatment. The increase in jejunal IGFBP-3 mrna after IGF-I treatment is consistent with a recent report demonstrating elevated jejunal IGFBP-3 mrna in IGF-I transgenic mice (17). Increased local synthesis of IGFBP-3 in the jejunum, but not the ileum, in response to endogenous and exogenous IGF-I may partially explain the apparent resistance to IGF-I-mediated adaptive growth in the jejunum. Further studies that evaluate IGFBP-3 protein levels and cellular sites of IG- FBP-3 mrna expression will help determine whether cellspecific expression of IGFBP-3 influences gut adaptation. In summary, this study confirms that different adaptive growth responses occur in proximal and distal residual small intestine after extensive small bowel resection. In addition, we have shown that IGF-I modestly increases ileal, but not jejunal, adaptation after resection. Specific alterations in the expression of IGF system mrnas were demonstrated in jejunum and ileum, suggesting potentially important roles for both the endogenous IGF system and exogenous IGF-I in mediating intestinal adaptation. Additional studies on IGF-I and IGFBP mrna expression in specific gut segments and

8 3126 IGF-I SYSTEM IN SMALL INTESTINAL ADAPTATION Endo 1998 Vol 139 No 7 cell types should help to define the role of the IGF system in intestinal growth, function, and postresection adaptation. Acknowledgments The authors gratefully acknowledge the assistance of Jane Ann Denney, Velta Ramolins, and Karen TenDyke of the Joslin Diabetes Center (Boston, MA), and Wei Zhang and Allison Bain of the Department of Surgery, Hospital of the University of Pennsylvania (Philadelphia, PA). The authors also thank the Chiron Corp. (Emeryville, CA) for the donation of IGF-I. References 1. Johnson LR 1988 Regulation of gastrointestinal mucosal growth. Physiol Rev 68: Dowling RH 1983 Small bowel adaptation and its regulation. Scand J Gastroenterol [Suppl 74] 17: Williamson RCN 1978 Intestinal adaptation. N Engl J Med 298: Byrne TA, Persinger RL, Young LS, Ziegler TR, Wilmore DW 1995 A new treatment for patients with the short bowel syndrome: growth hormone, glutamine and a modified diet. Ann Surg 222: Booth CC, Evans KT, Menzies T 1959 Intestinal hypertrophy following partial resection of the small bowel in the rat. Br J Surg 46: Mantell MP, Ziegler TR, Roth BA, Zhang W, Adamson WT, Bain A, Chow JC, Smith RJ, Rombeau JL 1995 Resection-induced colonic adaptation is augmented by IGF-I and associated with upregulation of colonic IGF-I mrna. Am J Physiol 269:G974 G Vanderhoof JA, McCusker RH, Clark R, Mohammadpour H, Blackwood DJ, Harty RF, Park JHY 1992 Truncated and native insulinlike growth factor I enhance mucosal adaptation after jejunoileal resection. Gastroenterology 102: Weser E, Hernandez MH 1971 Studies of small bowel adaptation after intestinal resection in the rat. Gastroenterology 60: Ziegler TR, Mantell MP, Chow JC, Rombeau JL, Smith RJ 1996 Gut adaptation and the insulin-like growth factor system: regulation by glutamine and insulin-like growth factor-i administration. Am J Physiol 271:G866 G Rountree DB, Ulshen MH, Selub S, Fuller CR, Bloom SR, Ghatei MA, Lund PK 1992 Nutrient-independent increases in proglucagon and ornithine decarboxylase messenger RNAs after jejunoileal resection. Gastroenterology 103: Taylor RG, Fuller PJ 1994 Humoral regulation of intestinal adaptation. Balliere Clin Endocrinol Metab 8: Taylor RG, Beveridge DJ, Fuller PJ 1992 Expression of ileal glucagon and peptide tyrosine-tyrosine genes: response to inhibition of polyamine synthesis in the presence of massive small bowel resection. Biochem J 286: Han VK, D Ercole AJ, Lund PK 1987 Cellular localization of somatomedin (insulin-like growth factor) messenger RNA in the human fetus. Science 236: Hill DJ, Clemmons DR 1992 Similar distribution of insulin-like growth factor binding proteins-1, -2, -3 in human fetal tissues. Growth Factors 6: Delhanty PJD, Hill DJ, Shimisaki S, Han VKM 1993 Insulin-like growth factor binding protein -4, -5 and -6 mrnas in the human fetus: localization to sites of growth and differentiation? Growth Regul 3: Winesett DE, Ulshen MH, Hoyt EC, Mohapatra NK, Fuller CR, Lund PK 1995 Regulation and localization of the insulin-like growth factor system in small bowel during altered nutrient status. Am J Physiol 268:G631 G Ohneda K, Ulshen MH, Fuller CR, D Ercole AJ, Lund PK 1997 Enhanced growth of small bowel in transgenic mice expressing human insulin-like growth factor-i. Gastroenterology 112: Rechler MM 1997 Growth inhibition by insulin-like growth factor (IGF) binding protein-3: what s IGF got to do with it? Endocrinology 138: Singh P, Dai B, Dhruva B, Widen SG 1994 Episomal expression of sense and antisense insulin-like growth factor (IGF)-binding protein-4 complementary DNA alters the mitogenic response of a human colon cancer cell line (HT-29) by mechanisms that are independent of and dependent upon IGF-I. Cancer Res 54: Crawford BA, Martin JL, Howe CJ, Handelsman DJ, Baxter RC 1992 Comparison of extraction methods for insulin-like growth factor-i in rat serum. J Endocrinol 134: Chomcyznski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: Pedrini MF, Giorgino F, Smith RJ 1994 cdna cloning of the rat IGF-I receptor: structural analysis of the rat and human IGF-I and insulin receptors reveals differences in alternative splicing and receptor-specific domain conservation. Biochem Biophys Res Commun 202: Lemmey AB, Martin AA, Read LC, Tomas FM, Owens PC, Ballard FJ 1991 IGF-I and the truncated analogue des-(1 3) IGF I enhance growth in rats after gut resection. Am J Physiol 260:E213 E Lemmey AB, Ballard FJ, Martin AA, Tomas FM, Howarth GS, Read LC 1994 Treatment with IGF-I peptides improves function of the remnant gut following small bowel resection in rats. Growth Factors 10: Ziegler TR, Almahfouz A, Pedrini MT, Smith RJ 1995 A comparison of rat small intestinal insulin and IGF-I receptors during fasting and refeeding. Endocrinology 136: Lo H-C, Ney DM 1996 GH and IGF-I differentially increase protein synthesis in skeletal muscle and jejunum of parenterally fed rats. Am J Physiol 271:E872 E Zhang W, Frankel WL, Adamson WT, Roth JA, Mantell MP, Bain A, Ziegler TR, Smith RJ, Rombeau JL 1995 Insulin-like growth factor I (IGF-I) improves mucosal structure and function in transplanted rat small intestine. Transplantation 59: Houle VM, Schroeder EA, Odle J, Donovan SM 1997 Small intestinal disaccharidase activity and ileal villus height are increased in piglets consuming formula containing recombinant human insulin-like growth factor-i. Pediatr Res 42: Wojnar MM, Fan J, Frost RA, Gelato MC, Lang CH 1995 Alterations in the insulin-like growth factor system in trauma patients. Am J Physiol 268:R970 R Wilson HD, Miller T, Ogesen B, Schedl HP, Failla ML, Loven DP 1986 Adaptation of the duodenum and ileum of the rat to mid-gut resection: enzyme activity and trace metal status. Am J Clin Nutr 43: Steeb CB, Trahair JF, Tomas FM, Read LC 1994 Prolonged administration of IGF peptides enhances growth of gastrointestinal tissues in normal rats. Am J Physiol 266:G1090 G Dvorak B, Stephana AL, Holubec H, Williams CS, Phillips AF, Koldovsky O 1997 Insulin-like growth factor-i (IGF-I) mrna in the small intestine of suckling and adult rats. FEBS Lett 388: Oh Y, Muller HL, Ng L, Rosenfeld RG 1995 Transforming growth factor- induced cell growth inhibition in human breast cancer cells is mediated through insulin-like growth factor binding protein-3 action. J Biol Chem 270: Zadeh JL, Binoux M 1997 The 16-kDa proteolytic fragment of insulin-like growth factor (IGF) binding protein-3 inhibits the mitogenic action of fibroblast growth factor on mouse fibroblasts with a targeted disruption of the IGF-I receptor gene. Endocrinology 138: Albiston AL, Taylor RG, Herington AC, Beveridge DJ, Fuller PJ 1992 Divergent ileal IGF-I and IGFBP 3 gene expression after small bowel resection: a novel mechanism to amplify IGF action? Mol Cell Endocrinol 83:R17 R20