Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

THE EFFECT OF INTESTINAL RESECTION ON THIRY-VELLA FISTULAE OF JEJUNAL AND ILEAL ORIGIN IN THE RAT: EVIDENCE FOR A SYSTEMIC CONTROL MECHANISM OF CELL RENEWAL

THE EFFECT OF INTESTINAL RESECTION ON THIRY-VELLA FISTULAE OF JEJUNAL AND ILEAL ORIGIN IN THE... Department o Cell Biology and Genetics, Erasmus Universio,Rotterdam, The Netherlands, f f and Radiation Research Laboratory, University o Iowa, Iowa City, Iowa, U.S.A. (Received 10 November 1916; revision received 8 February 1911) ABSTRACT Local and systemic control mechanisms have been postulated to explain the maintenance of steady state cell renewal in intestinal epithelium. Permanent alterations of cell renewal resulting in a new steady state imply alterations in control. Intestinal resection appears to cause such alterations resulting in hyperplasia of the residual intestine. To test the hypothesis of a systemic control, the effect of 60% mid-intestinal resection on Thiry-Vella fistulae of both jejunal and ileal origin was observed in rats. Results showed that hypoplasia occurred in fistulae without resection of the remaining intestine in continuity. Cell counts of crypt and villus columns and tritiated thymidine uptake in isolated whole crypts were reduced. Scanning electron microscopy showed marked hypoplastic alterations in villi. However, when 60% of the intestine in continuity was resected, hyperplasia occurred not only in the residual intestine but in the fistulae of both jejunal and ileal origin. Cell counts of villus and crypt columns were increased along with increased tritiated thymidine uptake per crypt. Neutral a-glucosidase and aon-specific esterase activities did not change as a result of resection but the activities of both enzymes were greater in ileal fistulae than in ileum in situ. Observations on the different resection response of the jejunal versus ileal fistulae lead to a distinction between inherent and induced differences within the small intestine. This study suggests a systemic control of cell renewal. A possible mechanism involving intestinal vascular physiology is discussed. Present address: Division of Biological and Medical Research, Argonne National Laboratory, Argonne, Illinois 60439, U.S.A. Correspondence: Dr J. W. Osborne, Radiation Research Laboratory, 14 Medical Laboratories, University of Iowa, Iowa City, Iowa 52242, U.S.A. 543 W. R.Hanson et al. INTRODUCTION The interest in mechanisms which control cell proliferation and renewal in the intestine has increased since the advent of intestinal by-pass surgery to treat obesity, hypercholesterolemia, and other metabolic diseases. Understanding the efficacy of treatment for diseases which necessitate the surgical manipulation of the gastrointestinal tract is important. However, another fundamental aspect of the interest in control mechanisms of cell proliferation in the intestine or any organ system is the possibility that perturbations of control systems cause or allow abnormal growth. A complete understanding of the mechanisms which control cell proliferation in any tissue does not exist; however, the ease of manipulation and the established techniques of analysis, coupled with rapid cell renewal, make the intestine a good model for study. Several control mechanisms have been proposed to explain the response of the intestine to such diverse experimental conditions as irradiation, surgical by-pass or blind-loop formation, intestinal resection, hyperphagia, intestinal ischemia, diabetes, and other disease states or abnormal conditions. The proposed control mechanisms can be divided into two major categories: local and systemic. Local control From the observation that after irradiation, the decline in villus cellularity and the increase in crypt proliferation occur simultaneously, Galjaard, van der Meer-Fieggen & Giesen ( 1972) suggested a feedback control of the functional villus compartment to the proliferative crypt compartment. Evidence to support this concept has come from studies involving intestinal damage due to ischemia which selectively damaged the villus epithelium and elicited an increased proliferative response in crypts (Rijke et al., 1976a). The suggestion of a local feedback control is further strengthened by observations on human diseases such as coeliac disease (Booth, 1970) or the gluten-sensitive enteropathy of dermatitis herpetiformis (Wright ef al., 1973) where the size and cell number of villi are greatly reduced and cell production in the crypts is higher. Another type of local control of cell renewal may be related to the luminal environment which includes proximal-distal variations in bacterial flora, enzyme content, chyme content and consistency, etc. The influence of intestinal contents on the maintenance or change of intestinal renewal is great, as is shown by marked hypoplasia after the formation of intestinal by-pass (Gleeson, Cullen & Dowling, 1972; Clarke, 1974; Menge et a/., 1974; Rijke et al., 1976b), starvation (Stevens-Hooper & Blair, 1958) or the cessation of oral intake with parenteral maintenance (Johnson, Copeland & Dudrick, 1975; Feldman et al., 1976). Systemic control Elood-borne factors have been postulated to explain hyperplasia after resection (Loran & Crocker, 1963) and ‘hyperplasia’ in unresected parabiotic rats in common blood circulation with resected rats (Loran & Carbone, 1968). Tilson & Wright (1970) showed hypertrophy in by-passed ileum after jejunectomy, which suggested a humoral factor similar to what Loran proposed. Wilmore & Dudrick (1969) observed villus hypertrophy in partially enterectomized beagles maintained on intravenous hyperalimenation, which also suggested a humoral factor, although a similar experiment (Feldman et al., 1976) yielded results which suggested that the presence of chyme was essential for intestinal adaptation. Elias & Dowling (1974). however, showed hyperplasia in the by-passed intestine of lactating rats, which suggested a possible hormonal influence on cell renewal. Other possible systemic controls are based on Response o Thiry-Vellafistulae to resection f negative feedback involving tissue mass or functional demand. Tilson (1972) tested these hypotheses and concluded that mechanisms which involved either luminal nutrition and/or functional demand most likely controlled intestinal adaptation. Tissue-specific inhibitory chalones have also been implicated as a control mechanism (Brugal, 1976). Resection is only one of many procedures which perturb the steady state cell renewal of the intestine. Resection, however, causes apparently permanent hyperplastic alterations in intestinal morphology and cell kinetics and must therefore act via permanent alterations of control mechanisms. The adaptive response of the residual intestine to resection as measured by increased proliferative activity and increased crypt and villus cell counts has been well documented (Hanson & Osborne, 1971; Weser & Hernandez, 1971; Hanson et al., 1976a, b). The purpose of this study was to investigate the response of a Thiry-Vella fistula to resection of the jejunum and ileum in continuity. The fistula is not subject to the same luminal environment as the intestine in continuity, although it has a common blood and neural supply. A response in the fistula would indicate a systemic control, whereas if the response was confined to the residual functioning intestine, local mechanisms related to function would be implicated. Further, fistulae of both jejunal and ileal origin were studied to observe inherent versus induced differences in these two regions. MATERIALS AND METHODS Surgical procedures Male Wistar rats (Glaxo) weighing 200-250 g were fasted 12 hr, then anesthetized with pentobarbital (40 mg/kg body weight i.p.). The peritoneal cavity was entered through a midline incision and one of the following four procedures was performed, five animals in each group. (1) Jejunal Thiry-Vella fistula control. The intestine was divided 15 and 23 cm distal to the ligament of Treitz, leaving the blood supply intact. The two ends of this 8 cm segment were brought through puncture wounds in the abdominal wall on the right side of the animal and sutured to the abdominal musculature using 6-0 silk, which created a Thiry-Vella fistula (Vella, 1882). The residual intestine was joined by an end-to-end anastomosis with 6-0 silk. The midline wound was closed in two layers, muscle with 3-0 silk and skin with wound clips. (2) Jejunal Thiry-Vella fistula with 60% resection. A Thiry-Vella fistula was made with an 8 cm jejunal segment from 15-23 cm distal to the ligament of Treitz as described above. The ileum was then divided 15 cm proximal to the cecum, the mesenteric vessels supplying the center 60% of the jejunum and ileum were ligated and the tissue was resected. The residual intestine was anastomosed and the abdomen was closed as above. (3) Ileal Thiry-Vella fistula control. An 8 cm intestinal segment 15-23 cm proximal to the cecum was used to form a Thiry-Vella fistula. The residual intestine was anastomosed and the abdominal wound was closed. (4) Ileal Thiry-Vella fistula with 60% resection. A Thiry-Vella fistula was made with an 8 cm ileal segment as in procedure 3. The jejunum was divided 15 cm distal to the ligament of Treitz. The vessels supplying the 60% mid-portion of the jejunum and ileum were ligated and the tissue was resected. The intestine was anastomosed and the abdominal wound closed. Throughout this report, treatment groups (1) and (3) are referred to as controls. W.R.Hanson et al. Caging and nutrition Animals were kept two per cage in a room of even temperature with a 12 hr light-dark cycle. For the first 24 hr post-operative, the animals were allowed 5% sucrose water only. On the second post-operative day, 5% sucrose and a small amount of rat chow were allowed, followed by rat chow and tap water at will from the third day on. Tissue sampling and preparation Thirty days after operation, the animals were lightly anesthetized with ether and injected through the exposed femoral vein with tritiated thymidine (3H-TdR), 1 pCi/g body weight (Amersham, specific activity 15.2 Ci/mM.) The small leg wound was closed with wound clips and 1 hr later, between 10.00 and 11.00 hours the animals were again anesthetized with ether and the abdominal viscera exposed. Three areas were sampled from each animal: (1) jejunum, 3 cm distal to the ligament of Treitz; (2) mid-fistula; and (3) ileum, 3 cm proximal to the cecum. Tissue from each area was quickly removed and frozen in liquid nitrogen for later analysis of neutral a-glucosidase and non-specific esterase. A second sample from each area was fixed in cold Carnoy’s fixative for whole-crypt isolation (Wimber et al., 1960) and the subsequent determination of disintegrations per min (DPM) per crypt. A third sample was fixed in 10% formol saline for histology and autoradiography, and a fourth sample from each area was fixed in Carnoy’s fixative for scanning electron microscopy. Quantitative microchemical determination o neutral a-glucosidase and non-specijic esterase f activities Liquid nitrogen-frozen segments were stored at -7OOC. Cryostat sections (1 2 pm) were freeze-dried for 4 hr at -25°C with P < 0.001 mmHg. Crypt and villus sections were separated and weighed on a quartz fiber balance (Galjaard et al., 1970). The activity of neutral a-glucosidase was determined fluorimetrically using 4-methyl-umbelliferyl-a-~glycopyranoside as a substrate at pH 6.5 in a final volume of 520 pl (de Both & Plaisier, 1974). The activity of non-specific esterase was determined spectrophotometrically using naphthol acetate as a substrate in a final volume of 350 pl (Galjaard et al., 1970). Activities were assessed for ten crypt and ten villus sub-samples from each area of each animal. 3H-TdRincorporation per crypt Tissues fixed in Carnoy’s were stored in 70% ethanol, hydrated, hydrolysed in 1 N HCI, and stained by the Feulgen reaction. Whole crypts were dissected free and placed in a liquid scintillation vial, fifty crypts per vial, four vials per sample. Soluene (0.5 ml; Packard) was added to each vial and after solubilization, 10 ml of scintillation mixture ( 5 g PPO* plus 0.5 g POPOP? per litre of toluene) were added. Tritium activity was counted in a Packard liquid scintillation spectrometer. An efficiency versus external standard ratio relationship was used to correct for quenching in individual samples. Histologic and autoradiographic techniques Tissues fixed in 10% formol saline were embedded in paraffin, sectioned at 4 ,urn, and placed on gelatin-subbed slides. Autoradiographs were prepared by the dipping technique using Kodak NTB emulsion. After 7 days storage at 4”C, the emulsion was developed in * PPO: 2.5-diphenyloxazole. t POPOP: 2,2’-phenylene-bis (5-phenyloxazole). Response o Thiry-Vellajstulae to resection f Kodak D-11 developer, fixed, and the tissue stained with hematoxylin and eosin. Intestinal cross-sections were selected in which the base, lumen, and top of crypts were present and the villus epithelium was present as a single layer of cells. The number of cells per crypt or villus column was determined for 20 columns from each area sampled of each animal. The number of labeled cells in each crypt column was determined along with the position of the topmost labeled cell. Scanning electron microscopy Small pieces of tissue were fixed in Carnoy's solution at 4OC for 24 hr, dehydrated in a graded series of ethanol, transferred to amyl acetate and dried in a critical point drying apparatus. After covering the tissues with gold using a sputter coater, the specimens were examined with a Cambridge M K IIA scanning electron microscope. RESULTS Neutral a-glucosidase and non-spec@ esterase activity Measurements of activity of brush border-associated neutral a-glucosidase as reflected by the production of methyl-umbelliferone showed expected differences between crypts and villi (Table 1). Values for crypts which have almost no brush border averaged about 0.05 pmol methyl-umbelliferone/hr/mg dry weight compared to values of brush border-rich villi which averaged about 0.8 pmol. Values of ileal and jejunal crypts were similar, but ileal villi gave values of about 0-25 pmol, which were considerably lower than jejunal villi. The formation of fistulae per se from either jejunum or ileum had no effect on a-glucosidase activity in crypt samples. Villus enzyme activity in the fistula from jejunal tissue was the same as control jejunum; however, ileal villus enzyme activity increased in the fistulae to a value of about 0.53 pmol methyl-umbelliferone/hr/mg dry weight, which is intermediate between control values of jejunum and ileum. Intestinal resection had little effect on enzyme activities. The crypt enzyme activity in the ileum of resected animals was statistically greater but not greatly increased. Resection caused slightly increased values for all villus samples from all resected groups. TABLEI. Neutral a-glucosidase activity in jejunum, ileum. and surgically-created fistulae of jejunal and ileal origin in normal and resected rats Jejunum Crypt Control 60% Resection Villus 0.75 k 0.09 0 . 9 0 i 0.18 Fistulae of jejunal origin Crypt Villus Crypt Ileum Villus 0.053 f 0.003' 0.054 f 0.004 0.053 & 0.008 0.78 i 0.04 0.059 & 0.01 0.87 i 0.14 Fistulae of ileal origin 0.051 i 0.003 0.29 0.12 0,060 i 0.004 0 . 4 2 & 0.09 Jejunum Ileum Crypt VillL \ crY Pt Control 60Yo Resection 0.051 f 0.004 Villus Crypt Villus 0.056 i 0,002 k 0.06 0.87 k 0.20 0.050 k 0,004 0.53 k 0. I 7 0.051 i 0,006 0.59 2 0.19 0.052 & 0,002 0.23 & 0.06 0.050 f 0.004 0.31 f 0. I I * Data are expressed in pnol methylumbelliferone/hr/mg dry weight i 1 standard deviation. W. R.Hanson et al. TABLE Non-specific esterase activity in jejunum. ileum, and surgically-created fistulae of jejunal and ileal 2. origin in normal and resected rats Jejunum Crypt Control 60%)Resection 2 1 f 2+ 23 f 2 Jejunum Crypt Control fiO% Resection Villus 80 k 6 83 + 6 Villus 19 f 9 Fistulae of jejunal origin Crypt 16i2 20 i 3 Ileum Crypt 8 f 0.6 li2 Ileum Crypt 7+1 7+1 VlllU\ Villus 89 f 9 9 2 k 13 28-t 3 22 3 h Fistulae of ileal origin c rY Pt 8k3 8+3 58 Villus Villus 30 t 5 2fl f 5 23 If- 2 22 + 2 + 12 62 f 9 * Data are expressed in f l o l a-naphthol/hr/mg dry weight k I standard deviation The activity of non-specific esterase reflected by the production of a-naphthol was different between crypt and villi locally and between proximal and distal intestine (Table 2). The activity of jejunal crypt samples averaged about 22 pmol a-naphthollhrlmg dry weight, whereas ileal crypt samples were about 8. Jejunal villus samples averaged about 80 but ileal villus activity averaged 29. The surgical procedure of forming a fistula had no effect on nonspecific esterase activity in the remaining functional intestine. The crypts and villi of the fistula created from the jejunum retained normal levels of esterase activity and did not differ from the functioning jejunum. The crypts of the fistula formed from the ileum retained their characteristic low activity; however, the activity of the villi went up from an average of 29 pmol a-naphthol/hr/mg dry weight to about 60. Intestinal resection had no effect on the activity of non-specific esterase. The increased activity seen in the villi of ileal fistulae, which was the only change, was present to an equal extent in both control fistulae and fistulae from resected animals. Jejunal fistulae Ileal fistuloe Jejunum Jejunal Ileum Jejunum fistula Ileal flstulo Ileum FIG. I . Disintegrations per min (DPM) per crypt in jejunum, jejunal fistulae, ileal fistulae, and ileum of control and 60% resected animals. Each value is mean (+ SEM) of five animals. Open, control; stippled, after resection. Response o Thiry- VellaJistulae to resection f 3H-TdRincorporation per whole crypt Disintegrations per min (DPM) per crypt, which reflects the number of cells in cycle, decreased in the jejunal fistulae from an average of about 10 2 0.8 DPM/crypt to 6 5 0.8 (Fig. 1). Values for the ileal fistulae also decreased from about 13 & 0.5 DPM/crypt to about 7 1.8. A 60% intestinal resection caused increases in DPM/crypt not only in jejunum and ileum of the residual intestine in continuity, but in the jejunal fistulae as well. The slight increase for the ileal fistulae was not statistically significant. Crypt and villus cell parameters The number of cells per crypt column was the same in jejunum and ileum and in fistulae made from either jejunal or ileal tissue (Fig. 2). The range was from 30-33 for all control tissue, although the lower values were in the fistulae. Intestinal resection caused markedly increased values not only in the residual functional intestine but in the fistulae as well. The increase, however, was less in the fistulae (20%) than in the residual intestine in continuity (30%). Jejunal flstuloe Ileal flstuloe 50r ~~ Jejunum Jejunal fistula Ileum Jejunum Ileal flstula Ileum FIG.2. Number of cells per crypt column in jejunum, jejunal fistulae, ileal fistulae, and ileum of control and 60% resected animals. Each value is mean (k SD) of five animals. Open, control; stippled, after resection. The number of 3H-TdR-labeled cells per crypt column counted from autoradiographs of all intestinal sites from resected animals increased proportionally to the increase in total cells per column and thus showed no change in any treatment group when expressed as a percentage of the total cells per crypt column (Table 3). The cell position of the leading edge of 3H-TdRlabeled cells expressed as a percentage of the total number of cell positions also showed no change in any treatment group (Table 3). The average number of epithelial cells per villus column was seventy-five in the control jejunum compared to forty-five in the control ileum (Fig. 3), reflecting the villus height gradient (Altmann, 1971). The height of the villi in fistulae made from jejunal segments decreased 36% to a value of fifty-five cells per villus column. However, in fistulae made from ileal segments, villus cell counts remained the same. After intestinal resection, jejunal villus cell counts increased from about seventy-five to ninety. Ileal villi increased from about forty- W. R. Hanson et al. TABLE The response of proliferative indices in fistulae and intestine in continuity to resection 3. Fistulae of jejunal origin Fistulae of ileal origin Jejunum Labeling index Control 60% Resection Leading edge of label Control 60% Resection Ileum 26 t 2*$ 28 f 3 25 f 2 21 f 3 25 f 3 23 f 5 21 2 26 & 3 50 & 2 t 52 f 2 53 2 53 f 2 56 f 4 53 f 3 55 -t 3 56 & 2 * The average percentage of labeled cells in crypt columns f 1 standard deviation. t The average position of top-most labeled cells in crypt columns f 1 standard deviation expressed as a percentage of the total crypt height. $ Values for jejunum in continuity from all treatment groups were not different and were averaged for brevity. Similarly. ileal values were averaged. five to seventy-two, a value similar to control jejunum. Villus size was not only increased in the residual intestine but also in the fistulae created from both jejunal and ileal segments. In jejunal fistulae, values increased from fifty-five to about seventy-eight cells per column, and in ileal fistulae. values were increased from forty-eight to sixty-seven. Jejunal fistuloe I l e a l fistulae € Jejunum Jejunal Ileum Jejunum Ileal Ileum FIG.3. Number of cells per villus column in jejunum, jejunal fistulae, ileal fistulae, and ileum of control and 60% resected animals. Each value is mean (+SEM) of five animals. Open, control; stippled, after resection. Scanning electron microscopy Representative morphology of the luminal surface of the intestine from all sampled areas of all treatment groups is presented in Fig. 4a-h. The villi of control jejunum (Fig. 4a) were broad and varied in shape with undulating ridges predominating. Villi of the fistulae from jejunal origin (Fig. 4b) showed a marked change in pattern. The villi appear as columns spaced apart so that one can view deep within the mucosa. These villi appear longer than those of the control jejunum but cell counts (Fig. 3) prove them to be shorter. The villi of control ileum (Fig. 4d) are uniform and leaf-shaped, with shorter and narrower dimensions than jejunal villi. The villi of the fistulae derived from the ileum became short, mound-like, and oval in appearance (Fig. 4c). The jejunal villi in the residual intestine from resected animals (Fig. 4e) appeared thicker and somewhat more undulating than in control jejunum, but in Response of Thiry-Vellajistulae to resection FIG. 4. Scanning electron microscopy of the luminal surface of the small intestine from all treatment groups: (a) jejunum control: (b) jejunal fistula control: (c) ileal fistula control; (d) ileum control: (e) jejunum after 60% resection: (0 jejunal fistula after 60% resection; (g) ileal fistula after 60% resection; and (h) ileum after 60% resection. Magnification = 5 8 x . Photographs presented are representative of tissues within each experimental group. Each value is mean (+SD) of five animals. W. R. Hanson et al. general the two were similar. However, the villi in the jejunal fistulae of resected animals (Fig. 4f) were changed greatly compared to control jejunal fistulae, and instead of widely-spaced columns, they were ridged and undulated similar to control or residual jejunum. Ileal villi in resected animals became very much like jejunal villi (Fig. 4h), changing from a very ordered pattern to the undulating ridge form. Cell counts showed them to be nearly as tall a s control jejunal villi (Fig. 3). Intestinal resection caused the short mound-like villi of the control ileal fistulae to become taller and more column-like but similar in appearance. DISCUSSION The rnorphologic response of the residual intestine to resection has been well documented (Dowling & Booth, 1967; Hanson & Osborne, 1971, Weser & Hernandez, 197 1 ; McDermott & Roudnew, 1976; Hanson et al., 1976a, b). Hyperplasia is seen both proximal and distal to mid-intestinal resections, with the ileal response being particularly great. In this study, the response of the residual intestine to resection was as expected. Hyperplasia was seen in jejunum and especially ileum. In general, DPM/crypt was increased along with an increase in number of cells per crypt and villus columns. Proliferative indices (labeled cells and the leading edge) increased in proportion to crypt enlargement. Microchemical analysis of neutral a-glucosidase and non-specific esterase did not show any remarkable change in the residual intestine. Scanning electron microscopy revealed that ileal villi became much like jejunal villi after resection, not only in height but in three-dimensional structure. One consequence of resection was to bring the ileum into closer proximity to all the environmental conditions normally affecting the jejunum. The list of luminal differences is indeed great, including differences in chyme content, consistency, enzymes, bile acids, pH, etc. Although there is good experimental evidence that pancreaticobiliary secretions (Altmann, 1971) influence villus morphology, the role of changed environment in the whole of the residual intestine is difficult to assess. For example, if food intake is the same in resected animals as in normal (Dowling, 1968), then the chyme content per length of gastrointestinal tract could be considerably increased and simulate the effects of hyperphagia which has been associated with hyperplasia (Campbell & Fell, 1964). The experimental design of the present study eliminated some of the variables and allowed some judgments to be made of the relative effects of luminal environment versus systemic factors. In this study, the hypoplasia observed in fistulae of both jejunal and ileal origin attests to the significant influence of chyme in the functional intestine. Similar changes were seen in bypassed loops (Clarke, 1974; Gleeson el a/., 1972) and fistulae (Keren e t a / . , 1975; Rijke et al., 1976b). The morphology of villi from jejunal fistulae remained distinct from fistulae of ileal origin (Fig. 4b, 0, indicating an inherent difference in the two areas, even when removed from luminal influences. Activity of a-glucosidase and esterase remained the same in jejunal fistulae compared to jejunum in situ, but in the ileum, values increased over control values for ileum in situ. This suggests that an inhibition of the production of these enzymes is normally present in the villi of the ileum. After a fistula was created, the inhibition was reduced and enzyme production increased to levels higher than normal. The major significant finding in this study was that fistulae of both jejunal and ileal origin responded to resection of the intestine in continuity. These results imply that there is either a stimulatory substance released into the circulatory system or that circulatory o r neural changes occur within the fistulae which stimulate hyperplasia. The hyperplastic changes seen Response of Thiry- VellaBstulae to resection in the fistulae were in general similar to those already described for the residual intestine. However, there were some basic differences in the resection response of jejunal compared with ileal fistulae. Crypts responded similarly in both fistulae, but jejunal villi responded markedly, changing from finger-like columns to convoluted ridges resembling normal jejunal villus structure. In the intestine in continuity, it is ileal villi which change to a greater degree. Villi in ileal fistulae changed from short mounds to taller mounds but did not change markedly in pattern. Intestinal chyme then must exert a relatively greater post-resection influence on ileal villi than on jejunal villi, which agrees with Dowling & Booth (1967). The results of the present study show that hyperplasia is induced by a systemic influence upon cell renewal, although the degree of hyperplasia and final structure of the hyperplastic intestine is influenced by the luminal environment. The fistulae and intestine in continuity have a common blood and neural supply. Neural stimulation has been shown to increase crypt cell proliferation in the jejunum of rats (Tutton, 1975); however, an increased neural activity after resection has not been investigated. Evidence suggesting a humoral factor released after resection (Loran & Crocker, 1963; Loran & Carbone, 1968) has been criticized by Clarke (1974) and by Kirchner (1975). The points they have raised are valid points of disagreement and cast doubt on the interpretation of the data of Loran & Carbone (1968). Kirchner designed a more detailed experiment to confirm the findings of Loran & Carbone and failed to d o so, although he used much younger rats. An alternative hypothesis to the release of a stimulatory factor by resection is the role that a physiologically altered intestinal circulatory system may play. Touloukian & Spencer (197 1) determined the mucosal perfusion of rats with 50% mid-intestinal resections using the 86Rb distribution technique (Sapirstein, 1958). They found increased perfusion of the ileum prior to hyperplasia and no change in perfusion or cell kinetics in the jejunum.They suggested that a relationship may exist between blood flow and intestinal compensation. Blood flow is directly related to pressure and inversely related to resistance, F P/R. Regulation of these parameters in the intestine is extremely complicated and i5 subject not only to systemic alterations such as cardiac rate, output, etc., but to autoregulation and autoregulatory escape as well (Folkow, 1967). Alterations in these parameters have not been studied i the residual n intestine after resection. Further, there is no proven relationship between blood perfusion rate, pressure or resistance and intestinal cell proliferation. However, it is intriguing to speculate that these parameters may influence cell renewal and that many of the experimental conditions which affect cell renewal act directly or indirectly via the circulatory system. Lundgren & Kampp (1966) suggested and have shown evidence (Kampp & Lundgren, 1966; Haglund, Jodal & Lundgren, 1973) of a countercurrent system in the intestinal mucosa which may result in an oxygen or nutrient concentration gradient. The existence of such a gradient may be the major impetus for the cessation of cell proliferation in crypts and may cause the sloughing of epithelial cells from villus tips. Experimental manipulation of the intestine may alter the gradient distribution within the mucosa and thus cause hyperplasia or hypoplasia, depending upon the condition created. The results of the study presented in this report may be explained by a physiological mechanism. Hypoplasia was seen in fistulae with no resection and since the amount of tissue involved was small, no changes were seen in the functional portion. However, when a large resection was done, blood flow alterations would not only occur in the residual functional portion but also in the fistulae, causing hyperplasia. The differences seen in proximal versus W.R . Hanson et al. distal fistulae are tissue-specific, whereas differences in proximal and distal intestine in continuity could result in part from the effect of the luminal environment on blood perfusion rate. The proposed physiological mechanism may provide a reasonable explanation for the apparent systemic control of intestinal cell renewal. This mechanism could allow gross adjustments to various perturbations, whereas the negative feedback of functional villus cells to crypt cell production could be a fine-tuning and local control. Direct evidence of a physiological control of intestinal cell renewal is lacking but the prospect that a relationship exists is promising. ACKNOWLEDGMENTS This study was supported by a grant from The Netherlands Foundation in Medical Research, and by National Institute of Arthritis and Metabolic Disease Research Grant AM- 15758. The authors thank Dr H. Galjaard, Dr R. J. M. Fry, and Dr H. P. Schedl for their advice and criticism. Mr N. H. C. Bruns is gratefully acknowledged for his technical assistance with the scanning electron microscopy, and Mr J. Bos for his excellent care and maintenance of the animals. We wish to thank Ms Cheryl Bedwell for her excellent secretarial help. We are also grateful to Ms Jo Ann Peiffer for secretarial assistance and for help with the figures. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Proliferation Wiley

THE EFFECT OF INTESTINAL RESECTION ON THIRY-VELLA FISTULAE OF JEJUNAL AND ILEAL ORIGIN IN THE RAT: EVIDENCE FOR A SYSTEMIC CONTROL MECHANISM OF CELL RENEWAL

Download PDF
13 pages

Loading next page...
 
/lp/wiley/the-effect-of-intestinal-resection-on-thiry-vella-fistulae-of-jejunal-zWut3XhR3G

References (41)

Publisher
Wiley
Copyright
1977 Blackwell Publishing Ltd
ISSN
0960-7722
eISSN
1365-2184
DOI
10.1111/j.1365-2184.1977.tb00311.x
Publisher site
See Article on Publisher Site

Abstract

Department o Cell Biology and Genetics, Erasmus Universio,Rotterdam, The Netherlands, f f and Radiation Research Laboratory, University o Iowa, Iowa City, Iowa, U.S.A. (Received 10 November 1916; revision received 8 February 1911) ABSTRACT Local and systemic control mechanisms have been postulated to explain the maintenance of steady state cell renewal in intestinal epithelium. Permanent alterations of cell renewal resulting in a new steady state imply alterations in control. Intestinal resection appears to cause such alterations resulting in hyperplasia of the residual intestine. To test the hypothesis of a systemic control, the effect of 60% mid-intestinal resection on Thiry-Vella fistulae of both jejunal and ileal origin was observed in rats. Results showed that hypoplasia occurred in fistulae without resection of the remaining intestine in continuity. Cell counts of crypt and villus columns and tritiated thymidine uptake in isolated whole crypts were reduced. Scanning electron microscopy showed marked hypoplastic alterations in villi. However, when 60% of the intestine in continuity was resected, hyperplasia occurred not only in the residual intestine but in the fistulae of both jejunal and ileal origin. Cell counts of villus and crypt columns were increased along with increased tritiated thymidine uptake per crypt. Neutral a-glucosidase and aon-specific esterase activities did not change as a result of resection but the activities of both enzymes were greater in ileal fistulae than in ileum in situ. Observations on the different resection response of the jejunal versus ileal fistulae lead to a distinction between inherent and induced differences within the small intestine. This study suggests a systemic control of cell renewal. A possible mechanism involving intestinal vascular physiology is discussed. Present address: Division of Biological and Medical Research, Argonne National Laboratory, Argonne, Illinois 60439, U.S.A. Correspondence: Dr J. W. Osborne, Radiation Research Laboratory, 14 Medical Laboratories, University of Iowa, Iowa City, Iowa 52242, U.S.A. 543 W. R.Hanson et al. INTRODUCTION The interest in mechanisms which control cell proliferation and renewal in the intestine has increased since the advent of intestinal by-pass surgery to treat obesity, hypercholesterolemia, and other metabolic diseases. Understanding the efficacy of treatment for diseases which necessitate the surgical manipulation of the gastrointestinal tract is important. However, another fundamental aspect of the interest in control mechanisms of cell proliferation in the intestine or any organ system is the possibility that perturbations of control systems cause or allow abnormal growth. A complete understanding of the mechanisms which control cell proliferation in any tissue does not exist; however, the ease of manipulation and the established techniques of analysis, coupled with rapid cell renewal, make the intestine a good model for study. Several control mechanisms have been proposed to explain the response of the intestine to such diverse experimental conditions as irradiation, surgical by-pass or blind-loop formation, intestinal resection, hyperphagia, intestinal ischemia, diabetes, and other disease states or abnormal conditions. The proposed control mechanisms can be divided into two major categories: local and systemic. Local control From the observation that after irradiation, the decline in villus cellularity and the increase in crypt proliferation occur simultaneously, Galjaard, van der Meer-Fieggen & Giesen ( 1972) suggested a feedback control of the functional villus compartment to the proliferative crypt compartment. Evidence to support this concept has come from studies involving intestinal damage due to ischemia which selectively damaged the villus epithelium and elicited an increased proliferative response in crypts (Rijke et al., 1976a). The suggestion of a local feedback control is further strengthened by observations on human diseases such as coeliac disease (Booth, 1970) or the gluten-sensitive enteropathy of dermatitis herpetiformis (Wright ef al., 1973) where the size and cell number of villi are greatly reduced and cell production in the crypts is higher. Another type of local control of cell renewal may be related to the luminal environment which includes proximal-distal variations in bacterial flora, enzyme content, chyme content and consistency, etc. The influence of intestinal contents on the maintenance or change of intestinal renewal is great, as is shown by marked hypoplasia after the formation of intestinal by-pass (Gleeson, Cullen & Dowling, 1972; Clarke, 1974; Menge et a/., 1974; Rijke et al., 1976b), starvation (Stevens-Hooper & Blair, 1958) or the cessation of oral intake with parenteral maintenance (Johnson, Copeland & Dudrick, 1975; Feldman et al., 1976). Systemic control Elood-borne factors have been postulated to explain hyperplasia after resection (Loran & Crocker, 1963) and ‘hyperplasia’ in unresected parabiotic rats in common blood circulation with resected rats (Loran & Carbone, 1968). Tilson & Wright (1970) showed hypertrophy in by-passed ileum after jejunectomy, which suggested a humoral factor similar to what Loran proposed. Wilmore & Dudrick (1969) observed villus hypertrophy in partially enterectomized beagles maintained on intravenous hyperalimenation, which also suggested a humoral factor, although a similar experiment (Feldman et al., 1976) yielded results which suggested that the presence of chyme was essential for intestinal adaptation. Elias & Dowling (1974). however, showed hyperplasia in the by-passed intestine of lactating rats, which suggested a possible hormonal influence on cell renewal. Other possible systemic controls are based on Response o Thiry-Vellafistulae to resection f negative feedback involving tissue mass or functional demand. Tilson (1972) tested these hypotheses and concluded that mechanisms which involved either luminal nutrition and/or functional demand most likely controlled intestinal adaptation. Tissue-specific inhibitory chalones have also been implicated as a control mechanism (Brugal, 1976). Resection is only one of many procedures which perturb the steady state cell renewal of the intestine. Resection, however, causes apparently permanent hyperplastic alterations in intestinal morphology and cell kinetics and must therefore act via permanent alterations of control mechanisms. The adaptive response of the residual intestine to resection as measured by increased proliferative activity and increased crypt and villus cell counts has been well documented (Hanson & Osborne, 1971; Weser & Hernandez, 1971; Hanson et al., 1976a, b). The purpose of this study was to investigate the response of a Thiry-Vella fistula to resection of the jejunum and ileum in continuity. The fistula is not subject to the same luminal environment as the intestine in continuity, although it has a common blood and neural supply. A response in the fistula would indicate a systemic control, whereas if the response was confined to the residual functioning intestine, local mechanisms related to function would be implicated. Further, fistulae of both jejunal and ileal origin were studied to observe inherent versus induced differences in these two regions. MATERIALS AND METHODS Surgical procedures Male Wistar rats (Glaxo) weighing 200-250 g were fasted 12 hr, then anesthetized with pentobarbital (40 mg/kg body weight i.p.). The peritoneal cavity was entered through a midline incision and one of the following four procedures was performed, five animals in each group. (1) Jejunal Thiry-Vella fistula control. The intestine was divided 15 and 23 cm distal to the ligament of Treitz, leaving the blood supply intact. The two ends of this 8 cm segment were brought through puncture wounds in the abdominal wall on the right side of the animal and sutured to the abdominal musculature using 6-0 silk, which created a Thiry-Vella fistula (Vella, 1882). The residual intestine was joined by an end-to-end anastomosis with 6-0 silk. The midline wound was closed in two layers, muscle with 3-0 silk and skin with wound clips. (2) Jejunal Thiry-Vella fistula with 60% resection. A Thiry-Vella fistula was made with an 8 cm jejunal segment from 15-23 cm distal to the ligament of Treitz as described above. The ileum was then divided 15 cm proximal to the cecum, the mesenteric vessels supplying the center 60% of the jejunum and ileum were ligated and the tissue was resected. The residual intestine was anastomosed and the abdomen was closed as above. (3) Ileal Thiry-Vella fistula control. An 8 cm intestinal segment 15-23 cm proximal to the cecum was used to form a Thiry-Vella fistula. The residual intestine was anastomosed and the abdominal wound was closed. (4) Ileal Thiry-Vella fistula with 60% resection. A Thiry-Vella fistula was made with an 8 cm ileal segment as in procedure 3. The jejunum was divided 15 cm distal to the ligament of Treitz. The vessels supplying the 60% mid-portion of the jejunum and ileum were ligated and the tissue was resected. The intestine was anastomosed and the abdominal wound closed. Throughout this report, treatment groups (1) and (3) are referred to as controls. W.R.Hanson et al. Caging and nutrition Animals were kept two per cage in a room of even temperature with a 12 hr light-dark cycle. For the first 24 hr post-operative, the animals were allowed 5% sucrose water only. On the second post-operative day, 5% sucrose and a small amount of rat chow were allowed, followed by rat chow and tap water at will from the third day on. Tissue sampling and preparation Thirty days after operation, the animals were lightly anesthetized with ether and injected through the exposed femoral vein with tritiated thymidine (3H-TdR), 1 pCi/g body weight (Amersham, specific activity 15.2 Ci/mM.) The small leg wound was closed with wound clips and 1 hr later, between 10.00 and 11.00 hours the animals were again anesthetized with ether and the abdominal viscera exposed. Three areas were sampled from each animal: (1) jejunum, 3 cm distal to the ligament of Treitz; (2) mid-fistula; and (3) ileum, 3 cm proximal to the cecum. Tissue from each area was quickly removed and frozen in liquid nitrogen for later analysis of neutral a-glucosidase and non-specific esterase. A second sample from each area was fixed in cold Carnoy’s fixative for whole-crypt isolation (Wimber et al., 1960) and the subsequent determination of disintegrations per min (DPM) per crypt. A third sample was fixed in 10% formol saline for histology and autoradiography, and a fourth sample from each area was fixed in Carnoy’s fixative for scanning electron microscopy. Quantitative microchemical determination o neutral a-glucosidase and non-specijic esterase f activities Liquid nitrogen-frozen segments were stored at -7OOC. Cryostat sections (1 2 pm) were freeze-dried for 4 hr at -25°C with P < 0.001 mmHg. Crypt and villus sections were separated and weighed on a quartz fiber balance (Galjaard et al., 1970). The activity of neutral a-glucosidase was determined fluorimetrically using 4-methyl-umbelliferyl-a-~glycopyranoside as a substrate at pH 6.5 in a final volume of 520 pl (de Both & Plaisier, 1974). The activity of non-specific esterase was determined spectrophotometrically using naphthol acetate as a substrate in a final volume of 350 pl (Galjaard et al., 1970). Activities were assessed for ten crypt and ten villus sub-samples from each area of each animal. 3H-TdRincorporation per crypt Tissues fixed in Carnoy’s were stored in 70% ethanol, hydrated, hydrolysed in 1 N HCI, and stained by the Feulgen reaction. Whole crypts were dissected free and placed in a liquid scintillation vial, fifty crypts per vial, four vials per sample. Soluene (0.5 ml; Packard) was added to each vial and after solubilization, 10 ml of scintillation mixture ( 5 g PPO* plus 0.5 g POPOP? per litre of toluene) were added. Tritium activity was counted in a Packard liquid scintillation spectrometer. An efficiency versus external standard ratio relationship was used to correct for quenching in individual samples. Histologic and autoradiographic techniques Tissues fixed in 10% formol saline were embedded in paraffin, sectioned at 4 ,urn, and placed on gelatin-subbed slides. Autoradiographs were prepared by the dipping technique using Kodak NTB emulsion. After 7 days storage at 4”C, the emulsion was developed in * PPO: 2.5-diphenyloxazole. t POPOP: 2,2’-phenylene-bis (5-phenyloxazole). Response o Thiry-Vellajstulae to resection f Kodak D-11 developer, fixed, and the tissue stained with hematoxylin and eosin. Intestinal cross-sections were selected in which the base, lumen, and top of crypts were present and the villus epithelium was present as a single layer of cells. The number of cells per crypt or villus column was determined for 20 columns from each area sampled of each animal. The number of labeled cells in each crypt column was determined along with the position of the topmost labeled cell. Scanning electron microscopy Small pieces of tissue were fixed in Carnoy's solution at 4OC for 24 hr, dehydrated in a graded series of ethanol, transferred to amyl acetate and dried in a critical point drying apparatus. After covering the tissues with gold using a sputter coater, the specimens were examined with a Cambridge M K IIA scanning electron microscope. RESULTS Neutral a-glucosidase and non-spec@ esterase activity Measurements of activity of brush border-associated neutral a-glucosidase as reflected by the production of methyl-umbelliferone showed expected differences between crypts and villi (Table 1). Values for crypts which have almost no brush border averaged about 0.05 pmol methyl-umbelliferone/hr/mg dry weight compared to values of brush border-rich villi which averaged about 0.8 pmol. Values of ileal and jejunal crypts were similar, but ileal villi gave values of about 0-25 pmol, which were considerably lower than jejunal villi. The formation of fistulae per se from either jejunum or ileum had no effect on a-glucosidase activity in crypt samples. Villus enzyme activity in the fistula from jejunal tissue was the same as control jejunum; however, ileal villus enzyme activity increased in the fistulae to a value of about 0.53 pmol methyl-umbelliferone/hr/mg dry weight, which is intermediate between control values of jejunum and ileum. Intestinal resection had little effect on enzyme activities. The crypt enzyme activity in the ileum of resected animals was statistically greater but not greatly increased. Resection caused slightly increased values for all villus samples from all resected groups. TABLEI. Neutral a-glucosidase activity in jejunum, ileum. and surgically-created fistulae of jejunal and ileal origin in normal and resected rats Jejunum Crypt Control 60% Resection Villus 0.75 k 0.09 0 . 9 0 i 0.18 Fistulae of jejunal origin Crypt Villus Crypt Ileum Villus 0.053 f 0.003' 0.054 f 0.004 0.053 & 0.008 0.78 i 0.04 0.059 & 0.01 0.87 i 0.14 Fistulae of ileal origin 0.051 i 0.003 0.29 0.12 0,060 i 0.004 0 . 4 2 & 0.09 Jejunum Ileum Crypt VillL \ crY Pt Control 60Yo Resection 0.051 f 0.004 Villus Crypt Villus 0.056 i 0,002 k 0.06 0.87 k 0.20 0.050 k 0,004 0.53 k 0. I 7 0.051 i 0,006 0.59 2 0.19 0.052 & 0,002 0.23 & 0.06 0.050 f 0.004 0.31 f 0. I I * Data are expressed in pnol methylumbelliferone/hr/mg dry weight i 1 standard deviation. W. R.Hanson et al. TABLE Non-specific esterase activity in jejunum. ileum, and surgically-created fistulae of jejunal and ileal 2. origin in normal and resected rats Jejunum Crypt Control 60%)Resection 2 1 f 2+ 23 f 2 Jejunum Crypt Control fiO% Resection Villus 80 k 6 83 + 6 Villus 19 f 9 Fistulae of jejunal origin Crypt 16i2 20 i 3 Ileum Crypt 8 f 0.6 li2 Ileum Crypt 7+1 7+1 VlllU\ Villus 89 f 9 9 2 k 13 28-t 3 22 3 h Fistulae of ileal origin c rY Pt 8k3 8+3 58 Villus Villus 30 t 5 2fl f 5 23 If- 2 22 + 2 + 12 62 f 9 * Data are expressed in f l o l a-naphthol/hr/mg dry weight k I standard deviation The activity of non-specific esterase reflected by the production of a-naphthol was different between crypt and villi locally and between proximal and distal intestine (Table 2). The activity of jejunal crypt samples averaged about 22 pmol a-naphthollhrlmg dry weight, whereas ileal crypt samples were about 8. Jejunal villus samples averaged about 80 but ileal villus activity averaged 29. The surgical procedure of forming a fistula had no effect on nonspecific esterase activity in the remaining functional intestine. The crypts and villi of the fistula created from the jejunum retained normal levels of esterase activity and did not differ from the functioning jejunum. The crypts of the fistula formed from the ileum retained their characteristic low activity; however, the activity of the villi went up from an average of 29 pmol a-naphthol/hr/mg dry weight to about 60. Intestinal resection had no effect on the activity of non-specific esterase. The increased activity seen in the villi of ileal fistulae, which was the only change, was present to an equal extent in both control fistulae and fistulae from resected animals. Jejunal fistulae Ileal fistuloe Jejunum Jejunal Ileum Jejunum fistula Ileal flstulo Ileum FIG. I . Disintegrations per min (DPM) per crypt in jejunum, jejunal fistulae, ileal fistulae, and ileum of control and 60% resected animals. Each value is mean (+ SEM) of five animals. Open, control; stippled, after resection. Response o Thiry- VellaJistulae to resection f 3H-TdRincorporation per whole crypt Disintegrations per min (DPM) per crypt, which reflects the number of cells in cycle, decreased in the jejunal fistulae from an average of about 10 2 0.8 DPM/crypt to 6 5 0.8 (Fig. 1). Values for the ileal fistulae also decreased from about 13 & 0.5 DPM/crypt to about 7 1.8. A 60% intestinal resection caused increases in DPM/crypt not only in jejunum and ileum of the residual intestine in continuity, but in the jejunal fistulae as well. The slight increase for the ileal fistulae was not statistically significant. Crypt and villus cell parameters The number of cells per crypt column was the same in jejunum and ileum and in fistulae made from either jejunal or ileal tissue (Fig. 2). The range was from 30-33 for all control tissue, although the lower values were in the fistulae. Intestinal resection caused markedly increased values not only in the residual functional intestine but in the fistulae as well. The increase, however, was less in the fistulae (20%) than in the residual intestine in continuity (30%). Jejunal flstuloe Ileal flstuloe 50r ~~ Jejunum Jejunal fistula Ileum Jejunum Ileal flstula Ileum FIG.2. Number of cells per crypt column in jejunum, jejunal fistulae, ileal fistulae, and ileum of control and 60% resected animals. Each value is mean (k SD) of five animals. Open, control; stippled, after resection. The number of 3H-TdR-labeled cells per crypt column counted from autoradiographs of all intestinal sites from resected animals increased proportionally to the increase in total cells per column and thus showed no change in any treatment group when expressed as a percentage of the total cells per crypt column (Table 3). The cell position of the leading edge of 3H-TdRlabeled cells expressed as a percentage of the total number of cell positions also showed no change in any treatment group (Table 3). The average number of epithelial cells per villus column was seventy-five in the control jejunum compared to forty-five in the control ileum (Fig. 3), reflecting the villus height gradient (Altmann, 1971). The height of the villi in fistulae made from jejunal segments decreased 36% to a value of fifty-five cells per villus column. However, in fistulae made from ileal segments, villus cell counts remained the same. After intestinal resection, jejunal villus cell counts increased from about seventy-five to ninety. Ileal villi increased from about forty- W. R. Hanson et al. TABLE The response of proliferative indices in fistulae and intestine in continuity to resection 3. Fistulae of jejunal origin Fistulae of ileal origin Jejunum Labeling index Control 60% Resection Leading edge of label Control 60% Resection Ileum 26 t 2*$ 28 f 3 25 f 2 21 f 3 25 f 3 23 f 5 21 2 26 & 3 50 & 2 t 52 f 2 53 2 53 f 2 56 f 4 53 f 3 55 -t 3 56 & 2 * The average percentage of labeled cells in crypt columns f 1 standard deviation. t The average position of top-most labeled cells in crypt columns f 1 standard deviation expressed as a percentage of the total crypt height. $ Values for jejunum in continuity from all treatment groups were not different and were averaged for brevity. Similarly. ileal values were averaged. five to seventy-two, a value similar to control jejunum. Villus size was not only increased in the residual intestine but also in the fistulae created from both jejunal and ileal segments. In jejunal fistulae, values increased from fifty-five to about seventy-eight cells per column, and in ileal fistulae. values were increased from forty-eight to sixty-seven. Jejunal fistuloe I l e a l fistulae € Jejunum Jejunal Ileum Jejunum Ileal Ileum FIG.3. Number of cells per villus column in jejunum, jejunal fistulae, ileal fistulae, and ileum of control and 60% resected animals. Each value is mean (+SEM) of five animals. Open, control; stippled, after resection. Scanning electron microscopy Representative morphology of the luminal surface of the intestine from all sampled areas of all treatment groups is presented in Fig. 4a-h. The villi of control jejunum (Fig. 4a) were broad and varied in shape with undulating ridges predominating. Villi of the fistulae from jejunal origin (Fig. 4b) showed a marked change in pattern. The villi appear as columns spaced apart so that one can view deep within the mucosa. These villi appear longer than those of the control jejunum but cell counts (Fig. 3) prove them to be shorter. The villi of control ileum (Fig. 4d) are uniform and leaf-shaped, with shorter and narrower dimensions than jejunal villi. The villi of the fistulae derived from the ileum became short, mound-like, and oval in appearance (Fig. 4c). The jejunal villi in the residual intestine from resected animals (Fig. 4e) appeared thicker and somewhat more undulating than in control jejunum, but in Response of Thiry-Vellajistulae to resection FIG. 4. Scanning electron microscopy of the luminal surface of the small intestine from all treatment groups: (a) jejunum control: (b) jejunal fistula control: (c) ileal fistula control; (d) ileum control: (e) jejunum after 60% resection: (0 jejunal fistula after 60% resection; (g) ileal fistula after 60% resection; and (h) ileum after 60% resection. Magnification = 5 8 x . Photographs presented are representative of tissues within each experimental group. Each value is mean (+SD) of five animals. W. R. Hanson et al. general the two were similar. However, the villi in the jejunal fistulae of resected animals (Fig. 4f) were changed greatly compared to control jejunal fistulae, and instead of widely-spaced columns, they were ridged and undulated similar to control or residual jejunum. Ileal villi in resected animals became very much like jejunal villi (Fig. 4h), changing from a very ordered pattern to the undulating ridge form. Cell counts showed them to be nearly as tall a s control jejunal villi (Fig. 3). Intestinal resection caused the short mound-like villi of the control ileal fistulae to become taller and more column-like but similar in appearance. DISCUSSION The rnorphologic response of the residual intestine to resection has been well documented (Dowling & Booth, 1967; Hanson & Osborne, 1971, Weser & Hernandez, 197 1 ; McDermott & Roudnew, 1976; Hanson et al., 1976a, b). Hyperplasia is seen both proximal and distal to mid-intestinal resections, with the ileal response being particularly great. In this study, the response of the residual intestine to resection was as expected. Hyperplasia was seen in jejunum and especially ileum. In general, DPM/crypt was increased along with an increase in number of cells per crypt and villus columns. Proliferative indices (labeled cells and the leading edge) increased in proportion to crypt enlargement. Microchemical analysis of neutral a-glucosidase and non-specific esterase did not show any remarkable change in the residual intestine. Scanning electron microscopy revealed that ileal villi became much like jejunal villi after resection, not only in height but in three-dimensional structure. One consequence of resection was to bring the ileum into closer proximity to all the environmental conditions normally affecting the jejunum. The list of luminal differences is indeed great, including differences in chyme content, consistency, enzymes, bile acids, pH, etc. Although there is good experimental evidence that pancreaticobiliary secretions (Altmann, 1971) influence villus morphology, the role of changed environment in the whole of the residual intestine is difficult to assess. For example, if food intake is the same in resected animals as in normal (Dowling, 1968), then the chyme content per length of gastrointestinal tract could be considerably increased and simulate the effects of hyperphagia which has been associated with hyperplasia (Campbell & Fell, 1964). The experimental design of the present study eliminated some of the variables and allowed some judgments to be made of the relative effects of luminal environment versus systemic factors. In this study, the hypoplasia observed in fistulae of both jejunal and ileal origin attests to the significant influence of chyme in the functional intestine. Similar changes were seen in bypassed loops (Clarke, 1974; Gleeson el a/., 1972) and fistulae (Keren e t a / . , 1975; Rijke et al., 1976b). The morphology of villi from jejunal fistulae remained distinct from fistulae of ileal origin (Fig. 4b, 0, indicating an inherent difference in the two areas, even when removed from luminal influences. Activity of a-glucosidase and esterase remained the same in jejunal fistulae compared to jejunum in situ, but in the ileum, values increased over control values for ileum in situ. This suggests that an inhibition of the production of these enzymes is normally present in the villi of the ileum. After a fistula was created, the inhibition was reduced and enzyme production increased to levels higher than normal. The major significant finding in this study was that fistulae of both jejunal and ileal origin responded to resection of the intestine in continuity. These results imply that there is either a stimulatory substance released into the circulatory system or that circulatory o r neural changes occur within the fistulae which stimulate hyperplasia. The hyperplastic changes seen Response of Thiry- VellaBstulae to resection in the fistulae were in general similar to those already described for the residual intestine. However, there were some basic differences in the resection response of jejunal compared with ileal fistulae. Crypts responded similarly in both fistulae, but jejunal villi responded markedly, changing from finger-like columns to convoluted ridges resembling normal jejunal villus structure. In the intestine in continuity, it is ileal villi which change to a greater degree. Villi in ileal fistulae changed from short mounds to taller mounds but did not change markedly in pattern. Intestinal chyme then must exert a relatively greater post-resection influence on ileal villi than on jejunal villi, which agrees with Dowling & Booth (1967). The results of the present study show that hyperplasia is induced by a systemic influence upon cell renewal, although the degree of hyperplasia and final structure of the hyperplastic intestine is influenced by the luminal environment. The fistulae and intestine in continuity have a common blood and neural supply. Neural stimulation has been shown to increase crypt cell proliferation in the jejunum of rats (Tutton, 1975); however, an increased neural activity after resection has not been investigated. Evidence suggesting a humoral factor released after resection (Loran & Crocker, 1963; Loran & Carbone, 1968) has been criticized by Clarke (1974) and by Kirchner (1975). The points they have raised are valid points of disagreement and cast doubt on the interpretation of the data of Loran & Carbone (1968). Kirchner designed a more detailed experiment to confirm the findings of Loran & Carbone and failed to d o so, although he used much younger rats. An alternative hypothesis to the release of a stimulatory factor by resection is the role that a physiologically altered intestinal circulatory system may play. Touloukian & Spencer (197 1) determined the mucosal perfusion of rats with 50% mid-intestinal resections using the 86Rb distribution technique (Sapirstein, 1958). They found increased perfusion of the ileum prior to hyperplasia and no change in perfusion or cell kinetics in the jejunum.They suggested that a relationship may exist between blood flow and intestinal compensation. Blood flow is directly related to pressure and inversely related to resistance, F P/R. Regulation of these parameters in the intestine is extremely complicated and i5 subject not only to systemic alterations such as cardiac rate, output, etc., but to autoregulation and autoregulatory escape as well (Folkow, 1967). Alterations in these parameters have not been studied i the residual n intestine after resection. Further, there is no proven relationship between blood perfusion rate, pressure or resistance and intestinal cell proliferation. However, it is intriguing to speculate that these parameters may influence cell renewal and that many of the experimental conditions which affect cell renewal act directly or indirectly via the circulatory system. Lundgren & Kampp (1966) suggested and have shown evidence (Kampp & Lundgren, 1966; Haglund, Jodal & Lundgren, 1973) of a countercurrent system in the intestinal mucosa which may result in an oxygen or nutrient concentration gradient. The existence of such a gradient may be the major impetus for the cessation of cell proliferation in crypts and may cause the sloughing of epithelial cells from villus tips. Experimental manipulation of the intestine may alter the gradient distribution within the mucosa and thus cause hyperplasia or hypoplasia, depending upon the condition created. The results of the study presented in this report may be explained by a physiological mechanism. Hypoplasia was seen in fistulae with no resection and since the amount of tissue involved was small, no changes were seen in the functional portion. However, when a large resection was done, blood flow alterations would not only occur in the residual functional portion but also in the fistulae, causing hyperplasia. The differences seen in proximal versus W.R . Hanson et al. distal fistulae are tissue-specific, whereas differences in proximal and distal intestine in continuity could result in part from the effect of the luminal environment on blood perfusion rate. The proposed physiological mechanism may provide a reasonable explanation for the apparent systemic control of intestinal cell renewal. This mechanism could allow gross adjustments to various perturbations, whereas the negative feedback of functional villus cells to crypt cell production could be a fine-tuning and local control. Direct evidence of a physiological control of intestinal cell renewal is lacking but the prospect that a relationship exists is promising. ACKNOWLEDGMENTS This study was supported by a grant from The Netherlands Foundation in Medical Research, and by National Institute of Arthritis and Metabolic Disease Research Grant AM- 15758. The authors thank Dr H. Galjaard, Dr R. J. M. Fry, and Dr H. P. Schedl for their advice and criticism. Mr N. H. C. Bruns is gratefully acknowledged for his technical assistance with the scanning electron microscopy, and Mr J. Bos for his excellent care and maintenance of the animals. We wish to thank Ms Cheryl Bedwell for her excellent secretarial help. We are also grateful to Ms Jo Ann Peiffer for secretarial assistance and for help with the figures.

Journal

Cell Proliferation – Wiley

Published: Nov 1, 1977

There are no references for this article.