1. Recent initiatives which advocate an increase in dietary polyunsaturated fat intake have led to the study of the effects of this upon gastrointestinal function. 2. Weanling rats were for 21 weeks fed diets containing 10% fat that were either high or low in polyunsaturated fats. Jejunal function was studied in vitro in an Ussing chamber. 3. Basal intestinal short-circuit was similar in both groups. 4. A decreased EC 50 for the non-neural electrogenic secretory responses to acetylcholine, bethanecol and isobutylmethylxanthine was apparent in the jejuna of rats fed a diet high in polyunsaturated fatty acids. 5. Submaximal electrogenic galactose absorption was increased in the rats fed a diet high in polyunsaturated fatty acids. 6. Changing the composition of dietary lipid resulted in a change in the fatty composition of the apical enterocyte membrane. 7. Diets high in polyunsaturated fatty acids may be both prosecretory and proabsorptive in the small intestine.

1. The potential contribution of enteric pathways to the extra-renal excretion of uric acid in chronic renal failure was evaluated by measuring [ 14 C]-urate and 36 Cl − transport in isolated, short-circuited segments of jejunum, ileum and distal colon from normal rats and rats with chronic renal failure (five-sixths nephrectomized) 6 weeks after surgery. Plasma and urine creatinine and urate concentrations were also determined to establish urate status in chronic renal failure. 2. A 40% reduction in creatinine clearance was observed in the group with chronic renal failure, resulting in a two-fold increase in plasma creatinine concentration (0.064 ± 0.001 mmol/l, n = 9, to 0.120 ± 0.008 mmol/l, n = 14). In contrast, there was no change in plasma urate concentration, despite a 75% reduction in urate clearance. 3. The transport studies demonstrated alterations in intestinal urate transport in rats with chronic renal failure compared with normal rats. Net absorption of urate was observed in the distal colon of control rats, whereas a significant secretory flux was observed in rats with chronic renal failure (from 0.88 ± 0.16 to −1.62 ± 0.81 nmol h −1 cm −2 , n = 12). Urate secretion was induced in the jejunum of the group with chronic renal failure (−0.15 ± 0.25 in control and −1.67 ± 0.35 nmol h −1 cm −2 in chronic renal failure, n = 11). Urate transport across the ileum was not altered in chronic renal failure and there was no net flux of urate in either the normal group or the group with chronic renal failure. Net Cl − absorption, which was observed in all segments of the control animals, was reversed to net secretion in chronic renal failure. 4. These studies indicate that urate homoeostasis is maintained in chronic renal failure and both the jejunum and colon may be actively involved in the extra-renal excretion of urate. These studies do not exclude the possibility that urate metabolic pathways are involved in urate homoeostasis in chronic renal failure.

1. A vicious cycle of malabsorption and malnutrition has been implicated in the pathogenesis of protracted diarrhoeal disease in infancy. Vitamin E deficiency is common in malnourished infants with protracted diarrhoea. We have studied the effects of chronic vitamin E deficiency on small-inestinal secretion and absorption in the rat. 2. Weanling rats were fed vitamin E-sufficient or -deficient diets for 21 weeks. Jejunal function was studied in vitro in an Ussing chamber after this period. 3. Steady-state isotopic flux experiments in unstimulated tissues demonstrated net Na + and Cl − secretion in vitamin E-deficient jejuna but net Na + and Cl − absorption in vitamin E-sufficient jejuna. 4. Basal intestinal short-circuit current was the same in both groups. 5. Cyclic nucleotide and maximal non-neuronal acetylcholine-mediated electrogenic secretion were increased in vitamin E-deficient jejuna. 6. Exogenous 5-hydroxytryptamine (serotonin) induced a smaller increment in electrogenic secretion in vitamin E-deficient jejuna. 7. Vitamin E-deficient jejuna were less responsive to exogenous noradrenaline, resulting in a smaller α 2 -adrenergic-mediated decrease in intestinal short-circuit current. 8. Fasting for 72 h produced a greater increment in intestinal short-circuit current in vitamin E-deficient jejuna. 9. Chronic vitamin E deficiency is prosecretory in the small intestine and may predispose to the perpetuation of protracted diarrhoeal diseases.

1. This paper reports a comparison of the kinetics of influx into hamster jejunum of a series of dipeptides of neutral, basic and acidic amino acids, and a tripeptide of neutral amino acids, with those of corresponding free amino acids. 2. K t , the substrate concentration at which the transport rate is half the maximal transport rate, and V max , the maximal transport rate, were more similar from one peptide to another than among amino acids, with the result that, over a wide range of concentrations, rates of influx of individual peptides varied much less than those of amino acids. 3. It is suggested that this may account for the rates of absorption of amino acids being closely related to the amino acid composition of the protein fed, instead of being widely dissimilar as with corresponding mixtures of free amino acids. 4. With neutral amino acids, both K t and V max. fell with increasing length of the side-chain, as observed on many previous occasions. This did not occur with the corresponding homologous dipeptides, which shows that the hypothesis that the apparent affinity for transport is related to the lipophilic properties of the side-chain cannot be applied to peptides.

1. The transport of 5-fluorouracil, uracil and thymine has been studied with isolated jejunal loops of rat small intestine. High performance liquid chromatography was used to identify the pyrimidines and measure their concentrations. 2. When the lumen of the intestine was perfused with 5-fluorouracil or uracil at 0.1 mmol/l or 0.2 mmol/l, the concentration in the serosal secretions was significantly higher than that in the lumen. For thymine the serosal concentration exceeded that in the lumen only at 0.1 mmol/l. 3. Analysis of the mucosal tissue water at the end of the perfusion demonstrated that when the intestinal lumen was perfused with any one of the three pyrimidines at 0.1 mmol/l or 0.2 mmol/l the concentration within the tissue was significantly above that in the lumen. 4. After an initial lag period linear rates of transport from the lumen to the serosal secretions were obtained for all three pyrimidines over a 10-fold concentration range from 0.1 mmol/l to 1 mmol/l. 5. Uracil and thymine inhibited the transmural transport of 5-fluorouracil. 6. The transport of 5-fluorouracil was also studied with a vascularly perfused preparation of rat small intestine. At 0.1 mmol/l the rate of transmural transport of the drug in this preparation was substantially higher than in the jejunal loops. This difference was eliminated by adding 5-fluorouracil to the vascular perfusate, suggesting that the higher transport rate in the vascularly perfused preparation was due to the lower serosal drug concentrations in the mesenteric circulation of the perfused intestine. 7. At a concentration of 5 mmol/l 5-fluorouracil inhibited water transport in the isolated loops and transmural D-galactose transport in the vascular perfusions.

1. The jejunal absorption of glucose from (1–4)-linked glucose oligomers including maltotriose has been compared with that from free glucose and sucrose in normal subjects. 2. A steady-state perfusion technique in vivo was used to study proximal jejunal assimilation of isotonic sugar-saline solutions isocaloric with 140 mmol/l glucose. Endogenous α-amylase was excluded from the test segment by proximal balloon occlusion. The glucose oligomer mixture consisted mainly of maltotriose, maltotetraose, maltopentaose and maltoheptaose. 3. Glucose absorption was significantly faster from maltotriose alone and from the glucose oligomer mixture than from 140 mmol/l glucose controls, whereas glucose absorption from 70 mmol/l sucrose was similar to that from 70 mmol/l glucose plus 70 mmol/l fructose. 4. Hydrolysis in vivo of maltotriose, the oligomer mixture and sucrose were similar, indicating that capture of glucose released by brush border sucrose hydrolysis was less efficient than that associated with (1–4)-linked oligomer hydrolysis. This suggests that the stoichiometric relationship of the active hydrolysis sites for sucrose to the glucose transport system is less advantageous than that of active sites for maltose hydrolysis. 5. Hydrolysis of oligomers larger than maltohexaose may be rate limiting for glucose absorption in the absence of luminal amylase activity.

1. The mode of action of acipimox (5-methyl-pyrazine carboxylic acid 4-oxide), an hypotriglyceridaemic agent, was examined in human adipose tissue and intestinal mucosa. 2. The rates of release of fatty acids and glycerol from human adipose tissue were measured in vitro. The release of fatty acids and glycerol from adipose tissue maximally stimulated by isoprenaline (10 −5 mol/l) fell by 40 and 25% respectively ( P <0.025 and P <0.025) in the presence of acipimox (10 −5 mol/l). In submaximally stimulated adipose tissue (isoprenaline 10 −7 mol/l) acipimox (10 −4 mol/l) fully inhibited release of fatty acids ( P <0.05) and glycerol ( P <0.025) to basal rates. In unstimulated adipose tissue acipimox (10 −3 mol/l) reduced the rate of glycerol release ( P <0.05), but not the rate of fatty acid release. 3. Cholesterol synthesis in jejunal mucosa was measured in vitro by the incorporation of [2- 14 C]-acetate into sterols. Addition of cholesterol to the incubation reduced [2- 14 C]acetate incorporation into sterols from 8.7 ± 2.1 (mean ± standard error) to 3.7 ± 1.0 pmol h −1 mg −1 of tissue ( P <0.01). Acipimox at 10 −4 -10 −2 mmol/l had no consistent effect on cholesterol synthesis. 4. Acipimox appears to exert its main hypolipidaemic effect by reducing lipolysis and free fatty acid flux to the liver, thereby reducing the precursor pool size of very low density lipoprotein (VLDL)-triglyceride and VLDL synthesis.

1. To compare the effects of intravenous and intraluminal hydrocortisone on jejunal transport, the proximal jejunum was perfused with glucose (28 mmol/l) in saline in two groups of normal subjects. 2. In the first group of seven subjects, compared with the control period results there were no changes in sodium, water and glucose absorption during the intravenous administration of hydrocortisone. In contrast, intraluminal hydrocortisone (100 mg/l) increased sodium, water and glucose absorption by 169%, 223% and 81% respectively ( P < 0.001 in each case) above the control values, when peripheral plasma cortisol levels were similar to those achieved with intravenous hydrocortisone. 3. In the second group of three subjects, intraluminal hydrocortisone (10 mg/l and 30 mg/l), followed by an intravenous infusion of hydrocortisone, had no effects on sodium, water and glucose absorption. 4. In a third group of six normal subjects perfused with fructose (28 mmol/l) in saline and bicarbonate (28 mmol/l) in saline intraluminal hydrocortisone (100 mg/l) had no effect on solute and water absorption. 5. These results suggest that intraluminal hydrocortisone stimulates glucose-coupled sodium transport by exerting a topical effect on the apical membrane of the jejunal mucosa.

1. To compare the bioavailability and the elimination of propranolol in seven untreated coeliac patients and six normal subjects, plasma concentrations were measured after oral and intravenous propranolol. The bioavailability and clearance of propranolol were similar in both groups. 2. With the use of a perfusion technique, propranolol absorption in the proximal jejunum was found to be decreased by 71% in five untreated coeliac patients, compared with the absorption in four normal subjects. 3. These results indicate that propranolol absorption is decreased in the proximal jejunum in untreated coeliac disease but overall absorption in the small bowel is not impaired.

1. We studied intestinal glucose transport in pigs during the acute and convalescent phases of an invasive viral enteritis, transmissible gastroenteritis. 2. When diarrhoea was severe 40 h after experimental infection, net absorption of glucose, Na + and water, measured by marker perfusion in the jejunum, was reduced; the enhancement of Na + and water absorption in response to increasing perfusate glucose concentrations up to 120 mmol/l was diminished compared with the response observed in control and convalescent pigs. 3. Measured in vitro , 40 h after infection, unidirectional fluxes of 3- O -methyl-d-glucose across the jejunal epithelium were reduced and net absorption of the sugar was obliterated. Phlorizin (0.05 mmol/l), which completely inhibited net 3- O -methyl-d-glucose absorption in control tissue, had no significant effect on transmissible gastroenteritis jejunum. 4. Our data suggest that in this invasive viral enteritis, which closely resembles human rotavirus enteritis, glucose absorption is impaired as a result of defects in both active and passive glucose flux. 5. Differences between the mechanisms of viral diarrhoea, demonstrated by our study and those of the enterotoxigenic diarrhoeas, should be taken into consideration in formulating active therapeutic measures for children with acute viral diarrhoea.

1. Analytical subcellular fractionation techniques have been applied to human jejunal homogenates to study the localization of serotonin and motilin. 2. Serotonin, detected by a fluorimetric assay, showed two distinct particulate distributions in the density gradients with modal densities of 1.14 and 1.20. Motilin, whether detected by N- terminal or C -terminal specific antibodies showed similar density distributions with a modal density of 1.20, coinciding with the denser peak of serotonin. 3. These results lend support to the histochemical suggestions that motilin is located in a serotonin-containing granule.

1. The kinetics of 2-min uptake of l-lysine and l-lysyl-l-lysine have been studied by using rings of everted hamster intestine in vitro , and values for K t and V max , established. 2. On a molar basis, mediated uptake was more rapid for the amino acid than for the peptide. Non-mediated uptake was more rapid for the peptide than for the amino acid. 3. Comparison of relative rates of uptake of lysine from equivalent solutions of lysine and lysyl-lysine showed that at low concentrations, uptake of lysine was less rapid from the peptide than from the amino acid, whereas at high concentrations, uptake of lysine was more rapid from the peptide than from the amino acid. This type of effect of concentration on relative rates of uptake from equivalent solutions of amino acid and peptide has not previously been described.

1. In a system in vitro , 65 Zn accumulation by jejunal mucosal biopsies from patients with acrodermatitis enteropathica was found to be markedly reduced compared with controls. 2. We suggest that defective uptake of zinc by enterocytes is the primary abnormality responsible for the zinc deficiency underlying this disorder.

1. Enterocytes, isolated from the proximal jejunum and distal ileum of normal and prednisolone-treated rats, were homogenized and fractionated by isopycnic centrifugation on sucrose density gradients. The distributions of marker enzymes for the principal subcellular organelles, RNA and protein were determined and related to the activities per enterocyte. 2. In enterocytes from the jejunum and ileum of prednisolone-treated animals the activities of particulate brush-border enzymes and of both soluble and mitochondrial malate dehydrogenase were increased compared with those of the control system. The equilibrium density of the brush borders was enhanced in the prednisolone-treated jejunum. The modal densities of the other organelles were unaltered by prednisolone administration. 3. There was a large increase in the total RNA content of enterocytes from the jejunum and ileum of prednisolone-treated animals. This was predominantly associated with a distinct particulate component, indicative of a proliferation of the rough endoplasmic reticulum and consistent with an enhanced rate of protein synthesis. 4. Studies of latent brush-border enzyme activities, the mechanical fragility of isolated brush borders and electron microscopy suggest that steroid administration results in no marked alterations in the gross conformation of the brush- border membrane or in the orientation of the enzymes within the membrane.

1. Rats were fed with the elemental diet Vivonex for 1 or 3 months and their jejunal histology was compared with that of an equal number of rats fed on a normal diet. 2. After 1 month of Vivonex feeding a significant reduction in the ratio of crypt height: villus height (CH: VH) was found in the Vivonex-fed rats ( n = 4) compared with the control rats ( n = 4) ( P < 0.05). 3. After 3 months the CH:VH ratio was also reduced in the Vivonex-fed rats ( n = 18) compared with control rats ( n = 18) ( P < 0.002). Villus height was significantly increased ( P < 0.002) and crypt height decreased ( P < 0.05). 4. Jejunal protein content, alkaline phosphatase and disaccharidase activity were also determined in 12 control and 12 Vivonex-fed rats from the 3 months study. 5. Alkaline phosphatase activity was increased from a control value of 201 ± 8 to 243 ± 15 munits/cm in the Vivonex-fed rats ( n = 12) ( P < 0.05) but no significant changes in lactase, sucrase or maltase activities were found. 6. The observed decrease in the CH:VH ratio suggested an improved survival of the mature enterocyte population during elemental diet feeding.

1. Enterocytes, isolated from the proximal jejunum and distal ileum of the rat, were homogenized and their organelles separated by isopycnic centrifugation on continuous sucrose density gradients. The distributions of marker enzymes for the principal organelles, RNA and protein were determined in the sucrose gradients and related to the activities per enterocyte. 2. In the jejunum the modal equilibrium densities of the various organelles were: brush borders (1.20), lysosomes (1.20), peroxisomes (1.19), mitochondria (1.17) and basal-lateral membranes (1.13). These values were not significantly different in the ileum. The activities of brush-border enzymes, soluble and mitochondrial malate dehydrogenase, soluble and membrane-associated lactate dehydrogenase and particulate protein content, however, were greater in the jejunal than the ileal enterocytes. 3. Detergent exposed latent alkaline phosphatase activity in jejunal enterocytes and indicated that this enzyme is present not only in the brush border but also in the basal-lateral membrane and soluble fractions of the cell. 4. Isolated jejunal brush-border preparations showed latent activities of both alkaline phosphatase and γ-glutamyltransferase whereas the activities of α-glucosidase and leucyl-β-naphthylamidase were not affected by detergent. Mechanical disruption of these preparations suggested the presence of two forms of alkaline phosphatase in the brush border and provides a technique to assess membrane fragility.

1. The subcellular distribution of peptidase activities in the normal human jejunum against glycine and leucine homopeptides has been investigated with an analytical fractionation technique. 2. An 8000 g -min supernatant was prepared from homogenates of Crosby capsule biopsy specimens and subjected to isopycnic centrifugation in a Beaufay automatic zonal rotor. 3. The distribution of subcellular organelles in the gradient was established by measurement of organelle-specific marker enzymes. 4. A sensitive fluorimetric assay for glycine peptidase was developed and used for the localization of peptidase activity with peptides composed of from two to five glycine residues as substrates. 5. Glycine peptidase activity was located in the cytosol and in the brush-border membrane but the distribution of activity varied markedly with the chain-length of substrate; the longer the peptide the greater the proportion of activity associated with the brush border. Leucine peptidase showed a similar variation in cytosol—brush border distributions. 6. The results are consistent with concepts that suggest absorption and intracellular hydrolysis of small peptides and brush-border digestion of larger peptides.

1. The validity of evidence for intact peptide absorption, derived from analysis of the relation of water and total solute absorption, has been tested. 2. Solute and water absorption from saline solutions of the disaccharide maltose have been studied in the normal human jejunum, using a double-lumen perfusion technique with a proximal occlusive balloon. It was expected that maltose would yield very different results from peptides, because maltose is virtually completely hydrolysed before absorption, whereas a proportion at least of some peptides is transported into the intestinal mucosal cells before hydrolysis. This expectation was not confirmed by experiment. 3. The assumption that the absorbate is always isotonic with plasma has been tested by altering the osmolality of glucose/saline solutions perfused in the jejunal lumen. This assumption was not substantiated by experiment, as when the luminal fluid was hypertonic to plasma, so was the absorbate. 4. It is suggested that our findings with peptides and saccharides could be explained by the production of a hypertonic absorbate by hydrolysis of these solutes to their monomer units. We therefore conclude that analyses of the relation of net solute and water absorption cannot be used to predict the form in which peptides enter the mucosal cells.

1. Portions of closed jejunal biopsies were gently homogenized in isotonic sucrose or sorbitol and centrifuged at 800 g for 10 min to prepare a cell extract. 2. The extract was fractionated in a single-step procedure by isopycnic centrifugation on a continuous sucrose or sorbitol density gradient with the Beaufay automatic zonal rotor. 3. The subcellular organelles were located in the density gradient by assay of marker enzymes and previously unassigned enzymes were localized to particular organelles. 4. The following organelles were characterized by their modal equilibrium densities in sucrose density gradients: brush borders (1·21), peroxisomes (1·18), mitochondria (1·16), endoplasmic reticulum (1·16), basal—lateral membranes (1·12). At least three distinct populations of lysosomes with different modal densities and enzyme content were demonstrated. 5. This analytical fraction technique can be used to study the subcellular pathology of human jejunum.

1. Pteroylmonoglutamic acid (PteGlu) absorption has been measured by using the technique of small-intestinal perfusion with tritiated PteGlu in normal subjects and in patients with coeliac disease. 2. At similar intrajejunal pH, patients with untreated coeliac disease have significantly less PteGlu than normal subjects and patients with treated coeliac disease. 3. The ‘resting’ pH in the jejunum did not differ markedly between normal subjects and patients with coeliac disease. 4. Increasing pH decreased PteGlu absorption in patients with coeliac disease and in normal subjects. 5. These findings suggest that PteGlu malabsorption in coeliac disease is not due to abnormally high pH in the jejunum.

1. The absorption in vivo of d-galactose by the rat small intestine has been examined in proximal jejunum and distal ileum by use of a recirculation—perfusion technique. 2. Multiple sequential perfusions over 4 h produced no subsequent functional or morphological damage in the perfused segments. 3. Absorption of galactose from 8 and 64 mmol/l solutions was found to be independent of flow rate over the range 1·0–6·5 ml/min. 4. Galactose absorption in both the jejunum and the ileum exhibited saturation kinetics of the Michaelis—Menten type, and phlorrhizin sensitivity. Sorbose was only absorbed minimally. These observations demonstrate that galactose is absorbed by carrier-mediated transport and that there is no significant passive diffusive component in vivo. 5. Under the stated experimental conditions, the maximum absorptive capacity was 4·5 times greater in the jejunum than in the ileum. The Michaelis constant for galactose was higher in the jejunum than in the ileum. 6. Enterocytes were isolated from perfused segments and quantified by DNA assay with a correction for yield. In this manner, the absorptive capacity per enterocyte was calculated. 7. The maximum absorptive capacity per enterocyte was 3·5 times greater in the jejunum than in the ileum.

1. The effects in rats of an oral pharmacological dose of prednisolone on mucosal function, enzymology, lysosomal membrane fragility, morphology and cell kinetics have been examined in proximal jejunum and distal ileum. 2. The maximum absorptive capacity for galactose was significantly greater in both the jejunum and the ileum of the steroid-treated animals. This was due to an increase in carrier-mediated transport in the individual enterocytes and not to a change in the cell population. The Michaelis constant for galactose was not significantly altered by prednisolone. 3. Activities of brush-border and mitochondrial enzymes and of RNA were increased in isolated enterocyte preparations from the jejunum and ileum of the steroid-treated group. 4. Lysosomal membrane fragility was unaltered in the prednisolone-treated group. 5. Morphometrical observations confirmed that the size of the enterocyte population was unaltered by prednisolone. Studies on cell kinetics indicate that the effects of prednisolone are due to a direct action on the enterocyte and not secondary to changes in migration rate.

1. The combination of a wire-mesh support with the roller-tube technique is described as a procedure for the culture of human jejunal mucosa in vitro. 2. The technique has been applied to fragments (approximately 10 mg) of jejunal biopsies from both normal subjects and patients with coeliac disease. 3. The cultured tissue has been shown by radioautography to incorporate [ 3 H]leucine into proteins of the villus epithelial cells and [ 3 H]thymidine into nucleic acid, predominantly by the enterobiasis. 4. Although the tissue protein and DNA contents fall during culture, it was found that the combined tissue and medium DNA content remained constant during culture and may be used as a reference for enzyme and biochemical studies on cultured intestinal biopsies.

1. This paper describes the results of a survey of the effects of peptides and amino acids on uptake by rings of everted hamster jejunum in vitro of glycylsarcosylsarcosine, a tripeptide which is taken up by an active mechanism but is very resistant to hydrolysis, appearing intact in the rings. The results of a small number of similar experiments with β-alanylglycylglycine, another tripeptide which is taken up with very little hydrolysis, are also described. 2. Uptake of the two tripeptides was inhibited by other di- and tri-peptides, but not by free amino acids. The results suggest that dipeptides and tri-peptides share a common uptake mechanism. The tetrapeptide glycylsarcosylsarcosylsarcosine did not inhibit uptake of glycylsarcosylsarcosine, and appears to be unable to utilize the uptake mechanism. 3. The results add to information about the influence of molecular structure on intestinal uptake of peptides by the system used by glycylsarcosylsarcosine, which is shared by a wide range of other di- and tri-peptides. In conjunction with previous results, they suggest that substitution of the N -terminal amino or C -terminal carboxyl groups reduces affinity for transport, that the presence of a β-amino acid residue in a peptide is tolerated by the transport system, and that the presence of a d-amino acid residue reduces affinity for transport. Some peptides containing or made up of basic or acidic amino acid residues appear to have a low affinity for the transport system used by glycylsarcosylsarcosine. 4. Of two biologically active peptides, one, cephalexin, a peptide antibiotic, inhibited uptake of glycylsarcosylsarcosine and is probably transported by the same system. The other, prolylleucylglycineamide, which has the action of a hypothalamic regulatory factor, did not, and its structural features may make it unsuitable for carrier-mediated transport by the small intestine.

1. This paper describes the uptake by rings of everted hamster jejunum in vitro of three peptides with structural features making them resistant to hydrolysis, glycylsarcosylsarcosine, glycylsarcosylsarcosylsarcosine and β-alanylglycylglycine. 2. Glycylsarcosylsarcosine was taken up by a saturable mechanism and accumulated intact in the intracellular compartment of the intestinal wall, apparently against an electrochemical gradient. Its uptake was reduced by Na + -replacement, anoxia and metabolic inhibitors. It was concluded that uptake of this peptide was the result of Na + -dependent active transport. 3. Glycylsarcosylsarcosylsarcosine was very poorly taken up and its uptake did not appear to be the result of active transport. 4. β-Alanylglycylglycine appeared intact in the intracellular compartment of the intestinal wall on a substantial scale though it was not concentrated. No satisfactory evidence of uptake by a saturable mechanism was obtained. Uptake was, however, inhibited by anoxia, 2,4-dinitrophenol and Na + -replacement. Reasons are given for supposing that uptake of this peptide may be the result of Na + -dependent active transport by the same carrier as that utilized by glycylsarcosylsarcosine. 5. The results suggest that provided that they escape brush-border hydrolysis, tripeptides, like dipeptides, are actively transported into the absorptive cells of the intestinal mucosa, but that the ability of these cells to take up peptides by an active mechanism is unlikely to extend to tetrapeptides.

1. Perfusion of the human jejunum with low concentrations of glycine-conjugated bile acids in physiological solutions induced net fluid flow that varied between absorption and secretion, with only slight variation of the luminal osmotic and ionic milieu. 2. Transmucosal net flux rates for water and the principal ions were calculated. Regression analysis of the flux data was consistent with the concept of a ‘net transported fluid’ which was iso-osmotic with respect to the lumen, with a superimposed fixed net anion exchange unaffected by the rate or direction of bulk flow. 3. Recalculation of earlier data from comparable studies of the human colon showed similar relationships consistent with varying iso-osmotic bulk flow and fixed ion exchange, the latter differing from that found in the proximal bowel. 4. Studies were performed also in the human ileum. Variable iso-osmotic bulk flow was again encountered, but ion exchange was of a lower magnitude than in the jejunum or colon. Qualitatively, ion exchange in the ileum was intermediate between the jejunum and colon. 5. These analyses suggest that transmucosal bulk flow and ion exchange may be quantitatively and spatially independent processes. They provide support for the hypothesis that bulk flow may be intercellular (and hence extracellular), while ion exchange may take place across the luminal face of the mucosal cell. 6. Secretory agents, such as the dihydroxy bile acids, provide a useful means of analysing solute-solvent flow relationships in greatly differing bulk flow conditions with relatively stable physicochemical parameters.

1. The effect of conjugated and unconjugated bile acids in various concentrations on the absorption of water, sodium and chloride in the proximal jejunum in man has been studied using a perfusion system incorporating a proximal occluding balloon. 2. Unconjugated cholic acid and its glycine and taurine conjugates had no significant effect on water and electrolyte absorption. Unconjugated deoxycholic acid significantly inhibited water and electrolyte absorption at 3 m m concentration and net secretion occurred at concentrations of 5 m m , 7 m m and 10 m m . Similar results were found when taurodeoxycholic and glycodeoxycholic acid solutions were used. A greater effect was noted when higher concentrations of bile acids were used. 3. Unconjugated chenodeoxycholic acid caused net water and electrolyte secretion at 3 m m concentration and this effect became more marked with higher bile acid concentrations. Inhibition of absorption also occurred with taurochenodeoxycholic acid in a concentration of 3 m m and net secretion occurred at a concentration of 10 m m . 4. The results show that dihydroxy bile acids, conjugated or unconjugated, inhibit water and electrolyte absorption in the proximal jejunum in low concentration and cause net secretion in high concentration.

1. Using a double-lumen tube perfusion system the rates of glycine, glycylglycine, and glycylglycine and glucose absorption from a 30-cm jejunal segment have been studied in vivo in a group of relatively normal Zambian African subjects. 2. To determine the kinetic curve for glycine absorption, four subjects were given consecutive perfusions of 50, 100 and 150 mm-glycine. 3. Six other subjects had consecutive perfusions of (1) a 100 mm-glycine and (2) a 50 mm-glycylglycine solution. Five of the six had a higher absorption rate of glycine from the glycylglycine solution. When data from a further six similar subjects in another study are included, the mean rate of glycine absorption is significantly greater from the glycylglycine compared with the glycine solution ( P < 0·001). 4. A further six subjects were given consecutive perfusions of (1) 50 mm-glycylglycine, (2) 50 mm-glycylglycine and 200 mm-glucose, and (3) 200 mm-glucose. The absorption rate of glycine from glycylglycine was lower in all subjects when glucose was present in the perfusing fluid ( P < 0·01). Although the mean glucose absorption rate was lower when glycylglycine was present in the perfusing fluid, the difference was not significant.

1. The transport of water, sodium and chloride across the jejunal mucosa was studied in fourteen patients with untreated adult coeliac disease by using a triple-lumen tube perfusion system. The results were compared with those obtained by identical methods in fifteen normal control volunteers. 2. Evidence was obtained suggesting that not only was absorption impaired in the adult coeliac disease group, but that in some patients there was secretion of water and electrolytes. 3. No significant improvement had occurred in six patients 2–4 months after withdrawal of gluten, nor in one patient after 8 months, in spite of histological improvement and clinical remission.

1. To investigate the effect of different intraluminal concentrations on the mutual inhibitive effect of glycine and glucose on their jejunal absorption rates, eighteen convalescent Zambian African patients who had no clinical evidence of intestinal disease or of malnutrition were given constant intrajejunal infusions with those solutes either together or alone. A double-lumen tube perfusion system was used, and three solutions containing (A) glycine, (B) glycine and glucose, and (C) glucose, all of which were rendered iso-osmotic with sodium chloride, were perfused in random order at 12·0 ml/min. The concentration of glycine in the perfusing fluid was either 10 or 20 mm, and that of glucose either 100, 200 or 280 mm. By reference to polyethylene glycol 4000, the absorption rates of the solutes and water were calculated for a 30 cm jejunal segment. 2. At a glucose concentration of 200 or 280 mm, but not 100 mm, the mean rate of glycine absorption was decreased by approx. 30%. Glucose absorption rates were not significantly altered by glycine. 3. These observations, taken in conjunction with those from a previous investigation, are consistent with the view that there are two mechanisms for the jejunal absorption of glycine in man, one of which is inhibited by glucose at high intraluminal concentration.