Intestinal enzymes activities in isolated villus and crypt cells during postnatal development of the rat

Hemidesmosome integrity protects the colon against colitis and colorectal cancer

OBJECTIVE

Epidemiological and clinical data indicate that patients suffering from IBD with long-standing colitis display a higher risk to develop colorectal high-grade dysplasia. Whereas carcinoma invasion and metastasis rely on basement membrane (BM) disruption, experimental evidence is lacking regarding the potential contribution of epithelial cell/BM anchorage on inflammation onset and subsequent neoplastic transformation of inflammatory lesions. Herein, we analyse the role of the α6β4 integrin receptor found in hemidesmosomes that attach intestinal epithelial cells (IECs) to the laminin-containing BM.

DESIGN

We developed new mouse models inducing IEC-specific ablation of α6 integrin either during development (α6^ΔIEC) or in adults (α6^ΔIEC-TAM).

RESULTS

Strikingly, all α6^ΔIEC mutant mice spontaneously developed long-standing colitis, which degenerated overtime into infiltrating adenocarcinoma. The sequence of events leading to disease onset entails hemidesmosome disruption, BM detachment, IL-18 overproduction by IECs, hyperplasia and enhanced intestinal permeability. Likewise, IEC-specific ablation of α6 integrin induced in adult mice (α6^ΔIEC-TAM) resulted in fully penetrant colitis and tumour progression. Whereas broad-spectrum antibiotic treatment lowered tissue pathology and IL-1β secretion from infiltrating myeloid cells, it failed to reduce Th1 and Th17 response. Interestingly, while the initial intestinal inflammation occurred independently of the adaptive immune system, tumourigenesis required B and T lymphocyte activation.

CONCLUSIONS

We provide for the first time evidence that loss of IECs/BM interactions triggered by hemidesmosome disruption initiates the development of inflammatory lesions that progress into high-grade dysplasia and carcinoma. Colorectal neoplasia in our mouse models resemble that seen in patients with IBD, making them highly attractive for discovering more efficient therapies.

Intestinal microbiota differentially affect brush border enzyme activity and gene expression in the neonatal gnotobiotic pig

To study microbial influence on intestinal development pertaining to nutrient digestion, two separate gnotobiotic experiments were performed, each with 16 piglets allocated to four treatment groups: germfree (GF), monoassociation with Escherichia coli, monoassociation with Lactobacillus fermentum or conventionalization with faecal bacteria (CV). Enzyme activity and gene expression of lactase phlorizin hydrolase (LPH) and aminopeptidase N (APN) were measured in isolated enterocytes, harvested on day 14, using specific substrates and quantitative PCR respectively. Enterocytes of CV pigs had reduced APN activity, but had increased gene expression relative to GF, making the specific activity:mRNA (A:G) ratio dramatically lower (p < 0.05). Similarly, LPH A:G ratio was significantly reduced (p < 0.05) in enterocytes of CV pigs as compared with GF. The results of co-incubation of L. fermentum, E. coli and faecal bacteria with APN indicate a direct relationship between enzyme inactivation and specific A:G ratio in enterocytes. We conclude that enterocyte up-regulation of APN expression occurs as either a direct response to microbial colonization or as a feedback mechanism in response to reduced enzyme activity through microbial degradation. This mechanism may play a role in ensuring effective competition of the host with the intestinal microbiota for available nutrients.

Role of the mucosal barrier in toxin/microbial attachment to the gastrointestinal tract

An important component of bacterial infection of the gastrointestinal tract, particularly when enterotoxin disruption of gut function occurs, relates to the interaction of the bacterium and toxin with the intestinal surface. Adherence or attachment of bacteria lead to colonization and toxic/invasive diarrhoeal states. The purpose of this review is to consider the intestinal mucosal barrier as a deterrent to microbial/toxin attachment, particularly emphasizing the mucosal surface itself, which includes the mucus coat and microvillous membrane. Consequences of mucosal barrier deficiency, particularly the incomplete development of the mucosal barrier, result in bacterially induced diarrhoeal states such as toxigenic diarrhoea and necrotizing enterocolitis. To illustrate the importance of the mucosal barrier as a factor in controlling the external environment containing bacteria and bacterial toxins, recent research in our laboratory on the development of the mucosal barrier is presented. This compares mucosal surface functional control of antigen/toxin attachment and penetration with mucosal surface compositional changes. Finally, evidence of a preliminary nature will be provided to suggest that modifications of the underdeveloped mucosal barrier of the immature intestine by the ingestion of breast milk may act to prevent pathological interactions between the gut and microbes/toxins.

Saccharomyces boulardii enhances N-terminal peptide hydrolysis in suckling rat small intestine by endoluminal release of a zinc-binding metalloprotease

Saccharomyces boulardii (S. boulardii), a biotherapeutic agent effective in acute and chronic enterocolopathies, produces trophic intestinal effects at least in part mediated by the endoluminal release of polyamines. However, the effects of the yeast on peptide hydrolysis have not yet been studied. The objectives of this study were to assess in suckling rats the endoluminal and mucosal aminopeptidase activities in response to S. boulardii treatment and to analyze their related mechanisms. Peptidase activities were assayed on yeast cells by using several L-amino acid-p-nitroanilide substrates in the pH range of 2 to 10. A marked hydrolytic activity was found for L-leucine-p-nitroanilide that peaked at pH = 8 (K(m) = 0.334 mM, V(max) = 44.7 micromol.min(-1).g(-1) protein). N-terminal peptide hydrolysis was confirmed using as substrate L-Leu-Gly-Gly (K(m) = 4.71 mM, V(max) = 18.08 micromol.min(-1).g(-1) protein). Enzyme reactions were inhibited in the presence of 1 mM Zn(2+). Oral treatment of sucklings with S. boulardii significantly enhanced jejunal and ileal mucosal leucine-aminopeptidase activities by 24 and 34%, respectively, over controls. In concordance, aminopeptidase activity was enhanced in jejunal and ileal endoluminal fluid samples by 47 and 105%, respectively. By use of an IgG-purified antibody raised against the zinc-binding domain common to metalloproteases, the yeast aminopeptidase was immunoprecipitated and detected as an heteromeric enzyme of 108 and 87-kD subunits. S. boulardii, when given orally to suckling rats, is able to significantly enhance hydrolysis of N-terminal oligopeptides in both endoluminal fluid and intestinal mucosa by the endoluminal release of a leucine aminopeptidase that appears to be a zinc-binding metalloprotease belonging to the M1 family of peptidases.

Occurrence of IgA and IgG autoantibodies to calreticulin in coeliac disease and various autoimmune diseases

Calreticulin (CRT), a high-affintiy calcium binding protein and chaperone, was recently identified as one of the targets of autoantibodies in coeliac disease. We evaluated the level of IgA and IgG antibodies to CRT in sera from patients with coeliac disease and various autoimmune diseases. The level of antibodies to gliadin (shown previously to cross-react with CTR), isolated enterocytes and tissue transglutaminase were determined for comparison. The mean level of IgA antibodies to CRT was significantly higher (P< 0.001) in sera from coeliac patients with active disease (139.9+/-11.2 AU/+/-SE) than in healthy controls (20.9+/-1.7 AU). In sera of patients with systemic lupus erythematosus (SLE), insulin dependent diabetes mellitus (IDDM), multiple sclerosis (MS) and autoimmune thyroiditis (AT) or inflammatory bowel disease (IBD) the mean level (25.8+/-3.7 to 38.1+/-5.6 AU) did not exceed the cut-off value. A low level of these antibodies, however, was detected in some sera of patients with MS and IBD. The level of IgG anti-CRT antibodies was increased in coeliac patients (mean 125.4+/-8.0 AU, P< 0.001) when compared to that in healthy controls (33.9+/-2.3 AU). The IgG anti-CRT antibodies were also detected in about 30% of SLE patients sera (54.1+/-3.6 AU, P< 0.001), but the mean level reached only half that detected in coeliac patients.

Modulatory effect of rat small intestinal epithelial cell-conditioned medium on lymphocyte proliferation

The small intestinal epithelium plays an important role in the mucosal host defense. Intestinal epithelial cells have been known to release substances that suppress lymphocyte proliferation, suggesting an immunoregulatory function. We investigated how intestinal epithelial cells affect lymphocyte proliferation. Serum-free medium that was conditioned by incubating epithelial cells, particularly crypt cells, of the rat small intestine affected proliferation of allogeneic spleen lymphocytes stimulated with concanavalin A, as assessed by measuring cellular [3H]thymidine incorporation. Less than 1% and greater than 2% of the conditioned medium enhanced and suppressed, respectively, lymphocyte proliferation. The causative substances found in the conditioned medium were dialyzable and heat-stable. Suppression was not due to toxicity to splenocytes. Exposure of splenocytes to a suppressive concentration of the conditioned medium beginning at 30 min before an onset of lectin stimulation decreased the suppression of lymphocyte proliferation. Splenocyte exposure to the suppressive concentration of the conditioned medium beginning at 30 min to 4 h after the onset of the stimulation inversely strengthened the suppression. A brief exposure of splenocytes to the conditioned medium for the last 4 h during a total 72-h culture period still suppressed lymphocyte proliferation. Thus, intestinal epithelial cells produce low-molecular-weight lymphocyte proliferation-modulating substances that suppress the proliferation of lectin-activated lymphocytes, but not resting ones, by affecting earlier intracellular events and the following DNA synthesis when incubated in culture medium.

A quantitative histochemical study of alkaline phosphatase activity in isolated rat duodenal epithelial cells

A simple separation method enabling the quantification of alkaline phosphatase activity in unfixed, isolated, individual, duodenal epithelial cells has been presented. The activity of intestinal brush border-bound alkaline phosphatase has been demonstrated using naphthol AS-BI phosphate as a substrate and hexazotized New Fuchsin as a simultaneous coupling agent. The amount of final reaction product, as measured cytophotometrically, increases linearly with incubation time (up to 10 min) and with substrate concentration (up to 0.4 mM). Maximum enzyme activity was obtained at pH 8.9. Variation of the substrate concentration revealed the kinetic parameters for naphthol AS-BI phosphate as Km = 0.17 i 0.015 and Vmax = 13.9 +/- 1.38. The specificity of the enzyme reaction was confirmed by the complete inhibition of the enzyme activity in the presence of L-cysteine (10 mm) and 80% inhibition with L-phenylalanine (30 mM). Comparison of alkaline phosphatase activity in 8-microm cryostat sections (beginning at the tip and proceeding to the cryptal part) along the villus axis, with the activity of individual cells obtained by successive separation, revealed similar values of the percentage quotient derived from the entire activities in these two different methods. This suggests that the presented separation procedure gives rise to isolation of the respective cells from the corresponding areas of the villus. Finally, the isolated cells can be used as a valuable tool for the quantitative analysis of alkaline phosphatase activity along the length of the villus.

Premature stimulation of rat sucrase-isomaltase (SI) by exogenous insulin and the analog B-Asp10 is regulated by a receptor-mediated signal triggering SI gene transcription

The mechanism(s) by which insulin enhance prematurely the activity of brush border membrane (BBM) hydrolases in rat immature intestine is unknown. Therefore, we have compared the responses of four BBM enzymes [sucrase-isomaltase (SI), maltase, lactase-phloridzine hydrolase (LPH), and aminopeptidase] with exogenous insulin, the analog B-Asp10, IGF-I, and antireceptor MAb [insulin-receptor (IR) MAb] given to preweaning pups. Low doses of insulin caused a precocious induction of SI and of SI mRNA and stimulated maltase activity without effect on LPH nor on aminopeptidase activities. IGF-I given at the same dose as that of insulin had no detectable effect on these enzymes. Administration to sucklings of IR MAb prevented the effect of endogenous insulin by inhibiting the expression of SI and maltase without effect on LPH activity. B-Asp10, an insulin analogue that exhibits in vitro a 3.5-fold increase in receptor affinity with sustained signaling of the receptor tyrosine kinase, caused an overexpression of SI by 3.5-fold and of maltase by 1.5-fold compared with equivalent doses of normal insulin. The premature increases in SI activity, SI mRNA, and maltase activity in response to insulin were dose-dependent and were associated with dose-dependent increases in intracellular spermine and spermidine concentrations. In conclusion, these data suggest that the premature induction of SI by insulin is mediated by a dose-dependent signal initiated by binding of the hormone to its intestinal receptor, which after transduction into the cell indirectly triggers the transcription of the SI gene, possibly by changes in intracellular polyamine concentrations.

Maturation status of small intestine epithelium in rats deprived of dietary nucleotides

We describe the changes of several brush-border enzymatic activities in different subpopulations of epithelial cells, separated sequentially from the villus tip-to-crypt axis of the small intestine, induced by deprivation of dietary nucleotides for different periods of time in adult rats. Deprivation of dietary nucleotides lead to a decrease in the content and specific activity of alkaline phosphatase, leucine-aminopeptidase, maltase, sucrase and lactase in the villus tip, but had little effect on the crypt zone. The effect of the nucleotide deprivation on the enzymatic activity progressively increased towards the tip of the villus. Since these enzymes are maturation markers of the intestinal cells, these results support the idea that dietary nucleotides affect the maturation status of small-intestine epithelium.

Peroxisomes and peroxisomal enzymes along the crypt-villus axis of the rat intestine

The development of peroxisomes and expression of their enzymes were investigated in differentiating intestinal epithelial cells during their migration along the crypt-villus axis. Sequential cell populations harvested by a low-temperature method were identified by microscopy, determination of alkaline phosphatase and sucrase activities and incorporation of [3H]-thymidine into DNA. Ultrastructural cytochemistry after staining for catalase activity, revealed the presence of peroxisomes in undifferentiated stem cells located in the crypt region. Morphometry indicated that the number of these organelles increased as intestinal epithelial cells differentiate. Catalase activity was higher in the crypt cells than in the mature enterocytes harvested from villus tips. On the other hand, an increasing gradient of activity was observed from crypts to villus tips for peroxisomal oxidases, i.e. fatty acyl coA oxidase, D-amino acid oxidase and polyamine oxidase. These findings indicate that biogenesis of peroxisomes occurs during migration of intestinal epithelial cells along the crypt-villus axis and that peroxisomal oxidases contribute substantially to the biochemical maturation of enterocytes.

Maturation of villus and crypt cell functions in rat small intestine. Role of dietary polyamines

To evaluate the role of dietary polyamines in maturation of the rat small intestine, spermine was given orally twice daily to suckling pups from day 10 to day 14 postpartum at different doses: 0, 0.2, 0.5, 1, 2.5, and 5 mumol/dose. Compared to saline treated controls, spermine (5 mumol) produced significant increases in mucosal mass parameters (+12 to +57%, P < 0.05), induced prematurely an adult pattern of microvillous enzymes, and enhanced, respectively, by 19- and 3.5-fold (P < 0.01 vs controls) the concentration of the secretory component of p-immunoglobulins in villous and crypt cells. The response of microvillous enzymes (lactase, sucrase, maltase, and aminopeptidase) to spermine was dose-dependent and -specific since oral administration of arginine (5 mumol) or ornithine (5 mumol) was without effect. Intestinal changes were found to be significant (P < 0.05) for doses of spermine exceeding 1 mumol/day, which is in the range of the amount of polyamines provided by solid pellets at weaning (0.4 mumol/g). However, intestinal changes were undetectable at the physiological amounts of polyamines consumed by pups from rat milk during the suckling period (less than 0.3 mumol/day). Consistent with a direct effect of spermine on the intestinal cell, the cytosolic activity of ornithine decarboxylase was depressed by 27-fold (P < 0.005 vs controls) in the jejunum, while inhibition of ornithine decarboxylase by alpha-difluoromethylornithine did markedly decrease but did not suppress the cell response to spermine. Alternately, plasma corticosteronemia, which was virtually absent by day 14 in controls, ranged between 1.4 and 4.6 micrograms/dl in 60% (N = 9) of the spermine-treated rats.(ABSTRACT TRUNCATED AT 250 WORDS)

Cholesterol synthesis and esterification in isolated enterocytes: regulation by cholesterol and cholestyramine feeding

The purpose of the present study was to investigate the physiological control of the main regulatory enzymes of cholesterol metabolism in isolated enterocytes obtained from chick duodenum, jejunum and ileum. Cholesterol feeding resulted in an inhibition of 3-hydroxy-3-methyl-glutaryl-CoA reductase and mevalonate 5-pyrophosphate decarboxylase, while cholestyramine feeding increased reductase activity in all the regions studied and decarboxylase activity only in duodenum. Cholesterol feeding markedly increased acyl-CoA:cholesterol acyltransferase, but the effects of cholestyramine were less clear. The effects on transferase activity cannot be due to differences in the availability of acyl-CoA as exogenous substrate as no significant differences were found in acyl-CoA hydrolase activity after any of the dietary treatments. The effects of cholesterol feeding were related to changes in the cholesterol content of epithelial cells, whereas in the case of cholestyramine this relationship was less apparent.

Intestinal permeability in pediatric gastroenterology

The role of the physiologic barrier function of the small bowel and its possible role in health and disease has attracted much attention over the past decade. The intestinal mucosal barrier for luminal macromolecules and microorganism is the result of non-immunologic and immunologic defense mechanisms. The non-immunologic mechanisms consist of intraluminal factors such as gastric acid, proteolytic activity, and motility and of mucosal surface factors like mucin and the microvillous membrane. The immunologic mechanisms include secretary IgA and cell-mediated immunity. Both types of mechanism are not completely mature at birth. Maturation of this barrier is not finished before the 2nd year of life. One of the aspects of the mucosal barrier function can be estimated by the intestinal permeability (IP) for macromolecules. We use the differential sugar absorption test (SAT), in which the ratio of urinary excretion of a relatively large molecule, lactulose, is compared with that of a relatively small molecule, mannitol, after oral ingestion. Although the small intestine is permeable to certain macromolecules in normal developmental conditions, an increased IP could be involved in the pathophysiology of several diseases, including infectious diarrhea, food allergy, celiac disease, and Crohn's disease. It can be concluded that IP, as measured with the SAT, reflects the state of the mucosal barrier and is altered in several gastrointestinal diseases. The SAT is a non-invasive IP test that can be of diagnostic help to demonstrate alterations in the small-mucosal barrier function and may be useful to evaluate therapeutic interventions.(ABSTRACT TRUNCATED AT 250 WORDS)

Changes in fatty acid composition during cell differentiation in the small intestine of suckling piglets

Alterations of phospholipid fatty acid composition in the renewing intestine were studied in the infant piglet. Newborn piglets were fed from birth to 2 weeks of age a concentrated cow's milk which defined a standard supply of dietary fatty acids. Phospholipids were isolated from the whole mucosa, isolated intestinal cells and purified brush border membranes. Intestinal cells were isolated according to their position along the crypt-villus axis and cell phospholipids were extracted at each step of differentiation. Changes in fatty acid composition of cell phospholipids were related to those of lactase activity in the corresponding cell homogenates. In cell phospholipids, the relative content of linoleic and linoleic acids increased about 2-fold from crypt base to villus tip. Substantial contents of alkenylacyl glycerophospholipids (plasmalogens) were found in crypt cell phospholipids and in purified brush border membrane phosphatidylethanolamine (11 and 14% of alkenyl groups by weight of total fatty acids, respectively). The proportion of alkenylacyl glycerophospholipids decreased as cells ascended the villus column and became more differentiated. The results show that fatty acid compositional changes in differentiating cell phospholipids occurred in the immature intestine (before weaning) and suggest that these alterations might be related to the appearance of specific functions.

Methods of investigating intestinal transport

This article provides a brief overview of well-established in vivo and in vitro methods that have contributed the most to the understanding of transport processes across the gastrointestinal epithelium. In vivo perfusion techniques in humans revolve around double- and triple-lumen per oral tubes. In animals, in vivo techniques include the single and recirculation perfusion techniques and the double-isotope technique for measurement of net absorption. In vitro methods of studying intestinal transport include the everted gut sac technique, the Ussing chamber, the use of isolated epithelial cells, and the use of brush border and basolateral membranes isolated from enterocytes. The use of fluorescent probes for the measurement of intracellular ionic concentrations is a new and powerful in vivo technique that is now being applied to the gastrointestinal tract.

Nuclear proteins and chromatin structure in liver and intestinal epithelial cells of young growing and adult rats

Nuclei from liver and intestinal epithelial cells of young growing rats (39 days old) and adult rats (98 days old) were isolated. After addition of DNase I, the chromatin was separated by centrifugation (1100 g) into two fractions; the pellet (P) and the supernatant (S). The amount of chromatin released into the S-fraction was the same for the two age groups. The intestinal epithelial cell nuclei underwent self-digestion (in the absence of added DNase I) which was significantly higher in the young rats than in the adults. Subsequent examination using immunotechniques established the presence of non-sarcomeric myosin heavy-chain indicating that active genes were present for that protein. Hybridization of DNA with cDNA specific for myosin heavy-chain revealed that, relative to total DNA, the DNA retained in the P-fraction of both tissues and age groups contained the same amount of hybridizable sequences. In liver, nuclear proteins decreased significantly with age per g wet weight of tissue. In the enterocyte tissue, total DNA and protein increased with age. SDS-polyacrylamide gel or acetic acid-urea gel electrophoresis gave no age-related differences in the pattern of the proteins within each tissue. The results show that both liver nuclear DNA and protein decrease with age per g wet weight but increase per total tissue. In intestinal epithelial cells changes in chromatin structure with age were inherent within the nucleus.

Regulation of cholesterol metabolism in the intestine

The small intestine is a major site of cholesterol biosynthesis and lipoprotein degradation. It is also the organ responsible for absorbing dietary and endogenously produced biliary cholesterol. Cholesterol metabolism in the intestine is regulated by factors that will alter cellular cholesterol requirements. Thus, during increased cholesterol flux, which occurs by bile acid-faciliated cholesterol absorption or by lipoprotein-mediated uptake of cholesterol, cholesterol synthetic rates decrease and esterification rates increase. The mechanisms by which dietary fats regulate intestinal cholesterol metabolism are complex. Dietary fats alter membrane fatty acid composition. Simultaneously, they also promote lipoprotein secretion and alter cholesterol absorption. Intestinal 3-hydroxyl-3-methylglutaryl coenzyme. A reductase activity is regulated by enzyme phosphorylation-dephosphorylation. The regulation of acylcoenzyme A-cholesterol acyltransferase activity by this mechanism remains controversial. Data on hormone regulation of intestinal cholesterol metabolism are not conclusive, although progesterone seems to be a potent inhibitor of acylcoenzyme A-cholesterol acyltransferase activity in intestinal cell culture and isolated cells. In a manner similar to the regulation of cholesterol metabolism in other cells, the enterocyte responds appropriately to factors that alter cholesterol flux. Therefore, changes that occur in the rates of cholesterol synthesis and esterification will reflect the cholesterol requirements of the cell.

Stimulation of secretory IgA and secretory component of immunoglobulins in small intestine of rats treated with Saccharomyces boulardii

Saccharomyces boulardii (S.b.) is largely used in Western European countries for the treatment of acute infectious enteritis and antibiotic-induced gastrointestinal disorders. To study the mechanisms of the protective effect of S.b. against enteral pathogen infection, we assessed the response of the intestinal secretion of secretory IgA (s-IgA) and of the secretory component of immunoglobulins (SC) to oral administration of high doses (0.5 mg/g body weight, three times per day) of S.b. cells in growing rats. S.b. cells (biological activity: 2.8 x 10(9) viable cells/100 mg) were administered daily by gastric intubation to weanling rats from day 14 until day 22 postpartum. Control groups received either 0.9% saline or ovalbumin following the same schedule. Expressed per milligram of cell protein, SC content was significantly increased in crypt cells isolated from the jejunum (48.5% vs saline controls, P less than 0.05) as it was in the duodenal fluid (62.8% vs saline controls, P less than 0.01) of rats treated with S.b. Oral treatment with S.b. had no effect on the secretion of SC by the liver. In the duodenal fluid of rats treated with S.b. cells, the mean concentration of s-IgA was increased by 56.9% (P less than 0.01) over the concentration of s-IgA measured in saline controls.(ABSTRACT TRUNCATED AT 250 WORDS)

Cytochemical analysis of single villus peptidase activities in pig intestine during neonatal development

The present work uses a new technique of whole tissue cytochemistry and automated scanning to obtain measurements of peptidase activity and surface structure in intact villi microdissected from the jejunum of newborn and 28-day-old pigs. Intact villi from 28-day-old pigs are shown by this method to contain 30% more aminopeptidase N and 400% more dipeptidylpeptidase IV activity than is found on villi taken from newborn pig intestine. Villi taken from 28-day-old pig intestine are also half as long and twice as wide at their base as those taken from newborn animals. These changes in shape take place without significantly affecting the total surface area of the villus. Increases in peptidase activities occurring during postnatal development can be further subdivided into those dependent upon induced changes in enterocyte biochemistry and those dependent upon the changing geometry of villi. Over 90% of the total increase in peptidase activities occurring during neonatal development was shown, by this analysis, to involve enterocyte reprogramming of enzyme production. The present method of whole tissue cytochemistry appears to combine successfully the ability to measure peptidase activities at the cellular level in undisrupted tissue with a facility to relate these results to the overall shape of a single villus. These and more general applications of the method now provide new ways to analyse a variety of changes taking place in intestinal structure and function in a quantitative manner.

Regulation of mevalonate 5-pyrophosphate decarboxylase in isolated cells from chick intestinal epithelium

The present studies were undertaken to determine whether mevalonate 5-pyrophosphate decarboxylase (EC 4.1.1.33) is subject to physiological regulation in the intestinal mucosa. Activity was determined in epithelial cells isolated in a villus-to-crypt gradient from chicks fed on different diets in order to vary the sterol flux across the intestinal epithelium. When animals were fed on cholesterol, decarboxylase activity was decreased in all the cell fractions studied, although percentages of inhibition were maximum in crypts of jejunum and ileum. In contrast, decreased sterol flux as a consequence of cholestyramine feeding stimulated decarboxylase activity, especially in villi of the duodenum, where values increased 3-fold with respect to controls. On the other hand, the total cellular sterol content was significantly increased by the cholesterol diet. In duodenum and jejunum, 20-30% of the total cholesterol was in the esterified form under these conditions. However, dietary cholestyramine did not significantly affect amounts of total cellular cholesterol in any of the cell fractions. These results demonstrate that mevalonate 5-pyrophosphate decarboxylase activity changes considerably under different dietary situations and that the existence of secondary sites in the physiological regulation of sterol synthesis in the intestinal mucosa should be considered.

Developmental regulation of cryptdin, a corticostatin/defensin precursor mRNA in mouse small intestinal crypt epithelium

Cryptdin mRNA codes for the apparent precursor to a corticostatin/defensin-related peptide that accumulates to high levels in mouse intestinal crypt epithelium during postnatal development. The primary structure, intestinal cell distribution, and developmental appearance of cryptdin mRNA have been determined. Cryptdin mRNA is 450-480 nucleotides long. Translation of the partial cryptdin cDNA sequence reveals a 70-amino acid open reading frame that includes 32 carboxy-terminal residues that align with those in the consensus sequence, C.CR...C....ER..G.C....CCR, which is a common feature of leukocyte defensins and lung corticostatins (Selsted, M. E., D. M. Brown, R. J. DeLange, S. S. L. Harwig, and R. I. Lehrer. 1985. J. Biol. Chem. 260:4579-4584; Zhu, Q., J. Hu, S. Mulay, F. Esch, S. Shimasaki, and S. Solomon. 1988. Proc. Natl. Acad. Sci. USA. 85:592-596). In situ hybridization of cryptdin cDNA to paraformaldehyde-fixed, frozen sections of adult jejunum and ileum showed intense and specific labeling of epithelial cells in the base of all crypts. Analysis of sections from suckling mice showed that cryptdin mRNA is detectable in 10-20% of crypts in 10-d-old mice, in approximately 80% of crypts in 16-d-old mice, and in all crypts of mice 20 d and older. During the fourth week, the sequence accumulates in crypts to the maximal adult level. Cryptdin mRNA content in adult small intestine is independent both of T cell involvement and luminal bacteria. The role of cryptdin in small bowel physiology remains to be determined: cryptdin may inhibit bacterial translocation, modulate intestinal hormone synthesis, influence hormonal sensitivity of the intestinal epithelium, or exhibit a multiplicity of related activities.

Distribution of 3-hydroxy-3-methylglutaryl-CoA reductase in isolated villus and crypt cells of chick duodenum, jejunum and ileum

3-Hydroxy-3-methylglutaryl-CoA reductase (EC 1.1.1.34), the major rate-limiting enzyme of cholesterogenesis, was studied in epithelial cells isolated in a villus to crypt gradient from chick duodenum, jejunum and ileum, in order to resolve the apparent controversy that exists on the anatomical localization of sterol synthesis in the intestine. Consistent separation was demonstrated by using the marker enzymes alkaline phosphatase, specific to the villus cells, and thymidine kinase, specific to the crypt cells. No relative difference in stability was observed, as shown by the equal distribution of acid phosphatase. Cells were 90-95 per cent viable. The highest specific activity of reductase was located in the microsomal fraction (41 per cent of the total). The mitochondria had lower specific activity (8 per cent of the total). The distribution of reductase activity in epithelial cells of the villus-crypt axis was also studied. The specific activity in each cell fraction from chick duodenum was clearly lower than that in jejunum and ileum. The jejunal and ileal crypt regions showed lower specific activity than the villus cells. About 70 per cent of total reductase activity was found in cells from the upper and the mid villus fraction in each intestinal segment.

Effect of hydrocortisone on sialyltransferase activity in the rat small intestine during maturation. Changes along the villus-crypt axis and in fetal organ culture

Sialyltransferase activity was assayed in rat intestinal cells isolated as fractions reflecting the villus-crypt axis of differentiation. In 13-day-old rats both endo- and exogenous sialyltransferase activity reached their maximum in undifferentiated crypt cells and their peaks overlapped. In contrast, sialyltransferase of the adult intestine was 4-fold lower than that of sucklings in the crypts, with slight tendency to be transferred to the villus cells. Hydrocortisone applied to 10-day-old rats caused three days later a precocious drop of sialyltransferase activity in the crypt cells. Unlike in vivo, glucocorticoid responsiveness was accompanied by increased sialyltransferase activity in fetal small intestine cultivated for 17 days.

Stability of enterocytes and certain enzymatic activities in suspensions of cells from the villous tip to the crypt of Lieberkühn of the mouse small intestine

A series of studies in this laboratory have focused on how an indigenous microbiota influences the activities of alkaline phosphatase, phosphodiesterase I, and thymidine kinase in the enterocytes of the upper small intestine of mice. To draw conclusions about the role of the microflora in determining levels of enzymatic activity, we found it necessary to develop a procedure by which cell suspensions could be obtained containing enterocytes isolated sequentially from the villous tip to the crypt of Lieberkühn. The procedure was modified from the one developed for rats by Weiser (J. Biol. Chem. 248:2536-2541, 1973), involved a minimum number of interfering factors (e.g., proteolytic enzymes and mechanical agitation), and worked reproducibly for mice. During development of the procedure, some variables affecting the assays of the enzymes known to be present in enterocytes were also explored. Rods to which were tied everted segments of gut were incubated in a series of tubes containing a solution of EDTA the concentration of which was changed from 1.5 to 5.0 mM, thus giving a greater yield of enterocytes at every step. The cells incubating in the chelating solution were most stable when 0.23 M sucrose was included in the EDTA solutions. Success in assaying enzymatic activities in the cell suspensions depended on (i) how the cells were isolated, (ii) the assay procedure for thymidine kinase, and (iii) whether cellular suspensions or extracts were assayed.

Intestinal development in the suckling rat: effect of insulin on the maturation of villus and crypt cell functions

The influence of insulin on the postnatal development of intestinal functions linked to villus cells (sucrase, lactase, maltase and aminopeptidase) and crypt cells (secretory component of immunoglobulins, SC) has been studied in suckling and weanling rats. At 9 days of age, the animals received a daily injection of insulin 12.5 mU g-1 body weight day-1 for 4 days. Compared with saline-treated controls, insulin had no effect on the development of the intestinal mucosal mass parameters determined in the jejunum, ileum and colon. A premature appearance of sucrase was noted in isolated jejunal villus and crypt cells, the level of activity reached by the enzyme being dependent of the amount of insulin injected. By 6 and 12 h after a single injection of the hormone (12.5 mU g-1 body weight), sucrase activity was detected in all the cell fractions along the villus-crypt axis. In villus cells of insulin-treated rats, maltase, lactase and aminopeptidase activities were significantly (P less than 0.001) increased (+201%, +50%, +207%, respectively, vs. controls), whereas the concentration of SC measured by a sensitive immunoradiometric assay was enhanced over the controls by 75% in villus cells, 83% in crypt cells and 172% in the liver. Weanling rats treated from day 10 to day 20 postpartum with 12.5 mU insulin also exhibited a higher intestinal production of SC (+93%, P less than 0.01) than did saline controls.(ABSTRACT TRUNCATED AT 250 WORDS)

Immunoglobulin A receptor of rat small intestinal enterocytes is unaffected by aging

The receptor for polymeric immunoglobulins is responsible for the transport of immunoglobulin A (IgA) through epithelial cells and its subsequent delivery to mucosal surfaces. We have extended our previous studies of the IgA receptor in the liver of the aging Fischer rat to include the small intestine. Basolateral membrane-enriched fractions prepared from rat small intestinal enterocytes exhibit a single binding site for dimeric IgA. This receptor is specific for molecules that interact with rat secretory component, e.g., rat dimeric IgA and IgM and human polymeric IgA but not human monomeric IgA or rat secretory IgA. Inhibition of binding by rabbit-antirat secretory component also indicated that binding is specific for secretory component. Both liver and intestinal membranes showed virtually identical binding specificity. Membranes from crypt cells show increased IgA binding (320 fmol bound per milligram protein) compared with villous cells (105 fmol bound per milligram protein); however, other than increased binding, crypt cells show the same binding characteristics as villous cells. In contrast to our previous findings, in which liver plasma membranes from old rats showed a four-fold decrease in IgA binding compared with young adult rats, membrane fractions from rat enterocytes showed no alterations in dimeric IgA binding with increased age.

Selective effects of PHA on rat brush border hydrolases along the crypt-villus axis

The mechanism of the toxicity of lectin from Phaseolus vulgaris seeds has been investigated on rat enterocytes. Cell isolation procedures showed a selectivity in the loss of brush border hydrolases; this indicated that the microvilli blebbing was not the only mechanism of action of lectins on rat enterocytes.

Mevalonate 5-pyrophosphate decarboxylase in isolated villus and crypt cells of chick intestine

Mevalonate 5-pyrophosphate decarboxylase was studied in isolated enterocytes obtained from duodenal, jejunal and ileal villi and crypts. In our assay conditions, decarboxylase activity was linear for 60 min and up to 0.3 mg of protein. The subcellular location of decarboxylase in chick enterocytes was investigated. About 94% of the total activity was recovered in the cytosol. The distribution of enzyme activity in epithelial cells also was studied. Maximal specific activity was found in cell fractions from jejunum followed by ileum and duodenum. About 80% of total activity was recovered in the villus cells, indicating an active role of these cells in cholesterogenesis. Ileal cells showed the highest cholesterol content. In all the intestinal epithelial cells assayed, free cholesterol represented about 95% of the total cholesterol.

Accumulation of abundant messenger ribonucleic acids during postnatal development of mouse small intestine

To describe the differentiation of the small bowel at the molecular level, intestinal messenger ribonucleic acids (mRNAs) from mice at different stages of fetal and postnatal development were investigated. On the basis of cell-free translation and complementary deoxyribonucleic acid cloning experiments, abundant mRNAs coding for small polypeptides of 6-12 kilodaltons (low-molecular-weight mRNAs) were found in adult small intestine but not in the fetal gut. These developmentally regulated low-molecular-weight mRNAs are uniquely abundant in jejunum and ileum of adult mice, but they are absent or occur only at low levels in the duodenum, colon, stomach, and all other mouse organs examined. Low-molecular-weight mRNAs begin accumulating in the small bowel at approximately 3 wk of age, coinciding with weaning and with profound changes in intestinal differentiation. One complementary deoxyribonucleic acid clone of a low-molecular-weight mRNA (asb4/134) is specific to the distal small bowel, specifically accumulates at weaning, and hybridizes to RNA from mouse testis and brain at approximately 2%-5% of the intestinal level. Low-molecular-weight mRNA sequences may provide important markers of intestinal differentiation at the genetic level, leading to a better understanding of the factors that contribute to its postnatal maturation.

Short-term effect of a high-protein/low-carbohydrate diet on aminopeptidase in adult rat jejunoileum. Site of aminopeptidase response

The short-term effects of high-protein/low-carbohydrate diet on aminopeptidase N activity were studied in the brush-border membranes of proximal jejunum and proximal ileum of adult rats. The animals were starved overnight and re-fed for 15 h either with a standard diet (20% protein, 55% carbohydrate, in terms of energy content) or with a high-protein/low-carbohydrate diet of equal energy content (70% protein, 5% carbohydrate). All rats consumed similar amounts of diet, and measurements were made 15 h after initiation of re-feeding. In the proximal jejunum a slight increase in aminopeptidase activity was observed after the high-protein intake. In contrast, considerable stimulation (52%) of the enzyme specific activity was obtained in the proximal ileum. This increase in ileal aminopeptidase activity was more prominent in the mature cells of the upper villus. To determine if the increase of aminopeptidase activity was due to an increased amount of enzyme protein, rocket immunoelectrophoresis was performed with detergent-solubilized brush-border protein from ileum on agarose gels containing anti-(rat brush-border) antiserum. When the same amount of enzyme activity was loaded on the gels, the peaks of immunoprecipitate for aminopeptidase were similar for animals fed on a standard or a high-protein diet. When the same amount of protein was loaded, the peak of immunoprecipitate for aminopeptidase was higher (81%) after a high-protein diet. These results showed that the high protein intake evoked an increase in aminopeptidase activity, with a concomitant increase in the amount of immunoreactive protein.

Calmodulin in epithelial intestinal cells during rat development

Calmodulin was immunocytochemically localized in the brush borders of rat intestinal epithelial cells from the tip to the base of the villi, from day 18 of fetal life up to the adult stage. The early (14th day) fetal cells, like the adult crypt cells, were not immunoreactive, although their calmodulin content was equal to that of the mature cells from the tips of the villi.

Can villin be used to identify malignant and undifferentiated normal digestive epithelial cells?

We have investigated the presence of villin (a Ca2+-regulated actin binding protein) in various tissues (normal or malignant) and in established cell lines by using sensitive immunochemical techniques on cell extracts and immunofluorescence analysis on frozen sections. Our results show that villin is a marker that can be used to distinguish normal differentiated epithelial cells from the simple epithelia lining the gastrointestinal tract and renal tubules. Villin is found in the absorptive cells of the small and large intestines, in the duct cells of pancreas and biliary system, and in the cells of kidney proximal tubules. Furthermore, undifferentiated normal and tumoral cells of intestinal origin in vivo and in cell culture express villin. Therefore, expression of villin is seen in cells that do not necessarily display the morphological features characteristic of their terminally differentiated state, such as the microvilli-lined brush border. We suggest the possible clinical implications of using villin as a marker in the diagnosis of metastatic adenocarcinomas.

Pre- and postnatal development of differentiated functions in rat intestinal epithelial cells

A panel of monoclonal antibodies to intestinal cell surface components has been used to compare the expression of differentiation-specific antigens in the epithelial cells of fetal, suckling, and adult rat small intestine. Indirect immunofluorescence staining, and immunopurification of detergent-solubilized membrane proteins, followed by single- and two-dimensional slab gel electrophoretic analysis, have demonstrated that fetal intestinal cells (at day 21 of gestation) express most differentiation-specific markers typical of adult absorptive villus cells. A marked heterogeneity in antigen expression was observed among different villus cell populations in suckling rat intestine, and three cell surface components were identified which are exclusively present during this period of intestinal development. Striking changes in the patterns of antigen expression in crypt and villus cells, and variations in the apparent isoelectric points for most luminal membrane components, were associated with the maturation of the intestinal mucosa at weaning. These changes could not be prematurely induced by cortisone injection in newborn rats, suggesting that factors other than glucocorticoids are responsible for the postnatal development of the intestinal epithelium. These results suggest that basic differences in biological properties and regulatory mechanisms exist among intestinal epithelial cells at different stages of pre- and postnatal maturation.

Study of intestinal cell differentiation with monoclonal antibodies to intestinal cell surface components

Monoclonal antibodies that react with antigens of the plasma membrane of rat intestinal villus and crypt cells have been prepared by fusion of mouse myeloma (NSI) cells with spleen cells of mice immunized with various intestinal cellular fractions, including the luminal membrane of adult villus and crypt cells, and of newborn rat intestinal cells. The antigenic targets of most antibodies have been identified. They include major protein components of the brush border (luminal) membrane of adult villus cells (sucrase-isomaltase, maltase, lactase, aminopeptidase N, alkaline phosphatase) and newly identified protein antigens specific for intestinal epithelial cells. Of 25 independently derived monoclonal antibodies prepared, 18 reacted exclusively with the brush border membrane of the villus cells, confirming its unique protein composition. Antibodies specifically staining the crypt cells, the newly differentiated epithelial cells present in the lower half of the villi, the top villus cells, and both villus and crypt cells were also obtained and characterized. These antibodies have been used to study the expression of cell- and tissue-specific functions during differentiation and development of the intestinal epithelium. Contrary to results obtained with polyclonal antisera, no inactive forms of the brush border enzymes have been detected in the crypt cells. The identification of cell surface components expressed at different levels of the villi, and in both undifferentiated and differentiated intestinal cells, suggests that cell differentiation in the intestinal epithelium is a continuous and gradual process involving both transcriptional and translational regulation of different sets of genes.

Changes in brush-border enzyme activities of intestinal epithelial cells isolated from the villus-crypt axis during the early phase of alloxan diabetes in rats

The sucrase activity in enterocytes isolated from the villus crypt axis was found to increase in all regions of the villus from day 2 after induction of diabetes, and the increase continued until day 4. In contrast, alkaline phosphatase activity increased mainly in the apical one-third of the villus-crypt column, and the increase occurred abruptly on day 4 with increase in food intake.

A new variant of serum leucine aminopeptidase in the mouse: its development and possible regulation

Serum leucine aminopeptidase (LAP) isozymes were compared in four strains of inbred mice during postnatal development, adult life, and pregnancy. In pregnancy, no changes in the maternal serum LAP pattern were observed, in contrast to human studies. One strain, DD/S, differs from the other three in serum LAP. Polymorphism in serum LAP has not been previously described in the mouse. Neonatal DD/S mice exhibit a single band of serum LAP upon starch gel electrophoresis; however, between 14 and 18 days of age, two distinct bands appear, which persist throughout adult life. In the strains C57BL/6J, BALB/cJ, and DBA/2J there is a single band of activity at all stages. Crosses and backcrosses between DD/S and C57BL/6J show that the double-band variant is inherited as an autosomal recessive. The variant is independent of both the supernatant malic enzyme (Mod-1) and the intestinal LAP (Lap-1) loci, which are known to be linked on chromosome 9. The serum LAP variant is linked to an intestinal alkaline phosphatase variant. The presence of a separate structural gene is suggested by the genetic independence of the serum LAP variant from Lap-1. Also, the two serum LAP bands of DD/S are not interconverted by treatment with neuraminidase, beta-mercaptoethanol, or heat or by mixing the sera of DD/S and C57BL/6J prior to electrophoresis. The level of serum LAP activity in DD/S is approximately twice that in C57BL/6J. While these observations imply two structurally distinct proteins, the absence of any trace of the second LAP band in the heterozygote strongly suggests that the LAP variant protein is not the result of a separate structural gene. Intestinal LAP in DD/S migrates with the same electrophoretic mobility as the serum LAP variant, implying that the variant might originate in the intestine and its appearance in the serum be modulated by some factor at an unlinked locus.

Biosynthesis of intestinal microvillar proteins. Expression of aminopeptidase N along the crypt-villus axis

The expression of pig small-intestinal aminopeptidase N (EC 3.4.11.2) along the crypt-villus axis was studied in tangential sections of [35S]-methionine-labelled, organ-cultured explants. The only detectable molecular forms of aminopeptidase N along the crypt-villus axis were polypeptides of Mr 140 000 and 166 000, representing the enzyme in a transient and mature form respectively. The synthesis was at a very low level in the crypt region in experiments with labelling periods ranging from 10 min to 3 h. These findings indicate that crypt cells are not fully committed to the expression of aminopeptidase N, either in its mature or in any other immunoreactive molecular form. The expression of aminopeptidase N was markedly stimulated by dexamethasone (1 microgram/ml). During labelling periods of 3 h, dexamethasone caused an approximately threefold increase in the expression of the enzyme in the crypt cells and a moderate increase of about 20% in the villus cells. Whereas the latter can possibly be ascribed to a general protective effect of dexamethasone on villus architecture, these experiments indicate that crypt cells of mucosa from adult individuals exhibit the same sensitivity to glucocorticoids as does the intestinal epithelium during the prenatal and early postnatal phase.

Localization of biotransformational enzymes along the crypt-villus axis of the rat intestine. Evaluation of two cell isolation procedures

Rat intestinal epithelial cells were isolated by EDTA-chelation, combined with gentle shaking (modified Weiser procedure) or with strong longitudinal vibration (Harrison/Webster procedure). Both methods yield large numbers of viable cells and are relatively easy to use. Electronmicroscopical and biochemical data indicate that cell fractions from different levels of the villous region can be obtained only by the modified Weiser procedure. When strong mechanical forces are involved (Harrison-Webster procedure) the villus epithelium is released according to an all-or-nothing process. The biotransformational capacity of cell fractions, obtained from different levels of the villi by the modified Weiser procedure, was investigated. It was shown that the rate of metabolism of 7-ethoxycoumarin and 1-naphthol was substantially higher in lower villous cells than in cells isolated from the upper villous region. O-Deethylation of 7-ethoxycoumarin decreases from 145 +/- 13 pmole/min mg cell protein (72 +/- 4% conjugated) in lower villous cells to 62 +/- 12 pmole/min mg cell protein (37 +/- 6% conjugated) in tip cells. Glucuronidation of 1-naphthol decreased from 495 +/- 23 pmole/min mg cell protein (lower villous cells) to 137 +/- 13 pmole/min mg cell protein (tip cells).

Distribution and biosynthesis of aminopeptidase N and dipeptidyl aminopeptidase IV in rat small intestine

The regional, cellular and subcellular distribution patterns of aminopeptidase N and dipeptidyl aminopeptidase IV were examined in rat small intestine. Aminopeptidase N of brush border membrane had maximal activity in the upper and middle intestine, while dipeptidyl aminopeptidase IV had a more uniform distribution profile with relatively high activity in the ileum. Along the villus and crypt cell gradient, the activity of both enzymes was maximally expressed in the mid-villus cells. However there was substantial dipeptidyl aminopeptidase IV activity in the crypt cells. Both enzymes were primarily associated with brush border membranes in all segments, however, in the proximal intestine, a significant amount of dipeptidyl aminopeptidase IV activity was associated with the cytosol fraction. The cytosol and brush border membrane forms of dipeptidyl aminopeptidase IV were immunologically identical and had the same electrophoretic mobility on disc gels. In contrast, the soluble and brush border membrane-bound forms of aminopeptidase N were immunologically distinct. When the total amount of aminopeptidase N and dipeptidyl aminopeptidase IV was determined by competitive radioimmunoassay, there were no regional or cellular differences in specific activity (enzyme activity/mg of enzyme protein) of either enzyme in brush border membrane and homogenate. The specific activity of both enzymes in a purified Golgi membrane fraction as measured by radioimmunoassay was about half that of the brush border membrane fraction. These results suggest that (1) aminopeptidase N and dipeptidyl aminopeptidase IV have different regional, cellular and subcellular distribution patterns; (2) there are enzymatically inactive forms of both enzymes present in a constant proportion to active molecules and that (3) a two-fold activation of precursor enzyme forms occurs during transfer from the Golgi membranes to the brush border membranes.

Cytochemical analysis of alkaline phosphatase and esterase activities and of lectin-binding and anionic sites in rat and mouse Peyer's patch M cells

M cells in Peyer's patch follicle epithelium endocytose and transport luminal materials to intraepithelial lymphocytes. We examined (1) enzymatic characteristics of the epithelium covering mouse and rat Peyer's patches by using cytochemical techniques, (2) distribution of lectin-binding sites by peroxidase-labeled lectins, and (3) anionic site distribution by using cationized ferritin to develop a profile of M cell surface properties. Alkaline phosphatase activity resulted in deposits of dense reaction product over follicle surfaces but was markedly reduced over M cells, unlike esterase which formed equivalent or greater product over M cells. Concanavalin A, ricinus communis agglutinin, wheat germ agglutinin and peanut agglutinin reacted equally with M cells and with surrounding enterocytes over follicle surfaces. Cationized ferritin distributed in a random fashion along microvillus membranes of both M cells and enterocytes, indicating equivalent anionic site distribution. Staining for alkaline phosphatase activity provides a new approach for distinguishing M cells from enterocytes at the light microscopic level. Identical binding of lectins indicates that M cells and enterocytes share common glycoconjugates even though molecular groupings may differ. Lectin binding and anionic charge similarities of M cells and enterocytes may facilitate antigen sampling by M cells of particles and compounds that adhere to intestinal surfaces in non-Peyer's patch areas.

Establishment and characterization of an epithelial intestinal cell line from rat fetus

An epithelial intestinal cell line has been established from explants of fetal rat small intestine. After the 9th passage (approx. 25 population doublings) epithelial-like cells acquired the properties of a permanent cell line. The epithelial nature of this cell line, and of clone IRD 98 subsequently isolated, is supported by morphological and ultrastructural criteria, and also by the presence of enzymes characteristic of enterocytes, such as aminopeptidase, alkaline phosphatase, gamma-glutamyl transferase, lactase and maltase. The occurrence of the triglyceride pathway enzyme monoacylglycerol acyltransferase and of apoproteins (Apo A1 and Apo E) can also be demonstrated. Taken together, the results presented here provide evidence that clone IRD 98 is an epithelial cell line, most likely originating from the relatively differentiated cell layer of fetal rat small intestine.