Search

To define a potential role for the angiotensin system in Crohn's colitis, the colonic mucosal levels of angiotensin I and II were measured in endoscopic biopsy samples from patients with active Crohn's colitis (n = 20), ulcerative colitis (n = 13), other forms of colitis (n = 3), and normal controls (n = 17). Colonic mucosal levels of angiotensin I and II were greater in patients with Crohn's colitis than in normal subjects (p less than 0.001 and p less than 0.001, respectively). Mucosal levels of angiotensin I and II were also higher in Crohn's colitis than in ulcerative colitis (p less than 0.001 and p less than 0.001, respectively), and levels of angiotensin II were higher in Crohn's than in other forms of colitis (p = 0.014). Mucosal levels of angiotensin I and II correlated well with the degree of macroscopic inflammation in Crohn's colitis (r = 0.86, p less than 0.001 and r = 0.68, p less than 0.001, respectively). Mucosal levels of angiotensin I correlated fairly well with the Crohn's Disease Activity Index (r = 0.46, p less than 0.05) while angiotensin II levels correlated poorly. These studies suggest that angiotensin I and II may have a role in the inflammation associated with Crohn's colitis.

Cell kinetic activity and adaptive response of rectal mucosa from patients with Hartmann's procedure were studied before and after restoration of colorectal continuity. Patients without colostomy and with normal rectal mucosa were used as controls. Autoradiography of in vitro labeled mucosal samples with [3H]thymidine was used. The proliferative activity in the rectal crypts was estimated by measuring labeling and mitotic indices, total DNA of isolated crypts, and total crypt cell numbers. One hundred forty days after creating a proximal end colostomy, labeling index (P less than 0.05), mitotic index (P less than 0.01), DNA content per crypt (P less than 0.05), and number of cells per crypt (P less than 0.05) decreased significantly compared to control values. Restoration of colorectal continuity resulted in a significant increase of the labeling index (P less than 0.05), the mitotic index (P less than 0.01), the DNA content per crypt (P less than 0.05), and the cell number per crypt (P less than 0.05). There were no significant differences between the postclosure values and the controls. These data indicated that excluding the human rectal mucosa from fecal stream determined a mucosal atrophy that could be reversed by restoration of colorectal continuity.

The presence of receptors for insulinlike growth factor I and II in human colonic epithelium is demonstrated. Scatchard analysis of binding data obtained with 125I-insulinlike growth factor I showed an insulinlike growth factor I receptor with a dissociation constant of 8.6 nM and a binding capacity of 0.8 pmol/mg membrane protein. Distinct insulinlike growth factor II receptors labeled with 125I-insulinlike growth factor II were also found with a dissociation constant of 6.9 nM and a binding capacity of 4.7 pmol/mg membrane protein. Two sets of observations make it possible to discriminate between the two types of insulinlike growth factor receptors. (a) Unlabeled insulinlike growth factor I was 3 times more potent than insulinlike growth factor II in inhibiting [125I] insulinlike growth factor I binding. Conversely, unlabeled insulinlike growth factor II was 10 times more potent than insulinlike growth factor I in competing with 125I-insulinlike growth factor II for binding to membranes. Insulin and proinsulin did not compete with either of the tracers. (b) Affinity labeling of membranes followed by sodium dodecylsulfate polyacrylamide gel electrophoresis under reducing conditions, revealed a radio-ligand-receptor complex of molecular weight 130,000 and 250,000 using 125I-insulinlike growth factor I and 125I-insulinlike growth factor II, respectively. These observations indicate that adult human colonic epithelium is abundantly equipped with two sets of receptors that recognize preferentially either insulinlike growth factor I or insulinlike growth factor II.

We report for the first time the concentrations of free amino acids in human intestinal biopsies obtained by routinely performed endoscopy. We studied 15 medical patients with no changes of the mucosa and six HIV-infected persons with duodenitis. The mean (and SD) sum of all amino acids, taurine excepted, was 61.9 (5.4) mmol/kg dry weight in duodenal biopsies of HIV-negative subjects (n = 11) and 82.9 (0.6) mmol/kg in colonic specimens: 50% (44%) of the total (minus taurine) consisted of aspartate and glutamate and 14% (12%), of the essential amino acids. The relative amino acid pattern in duodenum and colon differed completely from that for muscle: aspartate was fourfold higher; glutamate, phenylalanine, glycine, valine, leucine, and isoleucine were about twofold higher. In contrast, glutamine amounted only to 4% (duodenum) to 14% (colon) of muscle glutamine. In duodenal biopsies of the HIV-infected persons, we found significantly (P less than 0.01, except glutamine: P less than 0.025) increased concentrations of glutamate (24.1 vs 17 mmol/kg dry weight), ornithine (1.4 vs 0.4), valine (2.2 vs 1.7), and glutamine.

As part of a study of ulcer formation and healing, regeneration of colonic mucosa in rats was studied following placement of a surgical lesion. Alterations in mucosubstances and connective tissue were examined and their possible significance discussed. The sequence of events in healing was: 1) The mucosa adjacent to the lesion tipped into the lesioned area. The crypts in this mucosa became lined with cells which contained no mucus and had no striated borders. Later in the experimental period, these undifferentiated cells gave rise to cells containing carboxymucins. Cells containing sulfomucin, neutral mucin, or having striated borders arose from the carboxymucin cells. 2) An epithelial ledge of undifferentiated cells migrated onto a sulfated glycosaminoglycan, fibrous interface between necrotic and living tissue in the lesion. 3) Crypt formation began with the appearance of intraepithelial anlagen. 4) Crypts lengthened by a process of epithelial-connective tissue proliferation from the base of the crypt upwards. Following completion of connective tissue regeneration, crypts formed by invading the reestablished lamina propria. 5) The first mucous cells in the ledge contained carboxymucins. As crypt formation occurred, these cells gave rise to typical columnar absorptive cells, to cells containing sulfomucins, and to cells containing neutral mucins. 6) Lengthening of crypts ceased following the appearance of a sulfated acid glycosaminoglycan--collagenous layer deep in the submucosa.

Normal colon mucosa was found to produce carcinoembryonic antigen (CEA) quite actively as cancerous tissues do when maintained in an organ culture, although the fresh normal mucosae contained a very small quantity of CEA unlike cancerous tissues. This is consistent with an active expression of CEA mRNA in normal mucosa comparable to that in cancerous tissues actively producing CEA, and suggests that the normal cell product was rapidly released into the lumen of digestive tract and turned into normal fecal antigens (NFAs) previously found in feces. 3H-Labeled precursors of glycophospholipid such as ethanolamine and stearic acid were incorporated into CEA produced by both normal and cancerous tissues, suggesting that CEA in normal mucosa is anchored to the cell membrane through a glycophospholipid as in cancerous tissues.

The application of a morphometric method to the quantification of peptide-containing nerves in the gut is described. It allows a simple estimation of the nerve fiber supply per unit volume of tissue (length density) and the calculation of the total nerve fiber supply per unit of intestinal mucosa (villus/crypt unit).

The effect of adding cetylpyridinium chloride to the fixative on the preservation of sulphated glycosaminoglycans (SGs) was studied in human normal colonic mucosa. SGs were visualized at the ultrastructural level through the application of Spicer's High Iron Diamine (HID) technique followed by a post-fixation with potassium ferrocyanide-reduced osmium tetroxide. SGs were mainly localized in basement membranes of epithelium and capillary wall and along collagen fibers. The morphology of the reactive sites depended on the presence of cetylpyridinium chloride, SGs being granular in absence of the salt and more or less elongated when cetylpyridinium chloride was added to the fixative. We suggest that the use of cetylpyridinium chloride during fixation may help to preserve SG molecule at the ultrastructural level.

The results of clinical studies have suggested that inadequate preoperative mechanical preparation of the colon is a significant factor in the pathogenesis of disruption of colonic and colorectal anastomoses. A possible explanation for the adverse effect of poor mechanical preparation is that ammonia released from feces has a cytotoxic effect on the healing anastomosis. It has been suggested that acidification of the colonic lumen prevents the release of fecal ammonia and protects the anastomoses. The effects of mechanical preparation of the colon and colonic acidification were studied in anastomoses in the left colon of rats. Colonic acidification was produced by the oral administration of a synthetic disaccharide lactulose solution. Mechanical preparation resulted in a significant increase in the bursting wall tension in anastomoses on the seventh postoperative day. The oral administration of lactulose resulted in a reduction in the colonic mucosal pH, but this was not accompanied by improved healing of anastomoses, as judged by measurements of colonic bursting pressure, bursting wall tension and total challagen content. Mechanical preparation of the colon is an important factor in the safety of colonic anastomosis. However, the results do not suggest that acidification of the colonic lumen significantly improves the quality of colonic healing.

Purified human colonic mucin was separated into six distinct components by DEAE-cellulose chromatography, and the structures of oligosaccharide side chains from the three most abundant species were determined. Oligosaccharide side chains were isolated from colonic mucin species III, IV, and V after alkaline borohydride reductive cleavage in the presence of sodium borotritide. After initial separation of acidic and neutral oligosaccharides by ion exchange chromatography, individual oligosaccharides were isolated by sequential chromatography on Bio-Gel P-4 and Bio-Gel P-2 resins followed by preparative normal phase high performance liquid chromatography. Composition and structure of individual oligosaccharides were determined by combination of gas chromatography, methylation analysis, and sequential glycosidase digestion. Collectively, 21 discrete oligosaccharide structures were identified in the major human colonic mucin species including 10 acidic oligosaccharides and 11 neutral structures which ranged in size from 2 to 12 sugar residues. Although detailed structures were defined for each oligosaccharide, the majority of the structures identified were variations of a relatively small number of "basic" structures, and several generalizations pertained. First, many oligosaccharides represented variations of a biantennary structure in which branch chains arise in N-acetylglucosaminyl residues linked to C3 and C6 of a galactosyl residue linked in turn to a GlcNAc beta (1-3)GalNAc core; second, non-branched oligosaccharides appeared to be linear chain derivatives of the same core structure; third, all acidic oligosaccharides could be derived from neutral structures present in the mucin species; fourth, sialic acid substitution was limited to few sites and always included substitution in alpha 2-6 linkage to the reducing terminal N-acetylgalactosamine, and finally several structures contained both sialic acid and fucose residues. Individually, mucin species III, IV, and V were found to contain unique mixtures of 13, 14, and 10 oligosaccharide structures, respectively. These data demonstrate that human colonic mucin contain a wide range of oligosaccharides reflecting variations of common core oligosaccharide structures. The major chromatographically defined constituents of normal colonic mucin appear to possess characteristic and distinguishable combinations of oligosaccharide structures. These findings support the concept that colonic mucin contains structurally and functionally distinct subpopulations.

Colonic mucin was prepared by phenol-water partition extraction of th colonic mucosa, followed by ethanol precipitation of the water-soluble material, molecular sieve chromatography, and reductive sodium dodecyl sulfate-disc gel electrophoresis in an agarose-polyacrylamide gradient gel. Although this material as observed to be relatively homogeneous by physical criteria, it was shown to contain at least two components by immunodiffusion. Molecular sieve, ion exchange, adsorption, lectin affinity, and electrophorectic techniques failed to separate the two components. However, by protease digestion employing either 1% w/w pronase or papain, followed by molecular sieve chromatography, we were able to separate at least the portions of the molecules containing the immunodeterminants. Physicochemical and immunologic properties of both components were characteristic of mucins. Mucin A (pronase digest fraction 1 from molecular sieve chromatography) was enriched in threonine, proline, and sialic acid, while mucin B (pronase digest fraction 3 from molecular sieve chromatography) was enriched in serine, alanine, and fucose. Mucin A and mucin b were shown to be immunologically distinct. The evidence suggests that the protease-digested mucin fractions originate from two different mucins, one a sialomucin, the other a fucomucin; however, the alternative explanation, that these two mucin fragments may originate from a single mucin molecule, cannot be ruled out at present.

The distribution of glycosubstances in colonic epithelium using lectins that were rhodaminated for fluorescent microscopy or coupled to colloidal gold for electron microscopy has been studied. Findings, based on lectin binding patterns, indicate that the right, transverse, and left colon of the guinea pig differ in their content of glycoconjugates within goblet cells and along the brush border. Local variations of labeling were also observed within goblet cells between the upper and lower portions of crypts. Throughout the colon, corresponding to a region of supranuclear fluorescence, Ricin II-colloidal gold labeled the Golgi complex of both enterocytes and goblet cells. Ricin II-colloidal gold also labeled small (congruent to 800 A) vesicles in the apical portion of enterocytes in all colonic segments. Microvilli were labeled by Ricin II-colloidal gold in the right and transverse colon, a finding that correlated with the observed adhesion of bacteria in those segments.

Ampicillin and sulbactam were administered intravenously in 2 g and 1 g doses respectively every 8 h for two days to ten patients undergoing colorectal surgery. Drug concentrations were determined in serum, intestinal mucosa and faeces. The first dose was given at induction of anesthesia. The serum half-life was 2.1 h for ampicillin and 2.0 h for sulbactam on the first day. The concentrations in intestinal mucosa were in most cases 50% or more of the serum concentrations and there was no statistical difference between the two drugs. Faecal concentrations of both drugs were found in eight patients: a mean of 10.5 micrograms/g for ampicillin and a mean of 7.5 micrograms/g for sulbactam. Faecal samples were also collected during the investigation period for culture of aerobic and anaerobic bacteria. Among the aerobic bacteria enterococci, streptococci and enterobacteria decreased markedly during the prophylaxis period. Anaerobic bacteria were also suppressed significantly during the same period. The changes in colonic microflora were related to the concentrations of ampicillin and sulbactam in the intestinal mucosa and faeces. Overgrowth of yeasts was observed in six patients and overgrowth of Pseudomonas fluorescens in two others. These strains were resistant to the combination ampicillin and sulbactam. The microflora normalized in all patients after two weeks. No postoperative infections occurred.

A glycoprotein was isolated from rat-colonic mucosa. Analytical ultracentrifugation studies showed the glycoprotein to be homogeneous, having an apparent molecular weight of 9.0 X 10(5); no subunits could be detected in the presence of sodium dodecyl sulfate. It contained 14% of protein and 86% of carbohydrate. The principal sugars in the glycoprotein were galactose, fucose, sialic acid, 2-acetamido-2-deoxygalactose, and 2-acetamido-2-deoxyglucose. A small proportion of mannose was also present. The glycoprotein, apart from the usual carbohydrate constituents present in mucus glycoproteins, contained sulfate, but no uronic acid. High amounts of serine and threonine, and low contents of aromatic and traces of sulfur-containing amino acids, reflect a similarity of this glycoprotein to other mammalian mucus glycoproteins; it differs, however, by its high proportions of Asx + Glx (26 mol.%). Cleavage studies with alkaline borohydride indicated O-glycosidic linkages between N-acetylhexosamine and serine, and threonine, of the peptide core in the glycoprotein. Only about one third of the serine and threonine was linked to the carbohydrate side-chains, which averaged about 22 units in length and were apparently branched.

Because it is a common prerequisite for steroid responsiveness in target tissue, we investigated the presence of specific 1,25-DR in spontaneous human colorectal adenocarcinomas (ADC) and adjacent normal-appearing mucosa (NAM) from 23 operative specimens (12 male and 11 female patients). 1,25-DR was determined in cytosol by a DCC assay technique. 1,25-DR was present in 21 of 23 NAM and in only 4 of 23 HCRA. All positive ADC were well differentiated. Receptor content expressed in femtomoles/mg of protein (mean +/- SEM) was respectively 63.9 +/- 7.6 for right colon NAM and 51.3 +/- 12.9 for left colon or rectum NAM. When we compared all NAM specimens, receptor content was 56.7 +/- 8.0 femtomoles/mg of protein. No difference in 1,25-DR NAM level was observed between right colon and left colon or rectum. In adenocarcinoma the mean content was 66.5 +/- 14 fmoles/mg of protein. Scatchard analysis showed a single class of specific high-affinity saturable 1,25-DR with a dissociation constant (Kd) of 0.97 +/- 0.57 and 1.03 +/- 0.39 chi 10(-10) M in NAM and ADC respectively. These preliminary data represent the first demonstration of 1,25-DR throughout the entire human colon and indicate that the receptivity for this hormone is often lost during malignant transformation of the human colorectal mucosa. In addition, 1,25-DR could be a marker of differentiation in ADC. These preliminary results provide evidence supporting the addition of Vitamin D to the roster of developmental cancer chemopreventative agents.

Epithelial endocrine cells containing 5-hydroxytryptamine, substance P, somatostatin, enteroglucagon and vasoactive intestinal polypeptide-immunoreactivity were localized by immunocytochemistry in the mucosa of normal appendices, ileum and proximal colon, and in neurogenic appendicopathy. In neurogenic appendicopathy a large number of proliferating nerves were visualized independently of neurotransmitters by immunostaining for neuron-specific enolase. A large number of nerve fibers were shown to contain substance P-immunoreactivity and to be of intrinsic origin. Stromal endocrine cells containing 5-hydroxytryptamine, somatostatin- and possibly substance P-immunoreactivity, were observed in substantial numbers in neurogenic appendicopathy. Substance P may be involved as a neurotransmitter and/or as a paracrine/endocrine peptide in the pathogenesis of spastic contractions and abnormal peristalsis of the appendix, which are characteristic of neurogenic appendicopathy. Stromal endocrine cells may be considered to be the origin of certain carcinoids in the appendix.

The histochemical characteristics of colonic epithelial mucins were investigated in the chick embryo. At the 14th day of incubation it was possible to demonstrate the presence of glycogen. At the 15th day a few epithelial cells showed the presence of neutral and sialylated mucins. On the 16th day, also sulfated secretory material was detectable together with neutral and sialylated mucins in cells with the typical shape of goblet cells. From the 17th day to the 20th day of incubation the two types of acid mucins appeared in some cells to be placed in distinct zones of the supranuclear cytoplasm. At the 21st day, neutral, sialylated and sulfated mucins were all present in the majority of goblet cells, which were found mainly in the epithelium lining the crypts.

The histochemistry of mucin in normal large intestine and in experimental colorectal cancers induced in Wistar rats with 1,2-dimethylhydrazine was analyzed by modifications of the concanavalin A-horseradish peroxidase method (paradoxical concanavalin A-staining) and high-iron diamine-Alcian blue (pH 2.5) staining (HID-AB). Quantitative analysis of each mucin reaction was done by the use of an image analyzer. Regional variations in the percentage area of "labile class III mucin" were 1 to 50% in the distal colon and 50 to 90% in the proximal colon. Regional variations in the percentage area of sulfated sialomucin were 80 approximately 100% in the distal colon and under 20% in the proximal colon. The percentage area of "labile class III mucin" in primary colorectal cancers showed region-specific variation and that in metastatic colorectal cancers also showed similar region-specific variation of colorectal mucosa where primary tumors were located. In contrast, there were no region-specific mucin reactions for sulfated sialomucin in almost all primary and metastatic colorectal cancers.

Murine mAbs were produced against purified microvillus membranes of rat colonocytes in order to establish a marker protein for this membrane. The majority of antibodies binding to the colonic microvillus membrane recognized a single protein with a mean apparent Mr of 120 kD in both proximal and distal colon samples. The antigen is membrane bound as probed by phase-partitioning studies using Triton X-114 and by the sodium carbonate extraction procedure and is extensively glycosylated as assessed by endoglycosidase F digestion. Localization studies in adult rats by light and electron microscopy revealed the microvillus membrane of surface colonocytes as the principal site of the immunoreaction. The antigen was not detectable in kidney or liver by immunoprecipitation but was present in the small intestine, where it was predominantly confined to the apical membrane of crypt cells and much less to the microvillus membrane of differentiated enterocytes. During fetal development, the antigen appears first in the colon at day 15 and 1-2 d later in the small intestine. In both segments, it initially covers the whole luminal surface but an adult-like localization pattern develops soon after birth. The antibodies were also used to develop a radiometric assay for the quantification of the antigen in subcellular fractions of colonocytes in order to assess the validity of a previously developed method for the purification of colonic brush-border membranes (Stieger, B., A. Marxer, and H.P. Hauri. 1986. J. Membr. Biol. 91:19-31.). The results suggest that we have identified a valuable marker glycoprotein for the colonic microvillus membrane, which in adult rats may also serve as a marker for early differentiation of enterocyte progenitor cells in small-intestinal crypt cells.

A method has been developed for the simultaneous isolation of basolateral plasma membrane vesicles from surface and crypt cells of rabbit distal colon epithelium by sequential use of differential sedimentation, isopycnic centrifugation and Ficoll 400 barrier centrifugation. The protein yield was high (total 0.81 mg/g mucosa) and surface and crypt cell-derived basolateral membrane fractions have been purified 34- and 9-fold with respect to the homogenate. The pattern of marker enzyme enrichments revealed only minor contamination by subcellular organelles. Latency of ouabain-sensitive (Na+,K+)-ATPase activity prior and after trypsin treatment of membranes indicated a vesicle configuration of sealed right side-out: sealed inside-out: leaky of approximately 2:1:1. The presence of sealed vesicles was also evident from the osmotic sensitivity of the D-[1-14C] mannitol equilibrium space determined with either fraction. Although considerably different in protein profile, surface and crypt basolateral membranes were similar in cholesterol to phospholipid molar ratio and membrane fluidity as determined by steady-state fluorescence polarization. Stopped-flow light scattering experiments revealed a rather low water permeability of the membranes with a permeability coefficient of 6 microns/sec at 35 degrees C, which is one order of magnitude lower than reported for small intestinal plasma membranes. Both membrane fractions have been shown to effectively generate outward uphill potassium ion gradients, a process that is energized by ATP and inhibited by the membrane-permeant cardiac-glycoside digitoxin. These characteristics are consistent with the activity of a (Na+,K+) pump operating in inside-out vesicles.

Brush-border membranes were isolated from the mucosal surface of rabbit proximal colon epithelial cells by a procedure involving Ca2+ precipitation. Ouabain-insensitive K+-phosphatase, a marker enzyme for the colon brush-border membrane, was enriched 17-fold by this technique, while no enrichment was observed in the activity of ouabain-sensitive K+-phosphatase, a marker for the basal-lateral membrane. Insulin binding studies revealed a dose-dependent inhibition of 125I-insulin binding with porcine insulin and approximately 4 X 10(-9) M insulin was required to produce 50% inhibition of 125I-insulin binding, while desoctapeptide insulin, insulin-like growth factor I, and A chain of insulin had less effect on 125I-insulin binding. This is the first demonstration of the existence of high-affinity insulin binding sites on the brush-border membrane of mammalian colon epithelial cells. Subsequent studies with the cross-linking agent disuccinimidyl suberate confirmed the presence of insulin binding sites in these membranes and autoradiography of polyacrylamide gels revealed that the binding subunit of the colon epithelial cell brush-border insulin receptor is similar in size to that observed in hepatic tissue. Interestingly, the insulin binding capacity/mg of protein of this preparation is high, suggesting that large numbers of insulin receptors are present in vivo on the mucosal surface of colon epithelial cells. The potential physiological role of these previously unrecognized insulin receptors is discussed.

Cells of the normal colonic mucosa express several types of cytokeratins, the characteristic intermediate filament proteins of epithelial cells. An immunohistochemical study was designed to examine the expression of two distinct groups of cytokeratins, recognized by monoclonal antibodies AE1 and AE3, in the colonic mucosa and to compare the findings with those obtained with a large number of polypoid lesions (adenomatous and hyperplastic) and carcinomas of the colon. AE1 and AE3 immunostaining was found in the surface epithelium and upper portions of the crypts of Lieberkühn (functional zone) of normal colonic mucosa, whereas the lower portions of the crypts (proliferative compartment) were unreactive with both AE1 and AE3. Polypoid lesions of the colonic mucosa can be placed into two categories based on their patterns of cytokeratin expression. Solitary tubular adenomas and hyperplastic polyps are composed of AE1 and AE3 nonexpressing cells with only few, patchy areas of AE1 and AE3 expressing cells present within glands and in the surface epithelium. In contrast, villous adenomas show strong AE1 and AE3 reactivity throughout the glands. Furthermore, tubular and villous adenomas, and even histologically normal mucosa in patients with familial polyposis, show AE1/AE3 expression throughout the glands and surface epithelium. Colonic carcinomas show a predominance of AE1/AE3 expressing cells. Thus, cytokeratins recognized by monoclonal antibodies AE1 and AE3 represent molecular markers of cellular maturation in the normal colonic mucosa, that are expressed in colonic carcinomas and, in addition, serve as markers that distinguish colonic mucosa and adenomas with a high risk for development of cancers from those with a lower risk.

Modal DNA (ploidy) and sensitivity of DNA in situ to denaturation by acid have been analyzed by flow cytometry of 10 colorectal adenomas and 35 adenocarcinomas; 39 normal mucosa samples served as controls. A new method was developed to denature DNA in chromatin of the freshly isolated, intact, and unfixed individual cell nuclei from surgically resected material. The sensitivity of DNA denaturation (T alpha) was assayed by metachromatic staining with acridine orange and calculated as a ratio of the alpha t index of the tumor sample to the alpha t index of normal mucosa; the alpha t index is that fraction of DNA, following treatment at pH 1.4, that stains metachromatically with acridine orange at pH 2.6. All adenomas were diploid and in nine of 10 the T alpha value was close to 1.00, indicating no difference from control specimens in DNA sensitivity to denaturation. Forty-nine% of adenocarcinomas were aneuploid. Forty-six% of adenocarcinomas differed from normal in sensitivity of DNA to denaturation; the T alpha value was lower than 0.90 indicating that chromatin of the tumor cells was more resistant to denaturation than control cells. There was no correlation between sensitivity to denaturation of DNA and incidence of aneuploidy. However, there was a correlation between T alpha and the pathologically determined stage of disease. There was increased resistance to denaturation in 58% of tumors classified as Dukes' C/D stage, in 36% of tumors classified as Dukes' B, and in 20% classified as Dukes' A stage of the disease. Statistical analysis of these results revealed significant differences between distributions of T alpha in noninvasive (Adenomas and Dukes' A) versus invasive (Dukes' B and C/D) tumors with level of significance at P = 0.02. The data suggest that acid denaturation of DNA in situ may be a valuable adjunct in assessing the biology of colon cancer. The molecular basis for this phenomenon is discussed.

The structure of carbohydrate units of rat colonic mucus glycoproteins was investigated. Alkaline borohydride reductive cleavage of the glycoprotein resulted in the release of a series of reduced oligosaccharides. Eight of the oligosaccharides (1 to 8), ranging in length from 2 to 12 sugar units, have been purified to homogeneity and their structures determined. Based on the results of immunological assays, analysis of sugar composition, sequential degradation with specific glycosidases, and methylation analyses of the intact and desialyzed oligosaccharides, we propose the following structures for these compounds: (Formula: See text).

Full thickness specimens of normal colon (n = 15), and colon from patients with diverticular disease (n = 5) were obtained at operation or autopsy. In the isolated submucosa the ultrastructure of the constituent collagen fibres was examined by transmission electron microscopy. Collagen fibrils in the left colon become smaller (p less than 0.001) and more tightly packed (p less than 0.001) than those in the right colon with increasing age. This difference is accentuated in diverticular disease (p less than 0.01). Factors which contribute to the development of colonic diverticulosis, such as raised intraluminal pressure, may be responsible for premature change in submucosal structure.

The present study was designed to examine the effect of dietary fat composition on the structure of colon mucosal plasma membranes and fecal lipids. Rats were fed a purified diet containing 14% of either highly saturated fat (beef fat or butterfat) or highly polyunsaturated oil (safflower) in addition to 2% corn oil for 4 wk. Colon mucosal membranes were prepared and examined for lipid composition and protein pattern. Saturated fatty acid feeding resulted in the loss of some protein bands from plasma membranes compared to feeding polyunsaturated fatty acids. Within the saturated fatty acid--rich fats, feeding beef fat caused a greater loss than did feeding butterfat. Dietary fat composition had no effect on membrane content of phospholipid and cholesterol. Saturated fatty acid feeding resulted in an increase in the percentage of 18:1 in plasma membrane lipids compared to feeding safflower oil. The observed changes in the structure of colon mucosal membrane of animals fed the saturated fats were associated with an increase in fecal free fatty acids. There was a 4-fold and 2-fold increase in fecal free fatty acids with feeding the beef fat and butterfat diets, respectively, compared to the safflower oil diet. Alterations in fecal bile acid and free fatty acid composition were also noticed with feeding saturated fatty acids. The results obtained suggest that feeding saturated fatty acids as the main source of fat in the diet could influence the structure of colon mucosa, and this could be mediated through fecal free fatty acids.

The purpose of this study was to examine whether formal-dehyde-fixed tissue may afford reproducible and reliable immunhistochemical results when carcinoembryonic antigen (CEA), secretory component (SC), and epithelial IgA are evaluated semiquantitatively in normal and pathological colon specimens. Proximate tissue samples were processed by routine formalin fixation and by a cold-ethanol fixation method, respectively, and the immunofluorescence intensities obtained for the three antigens were scored. After formalin fixation SC and epithelial IgA were generally undetectable and also the staining for CEA was markedly reduced compared with that seen after ethanol fixation. Significant antigenic "unmasking" was obtained by enzyme treatment of the formalin-fixed tissue sections--resulting in enhanced staining for SC and epithelial IgA but not consistently so for CEA. With this modification scores from duplicate tissue samples processed by the two methods showed significant correlations for all the three epithelial markers; small amounts of CEA and epithelial IgA, and especially SC, nevertheless remained undetectable after formalin fixation. This result should be taken into account when epithelial markers are applied in studies of premalignant lesions of the colon where minor changes in the antigen pattern may be of diagnostic importance.

Substance P content was determined by radioimmunoassay in colonic mucosa from 24 patients with chronic severe constipation, 16 with active ulcerative colitis, and 28 normal controls. In patients with chronic severe constipation, the mean concentration of substance P (19.9 +/- 8.2 pg/mg) was significantly lower than in normal subjects (71 +/- 18 pg/mg). In patients with ulcerative colitis, colonic substance P concentration in inflamed mucosa (170 +/- 46 pg/mg) was significantly higher than its levels in normal subjects. Substance P may therefore have a role in the pathogenesis of clinical conditions associated with diarrhea and constipation.

beta-Endorphin-like immunoreactivity was detected in the mucosa and muscle layer of normal colon, adenocarcinomas derived from the colon mucosa, and colon polyps which were histologically confirmed to be adenoma without a focus of carcinoma or with in situ carcinoma. The contents of beta-endorphin-like immunoreactivity in adenocarcinomatous tissue (11.94 +/- 1.77 pmol/g wet wt) and colon polyps without focus of carcinoma (10.71 +/- 1.50 pmol/g wet wt) were found to be significantly higher than those in the mucosal layer (6.86 +/- 0.64 pmol/g wet wt) and muscle layer (8.30 +/- 0.68 pmol/g wet wt) of normal colon. These data suggest that the production of beta-endorphin-like immunoreactivity is specifically increased in some adenocarcinomas and adenomatous polyps and may be related to the alteration of bowel habits. Gel exclusion chromatography of beta-endorphin-like immunoreactivity revealed three peaks corresponding to beta-endorphin, beta-lipotropin, and an immunoreactive form between the two. In the mucosal layer and muscle layer of the colon, a broad major peak was eluted at the position of beta-endorphin, and minor peaks were eluted at the position of beta-lipotropin and between beta-endorphin and beta-lipotropin. In adenocarcinoma and polyp, the peak size corresponding to authentic beta-lipotropin was greater than that of beta-endorphin. This study demonstrated that beta-endorphin-like immunoreactivity existed at a high concentration in some colon adenocarcinomas and polyps whose elution patterns were different from those of normal colon tissue.

Extracts of human colon mucosa and muscle were fractionated by gel filtration and ion exchange chromatography and molecular forms of vasoactive intestinal peptide (VIP) estimated by radioimmunoassay. Total concentrations of immunoreactive VIP in mucosa were 237.1 +/- 53.9 and 119 +/- 26.0 pmoles per g in muscle. In muscle, over 90% of the immunoreactivity was accounted for by a single component indistinguishable from porcine VIP by gel filtration and ion exchange chromatography. In contrast, mucosa contained four components with VIP-like immunoreactivity. One of these had the chromatographic properties of porcine VIP; the others were less positively charged. Two of the new VIP-like components had gel filtration properties similar to those of VIP and are, therefore, probably of similar size to the porcine octacosapeptide, and the remaining component emerged later on Sephadex and is, therefore, probably a smaller peptide. Immunoreactive VIP in muscle extracts is believed to originate from nerve fibers, and this form of VIP is likely to be identical to the previously characterized form of VIP. However, immunoreactive VIP in mucosal extracts may originate either from endocrine cells or nerve fibers. The possibility arises, then, that there are different immunoreactive forms of VIP in nerves and endocrine cells.

The effect of the addition of pectin to an elemental diet on the healing of experimental colonic anastomoses was investigated. Transection and anastomosis of the ascending colon and feeding gastrostomy were performed in 24 Sprague-Dawley rats. All rats then received an elemental diet, and 12 of them had 1% (w/v) citrus pectin added to their diet. On the seventh postoperative day, animals that received pectin-supplemented diets had significantly greater bursting pressures at the anastomoses (266 versus 234 mm Hg, p less than 0.04) and significantly lower colonic mucosal pH (6.2 versus 6.8, p less than 0.001) than animals that received the elemental diet only. The colons from animals fed pectin also had significantly higher hydroxyproline content at the anastomosis than those of the control animals (46.6 versus 40.7 micrograms hydroxyproline nitrogen/mg tissue nitrogen, p less than 0.05). The decreased intracolonic pH is consistent with the hypothesis that improved healing is a local effect mediated by the presence of short-chain fatty acids resulting from the fermentation of pectin.

The ganglioside and neutral glycosphingolipid composition of adult monkey small and large intestinal mucosa were characterized and compared. GM3, GM2 and GD1A were found to be the principal gangliosides in each of these tissues. Dihexosylceramide was the major neutral glycosphingolipid of both organs. The total content of gangliosides and neutral glycolipids/ceramide, however, was ca. four-fold and two-fold higher, respectively, in small intestinal than colonic mucosa. While all glycosphingolipids examined contained hydroxy and nonhydroxy fatty acids, the former fatty acids accounted for 60-90% of the total fatty acids in both organs. Sphingosine was the predominant long chain base of ceramide, mono-, di-, tri- and tetrahexosylceramide, whereas phytosphingosine was the major base of GM3 in both tissues. The results of these studies demonstrate that while many similarities of monkey small and large intestinal glycosphingolipids exist, qualitative and quantitative differences are present along the length of the monkey gut. These differences may be at least partially responsible for certain of the well-recognized variations in normal physiological and pathological processes that occur in these organs.

Immunoenzyme staining for tissue antigens using glucose oxidase as marker enzyme was compared with immunoperoxidase techniques. Both the indirect, conjugated antibody method and the unlabeled antibody procedure employing preformed complexes of glucose oxidase-antiglucose oxidase (GAG) and peroxidase-antiperoxidase (PAP) were used to stain carcinoembryonic antigen (CEA) in paraffin tissue sections. The localization of CEA within a tissue specimen was the same in all cases, and background staining was minimal. The percentage of positive specimens detected with GAG and PAP was similar but was slightly greater with glucose oxidase compared to peroxidase conjugates. The glucose oxidase conjugates and GAG were similar to the comparable immunoperoxidase reagents in enzyme-antibody molar ratio, retention of antibody and enzyme activity, immunohistochemical staining dilutions, and stability. Immunoglucose oxidase and immunoperoxidase were combined to localize CEA and colon-specific antigen-p simultaneously. Excellent contrast and staining separation were shown between the enzymatic reaction products of the two systems. Since immunoglucose oxidase methods are as sensitive as the comparable immunoperoxidase techniques, they should be considered as a reliable alternative to the latter, especially when endogenous peroxidase activity may be a problem. Furthermore, the two methods can be conveniently combined for the simultaneous detection of two antigens.

1. Mucus samples obtained from colectomy colons from 10 patients with ulcerative colitis and four patients with Crohn's disease of the colon, together with samples from 10 normal postmortem colons, have been studied. 2. Two glycoprotein fractions were isolated from mucosal scrapings by gel-permeation chromatography. The high-molecular-weight fraction consisted largely of mucus glycoproteins and was further purified to the glycopolypeptides. The low-molecular-weight fraction contained a glycoprotein with a high mannose content. The carbohydrate content of both the mucus glycoprotein and glycopolypeptide fractions were significantly reduced in active colitis. 3. Oligosaccharides were prepared from the mucus glycoprotein material. In the normal material more than half the units contained eight or more monosaccharide residues, whereas in the disease material the bulk of the units were smaller than this.

Cryostat sections of rat descending colon were studied by fluorescence microscopy after exposure to conjugates of fluorescein isothicoyanate with lectins from Glycine max (soybean), Triticum vulgaris (wheat germ), Ricinus communis (castor bean), Ulex europaeus, (gorse), Dolichos biflorus (horse gram) and Canavalia ensiformis (concanavalin A) (Jack bean). No two lectins showed identical patterns of fluorescence. FITC-conjugates of soybean and D. biflorus lectins reacted strongly with the mucus present in the crypt lumens and with the surface (as well as cytoplasm) of the epithelial cells suggesting that these sites are rich in terminal, non-reducing, N-acetylgalactosamine residues. Wheat germ, R. communis, U. europaeus and concanavalin A-FITC conjugates did not stain mucus but showed fluorescence in the cytoplasm of absorptive cells as well as in the lamina propria and submucosa. The FITC-R. communis conjugate also reacted with structures in the apical portion of epithelial cells that may correspond to the Golgi apparatus.

Because inherent regional differences in colonic epithelium may determine the biologic behavior of tumors originating from different sites, and inasmuch as colonic epithelial cells secrete mucins that may reflect the state of cell differentiation, colonic goblet cell mucin was analyzed with the use of fluorescein isothiocyanate-conjugated lectins and fluorescence microscopy in normal fetal and adult mucosa and in cancers of the proximal and distal colon. In the adult proximal colon only the goblet cell mucin in the upper portion of the crypts was specifically labeled by the lectin Dolichos biflorus agglutinin (DBA), whereas this gradient was progressively lost distally; mucin in the upper and lower crypts of the sigmoid colon and rectum bound the label uniformly. In fetuses less than 22 weeks of age, DBA bound only to mucin in the crypts of the distal colon. Seven of 12 (58%) cancers originating from the proximal colon bound DBA, whereas only 2 of 23 (9%) from the distal colon bound this lectin (P less than .005). Logistic regression analysis suggested that this difference may reflect the occurrence of larger tumors in the proximal colon. Regional differences in the binding of Ulex europaeus agglutinin (UEA-I) to nonneoplastic mucosa was similar to that found by others; predominant binding occurred in the proximal colon in the adult. No difference was noted for UEA-I binding to tumors of the proximal (8 of 12; 66.6%) versus distal (11 of 23; 48%) colon (P = .48). These findings may reflect regional differences in normal and tumor-related carbohydrate structures in mucin of the human colon.

The effect of estrogen on cell proliferation in the descending colon of the mouse as an example of a non-target organ was investigated. Ovariectomized mice were given single or multiple injections of 10 ng/g body weight of 17 beta-estradiol and were killed 1 h after 3H-thymidine injection. Estrogen treatments decreased incorporation of 3H-thymidine into the DNA of colonic mucosa most markedly at 4 h after the single or the last of multiple injections. The inhibitory effect of estrogen on 3H-thymidine incorporation was greater and lasted longer after a single injection than after multiple ones. A similar inhibitory effect was observed in the colonic mucosa of male mice as well as in the mucosa of mice in which colonic epithelial cell proliferation was enhanced by refeeding after 48 h of fasting. However, the colonic mucosa of mice treated with estrogen implants for up to 4 days was not affected. Estrogen treatments caused no significant change in the DNA, RNA and protein contents of the colonic mucosa. The efficacy of estrogen treatments was verified by an increase in both the wet and dry weights of the uterine horns of ovariectomized mice.

Peptide YY (PYY)-like immunoreactivity was detected in the mucosa and muscle layer of normal human colon and rectum and in well to moderately differentiated adenocarcinoma derived from the mucosa of the colon and rectum, using a sensitive and specific radioimmunoassay for PYY. The content of PYY-like immunoreactivity in the mucosa was markedly higher than those in the muscle layer and adenocarcinomatous tissue of any part of the colon and rectum. A high concentrations of PYY-like immunoreactivity was demonstrated throughout the colon mucosa (ascending colon 94.14 +/- 15.34 pmol/g, transverse colon 137.19 +/- 13.44 pmol/g, descending colon 168.89 +/- 15.63 pmol/g, and sigmoid colon 223.69 +/- 35.31 pmol/g), the highest being observed in the rectum (313.15 +/- 45.90 pmol/g). The major molecular form of PYY-like immunoreactivity both in the mucosa and muscle layer of normal human colon and rectum and in adenocarcinomatous tissue was judged by gel exclusion chromatography to be identical to pure porcine PYY. This study revealed the presence of PYY-like immunoreactivity not only in normal tissue of the colon and rectum but also in adenocarcinomas with the same elution pattern, and the mucosal concentrations of PYY-like immunoreactivity were found to be increasing distally throughout the colon and rectum.

We studied four cases of collagenous colitis (CC), which is characterized by thickening of the subepithelial collagenous layer of the colorectal mucosa and is associated with chronic watery diarrhea. Eighteen cases of CC have been previously reported in the literature, to our knowledge. In our four patients, ultrastructural study and collagen immunotyping demonstrated that the basement membrane over the thickened collagen was normal. We found no similar collagenous thickening in a series of 96 biopsy specimens of chronic inflammatory bowel diseases, but collagenous thickening was noted in 17 (25%) of 68 colonic metaplastic polyps. The cause of CC is unknown. The pericryptal fibroblastic sheath of the colonic mucosa has a probable role in the development of CC and metaplastic polyps. The role of the numerous mast cells present in our four cases of CC is currently unknown.

Glycoproteins have been isolated from a 1 M sodium chloride extract of the colonic epithelial cells of man and rat by a combination of Agarose gel and DEAE-cellulose chromatography; The glycoproteins contain O-acetylated or O-esterified sialic acids of at least four types: (a) unsubstituted or possible at position C-9, (b) substituted at C-4 and possibly at C-9, (c) substituted at C-7 and/or C-8, and (d) substituted at C-4 as well as at C-7 and/or C-8.

Twenty patients undergoing colorectal surgery were given an initial dose of 2 g latamoxef (moxalactam) parenterally at the induction of anaesthesia. Ten of the patients received two subsequent doses at 8 h intervals. A series of serum and faecal specimens were taken for analysis of latamoxef concentrations. Tissue samples from intestinal mucosa and peritoneal fat were also taken at surgery. The mean serum peak level was 119.8 +/- 10.7 mg/l. The mean half life was 2.8 h and the mean area under the serum concentration curve was 345.0 +/- 25.7 mg h/l. The concentrations in intestinal mucosa were high when the tissue was removed within 4 h after the administration of latamoxef (22.6 +/- 1.7 micrograms/g, mean +/- S.E.). The concentrations declined when the tissues were removed later (7.8 +/- 0.7 micrograms/g). The content of latamoxef in peritoneal fat varied, range 0.8-23.4, mean 8.5 +/- 1.4 micrograms/g. Measureable levels of latamoxef in faeces were found in all patients except one. The values varied between 0.2 to 23.0, mean 8.9 +/- 1.5 micrograms/g. Faecal samples were also collected during the investigation period for cultivation of aerobic and anaerobic bacteria. The aerobic bacteria-streptococci, enterococci and enterobacteria--were suppressed significantly during the prophylaxis period. Among the anaerobic bacteria--cocci, lactobacilli, bifidobacteria, clostridia, bacteroides and fusobacteria--decreased markedly during the same period. After two weeks the microflora was normalized in all patients. There were no differences between the patients receiving one dose of latamoxef and those receiving three doses. No postoperative infections occurred.