Purification of total RNA from human stool samples.

While colonoscopy may detect early-stage colon tumors, a less invasive and more cost-effective technique would be beneficial. Stool, which picks up sloughed-off colonic epithelial cells, would be ideal for sampling the mucosa; shed tumor cells may display alterations in gene expression observed in intact tumors. It is first necessary, however, to show that RNA can be isolated from human feces and that this RNA contains human gene transcripts. We have therefore developed a method for the isolation of total RNA from freshly passed human stool, consisting of lysis in chaotropic agents, repeated extraction with phenol and phenol-chloroform, and absorption with an RNA-binding resin. After treatment with RNase-free DNase I, we assayed these preparations for the presence of human RNA by quantitative slot blotting, northern blotting, and reverse transcription-polymerase chain reaction (RT-PCR). We obtained 5-30 microg RNA per gram of stool from cancer patients, and about 5 microg RNA per gram of control stool. Quantitative slot blotting showed that about 10% of this RNA was of human origin. Both northern blotting and RT-PCR demonstrated the presence of human RNA in these samples. To unambiguously demonstrate the isolation of RNA from stool, we incubated a mixture of rat cells and control human stool at 37 degrees C for up to 24 hr. RT-PCR of the RNA isolated from this sample clearly revealed the presence of rat-specific mRNA. These experiments indicate that RNA can be isolated from human stool and that message encoded by human genes can be assayed in these preparations. This procedure may provide a powerful tool to identify patients at risk for colon cancer.

Expression of protooncogene-encoded mRNA by colonic epithelial cells in inflammatory bowel disease.

Protooncogenes are cell cycle-related genes that are involved in cell growth of proliferation. Alterations in the level of expression of these genes, or expression of aberrant gene productions, have been observed in tumors and precancerous conditions. To determine if expression of these genes is altered in patients with inflammatory bowel disease (IBD) --who are at risk for development of colon cancer--we assayed transcripts of 15 protooncogenes in colonic epithelial cells of IBD patients and controls. Nine of these genes (H-ras, c-myc, c-fos, c-jun, junB, N-myc, c-abl, c-yes, and p53) were expressed in epithelial cells, whereas two (RB1 and N-ras) were not. expression of four other genes (c-src, K-ras, c-raf, and c-myb) was observed, but the intensity of these bands was too low for densitometric analysis. The steady-state levels of transcripts of H-ras and five nuclear protooncogenes (c-myc, c-fos, c-jun, junB, and N-myc) were lower in epithelial cells from involved or uninvolved IBD samples than in normal epithelial cells from either sporadic colon cancer or diverticulitis patients. The level of c-fos mRNA was two- to threefold higher in involved than in uninvolved areas of the colons of two ulcerative colitis (UC) patients, but not in one Crohn's disease (CD) patient. Message abundance of c-abl transcripts was two- to threefold lower in UC epithelial cells than in either the CD or control samples. The steady-state level of c-yes-encoded mRNA was considerably higher in IBD patients resected for colon cancer than in patients resected for active chronic IBD or in controls. The level of p53 message was constant in these samples. Increased levels of c-fos mRNA in involved UC relative to uninvolved UC may be related to the disease process. Decreased expression of c-abl transcript in UC may be a diagnostic marker for UC and may be related to the rate of cell turnover in these diseases. Enhanced expression of c-yes in IBD patients with tumors compared to active chronic IBD and controls suggests that expression of this gene may be a marker for development of colon cancer in IBD.

Expression of growth factor receptor-encoded mRNA by colonic epithelial cells is altered in inflammatory bowel disease.

A link between inflammation of the colon in inflammatory bowel disease (IBD) and the increased risk of colon cancer in ulcerative colitis (UC) may be provided by growth factor receptor genes. Their expression may be altered in response to growth factors present in the mucosa, and this, in turn, may induce further genetic changes, linked to carcinogenesis, in the cells of the colonic epithelium. To test this hypothesis, we assayed steady-state levels of eight growth factor receptor mRNAs in colonic epithelial cells of IBD patients and controls. Four of these genes (EGF-R, IGFI-R, CSF1-R, and PDGF-R-beta) were expressed in epithelial cells, whereas four (erbB-2, erbB-3, NGF-R, and met) were not. The level of the former in involved or uninvolved IBD was considerably lower than in normal epithelial cells from either sporadic colon cancer or diverticulitis patients. In contrast, expression was much higher in IBD patients with colon tumors than in active chronic IBD. The level of PDGF-R-beta mRNA was two- to fourfold higher in involved than in uninvolved areas of the colons of two UC patients, but not in one Crohn's disease patient. Message abundance of its ligand, PDGF-beta, however, was the same in paired UC samples. The pattern of expression of PDGF-beta and cripto was identical to that of EGF-R, whereas the level of mRNA of amphiregulin was the same in active chronic IBD and IBD patients with tumors. A fourth growth factor, Kfgf, was not expressed. Increased levels of PDGF-R-beta mRNA in involved UC relative to uninvolved UC may be related to the disease process in UC. Decreased expression of growth factor- and growth factor receptor-encoded mRNA in active chronic IBD may be related to the disease process, or it may be an effect of steroid therapy undergone by these patients. Enhanced expression of these genes in IBD patients with tumors compared to those without tumors suggests that this may be a marker for development of colon cancer in IBD.

Beta adrenergic stimulation and blockade in cirrhosis: effects on azygos vein blood flow and portal hemodynamics.

It is unknown whether beta adrenergic stress has adverse hepatic hemodynamic effects. Therefore, the authors studied the hemodynamic effects of beta adrenergic stimulation and subsequent blockade in 10 patients with cirrhosis (6 Childs A, 3 Childs B, and 1 Childs C) with known or suspected portal hypertension. Free and wedged hepatic vein pressures, hepatic venous pressure gradient, heart rate, mean arterial pressure, cardiac output, and azygos vein blood flow were measured at rest and after isoproterenol infusion (mean dose = 7.3 micrograms/min: target heart rate = 150% to 200% of resting heart rate). Esmolol, an ultra-short-acting beta blocker, was then infused (dose titrated to return heart rate to baseline), and all measurements were repeated. Based on the results, the authors conclude that beta adrenergic stress provoked by isoproterenol infusion significantly increases azygos vein blood flow and hepatic venous pressure gradient. Beta blockade with esmolol reduces azygos vein blood flow and hepatic venous pressure gradient significantly below baseline.

K-ras oncogene mutations in rat colon tumors induced by N-methyl-N-nitrosourea.

We have been studying a rat model of colon cancer in which tumors are induced by direct application of N-methyl-N-nitrosourea (MNU) to discrete areas of the colonic mucosa for a limited period of time. Activation of the ras genes by point mutation has been observed in many experimental tumors, including tumors induced by MNU. To detect potential activating point mutations in the H-ras and K-ras oncogenes in MNU-induced rat colon tumors, DNA samples from 40 adenomas, nine carcinomas, and 14 histologically normal tissue samples from 14 rats--as well as from 16 foci induced on NIH3T3 cells by tumor DNAs--were amplified by the polymerase chain reaction and hybridized with allele-specific oligonucleotide probes. No H-ras point mutations were observed in any of these samples. We did detect K-ras point mutations, however, in four primary tumours--one adenoma (2.5%) and three carcinomas (33%); these mutations were all G----A transitions at the second nucleotide of codons 12 and 13. The absence of detectable ras mutations from the majority of tumors suggests that, in contrast to other animal models utilizing MNU, tumorigenesis in MNU-induced rat colon tumors may predominantly involve activation of genes other than ras.

Detection of transforming oncogenes in rat colon tumors induced by direct perfusion with N-methyl-N-nitrosourea.

We assayed rat colon tumors induced by N-methyl-N-nitrosourea (MNU) for transforming oncogenes by the NIH 3T3 transfection and nude mouse tumorigenicity assays. Transfection of DNA from 3 of 3 adenomas and 3 of 5 carcinomas induced transformed foci on NIH 3T3 cells. DNA from 2 of 3 primary foci also possessed focus-forming activity, and rat-specific sequences were observed in secondary focus DNAs. Furthermore, NIH 3T3 cells transfected with DNA from a carcinoma and from a primary focus derived from it, both positive in the focus-forming assay, induced tumors in nude mice. We found no evidence for rat H-ras, K-ras, or N-ras sequences in the DNA of any of 16 primary foci derived from 6 rat tumors; thus, in contrast to other animal tumor models induced by MNU, activation of the ras genes does not appear to predominantly occur in MNU-induced rat colon tumors. We also did not observe, in any of these foci, sequences corresponding to the rat neu, raf, fms, met, or hst genes, thus indicating that none of these is the transforming oncogene in our model. These results suggest that an as yet unidentified transforming oncogene may be activated in rat colon tumors induced by MNU.

Oncogene alterations in rat colon tumors induced by N-methyl-N-nitrosourea.

The authors assayed oncogene alterations in rat colon tumors induced by the direct-acting chemical carcinogen, N-methyl-N-nitrosourea (MNU). DNA isolated from 34 adenomas and eight carcinomas, as well as adjacent normal colon, of 11 rats was assayed by Southern blotting for restriction fragment length polymorphisms and gene amplifications and deletions in 13 oncogenes known to be involved in human or other animal tumors. In addition to finding apparent point mutations or other small alterations in the fos and abl genes in individual rat colon tumors, the authors observed what appear to be larger alterations (ie, rearrangements, or intragenic insertions or deletions) in the H-ras and myb loci in several tumors. In contrast, no changes in the K-ras, N-ras, myc, N-myc, neu, raf, fms, met, and hst genes were seen in any of these tumors. The frequency of myb gene alterations was higher in carcinomas than in adenomas, suggesting that these changes occurred relatively late during tumorigenesis and were not direct effects of the carcinogen. In addition, the finding of alterations in two or three oncogenes in several MNU-induced rat colon tumors suggests the possibility of more widespread genomic lesions in this model.

Enhanced expression of oncogene-encoded mRNA in a rat model of colon cancer.

We have studied the expression of oncogene-encoded mRNAs in a rat model of colon cancer. In this model, rats are intrarectally administered several low doses of the direct-acting carcinogen, N-methyl-N-nitrosourea (MNU). Tumors, predominantly adenomas, develop 5-7 months following administration of the carcinogen, and many of these progress to carcinomas. Upon assaying the steady-state levels of oncogene-encoded transcripts in normal rat colon, we found that fos and N-myc are highly expressed; H-ras, K-ras, myc, myb, and neu messages are present at lower levels; and N-ras, abl, and raf mRNAs are absent. When we compared transcript levels in rat tumors to those in normal colons from the same animal, we observed a 2-4 fold increase in both myc- and H-ras-encoded mRNAs and a 2-7 fold increase in myb message, but no change in expression of any of the 7 other genes. To test whether this increased expression is related to tumor production or is simply a result of the more rapid cellular turnover observed in tumor tissue, the level of oncogene-encoded transcripts was assayed in colonic mucosae of rats given two treatments known to enhance cell turnover and DNA synthesis in the colon. Neither acute application of MNU nor a diet containing 1% cholic acid caused any change in the level of oncogene-encoded mRNAs in rat colons, thus suggesting that the increased abundance of myc, myb, and H-ras messages in tumors is associated with tumor formation. The enhancement of expression of these genes in adenomas, as well as in carcinomas, further suggests that these alterations occur relatively early during the tumorigenic process.

Direct cholangiography.

The role of direct cholangiographic methods has evolved significantly over the last two decades as other high-resolution imaging modalities have become available. Most of the time, careful clinical evaluation of a patient combined with ultrasonography or CT will enable a physician to arrive at a correct diagnosis with a high level of precision. In this article we have attempted to indicate situations in which direct cholangiographic methods are necessary to diagnose and treat certain hepatobiliary problems. Considerable controversy exists concerning the application of these methods for treatment of biliary problems. However, in circumstances in which the issues are openly discussed, application of these techniques can be agreed on. Direct cholangiography, like coronary angiography, is a technology that provides considerable valuable information that at present cannot be obtained by other techniques, and, in well-defined circumstances, is necessary for precise diagnosis and therapy.

Oncogene alterations in primary human colon tumors.

We have examined alterations in six oncogenes--H-ras, K-ras, N-ras, myc, fos, and N-myc--in nine primary human colon tumors. Tumors were obtained within an hour of resection; as a control for each tumor, adjacent normal colon tissue was also obtained. Deoxyribonucleic acid extracted from each tissue sample was assayed by digesting with appropriate restriction endonucleases and, after gel electrophoresis and transfer to nitrocellulose, hybridizing with radiolabeled oncogene probes. Amplification of the myc locus, relative to adjacent normal colon tissue, was observed in two of these tumors; by dot-blotting, it was estimated that myc was amplified twofold to fivefold in each tumor. No rearrangements of myc, however, were observed in any of these tumors. Examination of the H-ras alleles of these nine tumors revealed that eight possess only "common" alleles of this gene, and that each was identical to its control. Normal colon DNA of the ninth patient, however, was found to possess both a "common" and a "rare" allele, and the "common" allele of H-ras appeared to be deleted in the tumor DNA of this patient. A restriction polymorphism indicative of a mutation in the 12th codon of K-ras was not found in any of these tumors, and we observed no evidence of rearrangement of amplification of the N-ras, K-ras, fos, or N-myc genes.

Hemorrhoids or rectal varices: defining the cause of massive rectal hemorrhage in patients with portal hypertension.

Identifying the source of lower gastrointestinal hemorrhage in patients with chronic liver disease and portal hypertension can be challenging. We present 2 cases of hemorrhage from rectal varices and a discussion on the differences between simple hemorrhoids and rectal varices. Evaluation of rectal bleeding in patients with portal hypertension is discussed and possible therapeutic options are described.

Intestinal perforation associated with cytomegalovirus infection in patients with acquired immune deficiency syndrome.

Gastrointestinal complications are an important aspect of the acquired immune deficiency syndrome. In this report we describe two male homosexuals with the acquired immune deficiency syndrome whose gastrointestinal symptoms culminated in the complication of intestinal perforation. Cytomegalovirus inclusions are seen prominently in the areas of perforation. The pathogenic role of cytomegalovirus in these cases is discussed. We propose that cytomegalovirus-induced enteritis may lead to bowel perforation in patients with the acquired immune deficiency syndrome.

Reduction of N-methyl-N-nitrosourea-induced colon tumors in the rat by cholesterol.

Patient populations with a propensity to develop colon cancer have increased amounts of fecal cholesterol (and/or cholesterol metabolites). In this study, we report the effect of increased colonic concentrations of cholesterol and its metabolites on colon tumor promotion. The chemical carcinogen N-methyl-N-nitrosourea was instilled intrarectally into rats to initiate colon tumor formation. Following initiation, a cholesterol-supplemented diet was given. Despite a 2-fold elevation of fecal cholesterol, the number of colon tumors found was significantly reduced. These studies suggest that under certain conditions cholesterol may inhibit colon carcinogenesis.

The kinetics of the protective effect of beta-sitosterol against MNU-induced colonic neoplasia.

The effect of dietary supplementation with beta-sitosterol (0.2% of diet) was followed in Fischer rats after both acute and chronic feeding. Compared to controls after 3 and 7 days, the number of epithelial cells per crypt column in both plant sterol fed groups was shifted to lower values; moreover, fewer cells above cell position 12 were engaged in DNA synthesis. Continued feeding of beta-sitosterol for 28 weeks revealed the number and position of 3HTdR-labeled cells per crypt column after 1 pulse labeling to be similar to that seen in the acute phase. However, differences were more marked after 24 h showing the maximum number of labeled cells per column to be at least 25% less than the untreated group and the leading edge of labeled cells moving more slowly up the crypt wall. Rats treated with N-methyl-N-nitrosourea (MNU) intrarectally (8 mg/animal) while simultaneously consuming beta-sitosterol demonstrated a reduction in the size of the proliferative compartment as well as the number of labeled cells per crypt column as compared to rodents receiving just the carcinogen (3.3 and 5.4 mean labeled cells per column, respectively). At this time, MNU-treated rats fed beta-sitosterol had a significantly decreased colonic tumor incidence (Raicht et al. 1980). This plant sterol which passes through the digestive tract relatively unabsorbed appears to slow colonic epithelial cell proliferation resulting in a reduced expression of neoplastic transformation.

Sterol metabolism in the rat. Effect of cholesterol and cholesterol-alpha-epoxide on sterol metabolism in rats fed liquid diets.

Certain oxygenated derivatives of cholesterol have dramatic effects on cholesterol synthesis. The present study compared the effects of cholesterol and an oxygenated metabolite, cholesterol-alpha-epoxide, on sterol metabolism in rats. Sterol balance measurements using isotopic and chromatographic techniques were carried out in rats fed liquid control diets, control diets + cholesterol (1 mg/ml), and control diets + cholesterol-alpha-epoxide (1 mg/ml). Sterol metabolism was affected by both cholesterol and cholesterol-alpha-epoxide. Cholesterol feeding decreased cholesterol synthesis (-9.57 +/- 7.23 mg/day), increased endogenous bile acid synthesis (7.71 +/- 1.18 mg/day), and increased cholesterol turnover (7.78 +/- 2.33 mg/day) compared to controls. Cholesterol-alpha-epoxide had no effect on cholesterol synthesis, endogenous bile acid synthesis and cholesterol turnover compared to controls. However, animals fed cholesterol-alpha-epoxide had large increases in total acidic steroid output (determined by chromatographic analysis, 12.33 +/- 4.05 mg/day). This finding suggests that cholesterol-alpha-epoxide is absorbed and converted to bile acids. Apparently, the epoxide enters the bile acid biosynthetic pathway distal to the rate-limiting step of 7 alpha-hydroxylation. As a result, large amounts of bile acids are formed from the epoxide without affecting endogenous cholesterol or bile acid synthesis. This was confirmed in a separate experiment by feeding [4-14C]cholesterol-alpha-epoxide and recovering labeled bile acids (hyodeoxycholic acid and lithyocholic acid) as well as the starting radioactively labeled epoxide in the feces.

Effects of bile acids on colon carcinogenesis in rats treated with carcinogens.

Primary bile acids were studied as possible colon tumor promoters or inhibitors in a rat model of chemically induced colon cancer. Cholic acid feeding increased the number of animals with tumors, the number of tumors per animal, and the number of tumors per tumor-bearing animal. Tumor enhancement was attributed to deoxycholic acid, the bacterial metabolite of cholic acid. When chenodeoxycholic acid was fed to the rats in our model, tumor incidence was increased, but the number of tumors per animal and the number of tumors per tumor-bearing animal were similar to controls. The different fecal bile acid pattern obtained with chenodeoxycholic acid may be responsible for the differences in tumor incidence. The methodology to characterize and identify all steroidal components of the feces requires extraction, thin-layer chromatography, gas-liquid chromatography, and gas-liquid chromatography-mass spectrometry. Newer techniques include LH-20 chromatography (for sulfated steroids) and high-pressure liquid chromatography.

Kinetic and morphologic alterations in the colon of a patient with multiple polyposis.

The histologic and proliferative characteristics of 16 colonic biopsies taken from an operative specimen of a 38-year-old female patient with multiple polyposis are presented. Kinetic measurements are based on 3HTdR incorporation accomplished in vitro, using both single and double labelling techniques. Epithelial cells in adenomatous tissue were more actively engaged in DNA synthesis than those normal-appearing mucosa (L.I.13.0 +/- 6.9 versus 9.3 +/- 1.6); however, because of extreme variability between biopsies, the difference was not significant. No difference in S phase duration was found, but a faster turnover time (Tg) than that in the normal appearing colonic mucosa was estimated (Tg 52.6 hours versus 74.2 hours). Only two of ten biopsies containing normal-appearing mucosa had a completely normal incorporation pattern with proliferative cells located only in the lower two thirds of crypts. Eight biopsies showed extension of the proliferative compartment to the luminal surface. One of these eight also expressed an additional proliferative defect, namely, a shift of the major zone of DNA synthesis to the middle and upper regions of the crypts. This specimen had been located adjacent to an area of microscopic adenoma. Subpopulations of labelled epithelial cells showing transition from normal to hyperplastic or to adenomatous appearance were in the otherwise normal-appearing crypts. The majority of labelled hyperplastic-appearing cells occupied the lower thirds of the crypts, whereas 68% of labelled adenomatous-appearing cells were located in the middle and upper zones of the glands. Based on these observations, the development of an adenoma is believed forecast by the redistribution of the proliferative compartment toward the surface. A further stage in tumorogenesis before the appearance of a focus of neoplasia is the emergence of actively proliferating cells transforming to a more adenomatous appearance primarily in that same location, that is, the middle and upper third of the colonic crypts.