Expression of SDF-1 alpha and nuclear CXCR4 predicts lymph node metastasis in colorectal cancer.

Although stromal cell-derived factor (SDF)-1 alpha and its receptor CXCR4 are experimentally suggested to be involved in tumorigenicity, the clinicopathological significance of their expression in human disease is not fully understood. We examined SDF-1 alpha and CXCR4 expression in colorectal cancers (CRCs) and their related lymph nodes (LNs), and investigated its relationship to clinicopathological features. Specimens of 60 primary CRCs and 27 related LNs were examined immunohistochemically for not only positivity but also immunostaining patterns for SDF-1 alpha and CXCR4. The relationships between clinicopathological features and SDF-1 alpha or CXCR4 expression were then analysed. Stromal cell-derived factor-1 alpha and CXCR4 expression were significantly associated with LN metastasis, tumour stage, and survival of CRC patients. Twenty-nine of 47 CXCR4-positive CRCs (61.7%) showed clear CXCR4 immunoreactivity in the nucleus and a weak signal in the cytoplasm (nuclear type), whereas others showed no nuclear immunoreactivity but a diffuse signal in the cytoplasm and at the plasma membrane (cytomembrane type). Colorectal cancer patients with nuclear CXCR4 expression showed significantly more frequent LN metastasis than did those with cytomembrane expression. Colorectal cancer patients with nuclear CXCR4 expression in the primary lesion frequently had cytomembrane CXCR4-positive tumours in their LNs. In conclusion, expression of SDF-1 alpha and nuclear CXCR4 predicts LN metastasis in CRCs.

Protective role of intracellular superoxide dismutase against extracellular oxidants in cultured rat gastric cells.

We examined the role of intracellular superoxide dismutase (SOD) as an antioxidant by studying the effect of diethyldithiocarbamate (DDC) on extracellular H2O2-induced damage in cultured rat gastric mucosal cells. 51Cr-labeled monolayers from rat stomachs were exposed to glucose oxidase-generated H2O2 or reagent H2O2, which both caused a dose-dependent increase in 51Cr release. DDC dose-dependently enhanced 51Cr release by hydrogen peroxide, corresponding with inhibition of endogenous SOD activity. This inhibition was not associated either with modulation of other antioxidant defenses, or with potentiation of injury by nonoxidant toxic agents. Enhanced hydrogen peroxide damage by DDC was significantly prevented by chelating cellular iron with deferoxamine or phenanthroline. Inhibition of cellular xanthine oxidase (possible source of superoxide production) by oxypurinol neither prevented lysis by hydrogen peroxide nor diminished DDC-induced sensitization to H2O2. We conclude that (a) extracellular H2O2 induces dose dependent damage to cultured gastric mucosal cells; (b) intracellular SOD plays an important role in preventing H2O2 damage; (c) generation of superoxide seems to occur intracellularly after exposure to H2O2, but independent of cellular xanthine oxidase; and (d) cellular iron mediates the damage by catalyzing the production of more reactive species from superoxide and H2O2, the process which causes ultimate cell injury.

Nitric oxide enhances cytotoxicity of cultured rabbit gastric mucosal cells induced by hydrogen peroxide.

While NO has been reported to act as a protective factor to gastric mucosa, it has been shown to be cytotoxic to various cells. NO also has been demonstrated to stimulate prostaglandin (PG) release and mucous glycoprotein secretion which could result in the activation of gastric defensive mechanisms. We examined the effect of NO on cytotoxicity induced by hydrogen peroxide, and mucous glycoprotein secretion and PGE2 release from cultured rabbit gastric mucosal cells. NO enhanced cytotoxicity induced by hydrogen peroxide. Defensive prostaglandin E2 release and mucous glycoprotein secretion were not altered by NO. Under certain circumstances, NO might behave as an aggressive factor in gastric mucosal injury.

Minimal change oesophagitis: a disease with characteristic differences to erosive oesophagitis.

BACKGROUND: The majority of gastro-oesophageal reflux disease (GERD) seems to be non-erosive reflux disease. Nonerosive reflux disease includes minimal change oesophagitis (whitish or reddish, oedematous change and erosion that is not regarded as mucosal break) and no endoscopic abnormalities. AIM: To investigate the accurate proportion of those with minimal change oesophagitis and to clarify its characteristics. In addition, we evaluated the effect of famotidine (40 mg/day) in those with minimal change. METHODS: Prospective endoscopic assessment was performed for consecutive 606 out-patients. Of the 582 patients suitable for analysis, 347 were non-treated. The latter were divided into those with erosive GERD or minimal change, and their endoscopic findings and characteristics were compared. RESULTS: Among 347 non-treated patients, 88 (25%) had erosive GERD and 249 (72%) had minimal change. Compared with patients who have erosive GERD and those with minimal change, the latter were less likely to have hiatal hernia or bile reflux, but more likely to have gastric atrophy. Symptomatic patients (n = 55) with minimal change oesophagitis were more likely to have hiatal hernia than those who were asymptomatic (n= 194). Most patients preferred taking famotidine on-demand, during a 4-week follow-up period. CONCLUSIONS: Most non-erosive reflux disease can be classified as minimal change oesophagitis, and that have different characteristics from erosive GERD. On-demand famotidine may be a suitable alternative treatment for patients with minimal change disease.

The effects of aspirin on antioxidant defences of cultured rat gastric mucosal cells.

BACKGROUND: Helicobacter pylori-associated inflammation leads to exposure of the gastric epithelium to reactive oxygen species (ROS) generated in the gastric mucosa. In some pathological conditions, such as those induced by nonsteroidal anti-inflammatory drugs, the gastric mucosa may become more susceptible to ROS. AIM: To examine the effects of aspirin on antioxidant defenses as well as on oxidant injury in cultured rat gastric mucosal cells. METHODS: Primary monolayer cultures of rat gastric fundic mucosa were exposed to an ROS-generating system, hypoxanthine/xanthine oxidase (XOD). Cytotoxicity was quantified by measuring 51Cr release from prelabelled cells. The effects of aspirin on antioxidants and on cellular injury brought about by the ROS-generating system were determined. RESULTS: XOD, in the presence of hypoxanthine, caused a dose-dependent increase in specific 51Cr release, which corresponded to the ability of XOD to produce ROS (as assessed by the production of uric acid from hypoxanthine). Incubation of cells with aspirin (1-100 microM) produced a dose-dependent increase in XOD-induced 51Cr release. Aspirin did not affect cellular glutathione content or activity of glutathione peroxidase, glutathione reductase or endogenous catalase. By contrast, aspirin caused a dose-dependent reduction in mucus synthesis. as assessed by incorporation of [3H]-glucosamine hydrochloride into the cells. CONCLUSIONS: Aspirin at therapeutically relevant concentrations rendered cultured gastric cells more susceptible to subsequent exposure to ROS. Aspirin affected neither the glutathione redox cycle nor catalase activity. Thus, the enhancement of ROS-induced injury by aspirin may be accomplished through diminished gastric mucus synthesis, since mucus is a potent scavenger of ROS. These findings provide insight into how gastric inflammation and injury (such as that induced by H. pylori infection) in human gastric mucosa is modulated by the administration of nonsteroidal anti-inflammatory drugs.

Up-regulation of TFF expression by PPARgamma ligands in gastric epithelial cells.

BACKGROUND: Although trefoil factor family peptides (TFF peptides) are assumed to play important roles in gastric mucosal protection, the regulatory mechanism of gastric TFF expression has not been fully understood yet. Recent reports showed gastric expression of peroxisome proliferator-activated receptor gamma (PPARgamma), a nuclear receptor known to be involved in the regulation of cell growth and differentiation in other cell types, such as adipocytes. AIM: To determine whether PPARgamma affects the expression of TFF in gastric epithelial cells. METHODS: MKN45 gastric cells were used as a model of gastric epithelial cells. DNA synthesis of the cells was determined by the measurement of BrdU incorporation. The effects of PPARgamma ligands, 15-deoxy-Delta12,14-prostaglandin J2 (15d-PGJ2) and troglitazone (TGZ) on TFF expression were assessed by real-time quantitative reverse transcription-polymerase chain reaction (RT-PCR). RESULTS: MKN45 cells expressed a significant amount of PPARgamma. Both 15d-PGJ2 and TGZ suppressed DNA synthesis of the cells in a dose-dependent manner. In the control condition, MKN45 cells most abundantly expressed TFF1 and the relative expression level of TFF1, TFF2, and TFF3 mRNA was 1700:32:1. TFF1 and TFF2 mRNA levels were significantly up-regulated by the incubation of the cells with 15d-PGJ2 (10 micro m) or TGZ (30 micro m), whereas TFF3 mRNA level was not affected. CONCLUSION: The results of the present study suggest a possible role of PPARgamma in the regulation of TFF expression in gastric epithelial cells.

Polaprezinc protects gastric mucosal cells from noxious agents through antioxidant properties in vitro.

BACKGROUND: Polaprezinc has been shown to exert an anti-oxidant property in a tube experiment, protect gastric mucosa from experimental ulcerations in vivo, and accelerate the healing of gastric ulcer in humans. AIM: To examine a possible protective effect of polaprezinc on oxidant-mediated injury in primary monolayer cultures of rat gastric fundic mucosa. METHODS: Cytotoxicity was quantified by measuring 51Cr release. Whether or not polaprezinc exerts an antioxidant property was investigated by determining the effect of this agent on hydrogen peroxide (H2O2)-induced injury. The effects of polaprezinc on superoxide (O2-. ) generation as well as on ethanol (EtOH)-induced injury were also examined. Generation of O2-. was assessed by the reduction in cytochrome c. RESULTS: H2O2 caused a time- and dose-dependent increase in 51Cr release. The dose-response curve of 51Cr release by H2O2 shifted to the right in the presence of polaprezinc. Polaprezinc, at submillimolar concentrations, prevented H2O2-induced 51Cr release. EtOH also caused a dose-dependent increase in 51Cr release, which was prevented by the addition of polaprezinc. The incubation of cells with EtOH caused an increase in cytochrome c reduction, as the concentrations of EtOH increased. Polaprezinc inhibited EtOH-induced cytochrome c reduction. Protection by polaprezinc was microscopically associated with the prevention of monolayer disruption. CONCLUSIONS: Polaprezinc is antioxidative and directly protects gastric mucosal cells from noxious agents through its antioxidant properties in vitro. This finding may provide the theoretical basis for the usage of an antiulcer drug with antioxidant properties for the treatment of gastric inflammation, such as that induced by ethanol.

Effect of PPARgamma ligands on the viability of gastric epithelial cells.

BACKGROUND: [corrected] Peroxisome proliferator-activated receptors (PPAR) are a family of three nuclear receptors (PPARalpha, PPARdelta, and PPARgamma). Although recent evidence suggests a role for PPARgamma in the regulation of colonic epithelial cell growth, the role for PPARgamma in the stomach has not been established. AIM: To examine the expression of PPARgamma and the effects of PPARgamma ligands on the viability of gastric epithelial cells. METHODS: MKN45 cells and primary cultured rat gastric epithelial cells were used. Troglitazone (TGZ) and 15-deoxy-Delta12, 14-prostaglandin J2 (15d-PGJ2) were used as PPARgamma ligands. Expression of PPARgamma was examined by RT-PCR and Western blot analysis. Cell viability was measured by WST-1 assay and TUNEL assay was performed to detect apoptosis. RESULTS: MKN45 cells expressed all subtypes of PPAR. PPARgamma ligands decreased cell viability and induced cell death in a dose-dependent manner, whereas ligands for PPARalpha and PPARdelta had no significant effect. TUNEL assay showed that this cell death is apoptosis. Primary cultured rat gastric epithelial cells also expressed PPARgamma and activation of PPARgamma decreased cell viability. CONCLUSION: These results suggest that PPARgamma plays an important role in the regulation of cell growth and cell death in gastric epithelial cells.

Indometacin up-regulates TFF2 expression in gastric epithelial cells.

BACKGROUND: Trefoil factor family peptides are expressed in gastrointestinal epithelial cells and play a critical role in maintaining mucosal integrity. Although non-steroidal anti-inflammatory drugs (NSAIDs) are important causative agents of gastric mucosal lesions, few data are available about the effect of NSAIDs on trefoil family peptides in gastric mucosa. AIM: To examine whether indometacin, a widely used NSAID, affects trefoil factor family expression in gastric epithelial cells. METHODS: MKN45, a cell line derived from human gastric cancer, was used. TFF1, TFF2, and TFF3 mRNA expression was assessed by real-time quantitative reverse transcription-polymerase chain reaction (RT-PCR). TFF2 gene transcription was also examined by luciferase reporter gene assay. RESULTS: Relative expression level of TFF1, TFF2, TFF3 mRNA was 616: 12: 1 in unstimulated MKN45 cells. Although indometacin (1-250 micro mol/L) had no significant effect on the expression of TFF1 and TFF3 mRNA, it up-regulated TFF2 mRNA expression in a dose- and time-dependent manner. Luciferase reporter gene assay confirmed the up-regulation of TFF2 gene transcription by indometacin. Indometacin-induced up-regulation of TFF2 expression was not antagonized by externally applied prostaglandin E2. CONCLUSION: These results suggest that indometacin up-regulates gastric epithelial cell TFF2 expression through a COX-independent mechanism. Since TFF peptides play an important role in gastric mucosal protection, indometacin-induced TFF2 may reduce the degree of gastric mucosal damage induced by indometacin.

Bovine liver phosphoamidase as a protein histidine/lysine phosphatase.

A 13-kDa phosphoamidase was isolated as a single band on SDS-PAGE from bovine liver. Its Stokes' radius, sedimentation coefficient, molecular mass, and optimal pH were estimated to be 1.6 nm, 1.8 s, 13 kDa, and 6.5, respectively. The enzyme released P(i) from 3-phosphohistidine, 6-phospholysine, and amidophosphate at rates of 0.9, 0.6, and 2.6 micromol/min/mg protein, respectively. However, it did not dephosphorylate phosphocreatine, N(omega)-phosphoarginine, imidodiphosphate, or O-phosphorylated compounds including inorganic pyrophosphate. It also dephosphorylated succinic thiokinase and nucleoside diphosphate kinase autophosphorylated at His residues, indicating that it works as a protein histidine phosphatase. A thiol reagent, 30 microM N-ethylmaleimide, depressed the activity by half, while a thiol compound, 2-mercaptoethanol, protected the enzyme from heat-inactivation. Five millimolar divalent cations, such as Mg2+ and Mn2+, and 5 mM EDTA, had no effect on the activity.

N(omega)-phosphoarginine phosphatase (17 kDa) and alkaline phosphatase as protein arginine phosphatases.

Seven synthetic polymers, (Glu4, Tyr)n, (Arg)n, (Arg, Pro, Thr)n, (Arg-Gly-Glu)6, (Arg-Gly-Phe)6, (Glu-Arg-Gly-Phe)5, and (Ala-Leu-Arg-Arg-Ile-Arg-Gly-Glu-Arg)2, were treated with phosphoryl chloride to phosphorylate their Tyr, Thr, and Arg residues. Protamines and histones were phosphorylated similarly. These phosphorylated peptides were examined as to whether or not they serve as substrates for intestinal alkaline phosphatase [EC 3.1.3.1] and liver N(omega)-phosphoarginine phosphatase [Kuba, M., Ohmori, H., and Kumon, A. (1992) Eur. J. Biochem. 208, 747-752]. Phosphorylated polyarginine was hydrolyzed with a lower Km with alkaline phosphatase than with N(omega)-phosphoarginine phosphatase, while the phosphorylated forms of (Arg-Gly-Phe)6 and culpeine were better substrates for N(omega)-phosphoarginine phosphatase. When (Arg, Pro, Thr)n and culpeine were phosphorylated chemically after treatment with phenylglyoxal, these phosphorylated peptides were worse substrates for N(omega)-phosphoarginine phosphatase than for alkaline phosphatase. Moreover, the results of proton-decoupled 31P NMR analysis indicated that N(omega)-phosphoarginine phosphatase released Pi from N(omega)-phosphoarginine residues of phosphopeptides. These results indicate that both phosphatases function as protein arginine phosphatases in different manners, and that N(omega)-phosphoarginine phosphatase is useful for selectively detecting N(omega)-phosphoarginine residue in peptides containing various kinds of phosphorylated amino acids.

Cell culture model for antiulcerogenic agents.

To elucidate the mechanisms of antiulcerogenic agents, we established the cell culture model derived from rat gastric epithelium. The cultured cells were identified as mucus-producing cells by using histological analysis. This culture model is useful for investigating the untiulcer effect of various agents and to reveal the mechanisms of the drug action. In particular, the ulcer-healing model using the cultured monolayer is promising and convenient for the study of several growth factors such as HGF as well as antiulcerogenic agents. The effect of polaporezinc in the cultured model is introduced.

Protection of cultured gastric cells against tert-butyl hydroperoxide by glutathione isopropyl ester.

Oxidants are involved in the pathogenesis of a variety of gastrointestinal disorders. Intracellular GSH protects rat gastric cells against oxidants. We examined whether GSH isopropyl ester (GSH ester) can protect against oxidants and whether a system of GSH ester uptake is present in these cells; we also investigated a possible role of GSH in inhibiting lipid peroxidation. Cytotoxicity was quantified by 51Cr release. Lipid peroxidation was assessed by measuring malondialdehyde production. tert-Butyl hydroperoxide caused a dose-dependent increase in 51Cr release. Treatment with GSH ester attenuated the toxicity of tert-butyl hydroperoxide. Incubation with GSH ester enhanced the cellular GSH content, which was prevented by DL-buthionine-[S,R]-sulfoximine, an inhibitor of gamma-glutamylcysteine synthetase. GSH ester prevented tert-butyl hydroperoxide-induced lipid peroxidation, corresponding with the degree of protection. Therefore, we concluded that GSH isopropyl ester protects gastric cells against oxidants through the accumulation of intracellular GSH. This protection is mediated in part by the prevention of hydroperoxide-induced lipid peroxidation. However, the gastric epithelial system of GSH ester uptake appears distinctly different from those observed in hepatocytes, lymphoid cells and fibroblasts in terms of mediation of gamma-glutamylcysteine synthetase activity.

Antioxidant defenses of cultured colonic epithelial cells against reactive oxygen metabolites.

Reactive oxygen metabolites produce colonic epithelial cellular injury. The present study evaluated the protective role of cellular superoxide dismutase, catalase, and glutathione (GSH) redox cycle in cultured rabbit colonic cells. Cultured rabbit colonic epithelial cells were exposed to reactive oxygen metabolites generated by hypoxanthine (1 mM) and xanthine oxidase (1 mU/ml) for up to 5 h. Cytotoxicity was quantified by measuring 51Cr release from prelabeled cells. Pretreatment with diethyldithiocarbamate (inhibitor of superoxide dismutase) reduced activity of cellular superoxide dismutase and increased 51Cr release caused by hypoxanthine/xanthine oxidase from colonic cells. Pretreatment with diethyl maleate (covalently binds GSH as catalyzed by GSH transferase), or buthionine sulfoximine (inhibitor of gamma-glutamylcysteine synthetase) decreased cellular GSH and enhanced reactive oxygen metabolites induced injury. Pretreatment with bis(chloroethyl)-nitrosourea (inhibitor of GSH reductase) inhibited activity of GSH reductase and increased 51Cr release from colonic cells. Preincubation with aminotriazole (inhibitor of catalase) reduced cellular catalase, but did not affect cellular injury. Therefore, we concluded that both cellular superoxide dismutase and the GSH redox cycle appeared to play a role in detoxifying reactive oxygen metabolites and that cellular catalase may be less important in rabbit colonic epithelial cells.

Sucralfate and Helicobacter pylori.

Sucralfate enhances the eradication rate of Helicobacter pylori when administered with antibiotics. Sucralfate acts on H. pylori through mucus gel. Whether sucralfate exerts its anti-ulcer action through the eradication of H. pylori is still controversial. As sucralfate itself is an effective, well-tolerated and cost-effective anti-ulcer agent, its mechanisms of action should be investigated further.

Hepatocyte growth factor accelerates restitution of intestinal epithelial cells.

Many cytokines are involved in the repair of damaged tissue, and one of these, hepatocyte growth factor (HGF), is involved not only with liver regeneration but also in the repair of other tissues. To investigate the importance of HGF in the repair of the small intestine, we evaluated its effect and that of other growth factors in IEC-6 cells, an intestinal epithelial cell line derived from normal rat small intestine. Round "wounds" were made in confluent monolayers of IEC-6 by silicon rubber-tipped steel rods and various cytokines; transforming growth factor alpha (TGF-alpha), transforming growth factor beta1 (TGF-beta1), keratinocyte growth factor (KGF), and HGF, were added. We photographed the repaired monolayers every 24 h and calculated the ratios of areas not covered by cells to initial areas. Cell proliferation with TGF-alpha, TGF-beta, KGF, or HGF was examined in terms of [3H]-thymidine uptake. Finally, we determined c-met (the HGF receptor) mRNA in the IEC-6 cells by Northern blot hybridization. HGF was the most potent of the cytokines in accelerating repair of the damaged monolayer of IEC-6. HGF was also 1.34 times more effective than control the medium for inducing cell proliferation of IEC-6. By Northern blot hybridization, three bands of mRNA bound to c-met cDNA. These results suggest that HGF is important in the repair of the small intestine.

Hepatocyte growth factor protects gastric epithelial cells against ceramide-induced apoptosis through induction of cyclooxygenase-2.

Forced overexpression of cyclooxygenase-2 (COX-2) in intestinal cells has been shown to be associated with resistance to apoptosis. However, the role of physiologically-induced COX-2 in the regulation of apoptosis remains unclear. In the present study, we examined whether hepatocyte growth factor (HGF)-induced COX-2 affects ceramide-induced apoptosis in RGM-1 gastric epithelial cells. An externally applied cell permeable ceramide analogue, C2-ceramide, caused RGM-1 cell death in a dose-dependent manner, whereas an inactive ceramide analogue, C2-dihydroceramide, did not. TdT-mediated dUTP nick end labeling (TUNEL) assay showed that the C2-ceramide-induced cell death was apoptosis. Application of HGF rapidly induced the expression of COX-2, and HGF prevented the apoptotic cell death induced by C2-ceramide. However, the anti-apoptotic action of HGF was antagonized by coapplication of NS-398, a selective inhibitor of COX-2. Thus, these results indicate that COX-2 is involved in the survival signaling from HGF in gastric epithelial cells, and suggest a role for physiologically-induced COX-2 in the protection of the cells from apoptosis.

Hydrogen peroxide-mediated cytotoxicity to cultured colonic epithelial cells.

Reactive oxygen metabolites (ROM) contribute to colonic cellular injury, in certain pathophysiological conditions. We investigated the role of iron and individual metabolites in their cytotoxicity to cultured colonic epithelial cells from adult white rabbits. Reactive oxygen metabolites, enzymatically generated by hypoxanthine/xanthine oxidase, have a direct cytotoxic effect on cultured colonic epithelial cells. This cellular injury was inhibited by catalase but not SOD. Damage was not aggravated by ferrous iron or EDTA-chelated iron. Such damage was prevented by chelating intracellular iron, but not extracellular iron. These results suggest that H2O2 is more toxic to colonic epithelial cells than 02.- and OH. in the extracellular space. H2O2 enter the intracellular space and is converted to the more reactive and harmful OH. leading to cellular injury in the presence of intracellular iron.

The role of cellular migration in the repair process of gastric epithelial cells.

The epithelial cells of stomach are continuously exposed to various toxic agents that may cause mucosal injury. The epithelial lining is rapidly replaced by cells that migrate from the proliferative zone of the gastric gland, to maintain the integrity of the gastric mucosa. Thus, cell migration is an essential part of the early process of gastric mucosal repair. After various forms of gastric injury, mucosal integrity is reestablished by the rapid migration of epithelial cells. However, the cellular mechanisms of the restitution remain unclear to date. In this report, we will review the role of cellular migration in the repair process of gastric epithelial cells in culture. It has been reported that hepatocyte growth factor (HGF) has the potency of acceleration of cellular repair process. In this review, we also report that HGF plays a leading role in the mucosal repair after damage by using a novel cell culture model.

Clinical evaluation of granulocyte/monocyte apheresis therapy for active ulcerative colitis.

BACKGROUND: Recent evidence suggests that peripheral blood granulocytes and monocytes/macrophages have a major role in the exacerbation of ulcerative colitis. AIMS: Our objective was to investigate if selective granulocyte and monocyte adsorptive apheresis with Adacolumn promotes remission of active ulcerative colitis and spares corticosteroid. SUBJECTS: Sixty patients with active ulcerative colitis were studied, of whom 39 had relapsing-remitting ulcerative colitis, 15 had chronic continuous and 6 had their first episode of ulcerative colitis. METHODS: Granulocytapheresis was done with an Adacolumn filled with cellulose acetate beads as apheresis carriers that adsorb FcgammaR and complement receptors bearing leucocytes (granulocytes, monocytes and a small fraction of lymphocytes). Patients received up to 10 Adacolumn sessions over 12 weeks, one session was 60-90 min at 30 mL/min. No additional medication was given. Efficacy was assessed with Seo's activity index (AI) [Seo M, Okada M, Yao T. An index of disease activity in patients with ulcerative colitis. Am J Gastroenterol 1992;87:971-6]. The mean AL was 197.5 and range 154.4-277.7. AI < 150 was considered significant improvement and AI < 100 was considered clinical remission. RESULTS: Of 60 patients, 50 (83.3 %) improved, 14 achieved remission, granulocytapheresis was most effective in steroid-dependent patients. At entry, the mean dose of prednisolone was 15.3 mg/day per patient and was reduced to 3.6 mg/day after 10 sessions. Granulocytapheresis was well tolerated and no serious side-effects were observed. CONCLUSION: Based on our experience in patients with diverse ulcerative colitis disease expression and long-term exposure to conventional drug therapy, we believe that granulocytapheresis is an effective adjunct to conventional medication for promoting remission and sparing steroids in patients with active ulcerative colitis.