Search

The aim of the present study was to evaluate messenger RNA expression of c-met, a hepatocyte growth factor receptor gene, after gastric mucosal injury in rats. Male Sprague-Dawley rats were fasted for 24 hours, received 0.6 N hydrochloric acid (HCl), and served for polyadenylated RNA extraction from the oxyntic gastric mucosa. The transcripts of rat c-met gene were analyzed by reverse-transcript polymerase chain reaction and Northern blotting. Although it was detected even before the HCl administration, the c-met expression increased 6, 24 and 48 hours after the HCl administration. Thereafter, gastric mucosal injury diminished and the c-met expression declined. Hepatocyte growth factor reportedly plays an important role in gastric cell proliferation. The increased c-met expression indicates that this gene may participate in the healing process of gastric mucosa after injury.

Gastric microcirculatory disturbances are involved in the pathogenesis of stress ulcers; however, vasomodulators causing this process are not fully understood. This study was conducted to investigate the role of endothelin 1 (ET-1), a potent vasoconstrictive peptide, in stress ulcers in critically ill patients.Using sandwich enzyme immunoassay, we measured ET-1 content in plasma and the gastric mucosa of 16 critically ill patients with traumatic head injury on admission and of 11 healthy subjects. Gastric mucosal samples were obtained endoscopically. When gastric drainage contained occult blood, endoscopic examination was performed again, and ET-1 concentrations in injured and adjacent normal mucosa were compared.Plasma and mucosal ET-1 concentrations were significantly higher in critically ill patients on admission (6.1 +/- 0.6 pg/ml and 13.8 +/- 1.6 ng/g, respectively) compared with values in control subjects (2.7 +/- 0.4 pg/ml and 8.2 +/- 0.5 ng/g, respectively) (p0.01). The mucosal ET-1 concentration tended to be elevated in patients who had experienced hypoxia compared with those who had not (p = 0.07). In five patients who were again examined endoscopically, the ET-1 concentration in the injured mucosa was significantly higher than that in adjacent mucosa (19.2 +/- 3.2 and 10.1 +/- 1.6 ng/g, respectively; p0.05).These results suggest that endogenous ET-1 plays an important role in the local pathogenesis of stress ulcers, especially those caused by hypoxia.

The aim of this study was to investigate the safety and long-term effects on serum lipid levels of low-dose simvastatin, an inhibitor of 3-hydroxy-3-methylglutaryl coenzyme A reductase, in Japanese patients with moderate primary hypercholesterolemia. We assigned 201 patients (68 men and 133 women; mean +/- SD age, 61.3 +/- 10.2 years) with serum total cholesterol levelsor = 220 mg/dL to receive simvastatin 5 mg each evening; the treatment period was 1 year. Serum total cholesterol, triglycerides, and low-density lipoprotein (LDL) cholesterol levels decreased significantly in response to simvastatin therapy, and the changes were maintained throughout the treatment period. Mean total cholesterol decreased from 269.9 +/- 35.4 mg/dL to 215.2 +/- 34.5 mg/dL (20.3%), triglycerides decreased from 183.0 +/- 110.2 mg/dL to 155.5 +/- 88.5 mg/dL (15.0%), and LDL cholesterol decreased from 180.0 +/- 33.1 mg/dL to 130.1 +/- 35.1 mg/dL (27.7%). Total cholesterol, triglycerides, and LDL cholesterol tended to decline when the pretreatment values were higher; the critical values and the bidirectional changes of the serum lipid levels were 188.1, 109.5, and 91.6 mg/dL, respectively. Although the serum level of high-density lipoprotein cholesterol did not change significantly, it tended to increase more when the pretreatment values were lower; the "critical value" was 70 mg/dL. Nine patients experienced mild adverse events, but none discontinued simvastatin during the 12-month treatment period. We found that low-dose simvastatin therapy is effective in achieving long-term decreases in serum lipid levels and is well tolerated by patients with moderate hypercholesterolemia. Simvastatin therapy may result in normalization of serum lipid levels.

The imbalance between matrix metalloproteinases (MMPs) and tissue inhibitors of matrix metalloproteinases (TIMPs) is considered to be an important determinant of extracellular matrix deposition and breakdown. We measured serum MMP-1, MMP-2, TIMP-1 and TIMP-2 levels using the respective one-step sandwich enzyme immunoassays in 98 patients with chronic hepatitis C treated with interferon beta to examine their clinical significance for assessment of liver histology and to determine whether they can be useful as predictors of the interferon response.Serum TIMP-1 levels showed a positive correlation with the degree of fibrosis (r(s)=0.30, p= 0.004). Serum MMP-2 levels revealed positive relationships with the degree of periportal necrosis (r(s)= 0.32, p=0.002), the degree of fibrosis (r(s)=0.26, p= 0.01) and total score of histological activity index (r(s)=0.24, p=0.02). Serum MMP-2 levels were significantly higher in patients with no response than in those with sustained and transient response (p0.01 and p0.05, respectively), while serum MMP-1 levels did not differ among the three groups. Compared with the levels in sustained responders, the total amounts of serum TIMP-1 were significantly lower in transient responders and non-responders (p0.01 and p0.001, respectively). As for serum TIMP-2 levels, a significant decrease was found in transient responders and non-responders (p0.01). The ratios of serum MMP-2 to TIMP-1 levels were significantly higher in transient responders and non-responders than in sustained responders (p0.001, respectively) even when HCV RNA levels were low in patients with HCV genome subtype 1b or when the HCV genome subtype was 2a or 2b. Sustained response was never found in type 1b patients with ratios of serum MMP-2 to TIMP-1 levels of over 6.0. In logistic multivariate regression analysis, the ratios of serum MMP-2 to TIMP-1 level (p=0.0001), HCV genome subtype (p=0.005) and serum TIMP-2 level (p=0.03) were the independent predictors for sustained response, while serum MMP-2 level (p=0.0006) was the only predictor for no response.Serum MMP-2 and TIMP-1 levels might be useful for estimating the degree of liver fibrosis. The ratio of serum MMP-2 to TIMP-1 levels may serve as a new predictor of interferon response in patients with chronic hepatitis C.

Peptic ulcer in the human stomach causes localized destruction of the gastric wall, which may be associated with focal vascular insufficiency. Endothelin-1, an extremely potent vasoconstrictor peptide, modulates regional blood flow in the vasculature of stomach, suggesting a role for endothelin-1 in peptic ulcer. We examined the relationship among endogenous plasma and mucosal endothelin-1 concentrations and the severity and area of ulcer in 19 patients with gastric ulcers and eight healthy adults. Endothelin-1 concentrations were measured by enzyme immunoassay in plasma and gastric mucosal specimens from ulcer margins, corpus, and antrum. The severity and area of ulcer were assessed endoscopically. Plasma endothelin-1 concentrations in active (P0.01 compared with normal) and healing (P0.05) stages of ulcer were significantly greater than those in normal subjects. Plasma endothelin-1 concentrations, but not mucosal endothelin-1 concentrations in the ulcer margin, were significantly associated with the severity of the ulcer. There was a significant positive correlation between plasma endothelin-1 concentrations and area of ulcer (r = 0.70, P0.01). In conclusion, locally increased endothelin-1 may be an important mediator contributing to the pathogenesis of peptic ulcer.

The heterogeneity of the hepatitis C virus (HCV) genome has been reported to be associated with the effectiveness of interferon therapy. We investigated the correlation of the viral and host factors, including the degree of sequence complexity of the HCV genome for responses to interferon-alpha in patients with chronic hepatitis C.Ninety-seven patients received a 26-week course of recombinant interferon-alpha 2a therapy. The sequence complexity of the envelope 1-2 region was evaluated by polymerase chain reaction-mediated single-strand conformation polymorphism (PCR-SSCP) analysis.Of the 85 patients who completed the treatment, 31 (36%) achieved a sustained response, and 28 (33%) showed a sustained loss of HCV RNA. A low HCV RNA level, determined by the branched DNA probe assay, and serotype group 2 HCV correlated with a sustained response. In patients with serotype group 1 HCV of more than the threshold of the branched DNA probe assay, a band number on PCR-SSCP analysis of more than 2 could be associated with inefficacy of interferon therapy. Multivariate analysis in the 50 patients whose sera were available for all the virologic tests showed that only the HCV RNA level is independently predictive of a sustained response.Determination of the HCV RNA level is most important for predicting the response before interferon therapy. PCR-SSCP analysis may be useful as an additional test for patients with a high HCV RNA level of serotype group 1 HCV.

Long-term storage of liver grafts results in increased adhesion of leukocytes onto the sinusoidal walls. This eventually leads to posttransplant graft damage through disturbances of hepatic microcirculation. Intracellular adhesion molecule-1 (ICAM-1) is known to be involved in attachment of leukocytes. This study was designed to examine whether ICAM-1 participated in the pathogenesis of posttransplant liver injury. Inbred Lewis rats were used as both donors and recipients to avoid immunoreactivity. Donor livers were stored for either 1 or 6 hr in ice-cold Euro-Collins solution and subsequently implanted. Expression of ICAM-1 was examined immunohistochemically. In some rats that received livers stored for 6 hr, the intact IgG (1.0 mg/kg) or the F(ab')2 fragment (0.5 mg/kg) of an anti-ICAM-1 mAb (1A29) was administered via the tail vein immediately after reperfusion of portal blood. In the group receiving livers stored for 6 hr, ICAM-1 began to be expressed on the sinusoidal endothelial cells as early as 15 min after reperfusion of the portal blood. Strong ICAM-1 expression was observed from 2 hr up to 24 hr after reperfusion. In contrast, expression of ICAM-1 was not evident at any time point after surgery in the 1-hr storage group as well as in untransplanted, normal livers. Serum alanine aminotransferase (ALT) levels were significantly higher in the 6-hr storage group compared with those of the 1-hr storage group (1-hr: 171 +/- 9 IU/L; 6-hr: 825 +/- 109 IU/L, P0.05; mean +/- SEM) 24 hr after transplantation. Serum ALT levels were markedly reduced by treatment with the F(ab')2 fragment of 1A29 (247 +/- 34 IU/L, P0.05 vs. 6-hr storage group). This was associated with reduced accumulation of leukocytes in the liver. In marked contrast, treatment with the intact IgG of 1A29 increased serum ALT levels dramatically (5297 +/- 634 IU/L, P0.05 vs. 6-hr storage group) and reduced serum complement. Histological examination revealed focal hepatocellular necrosis 24 hr after surgery in the 6-hr storage group. Treatment with the F(ab')2 fragment decreased the liver damage; in marked contrast, treatment with the intact IgG strikingly aggravated the injury, as characterized by massive necrosis throughout the liver. Liver damage caused by the intact IgG might be related to activation of the complement system by the Fc portion of the antibody. Taken together, these results indicate that ICAM-1 is involved in the mechanism of postoperative liver injury following liver transplantation.

In this retrospective study, we compared the effects of histamine H2-receptor antagonists to those of antacids and anticholinergics in patients with hemorrhagic ulcers with various endoscopic appearances of bleeding.Patients with hemorrhagic ulcers (n = 376) were examined by emergency endoscopy and were treated with 1) antacids and anticholinergic drugs or 2) H2-receptor antagonists.In ulcer patients with oozing or fresh red coagulation, H2-receptor antagonists ceased further hemorrhage more effectively (65.9% of the cases) than antacids and anticholinergic drugs (46.7%). In patients with projectile bleeding, both of the treatments failed to stop hemorrhage. There were no significant differences in favorable outcome in the patients only with old black coagulation between antacid and anticholinergic drugs-treated group and H2-receptor antagonists-treated group (94.4% and 93.8%, respectively).The results suggest that H2-receptor antagonists are more effective than antacids and anticholinergic drugs in patents with peptic ulcer with fresh coagulation or oozing, but not with projectile bleeding or old black coagulation. The results also indicate that endoscopic appearances of peptic ulcer bleeding are good predictors for the effects of medication.

The significance of immunoglobulin (Ig) M antibody to hepatitis C virus core protein (IgM anti-HCV core) was studied in 41 patients with chronic hepatitis C virus (HCV) infection diagnosed by the polymerase chain reaction (PCR). IgM anti-HCVcore was tested with a solid-phase enzyme-linked immunoassay. The results were correlated with clinical features, liver histology findings evaluated by the histological activity index, and virological features such as genotypes and viremic levels assessed by a branched DNA assay. IgM anti-HCVcore was found in 29 (71%) patients, and its occurrence was only related to viremic levels. A significant relationship was observed between viremic levels and IgM anti-HCVcore cut-off index (rs = .42, P.01). Of the eight low viremic (branched DNA-negative) patients, only two (25%) tested positive for IgM anti-HCVcore with a low cut-off index of3, whereas 27 (82%) of the 33 highly viremic (branched DNA-positive) patients had IgM anti-HCVcore (P.01). After a 28-week interferon-alpha course (IFN-alpha), sustained aminotransferase normalization after therapy withdrawal was achieved by only two (13%) of the 16 patients with IgM anti-HCVcore cut-off index3 compared with 11 (44%) of the 25 patients with that3 (P.05). IgM anti-HCVcore cut-off index decreased after therapy in patients who cleared the virus in sera but increased again after reappearance of viremia. These findings suggest that IgM anti-HCVcore may serve as a simple serological indicator of active virus replication and have relevance to the outcome of antiviral therapy.

The redox state of cytochrome aa3 was measured at microspots (20 microns diameter) within the lobule of perfused rat livers, using reflectance microspectroscopy, and the effect of ethanol infusion on sublobular distribution of the redox states was evaluated. A sigmoidal relationship was observed between oxygen delivery and the reduction of cytochrome aa3 in both the periportal and pericentral regions of the liver lobule when the influent O2 concentration was decreased in a graduated manner. This sigmoidal curve was shifted to the more reduced state by ethanol infusion, with ethanol (25-100 mM) increasing the degree of cytochrome aa3 reduction in a dose-dependent manner according to the distance from the periportal region along a sinusoid. This increase was spatially heterogeneous within a liver lobule. These data indicate that ethanol accelerates cytochrome aa3 reduction, with a distinct gradient of reduction along sinusoids but a heterogeneous distribution within the liver lobule.

The lack of a small animal model of hepatitis C virus (HCV) infection has impeded elucidation of the pathogenesis of HCV. The aim of this study was to develop an HCV-expressing animal model by means of cationic liposome-mediated in vivo gene transfer. To examine the feasibility of this strategy, pActLacZ, an expression vector composed of the LacZ gene driven by the beta-actin promoter, complexed with lipofectin, was injected retrogradely into the common bile ducts of adult rats. X-Gal histochemical staining clearly showed that the LacZ gene was expressed in hepatocytes, but not in biliary epithelial cells. Maximal expression was observed at a DNA to lipofectin ratio of 1:4. Based on this observation, pAGS3M091, an expression vector containing the full length of HCV complementary DNA (cDNA) preceded by the beta-actin promoter, was evaluated. Two days after in vivo intrabiliary administration of pAGS3M091 complexed with lipofectin, polymerase chain reaction (PCR) amplification of reverse-transcribed liver RNA demonstrated the 5' and 3' portions of HCV transcripts derived from pAGS3M091. Immunohistochemical analysis showed the HCV core protein in a small number of hepatocytes scattered in the hepatic lobules. We conclude that the full-length HCV genome was successfully expressed in adult rat liver by means of cationic liposome-mediated in vivo gene transfer. This model will be useful for determining the immunopathological role of HCV in vivo.

We investigated the role of gastric mucosal proliferation and mucosal transforming growth factor-alpha (TGF alpha) in the healing of peptic ulcers in human subjects. During endoscopic observation, mucosal biopsy specimens were taken from the ulcer margin for flow cytometric analysis of the mucosal cell cycle and ELISA for mucosal TGF alpha content. Immunohistochemistry confirmed that biopsy specimens contained proliferating gastric epithelial cells, TGF alpha, and the TGF alpha receptor. When the stage of ulcer healing was categorized according to Sakita and Miwa, mucosal cell proliferation was less at active stages and gradually increased during the healing process. The mucosal TGF alpha level at the ulcer margin was not detectable at the A1 stage but increased to within the normal range during ulcer healing. After re-establishment of the epithelial layer at the site of the ulcer, cell proliferation returned to normal and TGF alpha remained within the normal range. Immunohistochemistry showed coexistence of TGF alpha and its receptor within PCNA-positive proliferating cells. These results suggest that TGF alpha and its receptor play an important role in gastric epithelial cell replication. The increase in TGF alpha at the ulcer margin might indicate a role in stimulation of gastric cell replication during ulcer healing.

To evaluate the inflammatory changes during ulcer healing, we measured myeloperoxidase (MPO) activity and the inflammatory cytokine interleukin-8 (IL-8) in the gastric mucosa of 51 patients with gastric ulcers and 5 normal controls. MPO activity was measured by enzyme assay and IL-8 by ELISA, using biopsy samples taken from the ulcer margin and at 3 and 6 cm from the ulcer. Levels of MPO activity were significantly higher than in normal controls and peaked at the A2 (active) stage, and then gradually decreased and returned to basal levels at the S2 (scarring) stage. The area of increased MPO activity around the gastric ulcer was approximately 6 cm in diameter at the A2 stage, and this area decreased as the ulcer healed. IL-8 levels in gastric mucosa also increased significantly at the A2 stage, and the changes paralleled those of MPO activity during ulcer healing. MPO activity correlated well with the IL-8 level.

Hepatitis viruses and alcohol are major causes of liver disease. This study was aimed at investigating the effect of alcohol intake on the replication of hepatitis C virus and the efficacy of interferon therapy. Fifty-three patients who were histologically proved to have chronic hepatitis C were tested. Of these, 16 were diagnosed as habitual drinkers whose cumulative total consumption of alcohol was more than 100 kg or who had consumed at least 60 gm of ethanol daily for at least 5 yr. The quantities of hepatitis C virus RNA in serum were measured with a competitive assay that combined reverse transcription and polymerase chain reaction. The subjects received a 26-wk course of interferon-alpha therapy. There were no significant differences in age and ALT levels between habitual drinkers and nonhabitual drinkers. The titer of viral RNA (logarithmic transformed copy numbers per milliliter of serum) of habitual drinkers (8.5 +/- 0.5) was higher than that of nonhabitual drinkers (7.7 +/- 0.8) (p0.01). Neopterin levels in serum, a marker for the activation of cell-mediated immunity, were lower for habitual drinkers (5.7 +/- 1.5 pmol/ml) than for nonhabitual drinkers (8.1 +/- 5.0 pmol/ml) (p0.01). Eleven of the nonhabitual drinkers (30%) were long-term responders whose alanine aminotransferase levels remained within normal range during the 24 wk after interferon therapy, whereas only one (6%) of the habitual drinkers was a long-term responder (p = 0.06). These findings suggest that alcohol intake increases hepatitis C virus RNA levels in serum--at least in part--impairment of cellular immunity, and modulates the efficacy of interferon therapy.

The effect of an organoselenium compound, ebselen [2-phenyl-1,2-benzisoselenazol-3-(2H)-one], on ethanol-induced liver damage through perturbation of microcirculation was investigated in perfused livers from fed rats. Infusion of ethanol at concentrationsor = 25 mM into the portal vein increased portal pressure in a concentration-dependent manner. Release of lactate dehydrogenase (LDH) into the effluent perfusate was minimal at 30 min; thereafter LDH release began to increase gradually until the end of the experiment (60 min after the onset of ethanol infusion) and was dependent on ethanol concentration. Simultaneous infusion of ebselen at a concentration of 10 or 30 microM with ethanol reduced significantly this ethanol-induced increase in portal pressure by 50 to 75% (P.05) and LDH release by 70% (P.05). When endothelin-1 or phenylephrine was infused into the liver, portal pressure was increased, reaching maximal levels (50 +/- 17 and 46 +/- 7 mm of H2O, respectively) and then decreasing gradually. Ebselen reduced the maximal increase in portal pressure induced by endothelin-1 (18 +/- 2 mm of H2O) by 64% (P.05). In addition, ebselen decreased the maximal levels of portal pressure induced by phenylephrine (8 +/- 1 mm of H2O) by 83% (P.05). These data indicate that ebselen has a vasodilative effect on constriction of hepatic vasculature and diminishes ethanol-induced hepatic damage by offsetting ethanol-induced increase in portal pressure. Thus, ebselen may prove useful for treatment of alcoholic liver injury via improvement of microcirculatory disturbances.

Liver histology in volunteer blood donors positive for serum hepatitis C virus RNA was investigated in relation to hepatitis C virus viremia levels.Twenty-one volunteer blood donors positive for serum hepatitis C virus RNA by polymerase chain reaction were monitored for at least 1 yr by monthly routine liver function tests and underwent liver biopsy. Liver histology findings were correlated with hepatitis C virus viremia levels assessed by a quantitative branched DNA assay.Liver histology showed the features of chronic hepatitis in 20 (95%) patients. Only one of the seven patients with persistently normal aminotransferase levels during follow-up had normal liver histology, and the others had chronic hepatitis. Sera obtained the same day of the liver biopsy were shown to contain hepatitis C virus RNA of 10(5.7)-10(7.6) equivalent/ml (median 10(6.7)). The total histological activity index score (median 2, range 0-15) and the scores of portal inflammation (median 1, range 0-3), lobular inflammation (median 1, range 0-4) and piecemeal necrosis (median 0, range 0-5) correlated with viremia levels (r = 0.64, p0.01; r = 0.60, p0.01; r = 0.48, p0.05; and r = 0.49, p0.05, respectively).These findings suggest that chronic hepatitis is frequently caused by hepatitis C virus infection irrespective of the serum aminotransferase levels, and high level hepatitis C virus replication is a contributory cause for liver injury in volunteer blood donor populations.

Apoptosis is a type of cell death that occurs in acute or chronic hepatitis. It has been suggested to be mediated through Fas antigen. To evaluate the role of apoptosis on liver injury of chronic hepatitis C, we studied the expressions of Fas antigen and hepatitis C virus antigen (core antigen) immunohistochemically. Forty liver biopsy samples from patients with type C chronic liver disease were immunostained for Fas antigen and hepatitis C virus antigen. Expression of Fas antigen was found mainly in the cytoplasm of hepatocytes, and these positive cells were found particularly among infiltrating lymphocytes at the advancing edges of "piecemeal necrosis." The histological activity index showed inflammation of both portal and periportal areas to be more severe in the Fas antigen-positive samples than in the Fas antigen-negative ones (p0.05 and p0.001, respectively). Furthermore, semiquantitative analysis revealed more expression of Fas antigen in the liver tissues with active inflammation than in those without it (p0.01). The prevalence of Fas antigen expression in the hepatitis C virus antigen-positive group was higher than that in the hepatitis C virus antigen-negative group (p0.05). Our findings suggest that Fas antigen expression (apoptosis) plays an important role in inflammation in the hepatitis C virus-infected liver, particularly in the active inflammation of chronic hepatitis C.

UDP-N-acetylglucosamine: beta-D-mannoside beta-1,4-N-acetylglucosaminyl-transferase III (GnT-III) is a key enzyme in the branching of asparagine-linked oligosaccharides, which are present in surface membrane proteins of various tissues and in secretory glycoproteins. The activity of GnT-III was assayed in 2 human hepatoblastoma cell lines, Huh6, which was the parental cell line, and HB611, which was established by transfection of 3 tandem copies of the hepatitis B virus genome into Huh6. A significant difference in GnT-III activity was found between Huh6 and HB611 (136 +/- 18.3 pmol/h/mg versus 6.7 +/- 2.4 pmol/h/mg; mean +/- SD, P0.001), whereas levels of the glycosyltransferases alpha-3-D-mannoside beta-1,4-N-acetylglucosaminyltransferase IV, alpha-6-D-mannoside beta-1,6-N-acetylglucosaminyltransferase-V, and beta-1,4-galactosyltransferase were almost the same in both cell lines. Northern blot analysis indicated that the decreased activity of GnT-III in HB611 was due to the decreased transcript. When HB611 was treated with interferon-alpha, expression of hepatitis B virus-related mRNA decreased, and the activity of GnT-III increased from 8.5 +/- 3.8 to 22.0 +/- 7.2 pmol/h/mg (mean +/- SD, P0.05). This increase was not found in Huh6. Binding capacity with erythrocyte phytohemagglutinin in these cells using fluorescence-activated cell sorter analysis was different, suggesting that the structure of sugar chain on the cell surface might be altered by suppression of GnT-III activity. This is the first report that hepatitis B virus selectively suppressed the GnT-III activity in hepatoblastoma cells.

The technique of generating 3-D images from CT and MR data has made progress. Manipulation of the images has become simpler, the processing time shorter, and the quality better. Clinical application of this technique has been performed for images of brain (5, 7, 8) and bone (1, 2, 6), but only a few studies have been done of abdominal organs (3, 4).

Microcirculatory disturbance is one of the main pathogenetic features of alcoholic liver damage. We have demonstrated that ethanol increased portal pressure, leading to hepatic hypoxia and hepatocellular necrosis. In this ethanol-induced hepatic vasoconstriction, nitric oxide, an endothelial derived relaxing factor, dilated constrictive hepatic vasculature and reduced liver injury by improving the microcirculatory disturbance.

Endothelin-1 (ET-1), a vasoactive peptide, causes a significant rise in portal vein pressure, which is most likely a result of severe vasoconstriction in the liver. In this study, the effect of ET-1 on sinusoidal vasoconstriction in the liver after ischemia and reperfusion was directly investigated using intravital microscopy. In anesthetized female Sprague Dawley rats (200-250 g) ischemia of the median and left liver lobes was induced for 90 min by temporary ligation of the left pedicle. After declamping and a 90-min reperfusion period, the livers were exposed for intravital microscopy. Using a Nikon MM-11 fluorescence microscope (545 nm, 330x), a CCD camera (Cohu FK 6990), and a SVHS video recording unit, the hepatic microcirculation was directly investigated. Besides sham groups, two ischemia groups were studied, receiving ET-1 antiserum (anti-ET-1; 0.5 ml; Peptide Inst., Osaka, Japan) or NaCl 0.9% (0.5 ml) 5 min prior to reperfusion of the liver (n = 6/group). Following a transient drop in the mean arterial blood pressure in the anti-ET-1-treated groups, comparable systemic hemodynamic conditions among the four groups were noted during intravital microscopic assessment at the end of the 90-min reperfusion period. Reduction in the sinusoidal diameters during postischemic reperfusion (7.7 +/- 0.5 microm) was prevented by anti-ET-1 treatment (9.6 +/- 0.25 microm; P < 0.01; mean + SEM) back to control values (9.6 +/- 0.32 microm), while most of other microcirculatory parameters did not show significant differences. The results supported further the role of ET-1 in dysregulation of the sinusoidal vascular tone in the liver, e.g., after ischemia and reperfusion.

This study investigated the role of endogenous nitric oxide in the regulation of hepatic vascular tone in the presence of ethanol. In the perfused rat liver, upon the initiation of ethanol infusion into the liver, portal pressure was increased in a dose-dependent manner, reaching maximal levels in 2-5 min, then decreasing gradually. Simultaneous infusion of N(G)-monomethy 1-L-arginine, a nitric oxide synthesis inhibitor, enhanced this ethanol-induced increase in portal pressure. This enhancement was reversed by simultaneous infusion of a precursor of nitric oxide, L-arginine. These results suggest that endogenous nitric oxide acts as a vasodilator which reduces ethanol-induced vasoconstriction, thus improving the perturbation of hepatic microcirculation by ethanol.

The LEC (Long-Evans with a cinnamon-like color) rat is a mutant of the Long-Evans strain which develops hereditary hepatitis and hepatoma with age. Activities and mRNA levels of N-acetylglucosaminyltransferase III and V (GnT-III and GnT-V, respectively) were determined during hepatocarcinogenesis in this rat using a LEA (Long-Evans with an agouti color) rat as a control. GnT-III activity in LEC rat liver increased after 30 weeks of age, at the stage of chronic hepatitis, to about 2.5-11.5 times the level in LEC rats aged 1-9 weeks. GnT-V activity in the LEC rat liver increased after 20 weeks of age, at the stage of acute hepatitis, to about 1.5-2.5 times the level in LEC rats of 1-9 weeks of age and then remained elevated. Both enzymes showed more dramatic increases in males than in females. The mRNA levels of the enzymes increased in proportion with the enzyme activities. Furthermore, GnT-III and GnT-V mRNAs were highly expressed in both cancer lesion and adjacent tissues. In one case of hepatoma with lymph node metastasis, GnT-III and GnT-V mRNA expression was much higher in the metastatic lesion than in the original cancer. GnT-III and GnT-V levels in the original cancer lesions were similar to those in the cancer lesions of the other LEC rats. These results indicated that expression of GnT-III and GnT-V was induced by chronic liver damage and hepatocarcinogenic changes in the LEC rats.

Portal circulation changes due to the progression of chronic liver disease and portal venous flow are also affected by pharmacotherapy. Thus, noninvasive measurement of effective portal venous flow (EPVF) is highly desirable. We evaluated EPVF under steady-state conditions using echo-Doppler flowmetry combined with per jejunal portal scintigraphy in 32 patients with chronic liver disease. After introduodenal administration of 37 MBq (1 mCi) of 123I-iodoamphetamine, scintigraphy of the pulmonary and hepatic regions was performed and a portosystemic shunt index (SI) calculated. EPVF was calculated as follows: EPVF = PVFx (1-SI/100). EPVF in chronic hepatitis, compensated cirrhosis and decompensated cirrhosis was 12.0 +/- 1.8 ml/min/kg, 10.3 +/- 1.6 ml/min/kg and 8.0 +/- 2.5 ml/min/kg, respectively. There were significant differences in EPVF between all groups, although PVF was similar in each group. EPVF correlated with liver function tests and was a better indicator of liver function than PVF. Measurement of EPVF may provide useful information in the management of patients with chronic liver disease.

We quantified IgG antibodies to structural (core) and nonstructural (C100-3) hepatitis C virus proteins in 42 patients with chronic hepatitis C treated with a 6-mo course of interferon-alpha. Sera were drawn before and at the end of therapy and also 6 mo after therapy withdrawal; they were stored for later analysis of antibodies and serum hepatitis C virus RNA. Sustained virus clearance was observed at the end of therapy and 6 mo after therapy withdrawal in nine cases; it was accompanied with sustained reductions of antibody to hepatitis C virus core protein and antibody to C100-3 protein. A sustained reduction of antibody to hepatitis C virus core protein was specific to sustained virus clearance, although that of antibody to C100-3 protein was not. None of the patients who did not show reductions of levels of both antibodies at the end of therapy displayed sustained virus clearance. Five patients showed hepatitis C virus RNA negativity and normal aminotransferase levels at the end of therapy without reduction of antibody to hepatitis C virus core protein levels. Of these five patients, relapse of hepatitis occurred in four, and viremia was present 6 mo after therapy withdrawal in all cases. These results demonstrate that testing for antibody titers may add information for evaluating virus clearance after interferon therapy.

Antibodies to hepatitis C virus NS3 (C33C), NS4 (5-1-1 and C100-3), and core (C22-3) proteins were tested for by a second-generation recombinant immunoblot assay of serum samples from 42 chronic hepatitis C patients treated with interferon-alpha. Antibodies to 5-1-1 and C100-3 were found in 83% (35/42) before therapy, and those to C22-3 and C33C in all patients. At 6 months after therapy withdrawal in patients with sustained virus clearance, only antibodies to 5-1-1 and C100-3 showed significant decreases (p0.05) (5/5 and 7/7, respectively). Antibodies to C22-3 and C33C decreased less frequently in only 11% (1/9) (p0.01) and 44% (4/9) (p0.05), respectively. Thus, antibodies to 5-1-1 and C100-3 are prone to decrease following a decline in viremia. The antibody to the C22-3 (core) protein was least likely to decrease and may serve as a reliable marker of exposure to the virus.

Anti-HCV prevalence was high in chronic non-A, non-B liver disease. Anti-C100-3 was detected frequently in each category of chronic non-A, non-B liver disease. Antibody to HCV core protein was detected more frequently in 95% with high titers. A significant correlation was observed between the occurrence of high titers of antibody to HCV core and serum HCV RNA positivity. In patients negative by the anti-C100-3 assay, 69% were anti-HCV-positive by the second-generation enzyme-linked immunosorbent assay and also positive by the reverse transcription-polymerase chain reaction. Thus, HCV is a major causative agent of chronic non-A, non-B hepatitis in Japan.

Protein kinase C (PKC) plays an important role in intracellular signaling. PKC consists of a family of at least eight distinct enzymes. How PKC can be responded to liver regeneration and growth is largely unknown. In an in vivo experiment, regenerating liver after partial hepatectomy (Hx) and carbon tetrachloride (CCl4) administration, expression of the PKC alpha-type was increased, with a single peak, after Hx and CCl4 administration. On the other hand, the PKC epsilon-type gene was activated with biphasic peaks following Hx but was elevated with a single peak after the CCl4 administration. In an in vitro experiment using primary cultured hepatocytes, both of the PKC alpha-type and epsilon-type genes could be induced only by combination of insulin and serum. These results demonstrate that both isoforms of PKC are identically regulated by extra- and intracellular stimuli, thereby influencing growth control of hepatocytes.

We examined the changes in serum human hepatocyte growth factor (hHGF), also called "scatter factor," levels after transcatheter arterial embolization (TAE) and partial hepatectomy (PH) in patients with hepatocellular carcinoma and metastatic liver tumor. In most cases, the serum hHGF levels increased transiently 1-3 days after TAE or PH, and then decreased nearly to the basal levels in 1 wk, suggesting that hHGF may play an important role in liver regeneration in humans. The mean amount of increase in serum hHGF levels after PH was 0.38 ng/ml, which was greater than that after TAE (0.16 ng/ml). In three cases of TAE followed by PH, two showed a greater increase in serum hHGF levels with PH than with TAE, but the third showed the reverse result. Because the rate of increase in serum ALT levels did not affect that of serum hHGF levels, the degree of liver injury induced by TAE or PH does not seem to be a determinant in serum hHGF elevation.

We investigated the changes of serum pre-S1 and pre-S2 antigens before and after interferon therapy in 35 carriers with HBeAg, to examine their clinical significance in the therapy. Both antigens were measured quantitatively by solid-phase enzyme immunoassays using monoclonal antibodies specific to each antigen. The titers of these antigens of responders before therapy were significantly lower (p less than 0.001) than those of non-responders (6.23 +/- 1.09 versus 8.65 +/- 2.09 and 4.46 +/- 1.27 versus 6.50 +/- 1.50, respectively). The titers decreased significantly 6 months after interferon therapy in the responders (p less than 0.05), whereas they did not change in the non-responders. Our findings indicate that pre-S1 and pre-S2 antigens have a close correlation with hepatitis B virus replication and are valuable markers for predicting the responsiveness to interferon therapy.

Jejunal varices are not a common manifestation of portal hypertension. This report describes a 46-yr-old man with recurrent massive gastrointestinal bleeding from jejunal varices arising in an area of adhesions between the intestine and the omentum. The bleeding site was identified by exploratory laparotomy. Medical therapy, including vasopressin infusion via the superior mesenteric artery, was of limited success for controlling acute variceal bleeding. However, jejunal resection and anastomosis resulted in complete resolution of the bleeding, and the patient has experienced no recurrent bleeding over a 3-yr follow-up period. A review of the literature shows that this syndrome is characterized by portal hypertension, generally due to liver cirrhosis; frequently, there is a history of abdominal surgery, and the syndrome presents with hematochezia but without hematemesis. Accurate preoperative diagnosis is often difficult. We propose that bleeding from jejunal varices, though uncommon, should be considered under such clinical conditions.

Although many authors reported that proton pump inhibitor showed mucosal protective activity, the precise mechanism is still unclear. In this study, we investigated the effect of proton pump inhibit on gastric mucosal hemodynamics, using a reflectance spectrophotometry system. The proton pump inhibitors had no effect on the gastric mucosal hemodynamics prior to the state of hemorrhagic shock state. However, during the hemorrhagic shock, proton pump inhibitors significantly inhibited the decrease in tissue oxygenation and significantly reduced ulcer formation. In conclusion, proton pump inhibitors maintained tissue oxygenation resulting in reduction of ulcer formation.

We estimated HCV RNA in serum of 20 patients with chronic hepatitis C during interferon therapy using a combination assay of reverse transcription and polymerase chain reaction (RT-PCR). RNA was extracted from 200 microliters of serum. The primers were chosen from the NS3 region of prototype HCV RNA sequence. After 40 cycles of PCR, the products were analyzed by Southern hybridization. HCV RNA was detected in 18 of 20 (90%) patients before therapy. In cases with improvement of serum GPT level, HCV RNA became undetectable at 4 weeks of the therapy. However, HCV RNA reappeared within 4 weeks after the therapy in cases with relapse. In the no response group, HCV RNA did not disappear during the therapy. Interferon therapy is beneficial in improvement of viremia of HCV, and the detection of HCV RNA in serum is useful to evaluate the antiviral effect of interferon.

Gastrointestinal hemorrhage from stress ulceration is a life-threatening complication in a critically ill patient. We retrospectively studied 471 patients admitted to the Department of Traumatology of our hospital who developed shock in their clinical course. Forty-two patients (8.9%) developed gastrointestinal hemorrhage, most within 1 wk (76.2%). The hemorrhage lesion usually was located in the corpus of the stomach. The mortality rate of the shock patients with gastrointestinal hemorrhage was 33.3% (14 of 42). Comparison of different types of shock showed that the incidence, severity, and mortality rates of gastrointestinal hemorrhage were significantly higher in septic shock patients than in hemorrhagic shock patients. These findings show the importance of preventive therapy against progression of early mucosal damage and development of gastrointestinal hemorrhage in shock patients, especially those with sepsis.

To evaluate the morphological changes in chronic liver disease non-invasively we used the frequency-domain analysis of ultrasound. We assumed that liver tissue is a collection of semi-regularly arrayed small scatterers of ultrasound. We applied cepstral analysis to ultrasonic waveforms and evaluated the periodicity of scalloping of the power spectrum, caused by an interference effect among liver scatterers of a given spacing. In a phantom experiment, we could correctly measure the space among scatterers of a phantom consisting of a regularly spaced distribution of nylon filaments. In an experiment involving a freshly excised porcine liver, the distribution of the space among scatterers (SAS) determined on cepstral analysis reflected the size distribution of the hepatic lobules. In a human study, two statistical parameters, mode and kurtosis, were determined from the distribution of the SAS for each patient. In a previous study, we collected ultrasonic data during peritoneoscopy with a 10 MHz transducer directly contacting the liver and showed that we could evaluate morphological changes in chronic liver disease using cepstral analysis. In the present study, we obtained data in two ways: with a 3.5 MHz transducer through the abdominal wall and with a 10 MHz transducer directly contacting the liver during peritoneoscopy. The mode and kurtosis of the SAS obtained with the 3.5 MHz transducer were both well correlated with the mode and kurtosis of the SAS obtained with the 10 MHz transducer. These results suggest that we can evaluate morphological changes in chronic liver disease non-invasively.

We investigated expression of HBV markers in chronic liver disease positive for antibody to HCV (anti-HCV). Sera from 107 patients with chronic non-A, non-B liver disease, 65 HBs antigen carriers with chronic liver disease and 14 asymptomatic HBV carriers were tested for the presence of anti-HCV. Anti-HCV was detected in 83 (78%) patients with chronic non-A, non-B liver disease, irrespective of the past history of blood transfusion, and anti-HCV prevalence was similar in each category of chronic liver disease. Fifty-three (64%) out of these 83 sera positive for anti-HCV has also antibodies to HBV. Anti-HBc antibody was detected frequently in liver cirrhotics with hepatocellular carcinoma than in chronic persistent hepatitis, chronic active hepatitis and cirrhotics without hepatocellular carcinoma. In addition, titers of anti-HBc antibody were significantly higher in cirrhotics with hepatocellular carcinoma than in the other groups. On the other hand, anti-HCV was detected in 7 out of 65 patients with HBV-related liver disease. Four out of these 7 were patients with HBV-related hepatocellular carcinoma. Anti-HCV was detected in none of asymptomatic HBV carriers. These findings suggest that infection with both HBV and HCV is likely to cause more serious liver disease than infection with a single agent.

The incidence of gastrointestinal bleeding was studied in 433 patients with head injury. This complication has a strong correlation to the severity of the injury. Other factors such as shock, arterial pO2 and glucocorticoid administration were not related to incidence of gastrointestinal bleeding. Forty per cent of patients developed bleeding within 48 hours and approximately and additional 40% within one week. Only nine patients required more than 1,000 ml of blood, and eight of these died and one requiring emergency abdominal surgery lived.