Contribution of exogenous ethanol and endogenous substrates to acetaldehyde-adduct formation in liver injury.

Acetaldehyde was found to be transported in the blood mainly bound reversibly to 2 components of the red blood cells (RBC): a) hemoglobin, which provides binding of a high affinity, but low capacity, and b) a non-protein component (presumably cysteine), which has a lower affinity but a higher capacity. In alcoholics, the main increase in RBC-acetaldehyde binding occurred in the protein-free fraction, in association with a marked increase in RBC-cysteine. Elevated RBC-acetaldehyde persisted for at least 2 weeks after alcohol withdrawal in 83% of the blood samples from alcoholics. Even in the absence of alcohol consumption, liver injury increased acetaldehyde from sources other than ethanol. This was associated with high serum antibody titers against acetaldehyde adducts. In an animal model of non-alcoholic liver cirrhosis, a target protein for the formation of acetaldehyde adducts appears to be procollagen type I.

Acetaldehyde-collagen adducts in CCl4-induced liver injury in rats.

Circulating AC levels as well as antibodies against AC-protein adducts are increased in non-alcoholic liver injury. To identify the adducts, we used rats with CCl4-induced cirrhosis. Liver subcellular fractions were analyzed by immunochemical staining of protein slot blots and of electrophoretically separated proteins, transferred to nitrocellulose, using AC-protein adduct-specific antibodies. One reactive protein of about 200 kD was detected in the liver soluble fraction and in the cytosol of isolated hepatocytes and, to a lesser extent in the liver microsomes of CCl4-treated rats; in control animals, this reactivity was much weaker. The immunopositive AC adduct co-migrated with the beta 1,2 dimer of rat collagen type I; it was sensitive to digestion by a highly purified collagenase and also reacted with anti-rat collagen type I-specific IgG. In addition, comparison of peptides of the CNBr-digested, immunoprecipitated AC adduct with those of rat collagen type I revealed a high degree of similarity. Thus, AC adduct formation occurs in liver injury of non-alcoholic origin, and a target protein appears to be related to collagen type I, most likely the procollagen precursor.

Changes in carbohydrate-deficient transferrin levels after alcohol withdrawal.

Sequential serum levels of carbohydrate-deficient transferrin (CDT) were determined in 72 alcoholics at various intervals during detoxification. Before treatment, 57 patients (79%) had increased CDT values (Group A), whereas in 15 individuals (21%) (Group B), CDT levels were within the normal range. In 51 Group A patients, CDT decreased progressively after cessation of alcohol intake (half-life, 16 +/- 5 days), but fluctuated and remained abnormal in the remaining six. Nine Group B patients maintained normal CDT values throughout the follow-up period, but slightly or moderately increased levels were recorded on one occasion in the other six Group B subjects. Patients whose CDT levels had reached normal values after treatment, showed a recurrent increase in CDT after a relapse. gamma-Glutamyl transferase activities, which were elevated in 56% of Group A and in 80% of Group B alcoholics, showed a decrease after cessation of alcohol consumption in most patients with initially elevated values (Group A, 30 of 32; Group B, 10 of 12). Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activities, as well as mean corpuscular volumes (MCV) were normal in the majority of patients. CDT/total transferrin ratios correlated positively with CDT levels. CDT proved to be the most sensitive marker for chronic alcoholism (79%), whereas GGT activity levels were more useful only in patients with normal CDT levels before alcohol withdrawal. In the assessment of treatment outcome, the combination of CDT and GGT as markers yielded a sensitivity of 95%.

Formation of acetaldehyde adducts with ethanol-inducible P450IIE1 in vivo.

Hepatic microsomes, obtained from rats pair-fed liquid diets supplemented with either ethanol or an isocaloric amount of carbohydrates (for 4 weeks), were subjected to crossed immunoelectrophoresis. Anti-acetaldehyde adduct-specific immunoglobulin reacted on the protein blots with a single major 52,000 dalton polypeptide. This same protein was recognized by antibodies specific for P450IIE1, an ethanol-inducible P450 isozyme. Furthermore, a single protein, also reactive with anti-P450IIE1 IgG, was isolated from liver microsomes of ethanol-fed rats by immunoaffinity chromatography on Sepharose-conjugated anti-acetaldehyde adduct IgG. These results indicate that P450IIE1 is a target protein for acetaldehyde binding in liver microsomes in vivo.

Carbohydrate-deficient transferrin, a marker for chronic alcohol consumption in different ethnic populations.

Serum levels of carbohydrate-deficient transferrin (CDT) were determined in a racially mixed population of 107 alcoholics, 18 healthy, nonalcoholic control subjects, 62 abstinent alcoholics, and in 64 Caucasian patients with various nonalcoholic liver diseases. The upper limit of normal CDT levels was 80 mg/liter (2 SD above the mean). CDT values exceeding this level were found in more than 80% of Black, Puerto Rican, and Caucasian alcoholics who had consumed greater than or equal to 50 g of alcohol/day for 1 month or longer prior to testing. Puerto Rican alcoholics had higher CDT values than the Black and Caucasian ethnic groups; however, these differences were significant only when compared to the Black population. Of 64 patients with nonalcoholic liver diseases, one individual with chronic active hepatitis (CAH) with an alcohol consumption of 20 g/day, and 10 of 26 subjects with primary biliary cirrhoses (PBC), who claimed to consume either no or only occasional moderate amounts of alcohol, had CDT levels ranging from 81 to 144 mg/liter. Seven of these individuals were in advanced stages of PBC. Total transferrin levels were variable and not significantly different in all subject groups studied. CDT/total transferrin ratios were increased in most patients with abnormal amounts of CDT, and there was a significant correlation between these ratios and CDT levels in all study groups. Serum enzyme parameters as well as red blood cell mean corpuscular volumes did not correlate with CDT values.(ABSTRACT TRUNCATED AT 250 WORDS)

The role of alcoholism and liver disease in the appearance of serum antibodies against acetaldehyde adducts.

We recently presented preliminary data indicating the presence of antibodies against acetaldehyde adducts in sera of over 70% of alcoholic patients. To assess the respective roles of liver disease and alcohol consumption as well as the specificity of this immune response, 141 patients in various stages of alcoholic and nonalcoholic liver diseases were tested by a hemagglutination assay. Sixty-three (73%) of 86 alcoholics had antibody titers above control levels (p less than 0.0001). Alcohol consumption of these individuals was significantly higher (p less than 0.001) than that of those alcoholics with normal titers. Twenty-two patients (39%) with nonalcoholic liver diseases also had elevated levels of antibodies against acetaldehyde adducts (p less than 0.0005); of these, 8 had primary biliary cirrhosis (7 in Stages III and IV), 9 had chronic active hepatitis (6 with cirrhosis) and 5 had acute (virus- or drug-induced) hepatitis. Antibody titers did not correlate with levels of transaminase or alkaline phosphatase activity, nor with bilirubin, and albumin. However, in 52 alcoholics and in nonalcoholic patients with biopsy-confirmed liver disease, the highest titers were seen in the more advanced stages of liver damage. Thus, in addition to alcohol consumption, severity of liver disease may play a role in the appearance of circulating antibodies against acetaldehyde adducts.

Humoral immune response to acetaldehyde adducts in alcoholic patients.

Sera of alcoholic patients were found to contain antibodies to acetaldehyde adducts as determined by immunodiffusion. Differences in antibody levels were determined by hemagglutination with acetaldehyde adduct-conjugated red blood cells. Anti-acetaldehyde adduct immunoglobulin titers in 21 healthy, non-drinking individuals ranged from 10-80, whereas 25 of 34 alcoholics had titers of 160 or above (P less than 0.001). Rabbit-polyclonal and mouse-monoclonal antibodies against acetaldehyde adducts did not agglutinate red blood cells of alcoholic patients, indicating that no detectable acetaldehyde-altered epitopes are present on erythrocyte membranes. These results suggest that acetaldehyde-induced immunogenic determinants can initiate a humoral immune response which may differentiate alcoholics from non-alcoholics.

Elimination of the yeast Candida parapsilosis from lymphoid cells and monolayer cells in culture.

In our laboratory, airborne yeast contaminants of cell cultures have consistently been of the genus Candida (species Candida parapsilosis), which are difficult to control with fungicidal agents. To salvage cell lines that show the presence of this fungus, two effective methods may be employed. In early stages of infection, the addition of activated mouse peritoneal macrophages (5 X 10(5) cells/ml) to the culture medium containing 5 micrograms Fungizone /ml eliminates all spores by phagocytosis. More heavily contaminated cultures can be depleted of fungi by density centrifugation on a layer of 38% Percoll. Remaining single spores, often not detectable by light microscopy, can be removed by the addition of macrophages (2 X 10(5)/ml) and Fungizone (5 micrograms/ml) to the culture medium. Contaminated monolayer cells can be freed of blastospores by several washes with balanced salt solution and subsequent culturing for 4 d in medium containing 10 micrograms Fungizone /ml without any toxic effects to the cells. These procedures can rescue valuable cell lines and hybridomas that would otherwise be lost.

Liver-specific and shared cell membrane antigens. Studies by light- and electron microscopy.

Liver-specific and shared saline-insoluble cell surface antigens were localized by immunofluorescence as well as by light- and electron microscopic immunoenzyme techniques. Antisera against purified mouse liver cell membranes were surface membrane but not organ-specific. Variable quantities of shared antigens were present in endoderm- and mesoderm-derived organs but not in ectodermal nerve tissue. Species crossreactivity was observed for the rat. Repeated absorption produced liver-specific antisera that reacted with antigenic sites distributed along the entire hepatocyte and sinusoidal cell surfaces. For the precise localization as well as the detection of low concentrations of both liver-specific and nonspecific antigens, the ultrastructural visualization of reactive sites proved essential.

Cell membrane sialoglycopeptides of corticoid-sensitive and -resistant lymphosarcoma P1798.

Cells of corticoid-sensitive (CS) and corticoid-resistant (CR) lymphosarcoma P1798 were labeled in vivo with either [14C]- or [3H]fucose before and after treatment with 9-alpha-fluoroprednisolone (9-FP). Labeled glycopeptides, derived from isolated, Pronase-digested cell membranes of both tumors were separated by gel filtration on Bio-Gel P-10 by a double-label technique. Elution profiles of CS and CR fractions showed significant differences in early eluting material. Desialylation of glycopeptides by neuraminidase lowered the molecular weight of both CS and CR fractions, and altered 3H:14C ratios indicated that CS and CR sialoglycopeptides are different. 9-FP treatment for 7 hr increased the density of isolated P1798-CS and -CR cell membranes. All CS and CR glycopeptides from treated tumors eluted faster than did those of untreated preparations. Both CS and CR sialoglycopeptides were altered, although differences in CS and CR profiles persisted. Histochemical investigations indicated that negative charge, present on surfaces of untreated CS cells, is lost between 6 and 8 hr after exposure in vivo to 9-FP. CR cells had no or few anionic sites on their surfaces before and after steroid administration. We demonstrated that glycopeptides of both CS and CR tumors contain sialic acid, although only CS cells carry a surface-exposed negative charge that is lost after 9-FP treatment. Glucocorticoids alter both P1798-CS and -CR sialoglycopeptides, but the consistent differences between their chromatographic patterns suggest that steroid-induced changes in cell membranes of the two tumors are not identical. Cell death or survival of glucocorticoid-treated P1798 cells may, therefore, be influenced by specific structural characteristics involving cell surface sialoglycoproteins.