Twenty-one patients with typhoid fever were studied to evaluate the presence of endotoxin in peripheral blood and its relationship to the incidence and features of hepatic dysfunction which may occur during this disease. The limulus test for endotoxin was positive in the plasma samples of all patients prior to treatment. Liver dysfunction, as assessed by fasting and postprandial serum bile acid levels and by standard biochemical tests, occurred in 90% of patients. In seven, the injury was purely cholestatic (elevation of postprandial serum bile acid levels, alone); in 12, it was of mixed cholestatic-hepatocellular type (elevation of both serum bile acids and aminotransferase levels). After recovering, the limulus test was negative and liver function tests returned to normal values in all patients. The results demonstrate that endotoxemia is present in patients with typhoid fever. In addition, since endotoxin can impair bile secretion, our results suggest that endotoxin may have a pathogenetic role in the development of liver injury during typhoid fever.

We evaluated whether the exposure of Staphylococcus aureus to a subinhibitory concentration of penicillin and cefamandole could modify its susceptibility to rat serum factors and to the phagocytic activity of the isolated and perfused rat liver. Control or sub-MIC treated bacteria were added to the circulating medium which contained homologous serum, and the disappearance of bacteria from the perfusate and their recovery in the liver was determined during the 10 min experimental time. Sub-MIC treated bacteria were more susceptible to the bactericidal activity of serum present in the perfusate. However, the clearance rate of bacteria by the liver was decreased for penicillin-treated organisms and unchanged for cefamandole-treated bacteria. The data suggest that beta-lactam antibiotics at sub-MIC levels may modify S. aureus susceptibility to host defense mechanisms.

Using isolated and perfused rat livers we found an inhibitory effect on intracellular killing of hepatic macrophages versus E. coli, in presence of therapeutic levels of amphotericin B (5 micrograms/ml). Since an interference of this antibiotic with phagocytic functions of human neutrophils has also been reported, we suggest that amphotericin B may exert a toxic effect on intraphagocytic microbicidal function, possibly by an interaction with membrane sterols.

Growth of S. aureus in the presence of subinhibitory concentrations of cefamandole alone (at 1/5 the MIC) or with gentamicin (both at 1/5 the MIC) produced large, globular cells with reduced viability (50%) as compared to untreated bacteria. Gentamicin (1/5 the MIC) did not modify bacteria and also reduced viability of inoculum (57% vs. controls). Cefamandole treated bacteria were more susceptible to the rat serum bactericidal activity then control bacteria. The growth of S. aureus in the presence of cefamandole plus gentamicin did not modify further bacterial susceptibility to serum opsonins. The data suggest that even at subinhibitory concentrations cefamandole may exert an antibacterial effect by acting in cooperation with serum factors.

Our purpose was to study whether treatment of Escherichia coli with subinhibitory concentrations of either cefamandole or gentamicin could change bacterial susceptibility to the serum bactericidal effect and to the phagocytic and killing activity of the rat liver reticuloendothelial system. Bacteria were grown overnight with 1/5 or 1/10 of the MIC of each antibiotic. At one-fifth of the MIC, cefamandole induced filamentous elongated bacteria whose viability was decreased by 75%. The susceptibility of control and antibiotic-treated bacteria to serum was tested by measuring the survival of organisms exposed to different concentrations of rat serum in vitro. Susceptibility of bacteria to hepatic macrophage activity was tested by following the hepatic clearance of bacteria after they were added to the perfusate of the isolated rat liver. E. coli treated with subinhibitory concentrations of cefamandole or gentamicin appeared somewhat more resistant to the lytic activity of serum at a concentration of 4%, but not at 20%. Bacteria treated with 1/5 or 1/10 of the MIC of cefamandole or with 1/5 of the MIC of gentamicin were significantly more susceptible to phagocytosis and to the bactericidal activity of liver macrophages. Cefamandole appeared more potent than gentamicin in inducing these effects. The results suggest that subinhibitory levels of antibiotics may alter bacterial cell surface (cefamandole) or may impair the expression of antiphagocytic material (gentamicin), thus favoring phagocytosis and killing by macrophages. Our study provides evidence that antibiotics at subinhibitory concentrations may cooperate with host defence mechanisms against bacterial infections.

Incubation of E.coli and S. aureus with subinhibitory concentration (1/5 MIC) of cefamandole modified bacterial morphology and resistance to host defence mechanisms. In fact, cefamandole induced filamentous forms of E. coli, when added to the perfusing medium of the isolated rat liver system, were phagocytized at a much fortes rate then control bacteria, but appeared less sensitive to serum bactericidal activity. In contrast, S.aureus, after exposure to the antibiotic, was more sensitive to the bactericidal activity of serum then controls, while treated and untreated cells were phagocytized at the some rate. The data suggest that even at low doses some antibiotics may alter bacterial structure and increase their susceptibility to host factors.

Ionophore A23187, an agent that promoves Ca2+ influx into cells, significantly increases enzymatic leakage of AST and LDH into the medium by perfused rat livers. This work provides evidence that Ca2+ dependent hepatotoxicity can be studied conveniently in vitro whit isolated and perfused rat livers, since this model respects integrity of organ and duplicates many its in vivo functions.

The effect of chronic ethanol feeding (3 weeks) on the phagocytic and the bactericidal activity of hepatic RES versus viable Escherichia coli was studied using the isolated rat liver perfused with a serum-containing medium. Controls or ethanol-fed animals were used as liver or serum donors. The bactericidal activity of serum from ethanol-fed rats was similar to controls and accounted for the disappearance of nearly one-tenth of the bacterial inoculum from the system. When control livers were perfused with a medium containing serum from ethanol-fed animals, phagocytosis was comparable to controls while intracellular killing was greatly reduced. When livers were isolated from ethanol-fed rats, phagocytosis was significantly depressed and no killing occurred, irrespective of the source of serum. Levamisole was able to restore the macrophage activity depressed by ethanol. Our data indicate that the direct effect of ethanol on hepatic macrophages plays a major role in reducing the bactericidal activity of hepatic RES, although alterations of serum factors may contribute to an ineffective preparation of bacteria for intracellular killing.

Endotoxemia, measured by Limulus amebocyte lysate (LAL) assay, was found to be present in 34 (46%) out of 72 patients with liver cirrhosis and in 7 (22%) out of 32 patients with chronic active hepatitis (CAH). In cirrhotics, no difference in the alteration of liver function tests and renal function was found between the two groups. However, 18 months mortality was higher in the group with endotoxemia in respect to the group without endotoxemia (p less than 0.05). In CAH patients, the Limulus-positive group showed a higher level of serum gamma-globulins, compared to the Limulus-negative group (p less than 0.005). Moreover, CAH patients with a positive LAL test showed marked histological activity and bridging necrosis more frequently than those with a negative test. This suggests that in these patients the appearance of endotoxemia may indicate a more advanced stage of the disease.

The effect of acute and chronic endotoxin (LPS) treatment on the erythromycin estolate (EE) induced cholestasis, was studied using the isolated perfused rat liver. Addition of EE markedly reduced bile and perfusate flows in livers from control rats but did not alter these parameters in livers from endotoxin pretreated rats or in vitro treated with LPS. We suggest that changes in membrane organization induced by LPS may alter the diplay of EE toxicity.

Hepatic RES is depressed in chronic alcholic intoxication and we have previously observed a significant reduction of killing of E.coli by the Kupffer cells of the isolated rat liver. In the present study rats treated with ethanol for 3 weeks, were given levamisole and then the clearance of E. coli, by the isolated liver, was followed. Levamisole 10mg/kg i.p. for 3 days, restored the depression of intracellular killing induced by ethanol. At the dosage of 2,5 mg/kg of levamisole this effect was not present. Changes of humoral factors were not involved in the phenomenon because sera from untreated anaimals were present in the perfusate for all sets of experiments.

The effects of chronic ethanolic intoxication on the phagocytic and bactericidal activity of the hepatic reticuloendothelial system against viable Escherichia coli were examined using be in vitro perfused at liver system. Livers were isolated from control and ethanol treated animals (3 week) and perfused with a red cell-free medium containing 4% homologous serum from control rats. After addition of E. coli (2 X 10(7) cell/ml) to the perfusing medium hepatic clearance of bacteria was measured by determining the recovery of viable cells in the perfusate, liver and bile after 30 min experimental time. Livers from ethanol treated animal showed a slightly lower phagocytic activity respect control animals, but were not able to kill phagocytized bacteria (20% intraphagocytic killing occurred in control livers). Our date suggest that the higher susceptibility toward infections with gram negative bacteria during chronic ethanol intoxication might be mediated by a impairment of functional activity of hepatic macrophages which may be independent of changes in serum opsonins.