Interaction between lipopolysaccharides and host leucocytes

Abstract

Gram-negative septic shock is a life threatening disease associated with high mortality, particularly in immunocompromised patients. The mechanism by which Gram-negative bacteria induce septic shock is complex, with little direct experimental evidence to pinpoint a single toxic factor. Endotoxin, a lipopolysaccharide (LPS) outer membrane component found in all Gram-negative organisms, has been implicated as a primary mediator of the toxicity associated with Gram-negative infection. When purified from organisms and injected into experimental animals, this substance induces many of the same pathophysiological changes observed during Gram-negative infection. Based on an increasing body of data, it is now felt that many of the symptoms of septic shock are attributable to the action of macrophage-derived mediators released following exposure to LPS.

The aim of this project was to investigate the interaction of various forms of LPSfree, membrane associated and as a component of whole bacteria- with mammalian cells to determine the cell surface molecules involved and to study the spectrum of mediators released.

Initially the interaction of various forms of Salmonella minnesota Ra LPS with mouse peritoneal macrophages (M0s) was investigated. Bacteria were labelled with fluorescein isothiocyanate (FITC) and the binding to host cells measured by flow cytometry. The interaction between bacteria and cells was shown to be dose and time dependant, maximal at 37° C and required divalent cations. Varying the pH from 6 to 8 had little effect on the binding. M0s elicited by intraperitoneal injection of thioglycollate bound more bacteria than cells obtained from control animals. To investigate the role of LPS in the interaction of bacteria with M0s, different forms of LPS were used to try to inhibit binding. Pure free LPS and LPS attached to bacterial outer membranes extracted from S. minnesota inhibited the binding of bacteria to M0s but whole unlabelled bacteria (5. minnesota) failed to inhibit the binding of labelled bacteria to the M0s. This finding led to a microscopic study of the effect of FITC labelling of bacteria on the binding to M0s. FITC labelling enhanced the binding of bacteria to the M0s as compared to the binding of unlabelled bacteria.

The production of nitric oxide (NO) and other reactive nitrogen intermediates by cytokine-activated rodent cells is an important component of antimicrobial and/or antineoplasic activity of these cells. The pathway involves the oxidation of arginine to citrulline with concomitant release of NO by an inducible form of NO synthase. Although human cells such as hepatocytes and endothelial cells have been shown to express inducible NO synthase, the presence of such a pathway in human monocytes/macrophages has been questioned by many investigators and is the subject of great controversy. Therefore an attempt was made to investigate whether human monocytes/MOs produce NO in response to different forms of LPS. NO production was assayed by measuring nitrite, a stable metabolic product of NO, in macrophage culture supernatants. Preliminary work was performed using MOs from several mice strains and different forms of LPS. An immature human MO cell line (THP-l) was used to investigate NO production in humans. Free LPS failed to produce any measurable amounts of NO from human cells, but produced small amounts of NO from mouse cells. LPS attached to outer membranes and whole bacteria produced significant amounts of NO from both human and mouse cells. Human peripheral blood monocytes also produced significant amounts of NO when stimulated with whole bacteria and LPS attached to outer membranes.

Macrophages are the principle source of the endotoxin-induced mediator tumour necrosis factor (TNF). This cytokine is pyrogenic and when infused into animals v produces all the clinical and pathological features of septic shock. All forms of LPS used stimulated THP-1 cells to release TNF as measured by a L929 bioassay.

The characteristics of the production of TNF and NO suggest that a novel pathway independent of CD 14 and lipopolysaccharide binding protein might be involved.