Effect of lipopolysaccharide, lipid A and polysaccharide on phagocytic cell function

Abstract

During phagocytosis of opsonized particles by polymorphonuclear leukocytes (PMN) oxygen is consumed, hexose monophosphate shunt activity is increased and hydrogen peroxide and activated toxic oxygen species are produced. This metabolic burst is accompanied by the generation of chemiluminescence (CL) [1]. The function of PMN can be influenced by a variety of substances such as bacterial cell wall components. There is considerable information available concerning the effects of lipopolysaccharide (LPS) on phagocytes [2]. However, results of various authors are contradictory and these conflicting data prompted us to reinvestigate the effect of LPS preparations from different Escherichia col• strains on PMN functions. The use of mutant strains of E. col• is of significant value in elucidating the biological significance of LPS. E. col• J5, an UDP-gal-epimerase-deficient mutant strain of E. col• 0111B4, contains LPS with a relatively high content of lipid A compared to the parent strain, as it lacks the polysaccharide side chains. PMN were isolated by dextran sedimentation, differential density eentrifugation, and NH4CI lysis of contaminating red blood cells [3]. LPS of E. col• 0111B4 was prepared by phenol-water extractions [4] and LPS of E. col• J5 was isolated according to the phenol-chloroform-petroleum ether described by GALANOS et al. [5]. Phagocytosis was studied using 3H-thymidine-labeled bacteria [3]; generation of CL was measured in a liquid scintillation counter in the out-of-coincidence mode [6]; superoxide production was assayed by the reduction of ferricytochrome c [6]; oxygen consumption was measured with a Clark electrode [6] and chemotaxis of PMN under agarose was determined according to NELSON et al. [7]. PMN (107/ml) were incubated for 18 hours with and without LPS J5 or LPS 0111B4 at 37~ 5% CO 2. After this incubation the PMN were centrifuged, resuspended in fresh medium and tested for phagocytic, metabolic and chemotactic activity (see Table 1). PMN preincubated with LPS J5 phagocytized less Staphylococcus aureus (opsonized in 5% human pooled serum) than control PMN and neutrophils preincubated with LPS 0111B4 phagocytized more opsonized bacteria than control cells. The chemotactic activity of PMN preincubated with LPS J5 was zero. Control PMN or cells preincubated with LPS 0111B4 could still migrate over a small distance. PMN which had been preincubated with LPS J5 produced less superoxide than control PMN and PMN preincubated with LPS 0111B4 produced more superoxide than control PMN when stimulated with preopsonized staphylococci. Thus preincubation of PMN with LPS J5 causes loss of their ability to migrate and diminishes their phagocytic and metabolic activity. The part of LPS responsible for the diminished PMN function is possibly the lipid A region of LPS as incubation of PMN with lipid A for 18 hours caused a marked loss of their ability to produce superoxide. When PMN were directly stimulated with LPS or LPS derivatives and superoxide production measured, cells stimulated with LPS J5 or lipid A produced twice as much superoxide as PMN stimulated with LPS 0111B4 or polysaccharide (17.0 • 4.2 nmol vs. 7.1 • 3.9 nmol O-~/107 PMN/30 minutes;p < 0.01). Similar results were seen when the generation of CL and the amount of oxygen consumed by PMN upon direct stimulation with LPS J5 or LPS 0111B4 were measured. When thiourea, a scavenger of radicals, was added (100 raM) during the 18 hour incubation with LPS J5 the capacity of the PMN to produce superoxide remained normal. We therefore conclude that the effects of LPS found on phagocytic cells are dependent on the lipid A-polysaccharide ratio of the LPS used. When more lipid A is present phagocytic cell function is diminished by the production of toxic radicals upon stimulation by the lipid A portion of LPS. Thus after an 18 hour incubation with LPS J5 or lipid A ehemotactic, fagocytic and metabolic activity are decreased compared to control PMN.