BRE 20961 Systemic hypoglycemia following central injection of endotoxin in mice SHIMON AMIR and MICHAL HAREL Department of Isotope Research, The Weizmann Institute of Science, Rehovot (Israel) IAccepted March 12th, 1985) Key' words: endotoxin - - hypoglycemia - - CNS - - polymyxinB - - mouse Intracerehroventricular microinjection of endotoxin in mice resulted in powerful hypoglycemia.The effect was reproduced by the biologicallyactive moiety of endotoxin, lipid A, and prevented by coadministration of the polycationicpeptide antibiotic polymyxinB (PMB) or by detoxification of endotoxin by means of mild alkaline hydrolysis. Central treatment with PMB also attenuated the hypo- glycemicresponse to systemic administration of endotoxin or lipid A: These results suggest a direct role of the CNS in the mechanism of endotoxin hypoglycemia. Endotoxin mimicks the effect of insulin in stimulat- ing glucose oxidation in adipose tissue in vitro 24,31 and in producing hypoglycemia following systemic administration in vivol, 5. Insulin also causes hypogly- cemia when injected into the central nervous system (CNS)It, 25,33, and the CNS has been implicated in the mechanism of endotoxin action. Specifically, it has been shown that injection of endotoxin into the CNS in experimental animals could result in various pa- thophysiologic effects, including systemic hypoten- sion, respiratory depression, intestinal lesions, fever, and death 2,9,13,23. To explore possible links between the central action of endotoxin and its insulin-like hy- poglycemic effect we measured plasma glucose in mice, injected intracerebroventricularly (i.c.v.) with endotoxin. Experiments were carried out in normally fed male ICR mice, 28-30 g. Endotoxin (E. coli 0111:B4; Sig- ma) was administered i.c.v, in 5/A saline according to the method of Haley and McCormick 6. The mice were sacrificed by decapitation at different times af- ter endotoxin treatment, the trunk blood was col- lected and centrifuged to obtain plasma, and plasma glucose was determined using a Beckman Glucose Analyzer, model 2. As shown in Fig. 1A, i.c.v, injection of 2.5 ktg en- dotoxin produced a rapid, transient increase in plas- ma glucose followed by profound progressive hypo- glycemia. The hypoglycemic effect was most pro- nounced at 3 h post-treatment, and it could be repro- duced by much lower doses of endotoxin (Fig. 1B). Previous studies have shown that most of the biolo- gic effects of endotoxin are attributable to the hydro- phobic lipid region of the endotoxin molecule, termed lipid A 7,15. Specifically, it has been demon- strated that various endotoxic reactions including hy- poglycemia are inducible with purified lipid A prepa- rations ~6,29,30, and that the toxic effects of endotoxin are mitigated by the cyclic polycationic peptide anti- biotic polymyxin B (PMB) 3.t°,12.19,26, which binds specifically to lipid A 17. To determine the role of lipid A in the central hypoglycemic effect of endotoxin, mice were injected i.c.v, with 5/~1 saline containing 2.5 pg endotoxin, or 2.5 ug endotoxin plus 25 pg PMB (Sigma). Endotoxin and PMB were mixed 30 min prior to their coadministration. Plasma glucose was determined 3 h after treatment. As shown in Fig. 2, treatment with 25 ug PMB completely blocked the hypoglycemic effect of cen- tral endotoxin (2.5 pg). PMB had no effect on plas- ma glucose when administered alone, indicating that the inhibitory effect resulted from specific neutraliza- tion of the lipid A portion of endotoxin and not due to central activation of peripheral glucose counterregu- Correspondence: S. Amir, Department of Isotope Research. The Weizmann Institute of Science, Rehovot, Israel, 0006-8993/85/$03.30© 1985Elsevier Science Publishers B.V. (BiomedicalDivision)