The greater susceptibility to the lethal effects of bacterial endotoxin (heat-killed Salmonella typhimurium or Escherichia coli lipopolysaccharide, in mice infected with an attenuated strain of Mycobacterium tuberculosis (BCG) was confirmed. It reached a maximum at 2 weeks postinfection and gradually diminished for an additional 6 weeks.
At the time of maximum susceptibility several metabolic and physiological differences became apparent. BCG-infected mice die sooner (4 to 12 hours) and without the diarrhea, conjunctivitis, and general symptomatology associated with endotoxin deaths of normal animals. Reticuloendothelial blockade results in only a small change in reactivity to endotoxin, in contrast to normal mice. Subcutaneous injection of 2 units of ACTH is followed by no significant increase in urinary nitrogen excretion while in control animals it more than doubles. Plasma clearance of intravenously administered inulin is approximately normal in BCG-infected mice 17 hours after an LD50 dose of endotoxin but control mice similarly treated show renal impairment. In line with this result is the absence of elevated carcass non-protein nitrogen (NPN) following endotoxin poisoning or at the moment of death from endotoxemia in the hyperreactive animals in contrast to the two- to threefold increase in carcass NPN in normal mice under similar conditions. Body carbohydrate is at a minimum and becomes depleted to a level approximating that found at death more rapidly in BCG-infected mice given endotoxin than in controls. There is also a lower ratio of carbohydrate anabolized to protein catabolized following cortisone administration to BCG-infected mice than in control mice. This is found in adrenalectomized mice and in stressed animals and is reported elsewhere.
Some of the differences just described can be attributed to a refractory adrenal cortex. There is less depletion of adrenal cholesterol in vivo and lower corticoid synthesis in vitro than in normal mice yet this is not fundamentally responsible for the greater susceptibility of BCG-infected animals to endotoxin since adrenalectomized mice, which are even more susceptible, are metabolically and physiologically more comparable to normal mice than to BCG-infected mice. One can conclude, therefore, that the hyperreactivity of BCG-infected mice is more than an intensification of the normal response to endotoxin.