Sheldon E. Greisman

Comparative Pyrogenic Reactivity of Rabbit and Man to Bacterial Endotoxin

Summary Comparative reactivity of mature albino rabbits (1.8–2.2 kg) and of healthy man to the pyrogenic activity of purified S. typhosa, E. coli, and Pseudomonas endotoxins is presented. On a per kilogram basis, rabbit and man are approximately equally reactive to threshold pyrogenic quantities of endotoxin. When larger doses of endotoxin are employed, the dose-response relationships become considerably steeper for man. On a total dose basis, rabbits require smaller quantities of endotoxin to elicit threshold febrile responses, but as total toxin dose is increased, febrile responses of man rapidly exceed those of the rabbit. Subjective toxic responses of man parallel the pyrogenic responses. These data provide a more definitive base line for interpreting studies with bacterial endotoxin that involve extrapolation of rabbit febrile responses to man.

The role of endotoxin during typhoid fever and tularemia in man: IV. The integrity of the endotoxin tolerance mechanisms during infection

Volunteers infected with Salmonella typhosa develop a remarkable hyperreactivity to the pyrogenic and subjective toxic activities of homologous (S. typhos) and heterologous (Pseudomonas) endotoxins. The present studies quantitate this augmented reactivity and demonstrate by three differing approaches that significant tolerance to these endotoxins can be readily induced within the framework of the hyperreactive state. Thus, (a) tolerance induced before illness by repeated daily intravenous injections of the endotoxins remained demonstrable during overt illness, (b) daily intravenous injections of the endotoxins begun during overt illness evoked progressively increasing tolerance, and (c) continuous intravenous infusions of S. typhosa endotoxin during illness rapidly induced a pyrogenic refractory state. Despite unequivocal activation of the endotoxin tolerance mechanisms by any of the above methods, the febrile and toxic course of typhoid fever proceeded unabated. Similarly, in other volunteers with Pasteurella tularensis infection, continuous intravenous infusions of S. typhosa endotoxin evoked initial hyperreactive febrile and subjective toxic responses followed by rapid appearance of a pyrogenic refractory state without modification of the underlying clinical illness. These observations suggest that circulating endotoxin plays no major role in pathogenesis of the sustained fever and toxemia during typhoid fever and tularemia in man. The mechanisms responsible for the systemic hyperreactivity to endotoxin during typhoid fever and tularemia were further investigated. Low grade endotoxemia, nonspecific effects of tissue injury, impaired ability of the reticuloendothelial system to clear circulating endotoxin, and production of cytophilic antibodies capable of sensitizing leukocytes to endotoxin did not appear responsible. Inflammatory reactions to intradermal S. typhosa endotoxin increased significantly during typhoid fever. However, since no such dermal hyperreactivity developed to Pseudomonas endotoxin during typhoid fever nor to S. typhosa endotoxin during tularemia, the systemic hyperreactivity to bacterial endotoxins during typhoid fever and tularemia could not presently be ascribed to enhanced levels of acquired hypersensitivity.

Experimental gram-negative bacterial sepsis: reevaluation of the ability of rough mutant antisera to protect mice.

Summary Rabbit antisera to three rough Enterobacteriaceae mutants, the Rc chemotype of Escherichia coli J5, the Rd chemotype of Salmonella typhimurium SL 1032, and the Re chemotype of Salmonella minnesota 595, were administered iv or ip to outbred Swiss albino mice. Control animals were injected concomitantly with normal serum from the same donors obtained prior to immunization. One hour later, challenge was performed ip or iv with LD95-100 doses of viable Escherichia coli 018, Proteus mirabilis, or Klebsiella pneumoniae. Normal pre-immune rabbit sera, lacking detectable antibodies to the specific challenge bacterial strains or to Ra, Rc, Rd, or Re rough mutants of Enterobacteriaceae, exhibited definite abilities to reduce septic mortality when compared with physiologic sterile saline. Analysis of preimmune sera from individual rabbit donors revealed a wide spectrum of protective activity. Post-immune sera against the rough bacterial mutants, possessing high titers of Rc, Rd, or Re antibodies, conferred no protection above that afforded by the corresponding preimmune sera. Only antisera to the specific challenge bacterial strain proved more protective than the corresponding pre-immune sera. Normal horse sera from 6 different sources, obtained from healthy animals never immunized with gram-negative bacterial vaccines, all possessed agglutinating antibodies against Ra and Rc (but not Re) chemotypes of Enterobacteriaceae and all provided high levels of protection aginst iv K. pneumoniae sepsis in mice. However, naturally occurring specific antibodies, not the anti-rough mutant antibodies, appeared primarily responsible for such protection since protective activity was markedly reduced (>95%) by absorption with the homologous (K. pneumoniae) but not with a heterologous (P. mirabilis) smooth species and since this loss of protective activity was unaccompanied by any decline in anti-Ra or Rc titers. The findings fail to support the conclusion that antisera to rough gram-negative bacterial mutants confer broad spectrum protection to mice against parenteral challenge with smooth Enterobacteriaceae because of the rise in antibody titer to common core antigens.

Induction of Endotoxin Tolerance

Resistance to gram-negative bacterial endotoxins occurs in two general forms. One appears genetically determined, the affected species (or strain) exhibiting minimal responses to an initial intravenous injection of relatively massive quantities of endotoxin. Examples are the lethal and pyrogenic unresponsiveness seen in baboons, vervets, and C3H/ HeJ mice (Sultzer, 1968; Westphal, 1975). In contrast to such natural resistance, other species respond to an initial intravenous injection of endotoxin with an array of striking physiologic and biochemical alterations, man being one of the most highly responsive (Greisman and Hornick, 1973). Most of these responses decrease progressively when the endotoxin is readministered at appropriate intervals. Such acquired resistance, or tolerance, does not necessarily develop in the same temporal sequence for each response; indeed, particularly during the initial several days, hyperreactivity may occur to one while tolerance appears to another (Greer and Rietschel, 1978). This review will consider some of the mechanisms underlying the induction of endotoxin tolerance to two responses that have been most intensively studied—fever and lethality.

Experimental Gram-negative bacterial sepsis: optimal methylprednisolone requirements for prevention of mortality not preventable by antibiotics alone.

Abstract Outbed ICR mice were inoculated ip with one LD90-100 Escherichia coli 0:18, Proteus mirabilis, or Klebsiella pneumoniae. After carefully timed intervals, aminoglycoside antibiotics were begun at dosages and intervals predetermined to constitute optimal therapy. With progressive delay of antibiotic therapy, mortality rates increased progressively from 0 to 90-100%. Standardized models of infection were obtained by selecting delay periods before initiating antibiotic therapy such that 50 to 70% mortality rates resulted. In these models, 30 mg/kg methylprednisolone (MP) given prior to or concomitantly with the delayed antibiotic therapy and repeated 4 hr later was previously shown capable of preventing gram-negative septic mortality not preventable by the optimal antibiotic therapy alone. The present studies quantitate the optimal quantities of MP required for such protection. It was found that (1) in the absence of aminoglycoside antibiotics, MP consistently failed to reduce mortality; (2) in the antibiotic-treated animals, a single im injection of 30 mg/kg MP provided definite protection. A second and a third injection of 30 mg/kg of MP at 4-hr intervals provided additional, but decremental, increments of protection; (3) decreasing the dose of MP to 10 mg/kg or less consistently reduced its protective activity; (4) increasing the dose of MP to 60 mg/kg or greater consistently reduced its protective activity; (5) short additional increments in delay in initiating antibiotic and MP therapy annulled the protective activity of MP. The findings indicate that the ability of MP to reduce murine mortality from gram-negative bacterial sepsis is not only critically restricted by the requirements for its early administration in conjunction with appropriate antibiotics, but also by its relatively narrow optimal dose-response range and the decremental increments in its effectiveness upon repetitive administration.

Mechanisms of endotoxin tolerance. The role of the spleen.

Splenectomy markedly impaired the production of circulating anti-endotoxin antibodies during the initial 10 days after .v. administration of a Boivin preparation of Escherichia coli endotoxin (ET) in both rabbit and man. Increase in antibodies with secondary (flocculating and bactericidal) activities were virtually abolished, whereas increases in antibodies with primary (binding) activity were significantly reduced. On the basis of these findings, splenectomized rabbit and man were employed to test the hypothesis that the early phase (less than 72 h) of pyrogenic tolerance to endotoxin is independent of anti-endotoxin antibody but that such antibody contributes significantly to the later phase (less than or equal to 72 h) of tolerance. In the splenectomized rabbit, the initial pyrogenic reponses to ET and the subsequent tolerant responses at 24 and 48 h were comparable to sham-operated controls...

On the Demonstration of Circulating Human Endogenous Pyrogen

Summary Acclimatized rabbits were injected intravenously with 10 ml/kg plasma aliquots from patients febrile with Rocky Mountain spotted fever (104–105°F). Prior to assay, the febrile phase human plasma was processed by repeated absorptions with packed washed rabbit erythrocytes at 4° to remove all detectable heterophile agglutinins. The latter were found in all human plasma samples tested and accounted for their febrile and lethal activity in the rabbit. Ten of 11 absorbed febrile phase human plasma aliquots failed to evoke any febrile response. These negative findings could not be attributed to depression of rabbit thermoregulatory mechanisms or to absorption of human endogenous pyrogen during heterophile agglutinin removal. Since the negative findings might be related to relative insensitivity of the rabbit to human endogenous pyrogen, additional attempts to demonstrate circulating human endogenous pyrogen were performed with human recipients. Five hundred milliliters whole heparinized blood drawn from each of 2 volunteers 30 min before their peak febrile response to S. typhosa endotoxin failed, when reinfused the following morning, to induce an early febrile reaction compatible with an endogenous pyrogen response. Rather, a delayed rise in rectal temperature occurred compatible with residual circulating bacterial endotoxin. A third recipient was given 750 ml of pooled heparinized plasma drawn from 4 other volunteers 1 hr before their peak febrile reaction to S. typhosa endotoxin. Both an early and a delayed febrile response now ensued, compatible with responses to endogenous pyrogen and residual circulating endotoxin, respectively. Reasons for relating the early febrile response to circulating endogenous pyrogen are considered.

Activation of Histamine-Releasing Factor in Normal Rat Plasma by E. coli Endotoxin

Summary Employing isolated guinea pig ileum segments in aerated Tyrode solution, 5 to 250 μg aliquots of a physiologically active lipopolysaccharide component of E. coli endotoxin were found capable of activating a potent histamine-like releasing factor upon in vitro incubation with 0.5 ml diluted fresh heparinized normal rat plasma. The release of histamine by this rat plasma factor occurred within seconds after tissue contact. The E. coli endotoxin preparation per se, in the dose-time relations employed, did not liberate histamine directly from the ileum segments; indeed, 2000 μg E. coli endotoxin failed to release detectable histamine during 20 minute tissue contact periods. This lack of histamine-releasing activity could not be attributed to preexisting endotoxin contamination of the tissue bath nor to toxic depression of the histamine sensitivity of the assay system. The above findings, in conjunction with previous evidence, suggest that E. coli lipopolysaccharide behaves, at least in one manner, as do numerous other high molecular weight polysaccharides and justifies the hypothesis that activation of plasma histamine-releasing factors may account for certain of the pathophysiologic alterations which characterize the endotoxic response. Further studies are required to define the importance of the polysaccharide moiety of the endotoxin molecule for the histamine-releasing activity.

Simple Method for Removal of Intestinal Tubing from the Stomach

Abstract When intestinal tubing cannot be removed by firm traction on the proximal end, spontaneous passage through the intestinal tract should occur, provided the tube is beyond the stomach. Mista...

Hyperlipemia and Hemolysis I. Interaction of Sodium Oleate with Human Erythrocytes

Summary 1. The initial step in human erythrocyte lysis by oleate at pH 7.2 ± 0.1 and concentrations below 8 × 10-4 M involves erythrocyte-oleate binding: oleate binding by a standard number of erythrocytes in a fixed volume is a function of oleate concentration, temperature, and contact time. 2. Once lytic quantities of oleate are bound, hemolysis proceeds at pH 7.2 ± 0.1 as a function of quantity bound and of temperature. Lysis cannot be checked by repeated iced saline washings. A mean minimum of approximately 7 × 10-9 μg of sodium oleate is required/erythrocyte at 37°C, pH 7.1 for lysis at rates detectably faster than controls; the sodium oleate binding mechanism appears saturated with a mean of approximately 111 × 10-9 μg/erythrocyte. 3. Lowered surface tension of oleate solutions does not influence hemolysis rates under conditions tested. 4. Cyanide retards oleate hemolysis; other non-specific enzyme inhibitors fail to do so. 5. Following oleate hemolysis, hemolytic substances are released; these probably represent oleate or products of oleate-erythrocyte interaction. It is suggested that oleate possesses the potential of inducing a decelerating hemolytic chain reaction.