STUDIES ON THE PATHOGENESIS OF FEVER. 13. THE EFFECT OF PHAGOCYTOSIS ON THE RELEASE OF ENDOGENOUS PYROGEN BY POLYMORPHONUCLEAR LEUKOCYTES

ENDOGENOUS PYROGEN RELEASE FROM RABBIT BLOOD CELLS INCUBATED IN VITRO WITH PARAINFLUENZA VIRUS

Rabbit blood cells incubated in vitro with purified parainfluenza-5 virus (DA strain) released a rapidly acting pyrogen. Spleen and lymph node cells were inactive. The pyrogen resembled in behavior a pyrogen extracted from granulocytic exudates. Similar cells in the blood are believed to be activated by virus in vivo to produce the circulating endogenous pyrogen that mediates virus-induced fever.

STUDIES ON THE PATHOGENESIS OF FEVER. XII. ELECTROLYTIC FACTORYS INFLUENCING THE RELEASE OF ENDOGENOUS PYROGEN FROM POLYMORPHONUCLEAR LEUKOCYTES

The metabolic reactions responsible for the release of endogenous pyrogen from rabbit granulocytes incubated in 0.15 M NaCl are specifically inhibited by the presence of K⁺ (and by related alkali metal ions, Rb⁺ and Cs⁺) in the medium. The inhibitory action of K⁺ apparently involves penetration of the cell membrane and is directly antagonized by the cardiac glycoside, ouabain. It is concluded, therefore, that the inhibition of pyrogen release by extracellular K⁺ is due to transport of K⁺ into the cell. Although the precise molecular mechanisms which are responsible for the release of pyrogen from granulocytes incubated in K-free saline have not been elucidated, further study of the process has revealed: (a) that it is preceded by the accumulation of pyrogen within the cell, (b) that it depends upon the catalytic action of one or more sulfhydryl-containing enzymes, (c) that it does not require energy, either from glycolysis or from reactions depending on molecular oxygen, and (d) that its inhibition by K⁺ and by arsenite is qualitatively similar to the depression caused by these same reagents on the release of other leucocytic proteins; i.e., lysozyme and aldolase.

PRODUCTION OF INFLAMMATORY CHANGES IN THE MICROCIRCULATION BY CATIONIC PROTEINS EXTRACTED FROM LYSOSOMES

Lysosomal granules of rabbit exudate polymorphonuclear (PMN) leucocytes were isolated and then lysed by freezing-thawing. Topical application of this material to rat and rabbit mesentery produced sticking and emigration of leucocytes, stasis of blood flow, and petechial hemorrhage. The granule-free, supernatant fraction of the homogenized leucocytes failed to produce any of these reactions. Cationic proteins extracted from these granules by weak acid and precipitated by ethanol at concentrations of 20 and 45 per cent, were also tested on heterologous, homologous, and autologous mesenteric vessels. The 20 per cent ethanol-precipitated fraction produced all of the aforementioned injury reactions, whereas the 45 per cent fraction was inactive. The intensity of inflammatory changes produced by the active cationic protein fraction was greater than that produced by lysed whole granules. Both the 20 per cent and 45 per cent ethanol fractions of cationic protein induced clumping of rabbit platelets, in vitro. The 20 per cent ethanol fraction also caused a slight acceleration in rate of swelling of isolated rabbit liver mitochondria. The active material proved to be non-pyrogenic in rabbits. This material exhibited no kinin-like effects when tested on isolated smooth muscle preparations (rabbit aorta and guinea pig ileum). In the rat, the protein produced a transient vasodepression which was inhibited by pretreatment of the animal with an antihistamine. Ultraviolet absorption data and ribose assays showed that the 20 per cent ethanol fraction contained only 4 per cent or less of ribonucleic acid. Upon electrophoresis in starch gel, using acid buffer, this fraction separated into at least three major components which migrated towards the cathode. Precipitation of one of the slowly migrating components by titration of the fraction to pH 10.5 greatly increased the inflammatory activity of the material. The inflammatory basic protein fraction was essentially devoid of acid phosphatase, beta glucuronidase, acid ribonuclease, lysozyme, and catalase activity. The non-inflammatory basic protein fraction contained appreciable quantities of acid ribonuclease and lysozyme. The foregoing data demonstrate that certain of the cationic proteins present in lysosomes of rabbit exudate PMN leucocytes can reproduce one of the cardinal features of the inflammatory response; namely, adhesion and emigration of leucocytes in the microcirculation. These findings offer fresh support for the role of lysosomes in the pathogenesis of tissue injury, and may help to account for the propagation of leucocyte emigration to peak numbers during inflammatory reactions.

STUDIES ON TUBERCULIN FEVER. 3. MECHANISMS INVOLVED IN THE RELEASE OF ENDOGENOUS PYROGEN IN VITRO

In a search for the source of the circulating endogenous pyrogen (EP) that mediates tuberculin-induced fever, tuberculin was incubated in vitro with various tissues of rabbits sensitized by intravenous infection with BCG. Evidence was obtained that tuberculin specifically stimulates cells in the blood of sensitized rabbits to generate pyrogen in vitro, whereas both lymph node and spleen cells from the same donors were inactive. Since normal blood cells, incubated in plasma of sensitized donors, were similarly activated, it is postulated that circulating antibodies play a role in sensitizing cells (presumably granulocytes) to release pyrogen on contact with tuberculin) both in vitro and in vivo. Release of endogenous pyrogen in vitro may be a sensitive means of detecting immunologic reactions between antigen and specifically sensitized blood cells-in other allergic states accompanied by fever.

THE PRESENCE OF ENDOGENOUS PYROGEN IN NORMAL RABBIT TISSUES

Saline extracts of homogenized, uninfected, rabbit tissues produced febrile responses when injected intravenously into rabbits. Extracts of muscle, lung, and heart evoked fevers that were similar to those induced by leucocyte pyrogen; extracts of spleen, liver, and kidney caused more sustained fevers. The minimal pyrogenic dose appeared to be between 1.5 and 3 gm wet weight of tissue. Evidence is presented that neither Gram-negative bacterial endotoxin nor polymorphonuclear leucocytes (circulating or sequestered in the tissues) can be implicated as the source of pyrogen in tissue extracts. It seems likely, therefore, that a pyrogenic material of truly endogenous origin is widely distributed in tissues. Tissue pyrogen appears to be a large molecule which is relatively resistant to treatment with acid but not with alkali. Possible pathological roles for this endogenous agent (or agents) are briefly indicated.

Leukocytic pyrogen: a major mediator of the acute phase reaction

The acute phase reaction is initiated and regulated by leukocytic pyrogen which is released by activated host phagocytes at inflammatory sites during inflammatory reactions of any etiology. The physiological alterations which follow include, in addition to the characteristic acute phase protein changes, fever, granulocytosis, fall in serum iron and zinc, metabolic stimulation of blood granulocytes, activation of T-cell function and doubtless other effects as yet unrecognized. These responses are mediated by free or by protein-bound LP molecules and reflect both direct receptor coupled effects and more indirect mechanisms such as release of secondary mediators from cells or body fluids and involvement of neural or neurohumoral pathways. This unitarian hypothesis implies a functional interrelationship between the component parts, and allows a model for critical examination and new hypothesis testing. The wider availability of pure preparations of LP and of possible subtypes of LP and of a sensitive immunoassay would help us to prove or to expand this hypothesis and to advance our understanding of this important phase of acute inflammation. Earlier in this conference we heard that the mediator derepresses a message: I'd like to put forward the idea that in the acute phase reaction the mediator is the message.

Endogenous pyrogen production by human blood monocytes stimulated by staphylococcal cell wall components

To determine the properties of Staphylococcus aureus contributing to its pyrogenicity, we compared, in human monocytes, endogenous pyrogen production stimulated by heat-killed S. aureus with that stimulated by purified S. aureus cell walls or by particulate peptidoglycan prepared from the same strain. Peptidoglycan, but not the purified cell wall preparation, was found comparable to S. aureus as an endogenous pyrogen stimulus. This finding was associated with a more effective monocyte phagocytosis of S. aureus and peptidoglycan as compared with that of purified cell walls. Lysostaphin digestion of peptidoglycan markedly reduced its pyrogenicity. To test whether the chemical composition of the ingested particles is important, latex particles were tested as possible stimuli for monocyte endogenous pyrogen release. Although 40 to 68% of monocytes ingested latex particles during the first hour, there was no evidence of endogenous pyrogen activity in the supernatant even when supernatants equivalent to 5.2 X 10(6) monocytes were tested. This study demonstrates that the pyrogenic moiety of the S. aureus cell wall resides in the peptidoglycan component. Phagocytosis is not in itself a pyrogenic stimulus, but rather serves as an effective mechanism to bring about contact between the chemical stimulus and the monocyte.

Preliminary experience in treating lymphocytic leukaemia with antibody to immunoglobulin idiotypes on the cell surfaces

Tumour-specific antiserum was raised in sheep against idiotypic determinants on the surface immunoglobulin of neoplastic lymphocytes from a patient with chronic lymphocytic leukaemia (prolymphocytic variant). The complement-activating IgG1 subclass of the anti-idiotype was prepared from the serum in monodisperse form for infusion. Two treatments of 480 and 1200 mg caused the white-cell count to fall by one-third and one-half respectively. However, there was a rapid resurgence, so that by 8 days after each treatment the counts were restored to approximately 85% of their former levels. No change was noted in the size of spleen or lymph nodes. Each treatment probably destroyed 4-8 X 10(11) cells, some 10% of the total tumour load. The antibody was rapidly consumed, and there was evidence of heavy utilization of complement.

Phagocytosis of Borrelia recurrentis by blood polymorphonuclear leukocytes is enhanced by antibiotic treatment

The removal of Borrelia spirochetes from the blood in relapsing fever was studied by examining patients' blood phagocytic cells with the Dieterle silver stain. Polymorphonuclear leukocytes ingested Borrelia at increased rates for several hours after antibiotic treatment, during which time the total numbers of circulating plasma spirochetes were decreasing. Incubation of infected blood at 37 degrees C for 2 h resulted in a progressive increase in phagocytosis. Addition of penicillin G and tetracycline to infected blood caused a further enhancement of phagocytosis. Electron microscopy of polymorphonuclear leukocytes revealed spirochetes in phagosomes. These results indicated that blood polymorphonuclear leukocytes have a prominent role in removing Borrelia from the plasma and suggested that antibiotics act by altering the surface of spirochetes to render them more susceptible to phagocytosis.

The release of an endogenous pyrogen from guinea pig leukocytes in vitro: a new model for investigating the role of lymphocytes in fevers induced by antigen in hosts with delayed hypersensitivity

Guinea pig periotoneal exudate (PE) cells incubated overnight in vitro with heat-killed Staphylococci released an endogenous pyrogen (EP) that could be assayed by intravenous injection in rabbits. The febrile responses were linearly related to the dosage of EP over an eightfold range. PE cells derived from guinea pigs with delayed hypersensitivity (DH) to bovine gamma globulin (BGG), also released EP when incubated with antigen in vitro. This reaction was specific and did not occur withe PE cells from normal or complete Freund's adjuvant-sensitized guinea pigs. Studies indicated that monos and/or polymorphonuclear leukocytes rather than lymphocytes were the source of EP. However, when incubated with BGG and sufficient dosages of BGG-sensitized lymphocytes, normal PE cells released EP over a 42 h period. These results suggest that antigen stimulates specifically sensitized lymphocytes to release an agent (perhaps a lymphokine) that activates phagocytic cells to release EP. This model offers unique advantages for investigating in vitro the role of the lymphocyte in antigen-induced fever in DH as well as the relationship of this lymphocyte-induced activity to other known biologic activities mediated by antigen stimulated lymphocytes.

Differences in pyrogen production by mononuclear phagocytes and by fibroblasts or HeLa cells

Phagocytosis of bacteria stimulates "professional" phagocytes to produce and release endogenous pyrogen (EP), the protein that mediates fever. To determine whether "nonprofessional" phagocytes also have this capacity, mouse and human fibroblasts and HeLa cells were cultured after ingestion of latex or chicken erythrocytes (CE), and EP release into culture supernate measured by mouse assay. No detectable pyrogen was released by these cell types after phagocytosis, whereas both latex and CE stimulated EP production by cultured mouse macrophages. These studies support the hypothesis that only professional phagocytes of bone marrow origin synthesize EP and induce fever.

Ultrastructural damage of leukocytes procured by the Leukopak: vulnerability of leukocytes to mechanical injury

Because of the often observed undesired effects in patients transfused with leukocytes collected by continuous filtration leukapheresis (CFL), we have investigated the morphology of these cells in the hope that such studies may help evaluate their in vivo functions. Aliquots of CFL samples, procured by the Fenwal Leukopak and taken before and after tapping the filters, were examined. Leukocytes prepared by dextran sedimentation served as controls and were used to demonstrate the vulnerability of these cells to mechanical injury. The morphology of leukocytes harvested by filtration leukapheresis was markedly altered, ranging from formation of cytoplasmic projections and vacuoles to disintegration of entire cells. The severe changes were only observed in samples collected after the filters were tapped. The number of cells involved varied from donor to donor. In some cases more than half of the examined leukocytes showed signs of damage. Dextran sedimentation leukocytes retained their normal ultrastructure. However, substantial injuries were inflicted on dextran sedimentation collected leukocytes following brief periods of vortexing. Our results indicate that leukocytes are quite vulnerable to mechanical injury, and the major cause of damage found in Leukopak-isolated leukocytes is the tapping of the filters. The toxic reactions in some recipients transfused with Leukopak-collected leukocyte preparations may result from materials released from disrupted leukocytes. We believe that the employment of the continuous filtration to procure leukocytes with the present procedure should be reviewed.

Studies on the mechanism of endogenous pyrogen production. III. Human blood monocytes

The characteristics of pyrogen production and release by human blood monocytes were investigated. A dose-response assay of monocyte pyrogen in rabbits indicated a linear relationship of temperature elevation to dose of pyrogen at lower doses. Monocytes did not contain pyrogen when first obtained, nor did they release it spontaneously even after 5 days of incubation in vitro. Pyrogen production was apparent 4 h after stimulation by endotoxin or phagocytosis, and continued for 24 h or more. Puromycin, an inhibitor of protein synthesis, prevented both initiation and continuation of pyrogen production and release. Pyrogen-containing supernates retained most pyrogenic activity during overnight incubation even in the presence of activated cells. Lymphocytes appeared to play no role in either initiation or continuation of pyrogen production in these studies.

Elevation in set point for body temperature regulation after prolonged exercise

1. In nine healthy men, after prolonged exercise (approximately 37 km walk and 2.5 km ascent and descent in 9 hr) deep body temperature at rest stabilized at a higher level than in the absence of previous exercise.2. The increase, approximately 0.36 degrees C, was recorded at both oral and rectal sites. It persisted while the subjects rested in warm and in cool surroundings, and disappeared about 11 hr after the end of exercise.3. The increase in body temperature after exercise was accompanied by a similar increase in the body temperature at which sweating started during body heating.4. After exercise the subjects' metabolic response to cold air (13.6 degrees C) was normal or increased in spite of the elevation in body temperature.5. Blood glucose concentration was never lower than 3.1 mM during these experiments.6. The temperature changes are most easily explained by mild pyrexia due to the release of endogenous pyrogen associated with minor tissue damage.

A neutrophil-immobilizing factor derived from human leukocytes. I. Generation and partial characterization

A factor has been derived from human leukocytes which irreversibly inhibits the response of human neutrophils to diverse chemotactic stimuli without impairing their viability. It is released by both polymorphonuclear and mononuclear leukocytes during incubation in acidic medium, after endotoxin exposure and subsequent incubation in low potassium medium, and during phagocytosis of particles. It is extractable from both leukocyte types and therefore must be preformed. This chemotactic inhibitor is completely separable from contaminating chemotactic activity in the crude supernatants, has a mol wt of 5000, and is inactivated by digestion with trypsin or chymotrypsin. It has been termed a neutrophil-immobilizing factor because it inhibits neutrophils directly and independently of the chemotactic stimulus, and has relatively little effect on human monocyte chemotaxis.

Studies on the pathogenesis of fever. XIX. Localization of pyrogen in granulocytes

Only intact exudate granulocytes from rabbits generated large amounts of endogenous pyrogen when incubated in 0.15 M NaCl. No matter how whole-cell lysates or combinations of subcellular fractions were incubated, their yields of pyrogen never approached those of whole cells; at most, only minimal amounts of pyrogen were formed, once the integrity of the cells had been destroyed. Some pyrogen could be extracted from disrupted cells, but never more than a fraction (<25%) of that released from incubated whole cells. The yield could be slightly improved by lowering the pH (to 3.5) and by increasing the volume of extraction fluid. Virtually all of the preformed pyrogen that could be extracted from sucroselysed cells was found in their cytoplasmic fraction. Contrary to the results of Herion et al. (3), none could be detected in the granular (or lysosomal) fraction. Likewise, all efforts to recover pyrogen from the membrane-nuclear fraction were unsuccessful. In keeping with the finding that preformed pyrogen is contained in the cytoplasmic fraction were the observations that practically all of the aldolase, a cytoplasmic enzyme, and very little of the acid phosphatase, a granular enzyme, were lost from the cells during the release of pyrogen. Lysozyme, an enzyme stored in both the granules and the cytoplasm, was partially released from the cells under the same circumstances. Neither the release of pyrogen nor its slight intracellular buildup that precedes release (4) were affected by concentrations of puromycin that block protein synthesis in the cells and prevent their activation. Hence, it is concluded that the release process, which also involves the formation of active pyrogen (4), does not require protein synthesis, whereas activation of the cells, which may involve the synthesis of an inactive precursor (2), does.

Studies on the origin of human leukocytic pyrogen

Release of the protein molecule, leukocytic pyrogen, is one of the many reactions exhibited by leukocytes after phagocytosis. After the ingestion of heat-killed S. albus, a 3-4 hr latent period exists, during which human peripheral leukocytes release no pyrogen, yet cellular metabolism is altered in such a way that pyrogen output may subsequently occur in the absence of further phagocytosis. Transcription of messenger RNA and translation of new protein are initial events in the. activation process, since addition of the inhibitors, actinomycin D, and cycloheximide or puromycin, during this period markedly depressed or abolished subsequent pyrogen release. These effects were noted to be dependent upon the time of addition of the inhibitors. None of the inhibitor drugs interfered with cell viability as measured by phagocytosis and hexose monophosphate shunt activity, nor did they alter the pyrogenicity of preformed leukocytic pyrogen. Vincristine did not inhibit pyrogen formation, consistent with its reported failure to alter RNA synthesis in mature human granulocytes. The glycolytic inhibitor, sodium fluoride, blocked pyrogen release both when added prior to particle ingestion or 1 hr after the initiation of phagocytosis. Whereas inhibition of phagocytosis would explain the sodium fluoride effect prior to 1 hr, this was not observed in leukocyte preparations incubated for 1 hr with S. albus before adding sodium fluoride. When sodium fluoride was added to preparations 2 hr after the start of incubation, the LP production was unimpaired. Potassium cyanide had no effect on cell activation or pyrogen release. These findings suggest that the primary energy supply for the activation process is derived from high energy phosphate bonds provided by anaerobic glycolysis. Since the major amount of cell activation appears to occur in the 1st hr after phagocytosis, this energy might be involved in the induction of a genome leading to the transcription of m-RNA and its translation into new protein or is required for polysome integrity during protein synthesis. It is suggested that this new protein may be leukocytic pyrogen itself, or an enzyme responsible for cleaving it from an inactive precursor.

Studies on the pathogenesis of fever. XVII. The cationic control of pyrogen release from exudate granulocytes in vitro

Evidence has been presented that the release of active endogenous pyrogen from rabbit exudate granulocytes incubated in isotonic NaCl is a relatively prompt energy-dependent process that is preceded by a rise in intracellular pyrogen, and involves a rise in total intracellular cations and an increased permeability of the cell membranes, but does not require the synthesis of new proteins.

Studies on the pathogenesis of fever. 18. Activation of leukocytes for pyrogen production

Blood leukocytes, in contrast to exudate leukocytes, release little or no pyrogen when incubated in 0.15 M NaCl unless previously activated by exposure to endotoxin or to a protein activator that is present in acute exudate fluid. The activation process, which also occurs during phagocytosis, involves the synthesis of cellular protein, presumably related to the pyrogen molecule. Evidence is presented that generation of pyrogen in sterile inflammatory lesions depends on both the activator and the anaerobic conditions in the exudate fluid.

Species specificity of leukocytic pyrogens

Polymorphonuclear neutrophilic leukocytes of the dog, cat, and goat release leukocytic pyrogen under the same conditions as the heterophile polymorphonuclear leukocytes of the rabbit. The characteristics of the febrile response to an intravenous injection of homologous leukocytic pyrogen in all four species are very similar: a brisk monophasic fever reaching a peak between 30 and 50 min with smooth defervescence to the baseline by 3 hr. Shivering, which is not obvious in the rabbit, is noted in the dog, cat, and goat during the first 30 min. Quantitative differences in response reveal the cat to be the most sensitive of of these species to homologous leukocytic pyrogen, followed by the rabbit, dog, and goat. The response to heterologous pyrogen is in most cases markedly diminished compared to that after equal doses of homologous protein, suggesting the operation of species specificity, although canine and feline pyrogen behaved very similarly in all tests. Species specificity of leukocytic pyrogen is probably related to amino acid substitutions in different species of a common mammalian protein effector molecule.

Studies on the pathogenesis of fever. XVI. Purification and further chemical characterization of granulocytic pyrogen

Small quantities of highly purified granulocytic pyrogen have been separated from contaminating proteins by disc electrophoresis in polyacrylamide gel. The biologically active material thus isolated was shown to be electrophoretically homogeneous at pH 9 and pH 3.8. Earlier work on the chemical properties of the pyrogen molecule has been extended to include: (a) estimation of its molecular weight by gel filtration; (b) demonstration of free sulfhydryl groups essential for its biological activity; and (c) evidence that it is not inactivated by exhaustive extraction with ethanolether or n-heptane.

Release of an endogenous pyrogen in vitro from rabbit mononuclear cells

The capacity of rabbit mononuclear cells to release an endogenous pyrogen (EP) in vitro has been studied. After incubation with tuberculin, preparations of predominantly monocytic cells, derived from the respiratory passages of the lungs of rabbits sensitized with BCG, were activated to release EP. Pyrogen production occurred more slowly with lung monocytes than with blood leukocytes of similarly sensitized rabbits and 9 to 10 hr incubation in a fully supportive medium was required to produce clear-cut results. As previously reported with blood leukocytes, mononuclear cells from the lungs of normal animals were also activated by tuberculin but to a lesser degree than were those from specifically sensitized rabbits. Under a variety of conditions, mononuclear cells from either spleen or lymph nodes of the same sensitized rabbits failed to release detectable amounts of pyrogen when incubated with tuberculin in vitro but were activated in a majority of instances when phagocytosis of heat-killed staphylococci was used as the stimulus. Release of pyrogen from lung monocytes appears to be an active process that is both temperature-dependent and requires protein synthesis. Neither serum antibody nor complement appears to play a role in this process. Evidence is presented that the granulocyte is the main source of pyrogen evolved by blood leukocytes incubated in vitro with OT or heat-killed staphylococci, whereas the lung macrophage and/or monocyte is responsible for most of the pyrogen released from the lung cell preparations. From these studies, it is concluded that mononuclear cells can be activated in vitro by several microbial stimuli and must be considered an additional cellular source of EP. The clinical implications of these findings for the pathogenesis of fever in granulomatous diseases where the monocyte is the predominant cell are discussed.

Studies on the pathogenesis of fever. XV. The production of endogenous pyrogen by peritoneal macrophages

Macrophages from oil-induced peritoneal exudates in rabbits produce endogenous pyrogen when first activated by incubation in 4 hr exudate fluid and then stimulated by incubation in potassium-free isotonic sodium chloride solution. The failure of earlier investigators to obtain pyrogen from macrophages is explained, and the relevance of macrophage pyrogen to fevers of agranulocytosis and other diseases, in which mononuclear rather than granulocytic exudates predominate, is discussed.

Mechanisms of endotoxin tolerance. IV. Specificity of the pyrogenic refractory state during continuous intravenous infusions of endotoxin

The mechanisms underlying the pyrogenic refractory state which develops rapidly during a continuous intravenous infusion of bacterial endotoxin have been further explored. The findings demonstrate that: (a) rabbits rendered refractory to a continuous intravenous infusion of E. coli endotoxin at a standard rate (18 x 10(-4) microg/min) become highly refractory to a single intravenous test bolus of endotoxin, but remain fully responsive to preformed endogenous pyrogen and to substances known to release endogenous pyrogen, i.e. influenza virus, old tuberculin in specifically sensitized rabbits, and staphylococcal enterotoxin; (b) administration of fresh whole blood from normal donors containing an average of 1.6 - 10(8) granulocytes fails to restore febrile responsiveness to the continuing E. coli endotoxin infusion; (c) refractory phase plasma and liver homogenates exhibit no enhanced capacity to inactivate E. coli endotoxin pyrogenicity; (d) splenectomized animals readily develop the pyrogenic refractory state during E. coli endotoxin infusions and exhibit diminished, rather than the increased inflammatory responses to intradermal endotoxin seen in sham-operated controls; (e) continuous intravenous infusions of gelatin-stabilized, heat-killed pneumococci produce sustained fevers; and (f) continuous intravenous infusions of old tuberculin into specifically sensitized animals rapidly elicit a pyrogenic refractory state. The present observations, considered together with those of other investigators, support the hypothesis that pyrogenic unresponsiveness to endotoxin involves two distinct immunologic mechanisms. In terms of this hypothesis, the rapid reduction in febrile responsiveness to endotoxin is mediated by desensitization at the cellular level. With small doses of endotoxin, such as those employed in the present studies, this desensitization is primarily specific; with larger doses, nonspecific mechanisms are superimposed. So long as the subsequent doses of endotoxin are closely spaced or continuously infused, optimal conditions are provided for cellular desensitization and pyrogenic unresponsiveness to a given quantity of endotoxin can be induced rapidly and maintained without the requirement for antibody. However, as the interval between endotoxin challenge is lengthened, cellular desensitization wanes and tolerance becomes increasingly dependent upon those antibodies directed against the common toxophore groupings responsible for endotoxin pyrogenicity which assist the reticuloendothelial system in the clearance and destruction of this molecule.