Distribution of bacterial endotoxin in human and rabbit blood and effects of stroma-free hemoglobin.

Bacterial endotoxin (lipopolysaccharide [LPS]) is known to interact with numerous components of blood, including erythrocytes, mononuclear cells, platelets, neutrophils, lipoproteins, and plasma proteins. The relative affinities of LPS for these elements, and the distribution of LPS between them, are unknown. Cross-linked stroma-free hemoglobin (SFH), a potential substitute for erythrocyte transfusion, produces in vivo toxicity in animals consistent with significant LPS contamination. Therefore, we studied the distribution of LPS in human and rabbit blood and examined whether the presence of SFH altered LPS distribution. In either the presence or absence of SFH, LPS was associated predominantly with high-density lipoproteins and apoproteins. There was lesser binding to low- and very-low-density lipoproteins. Examination of the apoprotein pool by column chromatography and density centrifugation demonstrated that LPS in this fraction was predominantly protein bound. Binding of LPS to SFH resulted in dissociation of a portion of the LPS into low-molecular-weight complexes. Cell-bound LPS was only 2 to 16% of the total and was unaffected by SFH. The distribution among blood cells demonstrated predominant binding to platelets in human blood but predominant binding to erythrocytes in rabbit blood. Cellular distribution was not significantly altered by SFH.

Correlation of in vitro and in vivo biological activities during the partial purification of thrombopoietin.

We have evaluated three in vitro assays for megakaryocyte maturation as monitors of biological activities of thrombopoietin (TPO). The in vitro measurements, that is, potentiation of murine megakaryocyte colonies in soft agar cultures, growth of single murine megakaryocytes in soft agar, and acetylcholinesterase production in liquid cultures of murine bone marrow cells, were correlated with measurements of thrombopoiesis stimulatory activity in vivo, based on labeling of newly formed platelets with [75Se]selenomethionine. Protein fractions produced during the purification of TPO from the plasma of thrombocytopenic rabbits were used to evaluate the in vitro assays. Potentiation of murine megakaryocyte colony growth in soft agar was least valuable as a TPO assay, due to variability. Growth of single megakaryocytes in vitro, in a serum-free, agar culture system, correlated well with measurement of thrombopoiesis-stimulating activity in vivo, in regard to increases in specific activity during the purification, and was the most sensitive of the three assays. The serum-free, liquid culture assay also correlated well with the in vivo assay, and it had the additional advantage of being feasible for evaluation of the large numbers of protein fractions produced by high-resolution chromatographic procedures. Using the liquid culture system to assay fractions from gel permeation high performance liquid chromatography, a biologically active fraction with a molecular weight range of 40-47 kd was identified.

Multiple in vivo effects of interleukin-3 and interleukin-6 on murine megakaryocytopoiesis.

The in vivo effects of interleukin-3 (IL-3), interleukin-6 (IL-6), and a combination of IL-3 plus IL-6 on murine megakaryocytopoiesis and thrombopoiesis were examined. Human recombinant IL-6 was administered subcutaneously as 14 equal injections of 5,000 units each during a 102-hour period. Murine recombinant IL-3 was given as 8 injections of 80,000 units each during the first 54 hours. Megakaryopoiesis and thrombopoiesis were evaluated 120 hours after initial administration of the cytokines. Platelet levels increased by 20% following IL-3 alone, 35% following IL-6 alone and 61% after administration of both IL-3 and IL-6. Platelet production, as measured by 75Se-selenomethionine incorporation, increased by approximately 120% in animals that had received IL-6 or IL-3 plus IL-6. Megakaryocyte ploidy analysis by two-color flow cytometry showed a shift in the modal ploidy class from 16N to 32N and a significant increase in the frequency of 64N cells only in IL-6 treated animals. Both bone marrow and splenic megakaryocyte colony-forming cells were significantly increased following either IL-3 or IL-6. Bone marrow megakaryocyte size increased 18%, 43%, and 38%, respectively, after administration of IL-3, IL-6, or the combination of IL-3 plus IL-6. Leukocyte counts and hematocrits were unaffected by either cytokine. Additional groups of mice received the same injection schedule as above and the serial effects on peripheral blood cell levels were assessed for 30 days. Platelet levels, which had been elevated by IL-3 or IL-6, fell to control values within 4 days following the last injection. Animals given IL-6 or IL-3 plus IL-6 were subsequently thrombocytopenic relative to controls on days 7 through 9 following cessation of treatment. Temporary 'cycling' of platelet levels was observed for 3 weeks following treatment with IL-6 or the combination of IL-3 plus IL-6. We conclude that IL-6 and to a lesser extent IL-3 stimulate platelet production in vivo and that their combined effects on platelet levels are approximately additive. Following discontinuation of IL-3 or IL-6, the effects are rapidly reversed, presumably by negative feedback mechanisms, resulting in a period of 'rebound thrombocytopenia' in mice that had received IL-6.

Optimization of detection of bacterial endotoxin in plasma with the Limulus test.

Detection and quantification of bacterial endotoxin in plasma by the Limulus amebocyte lysate test (or other assays for endotoxins) is hindered by the presence of inhibitors. Treatment of plasma to overcome inhibitory activities is required before plasma can be successfully assayed for endotoxin. We have conducted an investigation comparing the three most commonly used procedures (dilution-heating, trifluoroacetic acid oxidation, and chloroform extraction) for treatment of plasma before its assay for endotoxin with the chromogenic Limulus test. Initially, conditions were optimized for treatment of plasma by each of these methods. Subsequently, a direct comparison of the three plasma treatment procedures was performed with plasma spiked with known concentrations of endotoxin. The optimized dilution-heating procedure resulted in the most sensitive detection of endotoxin, with sensitivity approximately 10 times greater than the optimized trifluoroacetic acid oxidation procedure and approximately 100 times greater than treatment of plasma by chloroform extraction. Maximal detection of low concentrations of endotoxin by the chromogenic Limulus test was obtained by dilution of plasma fourfold with 0.15 mol/L NaCl followed by heating at 60 degrees C for 30 minutes. This procedure was simple, rapid, and did not involve addition of any reagents to plasma that could potentially add contaminating endotoxin.

The effect of partially purified thrombopoietin on guinea pig megakaryocyte ploidy in vitro.

Enriched populations of guinea pig bone marrow megakaryocytes were prepared by density gradient and velocity centrifugation and maintained in liquid cultures for 24 or 48 h. The resulting megakaryocyte preparations were of 86.1% +/- 5.5% purity. After 24 or 48 h in liquid culture, recovery of viable cells was 77% +/- 11% or 83% +/- 13%, respectively. Megakaryocyte cultures were supplemented with 100-200 micrograms/ml of either a lectin-fractionated preparation of thrombopoietin (TPO) from the plasma of thrombocytopenic rabbits or an identically prepared protein fraction from non-thrombocytopenic animals. Addition of TPO resulted in a significant increase (p less than 0.05) in both the proportion and total numbers of 32N megakaryocytes and a significant decrease (p less than 0.05) in the relative frequency of 8N megakaryocytes. In most experiments, a decrease in the total number of 8N megakaryocytes also was noted. These results indicate that partially purified TPO is able to increase the ploidy (DNA levels) of megakaryocytes in vitro.

The effects of 5-fluorouracil on hematopoiesis: studies of murine megakaryocyte-CFC, granulocyte-macrophage-CFC, and peripheral blood cell levels.

the effect of 5-fluorouracil (5-FU) on megakaryocytopoiesis in mice was studied with assays of megakaryocyte colony-forming cells (Meg-CFC) in bone marrow and spleen and simultaneous determinations of peripheral blood counts, after a single intraperitoneal dose (150 mg/kg) of 5-FU. Although only moderate thrombocytopenia (platelet count 40% of control values) occurred at 7 days following administration of 5-FU, sustained rebound thrombocytosis (platelets 200-250% of control values) was observed from days 11 to 17. No rebound leukocytosis was detected despite comparable initial leukopenia. Megakaryocyte colony-forming cells (Meg-CFC) in bone marrow and spleen were decreased for 2 and 5 days, respectively, after administration of 5-FU. Subsequently, there was a prolonged rebound increase in the total number of Meg-CFC in the spleen from days 11 to 17 after 5-FU, a phenomenon which did not occur with Meg-CFC derived from the bone marrow. Granulocyte-macrophage colony-forming cells (GM-CFC) in bone marrow and spleen exhibited alterations which were similar to those of Meg-CFC, indicating similar sensitivities of GM-CFC and Meg-CFC to 5-FU. Normal feedback mechanisms which control platelet levels are perturbed for almost 3 wk after administration of 5-FU. The simultaneous occurrence of maximal thrombocytosis and increased splenic Meg-CFC suggests that increased platelet production after 5-FU is associated with concomitant stimulation of the megakaryocyte progenitor compartment in the mouse spleen. However, the concurrence of thrombocytosis and increased splenic Meg-CFC indicates that elevated levels of Meg-CFC did not initiate the period of thrombocytosis.

Effects of cyclophosphamide on murine bone marrow and splenic megakaryocyte-CFC, granulocyte-macrophage-CFC, and peripheral blood cell levels.

The effect of cyclophosphamide (CY) on megakaryocytopoiesis in mice was examined with assays of megakaryocyte colony-forming cells (Meg-CFC) in bone marrow and spleen and simultaneous determinations of peripheral blood counts, after a single intraperitoneal dose (200 mg/kg) of CY. Significant rebound thrombocytosis (170% of normal) occurred at day 11 after injection with CY, although only modest preceding thrombocytopenia (70% of normal) was observed. After an initial 3-5 day period of suppression, total megakaryocyte colony-forming cells (Meg-CFC) in both bone marrow and spleen of CY-treated mice demonstrated rebound increases at 5 and 7 days, respectively, after administration of the drug. Granulocyte-macrophage colony-forming cells (GM-CFC) exhibited alterations which were similar to those of Meg-CFC, suggesting similar sensitivities of Meg-CFC and GM-CFC to CY. The increase in Meg-CFC in both bone marrow and spleen preceded development of thrombocytosis by 4-6 days. This suggests that increased platelet counts in CY-treated mice are attributable, at least in part, to alterations in feedback mechanisms which control megakaryocytopoiesis, with resultant stimulation of the megakaryocyte progenitor compartment.

Measurement of ploidy distribution in megakaryocyte colonies obtained from culture: with studies of the effects of thrombocytopenia.

Microdensitometric measurement of the DNA content of individual megakaryocytes was performed using megakaryocyte colonies obtained following culture, in soft agar, of hematopoietic cells from C57BL/6J mice. Two types of colonies were detected. After 7 days of culture, the big cell type contained 16 /+- 2.3 acetylcholinesterase (AChE) positive cells/colony, with a mean ploidy level of 16.8 /+- 0.8/cell and the ploidy distribution characteristic of recognizable megakaryocytes in bone marrow. The heterogeneous type contained 44 /+- 9.6 cells/colony (some of which were AChE negative), with a mean ploidy level of 6.8 /+- 0.7/cell. The ploidy distribution of heterogeneous colonies differed markedly from big cell colonies, with preponderance of 2N and 4N cells. Colony-forming cells, obtained 4-5 days after induction of acute thrombocytopenia, gave big cell colonies with a marked increase in DNA content. Mean ploidy level increased to 21.5 /%- 1.8/cell; the frequency of 32N cells increased from 17% to 30% and 64N cells from 0% to 6%. This is the pattern of change observed in bone marrow, in vivo, 24 to 48 hr after induction of acute thrombocytopenia. The number of cells/colony did not increase. In contrast, acute thrombocytopenia did not alter the ploidy of heterogeneous colonies. The different responses to the stimulus of acute thrombocytopenia suggest that there are at least two types of Meg-CFC. The delayed appearance of altered Meg-CFC that produced big cell colonies indicates that the pool of stem cells, from which committed megakaryocyte precursors are derived, may respond indirectly to the stimulus of platelet depletion.

The effects of acute thrombocytopenia on megakaryocyte-CFC and granulocyte-macrophage-CFC in mice: studies of bone marrow and spleen.

The effects of acute thrombocytopenia, produced by platelet antiserum (PAS), on both megakaryocyte colony-forming cells (Meg-CFC) and granulocyte-macrophage colony-forming cells (GM-CFC) were studied. During the 1-hr to 14-day period following acute thrombocytopenia (platelet counts < 5% of normal), bone marrow and splenic cells of C57BL/6J mice were obtained and cultured for 7 days in 0.3% agar. Numbers of GM and Meg colonies were determined. At no times were alterations in frequency of GM-CFC and Meg-CFC detected in femoral bone marrow. In contrast, GM-CFC in spleen were increased from 3 to 7 days after PAS and from 4 to 7 days after normal serum (NS). Increase in Meg-CFC in the spleen occurred from 3 to 5 days after PAS with a lesser, not significant increase after NS. Alterations in white blood cells and hematocrit values were not detected. Similar responses were observed in germ-free mice and after rechallenge of animals that had received PAS or NS 14 days previously. The delayed increase in Meg-CFC indicates that they are unlikely to be responsible for the altered megakaryopoiesis previously reported in bone marrow after acute thrombocytopenia and was not due to inhibition by PAS. The increase in GM-CFC may reflect stimulation of the reticuloendothelial system by heterologous proteins.

Platelet and fibrinogen production: relative sensitivities to endotoxin.

The relative sensitivities of platelet and fibrinogen production to endotoxin were determined in male New Zealand rabbits. E. coli endotoxin was administered in single intravenous doses of 0.1 to 50.0 mu/g/kg body mass. 75SeM was injected 5, 12, or 18 hr after endotoxin, and the percent incorporation into platelets and fibrinogen was used to measure thrombopoiesis and fibrinogen synthesis. Leukopenia occurred after the infusion of endotoxin at all dose levels; the lowest dose that caused thrombocytopenia was 0.5 microgram/kg. Endotoxin was detected with the Limulus test in plasma of animals that received 5 to 50 microgram/kg. The stimulation of fibrinogen and platelet production, as well as the postinfusion decrease in platelet count, correlated directly with the log of the dose of endotoxin infused. The lowest dose of endotoxin that stimulated platelet and fibrinogen production was 0.5 microgram/kg. In animals that received this dose of endotoxin, increased fibrinogen production was detected only when 75SeM was injected 5 hr later. In animals injected 5, 12, or 18 hr later. Increased platelet production was detected at these two doses only when 75SeM was injected 18 hr after endotoxin. The data indicate that fibrinogen and platelet production have similar sensitivities to endotoxin, although the time course of stimulation is different for these two blood components.