Properties of the exchangeable splenic platelets released into the circulation during exercise-induced thrombocytosis.

The human spleen normally retains about one-third of the body's platelets in an exchangeable pool which can be released into the circulation by alpha-adrenergic stimulation. Some previous investigators concluded that the splenic platelet population was enriched in a subpopulation of large, young, dense platelets (megathrombocytes) but more recent research suggests that platelet size, age, and density are largely independent variables. In this investigation the properties of the splenic platelets were studied after their release into the circulation by acute strenuous exercise in 11 normal subjects. The exercise caused a rise in mean platelet count from 245 +/- 49 to 328 +/- 71 x 10(9)/L--a net increase of 24 +/- 6% after correction for haemoconcentration. The mean platelet volume (MPV) of citrated platelets increased from 6.38 +/- 0.78 to 6.59 +/- 0.68 fL after exercise (P less than 0.01)--a rise of 3.7 +/- 4.1% suggesting that the MPV of the splenic platelet population was about 20% greater than that of the normal circulating population. The age distribution of the platelets was studied by measuring the platelet monoamine oxidase (MAO) activity several days after irreversible inhibition by tranylcypromine, when the young platelets had normal MAO activity but the older platelets had only 20% of normal activity. The mean platelet MAO activity did not change after exercise, indicating that the age distributions of the circulating and splenic populations were very similar. The platelet contents of several putative markers of platelet age (sialic acid, serotonin, beta-thromboglobulin, beta-N-acetylglucosaminidase) were also unchanged after exercise. Modal platelet density decreased slightly but not significantly after exercise. The splenic platelet population has a larger MPV but appears to have similar age and density distributions to the basal circulating population.

Subcellular composition of platelet density subpopulations prepared using continuous Percoll gradients.

Normal human platelets have been separated by density on continuous Percoll gradients and the subcellular composition of platelets of different density has been analysed. The number and concentration of dense granules increased significantly with platelet density, as did the concentrations of the dense granule constituents calcium and serotonin. The amount of serotonin per granule in the low density (LD) platelets was only two-thirds of the corresponding amount in the high density (HD) platelets. Platelets of all densities were able to sequester exogenous serotonin and release it in response to thrombin stimulation with similar efficiencies. The concentrations of the alpha-granule constituents von Willebrand factor and beta-thromboglobulin increased significantly with platelet density but the concentrations of the lysosomal enzyme beta-N-acetylglucosaminidase and total sialic acid did not differ significantly in the density subpopulations. The concentrations of the cytosolic enzymes lactate dehydrogenase and glucose-6-phosphate dehydrogenase were slightly higher in the LD population than in the other density subpopulations. The concentration of glycogen showed a marked positive relationship with platelet density and calculations suggested that glycogen was an important determinant of platelet density heterogeneity. The findings of the present study are compatible with recent suggestions that LD platelets may be slightly younger than HD platelets in normal human subjects.

The relationship of human platelet density to platelet age: platelet population labeling by monoamine oxidase inhibition.

The relationship between platelet density and platelet age appears to vary between species with relatively few labeling studies in humans reported. In this study, irreversible monoamine oxidase (MAO) inhibitors were used to biochemically label the circulating platelet population in 15 humans. Platelet samples were then isolated during the 15 days after drug ingestion. The platelets were separated by density on continuous linear Percoll gradients and the density distributions were divided into five fractions containing approximately equal numbers of platelets. Baseline MAO activity was strongly correlated with platelet density. Twenty-four hours after a single dose of tranylcypromine, platelet MAO activities in the density subpopulations were reduced to 14% to 17% of the baseline values. During the first five days after inhibition, the rates of recovery of MAO activity (percentage per day) were inversely proportional to platelet density. The recovery rates in the two most dense fractions were initially slow but increased after five days. Percentage recovery of MAO activity in the least dense fraction was significantly greater than the percentage recovery in the most dense fraction on days 2, 3, 5, and 8 (P less than .01, sign test). These results support the hypothesis that normal human platelets show a small increase in density with age, but they do not exclude the additional possibility that human platelet lifespan is positively correlated with platelet density.

Morphometric analysis of density subpopulations of normal human platelets.

Platelets from seven normal subjects were fractionated on continuous Percoll density gradients and low density (LD), intermediate, and high density (HD) platelets were prepared for transmission electron microscopy followed by computerised morphometric analysis. Normal ultrastructural appearance and discoid shape were preserved by incubation of the platelets in nutrient medium at 37 degrees C immediately before fixation. HD platelet sections had a larger mean cross-sectional area but a lower ratio of the major to the minor axis compared to LD platelet sections. HD platelets contained more alpha granules, dense granules and mitochondria per square micron of section area than LD platelets. The percentage of section area occupied by open canalicular system was greater in the LD platelets while the percentage area occupied by glycogen fields was over ten-fold higher in the HD platelets. The mean cross-sectional areas of individual alpha granules and dense granules increased with density while the opposite trend was found for mitochondria. It is suggested that these ultrastructural differences mainly arise during thrombopoiesis and may indicate some functional specialization among platelets.

The use of monoamine oxidase inhibition to estimate megakaryocyte-platelet regeneration time (MPRT).

The megakaryocyte-platelet regeneration time (MPRT) was measured in 16 normal subjects after irreversible monoamine oxidase (MAO) inhibition. When ten subjects were given a single 20 mg dose of Parnate (tranylcypromine) the mean MPRT was 242 +/- 20 hours (linear model) or 212 +/- 22 hours (weighted mean model) with a mean %-linearity of 75 +/- 22%. The renewal of platelet MAO activity was initially retarded for about 2 days, suggesting that inhibition of megakaryocyte MAO had occurred. Six subjects who had taken MAO inhibitors daily for 1-6 months gave longer MPRT values (274 +/- 57 hours (lin.) or 246 +/- 67 hours (w. m.)), but the differences between the groups were not statistically significant. The single-dose MPRT results agree well with those obtained using the irreversible inhibition of cyclo-oxygenase by aspirin, but exceed the platelet life-spans estimated using radiochemical methods by about 24-36 hours. Platelet serotonin levels did not change significantly after a single dose of Parnate, but supine systolic blood pressure rose by 9 +/- 7 mm Hg (p less than 0.01) 24 hours after the dose. Estimation of MPRT by MAO inhibition is a useful alternative to the aspirin method but screening of subjects and clinical supervision is needed to avoid hypertensive complications.

Monoamine oxidase and other mitochondrial enzymes in density subpopulations of human platelets.

Normal human platelets have been separated according to density on continuous Percoll gradients and the platelet distribution divided into five fractions containing approximately equal numbers of platelets. The mean volumes and protein contents of the platelets in each fraction were found to correlate positively with density while the protein concentration did not differ significantly between the fractions. Four mitochondrial enzymes (monoamine oxidase, glutamate dehydrogenase, cytochrome oxidase and NADP-dependent isocitrate dehydrogenase) were assayed and their activities per unit volume were found to increase in a very similar monotonic fashion with platelet density. When MAO and GDH were assayed on the same set of density fractions the correlation between the two activities was very high (r = 0.94-1.00, p less than 0.001) and a similar close correlation was found between MAO and ICDH. The results support the hypothesis that high density platelets either have a higher concentration of mitochondria or have larger mitochondria than low density platelets.

Proplatelets and stress platelets.

The process of platelet formation by the fragmentation of megakaryocyte pseudopodia, termed proplatelets, demonstrable in the marrow sinusoids is poorly understood. "Stress" platelets produced under conditions of stimulated platelet production differ from normal circulating platelets with respect to volume and a number of functional characteristics. To clarify the relationship of stress platelets to proplatelets, rats were injected with heterologous platelet antiserum. Nondiscoid platelet forms, some characteristically beaded in appearance, strongly resembling bone marrow proplatelets, can be recovered in the circulation of normal rats. During the early period of recovery from acute thrombocytopenia, there was a substantial increase in the proportion of these elongated platelets in the citrated platelet rich plasma. Exposure to EDTA rendered them spherical. Circulating proplatelets may contribute significantly to the prompt increase in platelet volume during recovery from acute thrombocytopenia at a time prior to significant increase in megakaryocyte size and ploidy.

Triton X-114 phase separation of platelet membrane glycoproteins from normal subjects and a patient with type I thrombasthenia.

Surface-labelled normal and thrombasthenic platelets have been subjected to phase separation in Triton X-114. Triton-rich and Triton-poor fractions have been analysed by SDS-PAGE and IEF-SDS-PAGE. Partitioning characteristics of the major glycoproteins have been defined. The Triton-rich fraction contained GPIIb, III, IV, VI, VII, VIII, GP38 and the IIb beta subunit. In contrast, the Triton-poor fraction contained the HMWGP, GPIa, Ib, IIb, III, V and GPIX. Analysis of the platelet membrane glycoproteins of a patient with Type 1 thrombasthenia has been carried out using Triton X-114. The value of the method in diagnosis of this condition and differences between our findings and those published previously are discussed.

Variable expression of Qa-m7, Qa-m8, and Qa-m9 antigenic determinants on primitive hemopoietic precursor cells.

Using monoclonal antibodies, we have analysed the distribution of three recently described Qa antigenic determinants (Qa-m7, Qa-m8 and Qa-m9) on murine clonable hemopoietic progenitor cells and spleen colony-forming units (CFU-S). Cytotoxicity experiments showed that Qa-m7 was expressed on almost all the progenitor cells (colony-forming cells, CFC) of megakaryocytes (MEG-CFC), erythroid cells (E-CFC), B lymphocytes (BL-CFC), and mixed colonies (MIX-CFC) as well as day 13 CFU-S, and a major proportion of granulocyte-macrophage colony-forming cells (GM-CFC) and day 8 CFU-S. Experiments using four sources of granulocyte-macrophage colony-stimulating activity suggested differential expression of Qa-m7 on subpopulations of GM-CFC, those preferentially forming macrophage colonies having lowest Qa-m7 antigen density. Immune rosetting techniques demonstrated the selective expression of Qa-m8 on approximately 50% of MEG-CFC, MIX-CFC and day 13 CFU-S, a pattern similar to that of Qa-m2. In contrast, Qa-m9 was not detected on any of the primitive hemopoietic precursors assayed. The results demonstrate the complexity of the Qa antigenic system, and suggest a possible role for these antigens in hemopoietic differentiation.

Megakaryoblast precursors in rodent bone marrow: specificity of acetylcholinesterase staining.

Small cells staining diffusely for acetylcholinesterase, corresponding closely to the description of Jackson (1973, 1974), have been quantitated in the bone marrow of mouse and rat. These cells fulfil the requirements of megakaryoblast precursors, showing elevation in numbers prior to megakaryocytes in response to thrombocytopenia, and carrying platelet-specific antigens demonstrated by immunofluorescence. Some show a clearly defined nucleus whilst in others the nucleus is obscured; the femoral content of both types follows the same time course after acute thrombocytopenia and the latter can be distinguished from megakaryocyte cytoplasmic fragments. Other small cells present in rodent bone marrow contain granular positivity after acetylcholinesterase staining but also granules staining with Luxol-Fast-Blue, suggesting that these cells are eosinophils. These cells are greatly increased in number with stimulation of eosinophilia by parasite infestation but show no response to acute thrombocytopenia. Peripheral blood eosinophils exhibit both acetylcholinesterase and Luxol-Fast-Blue staining. A number of properties ascribed to granular-staining small acetylcholinesterase-positive cells previously presumed to be megakaryoblast precursors will need to be reassessed.

The effects of low-dose vincristine on megakaryocyte colony-forming cells and megakaryocyte ploidy.

The administration of low-dose vincristine (VCR) (0.1 mg/kg) to mice resulted in thrombocytosis without prior thrombocytopenia. No significant changes in marrow megakaryocyte numbers were found. However, after a minor early decrease, mean megakaryocyte ploidy increased, with a peak at 3 d. The number of megakaryocyte colony-forming cells (MEG-CFC) in bone marrow did not change significantly. In contrast with the effects on marrow, the concentration of megakaryocytes and the content of MEG-CFC in the spleen were significantly reduced for 1-2 d after VCR. This reduction was followed by a compensatory rise in the splenic content of MEG-CFC (peak 3-fold increase at 3 d), and 1-2 d later, an increase in splenic megakaryocytes which was concurrent with the increased platelet count. Culture of marrow and spleen cells in the presence of VCR resulted in inhibition of megakaryocyte colony formation at concentrations greater than 5 ng/ml and parallel reduction of the number of megakaryocytes per colony and the mean ploidy of colony megakaryocytes. The results suggest that the thrombocytosis induced by low-dose VCR does not result simply from an effect on platelets, but reflects compensatory changes in megakaryopoiesis secondary to toxic suppression of megakaryocytes and their progenitors.

An antigenic difference between cells forming early and late haematopoietic spleen colonies (CFU-S).

Murine pluripotent haematopoietic stem cells have generally been assayed by their ability to form macroscopic colonies of haematopoietic cells in the spleens of heavily irradiated recipient mice (colony-forming unit-spleen or CFU-S assay). However, recent evidence suggests that there are distinct subpopulations of CFU-S. Most spleen colonies present 7-8 days after injection consist of differentiated erythroid cells, contain no primitive myeloid or erythroid precursor cells and disappear from the spleen within 3 days, whereas the majority of colonies present at 10-12 days contain primitive precursors, are multilineal and cannot be detected at 7-8 days. Furthermore, many 10-day-old spleen colonies do not contain cells capable of forming spleen colonies in secondary irradiated recipients (an index of the self-renewal capacity of stem cells) whereas at day 14 almost all colonies contain these CFU-S. These studies suggest that only when colonies are scored at or later than 11-12 days is the spleen colony technique adequate for the assay of multipotential stem cells. We now report an antigenic difference between subpopulations of CFU-S forming early (day 8) and late (day 14) spleen colonies which has been used to purify multipotential haematopoietic stem cells from murine bone marrow.

Differential expression of Qa-m2 alloantigen on murine hemopoietic progenitor cells. Selective enrichment for megakaryocyte progenitors.

The selective expression of Qa-m2 antigen on murine megakaryocyte colony-forming cells (MEG-CFC) has been demonstrated. A monoclonal antibody defining the Qa-m2 alloantigen has been shown to inhibit, in the presence of complement, in vitro formation by murine bone marrow cells of megakaryocyte colonies, but not granulocyte-macrophage, erythroid or eosinophil colonies. Inhibition of megakaryocyte colony formation was shown, by mixing and other experiments, to be due to C-mediated lysis of MEG-CFC rather than lysis of accessory cells. B-lymphocyte colony formation was reproducibly decreased by approximately 50%, demonstrating the presence of Qa-m2+ and Qa-m2- subsets of B-lymphocyte colony-forming cells (BL-CFC). MEG-CFC and granulocyte-macrophage colony-forming cells (GM-CFC) were also found to be Thy-1-, Ly-1-, Ly-6- and Ly-5-. Fractionation of Thy-1-, Ig- marrow by rosetting Qa-m2+ cells and centrifuging them through an Isopaque-Ficoll gradient yielded up to 8-fold enrichment for MEG-CFC and greater than 20-fold separation of MEG-CFC from GM-CFC. The anti-Qa-m2 antibody should be useful for further purification of MEG-CFC, and investigation of hemopoietic differentiation.