International Journal of Cell Cloning 8:291-298(1990)
The Control of Megakaryocyte Ploidy and Platelet Production: Biology and Pathology Ann-Marie Gladwin, John E Martin Cardiovascular Research, The Wellcome Research Laboratories, Beckenham, Kent, United Kingdom
Key Words. Platelets Megakaryocytes ondary thrombocytosis
Polyploidy
Vascular disease
*
Sec-
Abstract. Following experimental platelet destruction in animals, large platelets, which are more hemostatically active, are produced before any change in bone marrow megakaryocyte DNA content. When platelet production is stimulated by administration of i.v. vincristine in rats, megakaryoqte ploidy is increased, but mean platelet volume is unchanged. When platelet production and destruction are both stimulated by chronic hypoxia or administration of anti-platelet serum, mean platelet volume and megakaryocyte DNA content are both increased. Since platelet volume is determined primarily at thrombopoiesis, these results imply that mean platelet volume and megakaryocyte DNA content are under separate hormonal control. Therefore, it has been postulated that changes in mean platelet volume occur following changes in platelet production rate, whereas changes in megakaryocyte ploidy are associated with an increased rate of platelet production. In myocardial infarction, platelets have increased mean volume and reduced bleeding time more than in controls. In addition, men with myocardial infarction have increased megakaryocyte size and increased DNA content when compared to controls. These changes are similar to those observed in rabbits follawing cholesterol feeding. If megakaryocyte polyploidy and mean platelet volume are under separate hormonal control, this suggests that in myocardial infarction, both hormones are active-one stimulating an increased platelet size, the other stimulating the increased megakaryocyte DNA content. In contrast, patients with lymphoma exhibiting a secondary thrornbocytosis have no change in mean platelet volume. However, these subjects also have larger bone marrow megakaryocytes when compared to controls. The relation between megakaryocyte size and ploidy implies that the DNA content of these cells is increased in lymphoma. When the dual control theory of platelet production is applied, it suggests that in malignant lymphoma there is activation of the control mechanism which governs ploidy, but not of the one which controls mean platelet volume. Correspondence: Ann-Marie Gladwin, Ph.D., Cardiovascular Research, The Wellcome Research Laboratories, Langley Court, South Eden Park Road, Beckenham, Kent BR3 3BS, UK. Received May 18, 1990;accepted for publication May 18, 1990. 0737-1454/90/$2.0010 0 AlphaMed Press
GladwinIMartin
292
Introduction Circulating blood platelets are heterogeneousin size, density and reactivity. In 1964, McDonald [I] suggested that size heterogeneity reflected platelet aging in the circulation, so that young platelets were large platelets which decreased in size as they aged. Since platelet size and density were found to be linked, Kalpatkin [2] extended the aging hypothesis of McDonald to suggest that platelet density decreased with platelet age. However, evidence has now been presented by a number of groups suggesting that platelets do not change their size or density as they age in the circulation. It is now widely believed that platelet heterogeneity is determined primarily at thrombopoiesis [3]. Several studies have demonstratedthat the mean volume of circulating platelets is altered in a variety of situations. For example, following myocardial infarction or coronary artery bypass grafting, and in idiopathic thrombocytopenia purpura, platelets have an increased mean volume [4-71. If platelet changes are determined at thrombopoiesis, such alterations in size might be expected to be preceded by changes in their parent cell, the rnegakaryocyte (meg). These large cells, produced in bone marrow, undergo endoreduplicationas they mature, resulting in a marked degree of heterogeneity, in both DNA content and size [8]. At times of decreased platelet number in man and in animal models, it has been observed that the meg population can increase their mean ploidy and size [9-121. The importance of these changes in regulating platelet production is not fully understood. However, studies of the changes in both megs and platelets in a variety of animal models is enabling elucidation of the control system of platelet production. This is of importance in allowing an understanding of the origin of platelet abnormalities in a variety of human pathological states.
The Relationship between Platelet Volume and Meg Ploidy in Animal Models It has been demonstrated that when platelets are destroyed in the circulation following a single injection of an anti-platelet serum, the new platelets which are produced 24 h afterwards have an increased mean size [lo]. These changes in animals are similar to those observed in man following a decreased platelet count associated with cardiopulmonary bypass [6]. Investigationof megakaryocyte nuclear DNA content following induction of immunothrombocytopeniahas also demonstrated an increase by 24 h and shows a maximal change at day 3 [9]. These observations were originally interpreted as suggesting that megs with an increased DNA content are responsible for the production of larger platelets. However, more recently, mean platelet volume and meg DNA content have been measured shortly after induction of immunothrombocytopenia. Trowbridge [13] and Corash [14] identified that large platelets were present by 2 h in the rat and by 8 h in the mouse, respectively. In both of these
The Control of Platelet Production
293
studies, the mean DNA content of the megs was unchanged at those times when compared to normal animals. If it can be discounted that this effect is mediated via release of larger platelets from the splenic pool, this indicates that the large platelets produced following thrombocytopeniaare produced from megs with an unchanged DNA content. Since changes in platelet volume are determined at thrombopoiesis, these studies would imply that the rapid increase in the size of platelets produced following thrombocytopenia must result from a change in the fragmentation of the meg cytoplasm. This may be part of the control system’s response to the hemostatic deficit of thrombocytopenia since such larger platelets produce more thromboxane B2per unit volume of cytoplasm and reduce bleeding time per unit volume more than controls [9]. When platelet destruction was prolonged by giving animals injectionsof antiplatelet serum daily for 6 days, the mean volume of the circulating platelets increased daily. On day 6, the bone marrow meg ploidy increased over that observed at 24 h, but was less than the percentage increase in the mean platelet volume [lo]. In addition, a study of chronic hypoxia as the condition causing thrombocytopenia in rats demonstrated that the platelet count decreased gradually over 35 days, while the mean platelet volume was increased. Over the same period, meg modal ploidy increased gradually, moving from a control modal value of 16N to 32N after 35 days [15]. Therefore, these studies demonstrate that a chronic stimulus causes a prolonged response in meg ploidy and mean platelet volume. When a single dose of vincristine was given i.v. to rats, thrombocytosis occurred without a preceding thrombocytopenia. This was maximal at day 5, and platelet count returned to normal by day 8. Mean platelet volume was not changed from control values at any time. Meg ploidy was increased from a control modal value of 16N to 32N by day 4. On day 3, 128N cells were found. The latter were never seen in control marrows. It is possible that the increased meg polyploidy induced by vincristine is mediated through its effect on microtubule assembly, since such agents have been demonstrated to promote formation of similar restitution nuclei in Allium cepa [16]. Therefore, the effect of vincristine is to increase platelet count and meg ploidy without a change in platelet volume. It may be suggested that the presence of new large platelets mixed with the already circulating platelets of normal size may be difficult to detect, as opposed to models where most of the circulating platelets are destroyed, allowing the easy detection of the size of the new entrants into the circulation. However, because of the magnitude of the thrombocytosis after administration of vincristine, it is unlikely that it was associated with an increase in platelet volume. In summary, these experiments indicate that when increased platelet destruction occurs alone (very soon after induction of immunothrombocytopenia)there is an increase in the mean volume of the circulating platelets, although there is no change in meg DNA content. When increased platelet production occurs alone (after administrationof vincristine), meg ploidy is increased, but the mean volume
GladwidMartin
294
of the circulating platelets is unaltered. However, when platelet production and destruction are both stimulated (in chronic hypoxia and repeated administration of anti-platelet serum), both mean platelet volume and meg ploidy are increased. From these experimental results, it has been postulated that platelet volume and meg ploidy are under separate hormonal control such that they may be stimulated independently or together [17]. Changes in platelet volume would occur only after an alteration in the rate of platelet destruction, whereas an altered meg ploidy, which is accompanied by production of copious amounts of cytoplasm, would be associated with a change in the rate of platelet production. This hypothesis is based upon the assumption that the changes in meg size and DNA content measured in the bone marrow are representative of those cells which have produced the circulating platelets. In particular, since there is evidence to indicate that platelets are produced in the pulmonary vasculature by fragmentation of megs which have migrated from the bone marrow, it is necessary to assume that the cells measured in the marrow are representative of pulmonary megs [18-201. Furthermore, the measured ploidy distribution might be influenced by any number of factors such as rates of polyploidization,cell maturation or transit time through the bone marrow. These variables cannot be ascertained from static measurements of meg ploidy and size, and any one may be altered by perturbation of the thrombokinetic equilibrium in the experimental approaches employed. Therefore, these variables require measurement in each experimental model in separate studies in the future.
The Control of Platelet Production in Pathology In men suffering myocardial infarction within 12 h of the onset of chest pain, mean platelet volume was found to be significantly greater than for control patients from the same coronary care unit suffering from chest pain, but with no myocardial infarction [21]. Bone marrow trephine biopsies of the posterior iliac crest were performed on patients 18 days after myocardial infarction and controls [11].It was found that men with infarctionhad larger bone marrow megs, with a wider size range than controls (for myocardial infarction patients [n = 71, mean meg size was 13,879fl f 2,188 fl [range 1,672 fl f 1,077 fl to 46,779 fl f 3,578 fl]; for controls [n = 61, mean meg size was 10,033 fl f 404 fl [range 896 fl f 186 fl to 23,949 fl f 938 fll. In order to elucidate whether the larger megs might be present at the time of the infarction, a second study was undertaken. Bone marrow was obtained from subjects who had suffered sudden unexpected cardiac death, and from age-matched controls who had undergone sudden unnatural death. All biopsies were obtained within 3 h of death [Ill. Mean meg volume and size range was greater in the sudden cardiac death group when compared to controls (n = 11, mean meg size
The Control of Platelet Production
295
12,030 fl f 404 fl; range 938 fl f 87 fl to 40,358 fl f 1,747 fl) for the sudden unnatural death group, mean meg size was 8,973 fl f 321 fl (range 835 fl f 244 fl to 23,559 fl f 2,082 fl). Therefore, from these results it may he hypothesized that in myocardial infarction, large platelets are circulating at the time of the infarction and that large megs may be present in the bone marrow before infarction. Since large platelets are more reactive than smaller ones, their presence in the circulation might be causally related to thrombosis [9]. Certainly, there is a shortened bleeding time in myocardial infarction [22]. In addition, when the bleeding time of myocardial infarction patients was measured following a single oral dose (300 mg) of aspirin, the bleeding time was prolonged more than that for control subjects [23]. This might be interpreted as providing further evidence that platelets in infarction patients are more hemostatically active than in normal subjects. Interestingly, Kristensen and colleagues [24] have demonstrated that even in normal subjects there is an inverse relationshipbetween the bleeding time, meg size and DNA content. If meg ploidy and mean platelet volume are under separate hormonal control, as has been suggested above from observations of perturbed states in animal models, then one can speculate as to the abnormality of the plateletheg control system in myocardial infarction. It is implied that both hormones are active in this pathology since both mean platelet volume and meg ploidy are increasedone stimulating increased meg DNA content and size, the other causing an increased mean platelet volume. This study of myocardial infarction has raised the possibility that an increased meg size and DNA content may be prothrombotic. However, it has also been demonstratedin rabbits and guinea pigs that similar changes in meg size and ploidy can be induced by fixding animals on a high cholesteroldiet [25,26]. In this model, exogenouscholesterolis not taken up by the circulatingplatelets, but by megs which then incorporate it into future platelets [26]. These events are accompanied by developmentof fatty streaks in the aorta. When platelet destruction was increased in these animals using an anti-platelet antibody, there was a further increase in the size and DNA content of the megs [27]. At the same time, atherogenesis is accelerated. Therefore, these observations suggest that meg changes may also be atherogenic. Platelet production was also studied in patients with malignant lymphoma who had no evidence of bone marrow involvement, infarction, splenomegaly, or anemia [28]. Only previously untreated patients were studied. Lymphoma patients had a secondary thrombocytosis when compared to controls (n = 26, mean platelet count 359 f 46 x 109L-', for control subjects n = 11, mean platelet count 297 f 130 x 109L-'). However, there was no difference in mean platelet volume between the groups (mean platelet volume was 6.4 fl f 0.4 fl for lymphoma patients and 6.7 fl f 0.8 fl for controls). Bone marrow trephine biopsies of the posterior iliac crest were then taken from patients suffering from lymphoma during routine staging of their disease.
GladwinlMartin
296
Control bone marrow was obtained from the same anatomical site as from patients undergoingthoracotomy for repair of hiatal hernia. DNA content of megs was not measured, but mean meg sizes were significantly increased in the lymphoma group when compared to controls. Since meg size and DNA content have been shown to be related, it can be assumed that ploidy was increased in this study [8]. If the theory of dual control of platelet volume and meg ploidy is applied to these observations, it can be speculated that the secondary thrombocytosis observed in lymphoma is associated with an increased activity of the hormone that controls ploidy, but not of the one that controls platelet volume.
Conclusions In experimental models of perturbed thrombopoiesis, it has been observed that changes in mean platelet volume and meg ploidy and size can occur independently or simultaneously. Therefore, it has been hypothesized that these variables are under independent hormonal control, with an increased mean platelet volume occurring following an increased rate of platelet destruction and an increased meg size and ploidy occurring in association with an increased demand for platelet production. In myocardial infarction, both mean platelet volume and meg DNA content and size are increased. If the dual control theory is applied, there may be increased activity of both hormones in this pathology. In secondary thrombocytosisassociated with malignant lymphoma, meg size was similarly increased. However, in contrast to myocardial infarction, mean platelet volume was unchanged when compared to controls. This suggests that in lymphoma, there may be increased activity of the hormone that controls meg ploidy, but not of the one that controls platelet volume. The evidence for the minimum of a dual control is derived from experiments in the whole animal. It will be of great interest in the future to see if this evidence can be married with the nature of thrombopoietin [29, 301 and factors that regulate megakaryocytopoiesis in vitro [31].
References 1 McDonald TP, Ode11 'IT Jr, Gossless DG.Platelet size in relation to platelet age. Proc Soc Exp B i d Med 1964;115:684-689. 2 Karpatkin S. Heterogeneity of human platelets. I. Metabolic and kinetic evidence suggestive of young adult platelets. J Clin Invest 1969;48:1073-1082. 3 Ma~tinJF. Platelet heterogeneity in vascular disease. In: Platelet Heterogeneity: Biology and Pathology. London: Springer-Verlag, 1990:205-226. 4 Martin JF, Plumb J, Kilbey RS, Kishk YT. Changes in volume and density of platelets in myocardial infarction. Br Med J 1983;287:456-459.
The Control of Platelet Production
297
5 Cameron HA, Philips R, Ibbotson RM, Carson PHM. Platelet size in myocardial infarction. Br Med J 1983;287:449-451. 6 Martin JF, Daniels TD, Trowbridge EA. Acute and chronic changes in platelet volume and count after cardiopulmonary bypass induced thrombocytopenia in man. Thromb Haemost 1987;57 55-5 8. 7 Paulus JM. Platelet size in man. Blood 1975;46:321-336, 8 Levine RF, Hazzard KC, tamberg JD. The significanceof megakaryocyte size. Blood 1982;60:1122-1131. 9 Martin JF,Trowbridge EA, Salmon GL, Plumb J. The biological significance of platelet volume: its relationship to bleeding time, platelet thromboxane Bz production and megakaryocyte nuclear DNA concentration. Thromb Res 1983;32:443-460. 10 Martin JF, Trowbridge EA, Salmon GL, Slater DN. The relationship between platelets and megakaryocyte volumes. Thromb Res 1982;28:447-459. 11 Trowbridge EA, Slater DN, Kishk YT, Woodcock BW, Martin IF. Platelet production in myocardial infarction and sudden cardiac death. Thromb Haemost 1984;52; 167-171. 12 Queisser U,Queisser W, Spiertz B. Fblyploidization of megakaryocytes in normal humans, in patients with idiopathic thrombocytopeniaand with pernicious anaemia. Br J Haematol 1971;20:489-501. 13 Trowbridge EA, Warren CW,Martin JF. Platelet volume heterogeneity in acute thrombocytopenia. Clin Phys Physiol Meas 1986;7:203-210. 14 Corash L, Chen HY, Levin J, et al. Regulation of thrombopoiesis; effects of degree of thrombocytopeniaon megakaryoqte ploidy and platelet volume. Blood 1987;70: 177-185. 15 Metcalf B, Warren C, Slater D,Trowbridge EA, Martin JF,Barer GR. Changes in platelets and megakaryocytes in simulated h& altitude in rats. Clin Sci l984;67(suppl9):76a. 16 Nag1 W. ‘Restitutioncycles.’ In: Endopolyploidy and Polyteny in Differentiation and Evolution. Amsterdam: ElsevierlNorth Holland, 1978:117-121. 17 Martin JF.The relationship between megakaryocyte ploidy and platelet volume. Blood Cells 1989;15:108-117. 18 Slater DN, Trowbridge EA, Martin JF. The megakaryocyte and thrombocytopenia. A microscopic study which supports the theory that platelets are produced in the pulmonary circulation. Thromb Res 1983;31:163-176. 19 Tmaaard-Pedersen N. The occurrence of megakaryocytes in various vessels and their retention in the pulmonary capillaries in man. Scand J Haematol 1978;21:369-375. 20 Trowbridge EA, Martin JF, Slater DN. Evidence for a theory of physical fragmentation of megakaryocytes implying that all platelets are produced in the pulmonary circulation. Thromb Res 1982;28:461-475. 21 Trowbridge EA, Martin JF. The platelet volume distribution: a signature of the prethrombotic state in coronary heart disease? Thromb Haemost 1987,58:714-717. 22 Milner PC, Martin JF. Shortened bleeding time in acute myocardial infarction and its relation to platelet mass. Br Med J 1985;290:1767-1770. 23 Kristensen SD, Bath PMW, Martin JF. Differences in bleeding time, aspirin sensitivity and adrenaline between acute myocardial infarction and unstable angina. Cardiowc Res 1990;24:19-23. 24 Kristensen SD, Bath PMW, Martin JF. The bleeding time is inversely related to megakaryocyte nuclear DNA content and size in man. Thromb Haemost 1988;59: 357-359. 25 Martin JF, Slater DN, Kishk YT, Trowbridge EA. Platelet and megakaryayte changes in cholesterol-induced experimental atherosclerosis.Arteriosclerosis 1985;5;604-612.
GladwidMartin
298
26 Schick PB, Schick PK. The effect of hypercholesterolaemia on guinea-pig platelets, erythrocytes and megakaryocytes. Biochem Biophys Acta 1985;833:291-302. 27 Kristensen SD, Roberts KM, Kishk YT, Martin JF. Platelet destruction increases megakaryocyte size and DNA content while enhancing atherosclerosis in the hypercholesterolaemic rabbit. Eur J Clin Invest 1990 (in press). 28 Gladwin AM, Trowbridge EA, Slater DN, Reardon D, Martin JF. The size and numbers of bone marrow megakaryocytes in malignant lymphoma and their relationship to abnormalities in platelet count. Am J Hematol 1990 (in press). 29 Hill R, Levin J. Partial purification of thrombopoietin using lectin chromatography. Exp Hematol 1986;14:752-759. 30 McDonald TP, Cottrell M,Clift R, Khouri JA, Long MD. Studies on the purification of thrombopoietin from kidney cell culture media. J Lab Clin Med 1985;106:162-174. 3 1 Williams N, Eger RR, Jackson HM, Nelson DJ. Two factor requirements for murine megakaryocyte colony formation. J Cell Physiol 1982;llO:101-104.
The Control of Megakaryocyte Ploidy and Platelet Production: Biology and Pathology Ann-Marie Gladwin, John E Martin Cardiovascular Research, The Wellcome Research Laboratories, Beckenham, Kent, United Kingdom
Key Words. Platelets Megakaryocytes ondary thrombocytosis
Polyploidy
Vascular disease
*
Sec-
Abstract. Following experimental platelet destruction in animals, large platelets, which are more hemostatically active, are produced before any change in bone marrow megakaryocyte DNA content. When platelet production is stimulated by administration of i.v. vincristine in rats, megakaryoqte ploidy is increased, but mean platelet volume is unchanged. When platelet production and destruction are both stimulated by chronic hypoxia or administration of anti-platelet serum, mean platelet volume and megakaryocyte DNA content are both increased. Since platelet volume is determined primarily at thrombopoiesis, these results imply that mean platelet volume and megakaryocyte DNA content are under separate hormonal control. Therefore, it has been postulated that changes in mean platelet volume occur following changes in platelet production rate, whereas changes in megakaryocyte ploidy are associated with an increased rate of platelet production. In myocardial infarction, platelets have increased mean volume and reduced bleeding time more than in controls. In addition, men with myocardial infarction have increased megakaryocyte size and increased DNA content when compared to controls. These changes are similar to those observed in rabbits follawing cholesterol feeding. If megakaryocyte polyploidy and mean platelet volume are under separate hormonal control, this suggests that in myocardial infarction, both hormones are active-one stimulating an increased platelet size, the other stimulating the increased megakaryocyte DNA content. In contrast, patients with lymphoma exhibiting a secondary thrornbocytosis have no change in mean platelet volume. However, these subjects also have larger bone marrow megakaryocytes when compared to controls. The relation between megakaryocyte size and ploidy implies that the DNA content of these cells is increased in lymphoma. When the dual control theory of platelet production is applied, it suggests that in malignant lymphoma there is activation of the control mechanism which governs ploidy, but not of the one which controls mean platelet volume. Correspondence: Ann-Marie Gladwin, Ph.D., Cardiovascular Research, The Wellcome Research Laboratories, Langley Court, South Eden Park Road, Beckenham, Kent BR3 3BS, UK. Received May 18, 1990;accepted for publication May 18, 1990. 0737-1454/90/$2.0010 0 AlphaMed Press
GladwinIMartin
292
Introduction Circulating blood platelets are heterogeneousin size, density and reactivity. In 1964, McDonald [I] suggested that size heterogeneity reflected platelet aging in the circulation, so that young platelets were large platelets which decreased in size as they aged. Since platelet size and density were found to be linked, Kalpatkin [2] extended the aging hypothesis of McDonald to suggest that platelet density decreased with platelet age. However, evidence has now been presented by a number of groups suggesting that platelets do not change their size or density as they age in the circulation. It is now widely believed that platelet heterogeneity is determined primarily at thrombopoiesis [3]. Several studies have demonstratedthat the mean volume of circulating platelets is altered in a variety of situations. For example, following myocardial infarction or coronary artery bypass grafting, and in idiopathic thrombocytopenia purpura, platelets have an increased mean volume [4-71. If platelet changes are determined at thrombopoiesis, such alterations in size might be expected to be preceded by changes in their parent cell, the rnegakaryocyte (meg). These large cells, produced in bone marrow, undergo endoreduplicationas they mature, resulting in a marked degree of heterogeneity, in both DNA content and size [8]. At times of decreased platelet number in man and in animal models, it has been observed that the meg population can increase their mean ploidy and size [9-121. The importance of these changes in regulating platelet production is not fully understood. However, studies of the changes in both megs and platelets in a variety of animal models is enabling elucidation of the control system of platelet production. This is of importance in allowing an understanding of the origin of platelet abnormalities in a variety of human pathological states.
The Relationship between Platelet Volume and Meg Ploidy in Animal Models It has been demonstrated that when platelets are destroyed in the circulation following a single injection of an anti-platelet serum, the new platelets which are produced 24 h afterwards have an increased mean size [lo]. These changes in animals are similar to those observed in man following a decreased platelet count associated with cardiopulmonary bypass [6]. Investigationof megakaryocyte nuclear DNA content following induction of immunothrombocytopeniahas also demonstrated an increase by 24 h and shows a maximal change at day 3 [9]. These observations were originally interpreted as suggesting that megs with an increased DNA content are responsible for the production of larger platelets. However, more recently, mean platelet volume and meg DNA content have been measured shortly after induction of immunothrombocytopenia. Trowbridge [13] and Corash [14] identified that large platelets were present by 2 h in the rat and by 8 h in the mouse, respectively. In both of these
The Control of Platelet Production
293
studies, the mean DNA content of the megs was unchanged at those times when compared to normal animals. If it can be discounted that this effect is mediated via release of larger platelets from the splenic pool, this indicates that the large platelets produced following thrombocytopeniaare produced from megs with an unchanged DNA content. Since changes in platelet volume are determined at thrombopoiesis, these studies would imply that the rapid increase in the size of platelets produced following thrombocytopenia must result from a change in the fragmentation of the meg cytoplasm. This may be part of the control system’s response to the hemostatic deficit of thrombocytopenia since such larger platelets produce more thromboxane B2per unit volume of cytoplasm and reduce bleeding time per unit volume more than controls [9]. When platelet destruction was prolonged by giving animals injectionsof antiplatelet serum daily for 6 days, the mean volume of the circulating platelets increased daily. On day 6, the bone marrow meg ploidy increased over that observed at 24 h, but was less than the percentage increase in the mean platelet volume [lo]. In addition, a study of chronic hypoxia as the condition causing thrombocytopenia in rats demonstrated that the platelet count decreased gradually over 35 days, while the mean platelet volume was increased. Over the same period, meg modal ploidy increased gradually, moving from a control modal value of 16N to 32N after 35 days [15]. Therefore, these studies demonstrate that a chronic stimulus causes a prolonged response in meg ploidy and mean platelet volume. When a single dose of vincristine was given i.v. to rats, thrombocytosis occurred without a preceding thrombocytopenia. This was maximal at day 5, and platelet count returned to normal by day 8. Mean platelet volume was not changed from control values at any time. Meg ploidy was increased from a control modal value of 16N to 32N by day 4. On day 3, 128N cells were found. The latter were never seen in control marrows. It is possible that the increased meg polyploidy induced by vincristine is mediated through its effect on microtubule assembly, since such agents have been demonstrated to promote formation of similar restitution nuclei in Allium cepa [16]. Therefore, the effect of vincristine is to increase platelet count and meg ploidy without a change in platelet volume. It may be suggested that the presence of new large platelets mixed with the already circulating platelets of normal size may be difficult to detect, as opposed to models where most of the circulating platelets are destroyed, allowing the easy detection of the size of the new entrants into the circulation. However, because of the magnitude of the thrombocytosis after administration of vincristine, it is unlikely that it was associated with an increase in platelet volume. In summary, these experiments indicate that when increased platelet destruction occurs alone (very soon after induction of immunothrombocytopenia)there is an increase in the mean volume of the circulating platelets, although there is no change in meg DNA content. When increased platelet production occurs alone (after administrationof vincristine), meg ploidy is increased, but the mean volume
GladwidMartin
294
of the circulating platelets is unaltered. However, when platelet production and destruction are both stimulated (in chronic hypoxia and repeated administration of anti-platelet serum), both mean platelet volume and meg ploidy are increased. From these experimental results, it has been postulated that platelet volume and meg ploidy are under separate hormonal control such that they may be stimulated independently or together [17]. Changes in platelet volume would occur only after an alteration in the rate of platelet destruction, whereas an altered meg ploidy, which is accompanied by production of copious amounts of cytoplasm, would be associated with a change in the rate of platelet production. This hypothesis is based upon the assumption that the changes in meg size and DNA content measured in the bone marrow are representative of those cells which have produced the circulating platelets. In particular, since there is evidence to indicate that platelets are produced in the pulmonary vasculature by fragmentation of megs which have migrated from the bone marrow, it is necessary to assume that the cells measured in the marrow are representative of pulmonary megs [18-201. Furthermore, the measured ploidy distribution might be influenced by any number of factors such as rates of polyploidization,cell maturation or transit time through the bone marrow. These variables cannot be ascertained from static measurements of meg ploidy and size, and any one may be altered by perturbation of the thrombokinetic equilibrium in the experimental approaches employed. Therefore, these variables require measurement in each experimental model in separate studies in the future.
The Control of Platelet Production in Pathology In men suffering myocardial infarction within 12 h of the onset of chest pain, mean platelet volume was found to be significantly greater than for control patients from the same coronary care unit suffering from chest pain, but with no myocardial infarction [21]. Bone marrow trephine biopsies of the posterior iliac crest were performed on patients 18 days after myocardial infarction and controls [11].It was found that men with infarctionhad larger bone marrow megs, with a wider size range than controls (for myocardial infarction patients [n = 71, mean meg size was 13,879fl f 2,188 fl [range 1,672 fl f 1,077 fl to 46,779 fl f 3,578 fl]; for controls [n = 61, mean meg size was 10,033 fl f 404 fl [range 896 fl f 186 fl to 23,949 fl f 938 fll. In order to elucidate whether the larger megs might be present at the time of the infarction, a second study was undertaken. Bone marrow was obtained from subjects who had suffered sudden unexpected cardiac death, and from age-matched controls who had undergone sudden unnatural death. All biopsies were obtained within 3 h of death [Ill. Mean meg volume and size range was greater in the sudden cardiac death group when compared to controls (n = 11, mean meg size
The Control of Platelet Production
295
12,030 fl f 404 fl; range 938 fl f 87 fl to 40,358 fl f 1,747 fl) for the sudden unnatural death group, mean meg size was 8,973 fl f 321 fl (range 835 fl f 244 fl to 23,559 fl f 2,082 fl). Therefore, from these results it may he hypothesized that in myocardial infarction, large platelets are circulating at the time of the infarction and that large megs may be present in the bone marrow before infarction. Since large platelets are more reactive than smaller ones, their presence in the circulation might be causally related to thrombosis [9]. Certainly, there is a shortened bleeding time in myocardial infarction [22]. In addition, when the bleeding time of myocardial infarction patients was measured following a single oral dose (300 mg) of aspirin, the bleeding time was prolonged more than that for control subjects [23]. This might be interpreted as providing further evidence that platelets in infarction patients are more hemostatically active than in normal subjects. Interestingly, Kristensen and colleagues [24] have demonstrated that even in normal subjects there is an inverse relationshipbetween the bleeding time, meg size and DNA content. If meg ploidy and mean platelet volume are under separate hormonal control, as has been suggested above from observations of perturbed states in animal models, then one can speculate as to the abnormality of the plateletheg control system in myocardial infarction. It is implied that both hormones are active in this pathology since both mean platelet volume and meg ploidy are increasedone stimulating increased meg DNA content and size, the other causing an increased mean platelet volume. This study of myocardial infarction has raised the possibility that an increased meg size and DNA content may be prothrombotic. However, it has also been demonstratedin rabbits and guinea pigs that similar changes in meg size and ploidy can be induced by fixding animals on a high cholesteroldiet [25,26]. In this model, exogenouscholesterolis not taken up by the circulatingplatelets, but by megs which then incorporate it into future platelets [26]. These events are accompanied by developmentof fatty streaks in the aorta. When platelet destruction was increased in these animals using an anti-platelet antibody, there was a further increase in the size and DNA content of the megs [27]. At the same time, atherogenesis is accelerated. Therefore, these observations suggest that meg changes may also be atherogenic. Platelet production was also studied in patients with malignant lymphoma who had no evidence of bone marrow involvement, infarction, splenomegaly, or anemia [28]. Only previously untreated patients were studied. Lymphoma patients had a secondary thrombocytosis when compared to controls (n = 26, mean platelet count 359 f 46 x 109L-', for control subjects n = 11, mean platelet count 297 f 130 x 109L-'). However, there was no difference in mean platelet volume between the groups (mean platelet volume was 6.4 fl f 0.4 fl for lymphoma patients and 6.7 fl f 0.8 fl for controls). Bone marrow trephine biopsies of the posterior iliac crest were then taken from patients suffering from lymphoma during routine staging of their disease.
GladwinlMartin
296
Control bone marrow was obtained from the same anatomical site as from patients undergoingthoracotomy for repair of hiatal hernia. DNA content of megs was not measured, but mean meg sizes were significantly increased in the lymphoma group when compared to controls. Since meg size and DNA content have been shown to be related, it can be assumed that ploidy was increased in this study [8]. If the theory of dual control of platelet volume and meg ploidy is applied to these observations, it can be speculated that the secondary thrombocytosis observed in lymphoma is associated with an increased activity of the hormone that controls ploidy, but not of the one that controls platelet volume.
Conclusions In experimental models of perturbed thrombopoiesis, it has been observed that changes in mean platelet volume and meg ploidy and size can occur independently or simultaneously. Therefore, it has been hypothesized that these variables are under independent hormonal control, with an increased mean platelet volume occurring following an increased rate of platelet destruction and an increased meg size and ploidy occurring in association with an increased demand for platelet production. In myocardial infarction, both mean platelet volume and meg DNA content and size are increased. If the dual control theory is applied, there may be increased activity of both hormones in this pathology. In secondary thrombocytosisassociated with malignant lymphoma, meg size was similarly increased. However, in contrast to myocardial infarction, mean platelet volume was unchanged when compared to controls. This suggests that in lymphoma, there may be increased activity of the hormone that controls meg ploidy, but not of the one that controls platelet volume. The evidence for the minimum of a dual control is derived from experiments in the whole animal. It will be of great interest in the future to see if this evidence can be married with the nature of thrombopoietin [29, 301 and factors that regulate megakaryocytopoiesis in vitro [31].
References 1 McDonald TP, Ode11 'IT Jr, Gossless DG.Platelet size in relation to platelet age. Proc Soc Exp B i d Med 1964;115:684-689. 2 Karpatkin S. Heterogeneity of human platelets. I. Metabolic and kinetic evidence suggestive of young adult platelets. J Clin Invest 1969;48:1073-1082. 3 Ma~tinJF. Platelet heterogeneity in vascular disease. In: Platelet Heterogeneity: Biology and Pathology. London: Springer-Verlag, 1990:205-226. 4 Martin JF, Plumb J, Kilbey RS, Kishk YT. Changes in volume and density of platelets in myocardial infarction. Br Med J 1983;287:456-459.
The Control of Platelet Production
297
5 Cameron HA, Philips R, Ibbotson RM, Carson PHM. Platelet size in myocardial infarction. Br Med J 1983;287:449-451. 6 Martin JF, Daniels TD, Trowbridge EA. Acute and chronic changes in platelet volume and count after cardiopulmonary bypass induced thrombocytopenia in man. Thromb Haemost 1987;57 55-5 8. 7 Paulus JM. Platelet size in man. Blood 1975;46:321-336, 8 Levine RF, Hazzard KC, tamberg JD. The significanceof megakaryocyte size. Blood 1982;60:1122-1131. 9 Martin JF,Trowbridge EA, Salmon GL, Plumb J. The biological significance of platelet volume: its relationship to bleeding time, platelet thromboxane Bz production and megakaryocyte nuclear DNA concentration. Thromb Res 1983;32:443-460. 10 Martin JF, Trowbridge EA, Salmon GL, Slater DN. The relationship between platelets and megakaryocyte volumes. Thromb Res 1982;28:447-459. 11 Trowbridge EA, Slater DN, Kishk YT, Woodcock BW, Martin IF. Platelet production in myocardial infarction and sudden cardiac death. Thromb Haemost 1984;52; 167-171. 12 Queisser U,Queisser W, Spiertz B. Fblyploidization of megakaryocytes in normal humans, in patients with idiopathic thrombocytopeniaand with pernicious anaemia. Br J Haematol 1971;20:489-501. 13 Trowbridge EA, Warren CW,Martin JF. Platelet volume heterogeneity in acute thrombocytopenia. Clin Phys Physiol Meas 1986;7:203-210. 14 Corash L, Chen HY, Levin J, et al. Regulation of thrombopoiesis; effects of degree of thrombocytopeniaon megakaryoqte ploidy and platelet volume. Blood 1987;70: 177-185. 15 Metcalf B, Warren C, Slater D,Trowbridge EA, Martin JF,Barer GR. Changes in platelets and megakaryocytes in simulated h& altitude in rats. Clin Sci l984;67(suppl9):76a. 16 Nag1 W. ‘Restitutioncycles.’ In: Endopolyploidy and Polyteny in Differentiation and Evolution. Amsterdam: ElsevierlNorth Holland, 1978:117-121. 17 Martin JF.The relationship between megakaryocyte ploidy and platelet volume. Blood Cells 1989;15:108-117. 18 Slater DN, Trowbridge EA, Martin JF. The megakaryocyte and thrombocytopenia. A microscopic study which supports the theory that platelets are produced in the pulmonary circulation. Thromb Res 1983;31:163-176. 19 Tmaaard-Pedersen N. The occurrence of megakaryocytes in various vessels and their retention in the pulmonary capillaries in man. Scand J Haematol 1978;21:369-375. 20 Trowbridge EA, Martin JF, Slater DN. Evidence for a theory of physical fragmentation of megakaryocytes implying that all platelets are produced in the pulmonary circulation. Thromb Res 1982;28:461-475. 21 Trowbridge EA, Martin JF. The platelet volume distribution: a signature of the prethrombotic state in coronary heart disease? Thromb Haemost 1987,58:714-717. 22 Milner PC, Martin JF. Shortened bleeding time in acute myocardial infarction and its relation to platelet mass. Br Med J 1985;290:1767-1770. 23 Kristensen SD, Bath PMW, Martin JF. Differences in bleeding time, aspirin sensitivity and adrenaline between acute myocardial infarction and unstable angina. Cardiowc Res 1990;24:19-23. 24 Kristensen SD, Bath PMW, Martin JF. The bleeding time is inversely related to megakaryocyte nuclear DNA content and size in man. Thromb Haemost 1988;59: 357-359. 25 Martin JF, Slater DN, Kishk YT, Trowbridge EA. Platelet and megakaryayte changes in cholesterol-induced experimental atherosclerosis.Arteriosclerosis 1985;5;604-612.
GladwidMartin
298
26 Schick PB, Schick PK. The effect of hypercholesterolaemia on guinea-pig platelets, erythrocytes and megakaryocytes. Biochem Biophys Acta 1985;833:291-302. 27 Kristensen SD, Roberts KM, Kishk YT, Martin JF. Platelet destruction increases megakaryocyte size and DNA content while enhancing atherosclerosis in the hypercholesterolaemic rabbit. Eur J Clin Invest 1990 (in press). 28 Gladwin AM, Trowbridge EA, Slater DN, Reardon D, Martin JF. The size and numbers of bone marrow megakaryocytes in malignant lymphoma and their relationship to abnormalities in platelet count. Am J Hematol 1990 (in press). 29 Hill R, Levin J. Partial purification of thrombopoietin using lectin chromatography. Exp Hematol 1986;14:752-759. 30 McDonald TP, Cottrell M,Clift R, Khouri JA, Long MD. Studies on the purification of thrombopoietin from kidney cell culture media. J Lab Clin Med 1985;106:162-174. 3 1 Williams N, Eger RR, Jackson HM, Nelson DJ. Two factor requirements for murine megakaryocyte colony formation. J Cell Physiol 1982;llO:101-104.
