Megakaryocytes (MKs) release large PLT-like intermediates called prePLTs into sinusoidal blood vessels. PrePLTs will convert into barbell-shaped proPLTs in vitro to undergo repeated abscissions that yield individual circulating PLTs. These observations predict the presence of circular-prePLTs and barbell-proPLTs in human blood and highlight two fundamental questions in PLT biology that remain unanswered: what forces determine barbell-proPLT formation and how is the final PLT size established. We present here three new insights into the terminal mechanisms of PLT production. First, we accurately quantify circular-prePLTs and barbell-proPLTs in human blood from high resolution fluorescence images (4±3% and 0.05±0.02% of PLT counts, respectively; n=4) by a novel laser scanning cytometry assay. This advancement overcomes the quantification limitations of microscopy and qualitative limitations of flow cytometry that have previously been employed, suggests that once prePLTs convert into barbell-proPLTs, abscission is rapid, and supports the paradigm that the final stages of PLT production occur in the vasculature. These values are comparable in mice, and precise measurements of mouse proPLT culture intermediates and PLT-rich plasma revealed a minimum diameter for circular-prePLTs and maximal diameter for PLTs of ∼2.5μm. Second, given that prePLTs (>2.5μm diameter) can undergo barbell-proPLT conversion and PLTs (1.5μm diameter) cannot, we explored the possibility that force constraints resulting from cortical microtubule (MT) band diameter and thickness determine barbell-proPLT formation. To test this hypothesis, we studied platelet fission under conditions that manipulated the diameter and thickness of the PLT MT coil. PLT intermediates were isolated from whole blood and used to examine the ultrastructure and MT cytoskeleton of these cells by thin-section and negative stain electron microscopy, respectively. Surprisingly, circular-prePLTs and barbell-proPLTs were shown to contain MT coils of equivalent thickness to those of circulating PLTs, and varied only in diameter and shape. To artificially increase PLT diameter, PLT-rich plasma was cultured at 37°C for 6 hours. PLTs stored at 22°C do not show MT coil enlargement. In both human and mice, artificially increasing MT coil diameter resulted in increased barbell-proPLT conversion. Likewise, PLTs from mice treated with rabbit anti-mouse platelet serum (RAMPS) to generate large PLTs (∼3 μm diameter) resulted in increased barbell-proPLT formation relative to non-RAMPS controls. In contrast, FilaminA knock-out PLTs (∼3 μm diameter), which contain thicker MT coils (12 MTs versus 7 in normal PLTs) revealed significantly fewer numbers of barbell-proPLTs. Taken together, these data predict that twisting MT-based forces driving circular-prePLT to barbell-proPLT conversion may be governed by two major biophysical properties: (1) MT coil diameter and (2) MT coil thickness. Finally, we provide a mathematical model of circular-prePLT to barbell-proPLT conversion that is based on the biophysical properties of the marginal MT coil and membrane-limited cell volume of the PLT. This model confirms that MT coil diameter, regulated by temperature-dependent MT polymerization and/or sliding, and MT coil thickness, regulated by marginal MT number, drive barbell-proPLT formation and limit maximal PLT size. A better understanding of the cytoskeletal mechanisms governing PLT formation will lead to improved therapies for thrombocytopenia, while the ability to control this process in vitro could result in an important source of PLTs for infusion.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.