Platelet factor 4 (PF4), a member of the C-X-C family of chemokines, is abundantly expressed in megakaryocytes (Megs) and was thought to be stored in alpha-granules until released by activated platelets. We now show that PF4 has a more complex life-cycle, being released during megakaryopoiesis and then undergoing reuptake into Megs. This recycling may underlie PF4's physiologic role as a negative paracrine regulator of megakaryopoiesis and contribute to the regulation of platelet count at steady state or after recovery from bone marrow injury. When cultured in serum-free media, both murine and human bone marrow CD34+-derived Megs release detectable PF4 into the culture media starting at Day 3 of culture. We examined whether released PF4 is taken up by Megs in culture and whether this process is low-density lipoprotein receptor related protein-1 (LRP1)-dependent, since our previous studies had identified LRP1 as a potential receptor for PF4 on the megakaryocyte surface. Consistent with the known role of PF4 as a negative paracrine regulator, exposure of PF4null murine (m) Megs in culture to hPF4 decreases final cell numbers (8 × 104 hPF4 exposed cells/mL vs. 14 × 104 control cells/mL, p<0.01) and increases LRP1 expression on the surface of megakaryocytes/megakaryocyte precursors by >2-fold after 48 hrs in culture (p<0.01). Treatment of cells with anti-PF4 or anti-LRP1 antibodies blocked the PF4 effect on megakaryopoiesis but did not interfere with the increased expression of LRP1 on Megs after PF4 exposure. Upon exposing PF4null bone marrow-derived Megs to hPF4 (25 μg/ml) for 24 hrs, hPF4 bound to the surface of the cells; however, washing these cells with 1000 units/mL of heparin removed this surface-bound hPF4 and allowed us to measure uptake of hPF4 into these PF4null Megs. PF4 uptake by Megs was LRP1-dependent, as treatment with anti-LRP1 antibody or Receptor Associated Protein (RAP, an inhibitor of LRP family proteins) markedly decreased uptake of PF4 (95±32 IU/106 cells for PF4 treatment alone vs. 32±13 IU/106 cells for anti-LRP1 treatment, p<0.05 and 49±26 IU/106 cells for RAP treatment, p<0.01). Knockdown of LRP1 by an anti-LRP1 shRNA, which decreases LRP1 but does not completely eliminate it, also decreased PF4 uptake by Megs by 79% (41±33 IU/106 cells for control treated cells vs. 9±8 IU/106 cells for anti-LRP1 treated cells, p=0.01) and partially blocked the PF4-mediated increase in surface LRP1 expression. Using immunofluorescence microscopy, we show that internalized hPF4 in PF4null Megs co-localizes with P-selectin, an alpha-granule marker. When the Megs are treated with thrombin, the internalized PF4 is released into the medium and depleted from the cell lysates. Platelet basic protein (PBP), a closely related chemokine that is also stored in alpha granules, was not taken up by Megs in similar analyses. In summary, our study demonstrates that PF4 is not only synthesized by Megs but is also released and available for LRP1-dependent reuptake into alpha-granules. Notably, this cycling was not observed for another alpha granule chemokine, PBP. Whether the uptake is related to PF4's negative paracrine effect on megakaryopoiesis, whether other alpha granule proteins undergo similar recycling, and whether newly synthesized PF4 is first stored in alpha granules or is constitutively secreted will be addressed in future studies.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.