Fig. 5: High ploidy LCM, and not SCM, generate platelets.

a Schematic of MK transplant and in vivo platelet formation assay. b Platelets were identified by size and CD41 expression; transplanted RFP⁺ MK gave rise to RFP⁺platelets; non-inject control. c Number of platelets formed from each transplanted MK assuming all MK made equal numbers of platelets (n = 1). d, e) Schematic of platelet challenge and subsequent recovery following anti-platelet serum (APS) administration (n = 3, biological replicates, from one experiment). f Representative flow cytometry plot of BM MK during platelet recovery following an APS challenge (day 4). 8N LCM = 0.02%, 8N SCM = 0.26%, 16N LCM = 0.24%, 16N SCM = 0.57%, 32N LCM = 0.57%, 32N SCM = 0.65%, 64N LCM = 0.13%, 64N SCM = 0.18%. g Frequency and h absolute number of LCM within each ploidy ± APS treatment (day 4 data, APS n = 5, 64N n = 6, other ploidies n = 10, biological replicates, from one experiment). i RFP⁺ platelets in the PB of mice transplanted with RFP⁺ LCM or SCM following APS treatment, assuming all MK made equal numbers of platelets (day 0 n = 1, day 3 n = 1 for SCM and 2 for LCM, day 4 n = 2, biological replicates, from two experiments). j Representative data: tdTomato expression in LCM and k SCM of Pf4-Cre driven tdTomato expressing mice versus control mice. l tdTomato expression in the LCM and SCM of Pf4-Cre⁺ tdTomato expressing mice. m tdTomato expression in platelets of Pf4-Cre driven tdTomato expressing mice versus controls (j–m: n = 9, biological replicates, from two experiments). n Platelet number in MK-LCM^WT/WT and MK-LCM^Δ/Δ PB (MK-LCM^WT/WT n = 22, MK-LCM^Δ/Δ n = 20, biological replicates, from eight experiments). o Number of platelets generated per MK if only LCM generate platelets. Data presented were normalised to MK-LCM^WT/WT mice and assumed a 20 g mouse contained 1.5 ml of PB⁹⁴, MK are randomly distributed in BM¹ and BM from one hind leg accounts for 15% of total BM⁹⁵ (MK-LCM^WT/WT n = 10, MK-LCM^Δ/Δ n = 6, biological replicates, from three experiments). p The relationship between MK-LCM^Δ/Δ reconstitution in the PB and platelet counts (n = 55, biological replicates, from four transplants). q Platelet counts in transplanted mice relative to the proportion of LCM in the BM (exponential 2-phase association, dotted line = 0.001%, n = 54, biological replicates, from four transplants). r RT-PCR analysis of MK mitochondrial and genomic DNA (All n = 4 except 16N LCM n = 3, biological replicates, from two experiments). s Representative Tom20 staining of prospectively isolated MK (2 independent experiments were performed). t Quantification of mitochondrial networks in MK, platelets and HSC (32N LCM n = 9 cells, 16N LCM n = 10 cells, 32N SCM n = 9 cells, 16N SCM n = 7 cells, platelets n = 9, LSK n = 10 cells). u Representative Tom20 staining in platelets and v LSK cells (2 independent experiments were performed). w Representative MitoTrackerGreen staining of sorted MK (2 independent experiments were performed). x Hypothesised platelet release and contribution of single ploidy MK under haemostatic conditions and following the induction of acute thrombocytopenia. y Summary hypothesis of MK maturing from ^MaturingSCM to LCM then ^SpentSCM following platelet release. Scale bar = 5 μm except (w) inset = 10 μm, percentages are of the parent. Statistical analysis used ordinary one-way ANOVA, p < 0.0001 (overall) with individual groups compared using Tukey’s multiple comparisons test, p values as shown (e), two-way ANOVA with Geisser–Greenhouse correction, p < 0.0001 (overall) with individual groups compared using Holm–Sidak multiple comparisons test, p values indicated for (g), two-way ANOVA with Geisser–Greenhouse correction, p = 0.0004 (overall) with individual groups compared using Holm–Sidak multiple comparisons test, p values indicated for (h), unpaired two-sided Student’s t-test, p values indicated for (n–o), linear regression statistics (p) and two-phase exponential association curve (q), ordinary one-way ANOVA, p = 0.0004 (overall) with individual groups compared using Tukey’s multiple comparisons test, p values indicated for (r, t). Source data are provided as a Source Data File. Error bars = SEM.