Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsAP conceived the study, enrolled patients, performed the analysis and wrote the manuscript. AL ran the experiments and measured mitochondrial membrane potential. FF measured respiratory chain enzyme activities and revised the manuscript. table 5 AA participated in the study design, data analysis and revised the manuscript. NG ran the experiments and helped to draft the manuscript. SV participated in data analysis and helped to draft the manuscript. GF performed electron microscopy and revised the manuscript. MM participated in study design, data analysis and revised the manuscript. GPC participated in study design, data analysis and revised the manuscript. GM participated in data analysis and helped to draft the manuscript.

BL ran the experiments and revised the manuscript. LF ran the experiments and revised the manuscript. LG participated in study design, data analysis and revised the manuscript. All authors read and approved the final version of the manuscript.Supplementary MaterialAdditional File 1:Time-dependent effects of a highly toxic dose of metformin on human platelet mitochondrial function. Platelets from healthy donors were incubated in plasma with metformin diluted in saline (16,600 mg/L). (a) Plasma lactate concentration (P = 0.002; one-way repeated measures ANOVA) and (b) the ratio between normally polarized and abnormally depolarized platelet mitochondria (JC-1 fluorescence ratio) (P = 0.035; one-way repeated measures ANOVA) were measured every 24 hours, up to 72 hours.

Data are mean and SD, from three experiments. *P < 0.05 versus time 0 (Holm-Sidak method). ANOVA, analysis of variance; SD, standard deviation.Click here for file(18K, PDF)Additional File 2:Relationship between platelet mitochondrial function and lactate production. Platelets from healthy donors were incubated for 72 hours in plasma with metformin diluted in saline (concentrations ranging from 0 to 16,600 mg/L). Correlation (linear regression analysis) between final plasma lactate levels and (a) platelet complex I (CI) activity expressed relative to citrate synthase (CS) activity (R2 0.54, P = 0.001; n = 16), (b) platelet JC-1 fluorescence ratio (R2 0.37, P = 0.001; n = 32), and (c) platelet oxygen use (R2 0.82, P < 0.001; n = 27) are shown.

Click here for file(17K, PDF)Additional File 3:Dose-dependent effects of metformin on human platelet respiratory chain complex activities. Platelets from healthy donors were incubated in plasma with saline (white bar) Dacomitinib or metformin diluted in saline (concentration: 1.66 mg/L, grey bar; 166 mg/L, dark grey bar; or 16,600 mg/L, black bar). After 72 hours, the activity of (a) complex I (CI) (P = 0.009; one-way ANOVA), (b) complex II and III (CII+III) (P = 0.767; one-way ANOVA) and (c) complex IV (CIV) (P = 0.