Advertisement

TNF-α–driven inflammation and mitochondrial dysfunction define the platelet hyperreactivity of aging

Pavel Davizon-Castillo, Brandon McMahon, Sonia Aguila, David Bark, Katrina Ashworth, Ayed Allawzi, Robert A. Campbell, Emilie Montenont, Travis Nemkov, Angelo D’Alessandro, Nathan Clendenen, Lauren Shih, Natalie A. Sanders, Kelly Higa, Allaura Cox, Zavelia Padilla-Romo, Giovanni Hernandez, Eric Wartchow, George D. Trahan, Eva Nozik-Grayck, Kenneth Jones, Eric M. Pietras, James DeGregori, Matthew T. Rondina and Jorge Di Paola

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    Aging associates with elevated TNF-⍺ plasma levels and platelet hyperreactivity in mice and humans. (A) TNF-⍺ plasma levels from heathy young (8-12 weeks) and old (>72 weeks) C57BL/6J mice (n = 6-8; mean plus or minus SEM). Hematological profile of (B) platelets, (C) hemoglobin (Hgb), and (D) leukocytes from healthy young and old C57BL/6J mice (n = 8-20 mice per group; mean plus or minus SEM). (E) Platelet activation analysis by flow cytometry of washed platelets from young (8-12 weeks) and old (>72 weeks) mice stimulated with thrombin (0.1 IU/mL) as determined by activation of the ⍺IIbβ3 integrin (n > 3 per experiment with 3-5 biological replicates). Mann-Whitney test. (F) Platelet phosphatidylserine (PS) exposure 5 minutes after activation with thrombin (0.1 IU/mL) of washed platelets from young (8-12 weeks) and old (>72 weeks) C57BL/6J mice (n = 6; mean plus or minus SEM) using bovine fluorescein isothiocyanate (FITC)-lactadherin. Student t test. (G) Assessment of platelet adhesion and clot formation over collagen-coated slides under physiological flow conditions (650 s−1) of reconstituted whole blood. Normal pooled murine plasma and erythrocytes were mixed with washed platelets from either young (8-12 weeks) and old (>72 weeks) mice at the same concentration. Representative TEM of clots formed over collagen-coated slides by TEM (n = 5-6; mean plus or minus SEM). Mann-Whitney test. (H) Activation profile of the ⍺IIbβ3 integrin in human washed platelets from younger (mean age, 35.0 ± 10.1 years; n = 25) and older (mean, 79.5 ± 8.8 years; n = 25) healthy volunteers. Mann-Whitney test. *P < .05; **P < .01; and ***P < .001. max, maximum; NS, not significant.

  • Figure 2.

    Analysis of the bone marrow compartments between young and old mice show significant cellular and transcriptomic differences. (A) Age-associated differences of the hematopoietic bone marrow compartments from young and old mice (n = 5 mice per group; mean plus or minus SEM). (B) Old mice exhibit higher numbers of megakaryocyte (MK)-committed progenitors (MkPs) (n = 5 mice per group; mean plus or minus SEM). (C) Megakaryocyte ploidy analysis of bone marrow megakaryocytes from young and old mice (n = 5 mice per group; mean plus or minus SEM). (D) scRNA-seq of native bone marrow megakaryocytes from young and old mice (n = 4-5 mice per group) identifies 7 clusters of cells at different maturation stages. (E) Identification of megakaryocyte maturation regions (MK regions 1-7) based on megakaryocyte-specific transcripts. (F) Representative heat map of the mitochondrial dysfunction IPA gene pathway across the 7 megakaryocyte regions. (G) IPA identifies significant differentially regulated mitochondrial, metabolic, and inflammatory gene pathways in the transcriptome of young and old bone marrow megakaryocytes. *P < .05; **P < .01; and ***P < .001. Two-tailed t test and Mann-Whitney test. LT-HSC, long-term hematopoietic stem cell; fMLP, N-Formylmethionyl-leucyl-phenylalanine; ST-HSC, short-term hematopoietic stem cell.

  • Figure 3.

    Platelets from old mice exhibit significantly higher mitochondrial mass. (A-B) Quantification of platelet mitochondrial mass by TEM, flow cytometry (C), and by (D) immunoblotting of the mitochondrial protein Tom20 (n = 3-6 mice per group; mean plus or minus SEM). MFI, mean fluorescence intensity.

  • Figure 4.

    Platelets from young and old mice have a distinctive mitochondrial respiration and metabolome. (A) Bioenergetics Seahorse analysis of platelets from young and old mice shows that platelets from old mice have significantly higher respiration upon activation with 0.1 U/mL thrombin (n = 4-6 mice per group; mean plus or minus SEM). (B) Platelet metabolomics experimental approach. (C) Partial least squares discriminant analysis (PLS-DA) of platelets from young and old mice at resting and activated state shows distinctive metabolomic profiles of platelets due to age and activation (n = 4 mice per group; plus or minus SEM). (D) Pentose phosphate pathway metabolites of platelets from old mice are elevated at baseline and show preferential glycolytic metabolism at baseline. (E) TCA cycle upstream metabolites glucose, pyruvate, and lactate between platelets of young and old mice. (F) Differences in the adenylates AMP, ADP, and ATP between platelets from young and old mice. *P < .05; **P < .01; and ***P < .001. Student t test and Mann-Whitney test. AU, arbitrary unit; FCCP, carbonylcyanide p-trifluoromethoxyphenylhydrazone; LC-MS, liquid chromatography–mass spectrometry; R/A, rotenone/antimycin A.

  • Figure 5.

    Platelets from young mice exposed to TNF-α in vivo for 20 consecutive days become hyperreactive. (A) Experimental approach. Representative flow cytometry assessment of the activation of the ⍺IIbβ3 integrin of washed platelets (plts) from young mice treated with TNF-⍺ or control (ctrl) for 20 days after activation with (B) thrombin (0.1 U/mL) or (C) convulxin (50 ng/mL), (n = 4-6; mean plus or minus SEM). (D) Whole blood microfluidics assay. Platelet adhesion to collagen-coated slides under flow of young mice treated with TNF-⍺ or control for 20 days (n = 4; mean plus or minus SEM). (E) Representative flow cytometry assessment of the activation of the ⍺IIbβ3 integrin of washed platelets from TNFΔARE or littermate controls (n = 3; mean plus or minus SEM). (F) Representative flow cytometry assessment of the activation of the ⍺IIbβ3 integrin of washed platelets from young mice treated with IL-1β or control for 20 days (n = 3; mean plus or minus SEM). (G) Whole blood microfluidics assay. Platelet adhesion to collagen-coated slides under flow of young mice treated with IL-1β or control for 20 days (n = 4; mean plus or minus SEM). (H) Flow cytometry assessment of the activation of the ⍺IIbβ3 integrin of washed platelets from young C57BL/6J and p55/p75 KO mice treated with TNF-⍺ or control for 20 days after activation with thrombin (0.1 U/mL) or convulxin (50 ng/mL; n = 4-6; mean plus or minus SEM). (I) Microfluidics analysis of whole blood flowed over collagen-coated slides of young C57BL/6J and p55/p75 KO mice treated with TNF-⍺ or control for 20 days (n = 4; mean plus or minus SEM). *P < .05; **P < .01; and ***P < .001. Student t test and Mann-Whitney test. IP, intraperitoneal.

  • Figure 6.

    Jak2 V617F+ MPN is associated with increased circulating levels of TNF-α and higher platelet mitochondrial mass. (A) Plasma TNF-⍺ levels of healthy controls and Jak2 V617F+ MPN patients. (B) Representative immunoblot (left) and biochemical quantification (right) of the mitochondrial protein Tom20 between healthy controls and Jak2 V617F+ MPN patients. (C) Representative TEM images of platelets from healthy controls and Jak2 V617F+ MPN patients with black arrows depicting the significantly higher mitochondrial content in platelets from Jak2 V617F+ MPN patients (n = 4-12; mean plus or minus SEM). *P < .05; **P < .01; and ***P < .001. Student t test and Mann-Whitney test.

  • Figure 7.

    In vivo neutralization of TNF-α decreases platelet hyperreactivity and restores mitochondrial mass in aged mice. (A) Experimental approach. (B) Decreased plasma levels of TNF-⍺ in old mice treated with anti–TNF-⍺ antibody (n = 4-5 mice per group; mean plus or minus SEM). (C) Platelet counts did not significantly change between old mice treated with anti–TNF-⍺–neutralizing antibodies or isotype control (n = 5 mice per group; mean plus or minus SEM). (D) Inhibition of the ⍺IIbβ3 integrin by anti–TNF-⍺–neutralizing antibodies as measured by FACS (n = 4 mice per group; mean plus or minus SEM). (E) Platelets from old mice treated with by anti–TNF-⍺–neutralizing antibodies show significantly decreased platelet adhesion over collagen-coated slides under flow at 650 s−1 (n = 4-5 mice per group; mean plus or minus SEM). (F) Mitochondrial mass assessment after anti–TNF-⍺–neutralizing antibodies (n = 3-4 mice per group; mean plus or minus SEM). *P < .05; **P < .01; and ***P < .001. Student t test and ANOVA.

Tables

  • Table 1.

    Characteristics of healthy younger and older adult controls

    Healthy youngerFrail/prefrail olderP
    Mean ± SDn = 25n = 25
    Age, y35 ± 578.0 ± 8.4<.001
    % Male43.4827.27
    BMI, kg/m225.82 ± 6.0726.46 ± 5.77.671
    Hemoglobin, g/dL14.46 ± 1.4213.04 ± 1.97.004
    Platelets, × 109/L245.71 ± 38.03250.64 ± 72.83.792
    WBC, × 109/L5.35 ± 0.587.64 ± 1.59<.001
    INR1.03 ± 0.081.45 ± 0.44<.001
    Prothrombin time, s13.42 ± 0.5816.07 ± 2.79<.001
    Aspirin use, %8.6963.64
    Statin use, %072.73
    Clopidogrel use, %04.54
    Warfarin use, %018.18
    • BMI, body mass index; INR, international normalized ratio; SD, standard deviation; WBC, white blood cell.