Advertisement

Lin28b regulates age-dependent differences in murine platelet function

Massiel Chavez Stolla, Seana C. Catherman, Paul D. Kingsley, R. Grant Rowe, Anne D. Koniski, Katherine Fegan, Leah Vit, Kathleen E. McGrath, George Q. Daley and James Palis

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    Reduced P-selectin content in embryonic and fetal platelets correlates with reduced transcripts for P-selectin in fetal megakaryocytes. (A) Integrin (αIIbβ3) activation in unactivated and thrombin-activated primary platelets isolated from E12.5 and E15.5 murine embryos, P1 pups, and adult female dams. Mean ± standard error of the mean (SEM). (B) Simultaneous analysis of samples in panel A for surface expression of P-selectin (CD62P) in unactivated and thrombin-activated primary platelets. Significance was determined using a 2-way ANOVA followed by a Bonferroni posttest. *P < .001 vs activated maternal sample, n ≥ 3. (C) Immunofluorescence microscopy of fetal (E15.5) and adult platelets reveals fewer punctae of P-selectin–positive granules in fetal platelets relative to adult controls. Deconvolved merged maximum intensity projection images are shown. Scale bar, 5 μm. Similar patterns were observed in 3 independent experiments. (D) Transcript levels of P-selectin (Selp) relative to platelet factor 4 (Pf4) in primary megakaryocytes isolated from E15.5 fetal livers, and adult bone marrow. Extremely low levels of P-selectin transcripts were found in primary megakaryocytes harvested from E15.5 fetal liver. Significance was determined using an unpaired 2-tailed Student t test, *P < .05; n = 4. MFI, median fluorescence intensity.

  • Figure 2.

    Fetal/neonatal platelets contain a-granules and secrete PF4 on stimulation. (A) Immunofluorescence microscopy of VWF and VEGF protein reveals punctate patterns of expression in resting adult and fetal (E15.5) platelets. Deconvolved merged maximum intensity projection images are shown. Scale bar, 5 μm. Similar patterns were observed in 3 independent experiments. (B) Significant amounts of Pf4 were released after activation of P1 and adult platelets. Platelet activation with thrombin (0.5 U/mL) increases the concentration of Pf4 similarly in supernatants from P1 and adult platelets when evaluated by enzyme-linked immunosorbent assay. Significance was determined using a 2-way ANOVA followed by a Bonferroni posttest, *P < .05 between unactivated and activated samples per age group. Mean ± standard error of the mean shown; n = 4.

  • Figure 3.

    Neonatal platelets do not readily associate with granulocytes in vivo or in vitro. (A) Significantly fewer endogenous neutrophil-platelet aggregates are found in the circulation of neonates compared with adults. n = 5. Significance determined using unpaired Student t test P < .05. Leukocytes were defined as CD45+, granulocytes were defined as CD45+/GR-1+ double-positive, and platelets were CD41+. (B) Representative images of neutrophil-platelet rosettes with 1, 3, and 5 platelet spots. CD41 PE is shown in yellow, and GR-1 fluorescein isothiocyanate is shown in purple. Images were collected at 40× magnification. Scale bar, 10 μm. (C) Fewer rosettes were formed when neutrophils were mixed with activated platelets from P1 pups (+P1) when compared with rosette formation with activated platelets from adults (+Adult). Platelets were activated with 0.5 units/mL thrombin before mixing in a 10:1 ratio with bone marrow-enriched neutrophils. Rosettes were defined as a single neutrophil with 2 or more platelets attached. Significance was determined using an unpaired 2-tailed Student t test *P < .05; n = 3.

  • Figure 4.

    Fetal megakaryocytes express high levels of Lin28B and Hmga2 transcripts, as well as several other known Lin28b/Let-7 target genes. (A) Transcript levels of Lin28B relative to small ribosomal subunit 18S (18S) in cultured megakaryocytes from E12.5 fetal livers or adult bone marrow. Significance was determined using an unpaired 2-tailed Student t test *P < .0001; n = 3. (B) Transcript levels of high-mobility group AT-hook 2 (Hmga-2) relative to 18S in cultured megakaryocytes from embryonic day 12.5 fetal livers or adult bone marrow. Significance was determined using an unpaired 2-tailed Student t test *P < .0001; n = 3. (C) Transcript levels of Igf2bp1, Igf2bp2, Igf2bp3, and Arid3 relative to 18S in cultured megakaryocytes from E12.5 fetal livers or adult bone marrow. Significance was determined using 2-way ANOVA. *P < .0001; n = 3.

  • Figure 5.

    Induction of human LIN28B in adult mice reduces platelet P-selectin transcript and cell surface expression. (A) Integrin activation in unactivated and thrombin-activated platelets from mice with doxycycline-induced expression of LIN28B (iLIN28B) and control mice. (B) Surface expression of P-selectin (CD62P) in thrombin-activated platelets from iLIN28B and control mice. Significance was determined using a 2-way ANOVA followed by a Bonferroni posttest ***P < .001; n ≥ 6. (C) Transcript levels of human LIN28B and murine Selp relative to murine PF4 in cultured megakaryocytes from mice with doxycycline-induced expression of human LIN28B (iLIN28B) and control mice. Significance was determined using 2-way ANOVA. *P < .05; **P < .005; n = 3.

  • Figure 6.

    P-selectin is intrinsically regulated in fetal hematopoietic progenitors. (A) Experimental scheme. Kit-positive hematopoietic progenitors were derived from E12.5 fetal liver or adult bone marrow and cultured for 3 days in the presence of TPO. Megakaryocytes were isolated by CD41 selection. For transplant studies, linscakit+ (HPC) cells were sorted from E14.5 fetal livers and from adult bone marrow of UBC-GFP mice and transplanted into GFP recipients. Whole blood from recipients was activated on days 5 and 10 posttransplant, with 0.5 U/mL thrombin in the presence of antibodies for activated αIIbβ3 (JON/A) and surface P-selectin (CD62P). Donor (GFP+) and recipient (GFP) platelet populations were evaluated individually for activated integrin αIIbβ3 and P-selectin surface expression. (B) Transcript levels of P-selectin (Selp) relative to 18S in cultured megakaryocytes. Significance was determined using an unpaired 2-tailed Student t test *P < .05; n = 5. (C) Integrin activation (αIIbβ3) in donor (GFP+) platelets. Analysis was performed using 2-way ANOVA with Bonferroni posttest, n ≥ 3. (D) P-selectin surface expression in donor (GFP+) platelets. Significance was determined using a 2-way ANOVA followed by a Bonferroni posttest, **P < .001; n ≥ 3. (E) Integrin activation (αIIbβ3) in donor (GFP) platelets. Analysis was performed using 2-way ANOVA with Bonferroni posttest, n ≥ 3. (F) P-selectin surface expression in donor (GFP) platelets. No significance was detected using a 2-way ANOVA with Bonferroni posttest, n ≥ 3.

  • Figure 7.

    The onset of P-selectin expression occurs postnatally and is regulated by a developmental transition in HSPCs. (A) P-selectin surface expression in unstimulated and stimulated platelets isolated from various gestational ages postnatally. Values shown are expressed as a percent of CD62P MFI on activated adult platelets. Platelets harvested from murine embryos (E12.5), pups (days 1, 7, 14, 21, and 42 after birth), and non-pregnant adult mice, were activated with thrombin (0.5 units/mL) in the presence of antibodies to detect activated αIIbβ3 (JON/A) and surface P-selectin (CD62P). Significance was determined using a 2-way ANOVA followed by a Bonferroni posttest *P < .01, ** P < .001 versus activated adult P-selectin surface expression; n ≥ 3. (B) Representative histogram of MFI of CD62P (P-selectin) on activated platelets from P1, P14, P42, and adult. Postnatally, a single platelet population transitions from expression low levels of P-selectin (P1) to higher levels of P-selectin (Adult). (C) Experimental scheme. Linsca+kit+ (HSPCs) cells were sorted from E14.5 fetal livers and adult bone marrow of UBC-GFP mice and transplanted into sublethally irradiated GFP recipients. Whole blood from recipients was activated on days 10, 14, 28, and 42 posttransplantation with 0.5 U/mL thrombin for 10 minutes in the presence of antibodies for activated αIIbβ3 (JON/A) and surface P-selectin (CD62P). HSPCs were taken from E14.5 fetal livers; therefore, days 10 (D10), D14, D28, and D42 posttransplant have equivalent gestational ages of P3.5, P7.5, P21.5, and P35.5, respectively. Donor (GFP+) and recipient (GFP) platelet populations were evaluated individually for activated integrin αIIbβ3 and P-selectin surface expression. (D) Integrin activation (αIIbβ3) in untreated and thrombin-treated donor (GFP+) platelets. Comparable amounts of integrin activation were seen in platelets produced from both bone marrow (BM)–derived and fetal liver (FL)–derived donor HSPCs on days 10, 14, 28, and 42 after transplantation. Analysis performed using 2-way ANOVA with Bonferroni posttest, n ≥ 3. (E) P-selectin surface expression in untreated and thrombin-treated donor (GFP+) platelets. On days 10 and 14 posttransplant, FL-derived HSPCs produce platelets with low levels of P-selectin surface expression after activation relative to activated donor platelets from BM-derived HSPCs. By D28 posttransplant, platelets produced from either BM-derived HSPCs or FL-derived LSKs have comparable P-selectin expression after activation. Platelets produced by BM-derived HSPCs maintain comparable levels of P-selectin surface expression on all days after transplantation, suggesting developmental changes in FL HSPC’s result in the production of platelets with gradually increasing levels of P-selectin surface expression. Significance was determined using a 2-way ANOVA followed by a Bonferroni posttest *P < .001; n ≥ 3. (F) Integrin activation (αIIbβ3) in untreated and thrombin-treated recipient (GFP) platelets. Comparable amounts of integrin activation were found on platelets of the recipient mouse after transplantation of bone marrow or fetal-liver derived LSKs. (G) P-selectin surface expression in untreated and thrombin-treated recipient (GFP) platelets. Comparable amounts of P-selectin surface expression were found on platelets of the recipient mouse after transplantation of BM- or FL-derived LSK (Linsca+kit+). No significance was detected using a 2-way ANOVA with Bonferroni posttest, n ≥ 3.