Reduced miR-26b Expression in Megakaryocytes and Platelets Contributes to Elevated Level of Platelet Activation Status in Sepsis

Abstract

In sepsis, platelets may become activated via toll-like receptors (TLRs), causing microvascular thrombosis. Megakaryocytes (MKs) also express these receptors; thus, severe infection may modulate thrombopoiesis. To explore the relevance of altered miRNAs in platelet activation upon sepsis, we first investigated sepsis-induced miRNA expression in platelets of septic patients. The effect of abnormal Dicer level on miRNA expression was also evaluated. miRNAs were profiled in septic vs. normal platelets using TaqMan Open Array. We validated platelet miR-26b with its target SELP (P-selectin) mRNA levels and correlated them with clinical outcomes. The impact of sepsis on MK transcriptome was analyzed in MEG-01 cells after lipopolysaccharide (LPS) treatment by RNA-seq. Sepsis-reduced miR-26b was further studied using Dicer1 siRNA and calpain inhibition in MEG-01 cells. Out of 390 platelet miRNAs detected, there were 121 significantly decreased, and 61 upregulated in sepsis vs. controls. Septic platelets showed attenuated miR-26b, which were associated with disease severity and mortality. SELP mRNA level was elevated in sepsis, especially in platelets with increased mean platelet volume, causing higher P-selectin expression. Downregulation of Dicer1 generated lower miR-26b with higher SELP mRNA, while calpeptin restored miR-26b in MEG-01 cells. In conclusion, decreased miR-26b in MKs and platelets contributes to an increased level of platelet activation status in sepsis.

Keywords: MEG-01; SELP; inflammation; megakaryocyte; microRNA; platelet activation; sepsis.

Figure 1

Evaluation of the level of platelet activation status in septic patients (n = 21) via surface P-selectin positivity (A) and soluble P-selectin levels (B) measured in some selected plasma samples (n = 10/group) in comparison to controls. Mean platelet volume values were also analyzed to estimate the pool of larger and younger platelets (C). All these parameters were significantly higher than controls, suggesting an enhanced level of platelet activation status after the development of sepsis. Dots represent single results, and median values are depicted. Mann-Whitney U test was performed for comparison.

Figure 2

Analysis of abnormal platelet miRNA expression induced by sepsis. First, miRNAs were profiled by TaqMan Open Array in three septic and control samples (A). Out of 390 detected, there were 121 significantly downregulated and 61 upregulated miRNAs in septic platelets. Platelet miR-26b (B) was validated in both study groups (n = 21/group), showing attenuated levels in sepsis. Alterations in miRNA expression were further observed in the septic cohort in relation to disease severity (C) and sepsis-related mortality (D). Levels of miR-26b reflected septic shock and early death. Dots represent single expression values. Median values are depicted. Mann–Whitney U test was performed for comparison.

Figure 3

Increased SELP and P-selectin protein expression in septic platelets. Platelets of septic patients (n = 21) showed elevated levels of IL1B (A) and SELP mRNA (B) levels compared to control platelets. There was a tendency for higher platelet SELP expression in those sepsis individuals who had septic shock (C) or died by sepsis (D). In addition, significantly elevated SELP mRNA levels were observed in platelets with larger than normal MPV values (n = 8) (E). Western blot analysis of platelet lysates obtained after 72 h of sepsis onset (n = 5) demonstrated increased expression of P-selectin protein compared to normal samples (F). Mean ± SEM was depicted, * p < 0.05. Mann–Whitney U test or unpaired t-test was performed for the comparisons.

Figure 4

Immunohistochemical staining and analysis of the nuclear factor kappa B (NF-κB) pathway activation in LPS or TNF-α treated MEG-01 cells. Megakaryocytes were stimulated with PBS (baseline), 100 ng/mL LPS or TNF-α for 4 h. Nuclear localization of the NF-κB p65 subunit was monitored by immunostaining. Green: p65 staining; blue: cell nuclei. Scale bar: 20 μm (A). The ratio of the fluorescence intensity of the NF-κB immunostaining in cell nuclei and cytosol was analyzed (B). Mean ± SEM, n = 6–10/group. ** p < 0.01 and *** p < 0.001 based on statistical analyses.

Figure 5

Toll-like receptor (TLR)4 pathway induced changes in mRNA and miRNA expression of MEG-01 cells. Volcano plot depicting the number of differently expressed genes in untreated and lipopolysaccharide (LPS)-treated MEG-01 cells (n = 3/condition) at 4 h, observed by RNA-seq analysis, quantified by the DESeq method and plotted based on Log10 (Fold Change) and –log (p-value). There were 354 significantly upregulated, and 1060 downregulated transcripts in LPS-activated MEG-01 cells compared to control cells at a fold change (FC) of ≥1.5. Dotted lines indicate FC cut-off (A). Clustered heat map of the top 50 differentially expressed genes in LPS-treated MEG-01 cells at 4 h. Genes were displayed as Log (Fold Change) (B). We then validated SELP expression in MEG-01 cells after LPS treatment (n = 6–8/experiment), and TNF-α stimulation was used as a positive control (C). In parallel, miR-26b was significantly attenuated by LPS or TNF-α vs. untreated cells (D). These results were in accordance with the findings in ex vivo septic platelets. The functional relationship between miR-26b and SELP expression was analyzed using specific miRNA mimic transfection in LPS-stimulated MEG-01 cell cultures. The overexpression of miR-26b by mimic (E) resulted in lowered SELP mRNA levels compared to samples with control NEG-01 mimic (F). Mean ± SEM was depicted, * p < 0.05, ** p < 0.01 based on statistical analyses.

Figure 6

Investigation of Dicer1 level in septic platelets and LPS-stimulated MEG-01 cells in regard to altered miRNAs. First, septic platelets (n = 5) within 24 h of the onset of sepsis were studied for the Dicer1 protein level by Western blotting (A). We found a significantly lower expression of Dicer1 in septic platelets. LPS-stimulated MEG-01 cell cultures were intracellularly analyzed for Dicer1 positivity with a fluorescence microscope, showing decreased intensity after LPS treatment (B). Fluorescence intensity of Dicer1 immunostaining in the cytosol was analyzed (n = 15/experiment) (C). Green: Dicer1 staining; blue: cell nuclei. Scale bar: 20 μm. Mean ± SEM was depicted, * p < 0.05, *** p < 0.001 based on statistical analyses.

Figure 7

Investigation of the functional role of Dicer enzyme in attenuated miRNA levels in sepsis. First, to mimic sepsis-induced alteration of the Dicer1 level, DICER1 expression was downregulated via siRNA transfection in MEG-01 cells for the analysis of miR-26b and its target SELP mRNA. Dicer1 specific siRNA was successfully transfected into MK cell cultures based on the highly elevated level of Dicer1 siRNA measured by RT-qPCR (A). This resulted in decreased DICER1 mRNA expression compared to control samples with NEG-01 siRNA (B). In these cells, reduced miR-26b (C) with elevated SELP mRNA levels were quantified (E). In addition, Dicer1-independent miR-451 (D) was analyzed to double-check the specificity of siRNA transfection that showed no change due to manipulation. Second, calpain inhibition was applied to alter miRNA levels in MEG-01 cells among inflammatory conditions in vitro. Calpeptin significantly restored miR-26b (F) levels when was used during LPS treatment, suggesting the role of abnormal Dicer function in miRNA expression of sepsis. Mean ± SEM was depicted (n = 4–6/experiment). * p < 0.05, ** p < 0.01 vs. NEG-01 or untreated samples based on statistical analyses.