Platelets and platelet-like particles mediate intercellular RNA transfer

Abstract

The role of platelets in hemostasis and thrombosis is clearly established; however, the mechanisms by which platelets mediate inflammatory and immune pathways are less well understood. Platelets interact and modulate the function of blood and vascular cells by releasing bioactive molecules. Although the platelet is anucleate, it contains transcripts that may mirror disease. Platelet mRNA is only associated with low-level protein translation; however, platelets have a unique membrane structure allowing for the passage of small molecules, leading to the possibility that its cytoplasmic RNA may be passed to nucleated cells. To examine this question, platelet-like particles with labeled RNA were cocultured with vascular cells. Coculture of platelet-like particles with activated THP-1, monocytic, and endothelial cells led to visual and functional RNA transfer. Posttransfer microarray gene expression analysis of THP-1 cells showed an increase in HBG1/HBG2 and HBA1/HBA2 expression that was directly related to the transfer. Infusion of wild-type platelets into a TLR2-deficient mouse model established in vivo confirmation of select platelet RNA transfer to leukocytes. By specifically transferring green fluorescent protein, we also observed external RNA was functional in the recipient cells. The observation that platelets possess the capacity to transfer cytosolic RNA suggests a new function for platelets in the regulation of vascular homeostasis.

Figures

Figure 1

PLPs coincubation with HUVECs and THP-1 cells. (A) THP-1 cells, treated with 1 μg/mL Pam3CSK4, were coincubated with PLPs for 24 hours and stained with CD41a-FITC, CD42b-FITC (open curve; solid line) for flow cytometry analysis. Isotype IgG (gray filled curve), control (no PLPs; open curve, dotted line). (B) The same experiment was carried out staining THP-1 cells for CD11b-APC, CD41a-FITC, and CD42b-FITC. Samples were then mounted on coverslips for confocal microscopy (PLP). IgG control (isotype) and THP-1 alone (control). Scale bar denotes 10 μm; 100× objective lens. (C) Confocal microscopy also was performed on PKH67-labeled PLPs (green fluorescence) cocultured for 24 hours with THP-1 cells to demonstrate PLP internalization. Blue nuclear staining was performed with 4,6-diamidino-2-phenylindole (DAPI). Scale bar denotes 10 μm; 60× objective. (D) The experiment was performed as described in panel A using CD11b-FITC. Untreated cells (control), activated cells (PAM), PLP (THP-1 cells cocultured with PLPs). (E) HUVECs were treated for 10 minutes with 0.5 U/mL thrombin and then coincubated with PKH67-labeled PLPs for 30 minutes (gray filled curve) and analyzed by flow cytometry. Control (open curve, dotted line), untreated HUVEC + PLPs (open curve, solid line). (F) Confocal microscopy of the experiment described in panel E with additional DAPI staining. HUVECs incubated with PKH67-labeled PLPs (ii-iv) and unlabeled PLPs (i). Scale bar denotes 20 μm; 40× objective.

Figure 2

RNA transfer from MEG-01 to HUVECs and THP-1 cells. (A) Flow cytometry analysis of 1 μg/mL PAM-treated THP-1 cells cultured in presence of BrUTP-labeled PLPs for 6, 12, and 24 hours. No fusion was observed at 1- and 3-hour time points (*P < .05 compared with PAM). (B) The same experiment was analyzed by confocal microscopy. THP-1 cells demonstrated BrUTP labeling after 24-hour incubation with RNA-labeled PLPs (iii-vi; iv and vi are in bright field; 100× objective). THP-1 control cells stained with IgG (i) and BrdU-FITC (ii). Scale bar denotes 10 μm; 100× objective. (C) HUVECs, treated with 0.5 U/mL thrombin, were coincubated for 1 hour with BrUTP-labeled PLPs and analyzed by flow cytometry (open curve, solid line). Isotype IgG (gray filled curve), control (no PLPs; open curve, dotted line). (D) PAM (1 μg/mL)–treated THP-1 cells cocultured for 24 hours with GFP-PLPs treated (RNase) or not with RNase (PLP). THP-1 + untreated-PLPs were washed, after 24 hours, to eliminate residual PLPs and cultured for additional 24 hours (No PLP 48 h; *P < .05 vs untreated PLPs). (E) HUVECs showed GFP fluorescence, by flow cytometry, after 1-hour coincubation with PLPs containing GFP (open curve, dotted line), cells were then washed and allowed to grow for 24 hours (open curve, solid line). Control (gray filled curve).

Figure 3

THP-1 HBG1/HBG2 expression after PLP RNase treatment. Cells cocultured for 24 hours with RNase-treated PLPs expressed less HBG1/HBG2 (RNaseA/T1 and RNase ONE) compared with the untreated PLP coculture (PLP). HBG1/HBG2, normalized against GAPDH, was absent in the control, THP-1 cells treated with 1 μg/mL PAM and 100 ng/mL TPO.

Figure 4

Quantitative RT-PCR of TLR2 in transfused mice.TLR2 fold changes (expressed as 2−ΔΔCt) in platelets and PBMCs. All samples were normalized against GAPDH. In each case we used 3 mice per group; the transfer of TLR2 mRNA to PBMC was considered significant (*P = .05).

Figure 5

SOD2 activity and intracellular ROS in THP-1 cells. (A) SOD2 activity assessed in THP-1 control cells, 1 μg/mL PAM-treated and in presence of PLPs (PLP). (B) Cells coincubated with (PLP-T) or without PLPs (PAM-T) were treated with 5μM 5-(and-6)-carboxy-2′,7′-difluorodihydrofluorescein diacetate for 30 minutes and then evaluated by flow cytometry versus controls with (PLP) and without PLPs (*P < .05 compared with PLP and **P < .001 compared with NT [NT, no treatment]).

Figure 6

PLPs interaction with HUVECs and THP-1 cells.