Platelet derived exosomes disrupt endothelial cell monolayer integrity and enhance vascular inflammation in dengue patients

Abstract

Background: Thrombocytopenia is the most notable phenomenon in dengue. Activation status of platelets and interaction of platelets with endothelium contribute towards dengue disease pathogenesis. Platelets are the major cell types known to release extracellular vesicles, especially exosomes in circulation. However, the role of platelet derived exosomes (PLT-EXOs) in endothelial dysfunction during dengue infection remains unknown.

Methods: In this study, we recruited 28 healthy subjects and 69 dengue patients categorized as WS- (n=31), WS+ (n=29) and SD (n=9). Platelets were isolated from platelet rich plasma of dengue patients and their activation was assessed by flow cytometry. PLT-EXOs were isolated by ultracentrifugation method. Western blot analyses were performed to characterize the exosomes. Exosome uptake experiment was carried out to see the internalization of exosomes inside endothelial cells (HUVECs). To observe the effect of exosomes on endothelial cells, exosomes were added on HUVECs and expression of adherens and tight junctional proteins were examined by immunofluorescence assay and western blot. Expression levels of vascular injury markers were measured in the culture supernatants of Exosome-HUVEC coculture and sera of dengue patients by MSD-multiplex assay.

Results: As compared to healthy subjects, CD41/CD61 expression was significantly reduced (p<0.0001) and CD62p expression was significantly increased (p<0.0001) on platelets in dengue patients. PLT-EXOs isolated from the dengue patients showed higher expression of CD63 and CD9 proteins than the healthy subjects. With in-vitro immunofluorescence assays, we illustrated the internalization of PLT-EXOs by the HUVECs and observed disruption of endothelial cell monolayer integrity in the presence of PLT-EXOs from WS+ and SD patients. Furthermore, the significant reduction in the expressions of ZO-2, VE-Cadherin and CD31 in endothelial cells following exposure to PLT-EXOs from the dengue patients provide direct evidence of PLT-EXOs mediated vascular permeability. PLT-EXOs stimulated the release of inflammatory markers CRP, SAA, sVCAM-1 and sICAM-1 in the supernatants of HUVEC cells. Importantly, significantly higher levels of CRP, sVCAM-1 and sICAM-1 in the sera of severe than mild dengue patients (p<0.0001) suggest their role in disease severity.

Conclusions: In summary, our data suggest that PLT-EXOs promote vascular leakage via release of proinflammatory mediators and compromise vascular barrier integrity in dengue patients.

Keywords: dengue virus; exosomes; platelet-derived exosomes; platelets; sVCAM-1; vascular inflammation.

Figure 1

Alteration of platelet surface markers in different clinical presentations of dengue patients. (A) Pie chart showing number of dengue patients of different categories as per WHO 2009 classification. (B) Gating strategy used to assess the platelet activation status in dengue patients and healthy subjects. Expression levels of (C) CD41/CD61 and (D) CD62p on platelets of healthy subjects (n=21) and dengue patients (n=65) of different clinical presentations, WS- (n=29), WS+ (n=27) and SD (n=9). HS denotes healthy subjects, DP denotes dengue patients, WS- denotes dengue patients without warning signs, WS+ denotes dengue patients with warning signs, SD represents severe dengue patients and SDG represents severe disease group. Data is represented as mean ± standard error. All the samples were analyzed by non-parametric Mann-Whitney test in which * represents p < 0.05, ** represents p < 0.01, *** represents p < 0.001 and **** represents p < 0.0001.

Figure 2

Platelet count and its correlation with platelet surface markers in dengue patients. (A) Platelet counts in dengue patients with different clinical presentations, WS- (n=29), WS+ (n=27) and SD (n=9). (B) Comparison of levels of CD41/CD61 in healthy subjects (n=21) with dengue patient having platelet count >50,000/μL (n=37) and dengue patient having platelet count <50,000/μL (n=28) respectively. (C) Correlation analyses of thrombocytopenia with CD41/CD61 surface expression on platelets. (D) Comparison of levels of CD62p in healthy subjects (n=21) with dengue patient having platelet count >50,000/μL (n=37) and dengue patient having platelet count <50,000/μL (n=28) respectively. (E) Correlation analyses of thrombocytopenia with CD62p surface expression on platelets. (F) Comparison of platelet counts in dengue patients without (HM-, n=53) and with hemorrhagic manifestations (HM+, n=12). Comparison of levels of (G) CD41/CD61 and (H) CD62p in healthy subjects (n=21) with dengue patients without (HM-, n=53) and with hemorrhagic manifestations (HM+, n=12). HS denotes healthy subjects, WS- denotes dengue patients without warning signs, WS+ denotes dengue patients with warning signs, SD represents severe dengue patients, HM- represents dengue patients without hemorrhagic manifestations and HM+ represents dengue patients with hemorrhagic manifestations. Data is represented as mean ± standard error. Data were analyzed by non-parametric Mann-Whitney test between the two groups. Correlation analyses were performed using two-tailed Spearman’s correlation coefficient test. * represents p < 0.05, ** represents p < 0.01, *** represents p < 0.001 and **** represents p < 0.0001.

Figure 3

Platelet derived exosomes isolated from dengue patients express CD63 and CD9 proteins. (A) Expression of CD41, CD63 and CD9 proteins in pooled fractions (n=5) of platelets-pellet, microvesicles and exosomes isolated from healthy subjects and each category of dengue patients as shown by western blot. (B) Size distribution of platelet specific exosomes though DLS. Fold change expressions of (C) CD63 and (D) CD9 in dengue patients (n=23), WS- (n=10), WS+ (n=10), SD (n=3), SDG (n=13) compared to healthy subjects (n=11). Comparisons of fold change expression of (E) CD63 and (F) CD9 in exosomes derived from platelets of primary (1° INF, n=5) and secondary (2° INF, n=18) dengue patients with healthy subjects (n=11). HS denotes healthy subjects, DP denotes dengue patients, WS- denotes dengue patients without warning signs, WS+ denotes dengue patients with warning signs, SD represents severe dengue and SDG represents severe disease group. EXO:HS denotes platelet derived exosomes from healthy subjects, EXO:DP denotes platelet derived exosomes from dengue patients, EXO:WS- denotes platelet derived exosomes from dengue patients without warning signs, EXO:WS+ denotes platelet derived exosomes from dengue patients with warning signs, EXO:SD denotes platelet derived exosomes from severe dengue and EXO:SDG denotes platelet derived exosomes from severe disease group. All the samples were analyzed by parametric unpaired t test in which * represents p < 0.05 and ** represents p < 0.01.

Figure 4

Effect of platelet derived exosomes on endothelial cells. (A) Internalization of exosomes isolated from platelets of healthy subjects and dengue patients of each categories (WS-, WS+ and SD). (B) Immunofluorescence images of HUVECs showing expression of cell adhesion molecules Claudin-1 (green) and VE-Cadherin (Red) with nucleus stained with DAPI (blue) on exposure of platelet derived exosomes isolated from dengue patients for 48 hr onto HUVEC cells. Only HUVEC denotes untreated HUVEC cells, EXO:HS denotes platelet derived exosomes of healthy subjects, EXO:WS- denotes platelet derived exosomes from dengue patients without warning signs, EXO:WS+ denotes platelet derived exosomes from dengue patients with warning signs and EXO:SD represents platelet derived exosomes from severe dengue patients. n denotes the number of patients from which platelet derived exosomes of different categories were tested on HUVEC cells. Representative images were shown here.

Figure 5

Vascular dysfunction associated with platelet derived exosomes in dengue patients. (A) Transwell assay of endothelial permeability using BSA-Evans blue dye. Graphical representation of absorbance readings at 620 nm by plotting optical density (OD) values of microtiter plate of transwell treated with PLT-EXOs isolated from healthy subjects and dengue patients (n=4 from each category). (B) Expression of ZO-2, VE-cadherin, CD31 on exposure of platelet derived exosomes from healthy subjects and dengue patients after 48 hr in HUVEC cells as shown by Immunoblotting. (C) Fold change expressions of ZO-2 in HUVEC cells after treatment of exosomes isolated from different clinical presentations (n=2 from each category). Fold change expressions of (D) VE-Cadherin and (E) CD31 in HUVEC cells after treatment of exosomes isolated from HS (n=9), DP (n=24), WS- (n=10), WS+ (n=8), and SD (n=6). Comparisons of fold change expression of (F) VE-Cadherin and (G) CD31 in HUVEC cells after addition of exosomes derived from primary (1° INF, n=7) and secondary (2° INF, n=17) dengue patients with healthy subjects (n=11). CC denotes untreated HUVEC cells. EXO:HS denotes platelet derived exosomes from healthy subjects, EXO:DP denotes platelet derived exosomes from dengue patients, EXO:WS- denotes platelet derived exosomes from dengue patients without warning signs, EXO:WS+ denotes platelet derived exosomes from dengue patients with warning signs, EXO:SD denotes platelet derived exosomes from severe dengue and EXO:SDG denotes platelet derived exosomes from severe disease group. Data is represented as mean ± standard error. Fold change expression analyses were carried out using parametric unpaired t test in which * denotes p < 0.05, **denotes p < 0.01 and *** denotes p < 0.001.

Figure 6

Enhancement of vascular injury markers in dengue patients through platelet derived exosomes. (A-D) Comparisons of CRP, SAA, sVCAM-1, sICAM-1 levels in supernatants of exosome-HUVEC coculture of EXO:HS (n=8), EXO:DP (n=24), EXO:WS- (n=11), EXO:WS+ (n=8), EXO:SD (n=5) with untreated HUVEC cells (n=5). (E-H) Comparisons of CRP, SAA, sVCAM-1, sICAM-1 levels in serum samples of healthy subjects (n=24) and dengue patients (n=54) with different clinical presentations, WS- (n=22), WS+ (n=23), SD (n=9). EXO:HS denotes platelet derived exosomes from healthy subjects, EXO:DP denotes platelet derived exosomes from dengue patients, EXO:WS- denotes platelet derived exosomes from dengue patients without warning signs, EXO:WS+ denotes platelet derived exosomes from dengue patients with warning signs, EXO:SD+ denotes platelet derived exosomes from severe dengue and EXO:SDG denotes platelet derived exosomes from severe disease group. Only HUVEC denotes untreated HUVEC cells. HS denotes healthy subjects, DP denotes dengue patients, WS- denotes dengue patients without warning signs, WS+ denotes dengue patients with warning signs, SD represents severe dengue patients and SDG represents patients in severe disease group. Data is represented as mean ± standard error. Mean concentrations were analyzed by non-parametric Mann-Whitney test in which * represents p < 0.05, ** represents p < 0.01, *** represents p < 0.001 and **** represents p < 0.0001.

Figure 7

Association of sVCAM-1 levels with thrombocytopenia and disease severity. (A-D) Comparisons of levels of CRP, SAA, sVCAM-1 and sICAM-1 in dengue patients having platelet count >50,000/µL (n=29) with dengue patients having platelet count <50,000/µL (n=25). (E) Correlation analysis of sVCAM-1 levels in sera of dengue patients with platelet counts. ROC curve presentations differentiating sVCAM-1 levels of (F) healthy subjects with dengue patients and (G) WS- with SD patients. (H) sVCAM-1 levels in sera of WS- and severe disease group of WS+ and SD patients at different days post onset of illness. At POD 1-5, n= 6 for WS- patients and n=13 for SDG respectively. At POD 6-10, n=14 for WS- patients, and n=12 for SDG respectively. HS denotes healthy subjects, DP denotes dengue patients, WS- denotes dengue patients without warning signs, WS+ denotes dengue patients with warning signs and SD represents severe dengue patients. Data is represented as mean ± standard error. Mean concentrations were analyzed by non-parametric Mann-Whitney test in which ** represents p < 0.01. Correlation analysis was performed using two-tailed Spearman’s correlation coefficient test.

Figure 8

Schematic representation showing the effect of platelet derived exosomes on endothelial cells in dengue patients. Dengue virus activates platelets which in turn release exosomes on to the bloodstream. Endothelial cells internalize exosomes which carry CD63 and CD9 proteins. After internalization, platelet secreted exosomes damage endothelial cells integrity and induce inflammation. Inflamed endothelial cells subsequently increases the disease severity in dengue patients.