Circulating cardiomyocyte-derived extracellular vesicles reflect cardiac injury during systemic inflammatory response syndrome in mice

Abstract

The release of extracellular vesicles (EVs) is increased under cellular stress and cardiomyocyte damaging conditions. However, whether the cardiomyocyte-derived EVs eventually reach the systemic circulation and whether their number in the bloodstream reflects cardiac injury, remains unknown. Wild type C57B/6 and conditional transgenic mice expressing green fluorescent protein (GFP) by cardiomyocytes were studied in lipopolysaccharide (LPS)-induced systemic inflammatory response syndrome (SIRS). EVs were separated both from platelet-free plasma and from the conditioned medium of isolated cardiomyocytes of the left ventricular wall. Size distribution and concentration of the released particles were determined by Nanoparticle Tracking Analysis. The presence of GFP + cardiomyocyte-derived circulating EVs was monitored by flow cytometry and cardiac function was assessed by echocardiography. In LPS-treated mice, systemic inflammation and the consequent cardiomyopathy were verified by elevated plasma levels of TNFα, GDF-15, and cardiac troponin I, and by a decrease in the ejection fraction. Furthermore, we demonstrated elevated levels of circulating small- and medium-sized EVs in the LPS-injected mice. Importantly, we detected GFP⁺ cardiomyocyte-derived EVs in the circulation of control mice, and the number of these circulating GFP⁺ vesicles increased significantly upon intraperitoneal LPS administration (P = 0.029). The cardiomyocyte-derived GFP⁺ EVs were also positive for intravesicular troponin I (cTnI) and muscle-associated glycogen phosphorylase (PYGM). This is the first direct demonstration that cardiomyocyte-derived EVs are present in the circulation and that the increased number of cardiac-derived EVs in the blood reflects cardiac injury in LPS-induced systemic inflammation (SIRS).

Keywords: Cardiomyocyte; Cardiomyopathy; Extracellular vesicles; Inducible transgenic mice; SIRS.

Fig. 1

Tissue-specific and inducible Cre-mediated expression of GFP in cardiomyocytes. a Generation of tamoxifen-inducible double transgenic animals (MerCreMer/mTmG). Schematic illustration of MerCreMer and mTmG construct to generate bi-transgenic animals. MerCreMer mice are crossed with mTmG animals to trace the lineage of GFP-positive cells in the cardiomyocyte. b Ubiquitous expression of tdTomato and cardiomyocyte-specific expression of GFP demonstrated in mouse tissue from a MerCreMer x mTmG cross at necropsy. Fluorescent, F1 generation littermates were euthanized and imaged at 12 weeks of age. All sections were recorded during a single imaging session using the same exposure times. GFP expression is limited to the cardiomyocyte whereas tdTomato is brightly expressed in all the other cells. The images of different muscle tissues were adjusted equally for brightness and contrast. All other images are shown without post-acquisition processing. Scale bars, 100 μm. c Cardiomyocytes were isolated from adult transgenic mice. The left panel shows the isolated GFP expressing cardiomyocytes under the 20X objective, the right panel shows the bright field image of the isolated single cells. Scale bars: 30 µm

Fig. 2

Effect of an LPS injection on inflammation, EV release, and cardiac function. a C57/Bl6 mice were injected intraperitoneally with 6 mg/kg of LPS/mouse. b TNF-α in platelet-free plasma (PFP) (n = 6–8). c GDF-15 in PFP (n = 6–8). ELISA results after 4 h/6 h and 24 h after LPS injection. d Cardiac Troponin I was measured by ELISA 6 h and 24 h after LPS injection. Data points represent values from individual mice (means ± SD are indicated). *P < 0.05; **P < 0.01; ***P < 0.005; ****P < 0.001). e, f mEVs were isolated from 200 µl of mouse PFP using differential centrifugation e) followed by ultracentrifugation to separate sEVs f n = 6. Particle concentration was measured by NTA 6 h and 24 h after the treatment. g The graph shows the number of infiltrating cells/fields for each condition. Quantitative results indicate the average values ± SD of n = 3 animals in each group. The results were analyzed by Student’s T-test, no significant difference was found. h Representative images of the echocardiography analysis of the mouse hearts 6 h after LPS injection. i, j LPS led to a significant reduction of EF (ejection fraction) and prolongation of IVRT (isovolumic relaxation time). Data were analyzed using two-way ANOVA followed by Tukey's multiple comparisons test. Data are presented as mean ± SD (*P < 0.05 vs vehicle 6 h, ^#P < 0.05 vs LPS baseline)

Fig. 3

Effect of LPS administration on GFP + EV release to the circulation. a Schematic protocol of Tamoxifen induction and LPS injection of mice. b, c Flow cytometry analysis. Calibration beads were used (0.2–1 µm). d The gating strategy for mEVs and the 1 μm counting beads. e As a negative control, PFP-derived mEVs were analyzed from C57Bl6 mice. f Cardiomyocyte-derived EVs as GFP⁺ events. mEVs secreted by all other cells were Tomato⁺. g A 95 ± 1% reduction of the event number after exposure to 0.1% Triton. h, i Events within the mEV gate were then analyzed, and GFP + events with/without LPS injection are shown on (h). i demonstrates Tomato⁺ events with/without LPS injection. The figure shows the results of three independent experiments (n = 9). The figure represents data from individual mEV samples (mice) as well the means ± SD (*P < 0.05, T-test). j Fold increase upon LPS treatment, n = 6 ns = nonsignificant (T-test). k The pie diagram represents the percentages of green fluorescent mEVs in control and LPS treated animals (n = 9 in each group)

Fig. 4

Ex vivo release of mEVs by primary adult murine cardiomyocytes upon in vivo administration of LPS. a Representative results of size distribution analysis (NTA) of mEVs from isolated cardiomyocytes from LPS-injected and control mice (24 h conditioned medium). b mEVs release was significantly elevated after 24 h. Data are obtained from three independent experiments (mean ± SD values are indicated in the figure) (*P < 0.05; **P < 0.01; n = 6, ANOVA). c, d Transmission electron microscopy of isolated mEVs showing morphological heterogeneity of secreted mEVs and mitochondria (the black arrows point to the mitochondria; scale bar = 500 nm). e Annexin V positive events were detected by flow cytometry and the mEV concentrations were determined using counting beads. Data are mean ± SD, n = 5, *P < 0.05 vs. control (T-test). f shows the flow cytometry (FCM) results of Annexin V staining before and after Triton treatment (light green and dark green, respectively); unstained mEVs are in grey. g CD63 positive events were detected by flow cytometry and were normalized to EV concentrations using counting beads. Data are mean ± SD, n = 3, *P < 0.05 vs. control (T-test). h FCM results for CD63 before and after Triton treatment (light and darker blue, respectively); grey is for the isotype control. i, j The mEV fraction was isolated from the conditioned medium of cardiomyocytes from in vivo LPS-induced mice and was stained with MitoTracker Red. The representative dot plot shows the presence of mitochondria in the control i and LPS-injected group j within the mEV gate. k The representative histograms of MitoTracker expression in gated mEV subsets isolated from the control or LPS group are shown green and blue, respectively; unstained mEVs are in grey. l Statistical analysis of MitoTracker expression in CMEVs (n = 6). Comparisons were performed using a T-test, *P < 0.05

Fig. 5

Expression of clusterin and PYGM in CMEVs. a Schematic illustration of the experiment. b, c Clusterin and PYGM levels were measured by flow cytometry. Data are mean ± SD, n = 5, **P < 0.01; ***P < 0.001 vs. control. d, e CMEVs were exposed either to trypsin or Triton X-100 and detected by flow cytometry. Data are mean ± SD, n = 6

Fig. 6

Expression of GFP and cTnI or PYGM on the CMEVs in plasma. a Schematic illustration of the experimental setup. b, c Flow cytometry and ELISA of extracellular vesicles isolated by differential centrifugation from PFP samples. cTnI and PYGM positive events were measured in the GFP gate by flow cytometry. EV concentrations were normalized to counting beads. d ELISA shows the results of 24 h LPS injection. Data are mean ± SD, n = 3, *P < 0.05 vs. control with Student T-test. e Representative plot showing the SEC elution profile according to CD63 + events/mL as an mEV markers of each SEC fraction. Fractions with the highest CD63 (4–6) were pooled together, to obtain the SEC purified mEV preparations for further investigations. f, g Flow cytometry and ELISA of extracellular vesicles isolated by SEC from PFP samples. cTnI positive events were measured in the GFP gate by flow cytometry. EV concentrations were normalized to counting beads. g ELISA shows the results of 24 h LPS injection. Data are mean ± SD, n = 4, ***P < 0.001 vs. control with Student T-test