Microvesicles from the plasma of elderly subjects and from senescent endothelial cells promote vascular calcification

Quantification and characterization of plasma-derived microvesicles

Flow cytometric analysis of the plasma-derived MVs from both the young and elderly subjects localized the MVs population around the forward scatter signal corresponding to particles with a diameter of around 1 μm (Fig. 1). Figure 1B shows that the plasma of the elderly subjects contained a greater number of total MVs.

As showed in Table 1, elderly subjects' MVs contained an increase total MVs (annexin A5+) compared with young subjects (6959±1187/µL vs 4868±547/µL; p<0.001). Additionally, significantly more of the elderly subjects' MVs carried annexin A5+/CD31+/CD42+ platelet-associated antigens (5254±1051/µL compared with 4176±661/µL of the young subjects' MVs; p=0.004). In addition, more carried endothelial-associated annexin A5+/CD31+/CD42- antigens (1850±338/µL compared with 692±260/µL; p<0.001).

Quantification and characterization of HUVEC-derived microvesicles

Cytometric characterization revealed most of these MVs showing a similar distribution from both young and senescent endothelial cells (representing a diameter of 1 µm; Fig. 2A1 and 2B1). The number of total MVs in the culture medium of the senescent human umbilical vein endothelial cells (HUVEC) was greater than in that of the young cells (Fig. 2A2, 2B2 and 2C). These senescent HUVEC-derived MVs expressed more annexin A5+/CD31+ than those derived from young HUVEC (Fig. 2D). Based on the absence of fluorescent nuclear staining by acridine orange, only a small percentage of MVs (<10%) showed characteristics of apoptotic bodies.

Microvesicles from senescent HUVEC and elderly subjects' plasma induce calcification in HASMC

To determine the ability of MVs to promote calcification, human aortic smooth muscle cells (HASMC) were cultured in the presence of MVs (50,000 MVs/mL) from senescent or young endothelial HUVEC, (Fig. 3A). Only the MVs from the senescent HUVEC induced calcification in the HASMC. In fact, alizarin red deposits were observed after 6 days of treatment. These deposits were not observed in HASMC grown with MVs from the young HUVEC. HASMC were incubated with inorganic P (Pi), a pro-calcific condition, as positive control; in the negative control HASMC were grown without MVs and Pi. This last control did not show red alizarin stain. A similar effect was observed using MVs (100,000 MVs/mL) from elderly and from young subjects.

In order to quantify the process of calcification, calcium content was analysed using a colorimetric method containing phenolsulphonephthalein dye (Figure 3B). The calcium content was significantly increased in HASMC after 6 and 9 days of treatment with senescent HUVEC MVs. A similar effect was observed in pro-calcific condition.

Capture of HUVEC-derived microvesicles by HASMC

HUVEC-derived MVs were labelled with CellTracker CM-Dil PI (red) and cultured for 48 h with HASMC to determinate whether they promote calcification by adherence to the cell surface and/or by being endocytosed by the latter cells. After that HASMC were stained with PKH67 phalloidin (green). Confocal microscopy showed the majority of the MVs to locate inside the cytoplasmic compartment of the HASMC (Fig. 4). This was true for the MVs from both the young and senescent HUVEC.

Senescent HUVEC-derived microvesicles carry Ca

SEM analysis showed the HUVEC-derived MVs to be a heterogeneous population of spherical and elongated structures ranging from 0.2 to 1.8 μm along their longest axis (Fig. 5A). Large MVs were more commonly produced by senescent HUVEC than young HUVEC. Microanalysis revealed the MVs from all sources to have a heterogeneous atomic composition (as demonstrated by the heterogeneous brightness of the MVs; in EDS analysis, elements with a higher atomic number appear brighter because they emit more backscatter signal) (Fig. 5B and C).

To determine the Ca content of the MVs, 60 young and senescent HUVEC-derived MVs were SEM-EDS-analyzed in spot mode. This involves focusing a narrow electron beam at a specific point (spot) on the MV (Fig. 6A). Some 67% of the senescent HUVEC-derived MVs had a high Ca content (i.e., >20% relative content). MVs with low Ca contents (3-10%) represented some 25% of the total number; only 8% had no Ca. In contrast, 84% of the young HUVEC-derived MVs had no Ca; the remaining 16% contained a little Ca. These results agree with the Ca mapping results for both these types of MV (Fig. 6B1 and 2).

Interestingly, TEM analysis revealed morphological differences between the MVs from the young and senescent HUVEC (Fig. 7). MVs from the former (Fig. 7A) were generally spherical and showed a very simple internal structure. In contrast, few of the MVs from senescent HUVEC were spherical; indeed, a large number of irregular MVs were seen to contain even smaller vesicles of differing electron density (Fig. 7B).

Senescent endothelial microvesicles contain pro-calcification proteins

Western blot assays showed the senescent HUVEC-derived MVs to have more annexin A2 and annexin A6 than those from the young cells. Similar results were seen for BMP2 (Fig. 8A). However, there was no change in protein expression of these proteins between young and senescent HUVEC (Fig. 8B). These findings suggest that MVs from senescent endothelial are involved in the appearance of calcification.

Study population

The study subjects were 26 non-smoking volunteers with normal body mass indices. Fifteen were aged 20–28 years, and 11 were aged 75–83 years (Table 1). All were healthy according to analytical criteria and their responses to a medical questionnaire; certainly none had any cardiovascular, autoimmune, metabolic, inflammatory or kidney disease. None took any medication (not even the elderly subjects). The local Scientific Ethics Committee approved the present research protocol. All subjects provided written, informed consent to be included.

Blood samples

Venous blood samples were drawn from both elderly and younger subjects into EDTA-coated tubes. Platelet-free plasma (PFP) was prepared at room temperature by serial centrifugation (15 min at 1500 x g, followed by 2 min at 13,000 x g). The supernatant was stored in 1.5 mL tubes at −80°C until use.

Cell cultures

Isolation, quantification and characterization of microvesicles

Plasma-derived and HUVEC-derived (isolated from the culture medium) MVs were isolated and counted using the quantification method based in the previously published for of the International Society on Thrombosis and Haemostasis [28], with minor modifications. Platelet-free plasma (PFP) was obtained as previous described ("Blood samples" in Methods section). Next, MVs from PFP was isolated by centrifugation (13,000 g for 30 min). Furthermore, MVs from HUVEC were obtained from supernatants were centrifuged at 13,000 g for 30 min as described previously.

Pooled MVs (i.e., from elderly subjects' plasma, from young subjects' plasma, from the medium of young HUVEC cultures, and from that of senescent HUVEC cultures) were characterized in terms of size using a Beckman Coulter Cytomic FC 500 flow cytometer running CXP software. MVs were understood to be those events gated with a size between 0.5-1.5 µm; this gate was established from the side scatter (SSC) vs. forward scatter (FSC) dot-plot produced in a standardization experiment using the SPHERO™ Flow Cytometry Nano Fluorescent Size Standard Kit (Spherotech). The latter has size-calibrated fluorescent beads ranging from 0.1-1.9 µm in diameter. Events below 0.2 µm were excluded in order to adequately distinguish true events from the background; events >1.9 µm were excluded to prevent possible confusions with apoptotic bodies. The absolute number of MVs (events) per microliter was determined using Flow Count Calibrator beads (10 µm; Beckman Coulter Inc.) according to the manufacturer's recommendations using CXP software: (MVs counted x standard beads/L)/(standard beads counted). Data were recorded as the mean of three independent measurements of the same sample.

Triple-fluorescent labelling was performed to characterise the protein profile of the plasma MVs and thus determine their cellular origin. This was done by incubating platelet-free plasma (100 µL) with fluorescein isothiocyanate-conjugated (FITC) labelled monoclonal anti-CD31 (BD Pharmingen), peridinin chlorophyll protein complex (PerCP) monoclonal anti-CD42b (Abcam), and phycoerythrin-annexin A5 (BD Pharmingen), together in annexin A5-binding buffer (10 mM HEPES, 7.4 pH, 140 mM NaCl, 2.5 mM CaCl₂). Isotype negative controls were also prepared. Since CD31 is expressed by both platelets and ECs, but CD42 occurs only in platelets, platelet MVs were defined by events CD31+/CD42+ and endothelial MVs by events CD31+/CD42-. As a control for the annexin A5 labeling, a sample with fluorescein-conjugated annexin A5 using a CaCl2-free solution was established.

MVs from HUVEC culture medium were isolated and characterized with respect to size and in terms of their protein profile using FITC-anti-CD31 and phycoerythrin-annexin A5, as above.

To confirm the exclusion of apoptotic bodies in MVs isolations, the DNA content of the HUVEC-derived MVs was determined by incubating them with acridine orange (Invitrogen) to a final concentration of 20 mM.

Western blot analyses

Extracts from young and from senescent endothelial cells, and MVs from young and from senescent HUVEC were lysed in CytoBuster Protein Extraction Reagent lysis buffer (Millipore) containing protease and phosphatase inhibitor cocktail (Roche). The total protein content of lysates was quantified using a BCA Protein Assay Kit (Pierce), with BSA as the standard. Briefly, equal amounts of protein (10-50 μg protein/lane) were diluted with reducing sample buffer and separated by SDS/PAGE (10% gel) under reducing conditions. Samples were then transferred onto nitrocellulose membranes (BioRad), blocked in TBS containing 0.1% Tween 20 and 5% dry non-fat milk for 1 h at room temperature, and incubated in the same buffer with different primary antibodies (BMP2, Abcam ab82511, dilution 1:1000, 46 kDa; annexin A6, Abcam ab31026, dilution 1:1,000, 76kDa; annexin A2, Abcam ab41803, dilution 1:1000, 38 kDa). Anti-β-actin (Santa Cruz) (sc-47778, dilution 1:2000, 43 kDa) or GAPDH (Millipore) MAB374, dilution 1:2000, 38 kDa) was used as a loading control for endothelial cells, and anti-HSP90 (Cell Signaling) (CST#4874, dilution 1:1000, 90 kDa) was used as a loading control for MVs. After washing, the membranes were incubated with Novex horseradish peroxidase-conjugated secondary antibodies (1:5000). Bands were visualized with Luminata Crescendo Western HRP substrate (Millipore). The quality of proteins and efficacy of protein transfer was evaluated by Red Ponceau staining. Bands were quantified using Image J software (NIH) and normalized to anti-HSP90 in MVs and anti-GAPDH or anti-β-Actin in HUVEC.

Calcification Determination

The qualitative staining of HASMC calcification was identified by alizarin red. HASMC (10000 cells/well, 24-well plates) were cultured with 100,000 MVs/mL from the plasma of young or elderly subjects, or with 50,000 MVs/mL from young or senescent HUVEC for 1, 3, 6 and 9 day in 2% FBS-supplemented DMEM. As a positive control, cells were exposure with 2% FBS-supplemented DMEM containing 2.0mM Pi (pro-calcific condition). Culture medium was renewed every 2 days. Detection and quantification of mineral deposition were performed using alizarin red staining. Briefly, samples were fixed (50% ethanol 5 min, 95% ethanol 5 min), stained for 5 min with alizarin red S (40 mM, pH 4.2), rinsed with 50% ethanol, and subsequently scanned.

Assessment of calcium deposition

After 6 and 9 days of incubation, calcification was quantified. Cells were decalcified with 0.6 N HCl overnight at 4°C. The Ca content of the supernatants was determined by spectrophotometer by a kit containing phenolsulphonephthalein dye (no. DICA-500, QuantiChrom calcium assay kit; BioAssay Systems). Then, the cells were washed three times with PBS (Sigma) and solubilized in 0.1 M NaOH/1% SDS. The protein content was measured using the BCA protein assay kit (Pierce), and the Ca content was normalized for total protein. Ca content of the cells was expressed as µg/mg protein.

Confocal microscopy

MVs from young and from senescent HUVEC were stained with CellTracker CM-Dil (LifeTechnologies) to visualize interactions between the MVs and HASMC. The latter cells were cultured with the stained MVs for 48 h, washed with PBS, and then stained with phalloidin-FITC (PKH67). They were then fixed with 4% paraformaldehyde solution in PBS, and mounted in gold antifade reagent plus DAPI. Fluorescence was detected using a confocal microscope (Leica Microsystems GmbH).

Electron microscopy

Clean, dry coverslips (12 mm diameter) were coated with poly-L-lysine (70,000-150,000 Da) using a 0.1% solution of the same (Sigma). A water-repellent circle of approximately 6 mm diameter was then drawn on the coated coverslip using a pap pen. One drop of suspension of MVs from young or senescent HUVEC (6000 MVs/mL) was then placed within this circle and incubated for 1 h in a humidified chamber at room temperature. After a brief wash in PBS, the MVs were fixed in 3% glutaraldehyde for 10 min, rinsed in buffer, dehydrated in an ethanol series and dried following the routine critical point drying procedure for scanning electron microscopy (SEM). For morphological analysis, the prepared MVs were gold-palladium coated (theoretical thickness 5 nm) in a Polaron E5400 metal evaporator (BioRad), and examined using a Zeiss DSM 950 SEM.

For Ca microanalysis, non-metal-coated MVs were prepared using a Hitachi TM-1000 SEM with an EDS system. For Ca mapping, MVs samples were fixed in 70% ethanol and deposited on a formvar-coated copper grid and analysed using a Zeiss LEO 906E transmission electron microscope (TEM) equipped with an elemental analysis environmental module.

For ultrastructual analysis, samples were fixed in 2.5% (v/v) glutaraldehyde and 0.05% (w/v) ruthenium red in 0.05 M sodium cacodylate buffer for 4 h at 4°C. They were then post-fixed with 1% (w/v) OsO₄ in 0.1 M sodium cacodylate buffer for 1 h at room temperature, dehydrated in ethanol, and infiltrated with resin. Ultrathin sections were cut using a Reichert Jung E ultramicrotome. Finally, the sections were stained for 10 min with 1% (w/v) aqueous uranyl acetate solution and lead citrate, and viewed using the same TEM as above.

Statistical analysis

Data are expressed as means±SEM and analysed by ANOVA followed by the Duncan test. Comparisons between pairs of means were performed using the Student t test or Mann–Whitney U test as required. Significance was set at p<0.05.