1
The interactions of HFpEF patients and control subjects extracellular
vesicles with endothelial cells
Coagulation assays in HUVECs
Yvette van Steen Student number: 10813233
Bachelorproject Biomedical Sciences – University of Amsterdam UMC Utrecht – experimental cardiology Supervision Associate professor Hester den Ruijter PhD candidate Robin Hartman PhD candidate Gideon Valstar
Abstract
Heart failure (HF) is the leading cause of hospitalizations in elderly and the prevalence has been growing with the increased wealth and accompanying age in populations of western countries. HF can be distinguished in a syndrome with reduced or preserved ejection fraction. Heart failure with preserved ejection fraction (HFpEF) is more often seen in women and HFpEF can develop after a systemic microvascular endothelial activation induced by co-morbidities. Extracelluclar vesicles (EV) are thought to be involved in several biological processes important in HFpEF development like coagulation, inflammation and endothelial dysfunction. The (sex-specific) effect of HFpEF patient (n=6) and healthy control (n=6) extracellular vesicles, on human umbilical vein endothelial cells (HUVECs) was examined by usage of a tissue factor-dependent thrombus formation assay. Vesicles were isolated (with the aim to obtain exosomes) from blood plasma by ultra-centrifugation and were exposed to HUVECs during a 5 hour stimulation. Compared to non-stimulation, no effects of vesicle exposure were observed. However, EVs seemed to inhibit the coagulatory effect of positive control tumor necrosis factor alpha (TNF-α), especially after addition of HFpEF patient vesicles to TNF-α. A few (contradicting) sex-differences in TF concentration in unstimulated HUVECS were seen in the TF factor assay, but these findings were not verified during qPCR analysis. Further research is needed to confirm these results and usage of more precise exosome isolation methods and more robust TF assays is recommended.
2
INTRODUCTION
Heart Failure in Cardiovascular Diseases
Despite advances in treatment, cardiovascular diseases (CVD) remain the leading cause of death in Europe and in many countries CVD still causes more than twice as many deaths as cancer (Nichols et al., 2014). Cardiovascular risk factors including diabetes mellitus, hypertension and smoking play a central role in secondary disease prevention, but even treatment of these risk factors cannot completely reduce the risk of CVD development. Although much research has been performed on CVD, a lot is still unknown about this complex range of diseases. In CVD, heart failure (HF) is the leading cause of hospitalization among elderly and the prevalence has been growing with the increased wealth and accompanying age in populations of western countries (Andersen & Borlaug, 2014). Projections of HF in the United States show that the prevalence of HF will increase 46% from 2012 to 2030. Hence, total medical costs of CVD are projected to triple in this period (Heidenreich et al., 2013). Approximately half of the patients with HF are women and the life time risk of developing the HF syndrome is higher for women than for men (Go et al., 2014).
HFpEF and HFrEF
Heart failure can be distinguished in phenotypes with reduced ejection fraction (HFrEF) or preserved ejection fraction (HFpEF). In a recent systematic review, Dutch researchers found a median prevalence estimate of 3.3% HFrEF and 4.9% HFpEF in people above 60. Men were more diagnosed with HFrEF and women more often had HFpEF (Van Riet et al., 2016). Prevalence of HFrEF has been decreasing in the last decades, probably because the major risk factors for HFrEF (smoking, exposure to second-hand smoke and the incidence of ischemic heart disease) are on decline (Paulus & Tschöpe, 2014). In contrast, the prevalence of HFpEF is suggested to grow approximately 1% per year relative to HFrEF partly due increasing comorbidities and risk-factors, and more awareness of this phenotype among physicians (Owan et al., 2006).
HF with preserved ejection fraction (EF) is associated with diastolic left ventricular dysfunction that involves reduced left ventricular relaxation and increased left ventricular stiffness, with an ejection fraction that is not diminished and remains to be higher than 50% (Go et al., 2014). Manifestations of HF in both HFrEF and HFpEF are dyspnea, fatigue, and exercise intolerance. However, HFpEF patients differ from HFrEF in that they are somewhat older and more likely to be female, obese and hypertensive (Andersen & Borlang, 2014). HFpEF has a high co-morbidity prevalence with obesity, diabetes mellitus, chronic obstructive pulmonary disease, and salt-sensitive hypertension. While for HFrEF patients prognosis and survival have improved due to successful therapies for heart failure, survival rates for HFpEF patients have remained unchanged in the last decades and no medical agents have been proven to improve survival among these patients (Owan et al., 2006). This implicates differences in biological origin and/or development of heart failure in HFrEF and HFpEF.
Endothelial dysfunction in HFpEF
The endothelium is under normal conditions, able to maintain vascular homeostasis and structure (for more information, see ‘The endothelium’). However, the vascular endothelium can transform into an non-adaptive functional state which can be termed as ‘endothelial dysfunction’ (Gimbrone, 1995). Others prefer to speak about endothelial activation or inflammation (Deanfield et al., 2007). Cardiovascular diseases are associated with endothelial dysfunction and
3 The endothelium
The thin layer of cells that controls the blood fluidity by the inner lining of the circulatory system is called the endothelium. It forms an interface between circulating blood or lymph in the lumen and the rest of the vessel wall (Wang et al., 2008). Endothelial cells (EC) in direct contact with blood are called vascular endothelial cells, whereas lymphatic endothelial cells are in direct contact with the lymph. Lymphatic vessels play a major role in immune surveillance and fat absorption and are thought to be involved in inflammatory diseases like atherosclerosis (Kutkut et al., 2015). Vascular EC line the entire circulatory system and these cells have unique functions in vascular biology which include fluid filtration, blood vessel tone, hemostasis, neutrophil
recruitment, and hormone trafficking (Deanfield, Halcox & Rabelink, 2007). By the production and release of several vasoactive molecules that relax or constrict the vessel, the endothelium is able to create vasomotion that plays a direct role in the balance of metabolic demand and tissue oxygen supply (Schechter & Gladwin, 2003).
the grade of endothelial dysfunction is currently seen as a predictor of cardiovascular outcomes (Higasi et al., 2009). Many cardiovascular risk factors (e.g. obesity, hypertension, diabetes mellitus) are able to activate molecular mechanisms which can result in the expression of chemokines and cytokines, and release of platelets which all can promote inflammation in specific tissues (Hansson, 2005). Endothelial activation or dysfunction is often related to an impairment of the endothelium dependent vasodilatation caused by the loss of NO bioactivity in the vessel wall (Lam & Brutsaert, 2012). This is relevant for the HFpEF syndrome because patients often suffer from vasodilator impairment, especially during exercise (Borlaug et al., 2006).
Recently more insight into myocardial structure and function led to a new paradigm on the development of HFpEF (Paulus & Tschöpe, 2014). Researchers presented a model in which co-morbidities can induce a systemic inflammatory state. This chronic inflammation results in coronary microvascular endothelial inflammation and promotes cardiomyocyte stiffening, and a decrease in left ventricular diastolic function (Fig. 1).
Comorbidities in HFpEF thus contribute to systemic microvascular endothelial dysfunction by induction of a systemic inflammatory state. If we want to reduce the inflammatory effect of comorbidities on the endothelium and cardiomyocytes, it is relevant to look at the mechanism behind this systemic inflammation.
The Role of Extracellular Vesicles in HFpEF and Endothelial Activation
Recently extracellular vesicles (EVs) including exosomes, microparticles (MPs) and apoptotic bodies have been given great attention in relation to endothelial dysfunction and CVD (see
‘Extracellular vesicles: subtypes and function’ for more information).
Extracellular vesicles are found to have great influences on blood coagulation, inflammation and endothelial function (Tushuizen et al., 2011), factors which are all important in HFpEF development. EVs circulating in the blood could induce endothelial dysfunction by directly acting on endothelial cells and decreasing NO production even as increasing expression of pro-inflammatory proteins like ICAM-1, E-selectin and integrins (Gaceb et al., 2014), proteins which are also present in the myocardial remodelling model (Paulus & Tschöpe, 2013). This promotes the idea that extracellular vesicles are involved in endothelial dysfunction, and therefore also in HFpEF development.
4 Extracellular vesicles: subtypes and functions
Extracellular vesicles can be generated from the endosome-derived multi-vesicular bodies (exosomes) or produced by budding from the plasma membrane which is the case for MPs and apoptotic bodies (Andaloussi et al., 2013). In 2012 reviewers concluded that size of apoptotic vesicles ranges between 1000 and 5000 nm whereas microparticles (or microvesicles) and exosomes are much smaller, 20-1000nm and 50-100nm respectively (Van der Pol et al., 2012). However, nomenclature is still a matter of debate (Gould & Raposo, 2013) and in recent literature exosome sizes are found to be between 30-100nm, 50-120nm or 50-150nm (Rashed et al., 2017; Xu et al., 2016; Davidson et al., 2017). In this research we will define exosomes as particles with a size ranging from 30-150 nm.
When present in the extracellular matrix or blood, EVs can bind target cells by adhesion molecules that are present in EVs. Bound vesicles may fuse directly with the plasma membrane or may be endocytosed by a target cell (Raposo & Stoorvogel, 2013). In this way EVs may transfer lipids, proteins, mRNA, miRNA, second messengers and cell organelle factions (Gaceb, Martinez & Andriantsitohaina, 2014). EVs are therefore capable of regulating a diverse range of biological processes in cells, and can be considered as signalosomes for several biological core processes.
Figure 1. Comorbidities Drive Myocardial Dysfunction and Remodeling in HFPEF.
A high prevalence of comorbidities induces a systemic proinflammatory state with elevated plasma levels of cytokines like IL-6 and tumor necrosis factor alpha (TNF-α), which cause coronary microvascular endothelial inflammation. Coronary microvascular endothelial cells produce reactive oxygen species (ROS), vascular cell adhesion molecule (VCAM), and E-selectin in a reaction towards this inflammation. In this activated endothelium, production of ROS reduces NO bioavailability, cGMP content and PKG activity in nearby cardiomyocytes. Hypertrophy development is favored by low PKG activity and resting tension increases. Both interstitial fibrosis and cardiomyocyte stiffness in the end contribute to left ventricular stiffness and development of HFpEF. Adapted from Paulus & Tschöpe, 2013, A novel paradigm for heart failure with preserved ejection fraction. Journal of the American College of Cardiology, 62(4), 263-271
As a subtype of EVs, exosomes are thought to be involved in several cardiovascular diseases (Davidson, Takov & Yellon, 2016). Next to other EVs, exosomes could be pathologically relevant because these vesicles are not only carrying waste products of their secreting cell. Exosomes transfer biological information to neighboring cells and are involved in several normal physiological functions such as cell-to-cell communication. Besides, exosomes are also shown to be engaged in the pathogenesis of diseases like thrombosis, diabetes, atherosclerosis and, more recently in cancer (Rashed et al., 2017).
5 Tissue factor and TFPI
Under normal conditions TF is mostly present on the surface of endothelial cells and very low levels of TF are present in the blood of healthy individuals. Circulating TF is mostly present in in its full length (flTF) form in and/or on MPs. Microparticle levels are often found to be increased in several diseases including CVD (Morel et al., 2006). TF is also present in the blood in its soluble, alternatively spliced (asTF) form. However, procoagulant activity is reduced in this alternative form of TF due to the absence of a transmembrane domain (Kasthuri, Taubman & Mackman, 2009). Tissue-specific regulation of splicing can determine expression of full length or asTS. TF can together with factor VIIa initiate blood coagulation by activating factor X and factor IX, which activates prothrombin to thrombin, the protease that converts fibrinogen into fibrin, what in the end will result in cross-linked fibrin clots (Rauch, 2000). At rest the extrinsic coagulation pathway is inhibited in endothelial cells by anti-thrombogenic factors like TF pathway inhibitor (TFPI) which blocks the activity of the TF-FVIIa complex. Due to alternative splicing TFPI can exists in two isoforms (α and β) which can both inhibit thrombus formation by interfering in the extrinsic coagulation pathway, although on different levels (Mast, 2015).
HFpEF and coagulation
HFpEF is associated with a procoagulant state (Jug et al., 2009) and coagulation is an indicator for several other (inflammatory) vascular diseases (Levi, Van der Pol & Büller, 2004). In vivo coagulation needs the availability of the transmembrane protein tissue factor (TF), which is the initiator of the extrinsic coagulation pathway (Van Hinsbergh, 2012). Tissue factor is therefore seen as the principle initiator of coagulation (see ‘tissue factor and TFPI’ for more information). During an inflammatory state, as in HFpEF, endothelial cells become activated and change their expression pattern for both TF as TFPI. This new phenotype disrupts the hemostatic balance and promotes procoagulability (Witkowski, Landmesser & Rauch, 2016). Besides its role in hemostasis, TF has also signaling activities and promotes inflammatory responses in concert with other coagulation factors (Witkowski et al., 2016). In this research we would like to examine the effect of healthy controls EVs and HFpEF patients EVs on TF concentration because this protein can serve as an indicator of a procoagulant (and inflammatory) state of endothelial cells.
Sex Differences in HFpEF
Although researchers have been aware of sex-differences in CVD for a long time (Sullivan, 1981), most clinical research has been done with male patients suffering from cardiovascular diseases. This was done although in many cardiovascular diseases woman are often as much, or even more often diagnosed with the disease than men (Nichols et al., 2014). However during the past decade more emphasis had been placed on the differences in prevalence and (molecular) development of CVD in men and women (Mendelsohn & Karas, 2005; Mosca et al., 2011). Women have a lower quality of life when diagnosed with HFpEF, although they have an over-all better survival prognosis (Lam et al., 2012). Because more and more women (and men) in the future will suffer from HFpEF and there are no decent biomarkers or treatments available yet, the Dutch Queen of Hearts program was set up. The Queen of Hearts program aims to contribute to the understanding of the underlying pathogenesis of diastolic dysfunction and HFpEF. By trying to discover novel biomarkers from blood-derived sources (miRNA, extracellular vesicles, micro-environment, and circulating cells) they ought to look for specific biomarker discovery based on microvascular pathology (Den Ruijter et al., 2015).
6 This research will, as part of the Queen of Hearts program, focus on the (sex-specific) effects of extracellular vesicles derived from the blood plasma of HFpEF patients and healthy subjects on endothelial cells by using TF as an indicator for endothelial activation.
METHODS
Cell Culture
HMEC-1
Human microvascular endothelial cells (HMEC-1) were used to demonstrate the effect of extracellular vesicle exposure on tissue factor activity. HMECs were cultured at 37°C in a 5% CO2 humidified incubator in endothelial growth medium (for media details, see supplementary material).. Cells were grown in T75 flasks and were coated with 0.1% gelatin before use. Trypsin was used to detach the confluent cells when they were transferred into a new T75 flask or 96 well plate. For the experiments cells up to passage 30 were used.
HUVEC
Human umbilical vein endothelial cells (HUVECs) were obtained from umbilical veins provided by the UMC Utrecht. Cells were isolated according to a HUVEC isolation protocol (see supplementary material). Cells were cultured at 37°C in 5% CO2 in full EGM2 medium (for medium details, see supplementary material). Accutase was used to detach the cell. Cells up to passage 6 were used for all the experiments. For all the tissue factor experiments pooled (n=4) HUVECs were used for both male as female cells to compensate for differences in individual cell cultures.
Isolation of Extracellular Vesicles
Blood plasma was isolated from whole blood by a 10 minute centrifuge step of 1850 x g at room temperature (RT). The aim of the isolation was to obtain exosomes, as stated before, a subcategory of extracellular vesicles with a size ranging from 30-150 nm.
Donors and subjects
In this experiment blood was obtained from 6 healthy controls. Donors were chosen to be representative of both sexes (3 men and 3 women) and ages ranged from 21 till 36. HFpEF patients diagnosed with diastolic dysfunction at the Cardiologie Centrum Utrecht were aged 71, 76 and 80 (woman) and 73, 82 and 88 (men). Female healthy controls were aged 21, 22 and 23 years old and male controls were aged 20, 25 and 36.
Ultra-centrifuge
Blood plasma was isolated from HfpEF patients (700 µl) and healthy control subjects (1.5 ml) and plasma was stored in -80°C before use. Samples were thawed and first centrifuged at 1,850 x
g at RT for 10 min to eliminate large dead cells and large cell debris. The supernatant was
transferred into an optiseal polyallomer tube (4,9 ml, Beckman Countler) and filled with PBS (Life Technologies) before it was centrifuged twice with a Beckman Countler Optima XE-90 ultracentrifugeusing a NVT 65.2 rotor (Beckman Counlter)at 33,500 rpm at 4°C for 30 minutes and finally dissolved in 1 ml PBS (for full protocol, see supplementary material). Samples were stored at -80°C and were before use, vortexed for 1 minute and centrifuged for 10 minutes at
7 1850 x g. By performing these steps extracellular vesicles or other larger particles that may have congealed against each other are loosened again and large particles would be spun down.
Nanoparticle Tracking Analysis (NTAs) for extracellular vesicles
NTA was performed using the NanoSight NS500 system (Malvern instruments) on extracellular vesicles (with a specific aim for exosomes) resuspended in PBS. The NanoSight system focuses a laser beam through the suspension of the particles of interest. Vesicles are visualized by light scattering in such a way that they can be visualized via a 20 x magnification microscope (Malvern, n.d.). After several control experiments a 1:20 PBS dilution was determined to result in a suitable particle concentration for analysis with NTA (1 x 108 to 1.5 x 109 particles/ml). A detection threshold of 3 was used for both healthy controls as HFpEF samples and camera level was set on 13 and 14, respectively. Results were analyzed with NanoSight NTA software version 3.1.
Tissue Factor Assay
To investigate whether extracellular vesicle exposure had an effect on TF concentration in endothelial cells, a tissue factor-dependent thrombin generation test was performed. By using a calibration curve of known TF concentrations, the amount of TF can be determined. Optical density was measured every 30 seconds for 90 minutes at 405nm using Soft Max Pro software (for principle, antibodies and reagents, and full protocol see supplementary material). TNF-α was used as a positive control for coagulation since it is known as an important inducer of the extrinsic coagulation pathway (Van Deventer, 1997).
HMEC.1 and HUVEC cells were used for the TF assays. Cells were seeded in 0.1% gelatin-coated 96 well plate and were seeded in a density of 2 x 104 cells per well (HUVEC) or in different concentrations (HMEC.1) and used next day. TF expression was induced by 5 hour stimulation with α (5ng/ml) and/or 2 hour starvation conditions (EGM2 without FBS) previous to TNF-α stimulation on the day of the TF assay. EVs were added to cells alone or together with TNF-TNF-α during the 5 hour incubation. For all conditions samples were tested in triplo.
RNA Isolation, cDNA Construction and qPCR for TF Gene Expression
Pooled HUVECs (female and male separated, 4 cultures per pool with p.3) cultured in a 12 well plate were stimulated for 2 hours with starved medium (EBM2 without any supplements). After incubation all cells received new full EGM2 medium and were incubated with TNF-α (5ng/ml) for 5 hours after which the cells were lysed with 350µl RA1 buffer containing 1µM DTT. Lysated samples were stored at -80°C. RNA was extracted from these samples with a NucleoSpin® RNA kit (Machery Nagel) according to manufacturer’s instructions. A qScript cDNA synthesis kit (Quantabio, 95047-100) was used to reverse-transcribe the RNA to cDNA and real time PCR was performed on Bio Rad CFX96 Real-Time PCR detection system. For RNA quality control, primers and the qPCR procedure see supplementary material.
Statistics
Data were analyzed by two-way ANOVA tests with a Tukey multiple testing correction (GraphPad Prism software version 7.02). A probability value of less than .05 was considered to be statistically significant. All data are shown as mean ± SD. Data are compared within each group (e.g. female HUVECs), unless described otherwise.
8
RESULTS
Exosome Isolation
In this research we ought to isolate exosomes from blood plasm of healthy controls and HFpEF/diastolic dysfunction patients. Before working with patient material several ready available blood plasma samples were used to optimize the exosome isolation procedure.
For healthy control samples which were used during the TF assay, mean particle size was 186.02 (mean SD =108.74), mean mode was 138.92 (SD=27.4) with a mean particle concentration of 8.342 x 108 particles per ml (1:20 dilution). For HFpEF patients EVs mean particle size was 168.9 (mean SD=88.4), mean mode was 84.8 (SD=29.1) with a mean particle concentration of 4.074 x 108 particles/ml (1:20 dilution). In controls, mean particle size, mode of particle size and concentration were higher in males than in females. In HFpEF samples, mean particle size was higher in males but mode of particle size and particle concentration were higher in females.
Ideally isolated vesicles must have a size from about 30-150nm to be categorized as exosomes. However in all samples particles bigger than 150nm were detected, for example peaks at 288 nm (Fig. 2). Therefore we could not conclude that our sample contained only exosomes but nevertheless we continued with the use of the samples in further experiments.
Figure 2: an example of a Nanosight analysis output (sample 1 helpful, 1:20 dilution in PBS) with on the horizontal axis particle size and on the vertical axis concentration in particles/ml. Highest peaks are visible at 83nm, 47 nm and 127 nm.
TNF-α Stimulation in HMECs
Human immortalized Microvascular Endothelial foreskin cells (HMEC) were used to validate the inflammatory effect of TNF-α which could induce higher tissue factor concentrations in microvascular endothelial cells. No significant results were seen between TNF- α and nonstimulation (for data, see supplementary material). Because earlier TF assays had been
9 performed with HUVECs we decided to start with assays using these cells. Furthermore, sex-differences can only be analyzed in HUVECs because there are no female HMECs.
EVs from Healthy Controls and TNF-α Stimulation in HUVECs
Cells were stimulated with TNF-α, EVs from healthy controls or a combination of these two (Fig. 3A). Unstimulated male HUVECs showed a significant higher TF concentration than female cells (P= .0013) and after TNF-α stimulation TF was higher in male HUVECs as well (P<.0001). For both female as male cells TNF-α stimulation significantly differed in TF concentration compared with non-stimulation (respectively P= .0095 and P<.0001). No significant differences were observed between unstimulated cells and EV exposure. Mean TF concentration for control wells without cells was 2,2545 pM. Addition of EVs to TNF-α did not had an effect in female cells but decreased the effect of TNF-α in male HUVECs (P<.0001).
Figure 3. TF concentrations (pM) in HUVECs. A: 50 µl of diluted (1:20 in PBS) sample 1 (female, 5.27 x 108 +/- 8.40 x 107 particles/ml) and sample 4 (male, 7.61 x 108 +/- 3.80 x 107 particles/ml) were used. Female EVs were added to pooled female HUVECs and male EVs to pooled male HUVECs. In this experiment no anti-TFPI was used. Male HUVEC show higher TF concentrations than female cells in the unstimulated and TNF-α stimulated condition (P= .0013 and P<.0001). TNF-α successfully induced TF expression. In males the effect of TNF-α was
reduced in combination with EV exposure(P<.0001). B: Pooled female culture of previous experiment was used together with the same female control vesicles. TNF-α combined with EV exposure led to the highest TF concentration in female HUVECs.
In a follow-up TF experiment a control with 50 µl PBS was supplemented to examine the effect of the PBS in which the extracellular vesicles are dissolved on HUVECs (Fig. 3B). Moreover a condition with serum-starved medium was introduced to make cells more stressed, a factor that
10 could increase TF concentration. HUVECs in this condition were incubated with serum-starved medium for 2 hours prior to the 5 hour TNF-α/EV/PBS stimulation.
Serum-starvation did not have an effect on TF concentration in these female HUVECs. TF concentration after TFN-α stimulation seemed to be increased but was not significant higher in full medium (P=.0576), but was higher under serum-starved conditions (P= .0193) compared to non-stimulation. For both full and serum-starved medium conditions no difference was observed in TF concentration between unstimulated cells and EV or PBS stimulated cells. Stimulation with TNF-α and EVs together seemed to increase the amount of TF present in the cells for both serum-starved as full medium conditions compared to TNF-α alone, still this effect was not found to be significant. However, unstimulated cells, EV stimulated cells and PBS control cells which did not significantly differ from regular TNF-α stimulation did show a significant lower TF concentration compared to TNF-α+EV stimulation (P=.0011, P=.0046, P=.0027, respectively) in full medium. Nonetheless TNF-α stimulation increased TF concentration compared to unstimulated cells when medium conditions were grouped together (P=.0018), although not as much as TNF-α and EVs together compared to non-stimulation (P<.0001).
Stimulation with Control and HFpEF Patient EVs in HUVECs
Because serum-starvation did not influence TF concentration, this condition was not implemented in further experiments. EV exposure in a 1:20 dilution did not seem to affect TF concentration in HUVECs. Therefore an experiment with undiluted samples was set-up to expose cells to a higher amount of vesicles. Samples most equal to another particle concentration were selected. Since earlier experiments seemed to show a slight increase in TF concentration after EV addition to TNF-α compared to this stimulator alone, this condition was included in the following experiment (Fig. 4A). The unstimulated condition was performed in triplo to give a more robust outcome for sex-differences.
In unstimulated cells, female HUVECs showed a higher TF concentration compared to male cells (P=.0169). No other significant sex-differences were observed although stimulation with patient EVs seemed to lead to a higher TF concentration in female cells than in male HUVECs (P=.0670). For male and female cells no significant differences were found between stimulation with control or patient EVs. Likewise no differences were observed between unstimulated cells and EV exposure in general. As a positive control, TNF-α for both male and female cells increased TF concentration (both P<.0001). In contrast to earlier findings, addition of EVs to TNF-α did not increase the effect of TNF-α on TF concentration for both control as patient vesicles. Instead, TF concentration was lower in female cells exposed to α and patient vesicles compared to TNF-α and PBS stimulation (P=.0346). In male HUVECs this effect was seen for both control and patients vesicles in combination with TNF-α compared with TNF- α and PBS (P=.0242 and
P=.0210 respectively).
In earlier experiments we did not account for the difference in vesicle concentration. In a follow-up (Fig. 4B) we calculated which amount of every sample that had to be used to get an equal amount of vesicles (1010) per well. Moreover we used for both males and females two samples of healthy controls and two samples of patients EVs in the only EV stimulation conditions. A condition with sonicated extracellular vesicles (30 sec, 24 amplitude microns) was included with the aim to expose cells not only to the intact vesicles, but also to only the content of the vesicles by destroying the vesicle membrane during sonication.
11 Looking at sex-differences, TF concentration was higher in male HUVECs after TNF-α stimulation + control vesicles compared to female cells (P=.0035). As expected, both in female as in male cells TNF-α stimulation increased TF concentration (both P<.0001). Addition of non-treated control and patient EVs did not affect TF concentration in both male as female cells compared to unstimulated cells. However a small decrease in TF concentration was observed after addition of sonicated female vesicles compared to non-stimulated female HUVECs (P=.0249). No further differences were seen between sonicated and non-treated vesicles.
Figure 4: TF concentration (pM) in male and female HUVECs after stimulation with TNFα, control EVs, patient EVs, PBS, sonicated EVs and combined conditions. A: Particle concentration: healthy controls: male sample 6.09 x 108 particles/ml (1:20 dilution measurement for all samples) and female sample of 7.25 x 108 particles/ml. Patient EV female sample of 6.11 x 108 particles/ml) and male sample of 2.92 x 108 particles/ml. TF concentration was under non-stimulation higher in female HUVECs than in male cells (P=.0169). In male HUVECs addition of control and patient vesicles together with TNF-α seemed to decrease TF concentration compared to TNF-α with PBS (P=.0242 and P=.0210) and for female cells only TNF-α with patient EVs had this effect (P=.0346). B: EV particle concentration was calculated to be equal for all samples. In male cells TNF-α stimulation together with patient vesicles led to a lower TF concentration compared to TNF-α stimulation with control vesicles (P=.0002). In female cells the stimulating effect of TNF-α was less reduced when sonicated patient vesicles were added compared to non-sonicated vesicles (P<.001).
In male cells, TF concentration was lower after TNF-α stimulation with patient vesicles compared to TNF-α stimulation with control vesicles (P=.0002). For female cells this trend is also visible, although not significant (P=.0564). Contradicting with earlier experiments, no differences were seen between TNF-α stimulation alone compared to a combination of TNF- α with vesicles. However, in female cells TF concentration was lower in a condition where TNF- α
12 no n-stim ulat ed TN F-a (5 ng/m l) star ved med ium 0.00 0.05 0.10 0.15 2 ^ c t v a lu e Female HUVECs Male HUVECs
****
****
and patient vesicles were combined compared to TNF-α with sonicated patient vesicles (P<.001).
TF gene expression in Pooled Male and Female HUVECs
A qPCR analysis was performed to examine sex-differences in TF that were observed in the TF assays. Conditions with unstimulated cells, stimulation with TNFα and starved medium were included (Fig. 5). No sex-differences were observed in TF expression under all three conditions. Stimulation with TNF-α upregulated TF in both male as well as female cells (P <.0001). A 2 hour medium starvation was included to look at TF expression in stressed cells, but did over-all not increase TF expression.
Figure 5: TF gene expression in non-stimulated, TNF-α stimulated and medium-starved HUVECs. Gene expression given as a 2∆CT value in which TF and B-actin expression are compared. TNF-α increased TF expression for both male as female cells (P <.0001).
DISCUSSION
Extracellular Vesicle Isolation
In this research we ought to isolate exosomes (a subset of extracellular vesicles) from control and HFpEF/diastolic dysfunction patient blood plasma to examine the effect of these vesicles on endothelial cells. In theory exosomes range in size from 30-150 nm (Lobb et al., 2015), but researchers admitted the fact that it is very hard to discriminate between exosomes and other extracellular vesicles like microvesicles (Colombo, Raposo & Théry, 2014). Also in this research we did not succeed in isolating pure exosomes. In every sample particles with a size above 150nm were detected, although most NTA results showed the highest concentration for small (<150nm) vesicles. Even from particles with a smaller particle size, one cannot conclude that these particles are exosomes. This is because other blood components such as lipoproteins, plasma proteins and macromolecular protein complexes can be of a similar size as exosomes and will (to some degree) contaminate the isolated vesicles (Davidson, Takov & Yellon, 2016). However, one could argue that large (>150nm) detected particles may have been exosomes that are congealed towards each other. Although some peaks in the Nanosight results supports this idea, it is impossible to verify this hypothesis by nanoparticle tracking analysis. Transmission electron microscopy (TEM) could be used to distinguish exosomes from other microparticles by its typical cup-shaped morphology. However, also other vesicles with a the same particle size as
13 exosomes are seen to show this morphology, making classification of vesicles still complicated (Van der Pol et al., 2012).
Because analysis and characterization of exosomes is quite complex, one must try to reduce contamination with other (larger) extracellular vesicles as much as possible. In this research only one step of 10 min centrifuging at 1850 x g was performed before two steps of 30 minutes ultracentrifuge at 33500 rpm, while in an often used protocol first three low-centrifuging steps of 300 x g, 2,000 x g and 10,000 x g are performed to get rid of (dead) cells and cell debris after which two ultra-centrifuging steps of 70 minutes at 100,000 x g follow (Théry et al., 2006). In future experiments one could examine if including these low-centrifuge steps and longer ultra-centrifuge steps will lead to a more pure exosome sample with less contamination of other microparticles. Furthermore a filtration step through 0.22 μm filters before ultracentrifugation could be included to remove undesirable plasma-contaminating components (Muller et al., 2014). Also separation by density gradients may be utilized to remove contaminating impurities (Van Deun et al., 2014).
Another point of discussion is the amount of extracellular vesicles which we used in our experiments. One ml of blood may contain about 1010 exosomes (Caby et al, 2005). For our HFpEF/diastolic dysfunction patients mean particle concentration was 4.074 x 108 particles/ml in a 1:20 dilution, which is 8.148 x 109 particles/ml for a non-diluted sample. In the last experiment we exposed (for every sample) 1010 vesicles to the wells which contained 2 x 104 cells in a total fluid amount of 200µl. Vesicle concentration was thus five times higher in our experiment than under normal blood conditions. In future experiments the effect of different vesicle concentrations can be studied. While doing this, researchers must take into account that no abnormal high vesicle concentrations are used, since this may not represent actual physiology in coronary heart vessels.
We found smaller particles for patient than for control EVs. This might however been due to different camera levels which must be kept constant in future experiments. In controls mean particle size, mode of particle size and concentration were higher in males than in females. Future experiments could further examine these findings for potential biological relevance.
Functional Tissue Factor Assays and qPCR analysis
In none of our experiments addition of extracellular vesicles from patients or healthy controls to endothelial cells resulted in a difference in measured tissue factor concentration compared to unstimulated cells. Also sonication of vesicles, with the aim to expose the content of vesicles to endothelial cells, had no significant effect on HUVECs. Earlier, researchers found that sonication (30 minutes in sonicator bath) abolished the effect of exosomes on HTR-8/SVneo and JEG-3-derived exosomes effect on VSMC migration (Salomon et al., 2014). Because our vesicles were sonicated using a ultrasonicator with even a higher frequency, the content of the vesicles probably lost their biological function. Therefore one could argue that sonicated vesicles in this research could serve as a negative control. For future experiments more research has to be performed on a correct sonication time and frequency whereby the vesicle membrane, but not the content of the vesicles is affected. Also other options including the use of membrane lysis buffers can be considered.
It was interesting that in some experiments the effect of TNF-α on TF concentration seemed to be slightly diminished when EVs were added (Fig. 4). Our data even gives the impression that
14 patient vesicles can more strongly inhibit the effect of TNF-α on TF concentration, when sonicated vesicles are seen as a negative control (Fig. 5B). From this experiment we could conclude that extracellular vesicles are not actively involved in TF concentration under normal conditions. However, when endothelial cells are activated extracellular vesicles might be involved in reduction of the inflammatory or coagulatory effect of our stimulator TNF-α. Our results indicate that EVs from HFpEF patients may even more strongly inhibit the coagulatory effect of TNF-α (Fig. 6).
Figure 6: Proposed model for inhibition of TF concentration in HUVECs by exposure of extracellular vesicles during TNF-α stimulation.Vesicles from HFpEF patients may have a larger effect than vesicles from healthy control subjects.
It would be promising to further examine this effect with more samples because earlier research has shown that exosomes can have anti-inflammatory functions (Rashed et al., 2017). However, others pointed to the fact that microparticles (which probably have been present in our EV samples) are membrane vesicles with procoagulant and proinflammatory properties, which then would support the idea contributing to pathology in diseases like CVD (Meziani, Tesse & Andriantsitohaina, 2008).
In future experiments one could combine coagulation assays with experiments in which the effect of exosomes, or vesicles, on gene expression can be analyzed. This can be done by qPCR analysis of genes involved in inflammation, coagulation and endothelial dysfunction in HUVECs or other (endothelial) cells after incubation with HFpEF patient and control EVs.
Next to analysis of up- or down-regulation of proteins involved in biological processes that are relevant in HFpEF development in endothelial cells after EV exposure, it would also be relevant to examine the content of these vesicles. Cells that take up EVs by endocytosis might for example be affected by the proteins, lipids, mRNA, and miRNA that EVs are loaded with by their secreting cells. Recently a lot of attention is given to the potential pathological role of micro RNAs (miRNAs) in heart failure (Vegter et al., 2016). If one can find structural differences in miRNA content in EVs form controls or patients, this could have great diagnostic, prognostic and/or therapeutic value. Also analysis of exosomal/microparticle source could be interesting during examination of HFpEF development to see if there are for example differences in cellular origin between pathogenic and protective vesicles.
Recommendations to optimize TF assays
In this study we observed contradicting results when sex-differences in tissue factor expression were examined, mainly in unstimulated conditions (Figures 3A and 4A). In this study often high
15 ranging values were measured between triplicates of the same condition. However, I do recommend to continue with the tissue factor assay as well because this functional assay may provide more information on actual blood coagulation that could result from EV exposure. This because the majority of TF resides on the cell surface in an active and encrypted form, which can be rapidly decrypted after stimulation with e.g. calcium, cell disruption or ionophores (Bach, 2006). These fast post-transcriptional effects might not be observed with a gene expression analysis.
Some changes can be made in the assay to make it more representative for actual physiological conditions. I would suggest to examine the effects of EV exposure on HUVECs during a tissue factor-dependent factor Xa thrombus generation test sinc This test is proven to be effective after microparticle exposure (Hisada et al., 2016). By using an assay without the TFPI antibody, a more representative TF concentration can be measured since, as mentioned before, TF is highly influenced by TFPI. Exposure of EVs to endothelial cell surfaces may not only directly, but also indirectly affect TF by TFPI. However, more research must be performed on this subject since still a lot is unknown about how EVs could affect TF or TFPI via e.g. posttranscriptional regulation in alternative splicing.
During the experiments performed in this study we ought to examine the effects of EV exposure on coagulation by measuring (indirectly) tissue factor. One must take into account that HUVECs are not representative for all endothelial cells. Conclusions from these experiments can therefore not be translated to e.g. coronary microvascular endothelial cells, which are more relevant in the research field of HFpEF/diastolic dysfunction. Furthermore HUVECs were pooled to account for inter-culture variability, but it could be that problems mostly occur in the endothelium of the individual. Moreover all these experiments were performed in vitro with apparently healthy HUVECs and it is hard to generalize results to an in vivo pathological situation . We also worked with EVs from healthy control subjects with a mean age of 24.5 while the HFpEF/diastolic dysfunction patients had an average age of 78.3. In future experiments it would be interesting to work with three groups (young healthy, older healthy and older HFpEF/diastolic dysfunction) of subjects to compensate for age effects.
CONCLUSION
Heart failure with preserved ejection fraction (HFpEF) can develop after a systemic inflammation of the microvascular endothelium, which results in cardiomyocyte stiffening and left ventricular dysfunction. In this study we wanted to examine if extracellular vesicles can play a role in the underlying (sex-specific )mechanism of HFpEF development. Vesicles (with a aim for exosomes) were isolated from HFpEF patient and control blood plasma and exposed to HUVECs. Thrombus formation was measured as an indicator for endothelial activation during a tissue factor assay. Some sex differences were observed in TF concentration, but experiments showed contradicting results. Serum-starvation did not have any added value to the TF assay. Results indicated that neither addition of patient nor addition of control EVs affected TF concentration in HUVECs compared to non-stimulation. Sonication of vesicles, with the aim to break down only the vesicle membrane, probably destroyed the whole vesicle including the biological functional content. Some experiments showed an inhibition of the coagulatory effect of TNF-α when EVs were added to this condition. Vesicles from HFpEF patients even seemed to
16 inhibit this effect more strongly. However, due to high variability in the assay it is hard to draw clear conclusions from our experiments.
In future research, the exosome isolation protocol can be optimized by addition of more centrifuging steps and inclusion of a filtration step to obtain a more pure exosomes. A factor Xa generation test can be used as an alternative tissue factor assay. Furthermore the content of vesicles can be analyzed to search in differences between HFpEF patients and controls. Next to TF, also gene expression of other proteins involved in coagulation, inflammation or endothelial dysfunction can be examined in endothelial cells after incubation with patient and control vesicles from both males as females. This all, might in the future lead to a better understanding of the (sex-specific) development of HFpEF. Fundamental knowledge of the origin and progress of this disease has the potential to be of great diagnostic and therapeutic value for this severe and upcoming type of heart failure.
AKNOWLEGDEMENTS
I would like to thank my professor Hester Den Ruijter from the UMC Utrecht for providing a motivating place to perform my research project. Next to her very helpful scientific insights she has the great talent to create a positive and inspiring ambiance. My daily supervisor Robin Hartman gave me a lot of freedom in choosing directions for experimental designs but was at the same time always available for questions, which I very much appreciated. Furthermore I want to thank Gideon Valstar for his help with the ultra-centrifuge and last but definitely not least Daniek Kapteijn for her help with the exiting, but sometimes stressful, tissue factor assays.
References
Andaloussi, S. E., Mäger, I., Breakefield, X. O., & Wood, M. J. (2013). Extracellular vesicles: biology and emerging therapeutic opportunities. Nature reviews Drug discovery, 12(5), 347-357.
Andersen, M. J., & Borlaug, B. A. (2014). Heart failure with preserved ejection fraction: current understandings and challenges. Current cardiology reports, 16(7), 1-11.
Bach, R. R. (2006). Tissue factor encryption. Arteriosclerosis, thrombosis, and vascular biology,
26(3), 456-461.
Bom, V. J., & Bertina, R. M. (1990). The contributions of Ca2+, phospholipids and tissue-factor apoprotein to the activation of human blood-coagulation factor X by activated factor VII.
Biochemical Journal, 265(2), 327-336.
Borlaug, B. A., Melenovsky, V., Russell, S. D., Kessler, K., Pacak, K., Becker, L. C., & Kass, D. A. (2006). Impaired chronotropic and vasodilator reserves limit exercise capacity in patients with heart failure and a preserved ejection fraction. Circulation, 114(20), 2138-2147.
Caby, M. P., Lankar, D., Vincendeau-Scherrer, C., Raposo, G., & Bonnerot, C. (2005). Exosomal-like vesicles are present in human blood plasma. International immunology, 17(7), 879-887.
Colombo, M., Raposo, G., & Théry, C. (2014). Biogenesis, secretion, and intercellular interactions of exosomes and other extracellular vesicles. Annual review of cell and developmental biology, 30, 255-289.
17 Combes, V., Simon, A. C., Grau, G. E., Arnoux, D., Camoin, L., Sabatier, F., & Dignat-George, F. (1999). In vitro generation of endothelial microparticles and possible prothrombotic activity in patients with lupus anticoagulant. The Journal of clinical investigation, 104(1), 93-102.
Davidson, S. M., Takov, K., & Yellon, D. M. (2016). Exosomes and cardiovascular protection.
Cardiovascular Drugs and Therapy, 1-10.
Deanfield, J. E., Halcox, J. P., & Rabelink, T. J. (2007). Endothelial function and dysfunction.
Circulation, 115(10), 1285-1295.
Del Conde, I., Shrimpton, C. N., Thiagarajan, P., & López, J. A. (2005). Tissue-factor–bearing microvesicles arise from lipid rafts and fuse with activated platelets to initiate coagulation.
Blood, 106(5), 1604-1611.
Den Ruijter, H., Pasterkamp, G., Rutten, F. H., Lam, C. S. P., Chi, C., Tan, K. H., & de Kleijn, D. P. V. (2015). Heart failure with preserved ejection fraction in women: the Dutch Queen of Hearts program. Netherlands Heart Journal, 23(2), 89-93.
Forstermann, U. & Munzel T. (2006). Endothelial nitric oxide synthase in vascular disease: from marvel to menace. Circulation. 113: 1708–1714.
Gaceb, A., Martinez, M. C., & Andriantsitohaina, R. (2014). Extracellular vesicles: new players in cardiovascular diseases. The international journal of biochemistry & cell biology, 50, 24-28. Gimbrone, M. A. (1995). Vascular endothelium: an integrator of pathophysiologic stimuli in atherosclerosis. The American journal of cardiology, 75(6), 67B-70B.
Ghosh, S., & Karin, M. (2002). Missing pieces in the NF-κB puzzle. cell, 109(2), S81-S96. Go, A. S., Mozaffarian, D., Roger, V. L., Benjamin, E. J., Berry, J. D., Blaha, M. J., & Fullerton, H. J. (2014). Heart disease and stroke statistics-2014 update. Circulation, 129(3).
Gould, S. J., & Raposo, G. (2013). As we wait: coping with an imperfect nomenclature for extracellular vesicles. Journal of extracellular vesicles, 2.
Govers, R., & Rabelink, T. J. (2001). Cellular regulation of endothelial nitric oxide synthase.
American Journal of Physiology-Renal Physiology, 280(2), F193-F206.
Gustafson, C. M., Shepherd, A. J., Miller, V. M., & Jayachandran, M. (2015). Age-and sex-specific differences in blood-borne microvesicles from apparently healthy humans. Biology of sex
differences, 6(1), 10.
Halcox, J. P., Narayanan, S., Cramer-Joyce, L., Mincemoyer, R., & Quyyumi, A. A. (2001). Characterization of endothelium-derived hyperpolarizing factor in the human forearm microcirculation. American journal of physiology. Heart and circulatory physiology, 280(6), H2470-7.
18 Hansson, G. K. (2005). Inflammation, atherosclerosis, and coronary artery disease. New England
Journal of Medicine, 352(16), 1685-1695.
Heidenreich, P. A., Albert, N. M., Allen, L. A., Bluemke, D. A., Butler, J., Fonarow, G. C., ... & Nichol, G. (2013). Forecasting the impact of heart failure in the United States. Circulation: Heart Failure,
6(3), 606-619.
Higashi, Y., Noma, K., Yoshizumi, M., & Kihara, Y. (2009). Endothelial function and oxidative stress in cardiovascular diseases. Circulation journal, 73(3), 411-418.
Hisada, Y., Alexander, W., Kasthuri, R., Voorhees, P., Mobarrez, F., Taylor, A., McNamara, C., Wallen, H., MacroWitkowski N.S., Key, U. & Rauch, U. (2016). Measurement of microparticle tissue factor activity in clinical samples: a summary of two tissue factor-dependent FXa generation assays. Thrombosis research, 139, 90-97.
Jug, B., Vene, N., Salobir, B. G., Šebeštjen, M., Šabovic, M., & Keber, I. (2009). Procoagulant state in heart failure with preserved left ventricular ejection fraction. International heart journal, 50(5), 591-600.
Kasthuri, R. S., Taubman, M. B., & Mackman, N. (2009). Role of tissue factor in cancer. Journal of
Clinical Oncology, 27(29), 4834-4838.
Kutkut, I., Meens, M. J., McKee, T. A., Bochaton‐Piallat, M. L., & Kwak, B. R. (2015). Lymphatic vessels: an emerging actor in atherosclerotic plaque development. European journal of clinical
investigation, 45(1), 100-108.
Lam, C. S., & Brutsaert, D. L. (2012). Endothelial dysfunction. Journal of the American College of
Cardiology, 60(18), 1787.
Lam, C. S., Carson, P. E., Anand, I. S., Rector, T. S., Kuskowski, M., Komajda, M., & Kitzman, D. W. (2012). Sex differences in clinical characteristics and outcomes in elderly patients with heart failure and preserved ejection fraction: the i-preserve trial. Circulation: Heart Failure,
CIRCHEARTFAILURE-112.
Levi, M., van der Poll, T., & Büller, H. R. (2004). Bidirectional relation between inflammation and coagulation. Circulation, 109(22), 2698-2704.
Libby, P., Ridker, P. M., & Maseri, A. (2002). Inflammation and atherosclerosis. Circulation,
105(9), 1135-1143.
Lobb, R. J., Becker, M., Wen Wen, S., Wong, C. S., Wiegmans, A. P., Leimgruber, A., & Möller, A. (2015). Optimized exosome isolation protocol for cell culture supernatant and human plasma.
Journal of extracellular vesicles, 4(1), 27031.
Malvern (n.d.). NanoSight NS500: How it works. Derived on 20-06-2017 from
19 Mast, A. E. (2015). Tissue Factor Pathway Inhibitor: Multiple Anticoagulant Activities for a Single Protein. Arteriosclerosis, Thrombosis, and Vascular Biology 36,9-14.
Mendelsohn, M. E., & Karas, R. H. (2005). Molecular and cellular basis of cardiovascular gender differences. Science, 308(5728), 1583-1587.
Meziani, F., Tesse, A., & Andriantsitohaina, R. (2008). Microparticles are vectors of paradoxical information in vascular cells including the endothelium: role in health and diseases. Pharmacol
Rep, 60(1), 75-84.
Morel, O., Toti, F., Hugel, B., Bakouboula, B., Camoin-Jau, L., Dignat-George, F., & Freyssinet, J. M. (2006). Procoagulant microparticles. Arteriosclerosis, thrombosis, and vascular biology, 26(12), 2594-2604.
Mosca, L., Barrett-Connor, E., & Wenger, N. K. (2011). Sex/gender differences in cardiovascular disease prevention. Circulation, 124(19), 2145-2154.
Muller, L., Hong, C. S., Stolz, D. B., Watkins, S. C., & Whiteside, T. L. (2014). Isolation of
biologically-active exosomes from human plasma. Journal of immunological methods, 411, 55-65. Nichols, M., Townsend, N., Scarborough, P., & Rayner, M. (2014). Cardiovascular disease in Europe 2014: epidemiological update. European heart journal, ehu299.
Owan, T. E., Hodge, D. O., Herges, R. M., Jacobsen, S. J., Roger, V. L., & Redfield, M. M. (2006). Trends in prevalence and outcome of heart failure with preserved ejection fraction. New England
Journal of Medicine, 355(3), 251-259.
Paulus, W. J., & Tschöpe, C. (2013). A novel paradigm for heart failure with preserved ejection fraction. Journal of the American College of Cardiology, 62(4), 263-271.
Pober, J. S., & Sessa, W. C. (2007). Evolving functions of endothelial cells in inflammation. Nature
Reviews Immunology, 7(10), 803-815.
Raposo, G., & Stoorvogel, W. (2013). Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol, 200(4), 373-383.
Rashed, M., Bayraktar, E., K Helal, G., Abd-Ellah, M. F., Amero, P., Chavez-Reyes, A., & Rodriguez-Aguayo, C. (2017). Exosomes: From Garbage Bins to Promising Therapeutic Targets.
International journal of molecular sciences, 18(3), 538.
Rauch, U., & Nemerson, Y. (2000). Tissue factor, the blood, and the arterial wall. Trends in
cardiovascular medicine, 10(4), 139-143.
Salomon, C., Yee, S., Scholz-Romero, K., Kobayashi, M., Vaswani, K., Kvaskoff, D., Illanes, S.E., Mitchell, M.D. & Rice, G. E. (2014). Extravillous trophoblast cells-derived exosomes promote vascular smooth muscle cell migration. Frontiers in pharmacology, 5.
20 Schächinger, V., Britten, M. B., & Zeiher, A. M. (2000). Prognostic impact of coronary vasodilator dysfunction on adverse long-term outcome of coronary heart disease. Circulation, 101(16), 1899-1906.
Schechter, A.N. , Gladwin, M.T. (2003). Hemoglobin and the paracrine and endocrine functions of nitric oxide. N Engl J Med. ;348:1483–1485.
Sharma, K., & Kass, D. A. (2014). Heart failure with preserved ejection fraction. Circulation
research, 115(1), 79-96.
Stamler, J. S., Lamas, S., & Fang, F. C. (2001). Nitrosylation: the prototypic redox-based signaling mechanism. Cell, 106(6), 675-683.
Sullivan, J. (1981). Iron and the sex difference in heart disease risk. The Lancet, 317(8233), 1293nil1294-1294.
Théry, C., Amigorena, S., Raposo, G., & Clayton, A. (2006). Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Current protocols in cell biology, 3-22.
Théry, C., Zitvogel, L., & Amigorena, S. (2002). Exosomes: composition, biogenesis and function. Nature Reviews Immunology, 2(8), 569-579.
Tushuizen, M. E., Diamant, M., Sturk, A., & Nieuwland, R. (2011). Cell-derived microparticles in the pathogenesis of cardiovascular disease. Arteriosclerosis, thrombosis, and vascular biology,
31(1), 4-9.
Van Deun, J., Mestdagh, P., Sormunen, R., Cocquyt, V., Vermaelen, K., Vandesompele, J., Bracke, M., De Wever, O. & Hendrix, A. (2014). The impact of disparate isolation methods for extracellular vesicles on downstream RNA profiling. Journal of extracellular vesicles, 3.
Van Deventer, S. J. (1997). Tumour necrosis factor and Crohn's disease. Gut, 40(4), 443. Van der Pol, E., Böing, A. N., Harrison, P., Sturk, A., & Nieuwland, R. (2012). Classification, functions, and clinical relevance of extracellular vesicles. Pharmacological reviews, 64(3), 676-705.
Van Hinsbergh , V. W. (2012). Endothelium—role in regulation of coagulation and inflammation. In Seminars in immunopathology (Vol. 34, No. 1, pp. 93-106). Springer-Verlag.
Van Riet, E. E., Hoes, A. W., Wagenaar, K. P., Limburg, A., Landman, M. A., & Rutten, F. H. (2016). Epidemiology of heart failure: the prevalence of heart failure and ventricular dysfunction in older adults over time. A systematic review. European journal of heart failure.
Vegter, E. L., van der Meer, P., de Windt, L. J., Pinto, Y. M., & Voors, A. A. (2016). MicroRNAs in heart failure: from biomarker to target for therapy. European journal of heart failure.
21 Wang, S., Aurora, A. B., Johnson, B. A., Qi, X., McAnally, J., Hill, J. A., Richardson, J.A., Bassel-Duby, R. & Olson, E. N. (2008). The endothelial-specific microRNA miR-126 governs vascular integrity and angiogenesis. Developmental cell, 15(2), 261-271.
Witkowski, M., Landmesser, U., & Rauch, U. (2016). Tissue factor as a link between inflammation and coagulation. Trends in cardiovascular medicine, 26(4), 297-303.
Wolinsky, H. A. R. V. E. Y. (1980). A proposal linking clearance of circulating lipoproteins to tissue metabolic activity as a basis for understanding atherogenesis. Circulation research, 47(3), 301-311.
Xu, R., Greening, D. W., Zhu, H. J., Takahashi, N., & Simpson, R. J. (2016). Extracellular vesicle isolation and characterization: toward clinical application. Journal of Clinical Investigation,
