Pulmonary hypertension (PH) is a fatal disease with no treatment options, characterised by elevated pulmonary vascular resistanzce and secondary right ventricular failure. The aetiology of pulmonary arterial hypertension is multiple and its pathogenesis is complex. Although the exact role of cellular microparticles remains partially understood, there is increasing evidence to suggest an active role for microparticles in PH pathophysiology. Patients with PH exhibited higher circulating levels of microparticles compared to control subjects and in vitro or in vivo generated microparticles can induce endothelial dysfunction, interfere with coagulation pathways or modulate inflammatory phenomenon. Whether or not these new conveyors of biological information contribute to the acquisition and/or maintenance of the altered endothelial phenotype is unexplored in PH and requires further study.
Pulmonary hypertension (PH) is characterised by widespread obstruction and obliteration of pulmonary arterioles, leading to a progressive elevation in pulmonary vascular resistance and subsequent right heart failure and death. The exact mechanisms of pulmonary arterial remodelling that lead to the onset and progression of PH are still largely unclear. However, many disease-predisposing factors and/or contributing factors have been identified, including inflammation, endothelial cell dysfunction, aberrant vascular wall cell proliferation and mutations in the bone morphogenetic protein-receptor type 2 (Bmpr2) gene [1–3].
Precise and dynamic crosstalk between cells is critical for cellular behaviour, including for cell proliferation, apoptosis, differentiation, migration and survival, and thus crucial for proper tissue organisation and homeostasis; abnormal cellular crosstalk can lead to the development of cancer and PH [4, 5]. These well-regulated processes to transfer information between cells occur through direct cell–cell contact, various soluble bioactive factors and through cellular microparticles [6, 7].
Microparticles are plasma membrane vesicle fragments (between 0.1 and 1 μm in size) released from various cell types during activation by agonists or physical or chemical stress, including apoptosis [6, 7]. Microparticles in body fluids constitute a heterogeneous population, differing in cellular origin (endothelial cells, platelets, leukocytes and erythrocytes), number, size, antigenic composition and functional properties. Their concentration increases specifically and differently during various pathological states, including cardiovascular diseases (table 1) . Indeed, patients with PH (group 1 and 3 with no associated cardiovascular disease) exhibit higher circulating levels of microparticles compared to control subjects [9–11]. This increase might indicate a potential involvement of microparticles in the disease onset and its progression. This article reviews the evidence supporting potential roles and importance of microparticles in the pathogenesis of PH.
General considerations on microparticles
Detectable levels of microparticles of different cellular origin (platelets, red blood cells, leukocytes, endothelial cells) circulate in the plasma of healthy subjects (fig. 1b). Circulating microparticles levels increase in a wide range of cardiovascular diseases, including uncontrolled cardiovascular risk factors, stable and unstable atherosclerotic lesions, heart failure, arrhythmias and inflammatory vascular diseases (table 1) . The presence of microparticles in plasma reflects an active balance between microparticle generation and clearance. Both phagocytic activities present in the liver and the spleen contribute to the removal of microparticles from plasma of healthy animals [12, 13]. Changes in circulating levels of microparticles might bear important clinical information in healthy subjects and in patients with cardiovascular disorders. Microparticles display various bioactive substances and receptors on their surface and harbour a concentrated set of cytokines, signalling proteins, lipids and nucleic acids. The composition of microparticles, which subtends their biological effect, is a function of the parent cell they stem from, as well as the stimulus used for their generation. Thus, the amount and nature of circulating microparticles influence their beneficial or deleterious biological effects in physiological or pathological conditions. Indeed, various experimental evidence reports significant inter-relationships between circulating microparticles (either generated in vivo or in vitro) and other protagonists of disrupted vascular homeostasis. Microparticles can modulate vascular reactivity (particularly endothelial dependent relaxation), interfere with coagulation pathways, enhance or decrease inflammatory phenomenon and influence neoangiogenesis processes. A large part of our knowledge regarding microparticles is based on data issued from the research on athero-thrombotic disease . However, recent data shed light on their role as vectors of cellular crosstalk with implications in the pathogenesis of non-atherosclerotic cardiovascular conditions, including pulmonary vascular diseases (table 1).
Circulating microparticles as regulators of endothelial dysfunction
Endothelial dysfunction is a pivotal element in the development and progression of PH, irrespective of disease origin . The dysfunctional endothelium displays, to varying degrees, an imbalanced production of several mediators (vasoconstrictors versus vasodilators; activators versus inhibitors of smooth muscle cell growth; pro-inflammatory versus anti-inflammatory signals; prothrombotic versus antithrombotic substances) leading towards an excess of vasoconstriction, smooth muscle hyperplasia and pulmonary vascular remodelling [1–3]. In addition, disturbances of the normal balance between endothelial cell proliferation and endothelial cell apoptosis have been reported in idiopathic pulmonary arterial hypertension (PAH) [16, 17]. Indeed, accumulating evidence supports the concept that increased endothelial apoptosis at the early stage and decreased endothelial apoptosis at later stages of the disease could contribute to PAH . Consistent with this hypothesis, we have recently reported that primary pulmonary endothelial cells generated from PAH lung specimens exhibit various intrinsic abnormalities and present a modified pro-proliferative, apoptotic-resistant phenotype [4, 17, 19]. Although we have shown that increased activity of the FGF2 autocrine loop is among the mechanisms needed to acquire this altered endothelial phenotype in PAH , the exact nature of pulmonary endothelial cell modification during PAH and the balance between apoptotic and anti-apoptotic phenotypes remain only partially understood. In addition, alterations of endothelial cell monolayer integrity in mice, such as a selective disruption of the endothelial peroxisome proliferator-activated receptor-γ or of the endothelial BMPR2 signalling, are sufficient to predispose to PH [20, 21].
Different groups have reported increased concentrations of different subsets of circulating endothelial microparticles (EMPs) in PH patients compared to control subjects [9–11], supporting the presence of an increased structural damage with enhanced cellular activation and/or apoptosis. Indeed, the levels of platelet endothelial cell adhesion molecule (PECAM+) and vascular endothelial (VE)-cadherin+ EMPs were also correlated with several haemodynamics pulmonary vascular remodelling parameters (fig. 1)  and inflammation biomarkers . The aetiology of the disease also appears to influence the microparticles levels: Amabile et al.  reported that PAH patients had higher levels of EMPs compared with chronic pulmonary disease-related PH subjects, whereas Diehl et al.  observed higher levels of EMPs bearing E-selectin in thromboembolic PH compared with non-thromboembolic PH subjects.
The origin of these circulating microparticles (systemic versus pulmonary endothelium) is still a matter of debate. Although a trans-pulmonary gradient of CD105+ (endoglin) EMPs has been observed, suggesting a local release of microparticles within the pulmonary vascular bed, no significant change in CD31+ (PECAM), CD144+ (VE-cadherin) and CD62e+ (E-selectin) EMP levels was observed through the pulmonary vasculature [9, 10]. These findings, observed in small samples of patients with various PH origins, could reflect the heterogeneity of the endothelial changes among patients with pulmonary vascular disease, depending on the underlying aetiology or the lesion development stage (early versus late stage). These findings might support the notion that some specific EMP subpopulations could be released by either systemic or pulmonary endothelium. Finally, it cannot be excluded that a lack of sensitivity of the measurement methods for detection of mild variations in concentrations of microparticles could also explain these results.
The circulating microparticles are essential vectors of biological information and are involved in cellular communication through a paracrine or endocrine action . Cellular effects of microparticles are directly influenced by the various antigens harboured at their surface, which is a function of the cellular origin and the stimulus involved in the vesiculation process . In vitro or in vivo generated microparticles can induce endothelial dysfunction, interfere with coagulation pathways or modulate inflammatory phenomenon . Accordingly, Tual-Chalot et al.  reported that circulating microparticles isolated from hypoxia-induced PH rats could affect the vascular tone. They showed that these microparticles, mainly of platelet and erythrocyte origin, impaired the endothelium-dependent vaso-relaxation in pulmonary arteries ex vivo and decreased the nitric oxide (NO) production by pulmonary endothelial cells through an increase in oxidative stress . The authors observed comparable results in aorta and pulmonary arteries . Taken together, all these experimental findings support the notion that circulating microparticles could contribute to both initiation and/or amplification of the pulmonary endothelial dysfunction during PH and encourage further studies of the precise pathogenic mechanism underlying this phenomenon. Moreover, we cannot exclude that the generation of microparticles and their migration to the systemic circulation could mediate a crosstalk between the pulmonary and systemic endothelium and, therefore, explain the peripheral endothelial dysfunction observed during PH [21, 24, 25].
Finally, measurement of circulating EMPs might also represent a potential new prognostic tool in patients with PH. Preliminary data reported that baseline values of circulating CD62e+ microparticles, measured before initiation of any specific therapy, predicted occurrence of adverse outcome (death and readmission for right heart failure) in a small cohort of patients with mixed aetiologies of PH . Whereas these results were not observed for other EMP subsets (CD144+ and CD31+ EMPs) and should be confirmed in a larger population sample, they are in line with data reported in coronary artery disease , heart failure  and chronic end-stage renal failure , and thus confirm their potential role as a biomarker for outcome prediction in patients with cardiovascular diseases .
Circulating microparticles as regulators of inflammation
It is now well established that inflammation plays an important role in PH in humans: histological studies of PAH lung specimens demonstrate a prominent perivascular infiltration composed of macrophages, dendritic cells, T- and B-lymphocytes, and mast cells; circulating levels of certain cytokines and chemokines are elevated including interleukin (IL)-1β, IL-6, IL-8, CX3CL1, CCL2, CCL5 and tumour necrosis factor-α serum levels, and these may correlate with a worse clinical outcome. Evidence supporting a role for inflammation in the progression of the disease has been reviewed recently by Price et al. . However, the initial process triggering inflammation onset remains unknown.
Cellular microparticles are key factors in the inflammatory process because they contribute to the endothelial production of various pro-inflammatory cytokines and chemokines. Microparticles have been shown to stimulate the release of IL-1β, IL-6, CCL2, and induce the expression of intercellular adhesion molecule (ICAM)-1, vascular cell adhesion molecule (VCAM)-1 and E-selectin [32–35]. In addition, microparticles provide a preferential substrate for non-pancreatic secretory phospholipase A2 and thus a source of lysophosphatic acid, a potent pro-inflammatory mediator and platelet agonist . Furthermore, a recent study has shown that microparticles from septic shock patients increase the expression of endothelial and inducible NO synthase, cyclooxygenase-2 and nuclear factor-κB in the heart and lung . In addition, Densmore et al.  have demonstrated that endothelium-derived microparticle sequestration in the lung is capable of initiating inflammatory cascades at pathophysiologically relevant concentrations [38, 39]. Altogether, these findings support the notion that circulating microparticles could actively contribute to the initiation and/or amplification of pulmonary vascular inflammation in the pathogenesis of PH.
Procoagulant properties of microparticles
Thrombosis is a key mechanism in onset and progression of cardiovascular diseases, including venous thromboembolic disease, atherosclerosis and PH. Chronic thromboembolic PH results from the obstruction of the pulmonary arteries by clot fragments that have migrated from deep venous thrombosis . However, PAH pathological specimens often display thrombotic lesions in the absence of clinical or pathological evidence of pulmonary embolism , suggesting an in situ clotting phenomenon . Furthermore, PH is associated with a hyper-coagulable phenotype  that includes vascular upregulation of tissue factor  and an increase in circulating levels of Von Willebrand factor or fibrinopeptide A .
Microparticles are involved in thrombus generation by different mechanisms. Microparticles can provide the anionic phospholipid surface necessary for the assembly of clotting enzyme complexes (VIIIa and Xa) promoting thrombin generation. In some cases, microparticles also directly expose tissue factor, the main cellular initiator of blood coagulation, at their surface. Furthermore, microparticles can also harbour other functional membrane or cytoplasmic effectors (such as selectins, GPIIb–IIIa, GPIb, Von Willebrand factor, arachidonic acid, thromboxane A2) that can promote prothrombotic responses . Therefore, microparticles can promote thrombin generation in vitro, but also favour thrombus propagation in vivo  or transfer their procoagulant potential to target cells. The thrombogenic potential of microparticles has been demonstrated for several subpopulations, including platelet- , endothelial- [48, 49] or leukocyte-derived microparticles .
Most of our knowledge regarding microparticles' procoagulant activity derives from data observed in atherosclerosis, cardiac surgery and conditions associated with cardiomyocyte inflammatory stimulation, such as cardiac ischaemia.
An increase in circulating global procoagulants as well as tissue factor (TF+) microparticle levels has been reported in patients with PAH . Furthermore, the levels of TF+ microparticles and Von Willebrand factor were correlated. Interestingly, the highest levels of TF+-microparticles were observed in patients with functional class III–IV and/or the weakest 6-min test performance, suggesting a potential correlation between microparticles and functional status. Comparable results were observed in patients with recurrent deep venous thrombosis, a potential trigger for acute pulmonary embolism development .
The precise molecular mechanisms governing these biological effects of microparticles are still under investigation, but several elements support the notion that procoagulant microparticles might interfere with the coagulation process in PH patients and promote the disease development in different ways. High levels of procoagulant circulating microparticles could favour recurrent deep venous thrombosis , which will in turn create silent pulmonary embolism and ultimately lead to post-thromboembolic PH [40, 53]. Conversely, the plexiform lesions and the dysfunctional endothelium could release procoagulant microparticles, which could lead to propagation of the thrombotic process among the pulmonary vessels. The resulting local thrombosis could worsen the microvascular obstruction and thus increase pulmonary artery pressure. This hypothesis is supported by the results from Bakouboula et al. , who observed an increase in the concentration of the global number of procoagulant microparticles between the jugular vein and the wedged pulmonary artery, suggesting a local release of procoagulant microvesicles. However, the authors reported no significant transpulmonary gradient of TF+ microparticles, which might be explained by the small size of the group (n=20) and the heterogeneous aetiologies of PH . Moreover, proteomics analysis in experimental models of post-thromboembolic PH has shown that this condition was associated with increased expression of coagulation proteins carried by circulating microparticles .
Potential implications of microparticles in PH pathophysiology and remaining questions
In the light of the data mentioned above, the involvement of microparticles in PH pathogenesis seems to be, at least, plausible. Experimental and clinical data suggest multiple roles of microparticles in the progression of the disease and one might hypothesise a possible global scheme (fig. 1c). Procoagulant microparticles can enhance local and systemic thrombogenicity leading to pulmonary artery clot. Moreover, pro-inflammatory microparticles could also increase endothelial inflammation, local cytokine and growth factor release. Furthermore, microparticles might also favour endothelial cell proliferation. Finally, circulating microparticles can enhance endothelial dysfunction, decrease NO release and deteriorate the vasodilation/vasoconstriction humoral balance. Altogether, these effects can support pulmonary vessel remodelling and vascular obstruction and should encourage further in vivo and in vitro studies to identify their precise pathogenic mechanisms. In addition, a very recent study has reported that epoprostenol could inhibit human platelet-leukocyte mixed conjugate and platelet microparticle formation in whole blood, demonstrating that treatments might interfere with microparticles and some of their biological roles . Since our knowledge of microparticles properties in vivo might also be influenced by the current methods used for their isolation and their measurement, there is a need to establish a consensus regarding the standard centrifugation method to prepare poor platelets plasma from whole blood samples in order to determine microparticle phenotypes and levels. To date, the appropriate method for microparticles measurement is still a matter of debate . Flow cytometry methods, which have been widely used during the past 20 years, need small sample volumes and allow complete phenotypical analysis. However, flow cytometry doesn't analyse microparticles function and is limited by its poor sensitivity, resulting from the biophysical and optical properties of the laser interaction with small vesicles (<500 μm). The “capture” and other purely biological methods inform on microparticles function but don't provide any direct quantification. Dynamic light scattering, atomic force microscopy, automated flow through single-particle optical microscopy or impedance based flow cytometry represent promising methods for the future, allowing more accurate microparticle quantification, but these methods are not currently adapted for clinical research as they are limited by their cost or their availability.
In addition to these potential pathogenic roles of microparticles in PH, microparticles are well-known modulators of different stages of angiogenesis. These particular effects of microparticles depend on the stimulation and activation status of the cell from which they originate and also on their concentration [57, 58]. Indeed, microparticles from platelets display pro-angiogenic properties by promoting capillary-like structures and pro-angiogenic factor production . By contrast, lymphocyte- and endothelial-derived microparticles exhibit pro- or anti-angiogenic properties depending on the stimuli used for their production [57, 60–63]. These additional properties of microparticles reinforce the idea that they could be potential key actors in PH pathogenesis in the early-, middle- and/or late-stage of the disease.
Whether or not these new conveyors of biological information contribute to the acquisition and/or maintenance of the altered endothelial phenotype is not fully explored in PH and requires further study. Thus, several questions remain regarding their nature and true importance in this disease. What are the compositions of microparticles in PH and what are their functions? What causes increased amounts of circulating microparticles in PH? Are they beneficial or deleterious in PH? Are they useful for disease recurrence, prognosis and monitoring of therapeutic efficacy? By which potential mechanism(s) do microparticles initiate and propagate PH?
Patients with PH exhibit higher circulating levels of microparticles compared to control subjects and it is unclear whether these new conveyors of biological information contribute to the acquisition and/or maintenance of the altered endothelial phenotype or are just a consequence of the disease progression. However, there is strong accumulating evidence that microparticles are key protagonists of the disease and they could serve as a worthy biomarker for prognosis stratification and therapeutic efficiency monitoring. Therefore, a better understanding of their importance in the disease is essential and efforts should be made in the future to improve our knowledge by further exploring cellular aspects as well as clinical implications of microparticles in pulmonary hypertension.
Conflict of interest: Disclosures can be found alongside the online version of this article at www.erj.ersjournals.com
- Received June 2, 2012.
- Accepted August 28, 2012.
- ©ERS 2013