|Home | About | Journals | Submit | Contact Us | Français|
More and more knowledge on blood microparticles (MPs) has been accumulated this last decade. However the role(s) of those small phospholipid vesicles still remain unclear. MPs, also named microvesicles or ectosomes, range from 0.1 to 1 μm in size, and are released in blood flow by various types of cells1. They are found in healthy individuals as well as in patients suffering from different diseases. The release of MPs is a highly controlled process prompted by various stimuli such as shear stress, complement attack, agonist (or pro-apoptotic) stimulation or damage1,2. If MPs were first described as cellular debris without any biological function3, nowadays they are known as cellular effectors involved in numerous physiological processes such as haemostasis and inflammation4, transfer of surface proteins5, angiogenesis6 or apoptosis7. Moreover, there is a link between MPs and several diseases such as thrombocytopenic disorder8, thrombosis1, cardiovascular diseases9, diabetes10, sickle cell disease11, haemolytic anaemia12 or sepsis13, where an increase of the number of MPs in plasma has been demonstrated. Interestingly, a decrease in MPs numbers in plasma of patients presenting with the Scott syndrome, a rare bleeding disorder, has been reported14.
MPs are also present, and accumulate, in blood products such as erythrocyte and platelet concentrates during storage15–17. Because the surface of MPs is made of negatively charged phospholipids, they may play a role in the coagulation cascade. Therefore, the purpose of this paper is to review the potential procoagulant effect of erythrocyte MPs and thus, their potential adverse effect after blood transfusion.
Blood coagulation is an essential mechanism that prevents bleeding; this highly controlled process is carefully regulated in order to maintain blood circulation in case of injury while preventing vessels obstruction by clot formation. Coagulation factors, calcium ions and procoagulant membrane surfaces are the primary components involved in coagulation activation. In eukaryotic cells, including blood cells, it is established that neutral phospholipids (such as phosphatidylcholine) are mainly present on the outer membrane, whereas negative phospholipids (such as phosphatidylserine) are at the inner side of the membrane. Concerning blood cells various translocases are activated upon stimulation and phospholipid asymmetry is disrupted. Phosphatidylserine (PhSer) is externalized, modifying the neutral membrane charge into negative18. Activated blood cells and their MPs expose negative membrane and are thus thought to be procoagulant. The presence of negatively charged phospholipid membranes and calcium ions is required for the assembly of coagulation complexes such as tenase or prothrombinase. The role of tenase complex is to convert factor X (FX) into activated factor X (FXa), in both extrinsic and intrinsic pathways. The extrinsic tenase complex is composed of tissue factor (TF) and FVIIa18 whereas the intrinsic tenase is composed of co-FVIIIa and factor FIXa19,20. The function of prothrombinase, constituted of co-FVa and FXa, is to convert prothrombin into thrombin. FIXa and FXa can convert their substrate without their respective cofactors, however negatively charged membranes catalyses the reaction, for example 3x105 fold faster in the case of prothrombinase21,22. Figure 1 presents the possible site of actions of MPs on the coagulation cascade. The importance of negative phospholipids is to facilitate complex association, speeding up their association by around 1,000 fold for efficient haemostasis23. Because MPs express PhSer on their membrane, they could contribute to coagulation by providing catalyzing negatively charged membranes. It has been demonstrated that platelets-derived MPs have from 50- up to 100-times more procoagulant activity than platelets24. MPs membranes support more efficiently thrombin formation than platelet membrane when corrected for unit of surface, at least in vitro4,25. Not only platelet MPs are involved in this process, but erythrocyte MPs and other MPs have also a procoagulant activity through their negatively charged membrane as well1,11,26.
Although controversial, some authors claim that MPs support clotting not only through their negatively charged membranes, but also by expressing an inactive form of tissue factor (TF) on their surfaces. The origin and mechanism of action of blood-born TF is still subject to debate, nevertheless authors have suggested that MPs may be a reservoir of blood-born TF27. Furie et al delineated a “microparticles accumulation pathway” as part of the coagulation cascade25. According to this model, some monocyte-derived MPs expressing TF in an “inactive” form bind to a forming clot in vessels and display active TF that helps to amplify coagulation. In an other hand, Connor et al proposed an activated factor X assay to measure clotting time of various samples containing MPs and their results show that there is a correlation between the number of MPs and clotting time; interestingly enough, the assay was insensitive to the presence or the absence of TF28. Other studies mention the presence of TF on endothelial-derived MPs29, platelet-derived MPs30,31, or even on erythrocyte-derived MPs32, while some others did not detect any MPs-exposing TF11. All these pieces of evidence are difficult to explain, in particular because results often depend on the study design, methods of MPs isolation, and various pre-analytical factors; therefore one cannot exclude that discrepancies might result from different experimental methodologies33.
Hypercoagulability state is associated with an increased risk of fibrin deposition in blood vessels, which could result in thrombus formation34. Hypercoagulable state has been often observed in case of chronic haemolytic anaemia’s (sickle cell disease, thalassemia, paroxysmal nocturnal haemoglobinuria, autoimmune anaemia)35. Frequently, there are erythrocyte membrane alterations with accumulation of negative procoagulant membranes in blood flow, platelet activation, and a decreased level of anticoagulant proteins, amplifying the risk of thrombosis35. In thalassemic patients, the main determinant of the thrombotic risk seems to be related to altered red blood cell membranes as well as to platelet activation, but not to plasma protein abnormalities36. Erythrocyte membrane alterations leads to the exposure of PhSer on the outer layer, cytoskeleton cleavage, membrane budding and release of MPs8. An elevated number of MPs in blood flow has been reported in patients suffering from different vascular diseases1. Consequently, MPs could have an essential causal role in hypercoagulable states, as reported by Van Beers et al in sickle cell disease11. The authors showed that, while the number of platelet-derived MPs was stable, there was an increase in erythrocyte-derived MPs during crisis compared to steady state.
Modification, alteration or even degradation of erythrocyte concentrates (ECs) occurs during storage (a phenomenon called the storage lesions)7,37, and accumulating MPs have been considered as being one of the expression of storage lesions15. Red blood cells’ ageing in blood bank conditions clearly differs from physiological in vivo ageing. Undeniably, conditions to which red blood cells are exposed during storage such as temperature and nature of the medium are dramatically different from in vivo conditions. Erythrocytes degrade and loose efficiency during storage. Indeed, erythrocytes change shape; their membranes become more rigid and there is a disruption of phospholipids asymmetry and release of fragments and MPs37. In addition, during the conservation of erythrocyte in plastic bags, not only physiological alterations occur, but also biochemical changes in the storage medium principally due to red cells metabolism. These modifications, physiological and biochemical are part of the storage lesions. There is an increase in the concentration of lipids, free haemoglobin, potassium, lactate and in contrast a reduction of pH, glucose, 2,3-diphosphoglycerate, sodium or adenosine triphosphate38. For example diminution of glucose and increase of lactate concentration reflects red blood cells glycolysis.
In blood flow, MPs are rapidly removed in the liver by Kupffer cells39. Release of MPs occurs throughout erythrocyte lifespan39 and continues during storage, therefore they accumulate during storage17. Precise reasons of erythrocyte vesiculation are not elucidated, however MPs is a possible mechanism preventing red blood cell removal. MPs may help to clear away the C5-9 complement attack complex, band 3 neoantigen, IgG or other potential harmful agent from the membrane, when the erythrocyte is still viable40–42. In addition, MPs formation could be a mechanism involved to clear out methaemoglobin which is no more functional37.
The precise consequences of storage, including storage lesions and MPs release, on transfusion efficiency are not clearly understood, and relatively little is known about their molecular bases and even less about their possible biomedical consequences on recipients. There is accumulating evidence that receiving “old” blood is not as advantageous as receiving fresher blood. In 2008, Koch et al showed a link between the age of the transfused ECs and post-transfusion complications. They claimed that life expectancy at 5 years was better for patients who received fresher blood (ECs stored for less than 14 days) as compared to older blood (stored for more than 14 days) in the course of cardiac surgery43. A recent study by Spinella et al concluded that ECs stored for more than 28 days was associated with an increased incidence of deep vein thrombosis and death from multi-organ failure44. Other studies on post-transfusion complications or mortality and storage time were also published and controversial45–47. However, although these studies raise important questions about transfusion safety and storage lesions, they are subjected to many discussions48–51.
Due to the fact that transfused people are sick and polytransfused recipients are often in poor condition, it is difficult to attribute clearly an event to transfusion rather than on the clinical situation of the recipient. In a recent review, Zimrin and Hess analyzed several papers on clinical studies examining the effect of storage of red blood cells in transfusion and concluded that well designed epidemiologic studies are needed to demonstrate that a clear clinical effect may be due to ECs storage lesions. Despite the lack of conclusive evidence that storing blood longer is worse, it is current practice to transfuse fresher blood to patient considered at high risk (such as in neonatology for example or cardiac surgery).
As stated before, MPs may have procoagulant activities; therefore, we may reasonably suggest that transfusion of “older” ECs containing a high number of MPs could increase the risk of adverse reactions, by inducing a hypercoagulable state leading to thromboembolic complications. Inversely, in many situations requiring blood transfusion, a hypercoagulable state may be useful to diminish or even helping to stop the bleeding.
Recent advances in proteomics provide new tools allowing studying protein content of MPs in order to elucidate their exact role. Several publications are focused on proteomics of MPs derived from various type of cells (see Table I). Results are different according to the MP cell types and the methodological aspects. To the best of our knowledge, most studies have adopted the same workflow for proteomic analysis of MPs as shown in the “proteomic” part of figure 1 (excepting for the initial sample that is not ECs in all cases). Briefly, once isolated by centrifugation, MPs are directly separated by one-dimensional gel electrophoresis (1D-GE) and finally protein bands are excised, and further digested by a proteolytic enzyme (usually trypsin) and identified by various mass spectrometry techniques. The main advantage of this technique is the compatibility with the analysis of hydrophobic proteins in contrast to two dimensional electrophoresis gel (2D-GE) approaches. Among different proteomic studies on MPs, various lists of MP proteins have been established. Mapping of MP proteins involved in coagulation is still lacking. In the study of platelet MPs proteome by Garcia et al52 when the list of proteins is submitted to a QuickGO search with the GO reference “blood coagulation”, 12 proteins out of 578 were found to be involved in blood coagulation. Recently, Bosman et al53 established a reference proteome of red blood cells MPs. When the protein list was submitted to a QuickGO search with the same parameters as mentioned above, 5 out of 308 proteins were involved in coagulation, namely phospholipid scramblase 1, plasminogen precursor, fibrinogen beta chain precursor, complement component C9 precursor and beta-2-glycoprotein 1. Except for phospholipid sramblase 1, proteins were neither identified in red blood cells membrane proteome established by Bosman et al53 nor by Pasini et al54. As procoagulant proteins of MPs were not found on red blood cells, we could hypothesize that those proteins were either of plasmatic origin with a non-specific binding to MPs or were proteins present on red blood cell membranes, but enriched on MPs microdomains.
The important issue raised by the presence of MPs in blood products is to determine their exact involvement for the recipient. It has been demonstrated that MPs are present in blood products, and that they number increases during storage. Series of articles and reviews were published on the association between red cell storage duration and complications after transfusion. However few of them mentioned MPs as potential cause of these complications. Therefore, there is an urgent need of large clinical studies with the aim to determine precise effects (if any) of MPs in blood recipients. Furthermore, it is important to determine how MPs are involved in the coagulation cascade, by either passively acting by providing additional negatively charged membrane surface, or rather by allowing the expression of TF or other proteins implicated in the coagulation process. Furthermore, studies of MPs from platelets, endothelial cells or monocytes and comparison with EC-derived MPs are needed. Proteomic analysis, as well as quantitative proteomic evaluation of they protein content will open new avenues in haematology, because it will help to decipher the complexity of the hypercoagulable states and the effect of EC transfusion in routine clinical practice. Combining various approaches that are haemostatic investigations, MPs characterisation either by cytometry or by proteomics, clinical studies will certainly help to secure transfusion medicine.
The authors would like to thank the Swiss Red Cross Humanitarian Foundation for the financial support.