E ﬀ ect of Physical Exercise on the Release of Microparticles with Angiogenic Potential

: Cellular communication has a fundamental role in both human physiological and pathological states and various mechanisms are involved in the crosstalk between organs. Among these, microparticles (MPs) have an important involvement. MPs are a subtype of extracellular vesicles produced by a variety of cells following activation or apoptosis. They are normally present in physiological conditions, but their concentration varies in pathological states such as cardiovascular disease, diabetes mellitus, or cancer. Acute and chronic physical exercise are able to modify MPs amounts as well. Among various actions, exercise-responsive MPs a ﬀ ect angiogenesis, the process through which new blood vessels grow from pre-existing vessels. Usually, the neo vascular growth has functional role; but an aberrant neovascularization accompanies several oncogenic, ischemic, or inﬂammatory diseases. In addition, angiogenesis is one of the key adaptations to physical exercise and training. In the present review, we report evidence regarding the e ﬀ ect of various typologies of exercise on circulating MPs that are able to a ﬀ ect angiogenesis.


Introduction
The capacity to communicate between cells, tissues, and organs is a fundamental requisite to ensure appropriate physiological functions [1]. One of the most important mechanisms governed by cell communication and signaling is represented by the formation of vasculature. It is well known that vascular formation and growth are processes strictly regulated by growth factors and bioactive molecules. Hypoxia represents an important stimulus for angiogenesis: it activates hypoxia-inducible factors (HIFs) that upregulate various angiogenic genes, including vascular endothelial growth factor (VEGF) [2]. On the other hand, other different factors (e.g., thrombospondin-1, statins) are involved in the inhibition of angiogenesis [3].
Neovascular growth can be a physiological (e.g., in injured tissues [4]) or a pathological process, as it occurs in diabetic retinopathy, maculopathy, in tumors, and in many other disorders. Physiological angiogenesis represents the results of an accurate balance between pro-angiogenic and anti-angiogenic signals [5]. In pathological conditions, this balance is compromised and the new vessels do not provide a functional vascular perfusion [6].
Small extracellular vesicles, named microparticles (MPs), can be involved in angiogenic processes [7]. MPs are defined as heterogeneous small membranous vesicular structures derived from different cell types [8]. They are lacking nucleus and ranging in size from 0.1 to 1 µm. The most

Extracellular Vesicles Classification
Human cells are able to release various types of membrane vesicles in response to different stimuli, generally called extracellular vesicles (EVs). EVs include exosomes, MPs (also called microvesicles), and apoptotic bodies [28] (Table 1). Exosomes are vesicles surrounded by a double layer of phospholipids and their size ranges between 50 and 100 nm in diameter. They are released in both physiological condition and upon activation by exocytosis of multivesicular bodies (MVBs) [29]. Exosomes can communicate with other cells either through a direct contact between surface proteins or by vesicles-cell membrane fusion or by endocytosis. They expose phosphatidylserine (PS) on the external membrane and various markers such as LAMP1, TSG101, CD63, CD81, and CD9 [30]. The main functions of exosomes include antigen presentation and immunostimulatory activity. In addition, exosomes are able to horizontally transfer mRNA and miRNA as well as oncogenic receptors [31].
MPs, formerly described as "platelet dust", are now well known cell-cell communicators and they act as signaling molecules in physiological homeostatic processes or as a consequence of pathological condition including diabetes, chronic kidney disease, rheumatoid arthritis, multiple sclerosis, and cardiovascular diseases [32]. MPs are found in most body fluids such as plasma, saliva, urine, and cerebrospinal fluid. They represent a heterogeneous population of vesicular-like structures composed by a phospholipid bilayer and their sizes range from 100 nm to 1 µm in diameter; however precise cut-off values remain to be established. MPs found in bloodstream may originate from various vascular cells (e.g., platelets, monocytes, erythrocytes, granulocytes) [33]. However, among all circulating MPs, platelet-derived MPs (Plt-MPs) are the most abundant population, representing 70-90% of the total [34]. They are produced by budding of the plasma membrane, in a process called ectocytosis. After their release, MPs carry proteins that are signature of their cell of origin, and transport various enzymes, RNA, and miRNA [35]. MPs generally expose PS on their outer leaflet (Annexin V + ), however some evidence shows that MPs can also be PS negative [36]. In addition, a subpopulation of various phenotypes of MPs expose tissue factor [37]. The release of MPs is induced upon cell activation or apoptosis.
Apoptotic bodies are EVs larger than MPs and exosomes. Their sizes range from 500 nm to 3 µm and they expose PS; however, their outer membrane differs from MPs due to its permeability [38]. Apoptotic bodies are released by membrane blebbing of cells undergoing apoptosis and they may horizontally transfer their content, such as cell organelles, fragmented DNA, and oncogenes.

Methods of Detection of Microparticles
Flow cytometry (FC) is one of the most common methods of detection, quantification, and size evaluation of MPs [39]. In addition, FC allows the phenotypic characterization of MPs using fluorescent antibodies against the diverse antigens expressed by the cell of origin [40]. However, FC has some limitations. For example, MPs with a size ranging from 0.1 to 0.4 µm are too undetected by most cytometers [41]. In addition, the type of instrument, settings, and sample preparation procedures can cause a high level of variability of the results [42].
The nanoparticle tracking analysis (NTA), on the other hand, is able to detect vesicles smaller than those recognized by FC. Such type of analysis correlates the speed of movements, tracked using a microscope, with the size of MPs [43]. Nevertheless, NTA is unable to discriminate vesicles derived from cells from other particles of small dimensions such as high-density lipoprotein (HDL) [44]. Fluorescent nanoparticle tracking analysis (F-NTA), diversely from NTA, is based on the fluorescence of the particles [45].
Other techniques to analyze the EVs are based on image analyses obtained by optical and electron microscopy. Optical microscopy permits to assess the size and morphological features of MPs and of apoptotic bodies, while cannot provide information on their phenotype. This kind of information can, anyway, be obtained by fluorescent microscopy labeling MPs with fluorescent probes that recognize specific surface markers [41]. However, this approach does not permit to specify the size of the EVs [41]. An accurate morphological analysis of MPs is obtained by transmission electron microscopy (TEM), that represents the most used method to assess the composition, morphology, size, and membrane structure of MPs [34]. However, TEM requires a considerable amount of time for both preparation and analysis of the sample.
Western blotting (WB) is a commonly used technique that allows to identify the cellular origin of MPs [46]; anyway, this methodological approach presents some important limitations: indeed, it requires a large amount of MPs to be performed and the analysis needs to be completed with complementary methods such as NTA or TEM [47,48].
Finally, another method for MPs detection is the enzyme-linked immunosorbent assay (ELISA). This is also a quantitative analysis, that measures MPs indirectly: indeed, such method relies on the binding of MPs to specific conjugate-antibodies adsorbed to a well plate [49]. ELISA permits the specification of MPs subtypes, but fresh plasma is indispensable, because the freezing-thawing process leads to an increase of annexin-binding MPs [50].

Production Mechanisms of MPs
It is well known that MPs can be released upon activation or apoptotic stimuli; however, most cells constitutively shed MPs by ectocytosis and significant concentrations of these vesicles can be detected in the plasma [57]. The common mechanism for the production of MPs is represented by a change in the asymmetry of the phospholipids composing the plasma membrane. The lipid bilayer of normal cells presents an asymmetrical composition of phospholipids [58]. PS and phosphatidylethanolamine (PE) are located in the internal layer of the plasma membrane whereas the external layer is composed of phosphatidylinositol (PI), phosphatidylcoline (PC), sphingomyelin (SM), and other glycolipids. In order to allow MPs formation, this precise distribution is disrupted. Such changes in membrane architecture are controlled by three enzymes: flippase, floppase, and scramblase [59]. Flippase is involved in the maintenance of the phospholipid asymmetry between the membrane layers. Floppase controls the ATP-dependent translocation of PS to the external layer in response to cellular activation. Finally, scramblase is responsible for the membrane randomization allowing phospholipids to flow down their concentration gradients. In resting cells, only flippase is active, contributing to the internalization of negatively charged phospholipids (e.g., PS and PE) and to the maintenance of the physiological membrane asymmetry. Nevertheless, various events providing cellular activation are able to increase intracellular calcium concentration, causing the flippase inhibition and the activation of floppase, scramblase, and other calcium-sensitive enzymes such as calpain and gelsolin, resulting in MPs release [60].
An important mediator of apoptotic MPs production is Rho-associated kinase (ROCK I), a protein kinase activated via caspase-mediated cleavage. ROCK I may allow the phosphorylation of myosin light-chains, triggering cytoskeletal rearrangement and formation of apoptotic MPs through the coupling of actin-myosin filaments to the plasma membrane [61].

MPs and Angiogenesis
Vesicle-like structures have recently emerged as important mediators for cellular communication in both physiological and pathological processes [30]; in particular, MPs released by various cell types may play an important role in angiogenesiss [62,63], by modulating the production of proangiogenic factors through changes in the secretome of endothelial cells [64,65]. Like exosomes, MPs can communicate with other cells through ligands on their membrane that interact with specific receptors of the target cells; alternatively, MPs can fuse with the plasma membrane of the target cells, delivering its cytoplasmic content (e.g., proteins, miRNAs, RNAs). In this way, MPs may elicit the synthesis of proangiogenic or antiangiogenic factors and affect the main processes involved in formation of new vessels, such as endothelial cells adhesion, migration, and proliferation [66].
Endothelial cells-derived MPs (E-MPs) are clearly involved in the formation of blood vessels. A pioneer in vitro study highlighted that vesicles released by endothelial cells can transport integrin-β1 and active matrix metalloproteinases-2 and 9 (MMP-2 and MMP-9), thus inducing endothelial cells invasion and tube-like structure [67]. In addition, it has been demonstrated that upon stimulation with the proinflammatory interleukin-3, endothelial cells release vesicles containing pro-angiogenic factors such as miR-126-3p and Stat5 [68]. Moreover, another study shows that endothelial MPs are able to transfer miR-126 from endothelial cells to vascular smooth muscle cells, corroborating the angiogenic properties of endothelial MPs [69].
Additionally, Plt-MPs are involved in angiogenesis. These vesicles can promote angiogenic effects by transferring various cytokines, such as PDGF, VEGF, FGF-2, and activating key proteins such as PI3k, ERK, and src kinase [70]. Various studies found that the angiogenic potential of the platelet-derived MPs can have both functional positive and negative effects: when Plt-MPs were infused in a rat model of myocardial ischemia, indeed, they increased the quantity of functional capillaries [70]; on the other hand, MPs shed by platelets can favor angiogenesis involved in tumor progression, upregulating important angiogenic factors, such as VEGF, HGF, and interleukin-8 [71].

Angiogenesis Induced by Physical Exercise
Physical activity has beneficial effects on human health [72][73][74][75], and regular physical exercise improves health in both physiological and pathological conditions [76][77][78][79]. Numerous evidence demonstrates the inverse relationship between regular exercise and cardiovascular diseases, hypertension, stroke, type 2 diabetes mellitus, metabolic syndrome, obesity, osteoporosis, and various types of cancer [76,[80][81][82][83]. Depending on the type of physical exercise, we can observe different effects on the organism: endurance (or aerobic) exercise is characterized by a high number of muscular contractions, relative low load, and an important involvement of the cardiorespiratory system, whereas resistance exercise is represented by low number of muscular contraction but with relative higher loads and thus it relies mainly on the neuromuscular system [84]. These two types of exercise can be considered two extremes in a line, and between these two, numerous types of exercise exist, based on the modulation of the exercise variables (e.g., intensity, volume, density, rest). Each type of exercise can increase different responses and adaptions, triggering diverse cellular pathways.
Cardiovascular endurance training is the strongest stimulus for exercise-induced angiogenesis; this results in an improved capillary to fiber ratio in the skeletal muscle to provide an enhanced blood flow and oxygen delivery to the exercising muscles [85][86][87][88][89]. Endurance exercise stimulates the angiogenesis activating several pathways ( Figure 1). It induces an important reduction of the partial pressure of oxygen that leads to a local tissue hypoxia. This modification in oxygen levels decreases the hydroxylation of HIF-1alpha trough the inhibition of prolyl hydroxylase domain enzymes (PHD); in this way, HIF-1alpha become more stable and translocate into the nucleus, where forms an active complex with HIF-1beta [90] and activates genes involved in angiogenesis such as VEGF [84].
Another key regulator of VEGF synthesis is the peroxisome proliferator-activated receptor gamma co-activator 1-alpha (PGC-1alpha), which is modulated by aerobic exercise in several ways: intense endurance exercise produces an energy deficit and an increase of the AMP/ATP ratio that activate the AMP-activated protein kinase (AMPK) responsible for the PGC-1alpha phosphorylation. Moreover, an acute bout of exercise increases the NAD + /NADH ratio that promotes the sirtuin (specifically SIRT1) activity and PGC-1alpha deacetylation [91]. Finally, aerobic exercise affects also the estrogen-related receptor gamma (ERR gamma) that controls the skeletal muscle vascularization by modifying the VEGF expression [92]. In the last years it was evidenced that physical exercise induces modification of the MPs release into the bloodstream [93]. These EVs carry various molecules important for the tissues' crosstalk during and post exercise [94,95] and can be also a cargo of soluble factors involved in endothelial cell proliferation, migration, and adhesion. This results in the generation of new vessels in response to exercise.

MPs and Physical Exercise
Physical exercise, particularly endurance exercise, is known to stimulate shear stress due to its hemodynamic effects [96]. Intraluminal shear stress has a beneficial effect on vasculature because it is able to ameliorate vascular dilation through mechanisms mediated by nitric oxide [97]. In addition, exercise-induced shear stress can modulate the production of both E-MPs and Plt-MPs. For example, it was seen that high shear stress can reduce the concentration of E-MP while low shear stress is prone to increase such concentration [98]. Moreover, both high and low shear stress can elicit the production of MPs from platelets [99] (Figure 2). Acute exercise leads to an augmentation of the sympathetic activity, increasing both epinephrine and norepinephrine concentrations [100]. In this regard, studies report that norepinephrine is able to stimulate Plt-MPs release [101]. Along with hormonal response, acute exercise stimulates the secretion of many cytokines from skeletal muscle, named myokines [102]. Among these myokines, IL-6 is released in the bloodstream and elicits multiorgan effects [103]. It has been reported that IL-6 is also able to stimulate Plt-MPs formation [104], thus exercise can be a potent stimulus to enhance such mechanism.
The following sections provide an overview of various studies (summarized in Table 2) investigating the effects of physical exercise and training on MPs concentration, with an emphasis on endothelial and platelet-derived MPs, due to their specific involvement in angiogenesis [17,[105][106][107].

The Effect of Physical Exercise on MPs with Angiogenic Potential in Athletes
Endurance athletes have the characteristics to present greater adaptation of the cardiorespiratory and neuromuscular systems resulting in the increase of the delivery and the consumption of oxygen [118]. Two of the sports that highly rely on these adaptations are marathon and half-marathon. Bittencourt et al. compared the amounts of E-MPs and Plt-MPs in a group of half-marathon runners with a control group of gender-matched healthy controls showing that CD51 + E-MPs and CD42 + /CD31 + Plt-MPs did not significantly differ between the two groups [52]. This data suggests that athletes exposed to chronic endurance training can benefit from exercise effects but that such type of training does not lead to further decrease in MPs compared with a healthy non-athlete population. On the other hand, E-MPs and Plt-MPs significantly increase following a marathon run and return to baseline values within 2 days; conversely, leukocyte and monocyte-derived MPs decrease after such run [114].
Like long distance running, road cycling and triathlon are two sports heavily relying on cardiorespiratory endurance [119]. In triathletes/cyclists, high-volume (2 h at 55% of peak power output) and high intensity (four sets of minutes at 90-95% of peak power output interspersed by 3 min of active recovery or four sets of 30 s at maximal effort separated by 7:30 of active recovery) trials on cycle ergometer, significantly decrease CD31 + /CD42b − apoptotic E-MPs 60 and 180 min after the termination of the trials [117]. It seems that such decrease was due to an increased uptake of MPs by cells. Indeed, triathletes/cyclists' serum collected 180 min after high intensities exercises promote the uptake of CD31 + /CD42b − apoptotic E-MPs in endothelial cells in vitro [117]. Another recent study involving male athletes showed that an incremental cycling protocol until exhaustion determined a significant increase of CD105 + , CD146 + E-MPs, and CD62P + Plt-MPs at the peak of the exercise session [108].

The Effect of Physical Exercise on MPs with Angiogenic Potential in Healthy Subjects
The physical exercise performed in a regular manner elicits beneficial effects that include improved cardiorespiratory fitness, body composition, decreased risk of noncommunicable diseases and of all-cause mortality [76,[120][121][122][123][124].
Data on the effects of aerobic exercise on E-MPs and Plt-MPs are controversial. After 1 h of treadmill run at 70% of VO 2max , a decrease in TF + Plt-MPs and neutrophil-derived MPs in healthy men [110] has been observed; moreover, a single session of 30 min of moderate intensity aerobic exercise (60-64% of their VO 2peak ) reduces the number of CD62E + activated E-MPs and CD31 + /CD42b − apoptotic E-MPs [115]. On the other hand, in prehypertensive male and female, aerobic exercise performed at 65% of predicted maximal heart rate for a duration of 6 months, with a frequency of three times per week, decreases the blood concentration of CD62E + E-MPs and CD31 + /CD42a − E-MPs [111]. Some authors also evidenced a possible sex/gender effect on the MPs release: indeed, an acute bout of aerobic exercise at 60-70% of the VO 2peak produced a significant increase in CD62E + E-MPs in men, and an increase in CD34 + MPs in women [51]. However, results of study involving both male and female are inconclusive. Indeed, another study showed CD62E + E-MPs decrease both during and after moderate intensity continuous exercise (65% of VO 2max ) in women but not in men [116].
One of the factors that must be taken into account analyzing the effects of physical exercise on the MPs is the timing of the analysis. In response to low to moderate intensity (33% of VO 2max ) concentric or eccentric cycling, men show higher concentration of CD41 + Plt-MPs 5 min following exercise, but these levels returned to baseline levels after 40 min of recovery, while CD62E + E-MPs do not show modifications [56]. Another experimental study performed in young males showed that CD41 + Plt-MPs and CD62E + E-MPs were unaffected by 1 h of cycling at ≈46% of VO 2max , while when the intensity were set ≈67% of VO 2max an increase of CD41 + Plt-MPs was observed [17]. Interestingly, the MPs released in response to such type of exercise enhanced endothelial proliferation, migration, and tubule-like formation in vitro, compared with MPs present in resting conditions [17].
Another modality of training that is of increasing interest in research is high intensity interval training (HIIT). HIIT is a useful exercise modality to enhance cardiorespiratory endurance, and such type of training can be performed also at strenuous intensities [125,126]. Supramaximal sprint cycling protocol consisting of 10 bouts of 15 s at 120% of peak power output separated by 45 s of active recovery lead to a significant increase of circulating CD105 + and CD106 + E-MPs 90 min after the exercise session, whereas at 180 min post-exercise MPs returned towards resting values [112]. On the other hand, no changes in endothelial MPs concentration was observed after a high intensity interval exercise (95% of VO 2max ) [116].
Physical inactivity and sedentary behaviors have detrimental effects on vasculature state, contributing to the development of cardiovascular and metabolic diseases [127][128][129]. After 5 days of reduced physical activity (<5000 steps/day), significant increase of CD31 + /CD42b − apoptotic E-MPs was registered, whereas CD62E + activated E-MPs remained unchanged, suggesting that inactivity fosters vascular dysfunction increasing endothelial apoptosis in recreationally active men [53].

The Effect of Physical Exercise on MPs with Angiogenic Potential in Pathologic Subjects
Physical exercise is a fundamental stimulus for improving metabolic and cardiovascular health. Among the various typology, HIIT has the same, or superior capacity to improve health status in pathological subjects (e.g., presenting diabetes, cardiovascular diseases, or cancer) [130,131]. In overweight/obese men, high-intensity continuous exercise (20 min of cycling at just above the subjective ventilatory threshold) and high-intensity interval exercise (10 bouts of 1 min at ≈90% peak aerobic power) reduce E-MPs concentration 18 h following exercise; different results were observed in overweight/obese women in which high-intensity continuous exercise increased activated CD62E + E-MPs, leaving unaffected CD31 + /CD42b − apoptotic E-MPs, regardless the intensity [15]. However, the chronic exposure of overweight/obese subjects to exercise (2 weeks of HIIT or moderate intensity continuous training) lowered the concentration of activated CD62E + E-MPs [54]. The effect of exercise on MPs release may also depend on the intensity of the physical activity. Indeed, in a recent study involving both over and normal weight subjects, a bout of moderate intensity (60% of their VO 2max ) aerobic exercise, reduced the total EVs in the MPs range (130-700 nm); however, the release of MPs in normal weight subjects was higher than in obese and in females than in males. [113].
Polycystic ovary syndrome (PCOS) is an endocrine disorder in which endothelial dysfunction and altered angiogenesis are common. Eight weeks of aerobic training consisting of three sessions per week of 1 h of running on a treadmill at 60% VO 2max led to a significant decrease in CD105 + MPs in women with PCOS compared with a control group of healthy women, while did not have an effect on the amount of CD106 + E-MPs [55].
Coronary artery diseases determine a diminished myocardial perfusion that may result in angina, myocardial infarction, and heart failure [132]. In stable coronary heart disease patients, neither moderate-intensity continuous exercise (28.7 min at 70% of peak power output) nor a single bout of high-intensity interval exercise (10 min session with work interval of 15 s at 100% of peak power output interspersed with 15 s of passive recovery) have an effect on the concentration of circulating E-MPs and Plt-MPs [109].
Overall, the numerous studies show contrasting results regarding the qualitative and quantitative changes of MPs in response to acute physical exercise and to exercise training as well. Such divergent conclusions could be due to the heterogeneity of the studies, as they involved different samples in terms of age, sex, training status, possible pharmacological therapies [133], and pathologies [93] ( Figure 3). More standardized and harmonized procedures are needed to compare the studies and obtain definitive data on the release of MPs with angiogenic potential in response to physical exercise.

Conclusions
Various modalities of cardiovascular endurance-based exercise can modify blood concentration of E-MPs and Plt-MPs, known to affect angiogenic processes. It is well known that physical exercise, among different adaptations, can lead to a functional angiogenesis resulting in a wider vascular bed that increases the blood flow to exercising muscles. Results regarding the effect of exercise on MPs are divergent, but it seems that acute exercise results in an increase of MPs while chronic exposure to exercise results in a decrease in most cases. In addition, MPs sensitivity to exercise can be affected by the training status, sex, age, as well as mode of exercise (e.g., run on a treadmill, road run, cycle ergometer), pharmacological therapies, diseases, and method of detection. While only one study directly showed the pro-angiogenic effects of MPs produced following exercise [17], numerous studies highlight the potential of E-MPs and Plt-MPs in communicating with other cells and stimulating angiogenesis [69,[105][106][107]. Thus, future evidence regarding the effect of both acute exercise and training on MPs affecting angiogenesis will be useful to add important insights on the mechanisms by which exercise elicits vascular adaptations and to design specific training programs for various populations.

Conflicts of Interest:
The authors declare no conflict of interest.