If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Inserm U1297 and Université Paul Sabatier, Institut des Maladies Métaboliques et Cardiovasculaires, Toulouse, FranceLaboratoire d'Hématologie, Centre de Référence des Pathologies Plaquettaires, Centre Hospitalier Universitaire, Toulouse, France
Corresponding author at: Inserm U1297 and Université Paul Sabatier, Institut des Maladies Métaboliques et Cardiovasculaires, 1 Avenue Jean Poulhes BP 84225, 31432 Toulouse Cedex 4, France.
Obesity is a major health issue that can increase the risk of cardiovascular morbidity and mortality. Obesity is characterized by an elevated prothrombotic state in which dysregulated platelet activation is a major component due to accumulation of adipose depots, insulin resistance, endothelial dysfunction, low-grade chronic inflammation and oxidative stress. Obesity also alters the homeostasis of the bone marrow, the site where megakaryopoiesis and platelet production take place, which include bone marrow adipose tissue as an important constituent of this microenvironment. Here, we provide an up-to-date review focusing on bone marrow adipose tissue and the influence of obesity on megakaryocytes and platelets that increases the cardiovascular risk.
Platelets are known to play a major role in hemostasis and thrombosis. They are rapidly recruited at the site of vascular injury to prevent bleeding and are now recognized as key elements in many pathophysiological processes. Platelets originate from megakaryocytes (MKs) that arise from a complex process of hematopoietic stem cell (HSC) maturation in the bone marrow (BM). Any dysfunctions in the finely-regulated process of platelet production and activation could have side effects in humans. This is the case for the metabolic perturbations of obesity, in which BM homeostasis is altered by, for instance, an amplification of bone marrow adipose tissue (BMAT) and platelets are prone to hyperactivation due to multifactorial perturbations such as accumulation of adipose depots, insulin resistance, endothelial dysfunction, low-grade chronic inflammation and oxidative stress [
]. Obesity is thus currently considered an important risk factor for cardiovascular morbidity and mortality due to platelet dysfunctions. Herein, we summarize the most recent and important findings regarding the impact of obesity on megakaryopoiesis and platelet production within the BM microenvironment and on platelet activation; all these increase the thrombotic risk in obesity.
2. Megakaryopoiesis, thrombopoiesis and platelet activation
2.1 Primary hemostasis involves the multistep activation of platelets
Platelets are small enucleated discoid cells (ranging from 1 to 3 μm) and the second-most abundant cell lineage in the circulation (150,000 to 400,000 platelets per μL of blood in healthy individuals) [
]. The abundance of platelets in circulation allows for constant surveillance of the endothelium and facilitates their role in primary hemostasis, as was described in the 19th century [
]. Since the discovery of platelet function in hemostasis, numerous studies have described the complex mechanisms underlying platelet activation. In intact blood vasculature, the endothelium constitutes a non-thrombogenic surface and releases prostaglandin I2 (PGI2) and nitric oxide (NO) to maintain circulating platelets in a resting state [
]. Following vascular injury, the wounded vessel wall leads to the initial and transient adhesion (rolling) of platelets, via the binding of the glycoprotein (GP) Ib-V-IX on platelets to the von Willebrand Factor (vWF) stabilized on collagen fibers, followed by platelet firm adhesion to collagen fibers through two collagen-specific platelet receptors: GPVI and GPIaIIa (also called integrin α2β1) [
]. When adhering, activated platelets release autocrine soluble mediators, by secreting the content of their granules or by generating thromboxane (TX) A2, to amplify the initial platelet responses, and recruit circulating platelets [
]. Platelet activation is reinforced by thrombin, which is locally produced by the coagulation cascade and binds to platelets through protease-activated receptors (PARs). The functional upregulation of GPIIb-IIIa, also called αIIbβ3 integrin, on activated platelets allows thrombi growth by fibrinogen binding to adjacent platelets. Platelet thrombi by sealing the vessel lesion arrest the hemorrhage [
2.2 Platelets originate from MKs after a multistep maturation process
In humans, approximately 1.1011 platelets are formed every day by MKs following a finely regulated and specific maturation process called megakaryopoiesis that occurs in the BM. MKs, with a size ranging from 15 to 100 μm, are rarest cells in the BM (0.01 % of nucleated cells among BM-resident cells) [
]. Although BM is considered the major site of platelet production, MKs found in the lungs participate in platelet production, as was shown by two-photon intravital microscopy in mice [
]. Using the same technique, a recent study showed that in response to an inflammatory trigger such as sepsis, platelets can be produced from MKs localized in the spleen [
]. These recent findings unveil a newly comprehended complexity of the site of platelet production that can impact the future function of platelets. Beyond that, recent single-cell RNA sequencing analysis in human and mouse also showed that within the BM, MKs can be divided into three subpopulations due to transcriptional heterogeneity. The first subpopulation is specifically associated with thrombopoiesis functions (i.e., it is enriched for genes implicated in platelet production and functions), with a physical proximity to sinusoidal blood vessels. The second subpopulation expresses various inflammatory markers, such as CD40L, and appears to be involved in innate and adaptive immune functions. The third subpopulation is thought to be implicated in osteoblast and HSC niche maintenance [
]. This recent advance in the knowledge of MK biology allows to decipher new functions for MKs beyond its first characterized role in platelet production.
To allow platelet release, HSCs undergo a complex maturation process transforming them into mature MKs. This maturation process is regulated at multiple levels by different cytokines, the most important of which is thrombopoietin (TPO) that binds to its specific receptor, c-Mpl [
]. During megakaryopoiesis, HSCs undergo successive endomitosis (DNA replication without cell division), resulting in a polyploid nucleus (16 N on average and 128 N maximum). In parallel, the cells amplify the number of organelles such as mitochondria and future platelet granules and create an invaginated membrane system (also known as the demarcation membrane system, DMS) that is continuous with the plasma membrane and provides a membrane reservoir for platelet production [
]. Once mature, MKs produce long and thin cytoplasmic extensions from the DMS, called proplatelets, which cross the endothelial cell barrier to release platelets into the circulation through a combination of cytoskeletal mechanical tensions and shear forces present in the blood vessel lumen [
]. A recent study challenges the classic schema of proplatelet-dependent platelet production and describes MK membrane budding as a direct source of platelets for daily chronic needs [
]. In response to acute platelet needs, another mechanism of platelet production is the MK rupture-type thrombopoiesis, which appears to be interleukin (IL)-1α-dependent and allows a twenty times higher platelet release compared to the proplatelet-type thrombopoietic process [
]. It has been recently reported in humans and mouse models that microparticles originating from circulating platelets (PMPs) directly influence MKs and their progenitors. Under conditions associated with thromboinflammation, these PMPs can support megakaryopoiesis and platelet production [
]. Any dysregulations in the maturation process of MKs can directly impact the quantity and/or the quality of platelets present in circulation. This is for example the case for obesity where metabolic/inflammatory dysfunctions and modification of the BM microenvironment, such as the accumulation of BMAT, dysregulates megakaryopoiesis and platelet production [
In the BM cavity, hematopoietic cells, marrow stromal cells, nerves, blood vessels and adipocytes reside in close contact with bone. BM adipocytes have long been considered as silent bystanders that fill “empty space” in the BM when bone mass is low or when hematopoiesis is impaired. Recent studies on the functional role of BM adipocytes reveal that this adipose tissue, called BMAT, is a secretory and metabolically active organ distinct from extramedullary adipose depots [
3.1 BM adipocytes originate from mesenchymal stromal cells (MSCs) and present characteristics that are both common to and distinct from white and brown adipocytes
BM adipocytes are translucent, yellow elliptical cells with a large central lipid globule with smaller fat globules in the peripheral cytoplasm and are smaller than extramedullary adipocytes, with a diameter between 30 and 60 μm depending on the physiopathological context [
]. Tavassoli first characterized the morphologic features of BM adipocytes in 1976 by histochemical staining of rat bones using performic acid-Schiff (PFAS) and distinguished two populations of fat cells within the bone cavity. One population was distributed throughout the hematopoietic red marrow and was positively stained by PFAS, and the other was packed in the yellow marrow and was not stained by PFAS [
]. Forty years later, using osmium tetroxide staining and microcomputed tomography (μCT), Scheller and colleagues confirmed the presence of two types of adipocytes within the mouse tibia cavity: constitutive adipocytes and regulated adipocytes [
]. Constitutive adipocytes develop shortly after birth in the distal tibia and caudal vertebrae. They are compact and large in size (38 to 39 μm diameter), have little influence on hematopoiesis, contain more unsaturated than saturated lipids, have elevated levels of the adipogenic transcription factors CCAAT/enhancer binding protein α and β (Cebpa and Cebpb), and remain preserved upon systemic challenges. Regulated adipocytes develop with age within the red marrow, and have widespread localization and an active influence on hematopoiesis. They are smaller in size (31 to 33 μm diameter), contain more saturated lipids, express lower levels of Cebpa and Cebpb and are modulated by systemic changes (Fig. 1). Because of its bone location, it is complicated to structurally characterize BMAT compared to other adipose depots. Historically, bone biopsies at the iliac crest have allowed analysis of the marrow space where adipocytes are localized. More recently, noninvasive imaging techniques have been developed, such as μCT (single- or dual-energy) and magnetic resonance imaging (MRI) (conventional T1-weighted, water-fat or proton magnetic resonance spectroscopy (1H-MRS)). MRI is currently the best method to distinguish yellow fat marrow from hematopoietic red marrow, and 1H-MRS is considered the gold standard technique for both quantitative and qualitative characterization of BMAT in vivo [
Reporting guidelines, review of methodological standards, and challenges toward harmonization in bone marrow adiposity research. report of the methodologies working group of the international bone marrow adiposity society.
Standardised nomenclature, abbreviations, and units for the study of bone marrow adiposity: report of the nomenclature working Group of the International Bone Marrow Adiposity Society.
Fig. 1Bone marrow adipose tissue (BMAT). Schematic representation of BMAT origin, characteristics, regulation and functions. Lineage tracing studies demonstrate that BM adipocytes originate from mesenchymal stromal cells (MSCs) after a hierarchical process of differentiation. BM adipocytes can be classified into two distinct subtypes: regulated BMAT and constitutive BMAT, with different characteristics. In addition to physiological BMAT formation, various conditions are associated with BMAT loss or BMAT expansion, predominantly in regulated BMAT. BMAT can release adipokines such as adiponectin and leptin, chemokines/cytokines and lipids to exert both local and systemic effects on health and diseases such as bone homeostasis and hematopoiesis. IL: interleukin; TNFα: tumor necrosis factor α; G-CSF: granulocyte colony-stimulating factor; MIP1: macrophage inflammatory protein; GM-CSF: granulocyte macrophage colony-stimulating factor; RANKL: receptor activator of nuclear factor kappa-B ligand; CXCL: chemokine (C-X-C motif) ligand; FABP4: fatty acid-binding protein 4; Dpp4: dipeptidyl peptidase-4, Pref1: preadipocyte factor 1; Scf: stem cell factor; PPAR: peroxisome proliferator-activated receptor; C/EBP: CCAAT/enhancer binding protein.
BM adipocytes originate from non-hematopoietic multipotent MSCs, which are common progenitors for adipogenic, osteoblastic and chondrocytic lineages and reside at the endosteal site of the BM space [
]. The differentiation process of BM adipocytes is finely tuned by transcription factors (such as peroxisome proliferator-activated receptor (PPAR) γ and CCAAT-enhancer-binding protein (c/EBP) α) and soluble mediators (such as adiponectin and leptin) [
] (Fig. 1). Although the exact nature of BM adipocyte progenitor cells is still being investigated, it is now clear that BM adipocytes arise from a different lineage than extramedullary white and brown adipocytes [
]. Lineage tracing analysis in mice (by fluorescence-activated cell sorting or fluorescence imaging using lineage-specific mT/mG reporter mice) has been key to understanding the origin of BM adipocytes. Ambrosi and colleagues showed that MSCs follow a developmental hierarchy program of differentiation into BM adipocytes, starting from a multipotent perivascular osteogenic/adipocyte stem cell-like cell (CD45−CD31−Sca1+PdgfRα+CD24+) that unilaterally commits to either an adipogenic or osteoblastic lineage. An adipogenic fate-committed progenitor cell (CD45−CD31−Sca1+PdgfRα+CD24−) then transforms into a mature CD45−CD31−Sca1−Zfp423+ preadipocyte precursor cell [
]. The BM adipocyte lineage does not express the cell surface marker profile of white adipocyte progenitors (Lin−; CD29+; CD34+; Sca1+; PdgfRα+; with or without CD24) [
The medullary adipocyte phenotype combines some features of both white and brown adipocytes but is molecularly and functionally distinct from both types of adipocytes. The yellow color of BMAT is distinct from that of white or brown adipose tissue, especially in humans, probably because of differences in the number of mitochondria in each adipocyte type [
]. Transcriptomic analysis of BM adipose tissue from rabbit tibia and radius and C57BL/6 mouse tibia showed that BM adipocytes express some markers of brown adipose tissue (such as Prdm16, deiodinase 2 (Dio2), PGC1α, Sfrp2, Ibsp and Pth1r) but do not express the brown adipose tissue-specific marker uncoupling protein (UCP) 1 [
]. Additionally, BM adipocytes isolated from hip surgery patients show a different gene expression profile compared to subcutaneous white adipocytes collected from abdominal surgery patients, especially in terms of lipid metabolism regulation, HSC development and white-to-brown differentiation pathways. BM adipocytes seem to be more closely related to MSCs than to white adipocytes and play an active role in BM niche homeostasis [
]. While cytokine/chemokine gene expression is upregulated in BM adipocytes compared to white adipocytes, these two types of adipocytes secrete IL-6, IL-1β, tumor necrosis factor (TNFα), macrophage inflammatory protein (MIP) 1, granulocyte colony-stimulating factor (G-CSF), granulocyte macrophage colony-stimulating factor (GM-CSF) and C-X-C motif chemokine ligand (CXCL) 1 and 2 [
]. Transcriptomic and functional studies of human BM adipocytes also have shown no activation by cold temperatures unlike the brown adipose tissue, and a different glucose metabolism and reduced insulin responsiveness compared with white adipose tissue [
]. Proteomic and lipidomic analysis of BM adipocytes isolated from hip surgery patients revealed that compared to subcutaneous white adipocytes, BM adipocytes display a cholesterol-orientated lipid metabolic profile. They have a higher cholesterol content and upregulated expression levels of proteins involved in cholesterol transport, hydrolysis and synthesis, but a reduced fatty acid metabolism and lipolytic activity with an increased triglyceride and monoglyceride levels and a decreased expression of lipases involved in triglyceride hydrolysis, particularly monoglyceride lipase (MGLL), lipid droplet protein (PLIN1) and fatty acid binding protein 4 (FABP4) [
]. Thus, the main adipocyte function, liberating lipid reserves, appears to be less important for BM adipocytes than white adipocytes.
3.2 BM adipocyte homeostasis is regulated by age, sex and metabolic conditions
The development of BMAT is dependent on age, sex and metabolism. In humans, BMAT appears in the bone cavities shortly after birth and expands with skeletal growth throughout childhood and adolescence in a distal-to-proximal manner. In healthy adults, BMAT represents 70 % of the BM volume and accounts for approximately 5 % to 10 % of the total fat mass. It is found in trabecular bones such as long bones (tibia and femur), vertebrae, sternum and ribs. Throughout adulthood, BM conversion into BMAT progresses slowly with aging and often correlates with bone loss. In women, BMAT increases dramatically between the ages of 55 and 65, whereas in males, it increases gradually throughout life. Women have 10 % less BMAT than age-matched men, a phenomenon that reverses after menopause, as estradiol is a key player in BMAT endocrine regulation [
]. This adipose tissue shows a similar development pattern and distribution in rodents, predominantly located in the arms and legs and spreaded across the spine and central skeleton. However, the BM adiposity is lower in rodents than in humans and varies by mouse strain. As an example, C3H/HeJ (C3H) mice have very high endogenous levels of BMAT, while C57BL/6 mice have very low endogenous levels [
BMAT consistently expands with obesity and decreases with weight loss in clinically obese patients; these trends have also been seen in diet-induced obese rodents [
]. It remains controversial whether type 1 and 2 diabetes influence BMAT fate; however, BMAT is positively associated with glycated hemoglobin HbA1c levels and lower lipid unsaturation in diabetic patients [
Does vertebral bone marrow fat content correlate with abdominal adipose tissue, lumbar spine bone mineral density, and blood biomarkers in women with type 2 diabetes mellitus?.
]. BMAT volume increases in chronic caloric restricted patients with anorexia nervosa and decreases with exercise, acute caloric restriction and in late stages of anorexia nervosa when marrow becomes gelatinous [
Expansion of bone marrow adipose tissue during caloric restriction is associated with increased circulating glucocorticoids and not with hypoleptinemia.
]. Glucocorticoids and parathyroid hormone (PTH), a potent osteoanabolic drug, increased BMAT in postmenopausal osteopenic women and ovariectomized rats [
3.3 BM adipocytes are critical for bone homeostasis and hematopoiesis
3.3.1 Bone homeostasis
The correlation between BMAT and bone mineral density is often cited as a negative regulator of bone mass, and BMAT expansion appears to be a relevant marker of compromised bone integrity. In African and Caucasian healthy adults, μCT and MRI measurements have shown an inverse correlation between BMAT mass and bone quality/density [
Relationship between MRI-measured bone marrow adipose tissue and hip and spine bone mineral density in african-american and caucasian participants: the CARDIA study.
]. Caloric restriction or starvation in growing mice leads to an increased accumulation of BM fat, decreased bone density, and increased bone resorption [
]. Although strain-, sex-, age- and diet-dependent, the expansion of BMAT is associated with decreased bone quality with reduced bone mass and impaired bone strength in experimental models of obesity in rodents [
]. However, diet-induced obesity in rodents, that increases BMAT volume, appears to be responsible for an initial increase in bone mass followed by an impaired bone formation [
High bone mass in adult mice with diet-induced obesity results from a combination of initial increase in bone mass followed by attenuation in bone formation; implications for high bone mass and decreased bone quality in obesity.
The link between BM adipocytes and bone might be attributed to the common origin of their BM progenitors, MSCs, and to the involvement of PTH1 and leptin signaling pathways, which are known regulators of the balance between adipogenesis and osteogenesis, as shown in Prx1-Cre; Lepr(fl/fl) and leptin-deficient ob/ob mice [
]. Additionally, BM adipocytes secrete factors that directly regulate bone remodeling, such as extracellular vesicles, adipokines (leptin and adiponectin) and inflammatory factors, such as IL-6 and TNFα [
]. BMAT also contributes to osteoclast-like cell differentiation and secretes the chemokines CXCL 1 and 2, promoting osteoclast maturation in metastatic prostate cancer [
] (Fig. 1). Through a three-dimensional analysis of the ultrastructure of the BMAT niche using focused ion beam scanning electron microscopy (FIB-SEM), Scheller and colleagues showed that in mouse tibia, BM adipocytes extend organelle- and lipid-rich cytoplasmic regions toward areas of active osteoblasts at the bone surface, allowing reception of lipid and protein signals from adipocytes to osteoblasts, and therefore linking BM adipocytes and bone [
BMAT was first described as a negative regulator of hematopoiesis in mice. In genetically modified lipoatrophic A-ZIP/F1 “fatless” mice and in C57BL/6 J mice treated with the adipogenesis PPARγ inhibitor bisphenol A diglycidyl ether (BADGE), BMAT loss improves hematopoietic recovery by enhancing HSC engraftment in marrow transplantation experiments after lethal irradiation or chemotherapy [
]. While excessive medullary adiposity is considered deleterious for hematopoiesis, adipocytes can support primitive hematopoiesis in vitro and are required for in vivo BM HSC mobilization or retention through the regulation of the retention factor CXCL12/CXCR4 pathway in steady-state hematopoiesis [
Adipocytic cells augment the support of primitive hematopoietic cells in vitro but have no effect in the bone marrow niche under homeostatic conditions.
]. Other studies have also revealed that BM adipocytes, as a source of stem cell factor (SCF), are positive regulators of HSC survival and hematopoietic regeneration in steady-state hematopoiesis and in metabolic stress contexts such as obesity and aging [
]. Ambrosi and colleagues demonstrated in mice that adipogenic transplants of adipogenic-committed progenitor cells (CD45−CD31−Sca1+PdgfRα+CD24−) or mature CD45−CD31−Sca1−Zfp423+ preadipocytes in irradiated mice reduced lineage-sca1+ c-kit+ (LSK) frequencies and impaired hematopoietic repopulation. However, the transplantation of multipotent CD45−CD31−Sca1+CD24+ cells to generate adipocytes increases LSK BM recovery, suggesting that according to their differentiation stage, adipocyte precursors can exert distinct effects on hematopoiesis [
]. Therefore, whether BMAT plays a beneficial or detrimental role in hematopoiesis requires further study but appears to be dependent on the experimental model, differentiation state, location and disease. A three-dimensional analysis of BMAT niche ultrastructure using FIB-SEM showed that in mouse tibia, BM adipocytes extensively interact with maturing myeloid/granulocyte cell lineages and with erythroblast islands (macrophage-like nurse cells that provide support for maturing erythroblasts and traffic mature erythrocytes to sinusoids) [
], emphasizing the role of BM adipocytes in erythropoiesis and myelopoiesis through direct cell contact and indirect signals through adjacent cell membrane communication. The disruption of the BMAT niche during acute myeloid leukemia results in impaired myeloerythroid maturation [
]. In vitro culturing of BM cells with adiponectin, a BMAT-secreted hormone, enhances myelopoiesis and is associated with an inhibitory effect on B lymphopoiesis [
Adiponectin, a fat cell product, influences the earliest lymphocyte precursors in bone marrow cultures by activation of the cyclooxygenase-prostaglandin pathway in stromal cells.
]. Diet-induced obesity in rodents also rapidly suppresses B lymphopoiesis in vivo by disrupting the supportive capacity of the BM niche, upregulating myelopoiesis [
] and inducing a shift of lymphoid to myeloid cell differentiation with enhanced canonical myeloid gene (Csf1r, Spi1, Runx1) expression and decreased lymphoid gene (Flt3, Tcf3, Ebf1) expression [
] (Fig. 1). Despite discrepancies in the impact of BMAT on hematopoiesis, additional investigation is needed to elucidate the soluble signals involved in the communication between BM adipocytes and hematopoietic cells.
4. Obesity influences megakaryopoiesis and platelet production
4.1 Obesity enhances megakaryopoiesis and alters platelet production
To date, no study linking megakaryopoiesis and obesity in humans has been reported. Nevertheless, some studies have shown an increased mean platelet volume (MPV) in humans positively correlated with body mass index (BMI), while others have reported no correlation [
How obesity affects the neutrophil/lymphocyte and platelet/lymphocyte ratio, systemic immune-inflammatory index and platelet indices: a retrospective study.
]. Some studies have also shown a normal platelet count in the obese population compared to healthy individuals, while others reported that the platelet count is slightly higher in obese individuals, particularly in women with visceral adiposity, and is positively correlated with BMI [
How obesity affects the neutrophil/lymphocyte and platelet/lymphocyte ratio, systemic immune-inflammatory index and platelet indices: a retrospective study.
White blood cell count in women: relation to inflammatory biomarkers, haematological profiles, visceral adiposity, and other cardiovascular risk factors.
]. The mechanism behind the increased MPV and platelet count in obese individuals is poorly understood, but it appears to be a reflection of an increased percentage of reticulated young platelets that is linked to BMI [
In an experimental high-fat diet-fed male murine model that induced obesity, platelets released into the circulation have an increased MPV and a normal lifespan [
]. In this mouse model, obesity influenced BM megakaryopoiesis by increasing the MK size and enhancing their nuclear maturation and DMS formation. This enhanced MK maturation induced by obesity in mice, accelerates proplatelet formation, and is responsible for the ectopic platelet release outside the sinusoidal vessels in the BM and the slight decreased in circulating platelet count. However, in male rats, diet-induced obesity induces a slightly increase in platelet count [
]. These current differences in platelet count observed in obese men/women and male rodent models of obesity might be attributed to interspecies differences or/and to sex dysmorphism. Whether the slight decrease or increase in platelet count in different species has an impact on thrombotic risk needs to be further investigated.
4.2 Molecular mechanisms of the impact of obesity on megakaryopoiesis and platelet production
The molecular impact of obesity on megakaryopoiesis and platelet production remains ill-defined, but it appears to be multifactorial (Fig. 2). In the BM from obese mice, where the BM adipocyte mass increased, no direct interaction between BM adipocytes and MKs has been observed [
], strongly suggesting the involvement of soluble factors in the influence of obesity on megakaryopoiesis and platelet production.
Fig. 2Factors influencing megakaryocyte and platelet biology in obesity. Obesity can influence megakaryopoiesis, the production of platelets, and platelet activation by increasing the level of circulating thrombopoietin (TPO), increasing free fatty acid uptake and synthesis, upregulating reactive oxygen species (ROS) production, inducing a chronic low-grade inflammatory state, and stimulating adipokine secretion. In obesity, the platelet size is correlated with platelet activation, BMI, and circulating levels of platelet activation markers such as soluble P-selectin, soluble CD40L, soluble GPVI and TXA2 and platelet-derived microparticles (PMPs). Platelets have higher surface expression of GPVI and CLEC-2 and display upregulated GPVI and αIIbβ3 signaling pathways that make platelets prone to GPVI-dependent adhesion and aggregation and less sensitive to anti-aggregating agents. ROS: reactive oxygen species; DNL: de novo lipogenesis; DMS: demarcation membrane system; TXA2: thromboxane A2; TPO: thrombopoietin, IL: interleukin; TNFα: tumor necrosis factor α; PMP: platelet microparticles, CCL5: chemokine (C C motif) ligand 5.
Several studies have highlighted an important role for fatty acids in MK differentiation and platelet production. Recent studies using human and murine MKs have shown that free fatty acids (released by adipocytes or generated by de novo lipogenesis) can directly improve MK maturation. In vitro coculture of adipocytes with murine BM hematopoietic progenitors committed to MK differentiation with TPO showed a direct dialog between MKs and adipocytes through which MK maturation is enhanced and adipocytes are delipidated. This dialog clearly involves a direct free fatty acid transfer from adipocytes to MKs, through the MK fatty acid translocase CD36, which increases ploidy levels by upregulating MK maturation-related signaling pathways, enhances DMS formation and accelerates proplatelet formation [
]. Therefore, excessive adipocyte accumulation and excessive fatty acid release in obesity could explain the enhanced MK maturation induced by obesity. In the same line of evidence, adding free fatty acids (docosahexaenoic acid, arachidonic acid and polyunsaturated fatty acids) to CD34+ human hematopoietic progenitor cells committed toward the MK lineage increases the number of differentiated MKs and reduces their apoptosis and reactive oxygen species (ROS) production, resulting in higher platelet production [
Enhanced generation of megakaryocytes from umbilical cord blood-derived CD34(+) cells expanded in the presence of two nutraceuticals, docosahexanoic acid and arachidonic acid, as supplements to the cytokine-containing medium.
]. A study by Kelly and colleagues demonstrated that abolishing free fatty acid production in MKs by inhibiting acetyl-CoA carboxylase (ACC) or fatty acid synthase (FASN), two enzymes involved in de novo lipogenesis from non-lipid substrates (such as glucose), impaired DMS formation and platelet production in humans and monkeys but not in rodents or dogs [
]. Although it is unclear why MK DMS formation and platelet production in humans and primates would be more dependent on de novo lipogenesis than rodents, the interspecies differences observed in the importance of de novo lipogenesis in other tissues could explain this phenomenon. This suggests that in mice, adipocytes from BM are the major source of fatty acids for MK maturation. Nevertheless, further work is needed to define the signals of megakaryocytic origin that regulate fatty acid release by adipocytes.
4.2.2 Adipokines
Excessive fat accumulation in obese individuals strongly increases adipokine secretion by adipocytes [
], which has a direct impact on megakaryopoiesis. Resistin, leptin, plasminogen activator inhibitor-1 (PAI-1) and retinol binding protein (RBP) 4 secreted by adipocytes induce insulin resistance in human MKs by interfering with insulin receptor substrate (IRS) 1 [
Induction of insulin resistance by the adipokines resistin, leptin, plasminogen activator inhibitor-1 and retinol binding protein 4 in human megakaryocytes.
]. Adipose tissue accumulation can also influence megakaryocytic progenitor proliferation and differentiation into MKs by oversecreting TPO, the main cytokine regulator of megakaryopoiesis and thrombopoiesis, as observed in the serum of obese women [
Oxidative stress is dysregulated during obesity, with increased ROS production and decreased NO bioavailability. A high ROS environment (20 % O2) enhances nuclear MK maturation and proplatelet-forming MKs in CD34+ cultures from human BM samples [
]. In MKs from human CD34+ cells or from murine BM cultures, when ROS production is prevented by the addition of different antioxidants (such as N-acetyl cysteine, Trolox, quercetin) or nicotinamide adenine dinucleotide phosphate (NADPH) oxidase (NOX) inhibitors, the activation of the main signaling pathways required for megakaryopoiesis, such as ERK, AKT and JAK2, is partly inhibited and MK polyploidization is affected [
]. Additionally, human MKs treated with antioxidant compounds extend fewer proplatelets, whereas MKs treated with prooxidant molecules produce more proplatelets [
]. The role of ROS in MK maturation and platelet production has also been shown by the effect of antioxidant molecules on counteracting the role of the peroxisome proliferator-activated receptor gamma (PPARgamma) ligand called 15-deoxy-delta(12,14)-PGJ(2) (15d-PGJ(2)), which is known to promote megakaryocytic maturation in the MEG-01 megakaryoblastic cell line and primary human and mouse MKs and to accelerate platelet production after in vivo radiation-induced BM injury in mice [
]. Thus, in obesity, increased ROS production could lead to increased MK maturation and platelet production. For NO, both exogenous and endogenous sources of NO suppress the growth and differentiation of MKs by promoting apoptosis in the megakaryoblastic cell lines MEG-01 and HEL, and conversely, superoxide inhibitors suppress NO-induced MK apoptosis. Treatment with NO facilitates platelet production by MEG-01 cells, which is enhanced in combination with TPO [
]. This suggests that MK apoptosis induced by NO is essential for platelet release and that a decrease in NO bioavailability could lead to an increase in MK maturation but not an increase in platelet release.
4.2.4 Inflammatory cytokines
Obesity is also characterized by a chronic low-grade inflammatory state with increased plasma levels of inflammatory cytokines such as IL-6, IL-1β, and CCL5 that enhance megakaryopoiesis. In vitro, in human CD34+ cells and murine hematopoietic progenitor cells, IL-6 alone stimulates MK maturation by enhancing MK number, size, ploidy levels and proplatelet formation. In synergy with SCF, TPO and IL-11 or low doses of IL-3 and IL-1, IL-6 allows the proliferation of MK progenitors differentiated from purified CD34+ cells. In vivo, IL-6 administration in mice and primates induces an increase in the size, number, ploidy and protein synthesis of MKs, and an increased platelet count [
Interleukin-6 and interleukin-11 act synergistically with thrombopoietin and stem cell factor to modulate ex vivo expansion of human CD41+ and CD61+ megakaryocytic cells.
]. IL-1β activates nuclear factor-κB and mitogen-activated protein kinase (MAP) signaling pathways, which increases the ploidy of MKs from obese high-fat diet-fed mice [
Interleukin 1 receptor 1 and interleukin 1beta regulate megakaryocyte maturation, platelet activation, and transcript profile during inflammation in mice and humans.
5. Obesity influences platelet activation and thrombosis
5.1 Platelet activation is enhanced during obesity increasing the thrombosis risk
It is now well known that obesity is associated with increased thrombosis risk due to platelet activation (Fig. 2). In humans, obesity leads to higher circulating levels of activated platelet markers such as soluble P-selectin, soluble CD40L and soluble GPVI and TXA2 [
]. A proteomic analysis of platelets from individuals with severe obesity showed higher expression levels of GPVI and CLEC-2 that correlated with BMI, and upregulated levels of αIIb integrin and fibrinogen isoforms. Deep phosphoproteomic analysis of these platelets revealed upregulated phosphorylation of the active form of the tyrosine kinase Src (pTyr418) in the basal state and hyperphosphorylation of proteins involved in αIIbβ3 integrin-, GPIb-IX-V complex-, GPVI- and CLEC-2-related signaling pathways, in cytoskeleton remodeling, in platelet degranulation and in ROS production [
GPVI surface expression and signalling pathway activation are increased in platelets from obese patients: elucidating potential anti-atherothrombotic targets in obesity.
]. An in vitro analysis of platelet features showed that obesity, both in humans and rodents, increases GPVI-dependent platelet aggregation and platelet adhesion to the collagen surface but not to the fibrinogen surface and induces the formation of larger thrombi on the collagen surface under arterial shear conditions, making GPVI a potential anti-thrombotic target in the context of obesity [
5.2 Molecular mechanisms of the impact of obesity on platelet hyperactivity
The causes for enhanced platelet activation in obesity appear to be multifactorial and are probably caused by hereditary (intrinsic) changes in platelets due to the influence of obesity on MKs and by acquired (extrinsic) changes within platelets due to systemic modifications such as increased inflammatory cytokine and adipokine plasmatic levels.
5.2.1 Intrinsic causes
Several studies have shown that increased thrombosis risk is related to hereditary changes in platelets from altered megakaryopoiesis and platelet production caused by obesity. Increased MPV is associated with an increased number of mRNA rich-reticulated young platelets in obese individuals, which correlates with upregulated adenosine diphosphate (ADP) and P-selectin expression in platelets, increased aggregation, and resistance to antiplatelet therapy. Increased MPV during obesity appears to be a good predictor of increased thrombotic risk [
]. Recent transcriptomic studies also showed that in obese women, obesity is associated with an altered platelet transcriptome and increased platelet activation, which partly returns to normal after weight loss following bariatric surgery. The changes observed in platelet gene expression in obese individuals after surgical weight loss implicate pathways improving cardiometabolic risk post-weight loss [
]. Although the mechanism behind this transcriptomic modification needs to be further analyzed, it appears that the altered platelet production by MKs in obesity impacts platelet activation by increasing thrombotic risk. In mice, the platelet ultrastructure as analyzed by transmission electron microscopy is unaltered by obesity [
Several systemic factors have a direct impact on platelets by increasing their activation and thrombosis. Insulin resistance, commonly linked with obesity, is responsible for reduced platelet sensitivity to insulin, leading to increased platelet aggregation in obese individuals [
Pathophysiology of platelet resistance to anti-aggregating agents in insulin resistance and type 2 diabetes: implications for anti-aggregating therapy.
]. Conversely, the plasma level of adiponectin is inversely correlated with soluble P-selectin and CD40L in obese patients, acting as an antithrombotic factor [
]. The upregulated levels of platelet protein disulfide isomerase (PDI) and NOX-1 that increase platelet ROS generation in obese individuals contribute to platelet activation dependent on the GPVI signaling pathway and increase the thrombotic risk of obesity [
]. In individuals with visceral obesity, a higher plasma level of 8-isoPGF2α, the most abundant isoprostane produced through a ROS-dependent mechanism, is positively correlated with circulating levels of soluble CD40L, 11-dehydro TXB2 and PMP and is able to dysregulate platelet functions such as adhesive functions and TXA2 receptor-dependent activation [
Increased levels of microparticles originating from endothelial cells, platelets and erythrocytes in subjects with metabolic syndrome: relationship with oxidative stress.
]. Obese patients also have higher circulating levels of C-reactive protein, which correlate with BMI (as adipose tissue is a source of C-reactive protein) and return to normal levels after weight loss [
]. The dissociation of pentameric C-reactive protein at the platelet surface into monomeric C-reactive protein enhances platelet activation and adhesion, thus promoting thrombosis and thrombus growth [
Monomeric C-reactive protein is prothrombotic and dissociates from circulating pentameric C-reactive protein on adhered activated platelets under flow.
]. C-reactive protein is also implicated in the fibrinolytic system by enhancing PAI-1 in human endothelial cells; this enhancement is known to decrease fibrin clearance and thus promotes thrombotic risk [
]. The increased circulating TNFα level in obese individuals leads to increased platelet adhesion to the endothelium and is responsible for platelet hyperactivity, TXA2 biosynthesis and CD40L shedding [
]. Finally, obese individuals display a decreased platelet sensitivity to physiological anti-aggregating agents such as glyceryl trinitrate (GTN) and sodium nitroprusside (SNP), PGI2, adenosine and cyclic nucleotides [
Impaired synthesis and action of antiaggregating cyclic nucleotides in platelets from obese subjects: possible role in platelet hyperactivation in obesity.
Complications arising from obesity, particularly thromboembolic incidents, lead to increased risks of cardiovascular morbidity and mortality. There is significant interplays between endothelial dysfunction, hypercoagulability, altered megakaryopoiesis and platelet production, platelet hyperactivity, inflammation and the accumulation of adipose tissue. While some diagnostic and treatment options are available, further research is necessary to increase our knowledge in the underlying pathophysiology of obesity in cardiovascular diseases involving platelets.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. This work was supported by Fondation de France (grant number 00075817), Fondation pour la Recherche Médicale (grant number DEQ20170336737) and Inserm/Région Occitanie (R21100BB).
References
Van Gaal L.F.
Mertens I.L.
De Block C.E.
Mechanisms linking obesity with cardiovascular disease.
Reporting guidelines, review of methodological standards, and challenges toward harmonization in bone marrow adiposity research. report of the methodologies working group of the international bone marrow adiposity society.
Standardised nomenclature, abbreviations, and units for the study of bone marrow adiposity: report of the nomenclature working Group of the International Bone Marrow Adiposity Society.
Does vertebral bone marrow fat content correlate with abdominal adipose tissue, lumbar spine bone mineral density, and blood biomarkers in women with type 2 diabetes mellitus?.
Expansion of bone marrow adipose tissue during caloric restriction is associated with increased circulating glucocorticoids and not with hypoleptinemia.
Relationship between MRI-measured bone marrow adipose tissue and hip and spine bone mineral density in african-american and caucasian participants: the CARDIA study.
High bone mass in adult mice with diet-induced obesity results from a combination of initial increase in bone mass followed by attenuation in bone formation; implications for high bone mass and decreased bone quality in obesity.
00425-X&id=gr1.jpg){kind=link}
00425-X&id=gr2.jpg){kind=link}