Net negatively charged cells such as white and red blood cells and platelets are repelled from the glycocalyx due to its negative charge. Interactions between leukocytes and endothelial cell (EC) are mediated by adhesion molecules that direct leukocyte rolling along and adhesion to the endothelial surface. Under resting (non-activated) conditions these selectins and integrins are expressed at low or non-detectable levels and are masked by the overlying glycocalyx (Reitsma et al., 2007). The specific glycocalyx constituents responsible for preventing leukocyte adhesion have not been fully elucidated. At the level of the mouse intestinal mesentery, heparinase treatment results in increased leukocyte adhesion, pointing to a role for Heparan sulfate proteoglycans (HSPGs) (Mulivor and Lipowsky, 2002). In addition, using human umbilical vein ECs in culture, our group has shown that the knockdown of endomucin (EMCN) using siRNA leads to increased leukocyte adhesion in non-activated EC and that the adhesion was mediated by constitutively expressed I-CAM (Zahr et al., 2016). In addition, using the same system, the overexpression of EMCN in tumor necrosis factor-alpha (TNF-α) (activated) EC was able to significantly suppress leukocyte adherence as did adenoviral expression of EMCN in a mouse model of ocular inflammation.

A number of molecules bound to the glycocalyx account for its anti-coagulant and anti-thrombotic properties, including anti-thrombin III, tissue factor pathway inhibitor, and dermatan sulfate (reviewed in Potje et al., 2020). In addition, thrombomodulin, one of the glycocalyx proteoglycans, is another major contributor to this activity, acting via binding to thrombin (reviewed in Martin et al., 2013).

Under normal conditions, the glycocalyx forms a barrier against vascular permeability, partly by acting as a negatively charged molecular sieve (Alphonsus and Rodseth, 2014). The glycocalyx impacts vascular permeability via its three-dimensional structure—interactions among the various components generate a “mesh-like” structure that can control passage of molecules based on their size. Its tight mesh structure excludes macromolecules greater than 70 kDa (Pries et al., 2000). In addition, the glycocalyx structure is negatively charged and thus can limit the transit of molecules based on charge. For instance, the glycocalyx is the primary site for the restriction of the passage of albumin across the endothelium (Curry and Adamson, 2012), which although net negatively charged binds to the glycocalyx due its amphoteric characteristics (reviewed in Aldecoa et al., 2020).

Heparan sulfate has been demonstrated to be a primary sensor of the directionality of shear stress. Exposure of cultured EC to heparinase increased the speed of flow-induced endothelial migration but prevented shear-stress stimulated directionality of migration and recruitment of phosphorylated focal adhesion kinase (Moon et al., 2005). The glycocalyx also transmits the effects of shear stress on endothelial nitric oxide synthase (eNOS), which is localized to caveolae on the endothelial surface. Enzymatic removal of heparan sulfate prevents sheer stress-induced NO production (Pahakis et al., 2007). Recent evidence indicates that it is heparan sulfate in the form of glypican-1 that mediates this signal, as depletion of glypican-1 using siRNA prevent the shear stress activation of eNOS (Ebong et al., 2014). Using syndecan-4 core protein as a model, a large-scale molecular dynamics computational experiment indicated the syndecan-4 core protein manifests as a scissors-like motion and transmits force as a main pathway of signal transmission. While blood flow velocity changes the shedding of proteoglycan sugar chain, its functional role is to protect the core protein from severe conformational changes (Jiang et al., 2020). More mechanistic roles of glycocalyx in mechanosensing and transduction have been thoroughly reviewed recently (Zeng et al., 2018).

A number of the proteoglycan core proteins span the plasma membrane and associate with the endothelial cortical cytoskeleton via linker proteins such as talin and dystonin, thus communicating extracellular forces though the cell (reviewed in Cosgun et al., 2020). Conversely, the cortical cytoskeleton can alter the conformation of the glycocalyx so that polymerized actin leads to a flattened glycocalyx and actin disassembly results in a more extended glycocalyx. A recent study has demonstrated cooperation between intracellular cytoskeletal components and constituents of the glycocalyx in how cells alter their shape in response to their microenvironment. Specifically, it was shown that “polymers” formed by glycocalyx mucins and polysaccharides “generate an entropic bending force to favor formation of curved membrane structures” (Shurer et al., 2019). The authors speculate that the morphological impact of the glycocalyx would have “broad consequences on membrane processes, ranging from absorption and secretion to cellular communication, signaling, and motility.”

With heparan sulfate comprising the largest proportion of the glycocalyx (Reitsma et al., 2007) and knowing the significant effect that heparan sulfate has on the bioavailability, activity, and stability of a wide number of growth factors (Ricard-Blum and Lisacek, 2017), it is not surprising that the glycocalyx has an essential role in vessel formation and maturation. For example, heparan sulfate containing GAGs have been shown to interact with both the angiopoietins and Tie receptors as well as serving as a ligand for Tie1, enhancing signaling and cell survival (Griffin et al., 2021). HSPGs interact with VEGF and promotes angiogenesis through increasing VEGF levels in the extracellular microenvironment and fine-tuning the interaction of VEGF with its receptor and co-receptor (Nagarajan et al., 2018). Another membrane glycoprotein, neuropilin, is known to function as a co-receptor for VEGF receptor 2 (VEGFR2) and guide developmental angiogenesis (Gelfand et al., 2014). Using zebrafish as a model, it was demonstrated that fish lacking perlecan failed to develop primary intersegmental vessel sprouts (Zoeller et al., 2008). In addition, work from our group has identified a role for EMCN, a type 1 transmembrane glycoprotein abundantly expressed by venous and capillary endothelium, in the regulation of VEGFR2 signaling (see Part 2 below on glycocalyx and angiogenesis). Chemokines-GAG interactions mediates chemokines membrane localization and surface presentation and fine-tune the function of chemokines (Proudfoot et al., 2017). Interaction between CXCL8 and extracellular HSPGs regulates the leukocyte recruitment and transmigration (Sarris et al., 2012).

One major regulator of glycocalyx thickness is flow, such that exposure of culture EC to laminar flow leads to the increased synthesis of hyaluronan (Maroski et al., 2011). Whereas disturbed flow induces an approximately 50% decrease in endothelial glycocalyx compared to uniform laminar flow in vitro (Mensah et al., 2020). This effect of shear stress on the glycocalyx becomes particularly relevant in the pathogenesis of atherosclerosis (see Part 3). Exposure to pro-inflammatory cytokines leads to a dramatic reduction in the glycocalyx, largely by endothelial-derived metalloproteinase-mediated shedding (Yang et al., 2020) as well as by heparinase secreted by mast cells and hyaluronidase produced by the EC (Bourguignon and Flamion, 2016).

Angiogenesis is the formation of blood vessels from pre-existing vessels, and components of the glycocalyx mediate elements of angiogenesis in diverse and multifaceted ways. GAGs play an active role in regulating angiogenesis. In general, the gel-like composition generated by the presence GAGs, in combination with identified cytokine interactions, mainly FGFs and VEGF, facilitate growth factor signaling in ECs by influencing their bioavailability, local concentrations, and stability (Muramatsu and Muramatsu, 2008; Huynh et al., 2012).

Proteoglycans have been implicated in modulating angiogenesis more specifically by influencing cell proliferation and migration under both normal and pathologic conditions (Iozzo and Sanderson, 2011; Gubbiotti et al., 2020). Heparan sulfate is a known regulator of angiogenesis, and its 6-O-sulfation specifically has been reported to modulate VEGF165-induced angiogenesis (Stringer, 2006). VEGF, a well-studied, angiogenic factor, is produced as a number of splice variations, which differ in their ability to bind to heparan sulfate proteoglycan. Whereas VEGF120 does not contain any heparan sulfate binding domains, it is freely soluble upon its secretions from cells, VEGF188, on the other hand, binds very tightly to heparan sulfate and remains cell-associated upon release, and VEGF164 displays intermediate properties. Mice genetically engineered to express only VEGF120, that is lacking both heparan sulfate binding isoforms, are lethal perinatally due to defective lung development (Galambos et al., 2002), indicating the importance of growth factor association with the glycocalyx to their function.

Members of syndecan family of heparan sulfate proteoglycans have been studied for their diverse roles in angiogenesis regulation. Various syndecans have distinct function in promoting or inhibiting angiogenesis depending on whether they are present on the EC extracellular membrane or shed (De Rossi and Whiteford, 2013; De Rossi et al., 2014). Syndecan-1 knockdown in both in vivo and in vitro has been shown to interrupt VEGFR2 internalization by preventing clathrin-mediated endocytosis (Jing et al., 2016). A decrease in VEGFR2 expression following syndecan-1 depletion has also been reported (Lamorte et al., 2012). Syndecan-2 is necessary for angiogenesis, potentially acting as a co-receptor for VEGF signaling, giving it a critical role in promoting angiogenesis whereas the shed variant of the protein plays an active role in inhibiting cell migration (Essner et al., 2006; De Rossi et al., 2014). Similarly, syndecan-4 binds VEGF “sequestering” and preventing it from exerting its angiogenic signaling (Lambert et al., 2020).

Hyaluronan (HA), an abundant glycocalyx GAG, exerts an anti-angiogenic effect in its native form, whereas its degradation products induce angiogenesis by activating HA receptors CD44 and CD168 (West and Kumar, 1989; Slevin et al., 2007; Ghose et al., 2018). HA has also been reported to prevent the binding of VEGF to its primary signaling receptor VEGFR2 in a sulfate-dependent manner (Rother et al., 2017).

Another prominent glycocalyx component is EMCN, a membrane spanning mucin-type glycoprotein that has been shown to be involved for VEGFR2 signaling. Initial studies demonstrated that the knockdown of EMCN in the developing vasculature led to impaired vascular development and reduced tip cell activity (Park-Windhol et al., 2017). Investigation of the mechanism underlying this effect using primary human retinal capillary EC in culture revealed that siEMCN knockdown of EMCN blocked VEGF-stimulated proliferation, migration, and tube formation. Subsequent studies showed that the lack of EMCN leads to a failure of VEGF-activated VEGFR2 to internalize, thereby preventing the sustained intracellular signaling that is necessary for VEGF to stimulate endothelial function. EMCN, however, is itself not internalized with VEGFR2 during normal signaling (LeBlanc et al., 2019). Structure-function analysis using EMCN truncation mutants has revealed the minimum extracellular 21–121 amino acids of EMCN necessary for modulating VEGFR2 signaling and indicated also that the activity is dependent on N-glycosylation (Hu et al., 2020).

Since the first description of the animal glycocalyx by Martínez-Palomo (1970), there has been growing interest in the role of the glycocalyx, including in pathologic angiogenesis. Recent reports demonstrate that the glycocalyx is important for not only developmental angiogenesis and its maintenance, but for cancer progression secondary to its vascularization as well. Angiogenesis is necessary for tumor growth beyond the limits of diffusion. The modification of the glycocalyx structure of cancer cells has been suggested to provide tumors with the ability to survive and overcome vascular limitations, for example by mimicking the vasculogenic nature of ECs with 3-dimentional organization, while lacking EC markers such as CD31 and CD34 (Tachi et al., 2019). FGF and VEGF contain heparan-binding domains and their signal transduction depends on its sulfation. Alteration of heparan sulfate’s ability to bind FGF and VEGF has been reported to reduce angiogenic signal transduction and weaken the vascular foundation for tumors (Fuster et al., 2007). As the specific components of the glycocalyx are actively involved in aberrant angiogenesis, they may represent promising targets for therapy.

The stability of the vasculature is also dependent on the integrity of the glycocalyx layer as seen in diabetic retinopathy. Hyperglycemia has been reported induce the shedding of various components of the glycocalyx, disrupting EC function by increasing permeability in glomerular ECs (Singh et al., 2011). Loss of EMCN, for example, has been suggested to be associated with diabetic retinopathy and the overexpression of EMCN leads to the restoration of the glycocalyx and the related EC function in streptozotocin-induced diabetic rats (Niu et al., 2019). VEGF165b, a non-angiogenic isoform of VEGFA, was reported to restore the glycocalyx and ultimately the endothelium after diabetic damage in isolated diabetic human and rat glomeruli. VEGF165b shares the protective prosurvival and antiapoptotic signaling of canonical VEGF isoforms and was reported to restore the glycocalyx in short and long-term conditions (Oltean et al., 2015).