Heparan sulfates and heparan sulfate proteoglycans in hematopoiesis Free

Hematopoiesis is the process of blood formation in which hematopoietic stem and progenitor cells (HSPCs) self-renew to maintain the hematopoietic pool as well as differentiate and commit toward specialized lineages that carry out a host of cell-type–specific functions. Hematopoietic tissue can be broadly divided into myeloid and lymphoid arms, each of which contains its own distinct partially committed progenitors. Master transcriptional regulators, distinct stromal niches, and specific cytokine cues are critical in the maintenance of HSPCs, as well as their differentiation toward fully mature, functional cell lineages. External cues such as cytokine/chemokine stimulation occur through the binding of ligands to their corresponding surface receptors, a process frequently facilitated by coreceptors, as well as molecular modifications.

Heparan sulfates (HSs) are linear polysaccharide chains attached to select surface proteins (HS proteoglycans [HSPGs]) that are present in all eukaryotic cells. They are part of a family of molecules known as glycosaminoglycans, which also include chondroitin/dermatan sulfate, keratan sulfate, and hyaluronic acid. HS chains are synthesized as repeating disaccharide units of hexuronic acid (glucuronic acid [GlcA] or iduronic acid [IdoA]) and N-acetylglucosamine (GlcNAc), which can be modified at various positions, allowing for enormous chain complexity. The synthesis of HS is a concerted process relying on multiple families of enzymes and occurs in the endoplasmic reticulum and/or golgi apparatus. First, a conserved tetra-saccharide linker (made of xylose-galactose-galactose-glucuronic acid) is sequentially added to serine residues of a core HSPG protein by specific glycosyltransferases. A GlcNAc residue is then added to initiate HS biosynthesis, a step that distinguishes HS biosynthesis from other types of glycosaminoglycan synthesis (ie, chondroitin sulfate). In the next step, which is known as chain polymerization, the exostoses family of enzymes adds repeating GlcA-GlcNAc units. The resulting 50 to 150 unit polysaccharide chains are additionally modified by sulfation (at O- and N-residues), acetylation, and epimerization at numerous points along the chain, which is carried out by an array of modification enzymes (Figure 1A).¹ This elaborate process results in highly diverse modification patterns owing to near-limitless combinatorial possibilities. This molecular diversity contributes to cell-specific functionalities in many physiologic and developmental processes.^2-4 

HS chains are attached to core proteins called HS proteoglycans (HSPGs). These include 3 major classes of proteins: syndecans (SDCs), a class of transmembrane proteins (Sdc1-4 in mammals); glypicans, a class of glycophosphatidylinositol–linked proteins (Gpc1-6 in mammals); and secreted proteins such as agrin, perlecan, and serglycin among others (Figure 1B). Specific proteoglycan families play roles in distinct signaling pathways: glypicans, for example, facilitate Hedgehog, Wnt, and fibroblast growth factor (FGF) signaling, betaglycan (also known as transforming growth factor β receptor 3 [TGFBR3]) mediates transforming growth factor β (TGF-β) signaling, and the SDC family acts as coreceptors for many receptor tyrosine kinases, including FGF receptors among others.^4,5 

Once on the cell surface, HS can be further altered through enzymatic removal of sulfation marks by sulfatases, Sulf1 and Sulf2, or cleaved by the enzyme heparanase (Figure 1A). These processes remodel the composition of HS and its modifications on the cell surface and facilitate the formation and reshaping of the extracellular matrix. Importantly, they influence cytokine/chemokine gradients, facilitate cell migration, and promote cellular signaling; all functions essential for the biology of hematopoietic cells.^6-8 

The importance of HS in cells that interact with hematopoietic cells (ie, bone marrow [BM] stroma, endothelial cells, etc) has been extensively documented and reviewed, but less is known about the roles of HS in hematopoietic cells themselves.^6,8 HS on stromal cells supports stem cell maintenance and hematopoietic cell differentiation, resulting in blood cell formation, and HS on the endothelial cell surface of tissues mediates leukocyte adhesion, trafficking, and migration. However, the surface of hematopoietic cells also contains HS, which confer unique cell-specific functionalities. There is a dearth of knowledge on the expression, function, and significance of HS that is expressed on many subtypes of hematopoietic cells when compared with cell types that comprise the hematopoietic niche (osteoblast/clasts, endothelial cells, etc). This review summarizes what is known about HS with regard to the regulation of hematopoiesis and blood cell function, focusing on each general cell type within hematopoietic tissue.

A multitude of cytokines and chemokines that are essential for hematopoietic cell differentiation and maintenance have been shown to bind to HS either in vitro or in vivo (Table 1). Many of these ligands are secreted by hematopoietic cells themselves and regulate normal hematopoiesis in an autocrine or paracrine manner.⁶⁰ Binding of cytokines and chemokines to HS can facilitate oligomerization and cell signaling, protect against proteolytic degradation, and create concentration gradients that fine-tune the signaling processes. Although much remains to be elucidated, biochemical studies have shown specific requirements of HS modifications (both specific sulfation requirements, as well as regional sulfation/charge distribution requirements) for the binding of certain cytokines and chemokines to HS.⁶¹ Many interleukin molecules selectively bind to distinctly modified HS, highlighting the specificity of HS-cytokine interactions (Table 1). With the expression of HS on the surface of hematopoietic cells and the abundance of HS-binding cytokines and chemokines, there is still much to be discovered regarding the role of HS-chemokine/cytokine interactions in hematopoiesis.

Perhaps the best-characterized HS interaction relevant to the hematologic system is the interaction between heparin (an on average shorter HS polysaccharide chain with higher sulfation and predominance of L-iduronic acid instead of D-glucuronic acid) and antithrombin, which mechanistically defines the anticoagulant properties of heparin. The binding of heparin activates antithrombin, which substantially increases the rate of inactivation of key coagulation factors, including thrombin, factor Xa, and factor IXa. It was initially observed that only a fraction of the isolated heparin molecules had high-affinity binding to antithrombin, and this small portion accounted for most of the anticoagulant activity of heparin.^62,63 Molecular studies later identified the pentasaccharide sequence with specific modifications, GlcNAc/NS(6S)-GlcA-GlcNS(3S,6S)-IdoA(2S)-GlcNS(6S), with a unique and necessary 3-O-sulfation, as the primary ligand for antithrombin.⁶⁴ The heparin-antithrombin interaction is the exemplar of the specificity requirements for functionally relevant HS ligand binding.

Numerous animal models deficient in specific HS-related genes have been developed, many of which show embryonic lethality and drastic phenotypes in nonhematopoietic tissues, such as skeletal, cardiac, and lung defects. Table 2 summarizes the hematologic findings (if relevant) of knockout animal models of HS-related genes. Other existing HS–related gene knockout models have either not been generated (eg, Gpc5 and Hs3st5), have not been viable for analysis, and no conditional models have been generated to date (eg, Gpc6), are viable but have not had their hematopoietic compartment studied (eg, Gpc1 and Gpc2), or show no overt hematopoietic phenotype (eg, Hs3st1 and Ndst4), revealing a substantial gap in knowledge. The fact that HS–related gene knockouts lead to hematologic phenotypes in animals highlights the importance of HS in hematopoietic development and maintenance in both the myeloid and lymphoid arms.

Specific HSPGs and HS-related genes (including specific HS modifications) have been functionally implicated in specific hematopoietic cell types (Figure 2A-B). Although cell-specific functions for HS have been described, some HS and HSPG functions are shared across multiple hematopoietic cell types. For example, secretory granules within various immune cells (eg, T cells, mast cells, and natural killer [NK] cells) are critical for immune function. Serglycin (once known as hematopoietic proteoglycan core protein), was initially characterized as a predominantly intracellular proteoglycan in the secretory granules of various immune cells as well as in platelets. Serglycin plays central roles in the integrity and content of granules within various hematopoietic cells, as specified in “Granulocyte (neutrophil/basophil/eosinophil) lineage,” “Mast cell lineage,” “Megakaryocyte lineage,” “The T-cell lineage,” and “NK cells.”

CD44 is expressed in nearly all hematopoietic cell types and has isoforms with various glycosylation patterns due to differential splicing. The CD44v3 isoform uniquely contains motifs available for HS attachment.⁸⁹ This specific isoform has been shown to contribute to key biological functions in monocytes, macrophages, T cells, B cells, and other hematopoietic cell types.

Multiple proteoglycans have been identified in HSPC biology in various species. Early gene expression studies implicated SDC2 as a regulator of long-term repopulating HSCs (LT-HSCs), whereas SDC1 was differentially expressed in multipotent progenitors (MPPs).⁹⁰ Recent studies have confirmed the role of SDC2 in hematopoietic stem cells in regulating stem cell quiescence and have shown surface SDC2 enriches for long-term repopulating HSCs.⁹¹ Glypican-3 expressed on mouse and human HSPCs appears to mediate HSPC residence in the BM as well as their trafficking through circulation, presumably by a mechanism of CD26 inhibition by tissue factor pathway inhibitor.⁷¹ Loss of the Drosophila homolog of perlecan (Trol) was found to diminish the proliferation of HSPCs and induce premature differentiation into mature blood cells, suggesting that pericellular perlecan is involved in HSPC maintenance and self-renewal (Table 2).⁷⁶ However, to date, no studies looking at perlecan in HSPC biology have been performed on other organisms.

The ApoE receptor, which has been shown to regulate the proliferation of HSPCs, was found in a proteoglycan-rich pool on the HSPC cell surface, implicating HS in the function of cholesterol transport into HSPCs.⁹² The FDCP-1 cell line, which mimics HSPCs was observed to produce and secrete HS proteoglycans.⁹³ 

Gfi1b, a master hematopoietic regulator, helps dictate erythroid and megakaryocytic lineage commitment from common bipotent megakaryocyte-erythrocyte progenitor (MEP), partly through direct repression of TGFBR3 proteoglycan (also called betaglycan), resulting in regulated TGF-β signaling.⁹⁴ 

Developmentally, the hematopoietic system in mammals arises from the mesodermal embryonic layer. Temporal expression of specifically modified HS has previously been shown to characterize the mesodermal cells that give rise to Flk1⁺ hemangioblasts, which ultimately contribute to the formation of the hematopoietic system.⁹⁵ Correspondingly, the addition of specifically modified HS (mainly N- and 6-O-sulfated HS) rescued hematopoietic differentiation in murine embryonic stem cell cultures lacking HS (Ext1^−/− ES cells).⁹⁶ The addition of O-sulfated HS along with canonical cytokines increased the long-term colony-initiating capacity (a measure of “stemness”) of human HSPCs in vitro.⁹⁷ Interestingly, induced pluripotent stem cells (iPSCs) derived from patients with Extl3 deficiency and cultured to promote HSPC and lymphocyte differentiation displayed diminished expansion of HSPCs and reduced ability to generate lymphoid progenitors.⁸⁰ These findings indicate that both the overall HS expression and temporal expression of distinct HS modifications have a significant impact on HSPC maintenance and differentiation ex vivo.

Moreover, the presence of distinct surface HS modification patterns can separate erythroid and megakaryocytic-primed MEPs, highlighting the expression of HS modifications as a contributor to hematopoietic cell fate decision, at least at the megakaryocyte-erythroid progenitor level.⁹⁸ 

Serglycin expression has been shown in the Golgi and granules of promyelocytes and other developing immature neutrophil populations.⁹⁹ The role and content of neutrophil granules seem to heavily rely on serglycin expression and function, including neutrophil elastase loading into granules and alpha-defensin retention during myelopoiesis.^100,101 

HS molecules on endothelial cells are critical for interaction with adhesion molecules such as L-selectin on neutrophils to promote migration and intravasation. Yet, much less is known about what roles HS might play on the neutrophil surface. Neutrophil migration in response to antithrombin III, but not fMLP (fMet-Leu-Phe) or interleukin-8 (IL-8) is mediated by HSPGs, suggesting the role of neutrophil HS in chemotaxis is ligand-specific.¹⁰² In mice, removal of 2-O-Sulfation on endothelial cells significantly reduced neutrophil migration in response to inflammatory triggers; however, removal of 2-O-Sulfation specifically on neutrophils caused no change in the inflammatory response, suggesting that HS (especially 2-O-Sulfation) on neutrophils is not a driving component in inflammation-mediated neutrophil migration.⁸⁴ Heparanase is found inside tertiary granules of neutrophils and is secreted to mediate matrix degradation and facilitate intravasation.^103,104 Although surface-HS on neutrophils seems to play a minor role in migration as compared with endothelial HS, HS on neutrophils facilitates their activation through binding to cytokines such as IL-8.¹⁶ 

Eosinophil chemotaxis also heavily relies on HS, in part through the interaction of HS with the eotaxin-1 (CCL11) and eotaxin-3 (CCL26) ligands (Table 1).^32,34 Eosinophil major basic protein, the predominant constituent of the crystalline core of eosinophilic granules and mediator of histamine release and protection against helminths and bacteria, not only binds HS but is also a natural inhibitor of heparanase activity (Table 1).^105-107 

Although basophils express HS on their surface and their granules contain an abundance of heparin, little is known about the function of HS and HSPGs in these cells. Circulating basophils have been shown to capture viruses such as HIV-1 partially through HS.¹⁰⁸ 

As with other granule-containing cells, serglycin is an essential component of mast cell integrity and storage and is the predominant HSPG expressed in mast cells.⁶⁷ 

Multiple pieces of evidence suggest that specific HS modifications are needed for proper mast cell function. Deletion of Ndst2 in mice, for example, results in mice harboring mast cells with minimal amounts of granules, which themselves lack heparin and decreased levels of histamine and mast cell proteases.^82,109 Furthermore, fetal skin-derived mast cells from mice deficient in 6-O-sulfation ability (Hs6st1^–/–, Hs6st2^–/–, or Hs6st1^–/–Hs6st2^–/–) show decreased protease activity and quantity of specific mast cell proteases (tryptase and carboxypeptidase A, but not chymase) at the protein level, suggesting that 6-O-sulfated HS prevents degradation of specific proteases.⁸⁵ Mast cells also contain and release heparanase.¹¹⁰ 

Primary human macrophages have been shown to shed SDC1 and SDC4 from their cell surface, a process accelerated by stromal derived factor-1 stimulation and mediated by matrix-metalloprotease 9 (MMP-9).¹¹¹ However, while still on the cell surface, these proteoglycans have been shown to bind to the RANTES (regulated upon activation, normal T-cell expressed and presumably secreted) ligand to facilitate signaling.¹¹² SDC2 is also expressed on the surface of macrophages but does not bind to RANTES. Instead, SDC2 regulates FGF2 signaling and the binding of other cytokines to facilitate macrophage–derived growth factor function.¹¹³ 

Agrin proteoglycan plays a key role in monocyte/macrophage lineage commitment through cell-autonomous interaction with the α-dystroglycan receptor and is required for monocyte/macrophage lineage development.⁶⁵ Serglycin is also secreted by macrophages and plays a role in tumor necrosis factor alpha (TNF-α) secretion in response to lipopolysaccharide.¹¹⁴ 

The polarization of macrophages to either a proinflammatory or adaptive state results in drastic gene expression changes of proteoglycans and the HS machinery.¹¹⁵ Inflammatory macrophages induce the expression of CD44v3, which interacts with FGF2 to facilitate the immune response.¹¹⁶ 

Loss of N-sulfation in tissue-resident macrophages of soft tissues has been shown to promote atherosclerosis and obesity, among other biological processes.¹¹⁷ 

HS is present in cells of the erythroid lineage and has been shown to play many roles in the biology of erythropoiesis and red blood cell function. Early work on human cell lines identified the presence of HSPGs on the surface of erythroid progenitors.¹¹⁸ TGFBR3 (betaglycan) proteoglycan is expressed during erythropoiesis and distinguishes early and late burst-forming unit erythroid progenitors (BFU-Es).¹¹⁹ The expression of TGFBR3 (and the HS attached to it) likely modulates erythroid development by impacting local TGF-β concentrations. Agrin has been identified on the surface of erythroid cells and induces clustering of EphB1 receptors on erythroblasts. Moreover, Agrin-deficient mice display severe anemia and poorly developed erythroblastic islands among other hematopoietic findings (Table 2),⁶⁶ suggesting a role in erythroid maturation.

Similarly, mice deficient in HS6ST1 (which catalyzes 6-O-sulfation) are embryonically lethal, and only scarce numbers of embryonic-derived erythroid cells are present in placental tissue sections, indicating the involvement of 6-O-sulfated HS in erythroid production during embryogenesis (Table 2).⁸⁶ In both mice and humans, erythroid cells display dynamic and temporal expression of specifically modified HS during terminal differentiation, implying that the levels of specific surface HS modifications (including 6-O-sulfation) vary depending on the stage of erythroid differentiation, which may be of functional significance (Figure 3).⁹⁸ 

Finally, HS on the surface of erythrocytes has also been shown to bind and mediate the entry of Plasmodium falciparum.^120,121 

Lysates from human megakaryocytes and megakaryocytic cell lines have been found to contain serglycin and betaglycan; however, the specific roles of these molecules in megakaryocyte biology are yet to be defined.¹²² Mass spectrometry analysis of platelet extracts identified betaglycan, serglycin, and SDC1, suggesting that proteoglycans produced in megakaryocytes are packaged into platelets.¹²³ Serglycin-deficient mice were found to have normal megakaryocyte function; however, emperipolesis of neutrophils at all stages of maturing megakaryocytes was observed in serglycin-deficient (SG^–/–) mice but not in wild-type mice.⁶⁸ Although platelet counts and the number and histological morphology of granules inside platelets were similar between wild-type and SG^–/– mice, mice lacking serglycin showed significantly diminished platelet activation, aggregation, and secretion.⁶⁸ The diminished platelet content of key platelet factors, such as platelet factor 4, β-thromboglobulin, and platelet–derived growth factor in serglycin-deficient mice is partly due to improper loading into granules.⁶⁸ SDC4 has been detected in platelets and impacts their aggregation.¹²⁴ 

Although studies on HS in the megakaryocytic lineage are limited, it appears that the lack of or decreased surface HS is a characteristic of megakaryocyte differentiation and commitment (Figure 3).^98,125 Consistent with this, transgenic mice overexpressing the heparanase enzyme show increased number of platelets, higher number and percentages of megakaryocytes in tissues, and an increased concentration of plasma thrombopoietin.¹²⁶ However, these effects may be related to systemic heparanase expression and additional tissue-specific experiments are needed. The functional importance of the lack of or diminished expression of HS on the megakaryocyte cell surface, as well as the role of heparanase in megakaryopoiesis, has yet to be elucidated. Notwithstanding, megakaryocytes are dependent on interactions with HS. For example, the G6b-B receptor is vital for megakaryocyte activity and platelet production and has been shown to function by binding with the HS chains of pericellular perlecan as well as exogenous HS.¹²⁷ Numerous studies have also demonstrated the importance of exogenous HS (especially sulfated HS) as a ligand for stimulating megakaryopoiesis and platelet activation in conjunction with canonical activators of megakaryopoiesis, such as thrombopoietin.¹²⁸ 

Heparanase is found in abundance in the lysosomes of platelets, where it is stored and enzymatically activated. Platelet heparanase levels are increased under pathogenic conditions, such as sepsis and cancer, and have been implicated in platelet adhesion and thrombogenicity.¹²⁹ 

HSPGs are expressed on the surface of B-lineage cells and their tight regulation is critical for numerous B-cell functions. Early pioneering work identified the transient and temporal expression of SDC1 (also known as CD138) within populations during B-cell development, providing the foundation for the use of CD138 as a marker for B-cell populations, including mature plasma cells.¹³⁰ B-cell activation has also been shown to induce the expression of CD44v3, whose associated HS chains bind HGF and induce Met phosphorylation.¹³¹ 

HS expressed on the surface of mouse B-cell precursors mediates IL-7–driven B lymphopoiesis, suggesting that HS acts as a binding target for IL-7 and a contributor to B-cell differentiation.¹³² Moreover, the exogenous addition of HS (either soluble or surface coculture expression) enhances IL-7-mediated early B-cell differentiation in mice.¹³² Interestingly, however, early B-cell development in humans is IL-7-independent, which corresponds to a lack of SDC1 expression in early human B cells.¹³³ 

It has also been noted that expression of HS-modifying enzymes are differentially expressed during specific stages of B-cell development, suggesting that unique HS-modification patterns with potential functional relevance exist within different B-cell subtypes.¹³⁴ Indeed, the presence of specific modification patterns, particularly epimerization by the C5-epimerase GLCE, has been shown to be dispensable in early B-cell development but critical in the transition between immature and mature B cells, primarily by the binding of APRIL (a proliferation-inducing ligand) to specifically modified HS (Table 1).⁴⁷ Moreover, specific B-cell subtypes show differential HS-modification patterns. Naïve B cells and antibody-secreting B cells have been shown to express higher levels of N-sulfation than germinal center B cells, and the lack of N-sulfation on germinal center B cells prevents IL-21 binding (Table 1).²³ These collective studies suggest that differential HS-modification patterns exist within unique B-cell subtypes that play cell-type–specific roles in hematopoiesis.

In addition to B-cell development, HS appears to also have roles in B-cell function. Surface HS expression in murine B cells was significantly upregulated upon infection by gammaherpesvirus or betaherpesvirus, as well as with stimulation by multiple potent inducers of inflammation (toll-like receptor ligands, B-cell receptor agonists, etc), an effect mediated by the type I interferon receptor.¹³⁵ 

Fetal thymocytes lacking TGFBR3 exhibit higher apoptosis and a delay in T-cell development ex vivo, implicating betaglycan in the maturation and survival of T cells in the thymus.¹³⁶ Additionally, betaglycan is differentially expressed in various T-cell populations, reduced in regulatory T cells, and is induced upon T-cell receptor activation.¹³⁷ 

The activation of T cells upon antigen presentation occurs at the immunological synapse (the interface between the T cell and antigen-presenting cell) and involves numerous interactions and structural organization. Agrin was found to be differentially expressed in resting and activated primary immune cells and shown to be involved in lipid raft reorganization during T-cell activation.¹³⁸ Agrin functions at the immunologic synapse, similar to that in monocytes and macrophages, and involves its interaction with the α-dystroglycan receptor.¹³⁹ Activated T cells also appear to upregulate the surface expression of CD44v3, suggesting HS-involvement in immune processes.¹⁴⁰ In contrast, it seems that surface SDC4 facilitates the suppression of T-cell activation through its interactions with dendritic cell-associated HS proteoglycan–dependent integrin ligand on antigen-presenting cells.¹⁴¹ 

Cytotoxic (killer) T cells carry toxic granules that facilitate cell destruction during the immune response. Not only is serglycin present in these granules, but it also plays a role in the secretory granule repertoire, regulating its composition.¹⁴² 

Several studies have suggested that the proliferation of T cells is at least partially dependent on surface HS, and proliferation cues through cytokine signaling are likely orchestrated by HS binding. In both in vitro and transplantation studies of Ext1-deficient fetal thymus, T cells were found to be significantly diminished in number, albeit with no perturbation of T-cell differentiation.¹⁴³ IL-2-mediated control of T-cell responses and propagation is controlled by HS-bound IL-2, which localizes to the lymphoid organs.^144,145 Consistent with this are recent findings that show regulatory T cells (FoxP3⁺) access local HS-bound IL-2 stores by secreting heparanase.¹⁴⁶ 

In mice whose T cells lack N-sulfation (through T-cell specific deletion of NDST1 and NDST2), development was not altered (Table 2). However, Ndst-deficient T cells were hyper-responsive to low-level activation, likely due to the release of sequestered IL-2 by surface HS and activation of the IL-2 receptor.⁷⁹ 

T-cell migration into tissues has been found to be heavily mediated by their secretion of heparanase.¹⁴⁷ Extracellular matrix degradation of solid tissues by heparanase most likely facilitates T-cell migration.

The interaction between surface HS and T-cell-infecting viruses, including HIV-1, human T-lymphotropic virus 1 (HTLV-1), and severe acute respiratory syndrome coronavirus 2 (SARS-COV-2), is essential for efficient viral infection.^148-150 In the case of HIV-1, it has been shown that a conserved arginine in the gp120 protein specifically interacts with 6-O-sulfated HS to facilitate viral entry.¹⁴⁸ In addition, HIV-1 Tat protein has been shown to bind HS in a size-dependent manner and requires 2-O-, 6-O-, and N-sulfation for binding.¹⁵¹ HS also binds to HTLV-1 in CD4⁺ T cells, and desulfation of HS on the surface of T cells drastically reduces infectivity, suggesting that certain sulfation patterns may also be required for HTLV-1 entry.¹⁵⁰ Recently, it was discovered that SARS-COV-2 can infect T lymphocytes in an ACE2-independent manner, suggesting alternative receptors for SARS-COV-2 on the surface of T cells, such as HS.¹⁵² Given the known interactions between SARS-COV-2 and HS in other cell types, it can be speculated that HS on the surface of T cells may also play a role in SARS-COV-2 infection.¹⁴⁹ 

Similar to cytotoxic T cells and platelets, the secretory granule repertoire of NK cells also depends on the presence of serglycin.¹⁴² 

NK cells execute their function through germ line-encoded activating and inhibitory surface receptors, including natural cytotoxicity receptors (NCRs), killer-cell lectin-like receptors, and killer-cell immunoglobulin-like receptors. HS not only plays a key role on the surface of NK-cell targets by binding to NCRs, such as NKp30, NKp44, and NKp46^153,154 but is also present on the surface of NK cells themselves. Importantly, fine specification of HS modifications on target cells is needed for NK-cell identification and activation. Interestingly, in-cis interactions between HS on the NK-cell surface and NCRs can act to dampen the trigger for NK-cell activation.¹⁵⁵ 

The orchestrated removal of HS has important implications for tissue homeostasis and function. Heparanase expression and secretion by neutrophils, platelets, T cells, mast cells, and macrophages have been reported.^{110,147,156-159} Heparanase secretion is required for macrophage activation as part of the inflammation process, both in normal immune function (likely through modulation of toll-like receptor function) and mediation of protumorigenic and prometastatic processes in cancer.^156,160 NK cells significantly upregulate heparanase upon activation, which plays a key role in NK-cell invasion into tumors and thereby in tumor progression and metastases.¹⁶¹ Regulatory T cells access the local IL-2 needed for their survival and activity through the secretion of heparanase.¹⁴⁶ As previously mentioned, heparanase expression may play a role in megakaryopoiesis and platelet maturation.¹²⁶ HSC mobilization and engraftment have been shown to be impacted by the lack of HS in the stroma by either the absence of Ext1 or by overexpression of heparanase.¹⁶² These collective findings highlight the important balance and regulation of HS expression in hematopoietic cells as part of their biological functions.

The functions of key regulatory pathways that are indispensable for hematopoiesis, such as TGF-β, Wnt, bone morphogenic protein (BMP), and interleukin-driven pathways, as well as the accompanying cytokines, are dependent on HS and the proteoglycans they decorate (Table 1). HSPGs expressed in hematopoietic populations serve cell type-specific functions, including directly facilitating differentiation (eg, SDCs in B cells), establishing concentration gradients needed for differentiation (eg, betaglycan in erythroid populations), promoting migration (glypicans in HSPCs), and maintaining intracellular granule integrity (eg, serglycin in granulocytes, T cells, and mast cells). Although there is no doubt that HS is important for the proper function of hematopoietic tissues, only a few knockout mouse models of HS deficiency/defects show drastic hematopoietic phenotypes (Table 2). The embryonic lethality of many HS-related genes may hamper the investigation of the true importance of specific HS/HSPGs in hematopoiesis, and much more work is needed to elucidate these roles, potentially through the generation of inducible or conditional animal models or the development of specific HS-binding blocking agents.

It is becoming increasingly clear that the composition of HS-modification patterns on the surface of cells is unique and plays a role in defining cell-specific functions, including those within the hematopoietic system (Figure 2B). Barriers in the high-resolution detection and characterization of specific HS-modification patterns on cell surfaces, both in vitro and in vivo, have hindered the understanding of how distinct modifications may play detailed roles in cell-cell interactions, cytokine binding, viral entry, and other key biological functions within hematopoietic cells. Technologies such as mass spectrometry are becoming gradually more successful in identifying sequences of HS modifications, and increasingly complex and diverse arrays of HS oligosaccharides with distinct modification patterns are being created to test binding/interactions. Single-chain variable fragment antibodies (scFvs) against specific HS-modification patterns were isolated from a phage display library and have been used to identify the presence of specific HS-modification patterns in tissues.^163-165 These studies revealed astounding specificity of cellular expression of HS-modification patterns, which in some instances appeared to be single-cell specific. Based on these observations, scFvs have recently been applied to the analysis of cells of the hematopoietic lineage. Specifically, they were incorporated into multiparameter flow cytometry in combination with established CD-marker gating schemes to establish cellular glycotypes within hematopoietic cells of mice and humans, providing a platform to characterize systematically HS-modification patterns in individual hematopoietic cell populations.⁹⁸ Furthermore, this portable HS glycotyping system can serve as a tool to isolate living cells based on distinct HS patterns for downstream functional utilization and characterization. Although this can be applied to all tissues across species, the strong reliance on flow cytometry in hematopoietic tissues, such as bone marrow and blood, makes HS glycotyping a tool with immense potential. With near limitless possible combinations of HS-modification patterns, one can speculate that even more specialized and rare subpopulations of hematopoietic cells exist that currently cannot be defined or purified using existing technologies, as suggested by current destructive technologies such as single-cell RNA sequencing (Figure 3A). The detection of HS-modification patterns through glycotyping not only holds potential in further isolating distinct populations but also allows for the interrogation of the importance of HS modifications in hematopoietic functions, including lineage commitment. For example, HS3A8 scFv, which recognizes distinct combinations of HS modifications, not only helps separate MEPs from other progenitor types but also illustrates the dynamic HS-modification patterning throughout megakaryocyte and erythroid differentiation and development (Figure 3B).⁹⁸ Deciphering the composition of HS modifications on the surface of cells through various approaches will play a transformative role in uncovering highly specialized and unique functions that contribute to the biology and pathobiology of hematopoietic and other tissues.

The authors acknowledge the Creative Services Department at the Albert Einstein College of Medicine, specifically Tatyana Harris, for assistance with figure illustrations.

Work conducted in the Bülow and Steidl laboratories is supported by grants from the National Institutes of Health (NIH), National Cancer Institute (U01CA241981 [H.E.B.], U01CA241981, and R35CA253127 [U.S.]), and the NIH, National Institute of Neurological Disorders and Stroke (R01NS125134 and R01NS129992 [H.E.B.]). U.S. holds the Edward P. Evans Endowed Professorship for Myelodysplastic Syndromes at Albert Einstein College of Medicine. The Endowed Professorship was supported by a grant from the Edward P. Evans Foundation.

Contribution: R.T.P., H.E.B., and U.S. wrote the manuscript.

Conflict-of-interest disclosure: The authors disclose the patent application “Antibody-based method to identify, purify, and manipulate cell types and processes” (US patent publication no.: US 2022/0227886 A1). U.S. has received research funding from GlaxoSmithKline, Bayer Healthcare, Aileron Therapeutics, and Novartis; has received compensation for consultancy services and for serving on scientific advisory boards from GlaxoSmithKline, Bayer Healthcare, Celgene, Aileron Therapeutics, Novartis, Stelexis Therapeutics, Pieris Pharmaceuticals, and Trillium Therapeutics; and has equity ownership in and serves on the board of directors of Stelexis Therapeutics.

Correspondence: Ulrich Steidl, Albert Einstein College of Medicine, Chanin Building Room #601-605, 1300 Morris Park Ave, Bronx, NY 10461; email: ulrich.steidl@einsteinmed.edu.