Sage Journals: Discover world-class research

Abstract

The ability of chondroitin/dermatan sulfate (CS/DS) to convey biological information is enriched by the presence of iduronic acid. DS-epimerases 1 and 2 (DS-epi1 and 2), in conjunction with DS-4-O-sulfotransferase 1, are the enzymes responsible for iduronic acid biosynthesis and will be the major focus of this review. CS/DS proteoglycans (CS/DS-PGs) are ubiquitously found in connective tissues, basement membranes, and cell surfaces or are stored intracellularly. Such wide distribution reflects the variety of biological roles in which they are involved, from extracellular matrix organization to regulation of processes such as proliferation, migration, adhesion, and differentiation. They play roles in inflammation, angiogenesis, coagulation, immunity, and wound healing. Such versatility is achieved thanks to their variable composition, both in terms of protein core and the fine structure of the CS/DS chains. Excellent reviews have been published on the collective and individual functions of each CS/DS-PG. This short review presents the biosynthesis and functions of iduronic acid-containing structures, also as revealed by the analysis of the DS-epi1- and 2-deficient mouse models.

Keywords

chondroitin sulfate CS/DS proteoglycans dermatan sulfate dermatan sulfate epimerase 1 and 2 dermatan sulfate 4-O-sulfotransferase 1

Structure and Biosynthesis of Chondroitin/Dermatan Sulfate

Chondroitin (CS) is a long, linear acidic polysaccharide consisting of the repeating disaccharide unit -4GlcAβ1–3GalNAcβ1- (Silbert and Sugumaran 2002). During assembly, dermatan sulfate (DS) can be generated via epimerization of glucuronic acid (GlcA) into iduronic acid (IdoA) by two DS-epimerases (DS-epi1 and DS-epi2). Subsequently, O-sulfation in various positions is carried out by specific sulfotransferases: Four add a sulfate to the 4-OH position of GalNAc (C4ST1–3 and D4ST1), three are specific for sulfation on the 6-OH position of GalNAc (C6ST1/2 and GalNAc4–6ST, the latter adding a 6-O sulfate to an already 4-O–sulfated GalNAc), and a single CS2ST/UST adds a sulfate to the 2-OH position of both GlcA and IdoA. The reactions described above do not go to completion (i.e., only a fraction of available substrate is converted by a specific enzyme). Highly variable structures are therefore present in CS/DS chains (Fig. 1). In this review, we will indicate the CS/DS structures with the nomenclature O, C, A, B, D, E and iO, iC, iA, iB, iD, iE for the iduronic-containing counterpart, respectively. However, the disaccharide structures depicted in Fig. 1 are just the building blocks of longer protein binding domains, which are usually five to eight monosaccharides long. The exact sequencing of these protein binding domains is still a major challenge and has only been successfully completed in some instances (Li F et al. 2010; Purushothaman et al. 2012). Noteworthy, the disulfated structures B/iB, D/iD, and E/iE, which usually constitute no more than 10% of the chains, often stand out as being part of protein binding domains in CS/DS. Presumably, their double sulfation favors protein interaction. Moreover, a single CS/DS may contain multiple protein binding domains and potentially have multiple functions. In addition to protein binding, CS/DS chains have self-adhesion properties (Fransson et al. 1982).

Figure 1.

Structures and biosynthetic pathways of chondroitin/dermatan sulfate (CS/DS) disaccharide units. Studies on the disaccharide composition of CS/DS chains and enzyme specificities are the basis of this scheme. In iduronic acid (IdoA)-containing units, epimerization always occurs before sulfation. In disulfated structures, 2-O-sulfation follows both 4-O and 6-O sulfation, whereas 4-O sulfation precedes 6-O sulfation when a 4S, 6-di-O–sulfated unit is generated. Dotted lines indicate putative pathways because iC and iD unit formation has not been clarified yet.

The conversion of GlcA into IdoA is variable and can range from one IdoA residue per chain to almost 100% IdoA. The term CS/DS indicates the glucuronic/iduronic acid hybrid nature of the galactosaminoglycan chain. IdoA can be found in blocks (stretch of ≥6 IdoA residues), in alternating IdoA/GlcA structures, or as isolated IdoA interspersed in unmodified GlcA residues (Fig. 2) (Malmstrom et al. 1975; Maccarana et al. 2009). High content of DS epimerases, especially of DS-epi1 in vivo, and a concomitant high content of the DS-specific 4-O-sulfotransferase D4ST1 are required for formation of IdoA blocks (Maccarana et al. 2009; Pacheco, Maccarana, and Malmstrom 2009; Pacheco, Malmstrom, and Maccarana 2009). Indeed, DS-epi1 and D4ST1 are co-localized in the Golgi apparatus, as seen by confocal staining (unpublished observation). The distribution of IdoA governs some of the subsequent O-sulfation reactions. For instance, the IdoA blocks are never found containing 6-O-sulfated GalNAc and are instead 4-O sulfated and subsequently good substrates for the 2-O sulfation reaction, giving rise to consecutive iB structures. Downregulation of DS-epi1, DS-epi2, or D4ST1 all resulted in a reduced amount of iduronic acid blocks and iB residues. In addition, downregulation of D4ST1 using siRNA decreased the E/iE structures. The amount and distribution of IdoA within a single chain are cell/tissue specific (Cheng et al. 1994). Furthermore, a cell can produce different CS/DS chains depending on the core protein. For example, human skin fibroblasts produce decorin/biglycan with 60% IdoA, mostly present in blocks, and versican with 7% IdoA, mostly as isolated residues (Pacheco, Malmstrom, and Maccarana et al. 2009). Both the amount and distribution of IdoA are subjected to physiological regulation—for example, transforming growth factor (TGF)-β1 considerably decreases IdoA blocks in decorin and biglycan produced by fibroblasts (Tiedemann et al. 2005).

Figure 2.

Hybrid structure of chondroitin/dermatan sulfate (CS/DS) and distribution of iduronic acid (IdoA). In vivo, IdoA is commonly found in clusters (IdoA blocks) or as isolated or alternating glucuronic acid (GlcA)/IdoA residues. High expression of DS-epimerases, in close collaboration with the dermatan-specific 4-O-sulfotransferase 1 (D4ST1), is necessary for IdoA block formation.

DS-Epimerase 1 and 2 Structure and Catalytic Mechanism

Only three groups of enzymes catalyze the stereochemical inversion of the C5 carboxyl group of a hexuronic acid at the polymer level (i.e., the conversion takes place after the monosaccharide has been incorporated in the polysaccharide chain). In vertebrates, two DS epimerases and a single HS epimerase convert GlcA into IdoA, whereas in algae and some bacteria, alginate epimerases convert mannuronic into guluronic acid (Valla et al. 2001). No primary sequence or three-dimensional (3D) similarities have been detected between the DS-epimerases and the HS epimerase, which therefore seem to be the result of convergent evolution. DS-epi1 is coded by the DSE gene on chromosome 6, whereas DS-epi2 is coded by the DSEL(-like) gene on chromosome 18 (Maccarana et al. 2006). Interestingly, in DSE, the protein coding sequence is divided into five exons, whereas in DSEL, a single exon contains all the protein coding sequence. The two enzymes show apparent domain similarities (Fig. 3). Both share an N-terminus epimerase domain (51% amino acid identity, identical secondary and 3D predicted structure). DS-epi1 has a C-terminal domain (amino acids 691–958), terminating with two membrane-spanning domains, whose function is currently unknown. No homolog of known structure has been found in 3D databases for this particular domain. Likewise, DS-epi2 has a central domain (amino acids 720–823, including the two transmembrane regions) with unknown function. These domains in DS-epi1 and 2 do not share sequence similarities.

Figure 3.

Three-dimensional structure of the DS-epi1 epimerase domain (above) and domain structure of DS-epimerases (below). A tetrasaccharide substrate is positioned in the groove formed by the two subdomains. The four N-glycosylation sites are indicated with arrows and labeled N1 to N4. The locations of signal peptides (SPs) and predicted transmembrane regions (TMs) are shown with asterisks.

Interestingly, to get enzymatic activity, the epimerase domain alone is not sufficient. In fact, recombinant 1–755-DS-epi1 is inactive, whereas 1–775-DS-epi1 is active (E. Tykesson, personal communication 2012). This is mirrored by the fact that 1–734-DS-epi2 is inactive, whereas the full size 1–1222-DS-epi2 is active (Pacheco, Malmstrom, and Maccarana 2009). In addition, DS-epi2 has a C-terminal sulfotransferase domain homologous to several CS/DS O-sulfotransferases. Recombinant DS-epi2 indeed has detectable sulfotransferase activity (unpublished results), but the position of the added sulfate has yet to be defined. The epimerase domains of both enzymes carry four conserved N-glycans, which are required for activity (Pacheco et al. 2009). Based on structural predictions, the epimerase domain of DS-epi1 can be further subdivided into an N-terminal domain composed of alpha-helices (amino acids 1–390; SCOP superfamily a.102.2) and a C-terminally located domain predominantly composed of beta-sheets (amino acids 391–690; SCOP superfamily b.30.5). Together, the two domains form a groove where the substrate is positioned during catalysis. Several HA-, HS-, CS-, and alginate-lyases show structural similarities to these two DS-epi1 domains. Modeling of DS-epi1 into the structure of heparinase II, as determined by crystal X-ray diffraction, showed conservation and striking similarities of the position of key amino acids shown to be essential in the catalytic mechanism of heparinase II. The importance of these amino acids has been validated through site-directed mutagenesis in recombinant DS-epi1. From these data, a mechanism of epimerization has been proposed where the C5 stereochemical inversion of the carboxyl group is achieved by the removal of the C5 proton from one side of the sugar followed by the readdition of another proton from the opposite side of the sugar plane. Four steps have been detailed: (1) His-450 removes the C5 proton from GlcA, and (2) the glycosidic linkage between the attacked GlcA and the GalNAc at the non-reducing side is broken by Tyr-261 with the subsequent formation of a C4–C5 double bond containing intermediate hexuronic acid. In (3), the intermediate hexuronic acid is protonated from the opposite side by His-205, and in (4), the glycosidic linkage is re-created by an unidentified residue. Steps 1 and 2 are in common with the mechanism of lyases, whereas subsequent steps 3 and 4 are DS-epimerase specific. In summary, due to structural similarities and catalytic mechanism, the two DS epimerases appear to have evolved from bacterial lyases.

Dermatan Sulfate Epimerase 1- and 2-Deficient Mice

DS-epi1- and 2-deficient mouse models have been generated, and the double DS-epi1/2 KO mouse, which should be a DS-null model, is under way (Maccarana et al. 2009; Bartolini et al. 2012). DS-epi1 and -2 are ubiquitously expressed in the body (Nakao et al. 2000; Goossens et al. 2003). Brain has the lowest expression of DS-epi1 and the highest expression of DS-epi2. DS-epi1 is the major epimerase in vivo and contributes to the major part of enzyme activity in extracts of most tissues. Ablation of DS-epi1 led to an approximate 80% reduction of IdoA in the entire animal. This reduction affected both decorin chains (73% reduction in skin decorin) and versican chains (89% reduction in skin versican). DS-epi1 is mainly responsible for the biosynthesis of the iduronic acid blocks, which decreased 90% in DS-epi1-deficient mice. A remaining single functional allele of DS-epi1 (i.e., in the condition of heterozygosity) results in half of the total protein amount and enzyme activity but interestingly leaves the CS/DS structure unaltered. DS-epi2 is the major epimerase in brain, accounting for approximately 90% of the epimerase activity. These data are in agreement with in situ hybridization expression patterns in the developing mouse brain (Akatsu et al. 2011). Nevertheless, brains in DS-epi2-deficient mice had a consistent, but not a dramatic, 38% reduction of IdoA in CS/DS, which in wild-type mice accounts for only 2% of total hexuronic acid residues. No phenotype in unchallenged DS-epi2-deficient mice could be observed, and brain extracellular matrix (ECM) features were normal. It thus remains to be explained why there are two DS-epimerases.

Although no apparent phenotype could be observed in DS-epi2-deficient mice, the phenotypes observed in DS-epi1-deficient mice are multiple. Mice lacking DS-epi1 having a mixed C57BL6/Sv129 genetic background were vital and fertile (Maccarana et al. 2009). However, in pure C57BL6 background, they died at a late stage of embryonic development and perinatally (unpublished observation), and in a pure NFR genetic background, 21% of them presented a birth defect called gastroschisis, consisting of a protrusion of the abdominal content due to a defect in the abdominal wall closure (R. Gustafsson, personal communication 2012). These mice are smaller, with a lower body weight at birth compared with wild-type littermates, and this phenotype persists in adulthood. This condition becomes aggravated when DS-epi2 is also inactivated (unpublished results). DS-epi1-deficient mice have a defect in skin collagen fibril formation, leading to skin fragility. This finding links the presence of high IdoA content in skin decorin CS/DS with its capacity to modulate collagen fibrils and deepens the previous knowledge implicating the decorin proteoglycan (Danielson et al. 1997) and its CS/DS chain (Ruhland et al. 2007) in this function. Also, iB structures, which decrease 70% in DS-epi1-deficient skin decorin, could take part in this function. Interactions between antiparallel CS/DS chains carried by two decorins bound to adjacent collagen fibrils could be altered in DS-epi1-deficient skin (Scott 1991). Another phenotype observed in DS-epi1-deficient mice is the alteration of the epidermal basement membrane, as seen by electron microscopy (S. Kalamajski, personal communication 2012). There is an abnormal distance between the basement membrane, which appears more convoluted, and the keratinocyte cellular membrane. This phenotype is reminiscent of alterations observed in the dermal-epidermal basement membrane of double nidogen-1 and 2-deficient mice (Mokkapati et al. 2008). Possibly linked to the altered basement membrane, the epidermal basal layer is also thicker (R. Gustafsson, personal communication 2012).

Genetic Diseases Related to DS-Specific Enzymes

A recently discovered genetic defect in the D4ST1 gene has been identified in 20 patients with a new type of Ehlers–Danlos syndrome (Zhang et al. 2010). The disorder affects various connective tissues and is the result of impaired assembly of collagen fibrils. The main features are joint dislocations and deformities, skin hyperextensibility and fragility, and recurrent large subcutaneous hematomas. The mutations lead to a loss of D4ST1 activity, and decorin produced by affected fibroblasts was devoid of IdoA (Miyake et al. 2010). Not surprisingly, given the fact that D4ST1 and the DS-epimerases are functionally closely linked in the biosynthesis of the IdoA blocks, the DS-epi1-deficient mouse presents similar defects in collagen fibril formation.

Also DS-epi2 could be the cause of a human disease. In fact, single-nucleotide polymorphism variants of DS-epi2 could be detected at very low frequency in a group of patients with type II bipolar disorder but not in the control group (Goossens et al. 2003).

Functions of Iduronic Acid in CS/DS Proteoglycans

Table 1 lists 29 CS/DS-PGs, 19 of which have not been analyzed in detail for the presence of IdoA. For instance, in studies of CS/DS in the nervous system, most research has analyzed only total CS/DS extracted from whole tissue and not from individual PGs. Further work comprising single PG purification and identification with specific antibodies followed by analysis of its chains with multiple chondroitinase ABC, AC, and B digestions is required.

Table 1.

CS/DS-PGs

	PG^a	IdoA	Functions of PG	CS/DS-bound Proteins Functions of CS/DS (CS/DS Units Involved)^b	Reference^c
BM or BM- connected	Perlecan	NA	BM stability, embryogenesis, FGF-2, TGF-β, PDGF regulator	NA	Couchman et al. 1996
	Bamacan	NA^d	BM stability, morphogenesis	NA	Iozzo and Clark 1986
	Leprecan	NA	BM stability, collagen assembly	NA	Wassenhove-McCarthy and McCarthy 1999
	Collagen XV	NA	BM organization	NA	Li D et al. 2000
Nervous system	Neurocan	NA	Neurite outgrowth, growth factors binding, brain ECM organization	N-CAM, HB-GAM, amphoterin	Rauch et al. 1991
	Phosphacan	+	Neurite outgrowth, growth factors binding, brain ECM organization and differentiation	HB-GAM, amphoterin, midkine (B, E)	Faissner et al. 1994
	Brevican	NA	Neurite outgrowth modulation, cell adhesion and motility, brain ECM maintenance	Neuritogenic activity (E)	Yamada H et al. 1994
	Appican (AβPP isofom)	NA	Neuronal cell adhesion and migration, neurite outgrowth	Midkine, pleiotrophin	Shioi et al. 1995
	Neuroglycan C	NA	Cerebral development, neuritogenesis	NA	Shuo et al. 2004
Extracellular space	Aggrecan	–	Cartilage weight bearing	Water retention (ratio C/A)	Kokenyesi and Silbert 1997
	Versican	+	Cell adhesion, migration, proliferation, differentiation	FGF family, L- and P-selectin, chemokines (iE)	Kawashima et al. 2000; Maccarana et al. 2009
	Decorin	+	Differentiation, collagen fibrillogenesis, proliferation and wound healing, migration, coagulation, growth factors binding	FGF-2, FGF-7 (iA), HGF/SF (iA), HCII (iB),VWF (A/iA, C/iC, E/iE), α2β1, tenascin-X, fibril formation (iA blocks)	Deakin et al. 2009; Elefteriou et al. 2001; Guidetti et al. 2004; Maccarana et al. 2009; Tollefsen 2010
	Biglycan	+	ECM assembly, differentiation, collagen fibrillogenesis, growth factors binding	HCII (iB), FGF family (iA)	Hou et al. 2007; Stocker et al. 1991
	Epiphycan	+	Chondrocyte differentiation, cartilage stability	NA	Shinomura et al. 1983
Cell surface	NG2	NA	Differentiation, proliferation and motility, PDGF-AA and FGF-2 binding, adhesion	NA	Stallcup 2002
	Betaglycan	NA	TGF-β co-receptor, growth factor reservoir	NA	Andres et al. 1991
	Syndecan-1	+	Adhesion, growth factors co-receptor	Midkine, pleiotrophin, FGF-7	Lee et al. 2004
	Syndecan-3	NA	Adhesion, growth factors co-receptor, neurite outgrowth	NA	Nakanishi et al. 1997
	Syndecan-4	NA	Adhesion, growth factors co-receptor	Midkine, pleiotrophin	Deepa et al. 2004
	CD44	+	Adhesion, migration	Migration	Clark et al. 2004
	Neuropilin-1	NA	Axonal guidance, angiogenesis	VEGF signaling	Shintani et al. 2006
	α5β1	NA	Fibronectin-dependent cell migration, focal adhesion	NA	Franco et al. 2009; Veiga et al. 1997
Others	SRPX2	+	Cell signaling modulation, endothelial cell migration and angiogenesis	HGF	Tanaka et al. 2012
	Serglycin	NA	Inflammatory reaction, storage of chemokines, cytokines and proteases, mast cell secretion	Maturation of granules (B, E)	Kolset and Pejler 2011
	Endocan	+	Mitogenic regulator, inflammation	HGF/SF (iA)	Bechard et al. 2001; Sarrazin et al. 2010
	Testican-1	NA	Inhibition of proteases, neurite extension	NA	Bonnet et al. 1992; Edgell et al. 2004
Collagens	Collagen IX	NA	Biomechanical stability of cartilage	NA	Bruckner et al. 1985
Collagens	Collagen XII and XIV	NA	Stability of cartilage and skin	NA	Watt et al. 1992

BM, basement membrane; CS/DS, chondroitin/dermatan sulfate; ECM, extracellular matrix; FGF, fibroblast growth factor; HB-GAM, heparin-binding growth-associated molecule; HCII, heparin cofactor II; HGF/SF, hepatocyte growth factor/scatter factor; IdoA, iduronic acid; NA, not analyzed; N-CAM, neural cell adhesion molecule; PDGF, platelet-derived growth factor; PG, proteoglycan; TGF-β, transforming growth factor–β; VEGF, vascular endothelial growth factor; VWF, von Willebrand factor.

Some are part-time CS/DS-PGs because they can be unglycanated or carry HS chains. The division in groups is arbitrary because some CS/DS-PGs can belong to more than one group.

For the symbols of the units, refer to Figure 1.

Each PG has a reference reporting the experimental evidence of CS or eventually the CS/DS chain.

Indirect evidence was reported.

One classic way through which CS/DS-PGs can regulate biological processes at or near the cell surface is their ability to bind several growth factors. The flexibility of iduronic acid, which can be found as ¹C₄ and ⁴C₁ chairs and ²S₀ skew-boat conformers as opposed to GlcA found only in ⁴C₁ chair conformation (Ferro et al. 1990), is thought to facilitate protein binding. For instance, hepatocyte growth factor/scatter factor (HGF/SC) requires IdoA residue flanked by sulfated hexosamine (Deakin et al. 2009). IdoA-containing domains have also been shown to interact with the fibroblast growth factor (FGF) family, thereby regulating wound healing and cell migration (Trowbridge and Gallo 2002).

Basement membrane CS/DS-PGs mainly regulate the architecture of the structure they are part of. There is no clear evidence of IdoA presence in these PGs.

In brain, the abundant CS/DS-PGs regulate plasticity during development and regeneration after injury. Most of the described CS/DS-PGs interact with specific brain molecules of the tenascin family, midkine, pleiotrophin, and neural cell adhesion molecules, in addition to growth factors (Yamada and Sugahara 2008).

The representative CS/DS-PG secreted in the extracellular space is decorin. It regulates collagen fibril formation and acts as a growth factor reservoir and as an anticoagulant factor, thanks to the ability of its CS/DS chain to interact with von Willebrand factor (VWF) (Guidetti et al. 2004) and with heparin cofactor II (HCII), the latter via iB structures (Tollefsen 2010). Decorin is therefore believed to be part of a complex acting at the subendothelial matrix regulating the clotting cascade and platelet activation.

The cell membrane CS/DS-PGs are co-receptors for several growth factors, and because of their transmembrane localization, they are involved in adhesion to substrates, migration, and intracellular signaling by connecting the outside environment with intracellular responses. IdoA presence is documented for CD44 after PDGF stimulation (Clark et al. 2004) and for syndecan-1 produced by fibroblasts grown on a 3D collagen I gel and during in vivo wound healing (Lee et al. 2004).

Endocan and SRPX2 have been recently described as IdoA-containing PGs. They show different functions, although they have been both connected with tumor progression. Endocan is a soluble PG bearing a single CS/DS chain expressed by vascular endothelium and released in the bloodstream. It is capable of binding HGF/SF via the CS/DS chain (Bechard et al. 2001), regulating the mitogenic activity of cells. In a recent work, its chains have been analyzed both by NMR and by chondroitinase ACI split products, showing 30% of IdoA present as isolated or two to four consecutive residues (Sarrazin et al. 2010). SRPX2 is upregulated in gastric cancer cells and affects cell adhesion, migration, and angiogenesis, and its IdoA content is evident given the partial digestion with chondroitinase B (Tanaka et al. 2012).

Finally, some members of the collagen family can also be decorated with CS/DS chains, although so far they are described as CS-PGs since a proper analysis for IdoA content is lacking. Collagen XV connects the basement membrane with underlying structures, and skin collagen XII and XIV have been described in specialized sites called anchoring plaques that stabilize the cohesion between the upper dermis and the basement membrane zone (Agarwal et al. 2012).

Dermatan Sulfate and Cancer

CS/DS proteoglycans have been extensively studied in cancer (Wegrowski and Maquart 2006). Focus in the past years has been on the role of the cancer stroma whose structure affects proliferation and migration of tumor cells (Mueller and Fusenig 2004). There are several mechanisms through which CS/DS could influence tumor biology, among which are cell proliferation and migration via binding of growth factors (Villena and Brandan 2004; Taylor et al. 2005; Radek et al. 2009) and hiding cancer cells from T-cell cytotoxic surveillance in blood through a selectin-mediated coating by platelets (Kozlowski et al. 2011), to mention a few. Cell surface CS/DS-PGs on human metastatic breast cancer cell lines have been suggested as receptors for P-selectin (Monzavi-Karbassi et al. 2007). Furthermore, removal of CS/DS chains by chondroitinase B has been shown to inhibit endothelial and melanoma proliferation and invasion (Denholm et al. 2001). DS-epi1 has been found upregulated in almost all cancers examined (Nakao et al. 2000). In a recent study, the hypothesis that IdoA-carrying proteoglycans affect tumorigenesis was directly tested (Thelin et al. 2012). DS-epi1 was downregulated in the esophageal cancer cell line TE-1. The consequent 80% to 90% reduction in IdoA in CS/DS chains hampered cellular invasion and migration. This effect was pronounced in the presence of hepatocyte growth factor (HGF) and mediated through the c-Met receptor. Moreover, DS-epi1-downregulated TE-1 cells have malfunctioning assembly of adhesion complexes and abnormal cytoskeleton architecture. In conclusion, iduronic acid in esophageal cancer cells is important for invasion, migration, and signaling through the c-Met receptor.

Future Directions

Biosynthesis of CS/DS still awaits the answer to several basic questions: How is the machinery producing structurally diverse chains controlled? Does it exist as a single macromolecular biosynthetic complex, or are there several specialized biosynthetic workstations made up of a few biosynthetic enzymes (e.g., DS-epi1 with D4ST1) to make the iduronic acid blocks? Where in the Golgi does the CS/DS biosynthesis take place? How does the core protein govern the variable structure of CS/DS chains? Which transcriptional/translational (and posttranslational) controls are exerted on the biosynthetic enzymes? Which proteoglycans contain iduronic acid? The answer to the latter question would also help to explain the largely unexplored phenotype of the DS-epi1-deficient mouse and the even more profound effect of the double ablation of DS-epi1 and -2. In fact, these models run longitudinally across the functions of each single CS/DS-PG. Finally, can the biological effects of iduronic acid be pharmacologically modulated by inhibition of DS-specific enzymes? For sure, future research on such a basic life component as CS/DS will surprise curious minds for years to come.

Footnotes

Acknowledgements

We thank Renata Gustafsson, Sebastian Kalamajski, and Emil Tykesson for sharing their unpublished observations; Giacomo Frati for helping with ; and Åke Oldberg for critical reading of the manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants from the Swedish Science Research Council; the Medical Faculty of Lund University; the Greta and Johan Kock, the Albert Österlund, the Anna-Greta Crafoord to [B.B.] foundations; the Mizutani Foundation for Glycoscience to [M.M. and A.M.].

References

Agarwal

Zwolanek

Keene

Schulz

Blumbach

Heinegard

Zaucke

Paulsson

Krieg

Koch

. 2012. Collagen XII and XIV—new partners of cartilage oligomeric matrix protein in the skin extracellular matrix suprastructure. J Biol Chem. 287:22549–22559.

Akatsu

Mizumoto

Kaneiwa

Maccarana

Malmstrom

Yamada

Sugahara

. 2011. Dermatan sulfate epimerase 2 is the predominant isozyme in the formation of the chondroitin sulfate/dermatan sulfate hybrid structure in postnatal developing mouse brain. Glycobiology. 21:565–574.

Andres

Ronnstrand

Cheifetz

Massague

. 1991. Purification of the transforming growth factor–beta (TGF-beta) binding proteoglycan betaglycan. J Biol Chem. 266:23282–23287.

Bartolini

Thelin

Rauch

Feinstein

Oldberg

Malmstrom

Maccarana

. 2012. Mouse development is not obviously affected by the absence of dermatan sulfate epimerase 2 in spite of a modified brain dermatan sulfate composition. Glycobiology. 22:1007–1016.

Bechard

Gentina

Delehedde

Scherpereel

Lyon

Aumercier

Vazeux

Richet

Degand

Jude

. 2001. Endocan is a novel chondroitin sulfate/dermatan sulfate proteoglycan that promotes hepatocyte growth factor/scatter factor mitogenic activity. J Biol Chem. 276:48341–48349.

Bonnet

Perin

Maillet

Jolles

Alliel

. 1992. Characterization of a human seminal plasma glycosaminoglycan-bearing polypeptide. Biochem J. 288(Pt 2):565–569.

Bruckner

Vaughan

Winterhalter

. 1985. Type IX collagen from sternal cartilage of chicken embryo contains covalently bound glycosaminoglycans. Proc Natl Acad Sci U S A. 82:2608–2612.

Cheng

Heinegard

Malmstrom

Schmidtchen

Yoshida

Fransson

. 1994. Patterns of uronosyl epimerization and 4-/6-O-sulphation in chondroitin/dermatan sulphate from decorin and biglycan of various bovine tissues. Glycobiology. 4:685–696.

Clark

Lin

Greiling

Couchman

. 2004. Fibroblast invasive migration into fibronectin/fibrin gels requires a previously uncharacterized dermatan sulfate-CD44 proteoglycan. J Invest Dermatol. 122:266–277.

10.

Couchman

Kapoor

Sthanam

. 1996. Perlecan and basement membrane-chondroitin sulfate proteoglycan (bamacan) are two basement membrane chondroitin/dermatan sulfate proteoglycans in the Engelbreth-Holm-Swarm tumor matrix. J Biol Chem. 271:9595–9602.

11.

Danielson

Baribault

Holmes

Graham

Kadler

Iozzo

. 1997. Targeted disruption of decorin leads to abnormal collagen fibril morphology and skin fragility. J Cell Biol. 136:729–743.

12.

Deakin

Blaum

Gallagher

Uhrin

Lyon

. 2009. The binding properties of minimal oligosaccharides reveal a common heparan sulfate/dermatan sulfate-binding site in hepatocyte growth factor/scatter factor that can accommodate a wide variety of sulfation patterns. J Biol Chem. 284:6311–6321.

13.

Deepa

Yamada

Zako

Goldberger

Sugahara

. 2004. Chondroitin sulfate chains on syndecan-1 and syndecan-4 from normal murine mammary gland epithelial cells are structurally and functionally distinct and cooperate with heparan sulfate chains to bind growth factors: a novel function to control binding of midkine, pleiotrophin, and basic fibroblast growth factor. J Biol Chem. 279:37368–37376.

14.

Denholm

Lin

Silver

. 2001. Anti-tumor activities of chondroitinase AC and chondroitinase B: inhibition of angiogenesis, proliferation and invasion. Eur J Pharmacol. 416:213–221.

15.

Edgell

BaSalamah

Marr

. 2004. Testican-1: a differentially expressed proteoglycan with protease inhibiting activities. Int Rev Cytol. 236:101–122.

16.

Elefteriou

Exposito

Garrone

Lethias

. 2001. Binding of tenascin-X to decorin. FEBS Lett. 495:44–47.

17.

Faissner

Clement

Lochter

Streit

Mandl

Schachner

. 1994. Isolation of a neural chondroitin sulfate proteoglycan with neurite outgrowth promoting properties. J Cell Biol. 126:783–799.

18.

Ferro

Provasoli

Ragazzi

Casu

Torri

Bossennec

Perly

Sinay

Petitou

Choay

. 1990. Conformer populations of L-iduronic acid residues in glycosaminoglycan sequences. Carbohydr Res. 195:157–167.

19.

Franco

Trindade

Rocha

da Silveira

Paludo

Chammas

Veiga

Nader

Dietrich

. 2009. Glycosaminoglycan chains from alpha5beta1 integrin are involved in fibronectin-dependent cell migration. Biochem Cell Biol. 87:677–686.

20.

Fransson

Coster

Malmstrom

Sheehan

. 1982. Self-association of scleral proteodermatan sulfate: evidence for interaction via the dermatan sulfate side chains. J Biol Chem. 257:6333–6338.

21.

Goossens

Van Gestel

Claes

De Rijk

Souery

Massat

Van den Bossche

Backhovens

Mendlewicz

Van Broeckhoven

. 2003. A novel CpG-associated brain-expressed candidate gene for chromosome 18q-linked bipolar disorder. Mol Psychiatry. 8:83–89.

22.

Guidetti

Bartolini

Bernardi

Tira

Berndt

Balduini

Torti

. 2004. Binding of von Willebrand factor to the small proteoglycan decorin. FEBS Lett. 574:95–100.

23.

Hou

Maccarana

Min

Strate

Pera

. 2007. The secreted serine protease xHtrA1 stimulates long-range FGF signaling in the early Xenopus embryo. Dev Cell. 13:226–241.

24.

Iozzo

Clark

. 1986. Biosynthesis of proteoglycans by rat embryo parietal yolk sacs in organ culture. J Biol Chem. 261:6658–6669.

25.

Kawashima

Hirose

Nagakubo

Plaas

Miyasaka

. 2000. Binding of a large chondroitin sulfate/dermatan sulfate proteoglycan, versican, to L-selectin, P-selectin, and CD44. J Biol Chem. 275:35448–35456.

26.

Kokenyesi

Silbert

. 1997. Immortalized, cloned mouse chondrocytic cells (MC615) produce three different matrix proteoglycans with core-protein-specific chondroitin/dermatan sulphate structures. Biochem J. 327(Pt 3):831–839.

27.

Kolset

Pejler

. 2011. Serglycin: a structural and functional chameleon with wide impact on immune cells. J Immunol. 187:4927–4933.

28.

Kozlowski

Pavao

Borsig

. 2011. Ascidian dermatan sulfates attenuate metastasis, inflammation and thrombosis by inhibition of P-selectin. J Thromb Haemost. 9:1807–1815.

29.

Lee

Trowbridge

Taylor

Morhenn

Gallo

. 2004. Dermatan sulfate proteoglycan and glycosaminoglycan synthesis is induced in fibroblasts by transfer to a three-dimensional extracellular environment. J Biol Chem. 279:48640–48646.

30.

Clark

Myers

. 2000. Basement membrane zone type XV collagen is a disulfide-bonded chondroitin sulfate proteoglycan in human tissues and cultured cells. J Biol Chem. 275:22339–22347.

31.

Nandini

Hattori

Bao

Murayama

Nakamura

Fukushima

Sugahara

. 2010. Structure of pleiotrophin- and hepatocyte growth factor-binding sulfated hexasaccharide determined by biochemical and computational approaches. J Biol Chem. 285:27673–27685.

32.

Maccarana

Kalamajski

Kongsgaard

Magnusson

Oldberg

Malmstrom

. 2009. Dermatan sulfate epimerase 1–deficient mice have reduced content and changed distribution of iduronic acids in dermatan sulfate and an altered collagen structure in skin. Mol Cell Biol. 29:5517–5528.

33.

Maccarana

Olander

Malmstrom

Tiedemann

Aebersold

Lindahl

Malmstrom

. 2006. Biosynthesis of dermatan sulfate: chondroitin-glucuronate C5-epimerase is identical to SART2. J Biol Chem. 281:11560–11568.

34.

Malmstrom

Carlstedt

Aberg

Fransson

. 1975. The copolymeric structure of dermatan sulphate produced by cultured human fibroblasts: different distribution of iduronic acid and glucuronic acid–containing units in soluble and cell-associated glycans. Biochem J. 151:477–489.

35.

Miyake

Kosho

Mizumoto

Furuichi

Hatamochi

Nagashima

Arai

Takahashi

Kawamura

Wakui

. 2010. Loss-of-function mutations of CHST14 in a new type of Ehlers-Danlos syndrome. Hum Mutat. 31:966–974.

36.

Mokkapati

Baranowsky

Mirancea

Smyth

Breitkreutz

Nischt

. 2008. Basement membranes in skin are differently affected by lack of nidogen 1 and 2. J Invest Dermatol. 128:2259–2267.

37.

Monzavi-Karbassi

Stanley

Hennings

Jousheghany

Artaud

Shaaf

Kieber-Emmons

. 2007. Chondroitin sulfate glycosaminoglycans as major P-selectin ligands on metastatic breast cancer cell lines. Int J Cancer. 120:1179–1191.

38.

Mueller

Fusenig

. 2004. Friends or foes—bipolar effects of the tumour stroma in cancer. Nat Rev Cancer. 4:839–849.

39.

Nakanishi

Kadomatsu

Okamoto

Ichihara-Tanaka

Kojima

Saito

Tomoda

Muramatsu

. 1997. Expression of syndecan-1 and -3 during embryogenesis of the central nervous system in relation to binding with midkine. J Biochem. 121:197–205.

40.

Nakao

Shichijo

Imaizumi

Inoue

Matsunaga

Yamada

Kikuchi

Tsuda

Ohta

Takamori

. 2000. Identification of a gene coding for a new squamous cell carcinoma antigen recognized by the CTL. J Immunol. 164:2565–2574.

41.

Pacheco

Maccarana

Goodlett

Malmstrom

. 2009. Identification of the active site of DS-epimerase 1 and requirement of N-glycosylation for enzyme function. J Biol Chem. 284:1741–1747.

42.

Pacheco

Maccarana

Malmstrom

. 2009. Dermatan 4-O-sulfotransferase 1 is pivotal in the formation of iduronic acid blocks in dermatan sulfate. Glycobiology. 19:1197–1203.

43.

Pacheco

Malmstrom

Maccarana

. 2009. Two dermatan sulfate epimerases form iduronic acid domains in dermatan sulfate. J Biol Chem. 284:9788–9795.

44.

Purushothaman

Sugahara

Faissner

. 2012. Chondroitin sulfate “wobble motifs” modulate maintenance and differentiation of neural stem cells and their progeny. J Biol Chem. 287:2935–2942.

45.

Radek

Taylor

Gallo

. 2009. FGF-10 and specific structural elements of dermatan sulfate size and sulfation promote maximal keratinocyte migration and cellular proliferation. Wound Repair Regen. 17:118–126.

46.

Rauch

Gao

Janetzko

Flaccus

Hilgenberg

Tekotte

Margolis

. 1991. Isolation and characterization of developmentally regulated chondroitin sulfate and chondroitin/keratan sulfate proteoglycans of brain identified with monoclonal antibodies. J Biol Chem. 266:14785–14801.

47.

Ruhland

Schonherr

Robenek

Hansen

Iozzo

Bruckner

Seidler

. 2007. The glycosaminoglycan chain of decorin plays an important role in collagen fibril formation at the early stages of fibrillogenesis. FEBS J. 274:4246–4255.

48.

Sarrazin

Lyon

Deakin

Guerrini

Lassalle

Delehedde

Lortat-Jacob

. 2010. Characterization and binding activity of the chondroitin/dermatan sulfate chain from Endocan, a soluble endothelial proteoglycan. Glycobiology. 20:1380–1388.

49.

Scott

. 1991. Proteoglycan: collagen interactions in connective tissues: ultrastructural, biochemical, functional and evolutionary aspects. Int J Biol Macromol. 13:157–161.

50.

Shinomura

Kimata

Oike

Noro

Hirose

Tanabe

Suzuki

. 1983. The occurrence of three different proteoglycan species in chick embryo cartilage: isolation and characterization of a second proteoglycan (PG-Lb) and its precursor form. J Biol Chem. 258:9314–9322.

51.

Shintani

Takashima

Asano

Kato

Liao

Yamazaki

Tsukamoto

Seguchi

Yamamoto

Fukushima

. 2006. Glycosaminoglycan modification of neuropilin-1 modulates VEGFR2 signaling. EMBO J. 25:3045–3055.

52.

Shioi

Pangalos

Ripellino

Vassilacopoulou

Mytilineou

Margolis

Robakis

. 1995. The Alzheimer amyloid precursor proteoglycan (appican) is present in brain and is produced by astrocytes but not by neurons in primary neural cultures. J Biol Chem. 270:11839–11844.

53.

Shuo

Aono

Matsui

Tokita

Maeda

Shimada

Oohira

. 2004. Developmental changes in the biochemical and immunological characters of the carbohydrate moiety of neuroglycan C, a brain-specific chondroitin sulfate proteoglycan. Glycoconj J. 20:267–278.

54.

Silbert

Sugumaran

. 2002. Biosynthesis of chondroitin/dermatan sulfate. IUBMB Life. 54:177–186.

55.

Stallcup

. 2002. The NG2 proteoglycan: past insights and future prospects. J Neurocytol. 31:423–435.

56.

Stocker

Meyer

Wagener

Greiling

. 1991. Purification and N-terminal amino acid sequence of a chondroitin sulphate/dermatan sulphate proteoglycan isolated from intima/media preparations of human aorta. Biochem J. 274(Pt 2):415–420.

57.

Tanaka

Arao

Tamura

Aomatsu

Furuta

Matsumoto

Kaneda

Kudo

Fujita

Kimura

. 2012. SRPX2 is a novel chondroitin sulfate proteoglycan that is overexpressed in gastrointestinal cancer. PLoS One. 7:e27922.

58.

Taylor

Rudisill

Gallo

. 2005. Structural and sequence motifs in dermatan sulfate for promoting fibroblast growth factor-2 (FGF-2) and FGF-7 activity. J Biol Chem. 280:5300–5306.

59.

Thelin

Svensson

Shi

Bagher

Axelsson

Isinger-Ekstrand

van Kuppevelt

Johansson

Nilbert

Zaia

. 2012. Dermatan sulfate is involved in the tumorigenic properties of esophagus squamous cell carcinoma. Cancer Res. 72:1943–1952.

60.

Tiedemann

Olander

Eklund

Todorova

Bengtsson

Maccarana

Westergren-Thorsson

Malmstrom

. 2005. Regulation of the chondroitin/dermatan fine structure by transforming growth factor-beta1 through effects on polymer-modifying enzymes. Glycobiology. 15:1277–1285.

61.

Tollefsen

. 2010. Vascular dermatan sulfate and heparin cofactor II. Prog Mol Biol Transl Sci. 93:351–372.

62.

Trowbridge

Gallo

. 2002. Dermatan sulfate: new functions from an old glycosaminoglycan. Glycobiology. 12:117R–125R.

63.

Valla

Ertesvag

Barbeyron

Lindahl

. 2001. Hexuronyl C5-epimerases in alginate and glycosaminoglycan biosynthesis. Biochimie. 83:819–830.

64.

Veiga

Elias

Gremski

Porcionatto

da Silva

Nader

Brentani

. 1997. Post-translational modifications of alpha5beta1 integrin by glycosaminoglycan chains: the alpha5beta1 integrin is a facultative proteoglycan. J Biol Chem. 272:12529–12535.

65.

Villena

Brandan

. 2004. Dermatan sulfate exerts an enhanced growth factor response on skeletal muscle satellite cell proliferation and migration. J Cell Physiol. 198:169–178.

66.

Wassenhove-McCarthy

McCarthy

. 1999. Molecular characterization of a novel basement membrane-associated proteoglycan, leprecan. J Biol Chem. 274:25004–25017.

67.

Watt

Lunstrum

McDonough

Keene

Burgeson

Morris

. 1992. Characterization of collagen types XII and XIV from fetal bovine cartilage. J Biol Chem. 267:20093–20099.

68.

Wegrowski

Maquart

. 2006. Chondroitin sulfate proteoglycans in tumor progression. Adv Pharmacol. 53:297–321.

69.

Yamada

Watanabe

Shimonaka

Yamaguchi

. 1994. Molecular cloning of brevican, a novel brain proteoglycan of the aggrecan/versican family. J Biol Chem. 269:10119–10126.

70.

Yamada

Sugahara

. 2008. Potential therapeutic application of chondroitin sulfate/dermatan sulfate. Curr Drug Discov Technol. 5:289–301.

71.

Zhang

Muller

Baenziger

Janecke

. 2010. Congenital disorders of glycosylation with emphasis on loss of dermatan-4-sulfotransferase. Prog Mol Biol Transl Sci. 93:289–307.

Iduronic Acid in Chondroitin/Dermatan Sulfate

Abstract

Keywords

Structure and Biosynthesis of Chondroitin/Dermatan Sulfate

DS-Epimerase 1 and 2 Structure and Catalytic Mechanism

Dermatan Sulfate Epimerase 1- and 2-Deficient Mice

Genetic Diseases Related to DS-Specific Enzymes

Functions of Iduronic Acid in CS/DS Proteoglycans

Dermatan Sulfate and Cancer

Future Directions

Footnotes

Acknowledgements

Declaration of Conflicting Interests

Funding

References