Fine-tuning the structure of glycosaminoglycans in living cells using xylosides

Fine-tuning the structure of glycosaminoglycans in living cells using xylosides Abstract Xylosides can induce the formation and secretion of xyloside-primed glycosaminoglycans when administered to living cells; however, their impact on the detailed glycosaminoglycan structure remains unknown. Here, we have systematically investigated how the xyloside concentration and the type of xyloside, as well as the cell type, influenced the structure of xyloside-primed glycosaminoglycans in terms of the heparan sulfate and chondroitin/dermatan sulfate proportion and disaccharide composition. We found that although greatest influence was exerted by the cell type, both the xyloside concentration and type of xyloside impacted the proportion of heparan sulfate and the complexity of chondroitin/dermatan sulfate. The disaccharide composition of the chondroitin/dermatan sulfate was influenced by the xyloside concentration and type of xyloside to a higher extent than that of the heparan sulfate; the proportion of 4S-sulfated disaccharides in the chondroitin/dermatan sulfate decreased and the proportions of 6S-sulfated and/or nonsulfated disaccharides increased both with increasing concentrations of xyloside and with increasing xyloside hydrophobicity, whereas the proportion of nonsulfated disaccharides was primarily altered in the heparan sulfate with increasing concentrations of xyloside. Our results indicate that it is feasible to not only produce large amounts of glycosaminoglycans in living cells but also to fine-tune their structures by using xylosides of different types and at different concentrations. Chondroitin sulfate, dermatan sulfate, glycosaminoglycan, heparan sulfate, xyloside Introduction Chondroitin/dermatan sulfate (CS/DS) and heparan sulfate (HS) glycosaminoglycans (GAGs) are linear anionic polysaccharides, commonly 25–100 kDa in size, linked to a proteoglycan (PG) core protein (Turnbull et al. 2001; Sugahara and Kitagawa 2002; Mikami and Kitagawa 2013). Their biosynthesis is a nontemplate-driven process, believed to be regulated by the specificity of the enzymes involved and their organization in the Golgi. CS/DS and HS share a linkage region tetrasaccharide comprised of one xylose, two galactoses and one glucuronic acid (GlcA), where xylose is linked to a serine residue of the PG core protein. The polysaccharides are composed of repeating units of N-acetylgalactosamine (GalNAc) or N-acetylglucosamine (GlcNAc) and GlcA or iduronic acid (IdoA); [-3GalNAcβ1-4GlcAβ1-] in CS, [-3GalNAcβ1-4IdoAα1-] in DS and [-4GlcNAcα1-4GlcAβ1/IdoAα1-] in HS. DS is commonly found as a copolymer of CS/DS, where the IdoA content can vary from one single IdoA per chain to almost 100% IdoA (Malmström et al. 2012). The polysaccharides can be modified by O-sulfation at positions 4 and 6 of GalNAc and at position 2 of GlcA or IdoA in CS/DS, and by N-deacetylation/N-sulfation and by O-sulfation at position 2 of GlcA or IdoA, and at position 6 and, more rarely, at position 3 of GlcNAc in HS, thereby rendering GAGs with high structural diversity and complexity. GAGs are produced by virtually all mammalian cells and are, as a part of PGs, primarily located in the extracellular matrix or at the cell surface where they are involved in many biological processes through direct and indirect interactions with various proteins, such as receptors, enzymes and cytokines (Xu and Esko 2014; Mizumoto et al. 2015). However, only a few of these interactions have been defined structurally (Xu and Esko 2014). The interactions between GAGs and proteins may be nonspecific or specific, in which case they may not be limited to one specific sequence in the GAG but rather to several sequences. In addition, the structural characterization of GAGs is extremely difficult due to the high number of theoretically possible structural variants of GAGs and, as yet, not fully understood biosynthesis. Furthermore, there are no standardized methods to screen for GAG–protein interactions, although a few successful attempts have been reported (Gesslbauer et al. 2016; Gray et al. 2017). Large quantities of GAGs with different structures would facilitate such investigations. More than 40 years ago, 4-nitrophenyl β-d-xylopyranoside (pNP-Xyl, Figure 1A) and methylumbelliferyl β-d-xylopyranoside (MU-Xyl, Figure 1B) were reported to induce the formation and secretion of large quantities of GAGs upon administration to living cells and to concomitantly inhibit GAG production on PG core proteins (Okayama et al. 1973; Schwartz et al. 1974). Since then, β-d-xylopyranosides, also called xylosides, comprising a xylose residue linked to an aglycon, have been used to study the effects of altered PG synthesis in different cellular processes, for example, nerve growth during development (Hashemian et al. 2014), hematopoiesis (Spooncer et al. 1983) and morphogenesis and differentiation (Smith et al. 1990). In the 1990s, xylosides with naphthalene-based aglycons, including 2-(6-hydroxynaphthyl) β-d-xylopyranoside (XylNapOH, Figure 1C) and 2-naphthyl β-d-xylopyranoside (XylNap, Figure 1D), were discovered to induce substantial amounts of HS (Lugemwa and Esko 1991; Fritz et al. 1994; Mani et al. 1998), unlike pNP-Xyl and MU-Xyl, which primarily induce synthesis of CS/DS. Fig. 1. View largeDownload slide Xylosides included in the study. The chemical structures of (A) 4-nitrophenyl β-d-xylopyranoside (pNP-Xyl), (B) 4-methylumbelliferyl β-d-xylopyranoside (MU-Xyl), (B) 2-(6-hydroxynaphthyl) β-d-xylopyranoside (XylNapOH), (D) 2-naphthyl β-d-xylopyranoside (XylNap) and (E) 2-naphthyl 1,5-dithio-β-d-xylopyranoside (dithio-XylNap). The figure was prepared using ChemDraw (Perkin Elmer Informatics, Inc. (cambridgesoft.com); version 13.0.0.3015). Fig. 1. View largeDownload slide Xylosides included in the study. The chemical structures of (A) 4-nitrophenyl β-d-xylopyranoside (pNP-Xyl), (B) 4-methylumbelliferyl β-d-xylopyranoside (MU-Xyl), (B) 2-(6-hydroxynaphthyl) β-d-xylopyranoside (XylNapOH), (D) 2-naphthyl β-d-xylopyranoside (XylNap) and (E) 2-naphthyl 1,5-dithio-β-d-xylopyranoside (dithio-XylNap). The figure was prepared using ChemDraw (Perkin Elmer Informatics, Inc. (cambridgesoft.com); version 13.0.0.3015). Fig. 2. View largeDownload slide Amount of recovered PG-derived GAGs and xyloside-primed GAGs from different cell lines. The amount of recovered secreted PG-derived and XylNap-primed GAGs (in nanogram per 106 cells) from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). No GAGs were detected in untreated pgsA-745 cells. Fig. 2. View largeDownload slide Amount of recovered PG-derived GAGs and xyloside-primed GAGs from different cell lines. The amount of recovered secreted PG-derived and XylNap-primed GAGs (in nanogram per 106 cells) from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). No GAGs were detected in untreated pgsA-745 cells. The type of xyloside has been suggested to determine the formation of CS/DS and HS; in addition, xylosides have been suggested to influence the sulfation pattern of the GAGs (Fritz et al. 1994; Victor et al. 2009), but detailed knowledge of their impact on the GAG structure is still lacking. Such information could provide insights into the GAG biosynthesis and broaden the applications of xylosides and xyloside-primed GAGs. Therefore, we have here systematically investigated the influence of the concentration of xyloside, the type of xyloside and the type of cell on the structure of PG-derived and xyloside-primed GAGs in terms of HS and CS/DS proportions and disaccharide composition. Results The type of cell and xyloside concentration influence the HS and CS/DS proportions xyloside-primed GAGs To investigate the impact of xyloside concentration on GAG structure, PG-derived GAGs were isolated from culture media of untreated cells and xyloside-primed GAGs were isolated from culture media of cells treated with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The studied cell lines included two fibroblastic cell lines: human breast fibroblasts, CCD-1095Sk cells, and human lung fibroblasts, HFL-1, and three epithelial cell lines: human breast carcinoma cells, HCC70, xylosyltransferase-deficient Chinese hamster ovary (CHO) cells, pgsA-745, and the corresponding wild-type CHO cells, CHO-K1. Disaccharide fingerprinting was performed as described in Materials and methods. Based on these data, the amount of recovered GAGs and the proportions of HS and CS/DS were calculated, including the proportion (out of the total GAGs) of IdoA present in CS/DS in blocks and as alternating or single IdoA-containing disaccharide units and GlcA present in CS/DS (summarized in Supplementary data, Tables SI–SV). The results showed that all investigated cell lines produced GAGs primed on XylNap at all concentrations of XylNap, and that XylNap treatment resulted in a higher amount of recovered GAGs than originally synthesized by the cells (as much as 5–200 times more; Figure 2). The amount of recovered GAGs varied between the different cell lines; in general, the fibroblast cell lines produced higher amounts of XylNap-primed GAGs than the epithelial cell lines (at most 8–12 μg per 106 cells and 1–3 μg per 106 cells, respectively). The amount of recovered XylNap-primed GAGs peaked at 100 μM XylNap for all cell lines, except for HFL-1 cells for which the amount was similar or slightly increased at 1000 μM XylNap (Figure 2). All investigated cell lines produced GAGs composed of both HS and CS/DS to some extent; however, the proportions of each component differed between the cell lines. As expected from previous reports (Cöster et al. 1991; Lugemwa and Esko 1991; Vassal-Stermann et al. 2012; Persson et al. 2016), the proportion of HS was lower in the xyloside-primed GAGs than in the PG-derived GAGs (Figure 3A). Furthermore, the influence of the xyloside concentration on the proportion of HS appeared to be cell-dependent; the proportion of HS decreased with increasing concentrations of XylNap in the XylNap-primed GAGs from HCC70 cells (from 58% to 21%), it increased with increasing concentrations of XylNap in the XylNap-primed GAGs from pgsA-745 cells (from 17% to 34%), and did not follow any trend based on concentration of XylNap in the XylNap-primed GAGs from CHO-K1 cells (Figure 3A). The proportion of HS was minor (<3%) in the XylNap-primed GAGs from CCD-1095Sk cells and HFL-1 cells irrespective of concentration of xyloside. The amount of recovered HS was highest after treatment with 100 μM XylNap in the epithelial cell lines, which had a substantial proportion of HS, whereas the amount of recovered HS was highest after treatment with 1000 μM XylNap in the fibroblastic cell lines, where the proportion of HS was minor (Supplementary data, Table SI). Fig. 3. View largeDownload slide HS and CS/DS proportions of PG-derived and XylNap-primed GAGs from different cell lines. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of secreted GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). Fig. 3. View largeDownload slide HS and CS/DS proportions of PG-derived and XylNap-primed GAGs from different cell lines. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of secreted GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of IdoA present in CS/DS in blocks was low (≤11%) in both the PG-derived and XylNap-primed GAGs from all cell lines, except for the skin-derived CCD-1095Sk cells, where it constituted up to 25% of the GAGs (Figure 3B). Nevertheless, a trend towards a decreased proportion of IdoA present in CS/DS in blocks with increasing concentrations of XylNap was observed in the XylNap-primed GAGs from CCD-1095Sk cells, pgsA-745 cells and HFL-1 cells. A similar trend was indicated in the XylNap-primed GAGs from CHO-K1; however, the proportion of IdoA present in CS/DS in blocks in the PG-derived GAGs was slightly lower than in the XylNap-primed GAGs at 10 μM XylNap. In the XylNap-primed GAGs from HCC70 cells where the proportion of IdoA present in CS/DS in blocks was minor (≤3%). Overall, this suggests that the xyloside-primed GAGs become slightly less complex in terms of consecutive IdoA-containing disaccharide units with increasing concentrations of XylNap. IdoA was present in the PG-derived and xyloside-primed GAGs not only in blocks but also as alternating or single IdoA-containing disaccharide units (Figure 3C). Interestingly, the proportion of IdoA present in CS/DS as alternating or single IdoA-containing disaccharide units in the XylNap-primed GAGs from the different cell lines did not decrease with increasing concentrations of XylNap as the proportion of IdoA present in CS/DS in blocks did. Instead, it peaked at 100 μM XylNap in the XylNap-primed GAGs from all cell lines except for HCC70 cells and HFL-1 cells, where it peaked at 1000 μM and 10 μM, respectively. Taking the amount of recovered GAGs into account, this suggests that the distribution of the IdoA in CS/DS is related to the amount of GAGs present; the greater the amount of recovered GAGs, the higher the proportion of IdoA present as alternating or single IdoA-containing disaccharide units. As expected based on the proportions of HS and IdoA present in CS/DS, the proportion of GlcA present in CS/DS was higher in the XylNap-primed GAGs than in the PG-derived GAGs and was the major component (>50%) of all XylNap-primed GAGs (Figure 3D), except for the XylNap-primed GAGs derived from HCC70 cell after treatment with 10 μM XylNap (38% GlcA present in CS/DS). To summarize, the amount of recovered XylNap-primed GAGs was higher than the amount of recovered PG-derived GAGs and, although the amounts differed widely between the cell lines, it tended to peak after treatment with 100 μM XylNap. The XylNap-primed GAGs were in general composed primarily of CS/DS, of which a distinct proportion was IdoA. Overall, the proportion of IdoA in blocks decreased with increasing concentrations of XylNap, and the proportion of IdoA as alternating or single IdoA-containing disaccharide units peaked after treatment with 100 μM XylNap. In contrast, the proportion of HS did not follow any general trend. The type of cell and xyloside concentration influence the CS/DS and HS disaccharide compositions of xyloside-primed GAGs The disaccharide-fingerprinting data showed that the disaccharide composition of the PG-derived and XylNap-primed CS/DS differed between the different cell lines (Figure 4) (summarized in Supplementary data, Table SII). In accordance with previously published data (Persson et al. 2016), the PG-derived CS/DS from HCC70 cells was primarily composed of ΔUA-GalNAc,6S (55%), and to a lesser extent of ΔUA-GalNAc,4S (40%) and ΔUA-GalNAc,4S,6S (5%) (Figure 4A), in contrast to the PG-derived CS/DS from CCD-1095Sk cells that was primarily composed of ΔUA-GalNAc,4S (69%), and to a lesser extent of ΔUA-GalNAc,6S (20%), ΔUA,2S-GalNAc,4S (7%), ΔUA,2S-GalNAc,6S (3%) and ΔUA-GalNAc4S,6S (2%) (Figure 4B). Fig. 4. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed CS/DS from different cell lines. (A–E) The disaccharide composition of chondroitinase ABC-degraded PG-derived GAGs from HCC70 cells (A) and CCD-1095Sk cells (B), XylNap-primed GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) after treatment with 10 μM XylNap and PG-derived GAGs from HFL-1 cells (E). (F-K) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (F), ΔUA-GalNAc,4S (G), ΔUA,2S-GalNAc,4S (H), ΔUA-GalNAc,6S (I), ΔUA,2S-GalNAc,6S (J) and ΔUA-GalNAc,4S,6S (K) in the secreted CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived CS/DS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed CS/DS after treatment with 10 μM XylNap. Fig. 4. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed CS/DS from different cell lines. (A–E) The disaccharide composition of chondroitinase ABC-degraded PG-derived GAGs from HCC70 cells (A) and CCD-1095Sk cells (B), XylNap-primed GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) after treatment with 10 μM XylNap and PG-derived GAGs from HFL-1 cells (E). (F-K) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (F), ΔUA-GalNAc,4S (G), ΔUA,2S-GalNAc,4S (H), ΔUA-GalNAc,6S (I), ΔUA,2S-GalNAc,6S (J) and ΔUA-GalNAc,4S,6S (K) in the secreted CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived CS/DS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed CS/DS after treatment with 10 μM XylNap. Due to the absence of PG-derived GAGs in pgsA-745 cells and the detection of only ΔUA-GalNAc,4S disaccharides in the PG-derived CS/DS from CHO-K1 cells, the CS/DS from pgsA-745 cells and CHO-K1 cells primed on XylNap after treatment with 10 μM XylNap were compared instead of the PG-derived CS/DS (Figure 4C and D). The XylNap-primed CS/DS from pgsA-745 cells and CHO-K1 cells were composed of ΔUA-GalNAc,4S (86 and 87%, respectively) (Figure 4C and D), but also to a low degree of ΔUA-GalNAc (8 and 2%, respectively), ΔUA-GalNAc,6S (3 and 5%, respectively), ΔUA,2S-GalNAc,4S (2 and 4%, respectively), ΔUA-GalNAc,4S,6S (1% in both) and ΔUA,2S-GalNAc,6S (<1% in both). To our knowledge, ΔUA-GalNAc,4S,6S and ΔUA,2S-GalNAc,6S have not previously been observed in CHO-K1 cells (Lawrence et al. 2008). Thus, the formation and amplification of xyloside-primed CS/DS that accompanies xyloside treatment of cells may facilitate detection of disaccharides present at low proportions in PG-derived CS/DS. The PG-derived CS/DS from HFL-1 cells were composed of ΔUA-GalNAc,4S (64%), ΔUA-GalNAc,6S (25%), ΔUA,2S-GalNAc,6S (4%), ΔUA,2S-GalNAc,4S (4%), ΔUA-GalNAc4S,6S (2%) and ΔUA-GalNAc (<1%), to some extent resembling the disaccharide composition of the PG-derived CS/DS from CCD-1095Sk cells (Figure 4E). To estimate the influence of xyloside concentration on the CS/DS disaccharide composition, the proportion of each disaccharide in the XylNap-primed CS/DS from each cell line after treatment with each concentration was compared to that of the PG-derived CS/DS from the corresponding cell line (Figure 4F-K). Although the disaccharide composition differed between the different cell lines, there was a general trend toward a decrease in the proportion of ΔUA-GalNAc,4S disaccharides with increasing concentrations of XylNap (Figure 4G). Concomitantly, either the proportion of ΔUA-GalNAc or ΔUA-GalNAc,6S disaccharides or a combination was increased (Figure 4F and I); the ΔUA-GalNAc,6S proportion was increased in the CS/DS from HCC70 cells and CCD-1095Sk cells, the ΔUA-GalNAc proportion was increased in the CS/DS from the pgsA-745 cells and CHO-K1 cells, the proportions of both were increased in the CS/DS from the HFL-1 cells. The disulfated disaccharides remained essentially unaffected by the increase in concentration of XylNap (Figure 4H, J, and K). The differences in the disaccharide composition of the PG-derived and XylNap-primed HS from the different cell lines were more subtle than the differences in the disaccharide composition of the corresponding CS/DS (Figure 5) (summarized in Supplementary data, Table SV), possibly due to the fact that the HS was composed primarily of one disaccharide, ΔUA-GlcNAc. The PG-derived HS from the different cell lines was composed to 45–60% of ΔUA-GlcNAc (Figure 5A–E) and to a lesser extent of ΔUA-GlcNS (19–26%), ΔUA-GlcNAc,6S (8–14%), ΔUA,2S-GlcNS (4–10%) and ΔUA-GlcNS,6S (2–7%). In addition, the PG-derived HS from the epithelial cell lines were composed of ΔUA,2S-GlcNS,6S (4–5%), and that from HCC70 also of ΔUA,2S-GlcNAc (3%) and ΔUA,2S-GlcNAc,6S (2%). Despite the similarities in HS disaccharide composition, the average number of different disaccharides constituting the PG-derived HS was higher than that constituting the PG-derived CS/DS, suggesting a higher structural diversity in the HS. Fig. 5. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed HS from different cell lines. (A–E) The disaccharide composition of heparinase II- and III-degraded PG-derived GAGs from HCC70 cells (A), CCD-1095Sk cells (B), GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) primed on XylNap after treatment with 10 μM XylNap, and PG-derived GAGs from HFL-1 cells (E). (F–M) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (F), ΔUA,2S-GlcNAc (G), ΔUA-GlcNAc,6S (H), ΔUA,2S-GlcNAc,6S (I), ΔUA-GlcNS (J), ΔUA,2S-GlcNS (K), ΔUA-GlcNS,6S (L) and ΔUA,2S-GlcNS,6S (M) in the secreted HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0–μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived HS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed HS after treatment with 10 μM XylNap. Fig. 5. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed HS from different cell lines. (A–E) The disaccharide composition of heparinase II- and III-degraded PG-derived GAGs from HCC70 cells (A), CCD-1095Sk cells (B), GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) primed on XylNap after treatment with 10 μM XylNap, and PG-derived GAGs from HFL-1 cells (E). (F–M) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (F), ΔUA,2S-GlcNAc (G), ΔUA-GlcNAc,6S (H), ΔUA,2S-GlcNAc,6S (I), ΔUA-GlcNS (J), ΔUA,2S-GlcNS (K), ΔUA-GlcNS,6S (L) and ΔUA,2S-GlcNS,6S (M) in the secreted HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0–μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived HS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed HS after treatment with 10 μM XylNap. The influence of the xyloside concentration on the HS disaccharide composition was estimated in the same way as that on the CS/DS disaccharide composition. Similarly to the CS/DS disaccharide composition, the HS disaccharide composition changed with increasing concentrations of XylNap (Figure 5F–M). However, the HS disaccharide composition did not follow any general trend related to increasing concentrations of XylNap. For example, the proportions of ΔUA-GlcNAc, the most abundant disaccharide and altered to the greatest extent, either increased, decreased or changed inconsistently with the increasing concentrations of XylNap. The lack of distinct trends suggests that the dominating influence by the concentration of xyloside on the HS disaccharide composition may be by the type of cell. Taken together, both the CS/DS and HS disaccharide compositions differed between the different cell lines, although the proportional differences of the disaccharides were more pronounced in the CS/DS. In the CS/DS GAGs, the proportion of ΔUA-GalNAc,4S decreased with the increasing concentrations of XylNap, and the proportions of ΔUA-GalNAc,6S and/or ΔUA-GalNAc simultaneously increased. The HS disaccharide composition also changed with the increasing concentrations of XylNap, but it did not follow any particular trend. The type of xyloside influences the HS and CS/DS proportions of xyloside-primed GAGs To investigate the impact of the type of xyloside on the GAG structure, xyloside-primed GAGs were isolated from culture media of HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment with 100 μM of xyloside of different types. The studied xylosides included pNP-Xyl and MU-Xyl, which are commercially available and commonly used in research, XylNapOH and XylNap, which are known to induce HS synthesis, and 2-naphthyl 1,5-dithio-β-d-xylopyranoside (dithio-XylNap; Figure 1E), which has been shown to serve as an efficient substrate for β1,4-galactosyltransferase 7 (Thorsheim et al. 2017) and may therefore induce synthesis of a higher amount of GAGs than typically induced by xylosides. Because the chemical structures of the studied xylosides differ and do not constitute an adequate series (Figure 1), the xylosides were studied based on their hydrophobicity, which was deduced from the estimated log P values (Supplementary data, Table SVI), resulting in the following order (starting with the lowest hydrophobicity): pNP-Xyl, MU-Xyl, XylNapOH, XylNap and dithio-XylNap. Disaccharide fingerprinting was performed as described in Materials and methods, and the amount of recovered GAGs and the HS and CS/DS proportions were calculated (summarized in Supplementary data, Tables SVII–SXI). Treatment with the different xylosides resulted in xyloside-primed GAGs in all cell lines (Figure 6). The amount of recovered xyloside-primed GAGs differed between the different cell lines; when derived from HCC70 cells and CCD-1095Sk cells, the amount of recovered xyloside-primed GAGs was highest after treatment with MU-Xyl and thereafter decreased with increasing xyloside hydrophobicity. In contrast, when derived from pgsA-745 cells and CHO-K1 cells, the amount of recovered GAGs was similar irrespective of xyloside. Fig. 6. View largeDownload slide Amount of recovered GAGs from different cell lines primed on different xylosides. The amount of recovered xyloside-primed GAGs from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 6. View largeDownload slide Amount of recovered GAGs from different cell lines primed on different xylosides. The amount of recovered xyloside-primed GAGs from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. We, and others, have previously described differences in the proportion of HS in GAGs primed on xylosides with different aglycons (Fritz et al. 1994; Mani et al. 1998, 2004; Holmqvist et al. 2013; Persson et al. 2016). Here, we found that all investigated xylosides were capable of inducing HS by calculating the amount of recovered HS in the xyloside-primed GAGs (Supplementary data, Table SVII). The proportion of HS in the xyloside-primed GAGs differed between the different cell lines; nevertheless, there was a general trend toward an increase in the proportion of HS with increasing xyloside hydrophobicity in all cell lines except CCD-1095Sk cells, where the overall proportions of HS were low (<3%) (Figure 7A). The observed trend confirms previous studies showing that the proportion of HS is influenced by the type of xyloside and, to a certain extent, by its hydrophobicity (Fritz et al. 1994). Fig. 7. View largeDownload slide HS and CS/DS proportions of GAGs from different cell lines primed on different xylosides. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter HCC70 cells and CCD-1095Sk cells only). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 7. View largeDownload slide HS and CS/DS proportions of GAGs from different cell lines primed on different xylosides. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter HCC70 cells and CCD-1095Sk cells only). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. A cell dependence was observed also in the proportion of IdoA present in CS/DS in blocks, but in contrast to the proportion of HS, it did not follow any general trend related to xyloside hydrophobicity (Figure 7B), except in the xyloside-primed GAGs from CCD-1095Sk cells, where it decreased with increasing xyloside hydrophobicity (from 23% to 14%). The proportions of IdoA in CS/DS present in blocks in the xyloside-primed GAGs from HCC70 cells, pgsA-745 cells and CHO-K1 cells were similar irrespective of type of xyloside. The proportion of IdoA present in CS/DS as alternating or single IdoA-containing disaccharide units in xyloside-primed GAGs did not follow any particular trend, except in the xyloside-primed GAGs from CCD-1095Sk cells, where it, similar to the proportion of IdoA present in CS/DS in blocks, decreased with increasing xyloside hydrophobicity (from 22% to 12%; Figure 7C). The pNP-Xyl-primed GAGs from pgsA-745 cells and the MU-Xyl-primed GAGs from CHO-K1 cells were composed to higher proportions of IdoA present in CS/DS as alternating or single IdoA-containing disaccharide units than the GAGs from the corresponding cell lines primed on the other xylosides. The changes in the proportions of IdoA in CS/DS suggest that the type of xyloside influences the complexity of CS/DS by impacting the distribution of IdoA in blocks and as alternating or single IdoA-containing disaccharide units. The proportion of GlcA in CS/DS constituted a majority (>50%) of all XylNap-primed GAGs and differed depending on the type of xyloside but did not follow any trend based on the type of xyloside or xyloside hydrophobicity (Figure 7D). To summarize, the impact of the type of xyloside on the amount of recovered xyloside-primed GAGs was dependent on the type of cell. Nevertheless, in cell lines where the amount of recovered xyloside-primed GAGs was influenced by the type of xyloside, there was a peak in amount of recovered GAGs at a certain xyloside hydrophobicity. There was a trend toward an increase in the proportion of HS with increasing xyloside hydrophobicity, in contrast to the proportion and distribution of IdoA in CS/DS, which, although influenced by the type of xyloside, did not follow any particular trend based on xyloside hydrophobicity. The type of xyloside influences the CS/DS disaccharide composition of xyloside-primed GAGs, in contrast to the HS disaccharide composition, which is influenced to a lesser extent To estimate the influence of xyloside hydrophobicity on the CS/DS and HS disaccharide composition, the proportion of each disaccharide from each cell line primed on each type of xyloside was compared to that of the CS/DS or HS from the corresponding cell line primed on pNP-Xyl, which has the lowest hydrophobicity (Figures 8 and 9; summarized in Supplementary data, Tables SVIII and SXI). Similar to the disaccharide composition of the PG-derived CS/DS, the disaccharide composition of the pNP-Xyl-primed CS/DS clearly differed between the different cell lines (Figure 8A–D). The difference in disaccharide composition between pgsA-745 cell and CHO-K1 cells was more pronounced after treatment with 100 μM pNP-Xyl (20 and 10% ΔUA-GalNAc, respectively; Figure 8C and D) than after treatment with 10 μM XylNap (8% and 2% ΔUA-GalNAc, respectively; Figure 4C and D). Fig. 8. View largeDownload slide Disaccharide composition of CS/DS from different cell lines primed on different xylosides. (A–D) The disaccharide composition of chondroitinase ABC-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C) and CHO-K1 cells (D). (E–J) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (E), ΔUA-GalNAc,4S (F), ΔUA,2S-GalNAc,4S (G), ΔUA-GalNAc,6S (H), ΔUA,2S-GalNAc,6S (I) and ΔUA-GalNAc,4S,6S (J) in the xyloside-primed CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment pNP-Xyl, MU-Xyl, XylNapOH, XylNap and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed CS/DS. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 8. View largeDownload slide Disaccharide composition of CS/DS from different cell lines primed on different xylosides. (A–D) The disaccharide composition of chondroitinase ABC-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C) and CHO-K1 cells (D). (E–J) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (E), ΔUA-GalNAc,4S (F), ΔUA,2S-GalNAc,4S (G), ΔUA-GalNAc,6S (H), ΔUA,2S-GalNAc,6S (I) and ΔUA-GalNAc,4S,6S (J) in the xyloside-primed CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment pNP-Xyl, MU-Xyl, XylNapOH, XylNap and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed CS/DS. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 9. View largeDownload slide Disaccharide composition of HS from different cell lines primed on different xylosides. (A-F) The disaccharide composition of heparinase II- and III-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C), and CHO-K1 cells (D) and that of XylNapOH-primed GAGs from pgsA-745 cells (E) and CHO-K1 cells (F). (G-N) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (G), ΔUA,2S-GlcNAc (H), ΔUA-GlcNAc,6S (I), ΔUA,2S-GlcNAc,6S (J), ΔUA-GlcNS (K), ΔUA,2S-GlcNS (L), ΔUA-GlcNS,6S (M) and ΔUA,2S-GlcNS,6S (N) in the xyloside-primed HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap, and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed HS when the GAGs were derived from HCC70 cells and CCD-1095Sk cells, and MU-Xyl-primed HS when the GAGs were derived from pgsA-745 cells and CHO-K1 cells. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 9. View largeDownload slide Disaccharide composition of HS from different cell lines primed on different xylosides. (A-F) The disaccharide composition of heparinase II- and III-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C), and CHO-K1 cells (D) and that of XylNapOH-primed GAGs from pgsA-745 cells (E) and CHO-K1 cells (F). (G-N) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (G), ΔUA,2S-GlcNAc (H), ΔUA-GlcNAc,6S (I), ΔUA,2S-GlcNAc,6S (J), ΔUA-GlcNS (K), ΔUA,2S-GlcNS (L), ΔUA-GlcNS,6S (M) and ΔUA,2S-GlcNS,6S (N) in the xyloside-primed HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap, and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed HS when the GAGs were derived from HCC70 cells and CCD-1095Sk cells, and MU-Xyl-primed HS when the GAGs were derived from pgsA-745 cells and CHO-K1 cells. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. As with the increasing concentrations of XylNap, the proportion of ΔUA-GalNAc,4S decreased with increasing xyloside hydrophobicity (Figure 8F), accompanied by an increase in proportion of either ΔUA-GalNAc,6S or ΔUA-GalNAc (Figure 8E and H). However, the trend was neither as explicit nor were the differences in the proportions of the CS/DS disaccharides as great as those observed based on xyloside concentration. The proportions of the disulfated disaccharides remained essentially the same irrespective of the type of xyloside (Figure 8G, I and J). These results suggest that hydrophobicity could be one parameter that influences the disaccharide composition of xyloside-primed CS/DS, but those other parameters may also be of importance. The disaccharide composition of the pNP-XylNap-primed HS differed from that of the PG-derived HS in that it was composed of fewer disaccharides (Figure 9B–F, compare to Figure 5B–E); only the pNP-Xyl-primed HS derived from HCC70 cells were composed of the same disaccharides as the PG-derived HS (Figure 9A, and Figure 5A). The pNP-Xyl-primed HS from CCD-1095Sk cells and CHO-K1 cells were composed of ΔUA-GlcNAc, ΔUA-GlcNS and ΔUA,2S-GlcNS, and those from pgsA-745 cells of only ΔUA-GlcNAc and ΔUA-GlcNS (Figure 9C and D). This may reflect the low proportion of HS in these GAGs, as the XylNapOH-primed GAGs from both pgsA-745 cells and CHO-K1 cells, which were composed to a higher proportion of HS than the pNP-Xyl-primed GAGs, were composed of additional disaccharides (Figure 9E and F). Comparison of the disaccharide composition of the HS primed on different xylosides showed little variation and did not display any distinct trends, with the exception of pNP-Xyl-primed HS from pgsA-745 cells and pNP-Xyl- and MU-Xyl-primed HS from CHO-K1 cells, as previously stated in Results (Figure 9G–N). However, the results do not preclude differences in the sequential order of the disaccharides. As previously mentioned in Results, the amount of recovered GAGs from HFL-1 cells did not follow the same trend based on concentrations of XylNap as the GAGs recovered from the other cell lines. Also when introducing other types of xylosides, the HFL-1 cells behaved differently; treatment with 100 μM XylNapOH resulted in similar amount of recovered xyloside-primed GAGs as treatment with 10 μM XylNap did (Figure 10A). Interestingly, these GAGs also had similar CS/DS and HS disaccharide compositions (Figure 10B and C), suggesting that in addition to the xyloside concentration and type of xyloside, the total amount of xyloside-primed GAGs produced by a certain cell influences the disaccharide composition. Fig. 10. View largeDownload slide Disaccharide composition of xyloside-primed GAGs from HFL-1 cells. (A) The amount of recovered GAGs from HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM) and that after treatment with 100 μM XylNapOH. (B, C) The disaccharide composition after chondroitinase ABC degradation (B) and heparinase II and III degradation (C) of xyloside-primed GAGs derived from HFL-1 cells after treatment with 10 μM XylNap (black) and 100 μM XylNapOH (white). Fig. 10. View largeDownload slide Disaccharide composition of xyloside-primed GAGs from HFL-1 cells. (A) The amount of recovered GAGs from HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM) and that after treatment with 100 μM XylNapOH. (B, C) The disaccharide composition after chondroitinase ABC degradation (B) and heparinase II and III degradation (C) of xyloside-primed GAGs derived from HFL-1 cells after treatment with 10 μM XylNap (black) and 100 μM XylNapOH (white). Overall, the differences between the different cell lines in the GAG disaccharide composition were observed also after treatment with different types of xylosides. In the CS/DS, the proportion of ΔUA-GalNAc,4S decreased with increasing xyloside hydrophobicity, whereas the proportions of ΔUA-GalNAc,6S and/or ΔUA-GalNAc increased. This trend was not as explicit as that based on the concentrations of xyloside, suggesting that other parameters than xyloside hydrophobicity may influence the CS/DS disaccharide composition. The HS disaccharide composition was similar irrespective of the type of xyloside except when there were large differences in amount of HS produced. Accordingly, the amount of xyloside-primed GAGs produced appeared to influence the disaccharide composition. Discussion The administration of xylosides to living cells most often results in a dramatic increase in GAG production, likely due to the increase in concentration of xylose substrate in the Golgi. The total amount of GAGs produced is, however, influenced by several parameters, such as type of cell, xyloside concentration, type of xyloside, cell batch and cell passage (Schwartz et al. 1974; Victor et al. 2009; Persson et al. 2016). If aiming to generate as much xyloside-primed GAGs as possible from a certain type of cell, our current data suggest that the xyloside concentration would be the primary parameter to adjust and that 100 μM of xyloside would serve as an adequate starting-point. The relative decrease in production of xyloside-primed GAGs at higher concentrations could be related to substrate inhibition of one or several enzymes in the GAG biosynthesis. For example, substrate inhibition of β-1,4-galactosyltransferase 7, the enzyme responsible for catalyzing the transfer of the first galactose in the linkage region, by different xylosides has been demonstrated (Siegbahn et al. 2015). Alternatively, administration of xylosides at higher concentrations could start to affect the proliferation of the cells (Kolset et al. 1990), thereby resulting in a decrease in GAG biosynthesis. The hydrophobicity of the xyloside is another parameter to consider when aiming to generate as much xyloside-primed GAGs as possible, as there may be an optimum in hydrophobicity for maximal GAG priming, either related to the uptake of the xylosides into the cells and the Golgi or to factors downstream that influence GAG polymerization. Although the structure of GAGs are influenced by several parameters, the type of cell is probably the most important parameter to take into consideration when aiming to produce GAGs with different structures, as the GAG structure appears to be both directly and indirectly related to the type of cell. A direct influence is exemplified by the demonstrated differences in structure of PG-derived and xyloside-primed GAGs derived from the different cell lines, even between pgsA-745 cells and CHO-K1 cells, which differ from each other by the presence or absence of xylosyltransferase (Esko et al. 1985). Indirect influences were evident in the extent to which the xyloside concentration and type of xyloside impacted the structure of the GAGs from each type of cell. In addition, some of the trends related to increasing concentrations of xyloside or xyloside hydrophobicity were exclusively observed in the GAGs of some of the investigated cell lines. Xylosides are known to primarily induce the formation of CS/DS; however, we have here shown that by adjusting the concentration of XylNap and/or the type of xyloside, it is possible to increase the amount of recovered xyloside-primed HS. In the fibroblastic cells, where the proportion of HS was low irrespective of XylNap concentration, the highest amount of recovered HS was obtained after treatment with 1000 μM XylNap. In contrast, in the epithelial cells, where the GAGs were composed of a substantial proportion of HS, the amount of recovered HS peaked at 100 μM XylNap. In addition, the proportion of HS increased with increasing xyloside hydrophobicity, and thereby the total amount of HS was influenced by the amount of GAGs produced. In those cells where the amount of GAGs was similar irrespective of xyloside hydrophobicity, the highest amount of recovered HS was obtained after treatment with the xyloside with the highest hydrophobicity. In those cells that had an optimum in the amount of recovered GAGs based on the xyloside hydrophobicity, the absolute amount of HS instead increased with increasing amount of recovered GAGs. Overall, the extent to which the xyloside concentration and the type of xyloside influenced the production of xyloside-primed HS was dependent on the type of cell, suggesting that empirical optimization is needed when aiming to obtain as much xyloside-primed HS as possible. The domain organization in the GAG chains is believed to differ somewhat between HS and CS/DS; HS has distinct nonmodified N-acetylated domains, highly modified N-sulfated domains and intermediately modified domains containing alternating N-acetylated and N-sulfated disaccharide units between the nonmodified and highly modified domains (Turnbull et al. 2001), whereas CS/DS can be O-sulfated throughout the chain, and may contain GlcA in blocks, IdoA in blocks and alternating or single IdoA-containing disaccharide units in between (Cheng et al. 1994). These differences could provide an explanation to some of our observations: although the average number of different disaccharides present in the HS was found to be higher than that in the CS/DS, the disaccharide composition of the HS varied less between different types of cells than the disaccharide composition of the CS/DS. In addition, the disaccharide composition of HS was affected to a lesser extent than that of CS/DS by the xyloside concentration and type of xyloside. We have recently reported cytotoxic effects of CS/DS from HCC70 cells primed on XylNap or XylNapOH and shown that the HS from the same cells primed on XylNap, but not on XylNapOH, can inhibit the cytotoxic effect (Persson et al. 2016). As confirmed here, the HS disaccharide composition of XylNap- and XylNapOH-primed GAGs is overall similar, suggesting either that minor differences are critical for biological function or that the orchestration of the disaccharides differs between the HS primed on XylNap and XylNapOH. Disaccharide fingerprinting is an appropriate method to obtain an overview of the GAGs produced by a certain type of cell, however, it may not be suitable to detect minor, and perhaps critical, differences between GAGs. This may particularly apply to HS, as it is commonly composed to a high degree of nonsulfated disaccharides (as observed here and in (Li et al. 2015)). Relatively little is known regarding the presence and distribution of IdoA in xyloside-primed CS/DS (Cöster et al. 1991; Vassal-Stermann et al. 2012; Persson et al. 2016). Here, we demonstrated that IdoA was present at various proportions both in blocks and alternating or single IdoA-containing disaccharide units in the xyloside-primed CS/DS from all the investigated cell lines. In general, the proportion of IdoA in blocks decreased with increasing xyloside concentration. The proportion of IdoA distributed as alternating or single IdoA-containing disaccharide units, on the other hand, tended to increase with increasing amount of XylNap-primed GAGs resulting from the increase in concentration of XylNap. The IdoA distribution was also influenced by the type of xyloside but did not follow any trend based on increasing xyloside hydrophobicity. Thus, the proportion and distribution of IdoA in CS/DS are not solely cell-specific but can be adjusted by the xyloside concentration and the type of xyloside. Due to the involvement of several parameters, empirical optimization is yet again needed when aiming for a certain proportion or distribution of IdoA in CS/DS. The disaccharide composition of CS/DS primed on pNP-Xyl has previously been shown to change with increasing concentrations of pNP-Xyl in skin fibroblasts, with the proportion of ΔUA-GalNAc,4S decreasing and those of ΔUA-GalNAc,6S and ΔUA-GalNAc increasing (Cöster et al. 1991). Here, we observed a similar change with increasing concentrations of XylNap in five different cell lines, despite the fact that the disaccharide composition differed between the xyloside-primed CS/DS from different cell lines. We have previously suggested that the disaccharide composition of xyloside-primed GAGs is cell-dependent rather than xyloside-dependent, based on the disaccharide composition of XylNap- and XylNapOH-primed GAGs from HCC70 cells and CCD-1095Sk cells (Persson et al. 2016). However, by including several xylosides with different aglycons or substituents, having a broader range of hydrophobicity, our current results suggest that the type of xyloside also influences the CS/DS disaccharide composition. A decrease in proportion of ΔUA-GalNAc,4S and increases in proportion of ΔUA-GalNAc,6S and/or ΔUA-GalNAc were observed also in GAGs primed on different types of xylosides with increasing hydrophobicity. In addition to the type of cell, xyloside concentration, xyloside hydrophobicity and the amount of GAGs produced, other parameters may be of importance for the GAG structure. To further investigate the impact of the type of xyloside on the disaccharide composition and the possibilities and limits in structural fine-tuning of GAGs using different types of xylosides, future studies should include other series based on, for example, substituents in the aglycon or a broader range of hydrophobicity. Our current data demonstrate that it is possible to produce large quantities of GAGs and to adjust their structure by using xylosides of different types and at different concentrations. In a forward-looking sense, this potentiates the use of xyloside-primed GAGs in different biotechnological applications, such as development of functional screening methods and methods for structural sequencing of GAGs. Materials and methods Cell culture Human breast carcinoma cells, HCC70, human breast fibroblasts, CCD-1095Sk, xylosyltransferase-deficient CHO cells, pgsA-745, CHO cells, CHO-K1, and human lung fibroblasts, HFL-1, were obtained from ATCC and cultured according to ATCC’s instructions. Xylosides 2-naphthol β-d-xylopyranoside, 2-(6-hydroxynaphthyl) β-d-xylopyranoside and 2-naphthyl 1,5-dithio-β-d-xylopyranoside were synthesized as previously described (Fritz et al. 1994; Mani et al. 1998; Thorsheim et al. 2017). 4-Methylumbelliferyl β-d-xylopyranoside and p-nitrophenyl β-d-xylopyranoside were obtained from Sigma-Aldrich. Isolation of xyloside-primed GAGs from culture media The procedure has been described in detail previously (Mani et al. 1998; Persson et al. 2016). Briefly, the cells were cultured in T25 or T75 flasks (Thermo Scientific) to ∼70% confluence and then preincubated in DME/F12 medium supplemented with 10 μg/mL insulin, 25 μg/mL transferrin (all from Sigma-Aldrich), 2 mM L-glutamine, 100 units/mL penicillin, 100 μg/mL streptomycin (from Thermo Scientific) and 10 ng/mL EGF (Corning) for 24 h. Subsequently, the cells were incubated in fresh medium without or with 10 μM, 100 μM or 1000 μM of xyloside. For CHO cells and pgsA-745 cells, EX-CELL™ 325 PF CHO medium (Sigma-Aldrich) supplemented with 2 mM L-glutamine, 100 units/mL penicillin and 100 μg/mL streptomycin were used. For radiolabeling of PG-derived GAGs, the medium was supplemented with 5 μCi/mL [35S]sulfate. The HCC70 cells, CCD-1095Sk cells and HFL-1 cells were treated for 48 h, whereas the pgsA-745 cells and CHO-K1 cells were treated for 24 h to assure their viability was not impaired. After the given incubation time, the cell media were collected and subjected to ion-exchange chromatography, and for the xyloside-supplemented media also hydrophobic interaction chromatography. The GAGs were precipitated and further purified on size-exclusion chromatography HPLC using a Superose 12 HR 10/30 column and collected based on fluorescence (xyloside-primed GAGs) or radioactivity/UV (nonfluorescent xyloside-primed GAGs and PG-derived GAGs). Finally, the GAGs were freeze-dried and quantified roughly using the 1,9-dimethylmethylene blue method (Farndale et al. 1982). Disaccharide fingerprinting of GAGs The method has been described previously (Stachtea et al. 2015; Persson et al. 2016). Briefly, the GAGs were degraded using either chondroitinase ABC (EC 4.2.2.20) (Seikagaku), chondroitinase AC-I and -II (EC 4.2.2.5) (Seikagaku), chondroitinase B (EC 4.2.2.19) (R&D Systems) or heparinase II (no EC number) and heparinase III (EC 4.2.2.8) (both from Flavobacterium heparinum and overexpressed in Escherichia coli, a gift from Prof. Jian Liu, University of North Carolina). Heparinase II and III degradation of heparin resulted in the corresponding degradation profile as degradation with heparinase I–III did (data not shown), and since heparinase I is active primarily toward highly modified regions such as those present in heparin (Linhardt et al. 1990), only the heparinase II and III were used. The disaccharides were then labeled with 2-aminoacridone before separation on an XBridge BEH Shield RP18 (2.1 × 100 mm, 2.5 μm) column (Waters). The experiments were performed once where each series were run at the same time. The identity, quantity and proportion of each disaccharide were determined using disaccharide standards (Iduron) subjected to the corresponding labeling and separation. The proportions of HS, IdoA in CS/DS present in blocks (IdoAAlt/single in CS/DS) and as alternating or single IdoA-containing disaccharide units (IdoAAlt/single in CS/DS), and GlcA in CS/DS were calculated using equations (1)–(4) %HS=(mHeparinaseII+III(mHeparinaseII+III+mChondroitinaseABC))×100 (1) %IdoAChBinCS/DS=(mChondroitinaseB(mHeparinaseII+III+mChondroitinaseABC))×100 (2) %IdoAAlt/singleinCS/DS=(mChondroitinaseABC−(mChondroitinaseACI+II+mChondroitinaseB)2×(mHeparinaseII+III+mChondroitinaseABC))×100 (3) %GlcAinCS/DS=100−%HS−%IdoAChBinCS/DS−%IdoAAlt/singleinCS/DS (4) where m is the mass (in ng) calculated based on the disaccharide data after degradation with the indicated enzymes (Persson et al. 2016). Eq. (2) and (3) were generated based on the cleavage sites of chondroitinase AC-I and -II and chondroitinase B (Linhardt et al. 2006). All equations are based on the assumption that the degradations have gone to completion. Supplementary data Supplementary data are available at GLYCOBIOLOGY online. Funding This work was supported by the Foundation of the Hedda and John Forssman Fund, Lund University, the Medical Faculty at Lund University, the Royal Physiographic Society, the Swedish Cancer Society, the Swedish Research Council and the Åhlen Foundation. Acknowledgements We would like to thank Daniel Willén for synthesis of the xylosides, and Sébastien Vidal and Jean-Pierre Praly for the kind donation of 2,3,4-tri-O-acetyl-5-thio-α-d-xylopyranosyl bromide for the synthesis of 2-naphthyl 1,5-dithio-β-d-xylopyranoside. Furthermore, we would like to thank Anders Malmström for valuable discussions. Conflict of interest statement The authors declare that they have no conflicts of interest with the contents of this article. Author contributions A.P. and K.M. designed and coordinated the study. A.P. performed the experiments and analyzed the data. A.P., U.E. and K.M. interpreted and reviewed the results. A.P. wrote the paper, which was reviewed and approved by U.E. and K.M. Abbreviations CS/DS chondroitin sulfate/dermatan sulfate HS heparan sulfate GAG glycosaminoglycan PG proteoglycan GlcA glucuronic acid GalNAc N-acetylgalactosamine GlcNAc N-acetylglucosamine IdoA iduronic acid CHO Chinese hamster ovary pNP-Xyl p-nitrophenyl β-d-xylopyranoside MU-Xyl 4-methylumbelliferyl β-d-xylopyranoside XylNap 2-naphthyl β-d-xylopyranoside XylNapOH 2-(6-hydroxynaphthyl) β-d-xylopyranoside dithio-XylNap 2-naphthyl 1,5-dithio-β-d-xylopyranoside. References Cheng F , Heinegård D , Malmström A , Schmidtchen A , Yoshida K , Fransson L-A . 1994 . Patterns of uronosyl epimerization and 4-/6–0-sulphation in chondroitin/dermatan sulphate from decorin and biglycan of various bovine tissues . Glycobiology . 4 : 685 – 696 . Google Scholar CrossRef Search ADS PubMed Cöster L , Hernnäs J , Malmström A . 1991 . Biosynthesis of dermatan sulphate proteoglycans. The effect of β-d-xyloside addition on the polymer-modification process in fibroblast cultures . Biochem J . 276 : 533 – 539 . Google Scholar CrossRef Search ADS PubMed Esko JD , Stewart TE , Taylor WH . 1985 . Animal cell mutants defective in glycosaminoglycan biosynthesis . Proc Natl Acad Sci . 82 : 3197 – 3201 . Google Scholar CrossRef Search ADS PubMed Farndale RW , Sayers CA , Barrett AJ . 1982 . A direct spectrophotometric microassay for sulfated glycosaminoglycans in cartilage cultures . Connect Tissue Res . 9 : 247 – 248 . Google Scholar CrossRef Search ADS PubMed Fritz TA , Lugemwa FN , Sarkar AK , Esko JD . 1994 . Biosynthesis of heparan sulfate on β-d-xylosides depends on aglycone structure . J Biol Chem . 269 : 300 – 307 . Google Scholar PubMed Gesslbauer B , Derler R , Handwerker C , Seles E , Kungl AJ . 2016 . Exploring the glycosaminoglycan–protein interaction network by glycan-mediated pull-down proteomics . Electrophoresis . 37 : 1437 – 1447 . Google Scholar CrossRef Search ADS PubMed Gray CJ , Sánchez-Ruíz A , Šardzíková I , Ahmed YA , Miller RL , Reyes Martinez JE , Pallister E , Huang K , Both P , Hartmann M et al. . 2017 . Label- free discovery array platform for the characterization of glycan binding proteins and glycoproteins. Anal Chem . 89 : 4444 – 4451 . Google Scholar CrossRef Search ADS PubMed Hashemian S , Marschinke F , Af Bjerkén S , Strömberg I . 2014 . Degradation of proteoglycans affects astrocytes and neurite formation in organotypic tissue cultures . Brain Res . 1564 : 22 – 32 . Google Scholar CrossRef Search ADS PubMed Holmqvist K , Persson A , Johnsson R , Löfgren J , Mani K , Ellervik U . 2013 . Synthesis and biology of oligoethylene glycol linked naphthoxylosides . Bioorg Med Chem . 21 : 3310 – 3317 . Google Scholar CrossRef Search ADS PubMed Kolset SO , Sakurai K , Ivhed I , Overvatn A , Suzuki S . 1990 . The effect of β-d-xylosides on the proliferation and proteoglycan biosynthesis of monoblastic U-937 cells . Biochem J . 265 : 637 – 645 . Google Scholar CrossRef Search ADS PubMed Lawrence R , Olson SK , Steele RE , Wang L , Warrior R , Cummings RD , Esko JD . 2008 . Evolutionary differences in glycosaminoglycan fine structure detected by quantitative glycan reductive isotope labeling. J Biol Chem . 283 : 33674 – 33684 . Google Scholar CrossRef Search ADS PubMed Li G , Li L , Tian F , Zhang L , Xue C , Linhardt RJ . 2015 . Glycosaminoglycanomics of cultured cells using a rapid and sensitive LC–MS/MS approach . ACS Chem Biol . 10 : 1303 – 1310 . Google Scholar CrossRef Search ADS PubMed Linhardt RJ , Avci FY , Toida T , Kim YS , Cygler M . 2006 . CS lyases: Structure, activity, and applications in analysis and the treatment of diseases . Adv Pharmacol . 53 : 187 – 215 . Google Scholar CrossRef Search ADS PubMed Linhardt RJ , Turnbull JE , Wang HM , Loganathan D , Gallagher JT . 1990 . Examination of the substrate specificity of heparin and heparan sulfate lyases . Biochemistry . 29 : 2611 – 2617 . Google Scholar CrossRef Search ADS PubMed Lugemwa FN , Esko JD . 1991 . Estradiol β-d-xyloside, an efficient primer for heparan sulfate biosynthesis . J Biol Chem . 266 : 6674 – 6677 . Google Scholar PubMed Malmström A , Bartolini B , Thelin MA , Pacheco B , Maccarana M . 2012 . Iduronic acid in chondroitin/dermatan sulfate: Biosynthesis and biological function . J Histochem Cytochem . 60 : 916 – 925 . Google Scholar CrossRef Search ADS PubMed Mani K , Belting M , Ellervik U , Falk N , Svensson G , Sandgren S , Cheng F , Fransson L-Å . 2004 . Tumor attenuation by 2(6-hydroxynaphthyl)-β-d-xylopyranoside requires priming of heparan sulfate and nuclear targeting of the products . Glycobiology . 14 : 387 – 397 . Google Scholar CrossRef Search ADS PubMed Mani K , Havsmark B , Persson S , Kaneda Y , Yamamoto H , Sakurai K , Ashikari S , Habuchi H , Suzuki S , Kimata K et al. . 1998 . Heparan/chondroitin/dermatan sulfate primer 2-(6-hydroxynaphthyl)-o-β-d-xylopyranoside preferentially inhibits growth of transformed cells . Cancer Res . 58 : 1099 – 1104 . Google Scholar PubMed Mikami T , Kitagawa H . 2013 . Biosynthesis and function of chondroitin sulfate . Biochim Biophys Acta . 1830 : 4719 – 4733 . Google Scholar CrossRef Search ADS PubMed Mizumoto S , Yamada S , Sugahara K . 2015 . Molecular interactions between chondroitin–dermatan sulfate and growth factors/receptors/matrix proteins . Curr Opin Struct Biol . 34 : 35 – 42 . Google Scholar CrossRef Search ADS PubMed Okayama M , Kimata K , Suzuki S . 1973 . The influence of p-nitrophenyl β-d-xyloside on the synthesis of proteochondroitin sulfate by slices of embryonic chick cartilage . J Biochem . 74 : 1069 – 1073 . Google Scholar PubMed Persson A , Tykesson E , Westergren-Thorsson G , Malmström A , Ellervik U , Mani K . 2016 . Xyloside-primed chondroitin sulfate/dermatan sulfate from breast carcinoma cells with a defined disaccharide composition has cytotoxic effects in vitro . J Biol Chem . 291 : 14871 – 14882 . Google Scholar CrossRef Search ADS PubMed Schwartz NB , Galligani L , Ho P-L , Dorfman A . 1974 . Stimulation of synthesis of free chondroitin sulfate chains by β-d-xylosides in cultured cells . Proc Natl Acad Sci USA . 71 : 4047 – 4051 . Google Scholar CrossRef Search ADS PubMed Siegbahn A , Thorsheim K , Stahle J , Manner S , Hamark C , Persson A , Tykesson E , Mani K , Westergren-Thorsson G , Widmalm G et al. . 2015 . Exploration of the active site of β4GalT7: Modifications of the aglycon of aromatic xylosides . Org Biomol Chem . 13 : 3351 – 3362 . Google Scholar CrossRef Search ADS PubMed Smith CI , Hilfer SR , Searls RL , Nathanson MA , Allodoli MD . 1990 . Effects of β-d-xyloside on differentiation of the respiratory epithelium in the fetal mouse lung . Dev Biol . 138 : 42 – 52 . Google Scholar CrossRef Search ADS PubMed Spooncer E , Gallagher JT , Krizsa F , Dexter TM . 1983 . Regulation of haemopoiesis in long-term bone marrow cultures. IV. Glycosaminoglycan synthesis and the stimulation of haemopoiesis by β-d-xylosides . J Cell Biol . 96 : 510 – 514 . Google Scholar CrossRef Search ADS PubMed Stachtea XN , Tykesson E , van Kuppevelt TH , Feinstein R , Malmström A , Reijmers RM , Maccarana M . 2015 . Dermatan sulfate-free mice display embryological defects and are neonatal lethal despite normal lymphoid and non-lymphoid organogenesis . PLoS One . 10 : e0140279 . Google Scholar CrossRef Search ADS PubMed Sugahara K , Kitagawa H . 2002 . Heparin and heparan sulfate biosynthesis . IUBMB Life . 54 : 163 – 175 . Google Scholar CrossRef Search ADS PubMed Thorsheim K , Willén D , Tykesson E , Ståhle J , Praly J-P , Vidal S , Johnson MT , Widmalm G , Manner S , Ellervik U . 2017 . Naphthyl thio- and carba-xylopyranosides for exploration of the active site of β-1,4-galactosyltransferase 7 (β4GalT7) . Chem Eur J . 23 : 18057 – 18065 . Google Scholar CrossRef Search ADS PubMed Turnbull J , Powell A , Guimond S . 2001 . Heparan sulfate: Decoding a dynamic multifunctional cell regulator . Trends Cell Biol . 11 : 75 – 82 . Google Scholar CrossRef Search ADS PubMed Vassal-Stermann E , Duranton A , Black AF , Azadiguian G , Demaude J , Lortat-Jacob H , Breton L , Vivès RR . 2012 . A new C-xyloside induces modifications of GAG expression, structure and functional properties . PLoS One . 7 : e47933 . Google Scholar CrossRef Search ADS PubMed Victor XV , Nguyen TKN , Ethirajan M , Tran VM , Nguyen KV , Kuberan B . 2009 . Investigating the elusive mechanism of glycosaminoglycan biosynthesis . J Biol Chem . 284 : 25842 – 25853 . Google Scholar CrossRef Search ADS PubMed Xu D , Esko JD . 2014 . Demystifying heparan sulfate–protein interactions . Annu Rev Biochem . 83 : 129 – 157 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Glycobiology Oxford University Press

Fine-tuning the structure of glycosaminoglycans in living cells using xylosides

Loading next page...
 
/lp/ou_press/fine-tuning-the-structure-of-glycosaminoglycans-in-living-cells-using-O2yWFrUyQ5
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0959-6658
eISSN
1460-2423
D.O.I.
10.1093/glycob/cwy049
Publisher site
See Article on Publisher Site

Abstract

Abstract Xylosides can induce the formation and secretion of xyloside-primed glycosaminoglycans when administered to living cells; however, their impact on the detailed glycosaminoglycan structure remains unknown. Here, we have systematically investigated how the xyloside concentration and the type of xyloside, as well as the cell type, influenced the structure of xyloside-primed glycosaminoglycans in terms of the heparan sulfate and chondroitin/dermatan sulfate proportion and disaccharide composition. We found that although greatest influence was exerted by the cell type, both the xyloside concentration and type of xyloside impacted the proportion of heparan sulfate and the complexity of chondroitin/dermatan sulfate. The disaccharide composition of the chondroitin/dermatan sulfate was influenced by the xyloside concentration and type of xyloside to a higher extent than that of the heparan sulfate; the proportion of 4S-sulfated disaccharides in the chondroitin/dermatan sulfate decreased and the proportions of 6S-sulfated and/or nonsulfated disaccharides increased both with increasing concentrations of xyloside and with increasing xyloside hydrophobicity, whereas the proportion of nonsulfated disaccharides was primarily altered in the heparan sulfate with increasing concentrations of xyloside. Our results indicate that it is feasible to not only produce large amounts of glycosaminoglycans in living cells but also to fine-tune their structures by using xylosides of different types and at different concentrations. Chondroitin sulfate, dermatan sulfate, glycosaminoglycan, heparan sulfate, xyloside Introduction Chondroitin/dermatan sulfate (CS/DS) and heparan sulfate (HS) glycosaminoglycans (GAGs) are linear anionic polysaccharides, commonly 25–100 kDa in size, linked to a proteoglycan (PG) core protein (Turnbull et al. 2001; Sugahara and Kitagawa 2002; Mikami and Kitagawa 2013). Their biosynthesis is a nontemplate-driven process, believed to be regulated by the specificity of the enzymes involved and their organization in the Golgi. CS/DS and HS share a linkage region tetrasaccharide comprised of one xylose, two galactoses and one glucuronic acid (GlcA), where xylose is linked to a serine residue of the PG core protein. The polysaccharides are composed of repeating units of N-acetylgalactosamine (GalNAc) or N-acetylglucosamine (GlcNAc) and GlcA or iduronic acid (IdoA); [-3GalNAcβ1-4GlcAβ1-] in CS, [-3GalNAcβ1-4IdoAα1-] in DS and [-4GlcNAcα1-4GlcAβ1/IdoAα1-] in HS. DS is commonly found as a copolymer of CS/DS, where the IdoA content can vary from one single IdoA per chain to almost 100% IdoA (Malmström et al. 2012). The polysaccharides can be modified by O-sulfation at positions 4 and 6 of GalNAc and at position 2 of GlcA or IdoA in CS/DS, and by N-deacetylation/N-sulfation and by O-sulfation at position 2 of GlcA or IdoA, and at position 6 and, more rarely, at position 3 of GlcNAc in HS, thereby rendering GAGs with high structural diversity and complexity. GAGs are produced by virtually all mammalian cells and are, as a part of PGs, primarily located in the extracellular matrix or at the cell surface where they are involved in many biological processes through direct and indirect interactions with various proteins, such as receptors, enzymes and cytokines (Xu and Esko 2014; Mizumoto et al. 2015). However, only a few of these interactions have been defined structurally (Xu and Esko 2014). The interactions between GAGs and proteins may be nonspecific or specific, in which case they may not be limited to one specific sequence in the GAG but rather to several sequences. In addition, the structural characterization of GAGs is extremely difficult due to the high number of theoretically possible structural variants of GAGs and, as yet, not fully understood biosynthesis. Furthermore, there are no standardized methods to screen for GAG–protein interactions, although a few successful attempts have been reported (Gesslbauer et al. 2016; Gray et al. 2017). Large quantities of GAGs with different structures would facilitate such investigations. More than 40 years ago, 4-nitrophenyl β-d-xylopyranoside (pNP-Xyl, Figure 1A) and methylumbelliferyl β-d-xylopyranoside (MU-Xyl, Figure 1B) were reported to induce the formation and secretion of large quantities of GAGs upon administration to living cells and to concomitantly inhibit GAG production on PG core proteins (Okayama et al. 1973; Schwartz et al. 1974). Since then, β-d-xylopyranosides, also called xylosides, comprising a xylose residue linked to an aglycon, have been used to study the effects of altered PG synthesis in different cellular processes, for example, nerve growth during development (Hashemian et al. 2014), hematopoiesis (Spooncer et al. 1983) and morphogenesis and differentiation (Smith et al. 1990). In the 1990s, xylosides with naphthalene-based aglycons, including 2-(6-hydroxynaphthyl) β-d-xylopyranoside (XylNapOH, Figure 1C) and 2-naphthyl β-d-xylopyranoside (XylNap, Figure 1D), were discovered to induce substantial amounts of HS (Lugemwa and Esko 1991; Fritz et al. 1994; Mani et al. 1998), unlike pNP-Xyl and MU-Xyl, which primarily induce synthesis of CS/DS. Fig. 1. View largeDownload slide Xylosides included in the study. The chemical structures of (A) 4-nitrophenyl β-d-xylopyranoside (pNP-Xyl), (B) 4-methylumbelliferyl β-d-xylopyranoside (MU-Xyl), (B) 2-(6-hydroxynaphthyl) β-d-xylopyranoside (XylNapOH), (D) 2-naphthyl β-d-xylopyranoside (XylNap) and (E) 2-naphthyl 1,5-dithio-β-d-xylopyranoside (dithio-XylNap). The figure was prepared using ChemDraw (Perkin Elmer Informatics, Inc. (cambridgesoft.com); version 13.0.0.3015). Fig. 1. View largeDownload slide Xylosides included in the study. The chemical structures of (A) 4-nitrophenyl β-d-xylopyranoside (pNP-Xyl), (B) 4-methylumbelliferyl β-d-xylopyranoside (MU-Xyl), (B) 2-(6-hydroxynaphthyl) β-d-xylopyranoside (XylNapOH), (D) 2-naphthyl β-d-xylopyranoside (XylNap) and (E) 2-naphthyl 1,5-dithio-β-d-xylopyranoside (dithio-XylNap). The figure was prepared using ChemDraw (Perkin Elmer Informatics, Inc. (cambridgesoft.com); version 13.0.0.3015). Fig. 2. View largeDownload slide Amount of recovered PG-derived GAGs and xyloside-primed GAGs from different cell lines. The amount of recovered secreted PG-derived and XylNap-primed GAGs (in nanogram per 106 cells) from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). No GAGs were detected in untreated pgsA-745 cells. Fig. 2. View largeDownload slide Amount of recovered PG-derived GAGs and xyloside-primed GAGs from different cell lines. The amount of recovered secreted PG-derived and XylNap-primed GAGs (in nanogram per 106 cells) from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). No GAGs were detected in untreated pgsA-745 cells. The type of xyloside has been suggested to determine the formation of CS/DS and HS; in addition, xylosides have been suggested to influence the sulfation pattern of the GAGs (Fritz et al. 1994; Victor et al. 2009), but detailed knowledge of their impact on the GAG structure is still lacking. Such information could provide insights into the GAG biosynthesis and broaden the applications of xylosides and xyloside-primed GAGs. Therefore, we have here systematically investigated the influence of the concentration of xyloside, the type of xyloside and the type of cell on the structure of PG-derived and xyloside-primed GAGs in terms of HS and CS/DS proportions and disaccharide composition. Results The type of cell and xyloside concentration influence the HS and CS/DS proportions xyloside-primed GAGs To investigate the impact of xyloside concentration on GAG structure, PG-derived GAGs were isolated from culture media of untreated cells and xyloside-primed GAGs were isolated from culture media of cells treated with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The studied cell lines included two fibroblastic cell lines: human breast fibroblasts, CCD-1095Sk cells, and human lung fibroblasts, HFL-1, and three epithelial cell lines: human breast carcinoma cells, HCC70, xylosyltransferase-deficient Chinese hamster ovary (CHO) cells, pgsA-745, and the corresponding wild-type CHO cells, CHO-K1. Disaccharide fingerprinting was performed as described in Materials and methods. Based on these data, the amount of recovered GAGs and the proportions of HS and CS/DS were calculated, including the proportion (out of the total GAGs) of IdoA present in CS/DS in blocks and as alternating or single IdoA-containing disaccharide units and GlcA present in CS/DS (summarized in Supplementary data, Tables SI–SV). The results showed that all investigated cell lines produced GAGs primed on XylNap at all concentrations of XylNap, and that XylNap treatment resulted in a higher amount of recovered GAGs than originally synthesized by the cells (as much as 5–200 times more; Figure 2). The amount of recovered GAGs varied between the different cell lines; in general, the fibroblast cell lines produced higher amounts of XylNap-primed GAGs than the epithelial cell lines (at most 8–12 μg per 106 cells and 1–3 μg per 106 cells, respectively). The amount of recovered XylNap-primed GAGs peaked at 100 μM XylNap for all cell lines, except for HFL-1 cells for which the amount was similar or slightly increased at 1000 μM XylNap (Figure 2). All investigated cell lines produced GAGs composed of both HS and CS/DS to some extent; however, the proportions of each component differed between the cell lines. As expected from previous reports (Cöster et al. 1991; Lugemwa and Esko 1991; Vassal-Stermann et al. 2012; Persson et al. 2016), the proportion of HS was lower in the xyloside-primed GAGs than in the PG-derived GAGs (Figure 3A). Furthermore, the influence of the xyloside concentration on the proportion of HS appeared to be cell-dependent; the proportion of HS decreased with increasing concentrations of XylNap in the XylNap-primed GAGs from HCC70 cells (from 58% to 21%), it increased with increasing concentrations of XylNap in the XylNap-primed GAGs from pgsA-745 cells (from 17% to 34%), and did not follow any trend based on concentration of XylNap in the XylNap-primed GAGs from CHO-K1 cells (Figure 3A). The proportion of HS was minor (<3%) in the XylNap-primed GAGs from CCD-1095Sk cells and HFL-1 cells irrespective of concentration of xyloside. The amount of recovered HS was highest after treatment with 100 μM XylNap in the epithelial cell lines, which had a substantial proportion of HS, whereas the amount of recovered HS was highest after treatment with 1000 μM XylNap in the fibroblastic cell lines, where the proportion of HS was minor (Supplementary data, Table SI). Fig. 3. View largeDownload slide HS and CS/DS proportions of PG-derived and XylNap-primed GAGs from different cell lines. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of secreted GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). Fig. 3. View largeDownload slide HS and CS/DS proportions of PG-derived and XylNap-primed GAGs from different cell lines. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of secreted GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells, CHO-K1 cells and HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of IdoA present in CS/DS in blocks was low (≤11%) in both the PG-derived and XylNap-primed GAGs from all cell lines, except for the skin-derived CCD-1095Sk cells, where it constituted up to 25% of the GAGs (Figure 3B). Nevertheless, a trend towards a decreased proportion of IdoA present in CS/DS in blocks with increasing concentrations of XylNap was observed in the XylNap-primed GAGs from CCD-1095Sk cells, pgsA-745 cells and HFL-1 cells. A similar trend was indicated in the XylNap-primed GAGs from CHO-K1; however, the proportion of IdoA present in CS/DS in blocks in the PG-derived GAGs was slightly lower than in the XylNap-primed GAGs at 10 μM XylNap. In the XylNap-primed GAGs from HCC70 cells where the proportion of IdoA present in CS/DS in blocks was minor (≤3%). Overall, this suggests that the xyloside-primed GAGs become slightly less complex in terms of consecutive IdoA-containing disaccharide units with increasing concentrations of XylNap. IdoA was present in the PG-derived and xyloside-primed GAGs not only in blocks but also as alternating or single IdoA-containing disaccharide units (Figure 3C). Interestingly, the proportion of IdoA present in CS/DS as alternating or single IdoA-containing disaccharide units in the XylNap-primed GAGs from the different cell lines did not decrease with increasing concentrations of XylNap as the proportion of IdoA present in CS/DS in blocks did. Instead, it peaked at 100 μM XylNap in the XylNap-primed GAGs from all cell lines except for HCC70 cells and HFL-1 cells, where it peaked at 1000 μM and 10 μM, respectively. Taking the amount of recovered GAGs into account, this suggests that the distribution of the IdoA in CS/DS is related to the amount of GAGs present; the greater the amount of recovered GAGs, the higher the proportion of IdoA present as alternating or single IdoA-containing disaccharide units. As expected based on the proportions of HS and IdoA present in CS/DS, the proportion of GlcA present in CS/DS was higher in the XylNap-primed GAGs than in the PG-derived GAGs and was the major component (>50%) of all XylNap-primed GAGs (Figure 3D), except for the XylNap-primed GAGs derived from HCC70 cell after treatment with 10 μM XylNap (38% GlcA present in CS/DS). To summarize, the amount of recovered XylNap-primed GAGs was higher than the amount of recovered PG-derived GAGs and, although the amounts differed widely between the cell lines, it tended to peak after treatment with 100 μM XylNap. The XylNap-primed GAGs were in general composed primarily of CS/DS, of which a distinct proportion was IdoA. Overall, the proportion of IdoA in blocks decreased with increasing concentrations of XylNap, and the proportion of IdoA as alternating or single IdoA-containing disaccharide units peaked after treatment with 100 μM XylNap. In contrast, the proportion of HS did not follow any general trend. The type of cell and xyloside concentration influence the CS/DS and HS disaccharide compositions of xyloside-primed GAGs The disaccharide-fingerprinting data showed that the disaccharide composition of the PG-derived and XylNap-primed CS/DS differed between the different cell lines (Figure 4) (summarized in Supplementary data, Table SII). In accordance with previously published data (Persson et al. 2016), the PG-derived CS/DS from HCC70 cells was primarily composed of ΔUA-GalNAc,6S (55%), and to a lesser extent of ΔUA-GalNAc,4S (40%) and ΔUA-GalNAc,4S,6S (5%) (Figure 4A), in contrast to the PG-derived CS/DS from CCD-1095Sk cells that was primarily composed of ΔUA-GalNAc,4S (69%), and to a lesser extent of ΔUA-GalNAc,6S (20%), ΔUA,2S-GalNAc,4S (7%), ΔUA,2S-GalNAc,6S (3%) and ΔUA-GalNAc4S,6S (2%) (Figure 4B). Fig. 4. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed CS/DS from different cell lines. (A–E) The disaccharide composition of chondroitinase ABC-degraded PG-derived GAGs from HCC70 cells (A) and CCD-1095Sk cells (B), XylNap-primed GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) after treatment with 10 μM XylNap and PG-derived GAGs from HFL-1 cells (E). (F-K) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (F), ΔUA-GalNAc,4S (G), ΔUA,2S-GalNAc,4S (H), ΔUA-GalNAc,6S (I), ΔUA,2S-GalNAc,6S (J) and ΔUA-GalNAc,4S,6S (K) in the secreted CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived CS/DS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed CS/DS after treatment with 10 μM XylNap. Fig. 4. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed CS/DS from different cell lines. (A–E) The disaccharide composition of chondroitinase ABC-degraded PG-derived GAGs from HCC70 cells (A) and CCD-1095Sk cells (B), XylNap-primed GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) after treatment with 10 μM XylNap and PG-derived GAGs from HFL-1 cells (E). (F-K) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (F), ΔUA-GalNAc,4S (G), ΔUA,2S-GalNAc,4S (H), ΔUA-GalNAc,6S (I), ΔUA,2S-GalNAc,6S (J) and ΔUA-GalNAc,4S,6S (K) in the secreted CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived CS/DS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed CS/DS after treatment with 10 μM XylNap. Due to the absence of PG-derived GAGs in pgsA-745 cells and the detection of only ΔUA-GalNAc,4S disaccharides in the PG-derived CS/DS from CHO-K1 cells, the CS/DS from pgsA-745 cells and CHO-K1 cells primed on XylNap after treatment with 10 μM XylNap were compared instead of the PG-derived CS/DS (Figure 4C and D). The XylNap-primed CS/DS from pgsA-745 cells and CHO-K1 cells were composed of ΔUA-GalNAc,4S (86 and 87%, respectively) (Figure 4C and D), but also to a low degree of ΔUA-GalNAc (8 and 2%, respectively), ΔUA-GalNAc,6S (3 and 5%, respectively), ΔUA,2S-GalNAc,4S (2 and 4%, respectively), ΔUA-GalNAc,4S,6S (1% in both) and ΔUA,2S-GalNAc,6S (<1% in both). To our knowledge, ΔUA-GalNAc,4S,6S and ΔUA,2S-GalNAc,6S have not previously been observed in CHO-K1 cells (Lawrence et al. 2008). Thus, the formation and amplification of xyloside-primed CS/DS that accompanies xyloside treatment of cells may facilitate detection of disaccharides present at low proportions in PG-derived CS/DS. The PG-derived CS/DS from HFL-1 cells were composed of ΔUA-GalNAc,4S (64%), ΔUA-GalNAc,6S (25%), ΔUA,2S-GalNAc,6S (4%), ΔUA,2S-GalNAc,4S (4%), ΔUA-GalNAc4S,6S (2%) and ΔUA-GalNAc (<1%), to some extent resembling the disaccharide composition of the PG-derived CS/DS from CCD-1095Sk cells (Figure 4E). To estimate the influence of xyloside concentration on the CS/DS disaccharide composition, the proportion of each disaccharide in the XylNap-primed CS/DS from each cell line after treatment with each concentration was compared to that of the PG-derived CS/DS from the corresponding cell line (Figure 4F-K). Although the disaccharide composition differed between the different cell lines, there was a general trend toward a decrease in the proportion of ΔUA-GalNAc,4S disaccharides with increasing concentrations of XylNap (Figure 4G). Concomitantly, either the proportion of ΔUA-GalNAc or ΔUA-GalNAc,6S disaccharides or a combination was increased (Figure 4F and I); the ΔUA-GalNAc,6S proportion was increased in the CS/DS from HCC70 cells and CCD-1095Sk cells, the ΔUA-GalNAc proportion was increased in the CS/DS from the pgsA-745 cells and CHO-K1 cells, the proportions of both were increased in the CS/DS from the HFL-1 cells. The disulfated disaccharides remained essentially unaffected by the increase in concentration of XylNap (Figure 4H, J, and K). The differences in the disaccharide composition of the PG-derived and XylNap-primed HS from the different cell lines were more subtle than the differences in the disaccharide composition of the corresponding CS/DS (Figure 5) (summarized in Supplementary data, Table SV), possibly due to the fact that the HS was composed primarily of one disaccharide, ΔUA-GlcNAc. The PG-derived HS from the different cell lines was composed to 45–60% of ΔUA-GlcNAc (Figure 5A–E) and to a lesser extent of ΔUA-GlcNS (19–26%), ΔUA-GlcNAc,6S (8–14%), ΔUA,2S-GlcNS (4–10%) and ΔUA-GlcNS,6S (2–7%). In addition, the PG-derived HS from the epithelial cell lines were composed of ΔUA,2S-GlcNS,6S (4–5%), and that from HCC70 also of ΔUA,2S-GlcNAc (3%) and ΔUA,2S-GlcNAc,6S (2%). Despite the similarities in HS disaccharide composition, the average number of different disaccharides constituting the PG-derived HS was higher than that constituting the PG-derived CS/DS, suggesting a higher structural diversity in the HS. Fig. 5. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed HS from different cell lines. (A–E) The disaccharide composition of heparinase II- and III-degraded PG-derived GAGs from HCC70 cells (A), CCD-1095Sk cells (B), GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) primed on XylNap after treatment with 10 μM XylNap, and PG-derived GAGs from HFL-1 cells (E). (F–M) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (F), ΔUA,2S-GlcNAc (G), ΔUA-GlcNAc,6S (H), ΔUA,2S-GlcNAc,6S (I), ΔUA-GlcNS (J), ΔUA,2S-GlcNS (K), ΔUA-GlcNS,6S (L) and ΔUA,2S-GlcNS,6S (M) in the secreted HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0–μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived HS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed HS after treatment with 10 μM XylNap. Fig. 5. View largeDownload slide Disaccharide composition of PG-derived and XylNap-primed HS from different cell lines. (A–E) The disaccharide composition of heparinase II- and III-degraded PG-derived GAGs from HCC70 cells (A), CCD-1095Sk cells (B), GAGs from pgsA-745 cells (C) and CHO-K1 cells (D) primed on XylNap after treatment with 10 μM XylNap, and PG-derived GAGs from HFL-1 cells (E). (F–M) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (F), ΔUA,2S-GlcNAc (G), ΔUA-GlcNAc,6S (H), ΔUA,2S-GlcNAc,6S (I), ΔUA-GlcNS (J), ΔUA,2S-GlcNS (K), ΔUA-GlcNS,6S (L) and ΔUA,2S-GlcNS,6S (M) in the secreted HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue), CHO-K1 cells (yellow) and HFL-1 cells (white) after treatment without (0–μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding PG-derived HS, except when the GAGs were derived from pgsA-745 cells where the disaccharide proportion was instead subtracted with the proportion of the corresponding XylNap-primed HS after treatment with 10 μM XylNap. The influence of the xyloside concentration on the HS disaccharide composition was estimated in the same way as that on the CS/DS disaccharide composition. Similarly to the CS/DS disaccharide composition, the HS disaccharide composition changed with increasing concentrations of XylNap (Figure 5F–M). However, the HS disaccharide composition did not follow any general trend related to increasing concentrations of XylNap. For example, the proportions of ΔUA-GlcNAc, the most abundant disaccharide and altered to the greatest extent, either increased, decreased or changed inconsistently with the increasing concentrations of XylNap. The lack of distinct trends suggests that the dominating influence by the concentration of xyloside on the HS disaccharide composition may be by the type of cell. Taken together, both the CS/DS and HS disaccharide compositions differed between the different cell lines, although the proportional differences of the disaccharides were more pronounced in the CS/DS. In the CS/DS GAGs, the proportion of ΔUA-GalNAc,4S decreased with the increasing concentrations of XylNap, and the proportions of ΔUA-GalNAc,6S and/or ΔUA-GalNAc simultaneously increased. The HS disaccharide composition also changed with the increasing concentrations of XylNap, but it did not follow any particular trend. The type of xyloside influences the HS and CS/DS proportions of xyloside-primed GAGs To investigate the impact of the type of xyloside on the GAG structure, xyloside-primed GAGs were isolated from culture media of HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment with 100 μM of xyloside of different types. The studied xylosides included pNP-Xyl and MU-Xyl, which are commercially available and commonly used in research, XylNapOH and XylNap, which are known to induce HS synthesis, and 2-naphthyl 1,5-dithio-β-d-xylopyranoside (dithio-XylNap; Figure 1E), which has been shown to serve as an efficient substrate for β1,4-galactosyltransferase 7 (Thorsheim et al. 2017) and may therefore induce synthesis of a higher amount of GAGs than typically induced by xylosides. Because the chemical structures of the studied xylosides differ and do not constitute an adequate series (Figure 1), the xylosides were studied based on their hydrophobicity, which was deduced from the estimated log P values (Supplementary data, Table SVI), resulting in the following order (starting with the lowest hydrophobicity): pNP-Xyl, MU-Xyl, XylNapOH, XylNap and dithio-XylNap. Disaccharide fingerprinting was performed as described in Materials and methods, and the amount of recovered GAGs and the HS and CS/DS proportions were calculated (summarized in Supplementary data, Tables SVII–SXI). Treatment with the different xylosides resulted in xyloside-primed GAGs in all cell lines (Figure 6). The amount of recovered xyloside-primed GAGs differed between the different cell lines; when derived from HCC70 cells and CCD-1095Sk cells, the amount of recovered xyloside-primed GAGs was highest after treatment with MU-Xyl and thereafter decreased with increasing xyloside hydrophobicity. In contrast, when derived from pgsA-745 cells and CHO-K1 cells, the amount of recovered GAGs was similar irrespective of xyloside. Fig. 6. View largeDownload slide Amount of recovered GAGs from different cell lines primed on different xylosides. The amount of recovered xyloside-primed GAGs from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 6. View largeDownload slide Amount of recovered GAGs from different cell lines primed on different xylosides. The amount of recovered xyloside-primed GAGs from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. We, and others, have previously described differences in the proportion of HS in GAGs primed on xylosides with different aglycons (Fritz et al. 1994; Mani et al. 1998, 2004; Holmqvist et al. 2013; Persson et al. 2016). Here, we found that all investigated xylosides were capable of inducing HS by calculating the amount of recovered HS in the xyloside-primed GAGs (Supplementary data, Table SVII). The proportion of HS in the xyloside-primed GAGs differed between the different cell lines; nevertheless, there was a general trend toward an increase in the proportion of HS with increasing xyloside hydrophobicity in all cell lines except CCD-1095Sk cells, where the overall proportions of HS were low (<3%) (Figure 7A). The observed trend confirms previous studies showing that the proportion of HS is influenced by the type of xyloside and, to a certain extent, by its hydrophobicity (Fritz et al. 1994). Fig. 7. View largeDownload slide HS and CS/DS proportions of GAGs from different cell lines primed on different xylosides. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter HCC70 cells and CCD-1095Sk cells only). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 7. View largeDownload slide HS and CS/DS proportions of GAGs from different cell lines primed on different xylosides. The proportions of HS (A), IdoA present in CS/DS in blocks (B) and as alternating or single IdoA-containing disaccharide units (C) and GlcA present in CS/DS (D) of GAGs isolated from HCC70 cells, CCD-1095Sk cells, pgsA-745 cells and CHO-K1 cells after treatment 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap or dithio-XylNap (the latter HCC70 cells and CCD-1095Sk cells only). The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. A cell dependence was observed also in the proportion of IdoA present in CS/DS in blocks, but in contrast to the proportion of HS, it did not follow any general trend related to xyloside hydrophobicity (Figure 7B), except in the xyloside-primed GAGs from CCD-1095Sk cells, where it decreased with increasing xyloside hydrophobicity (from 23% to 14%). The proportions of IdoA in CS/DS present in blocks in the xyloside-primed GAGs from HCC70 cells, pgsA-745 cells and CHO-K1 cells were similar irrespective of type of xyloside. The proportion of IdoA present in CS/DS as alternating or single IdoA-containing disaccharide units in xyloside-primed GAGs did not follow any particular trend, except in the xyloside-primed GAGs from CCD-1095Sk cells, where it, similar to the proportion of IdoA present in CS/DS in blocks, decreased with increasing xyloside hydrophobicity (from 22% to 12%; Figure 7C). The pNP-Xyl-primed GAGs from pgsA-745 cells and the MU-Xyl-primed GAGs from CHO-K1 cells were composed to higher proportions of IdoA present in CS/DS as alternating or single IdoA-containing disaccharide units than the GAGs from the corresponding cell lines primed on the other xylosides. The changes in the proportions of IdoA in CS/DS suggest that the type of xyloside influences the complexity of CS/DS by impacting the distribution of IdoA in blocks and as alternating or single IdoA-containing disaccharide units. The proportion of GlcA in CS/DS constituted a majority (>50%) of all XylNap-primed GAGs and differed depending on the type of xyloside but did not follow any trend based on the type of xyloside or xyloside hydrophobicity (Figure 7D). To summarize, the impact of the type of xyloside on the amount of recovered xyloside-primed GAGs was dependent on the type of cell. Nevertheless, in cell lines where the amount of recovered xyloside-primed GAGs was influenced by the type of xyloside, there was a peak in amount of recovered GAGs at a certain xyloside hydrophobicity. There was a trend toward an increase in the proportion of HS with increasing xyloside hydrophobicity, in contrast to the proportion and distribution of IdoA in CS/DS, which, although influenced by the type of xyloside, did not follow any particular trend based on xyloside hydrophobicity. The type of xyloside influences the CS/DS disaccharide composition of xyloside-primed GAGs, in contrast to the HS disaccharide composition, which is influenced to a lesser extent To estimate the influence of xyloside hydrophobicity on the CS/DS and HS disaccharide composition, the proportion of each disaccharide from each cell line primed on each type of xyloside was compared to that of the CS/DS or HS from the corresponding cell line primed on pNP-Xyl, which has the lowest hydrophobicity (Figures 8 and 9; summarized in Supplementary data, Tables SVIII and SXI). Similar to the disaccharide composition of the PG-derived CS/DS, the disaccharide composition of the pNP-Xyl-primed CS/DS clearly differed between the different cell lines (Figure 8A–D). The difference in disaccharide composition between pgsA-745 cell and CHO-K1 cells was more pronounced after treatment with 100 μM pNP-Xyl (20 and 10% ΔUA-GalNAc, respectively; Figure 8C and D) than after treatment with 10 μM XylNap (8% and 2% ΔUA-GalNAc, respectively; Figure 4C and D). Fig. 8. View largeDownload slide Disaccharide composition of CS/DS from different cell lines primed on different xylosides. (A–D) The disaccharide composition of chondroitinase ABC-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C) and CHO-K1 cells (D). (E–J) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (E), ΔUA-GalNAc,4S (F), ΔUA,2S-GalNAc,4S (G), ΔUA-GalNAc,6S (H), ΔUA,2S-GalNAc,6S (I) and ΔUA-GalNAc,4S,6S (J) in the xyloside-primed CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment pNP-Xyl, MU-Xyl, XylNapOH, XylNap and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed CS/DS. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 8. View largeDownload slide Disaccharide composition of CS/DS from different cell lines primed on different xylosides. (A–D) The disaccharide composition of chondroitinase ABC-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C) and CHO-K1 cells (D). (E–J) The difference in the proportion of each of the disaccharides ΔUA-GalNAc (E), ΔUA-GalNAc,4S (F), ΔUA,2S-GalNAc,4S (G), ΔUA-GalNAc,6S (H), ΔUA,2S-GalNAc,6S (I) and ΔUA-GalNAc,4S,6S (J) in the xyloside-primed CS/DS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment pNP-Xyl, MU-Xyl, XylNapOH, XylNap and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed CS/DS. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 9. View largeDownload slide Disaccharide composition of HS from different cell lines primed on different xylosides. (A-F) The disaccharide composition of heparinase II- and III-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C), and CHO-K1 cells (D) and that of XylNapOH-primed GAGs from pgsA-745 cells (E) and CHO-K1 cells (F). (G-N) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (G), ΔUA,2S-GlcNAc (H), ΔUA-GlcNAc,6S (I), ΔUA,2S-GlcNAc,6S (J), ΔUA-GlcNS (K), ΔUA,2S-GlcNS (L), ΔUA-GlcNS,6S (M) and ΔUA,2S-GlcNS,6S (N) in the xyloside-primed HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap, and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed HS when the GAGs were derived from HCC70 cells and CCD-1095Sk cells, and MU-Xyl-primed HS when the GAGs were derived from pgsA-745 cells and CHO-K1 cells. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. Fig. 9. View largeDownload slide Disaccharide composition of HS from different cell lines primed on different xylosides. (A-F) The disaccharide composition of heparinase II- and III-degraded pNP-Xyl-primed GAGs from HCC70 cells (A), CCD-1095Sk cells (B), pgsA-745 cells (C), and CHO-K1 cells (D) and that of XylNapOH-primed GAGs from pgsA-745 cells (E) and CHO-K1 cells (F). (G-N) The difference in the proportion of each of the disaccharides ΔUA-GlcNAc (G), ΔUA,2S-GlcNAc (H), ΔUA-GlcNAc,6S (I), ΔUA,2S-GlcNAc,6S (J), ΔUA-GlcNS (K), ΔUA,2S-GlcNS (L), ΔUA-GlcNS,6S (M) and ΔUA,2S-GlcNS,6S (N) in the xyloside-primed HS isolated from HCC70 cells (black), CCD-1095Sk cells (gray), pgsA-745 cells (blue) and CHO-K1 cells (yellow) after treatment with 100 μM of pNP-Xyl, MU-Xyl, XylNapOH, XylNap, and dithio-XylNap (the latter only in HCC70 cells and CCD-1095Sk cells). The proportion of each disaccharide was subtracted with the proportion of the same disaccharide from the corresponding pNP-Xyl-primed HS when the GAGs were derived from HCC70 cells and CCD-1095Sk cells, and MU-Xyl-primed HS when the GAGs were derived from pgsA-745 cells and CHO-K1 cells. The xylosides were ordered based on their hydrophobicity, going from the lowest to the highest hydrophobicity. As with the increasing concentrations of XylNap, the proportion of ΔUA-GalNAc,4S decreased with increasing xyloside hydrophobicity (Figure 8F), accompanied by an increase in proportion of either ΔUA-GalNAc,6S or ΔUA-GalNAc (Figure 8E and H). However, the trend was neither as explicit nor were the differences in the proportions of the CS/DS disaccharides as great as those observed based on xyloside concentration. The proportions of the disulfated disaccharides remained essentially the same irrespective of the type of xyloside (Figure 8G, I and J). These results suggest that hydrophobicity could be one parameter that influences the disaccharide composition of xyloside-primed CS/DS, but those other parameters may also be of importance. The disaccharide composition of the pNP-XylNap-primed HS differed from that of the PG-derived HS in that it was composed of fewer disaccharides (Figure 9B–F, compare to Figure 5B–E); only the pNP-Xyl-primed HS derived from HCC70 cells were composed of the same disaccharides as the PG-derived HS (Figure 9A, and Figure 5A). The pNP-Xyl-primed HS from CCD-1095Sk cells and CHO-K1 cells were composed of ΔUA-GlcNAc, ΔUA-GlcNS and ΔUA,2S-GlcNS, and those from pgsA-745 cells of only ΔUA-GlcNAc and ΔUA-GlcNS (Figure 9C and D). This may reflect the low proportion of HS in these GAGs, as the XylNapOH-primed GAGs from both pgsA-745 cells and CHO-K1 cells, which were composed to a higher proportion of HS than the pNP-Xyl-primed GAGs, were composed of additional disaccharides (Figure 9E and F). Comparison of the disaccharide composition of the HS primed on different xylosides showed little variation and did not display any distinct trends, with the exception of pNP-Xyl-primed HS from pgsA-745 cells and pNP-Xyl- and MU-Xyl-primed HS from CHO-K1 cells, as previously stated in Results (Figure 9G–N). However, the results do not preclude differences in the sequential order of the disaccharides. As previously mentioned in Results, the amount of recovered GAGs from HFL-1 cells did not follow the same trend based on concentrations of XylNap as the GAGs recovered from the other cell lines. Also when introducing other types of xylosides, the HFL-1 cells behaved differently; treatment with 100 μM XylNapOH resulted in similar amount of recovered xyloside-primed GAGs as treatment with 10 μM XylNap did (Figure 10A). Interestingly, these GAGs also had similar CS/DS and HS disaccharide compositions (Figure 10B and C), suggesting that in addition to the xyloside concentration and type of xyloside, the total amount of xyloside-primed GAGs produced by a certain cell influences the disaccharide composition. Fig. 10. View largeDownload slide Disaccharide composition of xyloside-primed GAGs from HFL-1 cells. (A) The amount of recovered GAGs from HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM) and that after treatment with 100 μM XylNapOH. (B, C) The disaccharide composition after chondroitinase ABC degradation (B) and heparinase II and III degradation (C) of xyloside-primed GAGs derived from HFL-1 cells after treatment with 10 μM XylNap (black) and 100 μM XylNapOH (white). Fig. 10. View largeDownload slide Disaccharide composition of xyloside-primed GAGs from HFL-1 cells. (A) The amount of recovered GAGs from HFL-1 cells after treatment without (0 μM) or with increasing concentrations of XylNap (10 μM, 100 μM and 1000 μM) and that after treatment with 100 μM XylNapOH. (B, C) The disaccharide composition after chondroitinase ABC degradation (B) and heparinase II and III degradation (C) of xyloside-primed GAGs derived from HFL-1 cells after treatment with 10 μM XylNap (black) and 100 μM XylNapOH (white). Overall, the differences between the different cell lines in the GAG disaccharide composition were observed also after treatment with different types of xylosides. In the CS/DS, the proportion of ΔUA-GalNAc,4S decreased with increasing xyloside hydrophobicity, whereas the proportions of ΔUA-GalNAc,6S and/or ΔUA-GalNAc increased. This trend was not as explicit as that based on the concentrations of xyloside, suggesting that other parameters than xyloside hydrophobicity may influence the CS/DS disaccharide composition. The HS disaccharide composition was similar irrespective of the type of xyloside except when there were large differences in amount of HS produced. Accordingly, the amount of xyloside-primed GAGs produced appeared to influence the disaccharide composition. Discussion The administration of xylosides to living cells most often results in a dramatic increase in GAG production, likely due to the increase in concentration of xylose substrate in the Golgi. The total amount of GAGs produced is, however, influenced by several parameters, such as type of cell, xyloside concentration, type of xyloside, cell batch and cell passage (Schwartz et al. 1974; Victor et al. 2009; Persson et al. 2016). If aiming to generate as much xyloside-primed GAGs as possible from a certain type of cell, our current data suggest that the xyloside concentration would be the primary parameter to adjust and that 100 μM of xyloside would serve as an adequate starting-point. The relative decrease in production of xyloside-primed GAGs at higher concentrations could be related to substrate inhibition of one or several enzymes in the GAG biosynthesis. For example, substrate inhibition of β-1,4-galactosyltransferase 7, the enzyme responsible for catalyzing the transfer of the first galactose in the linkage region, by different xylosides has been demonstrated (Siegbahn et al. 2015). Alternatively, administration of xylosides at higher concentrations could start to affect the proliferation of the cells (Kolset et al. 1990), thereby resulting in a decrease in GAG biosynthesis. The hydrophobicity of the xyloside is another parameter to consider when aiming to generate as much xyloside-primed GAGs as possible, as there may be an optimum in hydrophobicity for maximal GAG priming, either related to the uptake of the xylosides into the cells and the Golgi or to factors downstream that influence GAG polymerization. Although the structure of GAGs are influenced by several parameters, the type of cell is probably the most important parameter to take into consideration when aiming to produce GAGs with different structures, as the GAG structure appears to be both directly and indirectly related to the type of cell. A direct influence is exemplified by the demonstrated differences in structure of PG-derived and xyloside-primed GAGs derived from the different cell lines, even between pgsA-745 cells and CHO-K1 cells, which differ from each other by the presence or absence of xylosyltransferase (Esko et al. 1985). Indirect influences were evident in the extent to which the xyloside concentration and type of xyloside impacted the structure of the GAGs from each type of cell. In addition, some of the trends related to increasing concentrations of xyloside or xyloside hydrophobicity were exclusively observed in the GAGs of some of the investigated cell lines. Xylosides are known to primarily induce the formation of CS/DS; however, we have here shown that by adjusting the concentration of XylNap and/or the type of xyloside, it is possible to increase the amount of recovered xyloside-primed HS. In the fibroblastic cells, where the proportion of HS was low irrespective of XylNap concentration, the highest amount of recovered HS was obtained after treatment with 1000 μM XylNap. In contrast, in the epithelial cells, where the GAGs were composed of a substantial proportion of HS, the amount of recovered HS peaked at 100 μM XylNap. In addition, the proportion of HS increased with increasing xyloside hydrophobicity, and thereby the total amount of HS was influenced by the amount of GAGs produced. In those cells where the amount of GAGs was similar irrespective of xyloside hydrophobicity, the highest amount of recovered HS was obtained after treatment with the xyloside with the highest hydrophobicity. In those cells that had an optimum in the amount of recovered GAGs based on the xyloside hydrophobicity, the absolute amount of HS instead increased with increasing amount of recovered GAGs. Overall, the extent to which the xyloside concentration and the type of xyloside influenced the production of xyloside-primed HS was dependent on the type of cell, suggesting that empirical optimization is needed when aiming to obtain as much xyloside-primed HS as possible. The domain organization in the GAG chains is believed to differ somewhat between HS and CS/DS; HS has distinct nonmodified N-acetylated domains, highly modified N-sulfated domains and intermediately modified domains containing alternating N-acetylated and N-sulfated disaccharide units between the nonmodified and highly modified domains (Turnbull et al. 2001), whereas CS/DS can be O-sulfated throughout the chain, and may contain GlcA in blocks, IdoA in blocks and alternating or single IdoA-containing disaccharide units in between (Cheng et al. 1994). These differences could provide an explanation to some of our observations: although the average number of different disaccharides present in the HS was found to be higher than that in the CS/DS, the disaccharide composition of the HS varied less between different types of cells than the disaccharide composition of the CS/DS. In addition, the disaccharide composition of HS was affected to a lesser extent than that of CS/DS by the xyloside concentration and type of xyloside. We have recently reported cytotoxic effects of CS/DS from HCC70 cells primed on XylNap or XylNapOH and shown that the HS from the same cells primed on XylNap, but not on XylNapOH, can inhibit the cytotoxic effect (Persson et al. 2016). As confirmed here, the HS disaccharide composition of XylNap- and XylNapOH-primed GAGs is overall similar, suggesting either that minor differences are critical for biological function or that the orchestration of the disaccharides differs between the HS primed on XylNap and XylNapOH. Disaccharide fingerprinting is an appropriate method to obtain an overview of the GAGs produced by a certain type of cell, however, it may not be suitable to detect minor, and perhaps critical, differences between GAGs. This may particularly apply to HS, as it is commonly composed to a high degree of nonsulfated disaccharides (as observed here and in (Li et al. 2015)). Relatively little is known regarding the presence and distribution of IdoA in xyloside-primed CS/DS (Cöster et al. 1991; Vassal-Stermann et al. 2012; Persson et al. 2016). Here, we demonstrated that IdoA was present at various proportions both in blocks and alternating or single IdoA-containing disaccharide units in the xyloside-primed CS/DS from all the investigated cell lines. In general, the proportion of IdoA in blocks decreased with increasing xyloside concentration. The proportion of IdoA distributed as alternating or single IdoA-containing disaccharide units, on the other hand, tended to increase with increasing amount of XylNap-primed GAGs resulting from the increase in concentration of XylNap. The IdoA distribution was also influenced by the type of xyloside but did not follow any trend based on increasing xyloside hydrophobicity. Thus, the proportion and distribution of IdoA in CS/DS are not solely cell-specific but can be adjusted by the xyloside concentration and the type of xyloside. Due to the involvement of several parameters, empirical optimization is yet again needed when aiming for a certain proportion or distribution of IdoA in CS/DS. The disaccharide composition of CS/DS primed on pNP-Xyl has previously been shown to change with increasing concentrations of pNP-Xyl in skin fibroblasts, with the proportion of ΔUA-GalNAc,4S decreasing and those of ΔUA-GalNAc,6S and ΔUA-GalNAc increasing (Cöster et al. 1991). Here, we observed a similar change with increasing concentrations of XylNap in five different cell lines, despite the fact that the disaccharide composition differed between the xyloside-primed CS/DS from different cell lines. We have previously suggested that the disaccharide composition of xyloside-primed GAGs is cell-dependent rather than xyloside-dependent, based on the disaccharide composition of XylNap- and XylNapOH-primed GAGs from HCC70 cells and CCD-1095Sk cells (Persson et al. 2016). However, by including several xylosides with different aglycons or substituents, having a broader range of hydrophobicity, our current results suggest that the type of xyloside also influences the CS/DS disaccharide composition. A decrease in proportion of ΔUA-GalNAc,4S and increases in proportion of ΔUA-GalNAc,6S and/or ΔUA-GalNAc were observed also in GAGs primed on different types of xylosides with increasing hydrophobicity. In addition to the type of cell, xyloside concentration, xyloside hydrophobicity and the amount of GAGs produced, other parameters may be of importance for the GAG structure. To further investigate the impact of the type of xyloside on the disaccharide composition and the possibilities and limits in structural fine-tuning of GAGs using different types of xylosides, future studies should include other series based on, for example, substituents in the aglycon or a broader range of hydrophobicity. Our current data demonstrate that it is possible to produce large quantities of GAGs and to adjust their structure by using xylosides of different types and at different concentrations. In a forward-looking sense, this potentiates the use of xyloside-primed GAGs in different biotechnological applications, such as development of functional screening methods and methods for structural sequencing of GAGs. Materials and methods Cell culture Human breast carcinoma cells, HCC70, human breast fibroblasts, CCD-1095Sk, xylosyltransferase-deficient CHO cells, pgsA-745, CHO cells, CHO-K1, and human lung fibroblasts, HFL-1, were obtained from ATCC and cultured according to ATCC’s instructions. Xylosides 2-naphthol β-d-xylopyranoside, 2-(6-hydroxynaphthyl) β-d-xylopyranoside and 2-naphthyl 1,5-dithio-β-d-xylopyranoside were synthesized as previously described (Fritz et al. 1994; Mani et al. 1998; Thorsheim et al. 2017). 4-Methylumbelliferyl β-d-xylopyranoside and p-nitrophenyl β-d-xylopyranoside were obtained from Sigma-Aldrich. Isolation of xyloside-primed GAGs from culture media The procedure has been described in detail previously (Mani et al. 1998; Persson et al. 2016). Briefly, the cells were cultured in T25 or T75 flasks (Thermo Scientific) to ∼70% confluence and then preincubated in DME/F12 medium supplemented with 10 μg/mL insulin, 25 μg/mL transferrin (all from Sigma-Aldrich), 2 mM L-glutamine, 100 units/mL penicillin, 100 μg/mL streptomycin (from Thermo Scientific) and 10 ng/mL EGF (Corning) for 24 h. Subsequently, the cells were incubated in fresh medium without or with 10 μM, 100 μM or 1000 μM of xyloside. For CHO cells and pgsA-745 cells, EX-CELL™ 325 PF CHO medium (Sigma-Aldrich) supplemented with 2 mM L-glutamine, 100 units/mL penicillin and 100 μg/mL streptomycin were used. For radiolabeling of PG-derived GAGs, the medium was supplemented with 5 μCi/mL [35S]sulfate. The HCC70 cells, CCD-1095Sk cells and HFL-1 cells were treated for 48 h, whereas the pgsA-745 cells and CHO-K1 cells were treated for 24 h to assure their viability was not impaired. After the given incubation time, the cell media were collected and subjected to ion-exchange chromatography, and for the xyloside-supplemented media also hydrophobic interaction chromatography. The GAGs were precipitated and further purified on size-exclusion chromatography HPLC using a Superose 12 HR 10/30 column and collected based on fluorescence (xyloside-primed GAGs) or radioactivity/UV (nonfluorescent xyloside-primed GAGs and PG-derived GAGs). Finally, the GAGs were freeze-dried and quantified roughly using the 1,9-dimethylmethylene blue method (Farndale et al. 1982). Disaccharide fingerprinting of GAGs The method has been described previously (Stachtea et al. 2015; Persson et al. 2016). Briefly, the GAGs were degraded using either chondroitinase ABC (EC 4.2.2.20) (Seikagaku), chondroitinase AC-I and -II (EC 4.2.2.5) (Seikagaku), chondroitinase B (EC 4.2.2.19) (R&D Systems) or heparinase II (no EC number) and heparinase III (EC 4.2.2.8) (both from Flavobacterium heparinum and overexpressed in Escherichia coli, a gift from Prof. Jian Liu, University of North Carolina). Heparinase II and III degradation of heparin resulted in the corresponding degradation profile as degradation with heparinase I–III did (data not shown), and since heparinase I is active primarily toward highly modified regions such as those present in heparin (Linhardt et al. 1990), only the heparinase II and III were used. The disaccharides were then labeled with 2-aminoacridone before separation on an XBridge BEH Shield RP18 (2.1 × 100 mm, 2.5 μm) column (Waters). The experiments were performed once where each series were run at the same time. The identity, quantity and proportion of each disaccharide were determined using disaccharide standards (Iduron) subjected to the corresponding labeling and separation. The proportions of HS, IdoA in CS/DS present in blocks (IdoAAlt/single in CS/DS) and as alternating or single IdoA-containing disaccharide units (IdoAAlt/single in CS/DS), and GlcA in CS/DS were calculated using equations (1)–(4) %HS=(mHeparinaseII+III(mHeparinaseII+III+mChondroitinaseABC))×100 (1) %IdoAChBinCS/DS=(mChondroitinaseB(mHeparinaseII+III+mChondroitinaseABC))×100 (2) %IdoAAlt/singleinCS/DS=(mChondroitinaseABC−(mChondroitinaseACI+II+mChondroitinaseB)2×(mHeparinaseII+III+mChondroitinaseABC))×100 (3) %GlcAinCS/DS=100−%HS−%IdoAChBinCS/DS−%IdoAAlt/singleinCS/DS (4) where m is the mass (in ng) calculated based on the disaccharide data after degradation with the indicated enzymes (Persson et al. 2016). Eq. (2) and (3) were generated based on the cleavage sites of chondroitinase AC-I and -II and chondroitinase B (Linhardt et al. 2006). All equations are based on the assumption that the degradations have gone to completion. Supplementary data Supplementary data are available at GLYCOBIOLOGY online. Funding This work was supported by the Foundation of the Hedda and John Forssman Fund, Lund University, the Medical Faculty at Lund University, the Royal Physiographic Society, the Swedish Cancer Society, the Swedish Research Council and the Åhlen Foundation. Acknowledgements We would like to thank Daniel Willén for synthesis of the xylosides, and Sébastien Vidal and Jean-Pierre Praly for the kind donation of 2,3,4-tri-O-acetyl-5-thio-α-d-xylopyranosyl bromide for the synthesis of 2-naphthyl 1,5-dithio-β-d-xylopyranoside. Furthermore, we would like to thank Anders Malmström for valuable discussions. Conflict of interest statement The authors declare that they have no conflicts of interest with the contents of this article. Author contributions A.P. and K.M. designed and coordinated the study. A.P. performed the experiments and analyzed the data. A.P., U.E. and K.M. interpreted and reviewed the results. A.P. wrote the paper, which was reviewed and approved by U.E. and K.M. Abbreviations CS/DS chondroitin sulfate/dermatan sulfate HS heparan sulfate GAG glycosaminoglycan PG proteoglycan GlcA glucuronic acid GalNAc N-acetylgalactosamine GlcNAc N-acetylglucosamine IdoA iduronic acid CHO Chinese hamster ovary pNP-Xyl p-nitrophenyl β-d-xylopyranoside MU-Xyl 4-methylumbelliferyl β-d-xylopyranoside XylNap 2-naphthyl β-d-xylopyranoside XylNapOH 2-(6-hydroxynaphthyl) β-d-xylopyranoside dithio-XylNap 2-naphthyl 1,5-dithio-β-d-xylopyranoside. References Cheng F , Heinegård D , Malmström A , Schmidtchen A , Yoshida K , Fransson L-A . 1994 . Patterns of uronosyl epimerization and 4-/6–0-sulphation in chondroitin/dermatan sulphate from decorin and biglycan of various bovine tissues . Glycobiology . 4 : 685 – 696 . Google Scholar CrossRef Search ADS PubMed Cöster L , Hernnäs J , Malmström A . 1991 . Biosynthesis of dermatan sulphate proteoglycans. The effect of β-d-xyloside addition on the polymer-modification process in fibroblast cultures . Biochem J . 276 : 533 – 539 . Google Scholar CrossRef Search ADS PubMed Esko JD , Stewart TE , Taylor WH . 1985 . Animal cell mutants defective in glycosaminoglycan biosynthesis . Proc Natl Acad Sci . 82 : 3197 – 3201 . Google Scholar CrossRef Search ADS PubMed Farndale RW , Sayers CA , Barrett AJ . 1982 . A direct spectrophotometric microassay for sulfated glycosaminoglycans in cartilage cultures . Connect Tissue Res . 9 : 247 – 248 . Google Scholar CrossRef Search ADS PubMed Fritz TA , Lugemwa FN , Sarkar AK , Esko JD . 1994 . Biosynthesis of heparan sulfate on β-d-xylosides depends on aglycone structure . J Biol Chem . 269 : 300 – 307 . Google Scholar PubMed Gesslbauer B , Derler R , Handwerker C , Seles E , Kungl AJ . 2016 . Exploring the glycosaminoglycan–protein interaction network by glycan-mediated pull-down proteomics . Electrophoresis . 37 : 1437 – 1447 . Google Scholar CrossRef Search ADS PubMed Gray CJ , Sánchez-Ruíz A , Šardzíková I , Ahmed YA , Miller RL , Reyes Martinez JE , Pallister E , Huang K , Both P , Hartmann M et al. . 2017 . Label- free discovery array platform for the characterization of glycan binding proteins and glycoproteins. Anal Chem . 89 : 4444 – 4451 . Google Scholar CrossRef Search ADS PubMed Hashemian S , Marschinke F , Af Bjerkén S , Strömberg I . 2014 . Degradation of proteoglycans affects astrocytes and neurite formation in organotypic tissue cultures . Brain Res . 1564 : 22 – 32 . Google Scholar CrossRef Search ADS PubMed Holmqvist K , Persson A , Johnsson R , Löfgren J , Mani K , Ellervik U . 2013 . Synthesis and biology of oligoethylene glycol linked naphthoxylosides . Bioorg Med Chem . 21 : 3310 – 3317 . Google Scholar CrossRef Search ADS PubMed Kolset SO , Sakurai K , Ivhed I , Overvatn A , Suzuki S . 1990 . The effect of β-d-xylosides on the proliferation and proteoglycan biosynthesis of monoblastic U-937 cells . Biochem J . 265 : 637 – 645 . Google Scholar CrossRef Search ADS PubMed Lawrence R , Olson SK , Steele RE , Wang L , Warrior R , Cummings RD , Esko JD . 2008 . Evolutionary differences in glycosaminoglycan fine structure detected by quantitative glycan reductive isotope labeling. J Biol Chem . 283 : 33674 – 33684 . Google Scholar CrossRef Search ADS PubMed Li G , Li L , Tian F , Zhang L , Xue C , Linhardt RJ . 2015 . Glycosaminoglycanomics of cultured cells using a rapid and sensitive LC–MS/MS approach . ACS Chem Biol . 10 : 1303 – 1310 . Google Scholar CrossRef Search ADS PubMed Linhardt RJ , Avci FY , Toida T , Kim YS , Cygler M . 2006 . CS lyases: Structure, activity, and applications in analysis and the treatment of diseases . Adv Pharmacol . 53 : 187 – 215 . Google Scholar CrossRef Search ADS PubMed Linhardt RJ , Turnbull JE , Wang HM , Loganathan D , Gallagher JT . 1990 . Examination of the substrate specificity of heparin and heparan sulfate lyases . Biochemistry . 29 : 2611 – 2617 . Google Scholar CrossRef Search ADS PubMed Lugemwa FN , Esko JD . 1991 . Estradiol β-d-xyloside, an efficient primer for heparan sulfate biosynthesis . J Biol Chem . 266 : 6674 – 6677 . Google Scholar PubMed Malmström A , Bartolini B , Thelin MA , Pacheco B , Maccarana M . 2012 . Iduronic acid in chondroitin/dermatan sulfate: Biosynthesis and biological function . J Histochem Cytochem . 60 : 916 – 925 . Google Scholar CrossRef Search ADS PubMed Mani K , Belting M , Ellervik U , Falk N , Svensson G , Sandgren S , Cheng F , Fransson L-Å . 2004 . Tumor attenuation by 2(6-hydroxynaphthyl)-β-d-xylopyranoside requires priming of heparan sulfate and nuclear targeting of the products . Glycobiology . 14 : 387 – 397 . Google Scholar CrossRef Search ADS PubMed Mani K , Havsmark B , Persson S , Kaneda Y , Yamamoto H , Sakurai K , Ashikari S , Habuchi H , Suzuki S , Kimata K et al. . 1998 . Heparan/chondroitin/dermatan sulfate primer 2-(6-hydroxynaphthyl)-o-β-d-xylopyranoside preferentially inhibits growth of transformed cells . Cancer Res . 58 : 1099 – 1104 . Google Scholar PubMed Mikami T , Kitagawa H . 2013 . Biosynthesis and function of chondroitin sulfate . Biochim Biophys Acta . 1830 : 4719 – 4733 . Google Scholar CrossRef Search ADS PubMed Mizumoto S , Yamada S , Sugahara K . 2015 . Molecular interactions between chondroitin–dermatan sulfate and growth factors/receptors/matrix proteins . Curr Opin Struct Biol . 34 : 35 – 42 . Google Scholar CrossRef Search ADS PubMed Okayama M , Kimata K , Suzuki S . 1973 . The influence of p-nitrophenyl β-d-xyloside on the synthesis of proteochondroitin sulfate by slices of embryonic chick cartilage . J Biochem . 74 : 1069 – 1073 . Google Scholar PubMed Persson A , Tykesson E , Westergren-Thorsson G , Malmström A , Ellervik U , Mani K . 2016 . Xyloside-primed chondroitin sulfate/dermatan sulfate from breast carcinoma cells with a defined disaccharide composition has cytotoxic effects in vitro . J Biol Chem . 291 : 14871 – 14882 . Google Scholar CrossRef Search ADS PubMed Schwartz NB , Galligani L , Ho P-L , Dorfman A . 1974 . Stimulation of synthesis of free chondroitin sulfate chains by β-d-xylosides in cultured cells . Proc Natl Acad Sci USA . 71 : 4047 – 4051 . Google Scholar CrossRef Search ADS PubMed Siegbahn A , Thorsheim K , Stahle J , Manner S , Hamark C , Persson A , Tykesson E , Mani K , Westergren-Thorsson G , Widmalm G et al. . 2015 . Exploration of the active site of β4GalT7: Modifications of the aglycon of aromatic xylosides . Org Biomol Chem . 13 : 3351 – 3362 . Google Scholar CrossRef Search ADS PubMed Smith CI , Hilfer SR , Searls RL , Nathanson MA , Allodoli MD . 1990 . Effects of β-d-xyloside on differentiation of the respiratory epithelium in the fetal mouse lung . Dev Biol . 138 : 42 – 52 . Google Scholar CrossRef Search ADS PubMed Spooncer E , Gallagher JT , Krizsa F , Dexter TM . 1983 . Regulation of haemopoiesis in long-term bone marrow cultures. IV. Glycosaminoglycan synthesis and the stimulation of haemopoiesis by β-d-xylosides . J Cell Biol . 96 : 510 – 514 . Google Scholar CrossRef Search ADS PubMed Stachtea XN , Tykesson E , van Kuppevelt TH , Feinstein R , Malmström A , Reijmers RM , Maccarana M . 2015 . Dermatan sulfate-free mice display embryological defects and are neonatal lethal despite normal lymphoid and non-lymphoid organogenesis . PLoS One . 10 : e0140279 . Google Scholar CrossRef Search ADS PubMed Sugahara K , Kitagawa H . 2002 . Heparin and heparan sulfate biosynthesis . IUBMB Life . 54 : 163 – 175 . Google Scholar CrossRef Search ADS PubMed Thorsheim K , Willén D , Tykesson E , Ståhle J , Praly J-P , Vidal S , Johnson MT , Widmalm G , Manner S , Ellervik U . 2017 . Naphthyl thio- and carba-xylopyranosides for exploration of the active site of β-1,4-galactosyltransferase 7 (β4GalT7) . Chem Eur J . 23 : 18057 – 18065 . Google Scholar CrossRef Search ADS PubMed Turnbull J , Powell A , Guimond S . 2001 . Heparan sulfate: Decoding a dynamic multifunctional cell regulator . Trends Cell Biol . 11 : 75 – 82 . Google Scholar CrossRef Search ADS PubMed Vassal-Stermann E , Duranton A , Black AF , Azadiguian G , Demaude J , Lortat-Jacob H , Breton L , Vivès RR . 2012 . A new C-xyloside induces modifications of GAG expression, structure and functional properties . PLoS One . 7 : e47933 . Google Scholar CrossRef Search ADS PubMed Victor XV , Nguyen TKN , Ethirajan M , Tran VM , Nguyen KV , Kuberan B . 2009 . Investigating the elusive mechanism of glycosaminoglycan biosynthesis . J Biol Chem . 284 : 25842 – 25853 . Google Scholar CrossRef Search ADS PubMed Xu D , Esko JD . 2014 . Demystifying heparan sulfate–protein interactions . Annu Rev Biochem . 83 : 129 – 157 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

GlycobiologyOxford University Press

Published: May 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off