Glycobiology Advance Access originally published online on March 19, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Glycobiology vol 14 no 5 pp. 467-479, 2004
Glycobiology vol. 14 no. 5 © Oxford University Press 2004; all rights reserved.
Mapping critical biological motifs and biosynthetic pathways of heparan sulfate
Department of Biology, Massachusetts Institute of Technology, Cambridge, MA 02139, and Division of Molecular and Vascular Medicine, BIDMC, Harvard Medical School, Boston, MA 02215
Received on November 20, 2003; revised on December 29, 2003; accepted on January 11, 2004
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Heparan sulfate (HS) interacts with numerous proteins at the cell surface and orchestrates myriad biological events. Unraveling the mechanisms of these events at the molecular level calls for the structural analysis of these negatively charged and highly heterogeneous biopolymers. However, HS is often available only in small quantities, and the task of structural analysis necessitates the use of ultra-sensitive methods, such as mass spectrometry. Sequence heterogeneity within HS chains required us to identify critical functional groups and their spacing to determine structure–function relationships for HS. We carried out structural analysis of HS isolated from wild type, 3-OST-1, 3-OST-3A, or 3-OST-5 sulfotransferase-transduced Chinese hamster ovary cells and also from various tissues. In the context of tissue-specific HS, the data allowed us to map the biosynthetic pathways responsible for the placement of critical groups. As a means of determining the distance between critical groups within a motif, we determined the spacing of the rare GlcNAc-GlcA disaccharide sequence in the completely desulfated re-N-sulfated porcine intestinal heparin. These disaccharides are biosynthetic regulatory markers for 3-OST-1 modification and the partial structure of the antithrombin III binding site. They occur only at the distance of hexasaccharide, octasaccharide, decasaccharide, or dodecasaccharide. Thus this approach allowed us to map both the biosynthetic pathways for generating critical functional groups and their spacing within HS. Our new strategy removes two obstacles to rapid progress in this field of research.
Key words: Glycosaminoglycans / capillary liquid chromatography / mass spectrometry / stable isotope labeling / biosynthetic mechanisms / critical groups
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Heparan sulfate (HS), a highly anionic linear polysaccharide belonging to the family of glycosaminoglycans, is present predominantly at the cell surface or in the extracellular matrix. HS has been shown to exert numerous biological functions through specific interactions with a wide variety of proteins that are involved in morphogenesis, inflammation, wound repair, host defense, and energy metabolism (Bernfield et al., 1999
). Recent molecular cloning of many of the biosynthetic enzymes has led to improved understanding of the basis for HS chain structural complexity and tissue-specific differences. HS proteoglycans are synthesized by (1) formation of a linkage region linking the HS chain to the core protein, (2) generation of the HS chain from activated N-acetylglucosamine and glucuronic acid sugar residues, and (3) enzymatic modification of the chain to yield the specific saccharide structures responsible for regulating various biological systems.
Two such HS structures that have been studied by us are the antithrombin III (AT III) binding site which binds to the ATIII molecule and accelerates its action, and the glycoprotein D (gD) binding site which binds to this Herpes simplex virus-1 (HSV-1) envelope protein and facilitates viral entry. The biosynthesis of these two rare motifs requires specific glucosaminyl 3-O-sulfotransferase (3-OST) enzymes, of which six isoforms have been cloned and characterized. Anticoagulant-active HS (HSact) can be produced by the action of either 3-OST-1 or 3-OST-5 (Liu et al., 1996; Xia et al., 2002). The modifications necessary to complete formation of the gD binding site, however, can be catalyzed by any of four 3-OST isoforms (3-OST-2, 3-OST-3, 3-OST-4, and 3-OST-5) (Shukla et al., 1999; Shukla and Spear, 2001; Xia et al., 2002). Because these enzymes sulfate at the C3 position of glucosamine monosaccharides within the heparan chain, the differences demonstrated by these isoforms probably reflects diversity in the regulation and generation of precursor structures. Due to this precursor specificity, the detection of a unique subset of saccharide residues can be diagnostic for the expression of certain sulfotransferases and for the elaboration of HS with critical groups that have important biological activities. However, these reactions do not convert all of the available precursor structures, resulting in substantial sequence diversity in the final chain. Consequently, these biosynthetic pathways lead to enormous structural heterogeneity in which no two chains are identical and thus render structural analysis of HS challenging.
To address these difficulties, we developed an ultra-sensitive approach that relies on the coupling of capillary high-performance liquid chromatography (HPLC) and electrospray ionization time-of-flight mass spectrometry (ESI-TOF MS) to analyze saccharide residues derived from heparitinase digestion of HS chains. This technique allowed us to rapidly delineate the fine structural modifications elaborated on HS chains from a variety of sources as well as the spacing of critical groups within the motifs of these chains. We can now determine the expression of specific sulfotransferases and the biosynthetic pathways in which they participate within distinct tissues and individual cells.
The AT III binding site requires a minimal pentasaccharide sequence uniquely 3-O sulfated on the central glucosamine residue (Atha et al., 1984, 1987; Kusche et al., 1988). The specific pentasaccharide has been shown to be primarily responsible for the anticoagulant activity of heparin or heparan sulfate (Desai et al., 1998; Petitou et al., 1987). Commercial heparin is highly heterogeneous in nature, with different sulfation patterns along the chain and with relatively few glucuronic acids (accounting for less than 10% of the total uronic acid) (Sudo et al., 2001). One purpose of the present study was to define the distribution of the rare glucuronic acid units within the chains of heparin in the context of mapping critical functional groups. It is known that epimerase requires N-sulfated glucosamine units at the nonreducing end of uronic acid for that residue to be epimerized (Jacobsson et al., 1984; Kuberan et al., 2003). Normally, GlcNAc residues precede glucuronic acids and the disaccharide unit GlcNAc-GlcA constitutes the precursor site for 3-OST-1 action (Linhardt et al., 1992; Zhang et al., 2001a,b). Spacing of these rare GlcNAc-glucuronic acids is unknown. The determination of spacing between these residues would help delineate the biosynthetic pathway leading to the generation of precursor sites for 3-OST-1 action.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Effect of ion-pairing agent on metal ion adduction
The direct coupling of capillary HPLC (LC) and MS greatly enhances their potential for the analysis of HS fine structure (Kuberan et al., 2002
; Linhardt et al., 1992). Among the various modes of LC and MS available, ion-pair reverse-phase HPLC and ESI MS are the most amenable for the direct interfacing of LC and MS (Kuberan et al., 2002). The use of volatile bases such as dibutylamine (DBA) in the analyte solution as an ion-pairing agent can aid in the resolution of saccharide residues by suppressing cationization. DBA significantly attenuates the formation of complexes between sulfated oligosaccharides and alkali metal cations, a phenomenon that tends to broaden the ion envelope for longer oligosaccharides, making mass assignment uncertain. Figure 1 shows the mass spectrum of an AT III–binding synthetic pentasaccharide (GlcNS6S-GlcA-GlcNS3S6S-IdoA2S-GlcNS6S-OMe) in 1:1 mixtures of water:methanol with or without ion-pairing reagent (5 mM DBA). The spectra are plotted to the same scale and can readily be compared. The results are summarized in Table I. The broad, diffuse, cationized ion clusters in the absence of DBA are converted to molecular ion species with little alkali metal ion (Na+) adduction. This result is partly responsible for the improved saccharide characterizations reported herein.
|
|
In vitro modification of extracted HS chains with stable isotope containing sulfate
We characterized both the retention time and the ionic distribution of product residues produced by various members of the 3-OST family in HS isolated from wild-type Chinese hamster ovary (CHO) cells that do not normally express any of these isoforms. Recombinant sulfotransferases, including 3-OST-1, 3-OST-3A, 3-OST-4, and 6-OST-1, were used to modify HS chains in vitro, using PAP34S as a sulfate donor. The stable isotope acts as a mass label that identifies residues generated by in vitro modification (Kuberan et al., 2002
, 2003). Modified chains were then digested with heparitinases to component disaccharide and tetrasaccharide residues, which were subsequently analyzed by LC/MS.
The 3-OST-1 isoform is known to generate two specific disaccharides (Liu et al., 1999), one of which, GlcA-GlcNS3S6S, is readily detected as 
UA-GlcNS3S6S by LC/MS after enzymatic digestion (Figure 2A, inset). This disaccharide was observed as two species: one with a total labeled mass of 596 and the other 725. These values correspond to the combined mass of adduct ions formed between the disaccharide and either water or water plus the ion pairing reagent, DBA: [M-1H + H2O]–1 and [M-2H + DBA + H2O]–1, respectively. In addition, some smaller amount of saturated nonreducing-end terminal residues containing this 3-O-modification and having the same m/z maybe present as well. Also detected were two similar adduct ions with m/z of 594 and 723, which eluted within a peak having a slightly longer retention time (Figure 2A). These masses correspond to adduct ions formed between UA2S-GlcNS6S and water or water along with DBA: [M-1H + H2O]–1 and [M-2H + DBA + H2O]–1, respectively (Figure 2A). Verification that this disaccharide residue contains 6-O-sulfate and must be UA2S-GlcNS6S was proven by in vitro modification of CHO cell HS with recombinant 6-OST-1 and PAP34S (data not shown). This resulted in labeling of the unmodified compound in Figure 2A as well as an additional peak (predominantly m/z of 576: [M-1H]–1), which appears later and coelutes with the UA2S-GlcNS6S standard. Thus the resolution of these two trisulfated residues by LC/MS differs not only by their retention times but also by their adduct ion formation preferences (their ionic distribution). It is important to emphasize that the differences in the migration of modified disaccharides, independent of the interpretation of adduct ions, allow 3-OST-1-modified HS to be easily distinguished from unmodified HS.
|
|
Next, CHO cell HS was modified in vitro using recombinant 3-OST-3A or 3-OST-4 and PAP34S (Figure 2B–D). The resulting LC/MS disaccharide/tetrasaccharide profiles for HS modified by these two enzymes were very similar (data not shown). Four labeled product residues were detected, including two disaccharides:
UA2S-GlcNS3S (m/z of 578 and 707 corresponding to [M-1H]–1 and [M-2H+DBA]–1) and UA2S-GlcNS3S6S (m/z of 787 corresponding to [M-2H+DBA]–1); and two tetrasaccharides: Tetra-A (m/z of 497 corresponding to [M-2H]–2), and Tetra-B (m/z of 601.6 corresponding to [M-3H + DBA]–2). The masses of the two heparitinase resistant tetrasaccharides suggest that they have four and five sulfate groups, respectively.
Despite the identical masses for all HS-derived trisulfated disaccharides, it was still possible to differentiate between the 3-O-labeled trisulfated disaccharide and the 6-O-containing UA2S-GlcNS6S residue. In addition to the differences in retention times between these species, they had different tendencies in forming adduct ions with DBA (Figure 2B, insets).
Overall, we found that adduct ion formation was significant for saccharides containing three or more sulfates only. As we have already demonstrated, preferential differences in adduct ion formation can be an important structural determinant that complements the other data in characterizing specific residues. The retention times and ionic distributions for each of the saccharide residues observed in mature HS are tabulated in Table II. These data allow for the easy identification of sulfotransferase product residues in labeled and unlabeled samples using LC/MS without the further need for standards. This is important, especially for 3-O-sulfate-containing residues, because model compounds of known structure are not readily available.
In vivo analysis of 3-OST isoform activity using LC/MS
To further verify the detection of specific residues by LC/MS, in vivo modified HS was analyzed based on the information derived from the in vitro studies. CHO cells expressing the ecotropic receptor for retroviral expression constructs were transduced with constructs for 3-OST-1, 3-OST-3A, or 3-OST-5, and subsequent HS purified from these cells was analyzed by LC/MS. The extracted ion current (XIC) for residues with three or more sulfates is given in Figure 3. The UA2S-GlcNS6S species was the only highly sulfated disaccharide observed in HS purified from wild type cells (Figure 3A). In contrast, 3-OST-1-transduced cells exhibited a small additional peak with retention time, mass spectrum, and ionic distribution consistent with UA-GlcNS3S6S (m/z of 594 and 723, Figure 3B). Thus, the in vivo 3-OST-1 modification results for CHO cell–derived HS were as expected from the in vitro results described. The in vivo results for 3-OST-3A were also consistent with the corresponding in vitro results (Figure 3C). Peaks for UA2S-GlcNS3S, UA2S-GlcNS3S6S, Tetra-A, and Tetra-B were detected as before (in vitro modification with either 3-OST-3A or 3-OST-4). Along with these species eluting at their expected positions, an additional peak was observed eluting just prior to the UA2S-GlcNS6S + H2O peak (Figure 3C, inset). The resolved mass and ionic distribution of this additional peak are consistent with UA2S-GlcNS3S in adduction with water alone ([M-1H + H20]–1) or with DBA and water ([M-2H + H2O + DBA]–1). It is less likely that this peak corresponds to terminal saturated residues due to their lower abundance.
|
The newly discovered isoform 3-OST-5 differs from other members of the 3-OST family by its ability to generate both AT III and gD binding motifs (Xia et al., 2002
). As a result, a mixture of saccharide residues, consistent with the expression of a number of 3-OST isoforms, is expected from HS modified by this sulfotransferase. 3-OST-5-transduced CHO cells exhibited significant, albeit low, levels of UA2S-GlcNS3S, UA2S-GlcNS3S6S, Tetra-A, and Tetra-B (Figure 3D). In addition, a small peak eluting at the position expected for UA2S-GlcNS3S6S, just before the UA2S-GlcNS6S + H2O peak, was also detected (Figure 3D, inset). Because we found what appeared to be a peak representing UA2S-GlcNS3S + H2O in HS isolated from 3-OST-3A-transduced CHO cells that eluted at this same location (Figure 3C), we were not able to determine unambiguously whether one or both of these species were actually present. Therefore detection of U2-GlcNS3S6S from 3-OST-5-transduced CHO cell HS was inconclusive.
Purification and rechromatography of saccharide residues
For most saccharide residues, the use of LC/MS leads to adequate resolution. However, in cases when two residues are structurally very similar, separation between the two may be incomplete. We were able to enhance the detection of disaccharides that similarly elute with the use of a microfraction collector to capture residues for rechromatography in the absence of salt and other nonsaccharide contaminants that can perturb elution.
Figure 4 shows the results of a fractionation between two closely eluting trisulfated disaccharides, UA2S-GlcNS3S and UA2S-GlcNS6S, derived from 3-OST-3A-transduced CHO cell HS. The capillary HPLC-fractionated samples were equally split between the ESI source and a robotic fraction collector, PROBOT (baiGmbH, Germany). Specific fractions containing the desired saccharides were identified by referring to the MS trace. During the first run the retention times of these two residues were not sufficiently different to give baseline separation (Figure 4A). Despite this, we were able to collect and rechromatograph fractions that contained predominantly one or the other disaccharide (Figures 4B and 4C). Another consequence of rechromatography of the sample was an increase in the intensity of the first peak (Figure 4B). This has been observed with other samples and may be attributable to the purification of the species from much of the interfering salt present in the original sample. Thus in addition to being able to resolve residues to near purity, we were able to increase the intensity for at least one of the two species. Furthermore, the microfraction collector could be employed to accumulate residues for further biological or structural analysis.
|
Analysis of HS found in a human embryonic kidney cell line
LC/MS was used to ascertain the expression of sulfotransferases in a cell line not previously characterized. Figure 5 shows the comparison of the LC/MS analysis of HS isolated from wild-type CHO cells and a human embryonic kidney cell line (HEK293). The profiles for the saccharide residues from these two cell types shows some significant differences. In addition to the same peaks found in CHO cell HS, HEK293 HS contained most of the component residues produced by 3-OST isoforms that synthesize the gD binding site (isoforms 3-OST-2–5) (Figure 5B). The residues detected included
UA2S-GlcNS3S, UA2S-GlcNS3S6S, and Tetra-A. Furthermore, the data in Figure 5 also show that, in comparison to CHO cells, HEK293 express elevated levels of 6-OST activity, as indicated by the relatively large peaks for UA-GlcNS6S and UA2S-GlcNS6S when compared to that of UA2S-GlcNS.
|
Mapping biosynthetic pathways of HS synthesis in various adult mouse tissues
We have previously demonstrated the differential expression of the 3-OST isoforms in various tissues by northern blot (Shworak et al., 1997
). We now attempt to extend this analysis by using LC/MS to map the expression of the product of some of these isoforms and to ascertain HS-biosynthetic pathways for specific HS structures. Tissue samples were isolated from adult C57 mice and the results from lung, kidney, and spleen samples are shown in Figure 6. Both kidney and spleen exhibited product residues similar to those generated by 3-OST-3, -4, or -5 (Figures 6B and 6C). Because northern analysis shows strong expression of 3-OST-3A in kidney, the detection of significant levels of 3-OST-3A-like product in mouse kidney was anticipated. Unlike kidney and spleen, lung showed very little 3-O-sulfate containing residues in accordance with the northern blot data. Because the addition of 3-O-sulfates are relatively rare modifications found in mature HS, LC/MS analysis of 3-OST products provides a more meaningful assessment of HS fine structure and biosynthetic pathway than that deduced from steady-state levels of mRNA. More important, mapping product residues will give us insight into the concerted action of multiple HS modifying enzymes that leads to the biosynthesis of HS domains having vital biological activities on a tissue or cellular basis.
|
In Figure 7, the XIC chromatographs encompassing all of the HS product disaccharides present within these tissue samples are shown (Figure 7A). The corresponding spectrometric trace taken over the entire range from 0–73-min elution time reflects changes seen in the more abundant product residues. We have not examined the presence of nonsulfated disaccharide (
UA-GlcNAc) and its composition because this specific disaccharide component elutes at the very beginning of the gradient along with impurities, rendering it difficult to be resolved. In addition to the apparent lack of 3-O-sulfate-containing disaccharides in HS isolated from lung tissue, there was a significant reduction in 6-O-sulfate-containing residues when compared to other residues such as UA-GlcNS and UA2S-GlsNS. In addition, when analyzed against HS from either kidney or spleen, HS from lung contained lower levels of UA-GlcAc6S, UA-GlcNS6S, and UA2S-GlcNS6S. These results suggest that 6-OST activity is relatively lower in lung tissue. HS from kidney, on the other hand, showed significant increase in NDST activity when compared to that of lung and spleen HS. This is demonstrated by the elevated levels of UA-GlcNS, UA-GlcNS6S, and UA2S-GlcNS6S detected in kidney HS (Figure 7A). Finally, HS isolated from spleen exhibited a relatively lower level of 2-OST activity as demonstrated by lesser detected yields for UA2S-GlcNS and UA2S-GlcNS6S.
|
Differences in multiple enzyme activities in various organs will lead to differences in the HS biosynthetic pathways present within these tissues. As is illustrated in Figure 7B, the differences in the activities of 3-OST and 6-OST in lung tissue will effect the generation of downstream HS structures. Simultaneous scanning for the products from multiple sulfotransferases gives us the ability to map the biosynthetic pathways for specific HS modalities within various tissue sources.
Spatial distribution of GlcNAc-GlcA in heparin
We have shown in other studies (Zhang et al., 2001b) that the N-acetyl group is a critical structural feature that marks HS chains for the action of 3-OST-1 but not 3-OST-3A. It is not clear how HS biosynthetic pathways diverge to accommodate the needs for different precursor structures that ultimately lead to AT III or gD binding motifs. Moreover, because of the highly heterogeneous nature of the polymer, it is very difficult to determine this important spacing parameter between these rare residues. Therefore, we prepared heparin polymers with a more homogeneous sulfation pattern to aid in the determination of the spacing between GlcNAc-GlcA disaccharide units. This was accomplished by complete desulfation/re-N-sulfation of heparin polymer while preserving all other structural features as shown in Scheme 1 (Nagasawa and Inoue, 1974). This polymer was then cleaved with recombinant heparanase to prepare oligosaccharides containing glucuronic acid at the reducing end of the cleavage site (McKenzie et al., 2003). The resulting oligosaccharides were then subjected to LC/MS analysis to determine the size distribution of the products that correspond to the spacing between these rare disaccharide residues. The results revealed a series of oligosaccharides that range from hexasaccharides to dodecasaccharides with approximately equal amounts of each oligosaccharide. This indicates that the average spacing between glucuronic acids is approximately 8 or 10 residues (Figures 8A and 8B; Table III). The results are summarized in Table III for oligosaccharides and their molecular weights obtained from LC/MS analysis.
|
|
|
In conclusion, we have mapped the occurrence of rare glucuronic acid units, which shows that the mapping of critical residues for biosynthetic pathways is now possible. We acknowledge that extending this approach to HS chains, which are more heterogeneous than desulfated, N-sulfated heparin, may require additional MS probes, such as the use of stable isotopes and cloned biosynthetic and catabolic enzymes, to establish structure–function correlations.
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
In this study we explored two important aspects of HS fine structure. These are mapping of biological motifs in different tissues and cells and determining the spacing of these markers within HS chains.
We determined which HS modifying enzymes are expressed and which HS structures can be synthesized in cells and tissues. We used LC/MS as an effective and simple tool for the structural analysis of HS isolated from various sources. This is an important development because antibodies against specific isoforms of biosynthetic enzymes are not available. This technique allows us to establish the presence or absence of specific sulfotransferases within cells or tissues. Furthermore, we showed that this approach also permits the delineation of biosynthetic pathways within different tissues, which has not been possible prior to this study.
We characterized the fine structure of HS purified from HEK293 cells. These cells appear to express a sulfotransferase activity similar to 3-OST isoforms that can generate the HSV-1 gD binding site. This result is consistent with the known susceptibility of HEK293 cells to HSV infection (Israel et al., 1995; Skaliter et al., 1996). In addition to demonstrating 3-OST activity in HEK293 cells, the data in Figure 5 also show that in comparison to CHO cells, HEK293 cells express elevated levels of 6-OST activity, as indicated by the relatively large peaks for UA-GlcNS6S and UA2S-GlcNS6S when compared to that of UA2S-GlcNS. This could have significant implications on the use of these cells in studying the role of HS structure in important biological systems, such as the interaction between cell factors and their receptors.
We earlier observed differences in the steady-state levels of 3-OST isoform mRNAs in various tissues. As expected from these mRNA results, significant levels of 3-O-sulfate-containing residues were detected in HS isolated from kidney, whereas no significant levels were detected in HS extracted from lung tissue. Furthermore, spleen, which had not as yet been characterized for 3-OST expression, was shown herein to express significant levels of 3-O-sulfated HS residues. In addition to differences in 3-OST expression, we were also able to detect variations in NDST, 2-OST, and 6-OST activities within these tissues. Thus we have the ability to map relative differences in HS biosynthetic enzymes, which allows us to elucidate the HS biosynthetic pathways and determine the biological role of HS sulfotransferases within both tissues and cells.
A number of studies have established that CHO cells normally express NDST-1 and NDST-2 as well as 2-OST and 6-OST-1 (Aikawa and Esko, 1999; Bai and Esko, 1996; Zhang et al., 2001a). Together these enzymes act in a concerted fashion to produce the HS structures normally observed in these cells. Figure 9 summarizes the complement of HS sulfotransferase activities normally found in CHO cells and the resulting modifications in HS chains (mainly at the disaccharide level). In addition, the effects on HS structure remodeling by exogenous sulfotransferases are shown. It is understood that a strong correlation between precursor site and 3-OST isoform must exist. However, the scheme outlined in Figure 9 shows only the modifications that occur on disaccharides within the nascent chain without regard to these critical biological constraints.
|
We found no significant differences in the disaccharides produced by either 3-OST-3A or 3-OST-4 in CHO cells. All four of the 3-O-sulfate-containing product saccharides previously identified in HS modified with 3-OST-3A were observed in HS modified by in vitro labeling with either 3-OST-3A or 3-OST-4 using LC/MS (Liu et al., 1999a
,b). Thus on the disaccharide and tetrasaccharide levels there are no significant differences detected in the products produced by these two isoforms. Differences can be detected, however, within the extended precursor sites that ultimately control the actions of these sulfotransferases (Wu et al., 2003). We note that observed disaccharide peaks allowed us to distinguish the 3-OST-1 isoform from either 3-OST-3A or 3-OST-4 independent of our interpretation of the nature of adduct ions. Mapping of entire pathways within both individual cells and tissues is now possible and is expected to be a powerful means of defining source-specific HS modalities. Furthermore, with enzymatic dispersion of tissue samples or laser capture microdissection, it should be possible to map these structures on a cellular level and thus focus attention on specific cells within tissues.
Finally, we defined the spacing of critical biosynthetic regulatory markers along the HS chains. To this end, the positions of the relatively rare GlcNAc-GlcA units along the heparin chains were determined. This structural feature directs the action of 3-OST-1 in sulfonating glucosamine units to generate AT III binding sites. In a model system, completely desulfated and re-N-sulfated heparin with homogeneous sulfation pattern was cleaved with endoglucuronidase. This study permitted us to infer the position of the potential 3-OST-1 acceptor sites along the heparin chain. Although this was accomplished in a model system, we trust that it can now be expanded with fully sulfated HS chains to define the spacing of biosynthetic molecular markers, which leads to the placement of critical functional groups.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Preparation and endoglucuronidase digestion of completely desulfated, re-N-sulfated heparin
The preparation of completely desulfated re-N-sulfated (CDSNS) heparin was carried out as reported earlier with some minor modifications (Nagasawa and Inoue, 1974
). One microliter of CDSNS-heparin solution (1 mg/ml) was added to 1 µl 400 mM ammonium acetate buffer (pH 7.0) containing 33 mM calcium chloride. The volume was adjusted to 10 µl with distilled water, and 1 µl endoglucuronidase enzyme (0.73 mg/ml) was added. The digestion was carried out at 37°C overnight. An equal volume of ion-pairing solution (5 mM dibutylammonium acetate) was added to the reaction mixture for subsequent analysis by LC/MS.
Tissue culture and retroviral transductions
CHO cells were grown in growth medium A (Ham's F-12 containing 50 U/ml penicillin, 50 mg/ml streptomycin, 2 mM glutamine, and 10% fetal calf serum by volume). A human embryonic kidney cell line (HEK293) as well as the viral packaging line (Phoenix cells, ATCC SD 3444) were grown in growth medium B (Dulbecco's modified Eagle's medium containing 10% fetal calf serum, 50 U/ml penicillin, 50 mg/ml streptomycin, and 2 mM glutamine).
For retroviral infections (transductions), a stable CHO cell line CHO[ECO] expressing the ecotropic retrovirus receptor was used as previously described.(Gu et al., 2000) The retroviral vector MVT-37 was prepared from the MSCVpac retroviral vector (Hawley et al., 1994) by ligation of a new multiple cloning segment. The complete coding sequences for human clones of 3OST-1, 3OST-3A, 3OST-4, and 3OST-5 were subcloned into the appropriate restriction sites and the Phoenix packaging line, maintained in growth medium B, and transfected with these 3-OST constructs as previously described (Gu et al., 2000).
Isolation of cell- and tissue-specific HS
HS was extracted from cells as described in our prior publication (Zhang et al., 2001b). Tissue samples were dissected from adult (3 months old) C57 BL6 mice (Taconic) and subsequently homogenized in phosphate buffered saline. HS from tissue samples was isolated in the same way as HS extracted from cells.
In vitro modification of HS using stable isotope
3'-Phosphoadenosine-5'-phosphosulfate (PAPS) with stable isotope 34S was prepared as previously outlined (Kuberan et al., 2003). Baculovirus-expressed 3-OST-1, 3-OST-3A, 3-OST-4, and 6OST-1 enzymes were prepared as described earlier (Kuberan et al., 2003; Liu et al., 1999; Shworak et al., 1997). Purification of labeled HS chains was completed as mentioned earlier.
Enzymatic degradation of HS chains
Stable isotope-labeled or unlabeled HS (10 µl) was combined with 34-µl water, 5 µl 10x digest buffer (400 mM ammonium acetate, 33 mM calcium acetate), 1 µl 0.33 mU heparitinases I, heparitinases II, and heparinase (Seikagaku, Tokyo). Digestion was allowed to proceed overnight at 37°C. Digests were lyophilized and reconstituted in 7 µl water, 1 ml ion pairing reagent DBA and then analyzed by LC/MS for heparitinase-derived residues as reported earlier (Kuberan et al., 2002).
Data analysis
The ionic masses for disaccharide residues are determined by the formula 336 + x(42) + y(80), where x = number of acetyl groups and y = number of sulfates. In addition to the raw ionic weight for each residue, the combined weight for adduct ions formed with either water or the ion pairing reagent used for reverse-phase chromatography (DBA) were also analyzed. Extracted ion current was computed as described in the documentation for the Data Explorer software provided by Applied Biosystems.
|
Acknowledgements |
|---|
This work was funded by the NIH (1PO1 HL 66105, 5RO1HL 59479). We thank Dr. Edward D. Lamperti for providing mouse tissue samples, Dr. Z. Wu for the preparation of PAP34S, and Dr. Edward McKenzie for providing heparanase.
|
Footnotes |
|---|
2 To whom correspondence should be addressed; e-mail: rdrrosen{at}mit.edu
1 These authors contributed equally to this article. 
|
Abbreviations |
|---|
UA, 4-5-unsaturated uronic acid; UA2S, 4,5-unsaturated uronic acid 2-O-sulfate; AT III, antithrombin III; CDSNS, completely desulfated re-N-sulfated; CHO, Chinese hamster ovary; DBA, dibutylamine; ESI-TOF MS, electrospray ionization time-of-flight mass spectrometry; gD, glycoprotein D; HPLC, high-performance liquid chromatography; HS, heparan sulfate; HSV-1, herpes simplex virus 1; LC/MS, liquid chromatography/mass spectrometry; OST, O-sulfotransferase; PAPS, 3'-phosphoadenosine-5'-phosphosulfate; XIC, extracted ion current |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsReferences |
|---|
Aikawa, J. and Esko, J.D. (1999) Molecular cloning and expression of a third member of the heparan sulfate/heparin GlcNAc N-deacetylase/N-sulfotransferase family. J. Biol. Chem., 274, 2690–2695.
Atha, D.H., Stephens, A.W., and Rosenberg, R.D. (1984) Evaluation of critical groups required for the binding of heparin to antithrombin. Proc. Natl Acad. Sci. USA, 81, 1030–1034.
Atha, D.H., Lormeau, J.C., Petitou, M., Rosenberg, R.D., and Choay, J. (1987) Contribution of 3-O- and 6-O-sulfated glucosamine residues in the heparin-induced conformational change in antithrombin III. Biochemistry, 26, 6454–6461.[CrossRef][Medline]
Bai, X. and Esko, J. D. (1996) An animal cell mutant defective in heparan sulfate hexuronic acid 2-O-sulfation. J. Biol. Chem., 271, 17711–17717.
Bernfield, M., Gotte, M., Park, P.W., Reizes, O., Fitzgerald, M.L., Lincecum, J., and Zako, M. (1999) Functions of cell surface heparan sulfate proteoglycans. Annu. Rev. Biochem., 68, 729–777.[CrossRef][ISI][Medline]
Desai, U.R., Petitou, M., Bjork, I., and Olson, S.T. (1998) Mechanism of heparin activation of antithrombin. Role of individual residues of the pentasaccharide activating sequence in the recognition of native and activated states of antithrombin. J. Biol. Chem., 273, 7478–7487.
Gu, X., Lawrence, R., and Krieger, M. (2000) Dissociation of the high density lipoprotein and low density lipoprotein binding activities of murine scavenger receptor class B type I (mSR-BI) using retrovirus library-based activity dissection. J. Biol. Chem., 275, 9120–9130.
Hawley, R.G., Lieu, F.H., Fong, A.Z., and Hawley, T.S. (1994) Versatile retroviral vectors for potential use in gene therapy. Gene Ther., 1, 136–138.[ISI][Medline]
Israel, S., Mendelovitz, M., and Honigman, A. (1995) Transactivation of human T-cell leukemia virus type 1 by herpes simplex virus type 1. Virus Genes, 9, 269–276.[CrossRef][ISI][Medline]
Jacobsson, I., Lindahl, U., Jensen, J.W., Roden, L., Prihar, H., and Feingold, D.S. (1984) Biosynthesis of heparin. Substrate specificity of heparosan N-sulfate D-glucuronosyl 5-epimerase. J. Biol. Chem., 259, 1056–1063.
Kuberan, B., Lech, M., Zhang, L., Wu, Z.L., Beeler, D.L., and Rosenberg, R.D. (2002) Analysis of heparan sulfate oligosaccharides with ion pair-reverse phase capillary high performance liquid chromatography-microelectrospray ionization time-of-flight mass spectrometry. J. Am. Chem. Soc., 124, 8707–8718.[CrossRef][ISI][Medline]
Kuberan, B., Lech, M.Z., Beeler, D.L., Wu, Z.L., and Rosenberg, R.D. (2003) Enzymatic synthesis of antithrombin III-binding heparan sulfate pentasaccharide. Nat. Biotechnol., 21, 1343–1346.[CrossRef][ISI][Medline]
Kusche, M., Backstrom, G., Riesenfeld, J., Petitou, M., Choay, J., and Lindahl, U. (1988) biosynthesis of heparin. O-sulfation of the antithrombin-binding region. J. Biol. Chem., 263, 15474–15484.
Linhardt, R. J., Wang, H. M., Loganathan, D., and Bae, J. H. (1992a) Search for the heparin antithrombin III-binding site precursor. J. Biol. Chem., 267, 2380–2387.
Linhardt, R.J., Wang, H.M., Loganathan, D., Lamb, D.J., and Mallis, L.M. (1992b) Analysis of glycosaminoglycan-derived oligosaccharides using fast-atom-bombardment mass-spectrometry. Carbohydr. Res., 225, 137–145.[CrossRef][ISI][Medline]
Liu, J., Shworak, N.W., Fritze, L.M., Edelberg, J.M., and Rosenberg, R.D. (1996) Purification of heparan sulfate D-glucosaminyl 3-O-sulfotransferase. J. Biol. Chem., 271, 27072–27082.
Liu, J., Shriver, Z., Blaiklock, P., Yoshida, K., Sasisekharan, R., and Rosenberg, R.D. (1999a) Heparan sulfate D-glucosaminyl 3-O-sulfotransferase-3A sulfates N-unsubstituted glucosamine residues. J. Biol. Chem., 274, 38155–38162.
Liu, J., Shworak, N.W., Sinay, P., Schwartz, J.J., Zhang, L., Fritze, L.M., and Rosenberg, R.D. (1999b) Expression of heparan sulfate D-glucosaminyl 3-O-sulfotransferase isoforms reveals novel substrate specificities. J. Biol. Chem., 274, 5185–5192.
McKenzie, E., Young, K., Hircock, M., Bennett, J., Bhaman, M., Felix, R., Turner, P., Stamps, A., McMillan, D., Saville, G., and others. (2003) Biochemical characterization of the active heterodimer form of human heparanase (Hpa1) protein expressed in insect cells. Biochem. J., 373, 423–435.[CrossRef][ISI][Medline]
Nagasawa, K. and Inoue, Y. (1974) Solvolytic desulfation of 2-deoxy-2-sulfoamino-D-glucose and D-glucose 6-sulfate. Carbohyd. Res., 36, 265–271.[CrossRef][ISI][Medline]
Petitou, M., Duchaussoy, P., Lederman, I., Choay, J., Jacquinet, J.C., Sinay, P., and Torri, G. (1987) Synthesis of heparin fragments: a methyl alpha-pentaoside with high affinity for antithrombin III. Carbohydr. Res., 167, 67–75.[CrossRef][ISI][Medline]
Shukla, D. and Spear, P.G. (2001) Herpesviruses and heparan sulfate: an intimate relationship in aid of viral entry. J. Clin. Invest., 108, 503–510.[CrossRef][ISI][Medline]
Shukla, D., Liu, J., Blaiklock, P., Shworak, N.W., Bai, X., Esko, J.D., Cohen, G.H., Eisenberg, R.J., Rosenberg, R.D., and Spear, P.G. (1999) A novel role for 3-O-sulfated heparan sulfate in herpes simplex virus 1 entry. Cell, 99, 13–22.[CrossRef][ISI][Medline]
Shworak, N.W., Liu, J., Fritze, L.M., Schwartz, J.J., Zhang, L., Logeart, D., and Rosenberg, R.D. (1997) Molecular cloning and expression of mouse and human cDNAs encoding heparan sulfate D-glucosaminyl 3-O-sulfotransferase. J. Biol. Chem., 272, 28008–28019.
Skaliter, R., Makhov, A.M., Griffith, J.D., and Lehman, I.R. (1996) Rolling circle DNA replication by extracts of herpes simplex virus type 1-infected human cells. J. Virol., 70, 1132–1136.[Abstract]
Sudo, M., Sato, K., Chaidedgumjorn, A., Toyoda, H., Toida, T., and Imanari, T. (2001) (1)H nuclear magnetic resonance spectroscopic analysis for determination of glucuronic and iduronic acids in dermatan sulfate, heparin, and heparan sulfate. Anal. Biochem., 297, 42–51.[CrossRef][ISI][Medline]
Wu, Z.L., Lech, M., Beeler, D.L., and Rosenberg, R.D. (2004) Determining heparan sulfate structure in the vicinity of specific sulfotransferase recognition sites by mass spectrometry. J. Biol. Chem., 279, 1861–1866.
Xia, G., Chen, J., Tiwari, V., Ju, W., Li, J.P., Malmstrom, A., Shukla, D., and Liu, J. (2002) Heparan sulfate 3-O-sulfotransferase isoform 5 generates both an antithrombin-binding site and an entry receptor for herpes simplex virus, type 1. J. Biol. Chem., 277, 37912–37919.
Zhang, L., Beeler, D. L., Lawrence, R., Lech, M., Liu, J., Davis, J.C., Shriver, Z., Sasisekharan, R., and Rosenberg, R. D. (2001a) 6-O-sulfotransferase-1 represents a critical enzyme in the anticoagulant heparan sulfate biosynthetic pathway. J. Biol. Chem., 276, 42311–42321.
Zhang, L., Lawrence, R., Schwartz, J.J., Bai, X., Wei, G., Esko, J.D., and Rosenberg, R.D. (2001b) The effect of precursor structures on the action of glucosaminyl 3-O-sulfotransferase-1 and the biosynthesis of anticoagulant heparan sulfate. J. Biol. Chem., 276, 28806–28813.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. R. Studelska, K. Giljum, L. M. McDowell, and L. Zhang Quantification of glycosaminoglycans by reversed-phase HPLC separation of fluorescent isoindole derivatives Glycobiology, January 1, 2006; 16(1): 65 - 72. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||