Glycobiology Advance Access originally published online on September 29, 2008
Glycobiology 2008 18(12):1119-1128; doi:10.1093/glycob/cwn097
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Minimum substrate requirements of endoglycosidase activities toward dermatan sulfate by electrospray ionization-tandem mass spectrometry
Lysosomal Diseases Research Unit, Department of Genetic Medicine, Children, Youth and Women's Health Service, 72 King William Road, North Adelaide, SA 5006; and Discipline of Paediatrics, University of Adelaide, Adelaide, SA 5005, Australia
1 To whom correspondence should be addressed: Tel: +61-8-8161-6741; Fax: +61-8-8161-7100; e-mail: maria.fuller{at}adelaide.edu.au
Received on August 27, 2008; revised on September 1, 2008; accepted on September 24, 2008
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterial and methodsConflict of interest statementReferences |
|---|
The catabolism of dermatan sulfate (DS) commences with endohydrolysis of the polysaccharide to oligosaccharides by proposed endo-β-N-acetylhexosaminidase and endohexuronidase activities. To investigate the substrate specificities of these activities, we developed an assay to measure specific products of their action upon oligosaccharide substrates. Tetra- to tetradecasaccharides, rich in glucuronic acid (GlcA) or iduronic acid (IdoA), were obtained from chondroitinase ABC digests of chondroitin sulfate (CS)-A and DS, respectively, separated by gel-filtration chromatography and characterized by electrospray ionization-tandem mass spectrometry (ESI-MS/MS). Endo-β-N-acetylhexosaminidase and endohexuronidase cleavage of these oligosaccharides was then assessed by incubating with cell homogenate (source of endoglycosidase activity) and measuring di- to octasaccharide products derived from the nonreducing end of the substrate by ESI-MS/MS. We found that both activities preferentially degraded the GlcA-rich substrate, with minor activity toward the IdoA-rich substrate and that a minimum of four and five monosaccharides were required on the reducing side of the target glycosidic linkage for endo-β-N-acetylhexosaminidase and endohexuronidase cleavage, respectively. Thus, the minimum-sized substrates were a hexasaccharide for endo-β-N-acetylhexosaminidase and an octasaccharide for endohexuronidase. We observed that endo-β-N-acetylhexosaminidase sequentially removed tetrasaccharides from the nonreducing end of oligosaccharides when unrestricted by substrate length, whereas endohexuronidase activity was random and comparatively low. The activities displayed acidic pH optima and were shown by subcellular fractionation to reside in lysosomes and late endosomes. We suggest that these activities represent the known Hyal-1 and endo-β-glucuronidase enzymes and that these enzymes act in concert to degrade GlcA-rich domains of DS but are less active toward regions containing IdoA.
Key words: dermatan sulfate / endo-β-glucuronidase / endoglycosidase / Hyal-1 / mass spectrometry
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsConflict of interest statementReferences |
|---|
Dermatan sulfate (DS) is a sulfated glycosaminoglycan (GAG) that is widely distributed as proteoglycan throughout the extracellular matrix where it plays an important role in many key biological processes (Trowbridge and Gallo 2002
). DS exists as chains of 40–100 repeating disaccharide units of uronic acid (UA) alternating with β-linked (1,4) D-N-acetylgalactosamine (GalNAc) residues that may be O-sulfated at C4 and/or C6. The UA may be 
-linked (1,3) iduronic acid (IdoA) or β-linked (1,3) glucuronic acid (GlcA), unsulfated or 2-O sulfated. DS is defined as chondroitin sulfate (CS) by the presence of GalNAc; the presence of variable amounts of IdoA distinguishes it from CS-A and CS-C, which contain exclusively GlcA (Trowbridge and Gallo 2002). DS forms a domain structure, whereby extended sequences containing predominantly IdoA alternate with sequences where GlcA prevails (Mitropoulou et al. 2001).
DS is degraded in a two-step process. In the first step, the polysaccharides are partially degraded by endoenzymes to produce oligosaccharides. Two different endoglycosidases have been identified that may participate in this process. Hyal-1 is an acid-active hyaluronidase (endo-β-N-acetylhexosaminidase) of the somatic tissues that principally degrades hyaluronan, but also cleaves the GalNAc-GlcA linkages of CS. An acid-active endo-β-glucuronidase that degrades GlcA–GalNAc bonds has also been described (Takagaki et al. 1988). Following partial catabolism by endoenzymes, the second step of DS degradation operates from the nonreducing end by the sequential action of up to seven lysosomal exoenzymes, reducing the oligosaccharides to monosaccharides and inorganic sulfate for transport out of the lysosome and reutilization by the cell. In a group of inherited metabolic disorders known as the mucopolysaccharidoses (MPS), a deficiency in one of these lysosomal exoenzyme activities required to degrade GAG results in the lysosomal accumulation of partially degraded GAG. Consequently, DS oligosaccharide fragments have been identified in the urine of MPS patients, with the nonreducing residue reflecting the enzyme deficiency (Fuller et al. 2004, 2006). These DS fragments are presumably the products of endoglycosidase activities, followed by exoenzyme activities that terminate at the substrate for the deficient enzyme. Oligosaccharides containing GalNAc and UA at the reducing terminus have been identified, confirming both endo-β-N-acetylhexosaminidase and endohexuronidase activities toward DS. The aim of the present study was to investigate the substrate specificities of these activities by measuring the specific products of their action upon oligosaccharide substrates.
Several approaches have been utilized to measure the products of endoenzyme activity toward DS. Unsaturated disaccharides resulting from depolymerization of DS by the bacterial chondroitinases have commonly been detected by UV absorbance at a wavelength of 232 nm (Karamanos et al. 1994), but this method is not suitable for analysis of the saturated oligosaccharide products of mammalian endoglycosidase activity. DS disaccharides have also been measured as fluorescent derivatives (Kitagawa et al. 1995; Volpi 2000). However, the measurement of specific DS oligosaccharides from complex mixtures by UV absorbance or fluorescence requires their prior separation by methods such as HPLC (Imanari et al. 1996) or capillary electrophoresis (Brown et al. 1998; Theocharis AD and Theocharis DA 2002; Yang et al. 2005). In recent years, electrospray ionization-tandem mass spectrometry (ESI-MS/MS) has become a powerful tool for the simultaneous determination of particular oligosaccharide structures without the need for purification (Desaire and Leary 2000; Desaire et al. 2001; Zaia and Costello 2001; Zaia et al. 2001; Zhang et al. 2005). For the present study, a range of oligosaccharides representing the GlcA-rich and the IdoA-rich domains of DS was prepared, and ESI-MS/MS was employed to measure the specific products of endoglycosidase activity toward these substrates.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsConflict of interest statementReferences |
|---|
Preparation and characterization of oligosaccharide substrates
Oligosaccharides representing the GlcA-rich and the IdoA-rich domains of DS were prepared from chondroitinase ABC digests of CS-A and DS, respectively, and were partially separated by gel-filtration chromatography. The chromatograms of the oligosaccharides on Bio-Gel P6 (Figure 1A) show approximately 25% of the total UA-positive material from each digest in the void volume (92–104 mL). Tetradeca- to disaccharides eluted between 116 and 204 mL. From each peak in the chromatograms that eluted within the column fractionation range, the fraction containing the highest concentration of UA was selected, and the oligosaccharides in these selected fractions were derivatized with 1-phenyl-3-methyl-5-pyrazolone (PMP) and identified based on mass-to-charge ratios (m/z) by ESI-MS. For most of the oligosaccharides, multiple sulfated species were identified. The oligosaccharides displayed characteristic multiply charged ions and as such were identified from [M-H]–1 up to [M-6H]–6, with the most abundant charge state for the oligosaccharide ions corresponding to one charge per sulfate group. A representative ESI-MS spectrum for DS oligosaccharides eluting at 164 mL (Figure 1B) shows three oligosaccharide compositions: a pentasaccharide (
UA-GalNAc-UA-GalNAc-UA) with two sulfate groups (+2S), and two hexasaccharides (UA-GalNAc-[UA-GalNAc]2 (+2-3S)). Once identified, the oligosaccharides were further characterized by ESI-MS/MS to elucidate the partial structure and, for di- to octasaccharides, to also identify a suitable product ion for multiple reaction monitoring (MRM). As previously reported, oligosaccharides with reducing-end GalNAc and UA residues fragmented to give characteristic product ions at m/z 256 and 331, respectively (Fuller et al. 2004). Figure 1C shows the product ion spectrum of the DS pentasaccharide (UA-GalNAc-UA-GalNAc-UA (+2S)), with major product ions at m/z 331.1, corresponding to internal glycosidic backbone cleavage of reducing-end UA with one PMP moiety, m/z 624.3 [M-PMP-2H]–2, and m/z 837.3, corresponding to the internal glycosidic backbone fragment UA-GalNAc-UA-GalNAc (+S) [M-H]–1.
|
A total of 13 oligosaccharides ranging in size from di- to tetradecasaccharides was identified from each of the chondroitinase ABC digests of CS-A and DS (Table I). These oligosaccharides were composed of repeating UA-GalNAc disaccharide subunits, with up to the equivalent of one sulfate per disaccharide. Two types of oligosaccharide structures were identified; the first type had an even number of saccharide residues with
UA at the nonreducing end and GalNAc at the reducing end ("even" oligosaccharides) and the second type was composed of an odd number of residues with UA at the reducing terminus ("odd" oligosaccharides). The number of sulfates associated with the oligosaccharides increased with the size of the oligosaccharide, up to the equivalent of one sulfate per disaccharide. The nonreducing UA residue of the CS-A and DS oligosaccharides eluted from the Bio-Gel P6 column was not a substrate for lysosomal exoenzymes, as indicated by the complete resistance of the CS-A tetrasaccharide to digestion with bovine liver β-glucuronidase (data not shown), and these oligosaccharides were thus suitable for use as substrates for the endoglycosidase assay.
|
Substrate specificity of endo-β-N-acetylhexosaminidase activity
Endo-β-N-acetylhexosaminidase substrate specificity was investigated by incubating tetra- to tetradecasaccharides from CS-A and DS (Table I) with Chinese hamster ovary (CHO)-K1 homogenate (source of endoglycosidase activity) and measuring the even oligosaccharide products derived from the nonreducing end of the substrate by ESI-MS/MS, using MRM. The abbreviations used in reference to the oligosaccharides are listed in Table I. Figure 2A shows that after incubation with the GlcA-rich CS-A
tetra(+2S) substrate, no products were detected, i.e., CS-A tetra(+2S) was not digested. However, the slightly larger CS-A hexa(+3S) substrate was digested, generating a small amount of di(+S) exclusively. Increasing the length of the substrate by one disaccharide (CS-A hexa(+3S) to octa(+4S)) resulted in a 7-fold increase in the di(+S) product and the appearance of the tetra(+2S) product. The addition of another disaccharide to the length of the substrate did not further alter the level of the di(+S) product; it did, however, result in a 9-fold increase in the tetra(+2S) product. The amounts of both di(+S) and tetra(+2S) products were unchanged with subsequent disaccharide increases in substrate length. Small quantities of hexa(+2S), hexa(+3S), and octa(+3S) products were also detected from the CS-A dodeca(+6S) substrate. The levels of these products increased slightly as substrate length was increased by another disaccharide (CS-A tetradeca(+7S)). Incubation with the IdoA-rich substrate (DS dodeca(+6S)) produced di(+S), tetra(+2S), hexa(+2S), and hexa(+3S) in quantities equivalent to only 14%, 7%, 8%, and 16%, respectively, of those generated from CS-A dodeca(+6S) (Figure 2B).
|
Based on these findings, the sensitivities of the N-acetylhexosaminic linkages of the oligosaccharide substrates to endo-β-N-acetylhexosaminidase cleavage were established (Table II). Whereas the activity cleaved CS-A
hexa(+3S) at the ultimate nonreducing-end N-acetylhexosaminic bond exclusively, the larger CS-A substrates were also cleaved at the penultimate, and CS-A dodeca(+6S) and tetradeca(+7S) were further degraded at the antepenultimate and preantepenultimate linkages. Minor cleavage of DS dodeca(+6S) occurred at each of the ultimate, penultimate, and antepenultimate N-acetylhexosaminic linkages from the nonreducing end.
|
Substrate specificity of endohexuronidase activity
To investigate endohexuronidase substrate specificity, hexa- to tetradecasaccharides from CS-A and DS (Table I) were incubated with CHO-K1 homogenate, and odd oligosaccharide products derived from the nonreducing end were measured by ESI-MS/MS. The use of CS-A
tetra(+2S) as a substrate for the endohexuronidase was prevented by high levels of tri(+S) impurities in the CS-A tetra(+2S) substrate preparation. These impurities resulted from coelution of tri(+S) and tetra(+2S) from the Bio-Gel P6 column (Figure 1A and Table I). Figure 3A illustrates that when the GlcA-rich CS-A hexa(+3S) was used as a substrate, no digestion was observed and that whilst both CS-A octa(+4S) and deca(+5S) generated exclusively tri(+S), 7-fold more was detected from the latter. Further increasing the length of the substrate had no effect upon the level of tri(+S). However, it did result in the generation of penta(+2S) and hepta(+3S) products. The quantities of penta(+2S) and hepta(+3S) products increased marginally as substrate length was increased by a disaccharide (CS-A dodeca(+6S) to tetradeca(+7S)). The IdoA-rich DS dodeca(+6S) produced tri(+S) and penta(+2S) in amounts equivalent to 6% and 19%, respectively, of those generated from CS-A dodeca(+6S) (Figure 3B).
|
Table II indicates the uronidic linkages in the oligosaccharides that were degraded by the endohexuronidase activity. While CS-A
octa(+4S) and deca(+5S) were digested only at the penultimate nonreducing-end uronidic bond, the larger CS-A substrates were additionally cleaved at the antepenultimate and preantepenultimate linkages. Some degradation of DS dodeca(+6S) occurred at the penultimate and antepenultimate nonreducing uronidic linkages.
Functional properties of endoglycosidase activities
The influence of pH and buffer conditions upon the endo-β-N-acetylhexosaminidase and endohexuronidase activities was examined by incubating CS-A substrates with CHO-K1 homogenate under a range of conditions and measuring the products by ESI-MS/MS. CS-A dodeca(+6S) and deca(+5S) were selected as substrates for the pH and buffer experiments, respectively, as these were of sufficient length for the maximum generation of the major oligosaccharide products, i.e., di(+S), tri(+S), and tetra(+2S) (Figures 2A and 3A). Both activities degraded the CS-A dodeca(+6S) substrate only within a narrow acidic pH range, with maximum product generation occurring at around pH 3.5 and none above pH 4.0 (Figure 4). Table III shows that the generation of tri(+S) and tetra(+2S) products was maximum when the CS-A deca(+5S) substrate was incubated in the presence of sodium formate, compared to sodium acetate and 3,3-dimethylglutaric acid (DMG) buffers. Increasing the ionic strength of the sodium formate and DMG buffers from 50 to 100 mM increased the generation of these products; however, their levels were considerably reduced when the strength of the sodium acetate buffer was correspondingly increased.
|
|
Subcellular localization of endoglycosidase activities
The subcellular location of the endoglycosidase activities was determined by fractionation on a Percoll gradient. Figure 5A shows that the endosomes were contained in fractions 5–10, and lysosomes in fractions 13–20, as determined by the activities of the organelle markers, acid phosphatase, and β-hexosaminidase, respectively. The fractions containing microsomes (1–4), endosomes (5–10), and lysosomes (13–20) were each pooled and concentrated. Each combined organelle fraction was incubated with the CS-A
tetradeca(+7S) substrate and the products were measured by ESI-MS/MS. Endo-β-N-acetylhexosaminidase and endohexuronidase activities were detected in all three preparations, as determined by the generation of tri(+S) and tetra(+2S) products, with the generation of both products in the lysosome preparation approximately 3- to 7-fold and 2-fold higher than in the microsome and endosome preparations, respectively (Figure 5B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsConflict of interest statementReferences |
|---|
Characterization of oligosaccharide substrates
ESI-MS/MS analysis of the size-fractionated oligosaccharide substrates identified di- to tetradecasaccharides with an even and odd number of saccharide residues and up to the equivalent of one sulfate per disaccharide (Table I). These structures are consistent with the uniform GlcA-GalNAc(+S) and IdoA-GalNAc(+S) repeating disaccharide sequences that are reported to account for over 90% of the polysaccharide CS-A from bovine trachea (Lauder et al. 2000
) and DS from porcine intestinal mucosa (Sudo et al. 2001), respectively. Hence, the oligosaccharides were deemed representative of the GlcA-rich and IdoA-rich domains of a typical polysaccharide DS.
Based on the specificity of chondroitinase ABC for GalNAc-UA bonds (Yamagata et al. 1968), oligosaccharides with nonreducing-end UA and reducing-end GalNAc residues were expected. Oligosaccharides with UA at the nonreducing end and UA at the reducing end may represent the products of tissue endohexuronidase activity or oligosaccharides resulting from decomposition of the polysaccharide chains during commercial processing that have been further depolymerized by chondroitinase ABC. An oligosaccharide structure corresponding to the CS-A trisaccharide, UA-GalNAc-GlcA(+S), has previously been observed after the digestion of CS with chondroitinase ABC (Sugahara et al. 1994; Lauder et al. 2000).
Substrate specificity of endo-β-N-acetylhexosaminidase activity
The endo-β-N-acetylhexosaminidase activity showed a marked preference for the oligosaccharide substrate rich in GlcA, compared to IdoA, as evidenced by the relative resistance of IdoA-rich DS oligosaccharide to digestion (Figure 2B). This most likely reflects specificity for GalNAc-GlcA rather than GalNAc-IdoA glycosidic linkages, with the small amounts of products observed after incubation with the DS oligosaccharide reflecting the specific cleavage of the approximately 1 in 10 GalNAc-UA linkages of porcine intestinal mucosa DS that contain GlcA (Sudo et al. 2001). However, as the structures of the products derived from the reducing side of the endo-β-N-acetylhexosaminidase cleavage sites are not apparent from these data, the nonreducing products observed from the DS oligosaccharide could conceivably also result from, for example, a slow cleavage of the GalNAc-IdoA linkages, or from cleavage at specific, infrequent subsets of such linkages. In any case, these findings indicate that endohydrolysis of polysaccharide DS occurs predominantly in the GlcA-rich domains, where the GalNAc-GlcA linkages are concentrated.
The ultimate reducing-end N-acetylhexosaminic linkage of each CS-A oligosaccharide was resistant to cleavage by the endo-β-N-acetylhexosaminidase activity (Table II), which implies an absolute requirement for a minimum of two disaccharides on the reducing side of the target GalNAc, and the minimum-sized substrate requirement was thus a hexasaccharide (Figure 2A). The maximum generation of a specific product required at least three disaccharides on the reducing side of the target GalNAc. For example, although some di(+S) product could be liberated from the CS-A hexa(+3S) substrate by endohydrolysis of the first GalNAc-UA linkage from the nonreducing end according to the two-disaccharide minimum requirement, production of di(+S) reached a maximum when at least three disaccharides were present to the reducing side of this linkage, i.e., in the octa(+4S) and larger substrates (Figure 2A).
As deuterated internal standards were not available for each of the oligosaccharide species under investigation, the levels of the oligosaccharide products were calculated relative to a disaccharide internal standard, UA (1,4) N-acetylglucosamine-6-sulfate-d3 (UA-GlcNAc(+S)(d3)). A direct comparison between the relative levels of different oligosaccharides is complicated by their varying response factors when determined by ESI-MS/MS (Rozaklis et al. 2002). Analysis of the CS-A oligosaccharide substrates showed that each disaccharide addition to an oligosaccharide resulted in a successive decrease in response of approximately 50% (data not shown). Therefore, to allow a more accurate comparison between product levels, the relative levels of the products larger than di(+S) were successively doubled for each disaccharide equivalent increase in chain length. After this correction for approximate response factors, the major endo-β-N-acetylhexosaminidase product from CS-A hexa(+3S) and octa(+4S) substrates was di(+S), and from the CS-A substrates larger than octa(+4S), the major product was tetra(+2S) (data not shown). This indicates that notwithstanding restrictions imposed by the length of the substrate, as discussed above, the endo-β-N-acetylhexosaminidase activity preferentially cleaves the second N-acetylhexosaminic linkage from the nonreducing end of oligosaccharide substrates rather than other N-acetylhexosaminic bonds to liberate tetrasaccharides as the main reaction products.
The endo-β-N-acetylhexosaminidase activity toward GalNAc-UA linkages described here is consistent with the reported ability of the Hyal-1 hyaluronidase enzyme to cleave the GalNAc-GlcA linkages of CS (Stern 2003). Although the minimum-sized hyaluronan substrate requirement of recombinant human Hyal-1 was recently reported as an octasaccharide (Hofinger, Bernhardt, et al. 2007), this may simply reflect substrate-specific size minima. Our data indicate that Hyal-1 preferentially hydrolyses tetrasaccharides from the nonreducing end of oligosaccharide substrates, similar to a mechanism previously described for the PH20 hyaluronidase from bovine testes (Takagaki et al. 1994).
Substrate specificity of endohexuronidase activity
Likewise, the endohexuronidase activity preferentially cleaved the oligosaccharide substrate rich in GlcA, compared to IdoA (Figure 3B), indicating that activity toward DS polysaccharide would be directed principally to the GlcA-rich domains. This suggests specificity for GlcA-GalNAc rather than IdoA-GalNAc glycosidic linkages, but as the reducing residue of the oligosaccharide products measured could not be distinguished by ESI-MS/MS, this cannot be stated absolutely. No less than five monosaccharides were required on the reducing side of the target UA for endohydrolysis of the CS-A substrate (Table II), and the minimum-sized substrate requirement was thus an octasaccharide (Figure 3A). Although unlikely, the endohexuronidase activity could conceivably have cleaved the first glycosidic linkage from the nonreducing end of the substrates to liberate the monosaccharide, UA, which was not monitored in these experiments as monosaccharides would not bind to the solid-phase extraction columns used to extract GAG from the digestion mixtures. The maximum generation of a specific product required at least seven monosaccharides on the reducing side of the target UA (Figure 3A). Correction of the endohexuronidase product levels for their approximate response factors, as described above, indicated that the digestion of CS-A octa(+4S) and deca(+5S) generated only tri(+S). Subsequent additions to the length of the substrate resulted in increasing production of penta(+2S) and hepta(+3S) also, such that tri(+S), penta(+2S), and hepta(+3S) products were detected in equal amounts from the tetradeca(+7S) substrate (data not shown). However, the combined levels of the products of the endohexuronidase digestion of tetradeca(+7S) represented only 
15% of that of the tetra(+2S) produced by endo-β-N-acetylhexosaminidase. Therefore, the endohexuronidase activity appears a relatively minor activity that randomly cleaves UA-GalNAc linkages when unrestricted by substrate length. The substrate specificity of the endohexuronidase activity described here is consistent with the CS-degrading endo-β-glucuronidase enzyme described by Takagaki et al. (1988).
Properties of endoglycosidase activities
The pH profiles of the endo-β-N-acetylhexosaminidase and endohexuronidase activities (Figure 4) closely resembled those previously reported for Hyal-1 (Orkin and Toole 1980; Muckenschnabel et al. 1998; Hofinger, Spickenreither, et al. 2007) and endo-β-glucuronidase (Takagaki et al. 1988) and suggested an acidic milieu for both activities. Subcellular fractionation indicated that endohydrolysis occurs predominantly in the lysosomes and possibly late endosomes (Figure 5B), which collectively are known to contain the bulk of the acid hydrolase pool in most cell types (Storrie 1988; Claus et al. 1998). Activity in the microsome preparation presumably represents newly synthesized enzyme originating from the endoplasmic reticulum en route to the Golgi complex. Inhibition of Hyal-1 by sodium acetate (Table III) has previously been reported (Orkin and Toole 1980) and indicates that the use of sodium acetate as a digestion buffer for the hyaluronidases should be avoided.
Conclusions
Our results suggest that both endo-β-N-acetylhexosaminidase and endohexuronidase activities, probably representing the Hyal-1 and endo-β-glucuronidase enzymes described previously, preferentially degrade the GlcA-rich domains of DS, with only limited activity toward the regions of DS that contain IdoA. Endohydrolysis of DS appears to commence in the late endosome and/or lysosome with cleavages by these activities to generate two types of oligosaccharide intermediates. The first type of intermediate is GlcA-rich and results from cleavages within the susceptible domains of the polysaccharide; these GlcA-rich oligosaccharides are further endohydrolyzed, prior to complete degradation by lysosomal exohydrolases. The second type of intermediate is larger, IdoA-rich, and represents the regions of the polysaccharide resistant to the initial cleavages; this type of oligosaccharide is not a good substrate for endohydrolysis and is predominantly degraded by the lysosomal exohydrolases.
Although the endo-β-N-acetylhexosaminidase and endohexuronidase activities described here share a number of biochemical properties, it is likely that the endohexuronidase activity represents that of the discrete endo-β-glucuronidase enzyme reported by Takagaki et al. (1988) rather than one enzyme with dual specificity, since a recent report on the catalytic behavior of recombinant human Hyal-1 showed no evidence of endo-β-glucuronidase activity toward hyaluronan oligosaccharides (Hofinger, Bernhardt, et al. 2007). An extension of the present work to investigate endo-β-N-acetylhexosaminidase and endohexuronidase activities independent of one another (e.g., in Hyal-1-deficient human skin fibroblasts (Natowicz et al. 1996)) will allow more conclusive estimations of their substrate specificities.
Our strategy, utilizing ESI-MS/MS for the determination of specific endoglycosidase product oligosaccharides, has enabled the measurement and characterization of these activities toward DS. The distinct advantage of this approach is that multiple products differing in structure by as little as one sulfate group can be resolved and relative levels measured simultaneously by MRM without the need for prior chromatography. In addition, the preparation of exoenzyme-resistant unsaturated oligosaccharides for use as substrates is simple and enables the selective monitoring of nonreducing-end products. Application of this approach to measure and characterize endoglycosidase activities in MPS may provide insight into the mechanisms responsible for the array of stored oligosaccharide structures, which may contribute to disease pathology through their unique biological activities.
|
Material and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsConflict of interest statementReferences |
|---|
Materials
Cell culture products were obtained from MP Biomedicals Inc. (Aurora, OH) and JRH Biosciences (Lenexa, KS). CS-A from bovine trachea, DS from porcine intestinal mucosa, chondroitinase ABC from Proteus Vulgaris (EC 4.2.2.4 [EC] ), tributylamine (>98.5%), BSA, leupeptin, pepstatin, PMSF, 4-methylumbelliferyl-N-acetyl-β- D-glucosaminide, 4-methylumbelliferyl-phosphate, and 4-deoxy-L-threo-hex-4-enopyranosyluronic (1,4) glucosamine-6-sulfate were from Sigma-Aldrich (St. Louis, MO). Percoll was purchased from Amersham Biosciences (Uppsala, Sweden). Acetonitrile and chloroform were of HPLC grade and supplied by Ajax FineChem (Seven Hills, NSW, Australia). Bio-Gel P6 was from Bio-Rad (Hercules, CA). Solid-phase quaternary amine extraction columns (200 mg/3 mL) were obtained from United Chemical Technologies (Bristol, PA). PMP was purchased from Tokyo Kasei Kogyo (Tokyo, Japan). The HPLC column was a 3 µm Alltima C18-LL column (50 x 2.1 mm) from Alltech Associates (Deerfield, IL), with a 2 mm EXSIL ODS 5 µm guard column from SGE (Austin, TX).
Cell culture
CHO-K1 cells were obtained from the American Type Culture Collection and maintained in 75 cm2 culture flasks in Ham's F12 supplemented with 10% (v/v) FCS at 37°C in a humidified atmosphere containing 5% CO2. The medium was replenished every 2–3 days. At confluence, cells were trypsinized and harvested by centrifugation for 5 min at 1000 x g. The cell pellet was washed twice with PBS, resuspended in DMG (pH 2.5–7.5), HCOONa (pH 3.5) or CH3COONa (pH 3.5), at 50 or 100 mM, and sonicated for 10 s at 4°C. Total cell protein was determined by the method of Lowry et al. (1951).
Cell fractionation
Three flasks of CHO-K1 cells were harvested as described above, pooled, and then resuspended in 1.5 mL of sucrose solution (250 mM sucrose, 1 mM EDTA, 1 µM pepstatin, 1 µM leupeptin, 200 µM PMSF, 10 mM HEPES, pH 7.0). The cell suspension was drawn into a 5 mL syringe using a 23-gauge needle, subjected to three hypobaric shocks, and then centrifuged at 170 x g for 5 min to remove cellular debris. The postnuclear supernatant was diluted to 3 mL with sucrose solution, loaded onto 17 mL of 18% (v/v) Percoll in sucrose solution, and then centrifuged at 29,400 x g for 1 h at 4°C. Following centrifugation, 20 x 1 mL fractions were taken from the top of the gradient and each fraction was assayed for the organelle marker enzymes β-hexosaminidase and acid phosphatase using the fluorogenic substrates 4-methylumbelliferyl-N-acetyl-β-D-glucosaminide and 4-methylumbelliferyl-phosphate, respectively (Leaback and Walker 1961; Kolodny and Mumford 1976). One milliliter of 0.5 M NaCl/0.2% (v/v) nonidet P40 was added to the fractions containing microsomes, endosomes, and lysosomes, and these fractions were then subjected to seven freeze/thaw cycles and centrifuged at 100,000 x g to remove Percoll. The microsomal, endosomal, and lysosomal fractions were then each pooled and concentrated to 0.2 mL in an Amicon Ultra centrifugal filtration device (30,000 nominal molecular weight limit) (Millipore, Bedford, MA) and dialyzed against 0.1 M HCOONa, pH 3.5, at 4°C to remove the residual sucrose solution.
Preparation of oligosaccharide substrates
Oligosaccharides were prepared from CS-A and DS by limited digestion with chondroitinase ABC. For CS-A digestion, 300 mg was dissolved in 20 mL of 50 mM Tris–HCl/60 mM CH3COONa/0.02% (w/v) BSA (pH 8.0) and digested with 1.5 U chondroitinase ABC at 37°C for 50 min. For DS digestion, 100 mg was dissolved in 20 mL of 50 mM Tris–HCl/60 mM CH3COONa/0.02% (w/v) BSA (pH 8.0) and digested with 0.75 U chondroitinase ABC at 37°C for 50 min. Digests were terminated by boiling in a 100°C water bath for 15 min, lyophilized, and reconstituted in 4 mL of 0.5 M HCOONH4. Di- to tetradecasaccharides were then size-fractionated on a Bio-Gel P6 column (170 x 1.5 cm) in 0.5 M HCOONH4. Seventy fractions of 4 mL were collected and assayed for UA equivalents by the method of Blumenkrantz and Asboe-Hansen (1973).
Endoglycosidase product assay
To measure endoglycosidase products, 50 nmol of oligosaccharide substrate was added to samples containing endoglycosidase activity in a final volume of 100 µL and the reaction incubated at 37°C for 24 h. The samples containing endoglycosidase activity were cell homogenate (2.5 mg protein) or microsomal, endosomal, and lysosomal fractions. Negative controls, in which substrate was added to the samples after the 24-h incubation period, were included to correct for background interference. Following incubation, the samples were lyophilized and then resuspended in 1.5 mL of 100 mM CH3COONH4 (pH 5.0) containing 2 nmol of the internal standard. The internal standard used was a disaccharide, UA-GlcNAc(+S)(d3), prepared by N-acetylation of 4-deoxy-L-threo-hex-4-enopyranosyluronic (1,4) glucosamine-6-sulfate with deuterated acetic anhydride according to the method of Ramsay et al. (2003). Oligosaccharides were extracted on solid-phase quaternary amine extraction columns (200 mg/3 mL). The columns were primed with 3 mL each of CH3OH, H2O, and 100 mM CH3COONH4 (pH 5.0) after which the 1.5 mL sample was applied and allowed to enter the solid phase completely. Columns were washed with 2 x 3 mL of 100 mM CH3COONH4 (pH 5.0) and dried thoroughly on a Supelco Visiprep 24 vacuum manifold. Samples were then eluted from the columns with 500 µL of 1.2 M LiCl/100 mM CH3COONH4 (pH 5.0) and lyophilized.
Preparation of oligosaccharides for mass spectrometry
Oligosaccharides from the Bio-Gel P6 column (100 µg UA equivalents) and from the endoglycosidase product assay were derivatized with PMP as described previously (Fuller et al. 2006). PMP-oligosaccharides from the Bio-Gel P6 column were resuspended in 50% (v/v) CH3CN/0.025% (v/v) HCOOH in H2O for analysis by mass spectrometry. PMP-oligosaccharides from the endoglycosidase product assay were resuspended in 100 µL of 2.4% (v/v) tributylamine in H2O and desalted using an HPLC column. The loading solvent (1% (v/v) CH3CN in H2O) and the eluting solvent (80% (v/v) CH3CN in H2O) both contained 16 mM CH3COONH4, 24 mM acetic acid, and 0.12% (v/v) tributylamine and were delivered at 0.25 mL/min using an Agilent 1100 binary HPLC pump. Samples (20 µL) were injected with a Gilson 233 autosampler and loaded onto the column. After 1.0 min, oligosaccharides were eluted directly into the ESI source for quantification; di- to octasaccharides coeluted from the column and were detected at 4.0 min. Salts eluting from the column from 0 to 2.5 min were diverted to waste via a Valco switching valve. The column was reequilibrated with loading solvent from 5.1 to 8.0 min.
Mass spectrometry of oligosaccharides
Oligosaccharide analysis was performed by ESI-MS/MS in the negative-ion mode using a PE Sciex API 3000 triple-quadrupole mass spectrometer equipped with a turbo-ionspray source and Analyst 1.4 software. Nitrogen was used as nebulizer (NEB), curtain (CUR), and collision (CAD) gas. Optimization of instrument parameters and structural characterization of oligosaccharides were achieved by directly infusing PMP-oligosaccharide solutions using a Harvard Apparatus pump at 10 µL/min with NEB and CUR gas flows set at 10 and 6, respectively, ion spray voltage (IS) set at –4500 V and temperature (TEM) set at 200°C. Oligosaccharides were identified on the basis of m/z by ESI-MS scans (300–1000 amu in 1 s) and declustering, focusing, and entrance potentials were optimized for each. For ESI-MS/MS (product ion) analysis, the collision energy was ramped from –130 to –4 in 4 V increments with the collision cell exit potential set at –15 V, while Q3 was scanned from 0 to 1500 amu in 3 s. The collision energy and collision cell exit potential were then optimized for selected product ions.
Relative quantification of PMP-oligosaccharides was performed using the MRM mode, with NEB, CUR, and CAD set at 12, 6 and 4, respectively, IS set at –4000 V and TEM set at 400°C. The MRM transitions monitored were m/z 788/534 (di(+S)), 964/331 (tri(+S)), 632/496 (tetra(+2S)), 711/ 624 (penta(+2S)), 813/506 (hexa(+2S)), 568/546 (hexa(+3S)), 627/648 (hepta(+3S)), 694/610 (octa(+3S)), 540/516 (octa(+4S)), and 791/259 (UA-GlcNAc(+S)(d3)) (ISTD). Each MRM pair was monitored for 100 ms at unit resolution; for each measurement, consecutive scans over the injection period were averaged. Relative oligosaccharide levels were determined by relating the peak heights of the PMP-oligosaccharides to the peak height of the internal standard.
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsConflict of interest statementReferences |
|---|
None declared.
|
Acknowledgements |
|---|
The authors thank Tomas Rozek and Kerryn Mason for technical advice.
|
Footnotes |
|---|
2 Present address: Baker Heart Research Institute, 75 Commercial Road, Melbourne, Victoria 3006, Australia
|
Abbreviations |
|---|
CHO, Chinese hamster ovary; CS, chondroitin sulfate; DMG, 3,3-dimethylglutaric acid; DS, dermatan sulfate; ESI, electrospray ionization; GAG, glycosaminoglycan; GalNAc, N-acetylgalactosamine; GlcA, glucuronic acid; GlcNAc, N-acetylglucosamine; IdoA, iduronic acid; MPS, mucopolysaccharidosis; MRM, multiple reaction monitoring; MS/MS, tandem mass spectrometry; MS, mass spectrometry; m/z, mass-to-charge; PMP, 1-phenyl-3-methyl-5-pyrazolone; S, sulfate; UA, uronic acid
UA, unsaturated uronic acid |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterial and methodsConflict of interest statementReferences |
|---|
Blumenkrantz N, Asboe-Hansen G. New method for quantitative determination of uronic acids. Anal Biochem (1973) 54:484–489.[CrossRef][Web of Science][Medline]
Brown MP, West LA, Merritt KA, Plaas AH. Changes in sulfation patterns of chondroitin sulfate in equine articular cartilage and synovial fluid in response to aging and osteoarthritis. Am J Vet Res (1998) 59:786–791.[Web of Science][Medline]
Claus V, Jahraus A, Tjelle T, Berg T, Kirschke H, Faulstich H, Griffiths G. Lysosomal enzyme trafficking between phagosomes, endosomes, and lysosomes in J774 macrophages. Enrichment of cathepsin H in early endosomes. J Biol Chem (1998) 273:9842–9851.
Desaire H, Leary JA. Detection and quantification of the sulfated disaccharides in chondroitin sulfate by electrospray tandem mass spectrometry. J Am Soc Mass Spectrom (2000) 11:916–920.[CrossRef][Web of Science][Medline]
Desaire H, Sirich TL, Leary JA. Evidence of block and randomly sequenced chondroitin polysaccharides: Sequential enzymatic digestion and quantification using ion trap tandem mass spectrometry. Anal Chem (2001) 73:3513–3520.[Medline]
Fuller M, Chau A, Nowak RC, Hopwood JJ, Meikle PJ. A defect in exodegradative pathways provides insight into endodegradation of heparan and dermatan sulfates. Glycobiology (2006) 16:318–325.
Fuller M, Meikle PJ, Hopwood JJ. Glycosaminoglycan degradation fragments in mucopolysaccharidosis I. Glycobiology (2004) 14:443–450.
Hofinger ES, Bernhardt G, Buschauer A. Kinetics of Hyal-1 and PH-20 hyaluronidases: Comparison of minimal substrates and analysis of the transglycosylation reaction. Glycobiology (2007) 17:963–971.
Hofinger ES, Spickenreither M, Oschmann J, Bernhardt G, Rudolph R, Buschauer A. Recombinant human hyaluronidase Hyal-1: Insect cells versus Escherichia coli as expression system and identification of low molecular weight inhibitors. Glycobiology (2007) 17:444–453.
Imanari T, Toida T, Koshiishi I, Toyoda H. High-performance liquid chromatographic analysis of glycosaminoglycan-derived oligosaccharides. J Chromatogr A (1996) 720:275–293.
Karamanos NK, Syrokou A, Vanky P, Nurminen M, Hjerpe A. Determination of 24 variously sulfated galactosaminoglycan- and hyaluronan-derived disaccharides by high-performance liquid chromatography. Anal Biochem (1994) 221:189–199.[CrossRef][Web of Science][Medline]
Kitagawa H, Kinoshita A, Sugahara K. Microanalysis of glycosaminoglycan-derived disaccharides labeled with the fluorophore 2-aminoacridone by capillary electrophoresis and high-performance liquid chromatography. Anal Biochem (1995) 232:114–121.[CrossRef][Web of Science][Medline]
Kolodny EH, Mumford RA. Human leukocyte acid hydrolases: Characterization of eleven lysosomal enzymes and study of reaction conditions for their automated analysis. Clin Chim Acta (1976) 70:247–257.[CrossRef][Web of Science][Medline]
Lauder RM, Huckerby TN, Nieduszynski IA. A fingerprinting method for chondroitin/dermatan sulfate and hyaluronan oligosaccharides. Glycobiology (2000) 10:393–401.
Leaback DH, Walker PG. Studies on glucosaminidase: 4. The fluorimetric assay of N-acetyl-beta-glucosaminidase. Biochem J (1961) 78:151–156.[Web of Science][Medline]
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem (1951) 193:265–275.
Mitropoulou TN, Lamari F, Syrokou A, Hjerpe A, Karamanos NK. Identification of oligomeric domains within dermatan sulfate chains using differential enzymic treatments, derivatization with 2-aminoacridone and capillary electrophoresis. Electrophoresis (2001) 22:2458–2463.[Medline]
Muckenschnabel I, Bernhardt G, Spruss T, Dietl B, Buschauer A. Quantitation of hyaluronidases by the Morgan–Elson reaction: Comparison of the enzyme activities in the plasma of tumor patients and healthy volunteers. Cancer Lett (1998) 131:13–20.[CrossRef][Web of Science][Medline]
Natowicz MR, Short MP, Wang Y, Dickersin GR, Gebhardt MC, Rosenthal DI, Sims KB, Rosenberg AE. Clinical and biochemical manifestations of hyaluronidase deficiency. N Engl J Med (1996) 335:1029–1033.
Orkin RW, Toole BP. Isolation and characterization of hyaluronidase from cultures of chick embryo skin- and muscle-derived fibroblasts. J Biol Chem (1980) 255:1036–1042.
Ramsay SL, Meikle PJ, Hopwood JJ. Determination of monosaccharides and disaccharides in mucopolysaccharidoses patients by electrospray ionisation mass spectrometry. Mol Genet Metab (2003) 78:193–204.[CrossRef][Web of Science][Medline]
Rozaklis T, Ramsay SL, Whitfield PD, Ranieri E, Hopwood JJ, Meikle PJ. Determination of oligosaccharides in Pompe disease by electrospray ionization tandem mass spectrometry. Clin Chem (2002) 48:131–139.
Stern R. Devising a pathway for hyaluronan catabolism: Are we there yet? Glycobiology (2003) 13:105–115.[CrossRef]
Storrie B. Assembly of lysosomes: Perspectives from comparative molecular cell biology. Int Rev Cytol (1988) 111:53–105.[Medline]
Sudo M, Sato K, Chaidedgumjorn A, Toyoda H, Toida T, Imanari T. 1H nuclear magnetic resonance spectroscopic analysis for determination of glucuronic and iduronic acids in dermatan sulfate, heparin, and heparan sulfate. Anal Biochem (2001) 297:42–51.[CrossRef][Web of Science][Medline]
Sugahara K, Takemura Y, Sugiura M, Kohno Y, Yoshida K, Takeda K, Khoo KH, Morris HR, Dell A. Chondroitinase ABC-resistant sulfated trisaccharides isolated from digests of chondroitin/dermatan sulfate chains. Carbohydr Res (1994) 255:165–182.[CrossRef][Web of Science][Medline]
Takagaki K, Nakamura T, Izumi J, Saitoh H, Endo M, Kojima K, Kato I, Majima M. Characterization of hydrolysis and transglycosylation by testicular hyaluronidase using ion-spray mass spectrometry. Biochemistry (1994) 33:6503–6507.
Takagaki K, Nakamura T, Majima M, Endo M. Isolation and characterization of a chondroitin sulfate-degrading endo-beta-glucuronidase from rabbit liver. J Biol Chem (1988) 263:7000–7006.
Theocharis AD, Theocharis DA. High-performance capillary electrophoretic analysis of hyaluronan and galactosaminoglycan-disaccharides in gastrointestinal carcinomas. Differential disaccharide composition as a possible tool-indicator for malignancies. Biomed Chromatogr (2002) 16:157–161.[Medline]
Trowbridge JM, Gallo RL. Dermatan sulfate: New functions from an old glycosaminoglycan. Glycobiology (2002) 12:117R–125R.
Volpi N. Hyaluronic acid and chondroitin sulfate unsaturated disaccharides analysis by high-performance liquid chromatography and fluorimetric detection with dansylhydrazine. Anal Biochem (2000) 277:19–24.[CrossRef][Web of Science][Medline]
Yamagata T, Saito H, Habuchi O, Suzuki S. Purification and properties of bacterial chondroitinases and chondrosulfatases. J Biol Chem (1968) 243:1523–1535.
Yang Y, Breadmore MC, Thormann W. Analysis of the disaccharides derived from hyaluronic acid and chondroitin sulfate by capillary electrophoresis with sample stacking. J Sep Sci (2005) 28:2381–2389.[Medline]
Zaia J, Costello CE. Compositional analysis of glycosaminoglycans by electrospray mass spectrometry. Anal Chem (2001) 73:233–239.[Medline]
Zaia J, McClellan JE, Costello CE. Tandem mass spectrometric determination of the 4S/6S sulfation sequence in chondroitin sulfate oligosaccharides. Anal Chem (2001) 73:6030–6039.[Medline]
Zhang Y, Conrad AH, Tasheva ES, An K, Corpuz LM, Kariya Y, Suzuki K, Conrad GW. Detection and quantification of sulfated disaccharides from keratan sulfate and chondroitin/dermatan sulfate during chick corneal development by ESI-MS/MS. Invest Ophthalmol Vis Sci (2005) 46:1604–1614.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||