Glycobiology Advance Access originally published online on March 23, 2007
Glycobiology 2007 17(7):735-743; doi:10.1093/glycob/cwm032
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press 2007
Successive glycosyltransfer of sialic acid by Escherichia coli K92 polysialyltransferase in elongation of oligosialic acceptors
Laboratory of Bacterial Polysaccharides, Center for Biologics Evaluation and Research, FDA, Bethesda, MD 20892
1 To whom correspondence should be addressed; Tel: +301 496 2008; Fax: +301 402 2776; E-mail: wvann{at}helix.nih.gov
Received on November 10, 2006; revised on February 19, 2007; accepted on March 12, 2007
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Escherichia coli K92 produces a capsular polysialic acid with alternating
2,8 2,9 NeuNAc linkages. This polysaccharide is cross-reactive with the neuroinvasive pathogen Neisseria meningitidis Group C. The K92 polysialyltransferase (PST) catalyzes the synthesis of the polysialic acid with alternating linkages by the transfer of NeuNAc from CMP-NeuNAc to the nonreducing end of the growing polymer. We used a fluorescent-based high-performance liquid chromatography assay to characterize the process of chain extension. The PST elongates the acceptor GT3-FCHASE in a biphasic fashion. The initial phase polymers are characterized by accumulation of product containing 1–8 additional sialic acid residues. This phase is followed by a very rapid formation of high-molecular weight (MW) polymer as the accumulated oligosaccharides containing 8–10 sialic acids are consumed. The high-MW polymer contains 90–100 sialic acids and is sensitive to degradation by periodate and K1–5 endoneuraminidase, suggesting that the polymer contains the alternating structure. The polymerization reaction does not appear to be strictly processive, since oligosaccharides of each intermediate size were detected before accumulation of high-molecular weight polymer. Synthesis can be blocked by CMP-9-azido-NeuNAc. These results suggest that the K92 PST forms both 2,8 and 2,9 linkages in a successive and nonprocessive fashion. Key words: capsular polysaccharide / chain extension / polysialyltransferase / processivity / sialic acid
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Polysialic acids play numerous roles in cellular interactions. In pathogenic bacteria, these acidic polysaccharides serve as extracellular shields against the defense systems of their mammalian host. Human pathogens encapsulated with polysialic acids cause invasive disease such as meningitis and urinary tract infection (Jann et al. 1992
). There are three polysialic acid homopolymers identified in Nature, 2,8 NeuNAc (McGuire et al. 1964), 2,9 NeuNAc (Bhattacharjee et al. 1975), and the alternating structure 2,8 2,9 NeuNAc (Egan et al. 1977). Escherichia coli K92 produces the alternating polysialic acid and is serologically cross-reactive with the 2,9 NeuNAc capsular polysaccharide of the neuroinvasive pathogen Neisseria meningitidis Group C (Egan et al. 1977).
The genetic loci encoding the expression of bacterial polysialic acid have been well characterized (Whitfield and Roberts 1999). A single polysialyltransferase (PST)1 has been identified in the gene clusters of E. coli and Neisseria that produce polysialic acid capsule (Steenbergen et al. 1990). The K1 and K92 PSTs are membrane-bound enzymes and require the addition of an acceptor molecule to measure enzyme activity. These PSTs elongate the growing chains of polysialic acid from the nonreducing end. They transfer NeuNAc from CMP-ß-NeuNAc with inversion of the anomeric configuration and belong to the CAZY glycosyltransferase family 38 (Coutinho et al. 2003). The sequence-based family 38 glycosyltransferases consist entirely of the PSTs of E. coli and Neisseria. These PSTs are therefore expected to have different 3D structures from the recently reported sialyltransferase CSTII of Campylobacter (Chiu et al. 2004) belonging to family 42 (Coutinho et al. 2003). The enzymes in family 38 have not been purified and the mechanism of polymerization has not been characterized in detail.
We reported that the K92 PST can elongate all known polysialic acids and oligosialic acids when PST is incubated with these polymers in vitro as acceptors (McGowen et al. 2001). In vivo, the K92 PST is responsible for synthesis of both 2,8 and 2,9-linked NeuNAc. Our investigation of the acceptor specificity suggests that the PST prefers oligosaccharide acceptors with a hydrophobic aglycon. Although the PST does not elongate disialyloligosaccharides, it does elongate gangliosides possessing a NeuNAc 2,8 NeuNAc and the fluorescent ganglioside analog GD3-FCHASE (McGowen et al. 2001). In this report, we take advantage of these fluorescent ganglioside analogs as acceptors to characterize the K92 PST elongation reaction.
Polysaccharides are often degraded by processive enzymes (Horn et al. 2006). These enzymes bind to the polymer and perform multiple cleavages without dissociating. Chitin and cellulose degradation by processive enzymes is indicated by the pattern of accumulation of degradation products (Mello et al. 2002; Soerbotten et al. 2005; Sikorski et al. 2006). In a similar manner, the pattern of accumulation of products has been used to examine the processivity of plant and bacterial glycosyltransferases (Akita et al. 2002; Guillaumie et al. 2003; Forsee et al. 2006). The type 3 capsular polysaccharide of Streptococcus pneumoniae is synthesized by a processive glycosyltransferase (Forsee et al. 2006). A critical oligosaccharide length of approximately eight monosaccharides is required for recognition of the growing chain by the glycosyltransferase. Upon binding of the oligosaccharide–lipid acceptor to the carbohydrate recognition site of the glycosyltransferase, the polymerization reaction enters a highly processive phase to produce polymer of high molecular weight (MW). In the experiments described later, we show that the K92 PST elongates the trisialyl acceptor, GT3-FCHASE, rapidly with release of each intermediate from the enzyme suggesting a non-processive process.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Earlier reports have suggested that the K1 PST can extend short oligosaccharides of sialic acid by only a few residues (Steenbergen et al. 1990
; Ferrero et al. 1991). Cho and Troy (1994) showed that K1 PST can extend disialylgangliosides to polymeric material. Our laboratory has shown previously, that the K92 PST can elongate disialylgangliosides (McGowen et al. 2001). We demonstrated that a fluorescent analog of GD3 acts as an acceptor (primer) for K92 PST (McGowen et al. 2001). In the current report, we performed a more detailed analysis of the K92 elongation reaction. We took advantage of the fluorescent primer GT3-FCHASE to characterize the primer extension reaction of K92 PST. We wanted to determine whether the in vitro extension of oligosialic acid consists of the addition of a few residues or the formation of high-MW polymer, and to determine whether the product is an alternating polymer. Since the degree of processivity of this reaction is unclear, we investigated the mode of polysialic acid chain elongation by K92 PST.
The trisialylated oligosaccharide GT3-FCHASE was incubated with membranes harboring the K92 NeuS for 30 min and the soluble product contained in the high-speed supernatant was analyzed by ion exchange high-performance liquid chromatography (HPLC). As is shown in Figure 1. GT3-FCHASE was completely consumed and converted to a major species eluting near the end of the ion exchange gradient. This suggested that the fluorescent oligosaccharide had been converted into a high-MW polymer. We isolated this late eluting species and treated it with either sodium periodate or the 2,8-specific K1-5 endoneuraminidase to demonstrate that this peak was indeed a polysialic acid. The results in Figure 2 show that this late eluting species is sensitive to both treatments. Treatment of the isolated species with sodium periodate resulted in a marked decrease in the peak at 100 min. That polysaccharide chains were cleaved by periodate was confirmed in a separate experiment by oxidation of the high-MW peak and reduction with NaB3H4. A degradation product similar to that expected for oxidized sialic acid was detected in a hydrolysate of the oxidized and reduced high-MW fraction (data not shown). Characteristically the K1-5 endoneuraminidase degrades the high-MW species to oligomers of sialic acid. The K1-5 endoneuraminidase is specific for 2,8 NeuNAc linkages and does not cleave 2,9 linkages. This result suggests that this late eluting species is a polysialic acid containing both 2,8 and 2,9-linked NeuNAc.
|
|
Nakata and Troy (2005)
described a method for determining the chain length of polysialic acids in cell extracts based on the elution position of the oligosialic and polysialic acid on an ion exchange gradient. We used the ion exchange gradient described for this method to estimate the chain length of polysialic acid resulting from the extension of GT3-FCHASE after a 30- min reaction. As is shown in Figure 3, the product elutes at a position expected for polysialic acids with a degree of polymerization of 90–100 sialic acid residues. Polysialic acid formed after a 30-min incubation eluted as a single peak in the 90–100 degree of polymerization (DP) range. However, in Figure 3 the high-MW polymer formed after a 5 min reaction is smaller and more polydisperse than that formed after a 30 min reaction, suggesting that at earlier time points smaller size oligosialic acids are formed that are eventually elongated to the 90–100 DP range.
|
Time course of GT3-FCHASE extension
It appears from the experiments described in Figure 1, that the elongation reaction is very rapid. Expansion of the early portion of the chromatogram in Figure 1 suggests that oligosialic acid intermediates are formed during the polymerization reaction. We developed an assay to detect the early stages of the polymerization reaction and to determine if oligosialic acid intermediates are formed. We used ethanol to inactivate the enzyme and stop the reaction. Oligosialic acid intermediates were not detected in reaction mixtures containing the K1 PST of E. coli strain K235 by the radiolabel assay. Similarly, oligosialic acid intermediates were not detected in in vivo pulse reactions of E. coli K1 PST (Steenbergen and Vimr 2003
). This result has been interpreted by previous authors to suggest that the enzyme might polymerize sialic acid in a processive reaction. Oligosialic acids were observed immediately after initiation of extension of GT3-FCHASE. The chromatograms in Figure 4 show that oligosialic acids of increasing size are observed ranging from elongation of GT3-FCHASE by a few sialic acids to the high-MW polysialic acid. The size of the high-MW polysialic acid peak increases as the concentration of the low- and intermediate-MW oligosialic acids decreases, suggesting that oligosialic acid is consumed in the formation of polysialic acid. This suggests that the K92 PST is not strictly processive but allows the dissociation of the product after each addition of sialic acid to GT3-FCHASE. To confirm that the dissociated intermediates are being consumed and converted into polysialic acid we isolated several peaks from scaled up 2 min reactions. The purified intermediates were incubated in a second reaction with K92 PST for 5 min and applied to the ion exchange column. In Figure 5 are shown the chromatograms of these peaks, illustrating their conversion to polysialic acid after the second incubation.
|
|
The K92 PST also appears to have a preference for higher DP oligosialic acids. In Figure 4A and B at early time points there is a greater accumulation of oligosialic acids with a DP less than 10–12. Oligosialic acids with higher DP are in lower concentration and appear to be converted more rapidly into polysialic acid. With time, the lower DP oligosialic acid appears to be converted to higher DP material since the concentration of the lower DP species decreases as the concentration of the high DP species increases (Figure 4). We interpret these results to mean that the PST has a higher affinity for the higher DP polysialic acid.
Dependence of GT3-FCHASE extension on substrate concentration
The size distribution of reaction products of the K92 PST is very dependent on CMP-NeuNAc concentration. At 250 µM CMP-NeuNAc the major species formed is high-MW polymer. However, at low concentrations (5–50 µM) the predominant products represent the addition of only a few sialic acids to GT3-FCHASE. This concentration range (5–50 µM) is well below the reported Km for CMP-NeuNAc, 250–300 µM, (Ortiz 1989) (data not shown). We tested the ability of nascent GT3-FCHASE products to remain bound to the membrane complex after washing. We could not detect significant amounts of fluorescent product bound to K92 membranes after washing to remove substrates. These results support the conclusion that K92 PST does not elongate acceptors in vitro in a processive fashion.
Chain termination of GT3-FCHASE extension reaction
The K92 PST appears to be adding sialic acid in both 2,8 and 2,9 linkages based on the results described in Figure 2. This result implies that inclusion of appropriate substrate analogs would limit the growth of the polysaccharide chain by termination of GT3-FCHASE extension. As outlined in Scheme 1, when CMP-9-N3-NeuNAc is used as a substrate the transfer of sialic acid to the position-9 of the primer molecule occurs only once, thus blocking synthesis. We tested this hypothesis using the radiolabeled assay. Membranes of the sialic acid-negative strain construct, EV239:pWV213, were first incubated with either CMP-NeuNAc or CMP-9-N3-NeuNAc, then centrifuged and washed prior to the addition of radiolabeled substrate. As is shown in Table I, incubation with the 9-azido derivative inhibits incorporation of NeuNAc into polymer. Similarly, K92 PST activity was inhibited when CMP-9-N3-NeuNAc was added directly to assay mixtures.
|
If sialic acid is transferred in an alternating fashion then extension of GT3-FCHASE with CMP-9-azido-NeuNAc would result in synthesis of a truncated product. In Figure 6A, the product of CMP-9-N3-NeuNAc incubation with K92 PST and GT3-FCHASE eluted as a single peak at a position expected for the addition of 1–2 sialic acids to GT3-FCHASE. No species of higher DP are observed in this reaction mixture. That this observation is not due to simple inhibition was demonstrated by isolation of the CMP-9-N3-NeuNAc extended product and incubation with CMP-NeuNAc and active membranes. Again, the isolated product was not extended (Figure 6C). Products of similar DP isolated from CMP-NeuNAc containing reaction mixtures were converted to polysialic acid of the 90–100 DP during an equivalent time period (Figure 5).
|
Reaction mixtures containing both substrates, CMP-NeuNAc and CMP-9-N3-NeuNAc, resulted in apparent lower DP products. In Figure 7, the K92 PST was incubated with CMP-NeuNAc and CMP-9-N3-NeuNAc in a ratio of 1:1 and 1:5. A more extensive accumulation of oligosialic acids of lower DP was observed with the 1:1 mixture compared with the 1:5 mixture. Note that very little polysialic acid is formed even after extensive incubation of the reaction mixtures. This result suggests that the length of GT3-FCHASE chain extensions can be controlled by the ratio of modified substrate. These reagents are potentially useful tools for preparing tagged oligosialic acids of defined DP.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
The mechanism of chain extension by the E. coli PSTs is unclear from previous reports. It was concluded, based on the ability to extend a short oligosialic acid by a few residues, that the K1 PST catalyzes the successive addition of single sialic acid residues (Steenbergen and Vimr. 1990
; Ferrero et al. 1991). The thin layer chromatography (TLC) method used in these experiments was not very sensitive. Furthermore, it appeared from these experiments that high-MW polymer was not formed from these oligosialic acid acceptors, thus leaving in question whether these experiments reflected the mechanism in vivo. Chao et al. (1999) studied the elongation of K92 oligosialic acids by the K92 PST and concluded that the enzyme extends the growing chain by the addition of a disialic acid unit to the nonreducing end of the growing chain (Chao et al. 1999). In contrast, the time course of in vivo synthesis of K1 polysialic acid was measured by continuous pulse of the E. coli strain EV138 (nanA4, neuB25) culture exogenously with radioactive NeuNAc (Steenbergen and Vimr 2003). In this latter experiment no intermediate oligosialic acids were detected by TLC. The authors therefore concluded that the in vivo reaction occurs in a processive manner (Steenbergen and Vimr 2003). In this report, we addressed the question of extension of polysialic acid chains using a very sensitive method based on a fluorescent acceptor. This method has greater sensitivity than the radiolabel TLC method and has a high resolution. The properties of the assay enabled us to follow the extension reaction further than previously reported. Using this approach, we could show that the K92 PST extends its acceptor to form high-MW polymer in a rapid fashion. Furthermore, since intermediates of every size were detected, the addition of sialic acid to the growing chain probably occurs in a successive manner. This successive addition is further supported by the apparent conversion of intermediate oligosialic acids to polysialic acids of 90–100 DP. This result is not consistent with the processive dimer model or the in vivo processive model. Our observation that the high DP oligosialic acids are consumed rapidly would explain the inability to detect intermediates in the relatively insensitive in vivo TLC method. Oligosialic acid intermediates, if rapidly consumed, would be in low concentrations and below the limits of detection of TLC autoradiographs.
An alternative explanation is that the PST alone is nonprocessive and catalyzes successive additions. Most of the experiments in this report were done with membranes harboring NeuS and lacking other gene products of the K92 gene cluster. However, in the presence of an accessory protein from the K92 gene cluster the NeuS PST would become processive. This would explain the discrepancy between the interpretation of the in vivo and in vitro observations. One should also keep in mind that the results reported here were obtained with an artificial acceptor. Determination of how closely these experiments mimic in vivo synthesis awaits the identification and characterization of the natural acceptor.
The K92 PST elongated disialyl and trisialyloligosaccharides possessing a hydrophobic aglycon. This property was demonstrated for disialylgangliosides and the ganglioside analogs GD3-FCHASE and GT3-FCHASE. In previous experiments, transfer was not observed with disialyl or trisialyl acceptors lacking the hydrophobic aglycon. The sialyltransferase, CSTII, of Campylobacter jejuni transfers sialic acid to oligosaccharide acceptors attached to a hydrophobic residue. This is due to the fortuitous location of a hydrophobic region near the sugar acceptor binding site (Chiu et al. 2004). A similar situation could exist with the K92 PST.
The fact that the product dissociated from the enzyme after transfer of each sialic acid implies that specificity of transfer is influenced by the structure of the acceptor. After transfer of sialic acid to form a 2–9 linkage, this product dissociates and competes with oligosialic acids having 2,8 NeuNAc at the nonreducing end of the chain. In order to maintain the synthesis of an alternating 2,8-NeuNAc–2,9-NeuNAc structure, transfer must be to the 2,9 hydroxyl when the 2,8NeuNAc is bound and similarly, transfer must be to the 2,8 hydroxyl when the oligosialic acid end in a 2,9 linkage is bound. This observation rules out the two active site model proposed in our earlier publication (McGowen et al. 2001). Nevertheless the orientation of the 8- or 9- hydroxyl of sialic acid relative to the sugar nucleotide could dictate specificity of the enzymatic transfer. This question may be answered when crystallographic data are available for this enzyme.
CMP-9-N3-NeuNAc is a useful substrate to investigate the polymerization of the alternating polysialic acid structure. Attempts to use the 9-fluoroceinyl-NeuNAc derivative were unsuccessful since this analog is a poor substrate for the PST (Vann, unpublished results). CMP-9-N3-NeuNAc is a much better substrate, probably due to the relatively small size of the azido function. The use of this substrate blocks synthesis as predicted for alternating linkages.
Polysialic acids play important roles in normal cell function and disease. These polysaccharides are used as vaccines against infectious bacteria and markers of tumor development. Sensitive reagents for monitoring functions of polysialic acid would improve the therapeutic and diagnostic value of these polysaccharides. Overall the results in this report illustrate a path to the preparation of polysialic acids with selectively functionalized end groups. Jing and DeAngelis (2004) have also used a chemoenzymatic approach to prepare monodisperse polymer with tags at the reducing end of the medically important hyaluronic acid. Introduction of the azido group at the non-reducing terminus provides a means of controlling the length of the growing chain. In addition, it provides a site for chemically introducing probes or conjugating to proteins. 1,3 Dipolar cycloadditon of alkynes to azido is a mild and powerful route for attachment of oligosaccharides to microarrays, synthesis of multivalent neoglycoconjugates, and labeling of proteins and cell surfaces. The availability of this variety of defined reagents will facilitate characterization of antibody-antigen interaction, synthesis of vaccines, and characterization of cell interactions with polysialic acids.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Bacterial strains and plasmids
The bacterial strains and plasmids used in this study have been described (Andreishcheva and Vann 2006
). E. coli K1 hybrid strain EV239 (neuS::Tn10 nanA4, neuB25) was obtained from Dr. Eric Vimr (University of Illinois, Urbana). The plasmid pWV213 contains the K92 neuS gene and was described previously (McGowen et al. 2001).
CMP-9-N3-NeuNAc synthesis
9-Azido-NeuNAc was synthesized according to Han et al (2005). CMP-9-N3-sialic acid was synthesized according to Shames et al. (1991) with a few modifications. The synthesis reaction consisted of 100 mg of CTP (Roche Applied Science, Indianapolis, IN) and 50 mg of 9-azido-sialic acid in 5 mL of 50 mM Tris, 50 mM MgCl2, pH 8.5. The reaction was initiated by the addition of six units of E. coli CMP-NeuNAc synthetase and incubated at room temperature for 5 h with periodic removal of precipitate by centrifugation. The progress of the reaction was monitored by thiobarbituric acid (Vann et al. 1987). An additional 6 U of enzyme was added for 1 h incubation. CMP-9-N3-NeuNAc was precipitated with ethanol (9:1, v/v) and isolated by centrifugation at 15 000g, 10 min.
CMP-9-N3-NeuNAc was purified by Dowex- AG-1-X8 resin, 200–400 mesh (Bio-Rad Laboratories, Inc. Hercules, CA). After the sample was loaded, the column was washed with 1 mM ammonium bicarbonate until unbound ultraviolet-absorbing material was removed. The sugar nucleotide was then eluted with a gradient of 1 M ammonium bicarbonate. The fractions containing CMP-9-N3-NeuNAc were pooled and dried in a Speed-Vac with successive additions of water to remove residual ammonium bicarbonate.
Preparation of membranes
Bacterial cultures of DH5:pWV213, and EV239:pWV213, were grown on 1.5 L of Luria–Bertani (LB) broth. LB medium was supplemented with appropriate antibiotics. The cultures were grown at 37 °C until A600 = 0.6 and induced with 1 mM isopropylthiogalactopyranoside (IPTG) for 2 h. Membranes were isolated as described previously (McGowen et al. 2001). The membranes (6–12 mg protein/mL) were resuspended in a cryoprotectant buffer (Kempner et al. 1989) that contained 14% (v/v) glycerol, 1.4% (w/v) D-sorbitol, 150 mM KCI, 50 mM Tris-HCl, and 25 mM MgCl2 (pH 8). Membrane preparations were stored at –80 °C in cryoprotectant buffer until use. Membrane preparations were added directly to reaction mixtures without removal of cryoprotectant.
Polysialyltransferase activity
PST activity was assayed as described previously (McGowen et al. 2001). Substrate, cytidine 5'-monophosphate sialic acid [sialic-4, 5,6,7,8,9-14C] ([14C] CMP-NeuNAc) was obtained from American Radiolabeled Chemicals, Inc., (St. Louis, MO). The exogeneous acceptor, K92 PSA (K92 PS), was isolated and purified from E. coli Bos 12. Each 50 µL reaction mixture contains 25 µL of membrane preparation (150–300 µg protein); 2.7 nmol of [14C]-CMP-NeuNAc (specific activity = 55 µCi/µmole), and as needed, 10 µg of exogenous acceptor dissolved in 50 mM Tris, 25 mM MgCl2, pH 8.0. The reaction mixture is then incubated at 37 °C for 30 min. A parallel reaction mixture with 25 µL of boiled membrane preparation, acceptor, and labeled substrate was prepared as a negative control. The reaction was quenched by spotting 30 µL onto Whatman 3M chromatography paper. The chromatogram was developed overnight with a 7.5:3 ethanol: 1 M ammonium acetate pH 7.0 solvent system. The paper was dried and the radiolabel incorporated into PSA that remains at the origin was quantitated by liquid scintillation counting. Inhibition of PST was performed by the inclusion of CMP-9-N3-NeuNAc in the reaction mixture.
Polysialyltransferase polymerization assay
The polymerization activity of the K92 PST was monitored by the extension of a fluorescent labeled acceptor. Each 25 µL reaction mixture contains 75–150 µg of membrane protein; 250 µM CMP-NeuNAc (Sigma no. 27562), 10 ng of GT3-FCHASE (obtained from Warren Wakarchuk and Michel Gilbert of the National Research Council, Ottawa, Ontario), 35 mM Tris, 17.5 mM MgCl2, pH 8.0 and is then incubated at 37 °C for 0–60 min. The reaction is stopped by adjusting to 25% ethanol and dried in a Savant Speed-Vac for 30 min. The reaction is adjusted to 25 µL with water, centrifuged at 2000g for 30 min, and the supernatant is analyzed by HPLC as described later.
Products of the FCHASE extension reaction were chromatographed on a DNA Pac PA-100 column using a Dionex Summit P680 Pump HPLC system equipped with a Hitachi Fluorescence L7485 detector set at 488 nm for excitation and 520 nm for emission. Elution was performed at 1 mL/min with segments of linear gradient of NaNO3 (Inoue et al. 2001; Inoue and Inoue 2003) with minor modifications.
Degree of polymerization
The DP of the high-MW GT3-FCHASE product was estimated using the HPLC gradient described by Nakata and Troy (2005). The high-MW peak from the GT3-FCHASE reaction described in the Polysialyltransferase polymerization assay section was collected and purified on a Sep-Pak cartridge (Waters part no. 051910) prior to chromatography on this gradient. Sep-Pak cartridges were wetted with 20 mL of 100% acetonitrile followed by 20 mL of H2O. Peaks isolated from HPLC were immediately loaded onto the prepared Sep-Pak cartridges and allowed to flow through by gravity. The cartridge was washed with 4 mL of H2O followed by elution with 3 mL of 50% acetonitrile. The elution fractions were dried, taken up in water, and applied to a DNA Pac PA-100 column. The DNA Pac PA-100 column was eluted with a convex gradient of 0–2 M ammonium acetate. In this gradient, the concentration of ammonium acetate at 0, 5, 15, 20, 35, 55, 145, and 182.5 min was 0, 0, 20, 25, 32.5, 40, 62.5, and 100% of 2.0 M. The DP was estimated by comparing the elution position in the ammonium acetate gradient with the elution position of polysialic acids described by Nakata and Troy (2005).
Chain termination reactions
The extension of polysialic acid chains on GT3-FCHASE was terminated by the inclusion of 250 µM CMP-9-N3-NeuNAc in PST polymerization assay mixtures in the presence or absence of CMP-NeuNAc as described in the text (Andreishcheva and Vann 2006).
Endoneuraminidase treatment
Reaction mixtures (25 µL) prepared for the PST polymerization assay were incubated for 1 h prior to the addition of either 35 µg freshly purified K1-5 endoneuramidase (Andreishcheva and Vann 2006) in 50 µL or 50 µL of buffer as a control. The digestion mixtures were incubated at 37 °C for 1 h, centrifuged and the supernatant analyzed on the DNA Pac column as described in the Polysialyltransferase polymerization assay section.
Alternatively, individual peaks from the PST reaction were collected, purified by Sep-Pak, and redissolved with 50 µL H2O prior to digestion with endoneuraminidase. A 25 µL aliquot was incubated at 37 °C for 10 min with 10 µL of freshly purified endoneuraminidase, whereas the other 25 µL aliquot is incubated with buffer. The reactions are loaded directly onto the DNA Pac column for HPLC analysis.
NaIO4 treatment
The reaction mixture contained 50 µL of membrane preparation; 250 µM CMP-NeuNAc, 40 ng of GT3-FCHASE, 35 mM Tris, 17.5 mM MgCl2, pH 8.0 in 100 µL and was incubated at 37 °C for 1 h. The reaction mixture was stopped by adjusting to 25% ethanol, treated as described in the Endoneuraminidase treatment section and analyzed by HPLC. A single late eluting peak of high-MW material was collected, purified by Sep-Pak, and redissolved with 450 µL H2O. A 225 µL aliquot was mixed with 225 µL of either 100 mM NaIO4 or 225 µL of H2O and incubated for 20 h at room temperature. The reaction treated with NaIO4 was quenched by adjusting to 200 mM glycerol. Both reactions were analyzed by HPLC.
To confirm the identity of the periodate degradation products by paper chromatography, the earlier PST reaction was scaled up to a 250 µL reaction. The high-MW peak isolated by HPLC and Sep-Pak cartridge was dissolved in 100 µL water and processed as follows. A 50 µL aliquot was mixed with 50 µL of either 10 mM NaIO4 in 20 mM sodium acetate, pH 5.6 or 50 µL of 20 mM sodium acetate, pH 5.6, and incubated 2 h at 37 °C. Both reactions were quenched by adjusting to 20 mM glycerol and reduced with 10 mM NaB3H4 (specific acitivity 22.6 Ci/mmol). The reactions were acidified with Dowex 50 (H+), filtered and dried under nitrogen. The borate was removed by adding 0.5 mL aliquots of methanol and evaporating to dryness three times. The oxidation products were taken up in 2 M acetic acid and hydrolyzed at 80 °C for 2 h. This hydrolysate was dried under a stream of nitrogen. The residue was dissolved in 100 µL of H2O and analyzed by paper chromatography in 7.5:3 ethanol: 1 M ammonium acetate pH 7.0 solvent system. Radioactive spots were detected with a Molecular Dynamics model 445 SI phosphor-imager and evaluated with Image Quant software
1,2-Diamino-4,5-methylenedioxybenzene labeling
The reaction supernatants described in the Polysialyltransferase polymerization assay section were labeled with DMB (1,2-diamino-4,5-methylenedioxybenzene) (Inoue et al. 2001; Inoue and Inoue 2003) with minor modifications described here. DMB forms a fluorescent product with the reducing terminal sialic acid residue of the polymer. After the reaction, supernatants were incubated 48 h, 10 µL of 0.5 M NaOH was added and the reaction was incubated for 16 h at room temperature. The pH was adjusted to 8 with 0.1 M acetic acid. The fluorescence detector was set at 373 nm for excitation and 448 nm for emission according to Hara et al. (1989).
|
Conflict of interest statement |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
None declared.
|
|
Acknowledgements |
|---|
We wish to thank Warren Wakarchuk and Michel Gilbert of the National Research Council, Ottawa, Ontario for supplying samples for GD3-FCHASE and GT3-FCHASE, without which these experiments could not have been performed. We thank Ekaterina Andreishcheva for helpful discussions and critical review of the manuscript.
|
Abbreviations |
|---|
CMP-9-N3-NeuNAc, CMP-9-azido-N-acetylneuraminic acid; DMB, 1,2-diamino-4,5-methylenedioxybenzene; DP, degree of polymerization; FCHASE, 6-(fluorescein-5-carbaxamido) hexanoic acid, succinimidyl ester; GD3, neuNAc-
2,8-neuNAc-2,3-gal-ß1,4-glc-ceramide; GD3-FCHASE, neuNAc-2,8-neuNAc-2,3-gal-ß1,4-glc-ß-aminophenyl FCHASE; GT3-FCHASE, neuNAc-2,8-neuNAc-2,8-neuNAc-2,3-gal-ß1,4-glc-ß-aminophenyl FCHASE; HPLC, high-performance liquid chromatography; LB, Luria–Bertani; MW, molecular weight; PST, polysialyltransferase; TLC, thin layer chromatography. |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsConflict of interest statementReferences |
|---|
Akita K, Ishimizu T, Ando T, Hase S. Successive glycosyltransfer activity and enzymatic characterization of pectic polygalacturoate 4-a-galacturonosyltransferase solubilized from pollen tubes of Petunia axilaris using pyridylaminated oligogalacturonates as substrates. Plant Physiol (2002) 130:374–379.
Andreishcheva EN, Vann WF. Gene products required for de novo synthesis of polysialic acid in K1 Escherichia coli. J Bacteriol (2006) 188:1786–1797.
Bhattacharjee AK, Jennings HJ, Kenny CP, Martin A, Smith IC. Structural determination of the sialic acid polysaccharide antigens of Neisseria meningitidis serogroups B and C with carbon 13 nuclear magnetic resonance. J Biol Chem (1975) 250:1926–1932.
Chao CF, Chuang HC, Chiou ST, Liu TY. On the biosynthesis of alternating alpha-2,9/alpha-2,8 heteropolymer of sialic acid catalyzed by the sialyltransferase of Escherichia coli Bos-12. J Biol Chem (1999) 274:18206–18212.
Chiu CPC, Watts AG, Lairson LL, Gilbert M, Lim D, Wakarchuk WW, Withers SG, Strynadka NCJ. Structural analysis of the sialyltransferase CstII from Campylobacter jejuni in complex with a substrate analog. Nat Struct Mol Biol (2004) 11:163–170.[CrossRef][ISI][Medline]
Cho J-W, Troy FA. Polysialic acid engineering: synthesis of polysialylated neoglycosphingolipids by using polysialyltransferase from neuroinvasive Escherichia coli K1. Proc Natl Acad Sci USA (1994) 91:11427–11431.
Coutinho PM, Deleury E, Davies GJ, Henrissat B. An evolving hierarchical family classification of glycosyltransferases. J Mol Biol (2003) 328:307–317.[CrossRef][ISI][Medline]
Egan W, Liu T-Y, Dorow D, Cohen JS, Robbins JD, Gotschlich EC, Robbins JB. Structure studies on the sialic acid polysaccharide antigen of Escherichia coli strain Bos-12. Biochemistry (1977) 16:3687–3692.[CrossRef][Medline]
Ferrero MA, Luengo JM, Reglero A. H.p.l.c. of oligo(sialic acids): application to the determination of the minimum chain length serving as exogenous acceptor in the enzymic synthesis of colominic acid. Biochem J (1991) 280:575–579.[ISI][Medline]
Forsee WT, Cartee RT, Yother J. Role of the carbohydrate binding site of the Streptococcus pneumoniae capsular polysaccharide type 3 synthase in the transition from oligosaccharide to polysaccharide synthesis. J Biol Chem (2006) 281:6283–6289.
Guillaumie F, Sterling JD, Jensen KJ, Thomas ORT, Mohnen D. Solid-supported enzymatic synthesis of pectic oligogalacturonides and their analysis by MALDI-TOF mass spectrometry. Carbohydr Res (2003) 338:1951–1960.[CrossRef][ISI][Medline]
Han S, Collins BE, Bengtson P, Paulson JC. Homomultimeric complexes of CD22 in B cells revealed by protein-glycan cross-linking. Nat Chem Biol (2005) 1:93–97.[CrossRef][ISI][Medline]
Hara S, Yamaguch M, Takemori Y, Furuhata K, Ogura H, Nakamura M. Determination of mono-O-acetylated N-acetylneuraminic acids in human and rat sera by fluorometric high-performance liquid chromatography. Anal Biochem (1989) 179:162–166.[CrossRef][ISI][Medline]
Horn SJ, Soerbotten A, Synstad B, Sikorski P, Soerlie M, Varum KM, Eijsink VGH. Endo/exo mechanism and processivity of family 18 chitinases produced by Serratia marcescens. FEBS J (2006) 273:491–503.[CrossRef][Medline]
Inoue S, Inoue Y. Ultrasensitive analysis of sialic acids and oligo/polysialic acids by fluorometric high-performance liquid chromatography. Meth. Enzymol (2003) 362:543–560.[ISI][Medline]
Inoue S, Lin S-L, Lee YC, Inoue Y. An ultrasensitive chemical method for polysialic acid analysis. Glycobiology (2001) 11:759–767.
Jann B, Jann K. Capsules of Escherichia coli, expression and biological significance. Can J Microbiol (1992) 38:705–710.[ISI][Medline]
Jing W, DeAngelis PL. Synchronized chemoenzymatic synthesis of monodisperse hyaluronan polymers. J Biol Chem (2004) 279:42345–42349.
Kempner ES, Fleischer S. Radiation inactivation of membrane components and molecular mass determination by target analysis. Meth Enzymol (1989) 172:410–439.[ISI][Medline]
McGowen MM, Vionnet J, Vann WF. Elongation of alternating 2,8/2,9 polysialic acid by the Escherichia coli K92 polysialyltransferase. Glycobiology (2001) 11:613–620.
McGuire EJ, Binkley SB. The structure and chemistry of colominic acid. Biochemistry (1964) 170:247–251.
Mello LV, de Groot BL, Li S, Jedrzejas MJ. Structure and flexibility of Streptococcus agalactiae hyaluronate lyase complex with its substrate: insight into the mechanism of processive degradation of hyaluronan. J Biol Chem (2002) 277:36678–36688.
Nakata D, Troy FA II. Degree of polymerization (DP) of polysialic acid (polysia) on neural cell adhesion Molecules (N-CAMs): development and application of a new strategy to accurately determine the DP of polySIA chains on N-CAMS. J Biol Chem (2005) 280:38305–38316.
Ortiz AI, Reglero Angel, Rodriguez-Aparicio Leandro, Luengo Jose M. In vitro synthesis of colominic acid by membrane-bound sialyltransferase: kinetic properties of this enzyme and inhibition by CMP and other cytidine nucleotides. Eur. J. Biochem (1989) 178:741–749.[ISI][Medline]
Shames SL, Simon ESC, Christopher W, Schmid W, Whitesides GM, Yang L-L. CMP-N-acetylneuraminic acid synthetase of Escherichia coli: high level expression, purification and use in the enzymatic synthesis of CMP-N-acetylneuraminic acid and CMP-neuraminic acid. Glycobiology (1991) 1:187–191.
Sikorski P, Soerbotten A, Horn S, Eiljsink VGH, Varum KM. Serratia marcescens chitinase with tunnel-shaped substrate-binding grooves show endo activity and different degrees of pocessivity during enzymatic hydrolysis of chitosan. Biochemistry (2006) 45:9566–9574.[CrossRef][Medline]
Soerbotten A, Horn SJ, Eijsink VGH, Varum KM. Degradation of chitosans with chitinase B from Serratia marcescens: production of chito-oligosaccharides and insight into enzyme processivity. FEBS J (2005) 272:538–549.[CrossRef][Medline]
Steenbergen SM, Vimr ER. Mechanism of polysialic acid chain elongation in Escherichia coli K1. Mol Microbiol (1990) 4:603–611.[ISI][Medline]
Steenbergen SM, Vimr ER. Functional relationships of the sialyltransferases Involved in expression of the polysialic acid capsules of Escherichia coli K1 and K92 and Neisseria meningitidis groups B or C. J Biol Chem (2003) 278:15349–15359.
Vann WF, Silver RP, Abeijon C, Chang K, Aaronson W, Sutton A, Finn CW, Lindner W, Kotsatos M. Purification, properties, and genetic location of Escherichia coli cytidine 5'-monophosphate N-acetylneuraminic acid synthetase. J Biol Chem (1987) 262:17556–17562.
Whitfield C, Roberts IS. Structure, assembly and regulation of expression of capsules in Escherichia coli. Mol Microbiol (1999) 31:1307–1319.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||