| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 281, Issue 9, 5391-5397, March 3, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1
From the
Department of Biochemistry and Molecular Biology, Oklahoma Center for Medical Glycobiology, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73104 and Bijvoet Center, Department of Bio-Organic Chemistry, Utrecht University, P.O. Box 80.075, NL-3508 TB Utrecht, The Netherlands
Received for publication, September 22, 2005 , and in revised form, December 7, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
1,3-N-acetylglucosamine-1,4-glucuronic acid sugar chain by the sequential addition of single monosaccharides to the non-reducing terminus. Therefore, PmHAS possesses both GlcNAc-transferase and glucuronic acid (GlcUA)-transferase activities. The recombinant Escherichia coli-derived PmHAS enzyme will elongate exogenously supplied hyaluronan chains in vitro with either a single monosaccharide or a long chain depending on the UDP-sugar availability. Competition studies using pairs of acceptors with distinct termini (where one oligosaccharide is a substrate that may be elongated, whereas the other cannot) were performed here; the lack of competition suggests that PmHAS contains at least two distinct acceptor sites. We hypothesize that the size of the acceptor binding pockets of the enzyme corresponds to the size of the smallest high efficiency substrates; thus we tested the relative activity of a series of authentic hyaluronan oligosaccharides and related structural analogs. The GlcUA-transferase site readily elongates (GlcNAc-GlcUA)2, whereas the GlcNAc-transferase elongates GlcUA-Glc-NAc-GlcUA. The minimally sized oligosaccharides, elongated with high efficiency, both contain a trisaccharide with two glucuronic acid residues that enabled the identification of a synthetic, artificial acceptor for the synthase. PmHAS behaves as a fusion of two complete glycosyltransferases, each containing a donor site and an acceptor site, in one polypeptide. Overall, this information advances the knowledge of glycosaminoglycan biosynthesis as well as assists the creation of various therapeutic sugars for medical applications in the future. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
The bacterial Pasteurella multocida synthases are dual action enzymes that add both sugars of the glycosaminoglycan repeat to the non-reducing terminus of the acceptor oligosaccharides (1). The P. multocida Type A hyaluronan synthase PmHAS and the P. multocida Type F chondroitin synthase PmCS have been amenable to study due to their two active center architectures (2, 3), their ability to polymerize long chains in vitro (4, 5), and their ability to elongate certain exogenously added acceptor oligosaccharides (6). The Escherichia coli K4 chondroitin polymerase KfoC (7) is 
60% identical to PmHAS and PmCS; therefore, this system probably operates in a similar fashion. The recently described P. multocida heparosan synthases, one from Type D (PmHS1) (8) and another cryptic enzyme from Types A, D, and F (PmHS2) (9), promise to be interesting experimental models of dual action enzymes as well, but their overall amino acid sequence appears to differ from PmHAS and PmCS.
In general, enzymological studies of glycosyltransferases have focused on the catalytic residues, the donor binding site, and the acceptor binding site. Structural information on some "simple" glycosyltransferases that add only one sugar to a glycoconjugate has been obtained, but a structure has not been determined for a dual action enzyme or a polysaccharide synthase. PmHAS and PmCS each appear to possess an independent hexosamine donor transfer site and a glucuronic acid donor transfer site (2, 3), but the nature and the number of sugar acceptor sites has not been determined. As a first step to analyze the acceptor site(s), here we have tested a range of acceptor sugars that PmHAS will elongate with the HA chain. We speculate that the size of the synthase acceptor binding pocket corresponds roughly to the size of the smallest high efficiency substrate. Our findings have allowed us to determine the optimal length of the sugar polymer necessary for efficient chain elongation and to design a synthetic, artificial acceptor for the synthase enzyme. In addition, we have indirect evidence for two distinct acceptor sites in PmHAS.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
95% pure PmHAS based on staining of the gels. Sugar Acceptor Substrates—The HA4 tetrasaccharide (with GlcUA at the non-reducing terminus) was derived from exhaustive digestion of HA (streptococcal) with testicular hyaluronidase (Type V), chloroform solvent extraction, and gel chromatography on P2 resin (Bio-Rad) (11). Longer natural HA oligosaccharides (HA14, 15, 20, and 21) were synthesized chemoenzymatically from HA4 and UDP-sugars using immobilized enzyme reactors (11).
A series of HA-like oligosaccharides were synthesized by organic chemistry methodology; each sugar contained a para-methoxyphenyl group at the reducing end (12). For simplicity, a coding scheme was used to designate each monosaccharide: GlcUA, A; GlcNAc, N; methoxyphenyl, MP. For example, the compound AN-MP refers to GlcUA-GlcNAc-MP as read from non-reducing end to reducing end.
Various synthetic glycosides were purchased, and again the simple coding scheme was applied: fluorescein mono--D-glucuronide, A-F; fluorescein di--D-glucuronide, A-F-A (Molecular Probes, Eugene, OR); fluorescein di--D-glucopyranoside, G-F-G (Molecular Probes); fluorescein di--D-N-acetylgalactosamine, GalNAc-F-GalNAc (Marker Gene Technologies, Eugene, OR); 1-naphthyl -D-glucuronide, A-NAP; p-nitrophenyl--D-glucuronide, A-NP (Calbiochem, La Jolla, CA), 4-nitrophenyl--D-galacturonide, GalA-NP; 4-methylumbelliferyl -D-glucuronide, A-MUM; -trifluoromethylumbelliferyl -D-glucuronide, A-F3MUM (Molecular Probes); 3-carboxyumbelliferyl -D-glucuronide, A-CU (Molecular Probes); 4-methylumbelliferyl N-acetyl--D-glucosaminide, N-MUM; and 4-methylumbelliferyl -D-glucopyranoside, G-MUM.
Single Sugar Addition Assays—The assays monitored the transfer of either (a) a single GlcUA to an acceptor with a non-reducing end terminating in GlcNAc according to the reaction,
 |
 |
Enzyme activity was linear with respect to time, and the reactions consumed <5% of the UDP-sugar substrate. All assay points were performed in duplicate, and the values were averaged. The data were plotted using the Michaelis-Menten equation (V = Vmax [substrate]/Km + [substrate]) with Sigma Plot software (Rockware, Golden, CO), where the apparent Michaelis-Menten constants (Km) were derived from the concentration of sugar that yields 50% of maximal incorporation.
When the acceptor possessed a hydrophobic aglycone (e.g. MP, F, etc.), reversed phase solid phase extraction was employed to separate the products from the reactants. Descending paper chromatography was utilized when native HA was longer than HA14 (polymer size that is completely retained at the origin, thus facilitating single sugar transfer assays).
For solid phase extraction analysis, reactions were terminated by placing on ice and diluting with 275 µl of ice-cold 1 M NaCl. The free unincorporated UDP-[3H]sugar precursors were separated from the elongated reaction products using reversed phase cartridges (Strata-X polymeric 33-µm resin; 30 mg of sorbent/1-ml cartridge; Phenomenex, Torrance, CA) and a vacuum manifold. When passing solvents through the column, the bed was not allowed to dry until directly before and after the elution step. The columns were sequentially conditioned with 1 ml each of 100% methanol, 50% methanol, and water. The columns were equilibrated with 1 ml of 1 M NaCl, and then samples in 1 M NaCl were added to the sorbent bed. The columns were washed with 7 ml of 1 M NaCl allowing retention of the methoxyphenol or hydrophobic compounds and the removal of UDP-[3H]sugars. After completing the NaCl wash, the column bed was air-dried for 30 s with vacuum suction. To release the acceptors from the sorbent, the columns were eluted with 2 ml of 50% methanol. BioSafe II mixture (4 ml) (Research Products International, Chicago, IL) was added to 1 ml of the eluted sample, and incorporation of the [3H]sugars was quantitated by liquid scintillation counting.
For descending paper chromatography analysis, reactions were terminated by adding SDS to a final concentration of 2% and then spotted onto strips of Whatman No. 3MM paper, and the reaction products at the origin were separated from the free UDP-[3H]sugars by development with 65:35 ethanol/1 M ammonium acetate, pH 5.5, overnight (2). The origins were cut from the paper strips and eluted in 750 µl of water for 1 h. BioSafe II mixture (4 ml) was added, and incorporation of the [3H]sugars was quantitated by liquid scintillation counting.
HA Polymerization Assays—The PmHAS-catalyzed polymerization assay measured the incorporation of both GlcUA and GlcNAc onto acceptors to form longer HA chains as in
 |
Sugar Competition Assays—To assess the number of acceptor sites within the PmHAS polypeptide, we devised competition experiments between two distinct oligosaccharides, each having a different non-reducing terminal sugar; HA14 terminates in GlcUA, whereas HA15 terminates in GlcNAc. Both oligosaccharides were present simultaneously in a 3–4-min reaction, with only one type of radiolabeled UDP-sugar nucleotide. In this situation, one oligosaccharide served as the acceptor substrate molecule and the other as the potential competitor molecule that cannot be extended. For example, in one experiment with UDP-[3H]GlcNAc, HA14 (0.25 mM) functioned as the acceptor for the addition of the GlcNAc moiety, whereas HA15 served as the potential competitor. In the converse experiment with UDP-[3H]GlcUA, HA15 (0.15 mM) served as the acceptor for the addition of the GlcUA monosaccharide, whereas HA14 functioned as the potential competitor. The reactions without a potential competitor (only the acceptor with the appropriate non-reducing terminus) were run in parallel and served as the "100% activity" value. The products of the reactions were analyzed by paper chromatography.
Analysis of in Vitro Synthesized HA—The synthetic molecule A-F-A was used as the acceptor to synchronize the synthesis of HA. Reaction conditions were 50 mM Tris, pH 7.2, 5 mM MnCl2, 1 M ethylene glycol, 12.2 mM UDP-GlcUA, 12.2 mM UDP-GlcNAc, and 14 µM PmHAS plus A-F-A acceptor at three different concentrations (8, 80, and 800 µM) in a total volume of 25 µl. The reactions were incubated at 30 °C overnight. The size of the products was analyzed using agarose gel electrophoresis (1.2%; 1x TAE buffer (40 mM Tris acetate, 2 mM EDTA); 30 V) and Stains-All dye detection (0.005% w/v in ethanol) (13). Select-HA Lo and Hi Ladders composed of monodisperse HA polymers (10) were used as standards for size estimates (Hyalose, Oklahoma City, OK). To assess the authenticity of the HA linkages, portions of the reactions were treated with Streptomyces HA lyase, an enzyme that degrades no other glycosaminoglycan except HA. The pH value for the reaction was adjusted to pH 6 by the addition of sodium acetate (50 mM final). The reaction was boiled for 1 min at 95 °C and centrifuged to remove PmHAS. After overnight incubation with Streptomyces lyase, the sample was loaded onto the agarose gel.
To ascertain the presence of the aglycone in the product polymer chains, the reactions were adjusted to 0.2 M sodium nitrate and analyzed by high performance gel filtration chromatography on a Polysep 4000 column (1 ml/min, 0.2 M sodium nitrate; Phenomenex, Torrance, CA) with UV light absorbance detection at 272 nm for the A-F-A glycone. Fluorescent dextran standards with molecular masses of 4, 12, 50, and 580 kDa were used as calibrants (detection 490 nm).
The masses of products formed after reaction of A-F-A with PmHAS and UDP-GlcNAc were measured by mass spectrometry. Matrix-assisted laser desorption ionization time-of-flight spectra were obtained using a Voyager Elite DE mass spectrometer. The sample in water (1 µl of 0.01 µg/µl oligosaccharide) was spotted on a target plate followed by 1 µl of matrix solution (10 mg/ml 6-aza-2-thiothymine in 50% acetonitrile, 49.9% water, 0.1% trifluoroacetic acid) and then mixed and vacuum-dried. The samples were analyzed using a negative ion reflectron mode with the following parameters: acceleration, 20 kV; low mass gate, 400 Da; and delayed extraction, 200 ns.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
(1,4)-N-acetylglucosaminyl transferase = GlcNAc-transferase and UDP-glucuronic acid, N-acetylglucosaminyl: (1,3)glucuronic acidyltransferase = GlcUA-transferase) within a single polypeptide chain (2, 3). To directly compare the catalytic activities of the PmHAS GlcNAc-transferase site and GlcUA-transferase site, we performed parallel time course experiments monitoring the two single sugar addition reactions. For GlcNAc-transferase activity, HA22, an acceptor that possesses a GlcUA at the non-reducing terminus, was employed with the UDP-[3H]GlcNAc donor. HA21, an acceptor that terminates in GlcNAc, was utilized with the UDP-[3H]GlcUA donor for the GlcUA-transferase activity. The HA21 and HA22 oligosaccharides were chosen to mimic a long nascent HA polysaccharide chain but not cause the viscosity of the reaction mixture to increase. Portions of the reactions were removed at various times, quenched, and analyzed by descending paper chromatography (Fig. 1). The rate of each transferase activity corresponds to the slope (average 
V/time) at the initial phase of the reaction. The initial velocity of the GlcUA-transferase activity (6.5 nmol/min) was much more rapid (20-fold) than the GlcNAc-transferase activity (0.32 nmol/min). These values translate to an enzyme turnover number of 10 and 0.5 substrate molecules/s, respectively; for comparison, the turnover numbers for two well studied enzymes, DNA polymerase and hen egg lysozyme, are 15 and 0.5, respectively.
|
|
|
4GlcUA-3GlcNAc]n = [AN]n or [3GlcNAc-4GlcUA]n = [NA]n) oligosaccharides containing a MP group at the reducing termini (Fig. 2). The use of sugars with (a) a non-reducing termini ending in GlcUA or (b) a non-reducing termini ending in GlcNAc allows the probing of both putative acceptor sites in the model. The relative activity of the MP sugars, including AN-MP, ANA-MP, ANAN-MP, ANANAN-MP, N-MP, NA-MP, NAN-MP, NANA-MP, and NANAN-MP, were tested. The hydrophobicity of the methoxyphenol group of these HA-related oligosaccharides permits the use of solid phase extraction with a reverse phase sorbent for facile analysis of single sugars to small molecule acceptors. Fig. 3 depicts a uniform, representative data set of all of the methoxyphenol sugars from two independent experiments. For the GlcUA-transferase activity, the tetrasaccharide NANA-MP and longer served as efficient acceptors for PmHAS-catalyzed elongation (i.e. rapid reactions (4 min) and low concentrations (1–2 mM) (Fig. 3A). Conversely, for the GlcNAc-transferase activity, the trisaccharide ANA-MP and longer were efficient acceptors (Fig. 3B). In contrast, the two possible methoxyphenol disaccharides and NAN-MP were poor substrates that required longer times (30–60 min) and higher concentrations (10–30 mM) to detect sugar transfer. However, the worst acceptor, N-MP, may be elongated after extensive reactions (Table 2); thin-layer chromatography analysis of the solid phase extraction-purified product verified that this low level of activity was indeed true sugar addition and not simply an assay background problem (data not shown).
|
|
A collection of hydrophobic glycosides were tested as PmHAS acceptors: A-F, A-F-A, G-F-G, GalNAc-F-GalNAc, A-NAP, A-NP, GalA-NP, A-MUM, N-MUM, Gluc-MUM, A-F3MUM, and A-CU. Most of the substrates tested were poor PmHAS acceptors, as seen by the production of no or small amounts of elongation products, even after utilizing extensive reaction times and/or high concentrations. However, A-F-A (Fig. 2) generated a signal similar to ANA-MP (V =0.1 nmol/min at 2 mM) (Table 2). Although A-F elongation was detected (9% of the A-F-A value), the addition of the second GlcUA to produce A-F-A greatly boosted the velocity.
To verify that PmHAS was incorporating the GlcNAc sugar onto A-F-A, a single sugar addition reaction was analyzed by mass spectrometry. After a 2-h reaction incubation period, the major reaction product was a compound formed by the addition of a single GlcNAc residue (experimental = 683 Da; theoretical = 684 Da). Upon longer incubation, PmHAS added a GlcNAc to both sides of the substrate (experimental = 886 Da; theoretical = 887 Da), but the single addition product was still more abundant. Thus the increase in activity of A-F-A over A-F was not because of a simple doubling of the number of usable termini. The ability of PmHAS to extend both GlcUA groups of A-F-A (although at low levels in these experimental conditions) is probably the result of the initial NA-F-A product molecule rebinding to the enzyme and being elongated to form NA-F-AN.
Two other structurally similar compounds, G-F-G and GalNAc-F-GalNAc, however, did not show high activity (<2% of the A-F-A value). Therefore, the substrate containing both the hydrophobic component (potentially mimicking the acetyl moiety of the GlcNAc saccharide and/or hydrocarbon face of the pyranose ring) as well as two GlcUA sugars elicited the best enzyme activity, reinforcing the importance of both characteristics. The predicted spacing between the two carboxylate groups of the ANA and the A-F-A molecules are similar (estimated distances from C6-GlcUA to C6-GlcUA: ANA, 9.0 ± 0.8 Å in aqueous solution or 10.7 Å if totally extended; A-F-A 12–20 Å),4 but the actual bound conformation of the saccharides in the acceptor site is not yet known.
Synthesis of HA Polymers with Artificial Acceptors—The success of A-F-A as an acceptor substrate initiated the exploration of its utility as a primer for the synthesis of HA. An acceptor molecule will bypass the slow PmHAS initiation step, resulting in synchronized reactions that yield relatively monodisperse polymer products. Polymerization reactions with A-F-A at three different concentrations (8, 80, and 800 µM) were performed and analyzed by gel electrophoresis (Fig. 4). The size of the products were 1,500, 400, or 110 kDa for various reactions where higher concentrations of A-F-A yielded smaller chains as expected. Essentially, the presence of a low concentration of acceptor with a finite amount of UDP-sugars will synthesize large HA products; conversely, the presence of a high concentration of acceptor and the same finite amount of UDP-sugars will synthesize smaller HA products (10). Gel filtration analysis with ultraviolet absorbance detection proved that the polymer contained the fluorescein aglycone (not shown). Lyase degradation of the A-F-A reaction products proved that authentic HA chains were produced.
|
-glucuronidase cleaves the GlcUA groups proximal to the fluoresceine moiety. These degradation enzymes co-exist in the lysosome; thus, such probes should be suitable for tracking HA degradation following uptake via receptors in liver sinusoidal cells or lymph node cells. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
20-fold faster than the GlcNAc-transferase activity; GlcNAc-transfer does indeed occur, although at a much lower rate. This finding is interesting in light of the similar Km values of the UDP-sugar precursors for PmHAS (UDP-GlcNAc, 160 ± 60 µM; UDP-GlcUA, 140 ± 40 µM) (2). There may be two possible explanations for this disparity between the two transferase activities based on (a) biochemical or (b) evolutionary rationales. In the first explanation, the synthesis of the HA chain is driven by the hydrolysis of UDP-sugar precursors in which the activated bond between the oxygen and the sugar to be transferred is broken, and the sugar is then added to the growing polymer chain. The terminal phosphate of the UDP by-product gains an additional negative charge (two other negative groups are present in the original UDP-sugar, but these groups are probably chelated by the Mn2+ metal ion). In the case of UDP-GlcUA addition, it is possible that this negative charge on the UDP by-product repels the glucuronic acid sugar moiety on the terminus of the nascent HA chain because of the similar charge of the carboxylate group, thus enhancing reaction rate and/or aiding product departure from the enzyme active site. On the other hand, in the case of UDP-GlcNAc addition, the neutral hexosamine is at a relative disadvantage, because it is not subject to electrostatic repulsion with the UDP by-product.
In the second potential explanation, during evolution, PmHAS may have been forged by gene duplication and fusion events. Potential evidence for a duplication event is found within the PmHAS protein sequence itself; two tandem domains of 100 residues with similar motifs are present (2). Later, mutation of one transferase site into a new activity rendered the glycosyltransferase capable of producing a heteropolymer. It is not readily apparent if the fusion occurred before or after the mutation event. Essentially, it is possible that the adaptation of the transferase site to the new function is not complete and therefore nonoptimal, resulting in the observed disparate reaction speeds of the GlcUA-transferase and the GlcNAc-transferase activities.
Postulation of a Two-site Acceptor Binding Model—Competition between GlcUA- and GlcNAc-terminated oligosaccharides for PmHAS-mediated elongation was not observed. Therefore, Occam's razor (i.e. the simplest explanation is usually correct) was invoked to consider the possibility that PmHAS functions by utilizing at least two independent acceptor binding sites.
Several general hypothetical models of dual action glycosyltransferase structure are obvious including: 1) the "two acceptor binding sites/two donor binding sites" model, where the acceptor itself transitions from one binding site to the other site after each monosaccharide is added during elongation, (2) the "one acceptor binding site/two donor binding sites" model, where repetitive, cycling conformational change of the enzyme positions the non-reducing terminal sugar of the acceptor in the oligosaccharide binding site in close proximity to the appropriate transferase site, or (3) the "one acceptor binding site/one donor binding site" model, where repetitive conformational change creates/locates the appropriate catalytic machinery within the acceptor and donor binding sites for catalytic activity. Model 1 best depicts the observed PmHAS characteristics based on our new data and previous work by Jing and DeAngelis (2, 3).
Interestingly, the HA15 competitor appears to cause a slight increase in the GlcNAc-transferase activity (but the converse competitor HA14 does not stimulate GlcUA-transferase activity), suggesting potential conformational feedback or cooperativity between the two putative acceptor sites. This issue will require further study.
On basic physical premises, it is expected that polymerization catalyzed by a model 1 enzyme is intrinsically non-processive (i.e. the nascent chain is repeatedly released and rebound before the polymer is completed), whereas a model 2 or 3 enzyme is potentially processive (i.e. the nascent chain is not released until the polymer is completed). In several tests, including in vitro syntheses of oligosaccharides (11) and polysaccharides (10), PmHAS appears to operate in a non-processive fashion. The prototypical streptococcal HA synthase enzyme SpHAS (or HasA), which possesses a very different amino acid sequence, exhibits processive behavior in vitro (14).
Proposed Model of a Sugar Binding Pocket with Important Carboxylate Contacts—At this time, three-dimensional structures are not available for polysaccharide synthases. However, crystal structures have been solved in the presence of both donor and acceptor substrates for two non-homologous single action glycosyltransferases, GlcUAT1 and EXTL2 (15, 16); their amino acid sequences are not similar to the PmHAS sequence, but their transferase activities are similar. The natural acceptor substrates are the trisaccharide Gal-Gal-Xyl-O-Ser-protein for GlcUAT1 and the tetrasaccharide GlcUA-Gal-Gal-Xyl-O-Ser-protein (but a GlcUA-Gal-glycoside will work in vitro) for EXTL2, a Glc-NAc transferase. The structures of these acceptors are not identical to the repeating glycosaminoglycan backbone; however, the structural information indicates that interactions with the non-reducing disaccharide portion of the acceptor substrates determine acceptor specificity for these two enzymes (17). Essential oligosaccharide acceptor substrate binding sites have been determined for a number of carbohydrate-active enzymes. For example, the rate of lysozyme-catalyzed hydrolysis of oligomers of GlcNAc increases distinctively as the number of residues is increased from five to six; the substrate binding pocket accommodates a hexasaccharide as determined by crystallography (18, 19). A maltononaose binding site was observed in the crystal structure of the cyclodextrin transferase of Bacillus circulans strain 251 (20), illustrating that acceptors of nine sugars in length are not outside the realm of possibility.
Our experimental model system of PmHAS has allowed us to analyze protein-oligosaccharide interactions indirectly. Probing the active site of PmHAS with the series of methoxyphenol sugars of different lengths and other various acceptor substrates potentially reveals information about the acceptor specificity of PmHAS in the absence of a crystal structure. The kinetic data allow the ranking of various lengths of sugar acceptors to determine the optimal length of the sugar polymer necessary for efficient PmHAS chain elongation. The PmHAS GlcUA-transferase site efficiently elongates the tetrasaccharide NANA-MP at a low concentration during short incubation periods, whereas the PmHAS GlcNAc-transferase site efficiently elongated the trisaccharide ANA-MP. Therefore, our data suggest the acceptor binding sites of PmHAS contain pockets that can bind at least three or four monosaccharides for the GlcNAc-transferase or the GlcUA-transferase, respectively.
The minimal length acceptors demonstrating efficient elongation are oligosaccharides that contain the trisaccharide element ANA. The predilection for ANA-MP over NAN-MP suggests there are important contacts between the carboxylate groups of the two GlcUA sugars and the acceptor binding site of PmHAS. The substantial increase in the PmHAS elongation efficiency for the A-F-A acceptor, the simple proxy for ANA-MP, in comparison to the A-F acceptor also supports the hypothesis that the two GlcUA groups provide important enzyme contacts. Overall, this study should lead to synthesis of better analogs (e.g. higher efficiency, less expensive, animal-free manufacture) in the future.
|
FOOTNOTES |
|---|
1 To whom correspondence should be addressed: Dept. of Biochemistry and Molecular Biology, Univ. of Oklahoma Health Sciences Ctr., 940 Stanton L. Young Blvd., Oklahoma City, OK 73104. Tel.: 405-271-2227; Fax: 405-271-3092; E-mail: paul-deangelis{at}ouhsc.edu.
2 The abbreviations used are: HA, hyaluronan, hyaluronic acid, or hyaluronate; PmHAS, P. multocida hyaluronan synthase; PmCS, P. multocida Type F chondroitin synthase; GlcUA, glucuronic acid; MP, methoxyphenol.
3 K. J. Williams, F. M. Haller, and P. L. DeAngelis, unpublished data.
4 A. Almond, estimates from preliminary molecular modeling, dynamic simulations; personal communication.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. H. Weigel and P. L. DeAngelis Hyaluronan Synthases: A Decade-plus of Novel Glycosyltransferases J. Biol. Chem., December 21, 2007; 282(51): 36777 - 36781. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Sismey-Ragatz, D. E. Green, N. J. Otto, M. Rejzek, R. A. Field, and P. L. DeAngelis Chemoenzymatic Synthesis with Distinct Pasteurella Heparosan Synthases: MONODISPERSE POLYMERS AND UNNATURAL STRUCTURES J. Biol. Chem., September 28, 2007; 282(39): 28321 - 28327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. S. Tracy, F. Y. Avci, R. J. Linhardt, and P. L. DeAngelis Acceptor Specificity of the Pasteurella Hyaluronan and Chondroitin Synthases and Production of Chimeric Glycosaminoglycans J. Biol. Chem., January 5, 2007; 282(1): 337 - 344. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A. Kane, C. L. White, and P. L. DeAngelis Functional Characterization of PmHS1, a Pasteurella multocida Heparosan Synthase J. Biol. Chem., November 3, 2006; 281(44): 33192 - 33197. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
