FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2019487637> ?p ?o ?g. }

• W2019487637 endingPage "28327" @default.
• W2019487637 startingPage "28321" @default.
• W2019487637 abstract "Heparosan (-GlcUA-β1,4-GlcNAc-α1,4-)n is a member of the glycosaminoglycan polysaccharide family found in the capsule of certain pathogenic bacteria as well as the precursor for the vertebrate polymers, heparin and heparan sulfate. The two heparosan synthases from the Gram-negative bacteria Pasteurella multocida, PmHS1 and PmHS2, were efficiently expressed and purified using maltose-binding protein fusion constructs. These relatively homologous synthases displayed distinct catalytic characteristics. PmHS1, but not PmHS2, was able to produce large molecular mass (100–800 kDa) monodisperse polymers in synchronized, stoichiometrically controlled reactions in vitro. PmHS2, but not PmHS1, was able to utilize many unnatural UDP-sugar analogs (including substrates with acetamido-containing uronic acids or longer acyl chain hexosamine derivatives) in vitro. Overall these findings reveal potential differences in the active sites of these two Pasteurella enzymes. In the future, these catalysts should allow the creation of a variety of heparosan and heparinoids with utility for medical applications. Heparosan (-GlcUA-β1,4-GlcNAc-α1,4-)n is a member of the glycosaminoglycan polysaccharide family found in the capsule of certain pathogenic bacteria as well as the precursor for the vertebrate polymers, heparin and heparan sulfate. The two heparosan synthases from the Gram-negative bacteria Pasteurella multocida, PmHS1 and PmHS2, were efficiently expressed and purified using maltose-binding protein fusion constructs. These relatively homologous synthases displayed distinct catalytic characteristics. PmHS1, but not PmHS2, was able to produce large molecular mass (100–800 kDa) monodisperse polymers in synchronized, stoichiometrically controlled reactions in vitro. PmHS2, but not PmHS1, was able to utilize many unnatural UDP-sugar analogs (including substrates with acetamido-containing uronic acids or longer acyl chain hexosamine derivatives) in vitro. Overall these findings reveal potential differences in the active sites of these two Pasteurella enzymes. In the future, these catalysts should allow the creation of a variety of heparosan and heparinoids with utility for medical applications. Heparosan (N-acetylheparosan) (-GlcUA-β1,4-GlcNAc-α1,4-) is the repeating sugar backbone of the polysaccharide found in the capsule of certain pathogenic bacteria as well as the biosynthetic precursor of heparin or heparan sulfate found in animals from hydra to vertebrates (1DeAngelis P.L. Anat. Rec. 2002; 268: 317-326Crossref PubMed Scopus (87) Google Scholar, 2Esko J.D. Lindahl U. J. Clin. Investig. 2001; 108: 169-173Crossref PubMed Scopus (787) Google Scholar, 3Sugahara K. Kitagawa H. IUBMB Life. 2002; 54: 163-175Crossref PubMed Scopus (217) Google Scholar). In mammals, the sulfated forms bind to a variety of extremely important polypeptides including hemostasis factors (e.g. antithrombin III, thrombin), growth factors (e.g. EGF, VEGF), and chemokines (e.g. interleukin-8, platelet factor 4) as well as the adhesive proteins for viral pathogens (e.g. herpes, Dengue fever) (4Capila I. Linhardt R.J. Angew. Chem. Int. Ed. Engl. 2002; 41: 391-412Crossref PubMed Scopus (1544) Google Scholar, 5Bernfield M. Gotte M. Park P.W. Reizes O. Fitzgerald M.L. Lincecum J. Zako M. Annu. Rev. Biochem. 1999; 68: 729-777Crossref PubMed Scopus (2317) Google Scholar). Currently, heparin is extracted from animal tissue and used as an anticoagulant or antithrombotic drug. In the future, similar polymers and derivatives should also be useful for pharmacological intervention in a variety of pathologic conditions including neoplasia and viral infection. Several enzyme systems have been identified that synthesize heparosan. In bacteria, either a pair of two separate glycosyltransferases (Escherichia coli KfiA and KfiC) (6Hodson N. Griffiths G. Cook N. Pourhossein M. Gottfridson E. Lind T. Lidholt K. Roberts I.S. J. Biol. Chem. 2000; 275: 27311-27315Abstract Full Text Full Text PDF PubMed Google Scholar) or a single glycosyltransferase (Pasteurella multocida PmHS1 or PmHS2) 3The abbreviations used are: PmHSP. multocida heparosan synthaseGlcUAglucuronic acidGlcNAcN-acetylglucosamineMBPmaltose-binding proteinMALDI-TOF MSmatrix-assisted laser desorption ionization time of flight mass spectrometrySEC-MALLSsize exclusion chromatography-multiangle laser light scatteringHAhyaluronic acid or hyaluronanPPAinorganic pyrophosphataseA-F-Afluorescein di-β-d-glucuronideA-F-ANGlcUA-F-GlcUA-GlcNAcNA-F-ANGlcNAc-GlcUA-F-GlcUA-GlcNAc. 3The abbreviations used are: PmHSP. multocida heparosan synthaseGlcUAglucuronic acidGlcNAcN-acetylglucosamineMBPmaltose-binding proteinMALDI-TOF MSmatrix-assisted laser desorption ionization time of flight mass spectrometrySEC-MALLSsize exclusion chromatography-multiangle laser light scatteringHAhyaluronic acid or hyaluronanPPAinorganic pyrophosphataseA-F-Afluorescein di-β-d-glucuronideA-F-ANGlcUA-F-GlcUA-GlcNAcNA-F-ANGlcNAc-GlcUA-F-GlcUA-GlcNAc. (7DeAngelis P.L. White C.L. J. Biol. Chem. 2002; 277: 7209-7213Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar) has been shown to polymerize heparosan; the enzymes from both species are homologous at the protein level. In vertebrates, a pair of enzymes, EXT 1 and EXT 2 (2Esko J.D. Lindahl U. J. Clin. Investig. 2001; 108: 169-173Crossref PubMed Scopus (787) Google Scholar, 3Sugahara K. Kitagawa H. IUBMB Life. 2002; 54: 163-175Crossref PubMed Scopus (217) Google Scholar, 9Zak B.M. Crawford B.E. Esko J.D. Biochim. Biophys. Acta. 2002; 1573: 346-355Crossref PubMed Scopus (146) Google Scholar), that are not similar to the bacterial systems appear to be responsible for producing the repeating units of the polymer chain, which is then subsequently modified by sulfation and epimerization (2Esko J.D. Lindahl U. J. Clin. Investig. 2001; 108: 169-173Crossref PubMed Scopus (787) Google Scholar, 3Sugahara K. Kitagawa H. IUBMB Life. 2002; 54: 163-175Crossref PubMed Scopus (217) Google Scholar). P. multocida heparosan synthase glucuronic acid N-acetylglucosamine maltose-binding protein matrix-assisted laser desorption ionization time of flight mass spectrometry size exclusion chromatography-multiangle laser light scattering hyaluronic acid or hyaluronan inorganic pyrophosphatase fluorescein di-β-d-glucuronide GlcUA-F-GlcUA-GlcNAc GlcNAc-GlcUA-F-GlcUA-GlcNAc. P. multocida heparosan synthase glucuronic acid N-acetylglucosamine maltose-binding protein matrix-assisted laser desorption ionization time of flight mass spectrometry size exclusion chromatography-multiangle laser light scattering hyaluronic acid or hyaluronan inorganic pyrophosphatase fluorescein di-β-d-glucuronide GlcUA-F-GlcUA-GlcNAc GlcNAc-GlcUA-F-GlcUA-GlcNAc. The heparosan synthases from P. multocida possess both a hexosamine and a glucuronic acid transfer site in the same polypeptide chain, as shown by mutagenesis studies (10Kane T.A. White C.L. DeAngelis P.L. J. Biol. Chem. 2006; 281: 33192-33197Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar), and are therefore referred to as “dual action” or bifunctional glycosyltransferases. These enzymes are complex because they employ both an inverting and a retaining mechanism when transferring the monosaccharide from UDP precursors to the nonreducing terminus of a growing chain. The two Pasteurella heparosan synthases, PmHS1 and PmHS2, are ∼70% identical at the amino acid sequence level. The two genes are found in different regions of the bacterial chromosome: PmHS1 (hssA) is associated with the prototypical Gram-negative type II carbohydrate biosynthesis gene locus, but PmHS2 (hssB) resides far removed in an unspecialized region (8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar). Here we have demonstrated that the catalytic utility of PmHS1 and PmHS2 are very distinct as measured by various criteria including their ability to produce polymers either with monodisperse size distributions or with unnatural sugar compositions. Materials—Chemicals were purchased from Sigma-Aldrich or Fisher unless otherwise noted. Custom synthetic DNA oligonucleotides were from Integrated DNA Technologies (Coralville, IA). Various HA molecular weight standards were obtained from Hyalose LLC (Oklahoma City, OK). Construction and Purification of Recombinant Maltose-binding Fusion Construct Proteins—PmHS1 and PmHS2 were both expressed as a carboxyl-terminal fusion to maltose-binding protein (MBP) using the pMAL-c2X vector (New England Biolabs, Beverly, MA). Polymerase chain reaction was employed to subclone the open reading frames from our previous pET-Blue-1 constructs (7DeAngelis P.L. White C.L. J. Biol. Chem. 2002; 277: 7209-7213Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar). For PmHS1, new unique flanking restriction sites (BamHI and HindIII) were added with the primers used for amplification (15 cycles: 94 °C, 30 s; 52 °C, 30 s; 72 °C, 2 min) with Taq DNA polymerase. For PmHS2, restriction sites (BamHI and PstI) were added with the primers used for amplification (21 cycles: 94 °C, 30 s; 52 °C, 30 s; 72 °C, 3.5 min) with Pfu DNA polymerase (Stratagene, La Jolla, CA). The amplicons were gel-purified, restriction-digested with both appropriate enzymes (Promega, Madison, WI), and ligated to similarly double cut pMAL-c2X plasmid. E. coli One Shot® Top 10F′ (Invitrogen) was used for the initial transformation on LB ampicillin (100 μg/ml) plates and grown at 30 °C. All constructs were confirmed by DNA sequencing (OMRF Sequencing Facility, Oklahoma City, OK). To facilitate extracting the enzymes, the expression host E. coli XJa (Zymo Research, Orange, CA), which encodes a phage lysin enzyme, was employed allowing for simple freeze/thaw lysis. Cultures were grown in Superior Broth (AthenaES, Baltimore, MD) at 30 °C with ampicillin (100 μg/ml), and l-arabinose (3.25 mm; to induce the lysin enzyme). At mid-log phase, isopropyl β-d-1-thiogalactopyranoside (0.2 mm final) was added to induce fusion protein production. One hour after induction, the cultures were supplemented with fructose (12.8 mm final) and grown for ∼5–12 h before harvesting by centrifugation at 4 °C. The bacteria were resuspended in 20 mm Tris, pH 7.2, and protease inhibitors (p-[4–2-aminoethyl]benzenesulfonyl fluoride hydrochloride, pepstatin, benzamidine, N-[N-(l-3-trans-carboxyoxirane-2-carbonyl)-l-leucyl]-agmatine, leupeptin) on ice, and frozen and thawed twice allowing lysin to degrade the cell walls. The lysates was clarified by centrifugation. The synthases were affinity purified via the MBP unit using amylose resin (New England Biolabs). After extensive washing with column buffer (20 mm Tris, pH 7.2, 200 mm NaCl, 1 mm EDTA), the protein was eluted in column buffer containing 10 mm maltose. Protein concentration was quantitated by the Bradford assay (Pierce) using a bovine serum albumin standard. The purification was monitored by SDS-PAGE with copper negative staining or Coomassie Blue staining (protein molecular weight standards, Bio-Rad). Western blot analyses employed either (a) a rabbit anti-peptide antibody directed to the KGDIIFFQDSDDVCHHERIER sequence of PmHS1 and PmHS2 (8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar) or (b) a rabbit anti-MBP antibody (Santa Cruz Biotechnology, Santa Cruz, CA) followed by detection with protein A-alkaline phosphatase conjugate (Calbiochem) and 5-bromo-4-chloro-3′-indolyphosphate p-toluidine salt/nitro-blue tetrazolium chloride. Yields were about 1–2 mg of protein/liter of culture. Aliquots were stored at -80 °C; the activity of the synthases was stable to several freeze/thaw cycles. Western blotting was used to normalize the amounts of intact PmHS1 and PmHS2 polypeptide used in kinetic assays. Acceptor Preparations—The ∼55-kDa heparosan polysaccharide acceptor (8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar) was used as a positive control and a normalization factor for many experiments. Heparosan oligosaccharides ((GlcUA-GlcNAc)n-(GlcUA-anhydromannitol), n = 1, 2, or 3) were prepared by partial deacetylation with base, nitrous acid hydrolysis, and reduction (11Chen M. Bridges A. Liu J. Biochemistry. 2006; 45: 12358-12365Crossref PubMed Scopus (43) Google Scholar); these polymers contain intact nonreducing termini but an anhydromannitol group at the reducing end. The fragments were purified by gel filtration on a P2 column (Bio-Rad) in 0.2 m ammonium formate followed by normal phase thin layer chromatography (TLC) on silica plates (Whatman) with butyl alcohol/acetic acid/water (1:1:1). The bands were detected by staining the side lanes with napthoresorcinol (12DeAngelis P.L. Oatman L.C. Gay D.F. J. Biol. Chem. 2003; 278: 35199-35203Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar). The size and purity of oligosaccharides were verified by matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS) (12DeAngelis P.L. Oatman L.C. Gay D.F. J. Biol. Chem. 2003; 278: 35199-35203Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar). Fluorescein di-β-d-glucuronide (A-F-A; Molecular Probes, Eugene, OR), a commercially available synthetic acceptor that mimics a glycosaminoglycan trisaccharide that was identified in previous work with the Pasteurella HA synthase (13Williams K.J. Halkes K.M. Kamerling J.P. DeAngelis P.L. J. Biol. Chem. 2006; 281: 5391-5397Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar), was used as the starting material to prepare the A-F-AN (GlcUA-F-GlcUA-GlcNAc) and NA-F-AN (GlcNAc-GlcUA-F-GlcUA-GlcNAc) acceptors using UDP-GlcNAc with PmHS2 under conditions described later under “Single Sugar Addition Assays.” These longer acceptors were purified by TLC (silica plates and butyl alcohol/acetic acid/water, 2:1:1). In addition to lower costs than authentic oligosaccharides, the A-F-A-based acceptors also yielded better signals in mass spectrometric analyses. Polymerization Assays—Radiolabeled sugar incorporation assays using UDP-[3H]GlcUA or UDP-[3H]GlcNAc (PerkinElmer Life Sciences) (13Williams K.J. Halkes K.M. Kamerling J.P. DeAngelis P.L. J. Biol. Chem. 2006; 281: 5391-5397Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar) were employed to monitor substrate usage and pH dependence. Polymerization reactions (40 μl) typically contained 50 mm Tris, pH 7.2, 1 mm MnCl2, 0.5–1.0 mm UDP-GlcNAc, 0.5–1 mm UDP-GlcUA, 0.1 μCi of UDP-[3H]sugar, and ∼4–5 μg of enzyme (unless noted). The reactions were typically incubated at 30 °C for 15–30 min and then stopped with SDS (2% final). Descending paper chromatography was used to separate unincorporated UDP-sugars from the polymer product (i.e. longer than ∼14 sugars) (13Williams K.J. Halkes K.M. Kamerling J.P. DeAngelis P.L. J. Biol. Chem. 2006; 281: 5391-5397Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar). Three separate experiments were completed for each data set, and each assay was verified to be in the linear range with respect to time and enzyme concentration. Less than 5% of the UDP-sugar was consumed. For acceptor usage tests, a “no acceptor” control was performed for all assays to determine the de novo initiation heparosan synthesis for each enzyme; this value was subtracted from the value attained in parallel assays with an acceptor. For initial tests querying the ability of a synthase to misincorporate a non-authentic UDP-sugar donor, the test compound was used at ∼0.5–1.5 mm in the presence of carrier-free UDP-[3H]GlcUA or UDP-[3H]GlcNAc (0.1 μCi) without acceptor for 12–48 h. Subsequent assays to obtain relative kinetic rates employed radiolabeled authentic UDP-sugar diluted to 0.6 mm for 90 or 180 min. Monodisperse Heparosan Synthesis—A mixture of heparosan tetrasaccharide and hexasaccharide (∼1:4 by mass, respectively) acceptors was used to prime synthesis of narrow size distribution heparosan polymers. The length of the chain was controlled by altering the stoichiometry of the UDP-sugar to acceptor as for the nonhomologous PmHAS, the Pasteurella HA synthase (14Jing W. DeAngelis P.L. J. Biol. Chem. 2004; 279: 42345-42349Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar). These reactions typically contained 5 mm UDP-sugars in the same reaction buffer used for polymerization assays and were incubated overnight at 30 °C. Single Sugar Addition Assays—Tests were performed under the same buffer conditions as for polymerization assays but contained only a single UDP-sugar substrate (∼1–3 mm final) and an appropriately terminated acceptor (e.g. heparosan oligosaccharide, A-F-A or A-F-AN). The incorporation of sugar from the donor substrate was also detected by the increase in mass to the appropriate predicted molecular weight product by MALDI-TOF MS. Sugar Analysis Techniques—The sizes of the heparosan polymers were analyzed with agarose gels (1.2–3%, 1× Tris acetate-EDTA buffer, 30 V for 5–6 h) and Stains-All (1-ethyl-2-[3-(1-ethylnaphtho[1,2-d]thiazolin-2-ylidene)-2-methylpropenyl]naphtho-[1,2-d]thiazolium bromide) detection (15Lee H.G. Cowman M.K. Anal. Biochem. 1994; 219: 278-287Crossref PubMed Scopus (260) Google Scholar) or with polyacrylamide gels (15%, 29:1 monomer:bis, 1× Tris borate-EDTA, 250 V for 30 min) and Alcian blue staining (16Ikegami-Kawai M. Takahashi T. Anal. Biochem. 2002; 311: 157-165Crossref PubMed Scopus (45) Google Scholar). The size distribution of the polymers was determined by high performance size exclusion chromatography-multi-angle laser light scattering (14Jing W. DeAngelis P.L. J. Biol. Chem. 2004; 279: 42345-42349Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar). For yield estimates, the polymers were purified by heat treatment (95 °C, 1 min) and centrifugation (14,000 × g) to remove protein followed by desalting on PD10 columns (Sephadex G25; GE Healthcare) eluted with 0.2 m ammonium formate, a volatile buffer. The appropriate void volume fractions were pooled and lyophilized three times. The concentration of authentic heparosan polymers was assayed using the carbazole assay for uronic acid with GlcUA as the standard (17Bitter T. Muir H.M. Anal. Biochem. 1962; 4: 330-334Crossref PubMed Scopus (5192) Google Scholar). The unnatural GlcUA analogs used here did not have significant color yield in this assay (due to substitution of the pyranose ring), and therefore relative comparisons were made by densitometric analysis (ImageJ, version 1.37; rsb.info.nih.gov/ij/) of Alcian blue-stained gels with a titration of matched sets of natural and unnatural polymers. Sugar molecular masses were measured by MALDI-TOF MS (Voyager Elite DE, Applied Biosystems Inc.; reflector mode, negative ions) using 6-aza-2-thiothymine or 2,4,6-trihydroxy-acetophenone matrix (5–10 mg/ml in 50% acetonitrile, 0.1% trifluoroacetic acid) (12DeAngelis P.L. Oatman L.C. Gay D.F. J. Biol. Chem. 2003; 278: 35199-35203Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar). To assess the linkages of the polymers, samples were treated with heparin lyase III (recombinant E. coli-derived Flavobacterium heparinum, EC 4.2.2.8) (18Duncan M.B. Liu M. Fox C. Liu J. Biochem. Biophys. Res. Commun. 2006; 339: 1232-1237Crossref PubMed Scopus (37) Google Scholar) in 50 mm potassium phosphate, pH 7, at 37 °C for ∼6 h. UDP-glucosamine Synthesis—UDP-glucosamine (UDP-GlcN) was chemoenzymatically synthesized from GlcN-1-PO4 and UTP using immobilized recombinant GalU (UDP-glucose pyrophosphorylase) and PPA (inorganic pyrophosphatase) enzymes from E. coli. GalU and PPA were both expressed as amino-terminal His6-tagged proteins with pET15b (Novagen) constructs (19Liu Z. Zhang J. Chen X. Wang P.G. Chembiochem. 2002; 3: 348-355Crossref PubMed Scopus (94) Google Scholar). Enzymes were freeze/thaw extracted from the cell paste of E. coli XJa (DE3) (Zymo Research) in 50 mm Tris, pH 8.0, 2 mm β-mercaptoethanol, 10% glycerol, 100 mm NaCl, and 1 mm imidazole in the presence of protease inhibitors and DNase/RNase (10 μg/ml). Each enzyme was immobilized in a purified state via the His6 tags using nickel-nitrilotriacetic acid SuperFlow resin (Qiagen, Valencia, CA). After washing with extraction buffer containing 10 mm imidazole, the resins were equilibrated with a reaction buffer composed of 50 mm Tris, pH 7.8, 2 mm K2HPO4, and 10 mm MgCl2. A mixture of resins containing bound GalU and PPA enzymes was incubated overnight at room temperature with 15–39 mm GlcN-1-PO4 and 15–39 mm UTP in reaction buffer. UDP-GlcN was purified by preparative TLC (silica with butyl alcohol/acetic acid/water, 1.75:1:1) with UV shadowing (254 nm) detection followed by gel filtration on a P2 column in 0.2 m ammonium formate. The pure compound was lyophilized three times from water, and the negative ion molecular mass was verified by MALDI-TOF MS (predicted 564.03 Da, observed 564.13 Da). UDP-N-propionylglucosamine and UDP-N-butyrylglucosamine Synthesis—UDP-GlcN (15–25 mg/ml) was mixed with saturated NaHCO3 (pH∼10, 2–3 volumes), and any sodium silicate precipitate was removed by centrifugation. For UDP-N-propionylglucosamine (UDP-GlcNPro) and UDP-N-butyrylglucosamine (UDP-GlcNBut) preparation, a 5% solution of propionic anhydride or butyric anhydride in water (65 molar excess) was added and reacted at room temperature for 5 or 15 min, respectively. Reactions were desalted using a P2 column and lyophilization as for UDP-GlcN. The product negative ion molecular masses were verified by MALDI-TOF MS (UDP-GlcNPro predicted 620.06 Da, observed 620.38 Da; UDP-GlcNBut predicted 634.08 Da, observed 634.12 Da). UDP-α-d-N-acetylglucosaminuronic Acid, UDP-2,3-diacetamido-2,3-dideoxy-α-d-glucose, and UDP-2,3-diacetamido-2,3-dideoxy-α-d-glucuronic Acid Synthesis—As we reported recently (20Rejzek M. Mukhopadhyay B. Wenzel C.Q. Lam J.S. Field R.A. Carbohydr. Res. 2007; 342: 460-466Crossref PubMed Scopus (16) Google Scholar), UDP-α-d-N-acetyl-glucosaminuronic acid (UDP-GlcNAcUA), an analog possessing identical functional groups at the same positions as the authentic pair of donors, was prepared by direct oxidation of commercial UDP-α-d-N-acetyl-glucosamine (UDP-GlcNAc) using platinum-catalyzed oxidation with molecular oxygen. Following purification by strong anion exchange chromatography, the resulting triammonium salt of UDP-GlcNAcUA was converted to the more water-soluble trisodium salt by treatment with a cation exchange resin (Dowex 50WX8–200, Na+ form). UDP-2,3-diacetamido-2,3-dideoxy-α-d-glucose (UDP-GlcdiNAc), an analog with an additional amide group at the C-3 position as compared with authentic UDP-GlcNAc, was synthesized from commercially available N-acetyl-d-glucosamine by a multistep synthesis (full details will be published elsewhere). UDP-2,3-diacetamido-2,3-dideoxy-α-d-glucuronic acid (UDP-Glcdi-NAcUA) was prepared by direct oxidation of UDP-GlcdiNAc using the platinum-catalyzed oxidation procedure described above (20Rejzek M. Mukhopadhyay B. Wenzel C.Q. Lam J.S. Field R.A. Carbohydr. Res. 2007; 342: 460-466Crossref PubMed Scopus (16) Google Scholar). The triammonium salt was again converted into the trisodium salt for use in enzymatic reactions. Improved Catalysts with Distinct Kinetic Properties—The maltose-binding protein/heparosan synthase fusion constructs had greatly increased protein expression in comparison with the earlier generation thioredoxin PmHS1 fusion construct (10Kane T.A. White C.L. DeAngelis P.L. J. Biol. Chem. 2006; 281: 33192-33197Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar). The MBP also allowed for efficient purification as depicted in Fig. 1; in contrast, the thioredoxin affinity handle was observed to leach substantial amounts of target protein, thus thwarting purification attempts (data not shown). Furthermore, both MBP-PmHS constructs possessed increased stability at useful reaction temperatures (e.g. active at pH 7.2 at 30 °C for 24 h). Previous studies on the efficiency of cognate acceptor utilization by the crude native sequence enzymes suggested that these relatively homologous Pasteurella synthases had different catalytic properties; the acceptor stimulated PmHS1 sugar incorporation ∼7–25-fold (by serving as a primer to circumvent the slow initiation step), whereas PmHS2 is boosted only by ∼2.5-fold (7DeAngelis P.L. White C.L. J. Biol. Chem. 2002; 277: 7209-7213Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar). These levels of acceptor stimulation were also observed for the purified fusion enzymes. In polymerization assays without acceptor, it also appears that purified PmHS2 has an ∼2-fold higher level of de novo initiation of sugar chains compared with purified PmHS1 (∼5.2 versus ∼2.6 pmol of monosaccharide transferred/min/μg protein). On the other hand, it appears that PmHS1 has an elongation rate that is ∼3-fold faster than PmHS2 (∼76 versus ∼28 pmol of monosaccharide transferred/min/μg protein). The pH profiles as determined by polymerization assays were also different; the purified PmHS1 catalyst preferred a neutral pH, whereas purified PmHS2 preferred acidic (pH ∼4–5) conditions (Fig. 2). This result of differential activity was not expected considering that the protein sequences of PmHS1 and PmHS2 are relatively homologous. Simplistically, based on expected amino acid side-chain pKa values, it may be likely that one or more histidine residues in PmHS2 (but not present in PmHS1) are protonated at the lower pH, thus gaining a positive charge, making a better contact, or providing improved electrostatic steering for a negatively charged substrate (either a heparosan oligosaccharide or a UDP-sugar). Alternatively, one or more glutamate or aspartate residues in PmHS2 (but not PmHS1) are protonated at lower pH and thus neutralized, reducing potential electrostatic repulsion of a negatively charged substrate. It is important to note that even though PmHS2 prefers the acidic pH for maximal activity, this catalyst is not very stable in those conditions. The PmHS2 protein did not demonstrate noticeable additional proteolysis at low pH as assessed by Western blotting (not shown); thus the loss of activity must be due to denaturing via an unfolding event. We examined the size of acceptor oligosaccharide preferred by each of the synthases. The heparosan tetrasaccharide was about ∼150- and ∼8-fold better than the corresponding disaccharide for PmHS1 or PmHS2, respectively. In addition, the synthetic glycoside, A-F-A, was also a useful acceptor for PmHS1 and PmHS2. These findings suggest that the size of the active site pockets of the heparosan synthases may be similar to those hypothesized for the acceptor sites of PmHAS, the HA synthase (13Williams K.J. Halkes K.M. Kamerling J.P. DeAngelis P.L. J. Biol. Chem. 2006; 281: 5391-5397Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar); a site that appears to bind 3 or 4 monosaccharides is hypothesized to make contact with the nascent HA chain. Monodisperse Heparosan—Synchronized polymerization reactions should result in monodisperse polymers as observed previously for PmHAS (14Jing W. DeAngelis P.L. J. Biol. Chem. 2004; 279: 42345-42349Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar). The formation of heparosan with a narrow size distribution is dependent on the ability of the glycosyltransferase to be primed by acceptors (thus avoiding a slow de novo initiation event yielding out-of-step elongation events) and to efficiently transfer monosaccharides from UDP-sugars. It is likely that PmHS1 catalyzes the synthesis of higher molecular weight monodisperse polymer when compared with PmHS2 because of its better ability to utilize and rapidly elongate exogenously supplied acceptors. As determined by agarose gel and SEC-MALLS analyses (Fig. 3), PmHS1 produced various sizes of monodisperse high molecular weight heparosan (∼70% average yield based on starting UDP-sugars), whereas under identical conditions PmHS2 did not. Such monodisperse heparosan polymer may serve as the starting material for the creation of defined molecules that are more predictable with respect to biological responses and potency. Polymers with Unnatural Structures—The testing of a variety of different UDP-sugar donor substrates is a means of determining the tolerance of the synthase active sites for a variety of donor and acceptor functional groups. Characterizing donor preference is obvious, but in the case of a polymer with repeating saccharide units, once a sugar is added onto the nonreducing terminus of the nascent chain, the unnatural sugar then serves as an acceptor substrate. Therefore, a successful analog must be able to play multiple roles to produce a polysaccharide chain. MALDI-TOF MS analyses and radiolabeled sugar incorporation assays revealed that PmHS2 has the ability to catalyze the incorporation of several unnatural donor sugar analogs, whereas PmHS1 appears much more strict (Table 1, Fig. 4). The GlcNAc-transferase site of PmHS2 will accept different acyl chain lengths at the C-2 position as long as the amine is acylated but does not appear to tolerate substitution at the C-3 or C-5 positions. Interestingly, UDP-GlcNPro, the UDP-GlcNAc analog with an extra methylene group in the acyl chain, is preferred by both enzymes more than the authentic substrate; perhaps a hydrophobic pocket is responsible for this catalyst/substrate contact. However, this pocket must have limited dimensions because UDP-GlcNBut, a molecule with two more additional methylene units than the authentic donor, is a worse substrate. By analogy, a hydrophobic pocket on glycosaminoglycan-binding proteins or receptors such as the HA-binding site of TSG-6 (21Blundell C.D. Almond A. Mahoney D.J. DeAngelis P.L. Campbell I.D. Day A.J. J. Biol. Chem. 2005; 280: 18189-18201Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar) may bind with higher affinity to polymers containing hexosamines with longer acyl chains, and thus new sugar ligand derivatives with more potent inhibition or signaling effects may be possible.TABLE 1Donor substrate usage by PmHS1 and PmHS2 Each UDP-sugar analog was tested for its ability to substitute for UDP-GlcNAc or UDP-GlcUA by radiolabeled sugar polymerization assays and paper chromatography. An authentic UDP-sugar and the appropriate second UDP-sugar analog (e.g. UDP-GlcUA and a potential UDP-GlcNAc substitute) were co-incubated with enzyme. The rates for the combination of both authentic donors, UDP-GlcNAc and UDP-GlcUA, are set to 100%. Analogs are presented as: +++, >200%; ++, 100-11%; +, 10-1%; -, <∼0.2%; NA, not applicable. All positive compounds were verified by single sugar addition assays with mass spectrometry (Fig. 4) except for UDP-GlcdiNAcUA because of low transfer efficiency. Overall, PmHS2 can misincorporate several analogs, but PmHS1 appears to have more restricted donor usage.UDP-sugar analogPmHS1 substitutes for:PmHS2 substitutes for:UDP-GlcUA?UDP-GlcNAc?UDP-GlcUA?UDP-GlcNAc?UDP-GlcNNA–NA–UDP-GlcNAcUA––+–UDP-GlcdiNAcNA–NA–UDP-GlcdiNAcUA––+–UDP-GlcNButNA–NA+UDP-GlcNProNA+++NA+++ Open table in a new tab The GlcUA-transferase site of PmHS2, but not PmHS1, is tolerant of extra chemical groups at the C-2 or C-3 position (Table 1). The UDP-GlcNAcUA analog possesses within a single pyranose unit both the C-6 carboxylate and the C-2 acetylated amide groups (normally found separately on two adjacent pyranose units in native heparosan). It was observed that the PmHS2 enzyme utilizes this analog to only substitute for the uronic acid unit of the disaccharide repeat and not the hexosamine (Table 1). At this time, it is difficult to predict whether this analog fails as a hexosamine because it is a poor donor and/or a poor acceptor. Preparative syntheses employing PmHS2 catalyst with no acceptor gave average polymer yields of ∼60% or ∼22% for authentic heparosan versus unnatural GlcNAcUA-containing heparosan, respectively. Higher concentrations of UDP-sugar precursor helped compensate for the slower incorporation rates of some unnatural analogs. The relaxed specificity of PmHS2 could be because of different active site geometry or different surrounding residues other than PmHS1, which facilitates the favorable binding interactions and/or avoids certain hindrances (e.g. steric, electrostatic) with the analogs. Overall, these results help to elucidate the nature of the synthase active site without an experimentally determined three-dimensional enzyme structure. From previous work (7DeAngelis P.L. White C.L. J. Biol. Chem. 2002; 277: 7209-7213Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar, 8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar) and here with purified PmHS1 and PmHS2, the isomeric state of the C-4 hydroxyl of the UDP-sugar precursors appears to be critical for these synthases because the C-4 epimers of the authentic substrates, UDP-Gal-NAc and UDP-GalUA, are not functional analogs in the polymerization assay (not shown). This observed stringency is probably because of the importance of the hydroxyls forming the glycosidic linkages of the heparosan chain (-GlcUA-β1,4-Glc-NAc-α1,4-) residing in the correct orientation for catalytic residues to couple the saccharide units. The evolutionary history of PmHS1 and PmHS2 is not yet known. Two opposing hypotheses are possible: i) the “traditional” scenario, where a gene encoding a substrate selective PmHS1 progenitor is duplicated, and the resulting PmHS2 ancestor, unfettered from its normal duty of making heparosan, becomes less specific for a potential hitherto unknown function; or ii) a more recently recognized scenario (22Jensen R.A. Annu. Rev. Microbiol. 1976; 30: 409-425Crossref PubMed Scopus (823) Google Scholar), where a gene encoding a nonspecific PmHS2 progenitor is duplicated resulting in a PmHS1 ancestor that becomes more substrate-specific in order to make heparosan. As PmHS2 (but not PmHS1) occurs in many type A and type F strains (HA or chondroitin capsule producers, respectively) (8DeAngelis P.L. White C.L. J. Bacteriol. 2004; 186: 8529-8532Crossref PubMed Scopus (41) Google Scholar), model ii may be more likely. More DNA sequence information from other isolates and species may be required to establish whether PmHS1 or PmHS2 was the primordial enzyme. Pathogenic bacteria are under extreme selective pressure from host defenses, and thus the potential to alter capsule composition and maintain virulence is a valuable asset. The promiscuity of PmHS2 makes it a useful catalyst for preparing glycosaminoglycan polymer analogs with new biological or chemical properties. For example, unnatural polymers containing the GlcNAcUA monomer are not digested by heparin lyase III, an enzyme known to digest most other heparinoids (Fig. 5). Depending on the substitutions, similar heparinoids may have a slower turnover rate potentially making it a longer acting therapeutic. These new polymers should also prove to be very useful in the pursuit of understanding the structure/function relationships of the polymer and the interaction of heparinoids with various binding proteins including receptors, growth factors, and coagulation factors. We thank Leonard C. Oatman and Carissa L. White for general laboratory support; Bruce Baggenstoss for SEC-MALLS data analysis; Dr. F. Michael Haller (Hyalose, LLC) for predicted structure and mass determination; Dr. Peng George Wang (Ohio State University) for the GalU and PPA plasmids; Dr. Jian Liu (University of North Carolina) who provided the initial heparosan oligosaccharides and the heparin lyase III; and Dr. Joe Lam (Guelph University) for assistance with UDP-sugars." @default.
• W2019487637 created "2016-06-24" @default.
• W2019487637 creator A5044520673 @default.
• W2019487637 creator A5046076280 @default.
• W2019487637 creator A5048398380 @default.
• W2019487637 creator A5049760871 @default.
• W2019487637 creator A5051354705 @default.
• W2019487637 creator A5083808846 @default.
• W2019487637 date "2007-09-01" @default.
• W2019487637 modified "2023-10-06" @default.
• W2019487637 title "Chemoenzymatic Synthesis with Distinct Pasteurella Heparosan Synthases" @default.
• W2019487637 cites W1969109690 @default.
• W2019487637 cites W1970976508 @default.
• W2019487637 cites W1982423863 @default.
• W2019487637 cites W1983607795 @default.
• W2019487637 cites W1990306679 @default.
• W2019487637 cites W1999652145 @default.
• W2019487637 cites W2012247703 @default.
• W2019487637 cites W2016156311 @default.
• W2019487637 cites W2017606019 @default.
• W2019487637 cites W2030593951 @default.
• W2019487637 cites W2043988634 @default.
• W2019487637 cites W2052052749 @default.
• W2019487637 cites W2052578550 @default.
• W2019487637 cites W2087788631 @default.
• W2019487637 cites W2087885447 @default.
• W2019487637 cites W2103731920 @default.
• W2019487637 cites W2120903363 @default.
• W2019487637 cites W2123429740 @default.
• W2019487637 cites W2149944903 @default.
• W2019487637 cites W2169829977 @default.
• W2019487637 cites W3118986828 @default.
• W2019487637 cites W4238813159 @default.
• W2019487637 doi "https://doi.org/10.1074/jbc.m701599200" @default.
• W2019487637 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/17627940" @default.
• W2019487637 hasPublicationYear "2007" @default.
• W2019487637 type Work @default.
• W2019487637 sameAs 2019487637 @default.
• W2019487637 citedByCount "82" @default.
• W2019487637 countsByYear W20194876372012 @default.
• W2019487637 countsByYear W20194876372013 @default.
• W2019487637 countsByYear W20194876372014 @default.
• W2019487637 countsByYear W20194876372015 @default.
• W2019487637 countsByYear W20194876372016 @default.
• W2019487637 countsByYear W20194876372017 @default.
• W2019487637 countsByYear W20194876372019 @default.
• W2019487637 countsByYear W20194876372020 @default.
• W2019487637 countsByYear W20194876372021 @default.
• W2019487637 countsByYear W20194876372022 @default.
• W2019487637 countsByYear W20194876372023 @default.
• W2019487637 crossrefType "journal-article" @default.
• W2019487637 hasAuthorship W2019487637A5044520673 @default.
• W2019487637 hasAuthorship W2019487637A5046076280 @default.
• W2019487637 hasAuthorship W2019487637A5048398380 @default.
• W2019487637 hasAuthorship W2019487637A5049760871 @default.
• W2019487637 hasAuthorship W2019487637A5051354705 @default.
• W2019487637 hasAuthorship W2019487637A5083808846 @default.
• W2019487637 hasBestOaLocation W20194876371 @default.
• W2019487637 hasConcept C185592680 @default.
• W2019487637 hasConcept C55493867 @default.
• W2019487637 hasConcept C86803240 @default.
• W2019487637 hasConcept C89423630 @default.
• W2019487637 hasConceptScore W2019487637C185592680 @default.
• W2019487637 hasConceptScore W2019487637C55493867 @default.
• W2019487637 hasConceptScore W2019487637C86803240 @default.
• W2019487637 hasConceptScore W2019487637C89423630 @default.
• W2019487637 hasIssue "39" @default.
• W2019487637 hasLocation W20194876371 @default.
• W2019487637 hasOpenAccess W2019487637 @default.
• W2019487637 hasPrimaryLocation W20194876371 @default.
• W2019487637 hasRelatedWork W1531601525 @default.
• W2019487637 hasRelatedWork W1990781990 @default.
• W2019487637 hasRelatedWork W2319480705 @default.
• W2019487637 hasRelatedWork W2384464875 @default.
• W2019487637 hasRelatedWork W2606230654 @default.
• W2019487637 hasRelatedWork W2607424097 @default.
• W2019487637 hasRelatedWork W2748952813 @default.
• W2019487637 hasRelatedWork W2899084033 @default.
• W2019487637 hasRelatedWork W2948807893 @default.
• W2019487637 hasRelatedWork W2778153218 @default.
• W2019487637 hasVolume "282" @default.
• W2019487637 isParatext "false" @default.
• W2019487637 isRetracted "false" @default.
• W2019487637 magId "2019487637" @default.
• W2019487637 workType "article" @default.