No description available.

Leukocyte type core 2 {beta}1,6-N-acetylglucosaminyltransferase (C2GnT-L) is a key enzyme in the biosynthesis of branched O-glycans. It is an inverting, metal ion-independent glycosyltransferase that catalyzes the formation of the core 2 O-glycan (Gal{beta}1,3[GlcNAc{beta}1,6]GalNAc-O-Ser/Thr) from its donor and acceptor substrates, UDP-GlcNAc and the core 1 O-glycan (Gal{beta}1,3GalNAc-O-Ser/Thr), respectively. The primary objective of the work described in this thesis is to shed light on the structure, catalytic mechanism and substrate specificity of C2GnT-L. Since glycosyltransferases are membrane bound glycoproteins that possess disulphide bonds, the first challenge was to produce sufficient quantities of C2GnT-L for biochemical and structural characterization. To this end, C2GnT-L and various active site mutants were expressed and purified from stably transformed mammalian cell lines. The x-ray crystal structure of wild-type C2GnT-L was solved, in both its apo and acceptor substrate complexed forms. The structures, along with revealing the structural basis for acceptor substrate specificity, showed that C2GnT-L belongs to the GT-A glycosyltransferase fold type. This was a surprising result given that, to this day, C2GnT-L is the only GT-A glycosyltransferase that does not possess a DXD motif and does not require a divalent metal ion for catalysis. The mechanism of the metal ion-independent C2GnT-L activity, within the context of the metal ion-dependent GT-A fold, was probed further using site directed mutagenesis in conjuncition with x-ray crystallography, enzyme assays, and frontal affinity chromatography. It was found that positively charged side chains in C2GnT-L functionally replace the divalent metal ion found in other GT-A glycosyltransferases, providing evidence for a convergence of metal ion-independent activity between GT-A and GT-B glycosyltransferase fold types.

Leukocyte type core 2 {beta}1,6-N-acetylglucosaminyltransferase (C2GnT-L) is a key enzyme in the biosynthesis of branched O-glycans. It is an inverting, metal ion-independent glycosyltransferase that catalyzes the formation of the core 2 O-glycan (Gal{beta}1,3[GlcNAc{beta}1,6]GalNAc-O-Ser/Thr) from its donor and acceptor substrates, UDP-GlcNAc and the core 1 O-glycan (Gal{beta}1,3GalNAc-O-Ser/Thr), respectively. The primary objective of the work described in this thesis is to shed light on the structure, catalytic mechanism and substrate specificity of C2GnT-L. Since glycosyltransferases are membrane bound glycoproteins that possess disulphide bonds, the first challenge was to produce sufficient quantities of C2GnT-L for biochemical and structural characterization. To this end, C2GnT-L and various active site mutants were expressed and purified from stably transformed mammalian cell lines. The x-ray crystal structure of wild-type C2GnT-L was solved, in both its apo and acceptor substrate complexed forms. The structures, along with revealing the structural basis for acceptor substrate specificity, showed that C2GnT-L belongs to the GT-A glycosyltransferase fold type. This was a surprising result given that, to this day, C2GnT-L is the only GT-A glycosyltransferase that does not possess a DXD motif and does not require a divalent metal ion for catalysis. The mechanism of the metal ion-independent C2GnT-L activity, within the context of the metal ion-dependent GT-A fold, was probed further using site directed mutagenesis in conjuncition with x-ray crystallography, enzyme assays, and frontal affinity chromatography. It was found that positively charged side chains in C2GnT-L functionally replace the divalent metal ion found in other GT-A glycosyltransferases, providing evidence for a convergence of metal ion-independent activity between GT-A and GT-B glycosyltransferase fold types.

Thesis (Ph. D.)--Hong Kong University of Science and Technology, 2002. / Includes bibliographical references (leaves 196-212). Also available in electronic version. Access restricted to campus users.

Cyanobacteria of the Nostoc genus are known for the thick, mucilaginous carbohydrate coatings that they produce. In this work, two examples of cyanobacterial glycobiology are considered, each of which involves a cyanobacterium of the Nostoc genus. The first portion of this work details attempts to obtain amino acid sequence information from the enzymes (glycosyltransferases) that are responsible for producing the extracellular polysaccharide (EPS) of Nostoc commune DRH1, ultimately to allow the transfer of this capacity to another organism. Two artificial substrates were synthesized for use in a capillary electrophoresis-based enzyme assay, which was used to look for glycosyltransferase activity in Nostoc commune DRH1 cell extracts. Glucuronosyltransferase activity was detected in association with Nostoc commune membrane material. The active enzyme displayed a divalent cation metal dependence (Mg+2) that is typical of glycosyltransferase enzymes purified from other organisms. Because the enzyme responsible for this activity held the potential to be EPS-related, its purification was attempted. The capillary electrophoresis-based enzyme assay and a 32P-labeled affinity tag were utilized to follow the glucuronosyltransferase enzyme through successive purification steps. The active enzyme was extracted from Nostoc commune membrane material using Triton X-100, and then purified by anion exchange chromatography. The active detergent extract was extremely unstable, and consequently, other purification techniques tested were unsuccessful in enriching activity. Affinity-labeling experiments indicated that the active enzyme was forming protein aggregates during these procedures, which were not amenable to in-gel protease digestion and peptide analysis by tandem mass spectrometry. The second portion of this work describes an investigation of an Anabaena (Nostoc) PCC 7120 soluble cell extract. Upon separation by sodium dodecyl sulfate ¡V polyacrylamide gel electrophoresis (SDS-PAGE) and subsequent periodic acid-Schiff (PAS) staining of the resulting gel, the components of this cellular fraction produce a ladder-like pattern, which suggests that the extract may contain glycosylated protein. Analyses of several samples that were taken from within the PAS-staining region of such a gel revealed surface layer homology (SLH) domain-containing proteins, likely candidates to be covalently attached to or non-covalently interacting with carbohydrate. Various protein sequence analyses indicated that the detected SLH domain containing proteins belong to a family of (putative) cyanobacterial porins. Proteins in this family possess features that include a N-terminal signal sequence, a single SLH domain motif, followed by a coiled-coil region, and a C-terminal region that is homologous to the b-barrel-forming region of bacterial porins. All of these features were identified in the detected Anabaena (Nostoc) PCC 7120 SLH domain-containing proteins. Smith degradation was performed on a sample that was electroeluted from the PAS-staining region of a preparative-scale SDS-PAGE gel of the soluble cell extract. Subsequent analyses of the resulting sample by SDS-PAGE and mass spectrometry indicated that at least two SLH domain-containing proteins, encoded by all4499 and alr4550, were non-covalently interacting with the PAS-staining material. Following degradation, the PAS-staining material was still of sufficient size to detected by gel electrophoresis, and it continued to migrate in the absence of an interacting protein component. Protease digestion of a similarly prepared sample, and then subsequent analysis by SDS-PAGE and mass spectrometry, revealed that the region between amino acid residues #504 and #536, in the protein encoded by the alr4550 open reading frame, was interacting with the PAS-staining material. Monosaccharide composition analyses of this material revealed more carbohydrate constituents than are found in cyanobacterial primary (peptidoglycan) cell wall polymer alone, indicating that it contained a significant secondary cell wall polymer component as well. / Ph. D.

The isoflavones in kudzu roots, especially the C-glycosylated isoflavone puerarin, have been linked to many health benefits. Puerarin contains a carbon-carbon glycosidic bond that can withstand hydrolysis. The C-glycosylation reaction in the biosynthesis of puerarin has not been thoroughly investigated, with conflicting reports suggesting that it could take place on daidzein, isoliquiritigenin, or 2,7,4ʹ-trihydroxyisoflavanone. Kudzu species were identified for use in comparative transcriptomics. A non-puerarin producing kudzu was identified as Pueraria phaseoloides and a puerarin producing kudzu was identified as Pueraria montana lobata. Through the use of the plant secondary product glycosyltransferase (PSPG) motif, glycosyltransferases (UGTs) were identified from the transcriptomes. The UGTs that had higher digital expression in P. m. lobata were examined further using additional tools to home in on the UGT that could be responsible for puerarin biosynthesis. One of the UGTs identified, UGT71T5, had previously been characterized from kudzu as a C-glycosyltransferase involved in puerarin biosynthesis through in vitro enzyme activity (with daidzein) and a gain of function approach in soybean hairy roots. Previous studies have not supported the end-product of a pathway such as daidzein as the target for C-glycosylation, and no genetic analysis of UGT function had been conducted in kudzu. The activity of recombinant UGT71T5 with daidzein was confirmed in the present work. Following the development of a kudzu hairy root system, UGT71T5 expression was then knocked down by RNA interference (RNAi). When compared to control hairy roots there was a large reduction in puerarin content in the UGT71T5-RNAi roots, confirming the role of this enzyme in puerarin biosynthesis. Isotopic labeling of kudzu plants revealed that labeled daidzein could be directly incorporated into puerarin; however, the percent incorporation of daidzein was substantially lower than that of L-phenylalanine, a compound at the start of the pathway to isoflavone synthesis. The knockdown of 2-hydroxisoflavanone synthase (2-HIS) in kudzu hairy roots blocked formation of puerarin and daidzin (7-O-glycosyldaidzein), and was accompanied by accumulation of C-glycosylated isoliquiritigenin and C-glycosylated liquiritigenin. These compounds were found in low amounts in control hairy roots, but were virtually absent in UGT71T5 knockdown hairy roots. The knockdown of 2-hydroxyisoflavanone dehydratase (2-HID) in kudzu hairy roots resulted in a slight reduction in puerarin but no change to daidzin levels, suggesting that C-glycosylation might stabilize the substrate for 2-HID which can otherwise spontaneously dehydrate. Taken together these results reveal that UGT71T5 is likely the major C-glycosyltransferase involved in puerarin biosynthesis in kudzu. They also provide evidence for an alternative pathway to puerarin biosynthesis through the C-glycosylation of isoliquiritigenin or its immediate precursor. In one pathway, UGT71T5 acts as an operationally soluble enzyme that can directly C-glycosylate daidzein, and in the other pathway UGT71T5 acts as part of a metabolic channel for conversion of a C-glycosylated earlier precursor to puerarin. Other UGT enzymes identified in this work did not show C-glycosyltransferase activity; however, three enzymes showed activity in vitro that could be useful for introducing novel regiospecificity in biochemical synthesis of flavonoid glycosides.

MshA is a glycosyltransferase that synthesizes the precursor of mycothiol, a low-molecular-weight thiol found exclusively in Actinomycetes, including the virulent pathogen Mycobacterium tuberculosis (Mtb). The structure of MshA from Mtb (herein coined as TbMshA) and its complex with uridine diphosphate N-acetyl-glucosamine (UDP-GlcNAc) have been solved to resolutions of 2.32 A and 2.89 A respectively. Both structures form two monomers in the asymmetric unit cell and exhibit typical beta/alpha/beta Rossmann folds. Upon binding of UDP-GlcNAc, the C-terminal domain of TbMshA undergoes conformational changes in order to interact with UDP-GlcNAc at the binding site. In addition, ligand-bound TbMshA structure enables the identification of critical residues for enzymatic interactions, especially the residue Glu-353 (E353) at the active site that is believed to serve as a nucleophile in the sugar transfer of TbMshA. In order to verify this, a mutant of TbMshA with a single amino acid mutation from glutamate to glutamine at residue 353 is generated. The mutant (E353Q) has shown reduced enzyme activity by more than four-fold compared to the wild-type TbMshA (Vmax for wild-type is 0.17 plus/minus 0.02 microM sec^-1, whereas Vmax for E353Q is 0.04 plus/minus 0.01 microM sec-1). The kcat/Km for wild-type TbMshA (3.5 plus/minus 1.1 * 10^3 M^-1 sec^-1) is an order of magnitude higher than that of the mutant (0.3 plus/minus 0.1 * 10^3 M^-1 sec^-1), indicating the catalytic efficiency is greatly suppressed by the mutation. Mass spectrometry data also reveals that E353Q is unable to form the product of the reaction catalyzed by the wild-type TbMshA. These findings suggest the important role of Glu-353 in the structure and activity of TbMshA.

No description available.

No description available.

Peptidoglycan glycosyltransferases (PGTs) are highly conserved bacterial enzymes that catalyze the polymerization of the lipidic disaccharide, Lipid II, to form individual peptidoglycan (PG) strands which are subsequently cross-linked to form mature PG, the major skeletal component of the bacterial cell wall. Recent advances in the preparation of well-defined PGT substrates have enabled the biochemical characterization of Lipid II polymerization by the PGTs. In the course of these studies, we have observed that a distinctive lag phase in the initial rate of PG synthesis by the PGTs can be abrogated if the enzyme is preincubated with Lipid IV, the shortest PG fragment. The origins of this lag phase are intriguing because the chemical transformation involved in coupling Lipid II to yield Lipid IV is identical to the transformation involved in the synthesis of longer PG fragments from Lipid II. Crystallographic structures of the PGTs with Moenomycin A, an inhibitor that is believed to bind to the same site as Lipid IV, suggest that the PGTs possess flexible regions near the putative active site that can undergo substrate-induced conformational changes to accelerate PG synthesis. However, there is currently no structural evidence on how the PGTs interact with its substrates. The work in this thesis lays the foundation for pursuing a solution structure of a Lipid IV bound PGT complex by Nuclear Magnetic Resonance (NMR) spectroscopy, enabling the study of important enzyme conformational states and structural dynamics involved in PG synthesis. Specifically, Chapter 2 of this thesis presents the biochemical evidence that the preincubation of the PGTs with a Lipid IV derivative, Gal-Lipid IV abrogates the lag phase and accelerates the initial rate of PG synthesis. Chapter 3 presents a robust methodology for obtaining multimilligram quantities of isotope labeled, monodisperse and monomeric SgtB, a PGT from a clinically relevant pathogen, Staphylococcus aureus for solution structural studies. Chapter 4 describes the systematic development of a methodology for producing a well-behaved, stable sample of Moenomycin A bound SgtB for NMR spectroscopy. Chapter 5 delineates the adaptation of the methodology described in Chapter 4 for pursuing the solution structure of Lipid IV bound SgtB. / Chemistry and Chemical Biology