Tetrameric Structure of the GlfT2 Galactofuranosyltransferase Reveals a Scaffold for the Assembly of Mycobacterial Arabinogalactan

Biosynthesis of the mycobacterial cell wall relies on the activities of many enzymes, including several glycosyltransferases (GTs). The polymerizing galactofuranosyltransferase GlfT2 (Rv3808c) synthesizes the bulk of the galactan portion of the mycolyl-arabinogalactan complex, which is the largest component of the mycobacterial cell wall. We used x-ray crystallography to determine the 2.45-Å resolution crystal structure of GlfT2, revealing an unprecedented multidomain structure in which an N-terminal β-barrel domain and two primarily α-helical C-terminal domains flank a central GT-A domain. The kidney-shaped protomers assemble into a C₄-symmetric homotetramer with an open central core and a surface containing exposed hydrophobic and positively charged residues likely involved with membrane binding. The structure of a 3.1-Å resolution complex of GlfT2 with UDP reveals a distinctive mode of nucleotide recognition. In addition, models for the binding of UDP-galactofuranose and acceptor substrates in combination with site-directed mutagenesis and kinetic studies suggest a mechanism that explains the unique ability of GlfT2 to generate alternating β-(1→5) and β-(1→6) glycosidic linkages using a single active site. The topology imposed by docking a tetrameric assembly onto a membrane bilayer also provides novel insights into aspects of processivity and chain length regulation in this and possibly other polymerizing GTs.     

STD-NMR studies of two acceptor substrates of GlfT2, a galactofuranosyltransferase from Mycobacterium tuberculosis: Epitope mapping studies

The major structural component of the mycobacterial cell wall, the mycolyl–arabinogalactan–peptidoglycan complex, possesses a galactan core composed of approximately 30 galactofuranosyl (Galf) resides attached via alternating β-(1→6) and β-(1→5) linkages. Recent studies have shown that the entire galactan is synthesized by two bifunctional galactofuranosyltransferases, GlfT1 and GlfT2. We report here saturation transfer difference (STD) NMR studies GlfT2 using two trisaccharide acceptor substrates, β-d-Galf-(1→6)-β-d-Galf-(1→5)-β-d-Galf-O(CH₂)₇CH₃ (2) and β-d-Galf-(1→5)-β-d-Galf-(1→6)-β-d-Galf-O(CH₂)₇CH₃ (3), as well as the donor substrate for the enzyme, UDP-Galf. Epitope mapping demonstrated a greater enhancement toward the ‘reducing’ ends of both trisaccharides, and that UDP-galactofuranose (UDP-Galf) made more intimate contacts through its nucleotide moiety. This observation is consistent with the greater flexibility required within the active site of the reaction between the growing polymer acceptor and the UDP-Galf donor. The addition of UDP-Galf to either 2 or 3 in the presence of GlfT2 generated a tetrasaccharide product, indicating that the enzyme was catalytically active.     

Bifunctional glycosyltransferases catalyze both extension and termination of pectic galactan oligosaccharides

Pectins are the most complex polysaccharides of the plant cell wall. Based on the number of methylations, acetylations and glycosidic linkages present in their structures, it is estimated that up to 67 transferase activities are involved in pectin biosynthesis. Pectic galactans constitute a major part of pectin in the form of side‐chains of rhamnogalacturonan‐I. In Arabidopsis, galactan synthase 1 (GALS1) catalyzes the addition of galactose units from UDP‐Gal to growing β‐1,4‐galactan chains. However, the mechanisms for obtaining varying degrees of polymerization remain poorly understood. In this study, we show that AtGALS1 is bifunctional, catalyzing both the transfer of galactose from UDP‐α‐d ‐Gal and the transfer of an arabinopyranose from UDP‐β‐l ‐Ara_p to galactan chains. The two substrates share a similar structure, but UDP‐α‐d ‐Gal is the preferred substrate, with a 10‐fold higher affinity. Transfer of Ara_p to galactan prevents further addition of galactose residues, resulting in a lower degree of polymerization. We show that this dual activity occurs both in vitro and in vivo. The herein described bifunctionality of AtGALS1 may suggest that plants can produce the incredible structural diversity of polysaccharides without a dedicated glycosyltransferase for each glycosidic linkage.     

Galactosyl transferases in mycobacterial cell wall synthesis

Two galactosyl transferases can apparently account for the full biosynthesis of the cell wall galactan of mycobacteria. Evidence is presented based on enzymatic incubations with purified natural and synthetic galactofuranose (Galf) acceptors that the recombinant galactofuranosyl transferase, GlfT1, from Mycobacterium smegmatis, the Mycobacterium tuberculosis Rv3782 ortholog known to be involved in the initial steps of galactan formation, harbors dual β-(1→4) and β-(1→5) Galf transferase activities and that the product of the enzyme, decaprenyl-P-P-GlcNAc-Rha-Galf-Galf, serves as a direct substrate for full polymerization catalyzed by another bifunctional Galf transferase, GlfT2, the Rv3808c enzyme.     

Crystal structure of the bifunctional ATP sulfurylase-APS kinase from the chemolithotrophic thermophile Aquifex aeolicus

The thermophilic chemolithotroph, Aquifex aeolicus, expresses a gene product that exhibits both ATP sulfurylase and adenosine-5'-phosphosulfate (APS) kinase activities. These enzymes are usually segregated on two separate proteins in most bacteria, fungi, and plants. The domain arrangement in the Aquifex enzyme is reminiscent of the fungal ATP sulfurylase, which contains a C-terminal domain that is homologous to APS kinase yet displays no kinase activity. Rather, in the fungal enzyme, the motif serves as a sulfurylase regulatory domain that binds the allosteric effector 3'-phosphoadenosine-5'-phosphosulfate (PAPS), the product of true APS kinase. Therefore, the Aquifex enzyme may represent an ancestral homolog of a primitive bifunctional enzyme, from which the fungal ATP sulfurylase may have evolved. In heterotrophic sulfur-assimilating organisms such as fungi, ATP sulfurylase catalyzes the first committed step in sulfate assimilation to produce APS, which is subsequently metabolized to generate all sulfur-containing biomolecules. In contrast, ATP sulfurylase in sulfur chemolithotrophs catalyzes the reverse reaction to produce ATP and sulfate from APS and pyrophosphate. Here, the 2.3 A resolution X-ray crystal structure of Aquifex ATP sulfurylase-APS kinase bifunctional enzyme is presented. The protein dimerizes through its APS kinase domain and contains ADP bound in all four active sites. Comparison of the Aquifex ATP sulfurylase active site with those from sulfate assimilators reveals similar dispositions of the bound nucleotide and nearby residues. This suggests that minor perturbations are responsible for optimizing the kinetic properties for the physiologically relevant direction. The APS kinase active-site lid adopts two distinct conformations, where one conformation is distorted by crystal contacts. Additionally, a disulfide bond is observed in one ATP-binding P-loop of the APS kinase active site. This linkage accounts for the low kinase activity of the enzyme under oxidizing conditions. The thermal stability of the Aquifex enzyme can be explained by the 43% decreased cavity volume found within the protein core.     

Crystal structure of truncated Fibrobacter succinogenes 1,3-1,4-beta-D-glucanase in complex with beta-1,3-1,4-cellotriose

Fibrobacter succinogenes 1,3-1,4-beta-D-glucanase (Fsbeta-glucanase) catalyzes the specific hydrolysis of beta-1,4 glycosidic bonds adjacent to beta-1,3 linkages in beta-D-glucans or lichenan. This is the first report to elucidate the crystal structure of a truncated Fsbeta-glucanase (TFsbeta-glucanase) in complex with beta-1,3-1,4-cellotriose, a major product of the enzyme reaction. The crystal structures, at a resolution of 2.3 angstroms, reveal that the overall fold of TFsbeta-glucanase remains virtually unchanged upon sugar binding. The enzyme accommodates five glucose residues, forming a concave active cleft. The beta-1,3-1,4-cellotriose with subsites -3 to -1 bound to the active cleft of TFsbeta-glucanase with its reducing end subsite -1 close to the key catalytic residues Glu56 and Glu60. All three subsites of the beta-1,3-1,4-cellotriose adopted a relaxed C(1)4 conformation, with a beta-1,3 glycosidic linkage between subsites -2 and -1, and a beta-1,4 glycosidic linkage between subsites -3 and -2. On the basis of the enzyme-product complex structure observed in this study, a catalytic mechanism and substrate binding conformation of the active site of TFsbeta-glucanase is proposed.     

Synthesis of galactofuranose-based acceptor substrates for the study of the carbohydrate polymerase GlfT2

Despite the prevalence and importance of carbohydrate polymers, the molecular details of their biosynthesis remain elusive. Many enzymes responsible for the synthesis of carbohydrate polymers require a ‘primer’ or ‘initiator’ carbohydrate sequence. One example of such an enzyme is the mycobacterial galactofuranosyltransferase GlfT2 (Rv3808c), which generates an essential cell wall building block. We recently demonstrated that recombinant GlfT2 is capable of producing a polymer composed of alternating β-(1,5) and β-(1,6)-linked galactofuranose (Galf) residues. Intriguingly, the length of the polymers produced from a synthetic glycosyl acceptor is consistent with those found in the cell wall. To probe the mechanism by which polymer length is controlled, a collection of initiator substrates has been assembled. The central feature of the synthetic route is a ruthenium-catalyzed cross-metathesis as the penultimate transformation. Access to synthetic substrates has led us to postulate a new mechanism for length control in this template-independent polymerization. Moreover, our investigations indicate that lipids possessing but a single galactofuranose residue can act as substrates for GlfT2.     

Development of a coupled spectrophotometric assay for GlfT2, a bifunctional mycobacterial galactofuranosyltransferase

As a key constituent of their protective cell wall all mycobacteria produce a large structural component, the mycolyl-arabinogalactan (mAG) complex, which has at its core a galactan moiety of alternating beta-(1-->5) and beta-(1-->6) galactofuranosyl residues. Galactan biosynthesis is essential for mycobacterial viability and thus inhibitors of the enzymes involved in its assembly are potential drugs for the treatment of mycobacterial diseases, including tuberculosis. Only two galactofuranosyltransferases, GlfT1 and GlfT2, are responsible for the biosynthesis of the entire galactan domain of the mAG and we report here the first high-throughput assay for GlfT2. Successful implementation of the assay required the synthesis of multi-milligram amounts of the donor for the enzyme, UDP-Galf, 1, which was achieved using a chemoenzymatic approach. We also describe an improved expression system for GlfT2, which provides a larger amount of active protein for the assay. Kinetic analysis of 1 and a known trisaccharide acceptor for the enzyme, 2, have been carried out and the apparent K(m) and k(cat) values obtained for the latter are in agreement with those obtained using a previously reported radiochemical assay. The assay has been implemented in 384-well microtiter plates, which will facilitate the screening of large numbers of potential GlfT2 inhibitors, with possible utility as novel anti-TB drugs.     

Identification and characterization of large galactosyltransferase gene families: galactosyltransferases for all functions

Enzymatic glycosylation of proteins and lipids is an abundant and important biological process. A great diversity of oligosaccharide structures and types of glycoconjugates is found in nature, and these are synthesized by a large number of glycosyltransferases. Glycosyltransferases have high donor and acceptor substrate specificities and are in general limited to catalysis of one unique glycosidic linkage. Emerging evidence indicates that formation of many glycosidic linkages is covered by large homologous glycosyltransferase gene families, and that the existence of multiple enzyme isoforms provides a degree of redundancy as well as a higher level of regulation of the glycoforms synthesized. Here, we discuss recent cloning strategies enabling the identification of these large glycosyltransferase gene families and exemplify the implication this has for our understanding of regulation of glycosylation by discussing two galactosyltransferase gene families.     

Reduction precedes cytidylyl transfer without substrate channeling in distinct active sites of the bifunctional CDP-ribitol synthase from Haemophilus influenzae

CDP-ribitol synthase is a bifunctional reductase and cytidylyltransferase that catalyzes the transformation of D-ribulose 5-phosphate, NADPH, and CTP to CDP-ribitol, a repeating unit present in the virulence-associated polysaccharide capsules of Haemophilus influenzae types a and b [Follens, A., et al. (1999) J. Bacteriol. 181, 2001]. In the work described here, we investigated the order of the reactions catalyzed by CDP-ribitol synthase and conducted experiments to resolve the question of substrate channeling in this bifunctional enzyme. It was determined that the synthase first catalyzed the reduction of D-ribulose 5-phosphate followed by cytidylyl transfer to D-ribitol 5-phosphate. Steady state kinetic measurements revealed a 650-fold kinetic preference for cytidylyl transfer to D-ribitol 5-phosphate over D-ribulose 5-phosphate. Rapid mixing studies indicated quick reduction of D-ribulose 5-phosphate with a lag in the cytidylyl transfer reaction, consistent with a requirement for the accumulation of K(m) quantities of D-ribitol 5-phosphate. Signature motifs in the C-terminal and N-terminal sequences of the enzyme (short chain dehydrogenase/reductase and nucleotidyltransferase motifs, respectively) were targeted with site-directed mutagenesis to generate variants that were impaired for only one of the two activities (K386A and R18A impaired for reduction and cytidylyl transfer, respectively). Release and free diffusion of the metabolic intermediate D-ribitol 5-phosphate was indicated by the finding that equimolar mixtures of K386A and R18A variants were efficient for bifunctional catalysis. Taken together, these findings suggest that bifunctional turnover occurs in distinct active sites of CDP-ribitol synthase with reduction of D-ribulose 5-phosphate and release and free diffusion of the metabolic intermediate D-ribitol 5-phosphate followed by cytidylyl transfer.     

Insect chitin synthases: a review

Chitin is the most widespread amino polysaccharide in nature. The annual global amount of chitin is believed to be only one order of magnitude less than that of cellulose. It is a linear polymer composed of N-acetylglucosamines that are joined in a reaction catalyzed by the membrane-integral enzyme chitin synthase, a member of the family 2 of glycosyltransferases. The polymerization requires UDP–N-acetylglucosamines as a substrate and divalent cations as co-factors. Chitin formation can be divided into three distinct steps. In the first step, the enzymes‘ catalytic domain facing the cytoplasmic site forms the polymer. The second step involves the translocation of the nascent polymer across the membrane and its release into the extracellular space. The third step completes the process as single polymers spontaneously assemble to form crystalline microfibrils. In subsequent reactions the microfibrils combine with other sugars, proteins, glycoproteins and proteoglycans to form fungal septa and cell walls as well as arthropod cuticles and peritrophic matrices, notably in crustaceans and insects. In spite of the good effort by a hardy few, our present knowledge of the structure, topology and catalytic mechanism of chitin synthases is rather limited. Gaps remain in understanding chitin synthase biosynthesis, enzyme trafficking, regulation of enzyme activity, translocation of chitin chains across cell membranes, fibrillogenesis and the interaction of microfibrils with other components of the extracellular matrix. However, cumulating genomic data on chitin synthase genes and new experimental approaches allow increasingly clearer views of chitin synthase function and its regulation, and consequently chitin biosynthesis. In the present review, I will summarize recent advances in elucidating the structure, regulation and function of insect chitin synthases as they relate to what is known about fungal chitin synthases and other glycosyltransferases.     

Ubiquitin-Like Protein SAMP1 and JAMM/MPN+ Metalloprotease HvJAMM1 Constitute a System for Reversible Regulation of Metabolic Enzyme Activity in Archaea

Ubiquitin/ubiquitin-like (Ub/Ubl) proteins are involved in diverse cellular processes by their covalent linkage to protein substrates. Here, we provide evidence for a post-translational modification system that regulates enzyme activity which is composed of an archaeal Ubl protein (SAMP1) and a JAMM/MPN+ metalloprotease (HvJAMM1). Molybdopterin (MPT) synthase activity was found to be inhibited by covalent linkage of SAMP1 to the large subunit (MoaE) of MPT synthase. HvJAMM1 was shown to cleave the covalently linked inactive form of SAMP1-MoaE to the free functional individual SAMP1 and MoaE subunits of MPT synthase, suggesting reactivation of MPT synthase by this metalloprotease. Overall, this study provides new insight into the broad idea that Ub/Ubl modification is a post-translational process that can directly and reversibly regulate the activity of metabolic enzymes. In particular, we show that Ub/Ubl linkages on the active site residues of an enzyme (MPT synthase) can inhibit its catalytic activity and that the enzyme can be reactivated through cleavage by a JAMM/MPN+ metalloprotease.     

Molecular dynamics simulations of a cyclic-beta-(1-->2) glucan containing an alpha-(1-->6) linkage as a 'molecular alleviator' for the macrocyclic conformational strain

The conformational preferences of a cyclic osmoregulated periplasmic glucan of Ralstonia solanacearum (OPGR), which is composed of 13 glucose units and linked entirely via beta-(1-->2) linkages excluding one alpha-(1-->6) linkage, were characterized by molecular dynamics simulations. Of the three force fields modified for carbohydrates that were applied to select a suitable one for the cyclic glucan, the carbohydrate solution force field (CSFF) was found to most accurately simulate the cyclic molecule. To determine the conformational characteristics of OPGR, we investigated the glycosidic dihedral angle distribution, fluctuation, and the potential energy of the glucan and constructed hypothetical cyclic (CYS13) and linear (LINEAR) glucans. All beta-(1-->2)-glycosidic linkages of OPGR adopted stable conformations, and the dihedral angles fluctuated in this energy region with some flexibility. However, despite the inherent flexibility of the alpha-(1-->6) linkage, the dihedral angles have no transition and are more rigid than that in a linear glucan. CYS13, which consists of only beta-(1-->2) linkages, is somewhat less flexible than other glycans, and one of its linkages adopts a higher energy conformation. In addition, the root-mean-square fluctuation of this linkage is lower than that of other linkages. Furthermore, the potential energy of glucans increases in the order of LINEAR, OPGR, and CYS13. These results provide evidence of the existence of conformational constraints in the cyclic glucan. The alpha-(1-->6)-glycosidic linkage can relieve this constraint more efficiently than the beta-(1-->2) linkage. The conformation of OPGR can reconcile the tendency for individual glycosidic bonds to adopt energetically favorable conformations with the requirement for closure of the macrocyclic ring by losing the inherent flexibility of the alpha-(1-->6)-glycosidic linkage.     

Molecular modeling of sialyloligosaccharide fragments into the active site of influenza virus N9 neuraminidase

Molecular modeling studies have been carried out to investigate the interactions between substrate sialyloligosaccharide (SOS) fragments bearing different glycosidic linkages and influenza virus N9 neuraminidase, a surface glycoprotein of influenza virus subtype N9. The studies revealed that the allowed orientation for sialic acid (SA) is less than 1% in the Eulerian space at the active site. The active site of this enzyme has enough space to accommodate various SOS fragments, NeuNAcalpha(2-3)Gal, NeuNAcalpha(2-6)Gal, NeuNAcalpha(2-8)NeuNAc and NeuNAcalpha(2-9)NeuNAc, but on specific conformations. In the bound conformation, among these substrates there exists a conformational similarity leading to a structural similarity, which may be an essential requirement for the cleavage activity of the neuraminidases irrespective of the type of glycosidic linkage.     

Structural basis of UDP-galactose binding by alpha-1,3-galactosyltransferase (alpha3GT): role of negative charge on aspartic acid 316 in structure and activity

alpha-1,3-Galactosyltransferase (alpha3GT) catalyzes the transfer of galactose from UDP-galactose to form an alpha 1-3 link with beta-linked galactosides; it is part of a family of homologous retaining glycosyltransferases that includes the histo-blood group A and B glycosyltransferases, Forssman glycolipid synthase, iGb3 synthase, and some uncharacterized prokaryotic glycosyltransferases. In mammals, the presence or absence of active forms of these enzymes results in antigenic differences between individuals and species that modulate the interplay between the immune system and pathogens. The catalytic mechanism of alpha3GT is controversial, but the structure of an enzyme complex with the donor substrate could illuminate both this and the basis of donor substrate specificity. We report here the structure of the complex of a low-activity mutant alpha3GT with UDP-galactose (UDP-gal) exhibiting a bent configuration stabilized by interactions of the galactose with multiple residues in the enzyme including those in a highly conserved region (His315 to Ser318). Analysis of the properties of mutants containing substitutions for these residues shows that catalytic activity is strongly affected by His315 and Asp316. The negative charge of Asp316 is crucial for catalytic activity, and structural studies of two mutants show that its interaction with Arg202 is needed for an active site structure that facilitates the binding of UDP-gal in a catalytically competent conformation.     

Identification and molecular cloning of a heparosan synthase from Pasteurella multocida type D.

Pasteurella multocida Type D, a causative agent of atrophic rhinitis in swine and pasteurellosis in other domestic animals, produces an extracellular polysaccharide capsule that is a putative virulence factor. It was reported previously that the capsule was removed by treating microbes with heparin lyase III. We molecularly cloned a 617-residue enzyme, pmHS, which is a heparosan (nonsulfated, unepimerized heparin) synthase. Recombinant Escherichia coli-derived pmHS catalyzes the polymerization of the monosaccharides from UDP-GlcNAc and UDP-GlcUA. Other structurally related sugar nucleotides did not substitute. Synthase activity was stimulated about 7-25-fold by the addition of an exogenous polymer acceptor. Molecules composed of approximately 500-3,000 sugar residues were produced in vitro. The polysaccharide was sensitive to the action of heparin lyase III but resistant to hyaluronan lyase. The sequence of the pmHS enzyme is not very similar to the vertebrate heparin/heparan sulfate glycosyltransferases, EXT1 and 2, or to other Pasteurella glycosaminoglycan synthases that produce hyaluronan or chondroitin. The pmHS enzyme is the first microbial dual-action glycosyltransferase to be described that forms a polysaccharide composed of beta4GlcUA-alpha4GlcNAc disaccharide repeats. In contrast, heparosan biosynthesis in E. coli K5 requires at least two separate polypeptides, KfiA and KfiC, to catalyze the same polymerization reaction.     

Tryptophan synthase, an allosteric molecular factory

Tryptophan synthase (TrpS) is a pyridoxal phosphate-containing bifunctional enzyme that catalyzes the last two steps in the biosynthesis of L-tryptophan. Indole, an intermediate generated at the active site of the alpha-subunit is channeled via a 25 A long tunnel to the beta-active site where it reacts with an aminoacrylate intermediate derived from L-serine. The two reactions are kept in phase by allosteric interactions between the two subunits. The recent development of novel alpha-site ligands and alpha-reaction transition state analogs combined with kinetic and crystal structure analysis of Salmonella typhimurium tryptophan synthase has provided new insights into the allosteric regulation of substrate channeling, the reaction mechanisms of the alpha and beta active sites, and the influence of structural dynamics.     

Bifunctional abietadiene synthase: mutual structural dependence of the active sites for protonation-initiated and ionization-initiated cyclizations

Abietadiene synthase from grand fir catalyzes two sequential, mechanistically distinct cyclizations, of geranylgeranyl diphosphate and of copalyl diphosphate, in the formation of a mixture of abietadiene isomers as the committed step of diterpenoid resin acid biosynthesis. Each reaction is independently conducted at a separate active site residing in what were considered to be structurally distinct domains typical of terpene cyclases. Despite the presence of an unusual 250-residue N-terminal insertional element, a tandem pair of charged residues distal to the insertion was shown to form a functional part of the C-terminal active site. Because abietadiene synthase resembles the ancestral plant terpene cyclase, this observation suggests an early evolutionary origin of catalytically important positively charged residues at the N-terminus of enzymes of this general class. A series of N- and C-terminal truncations of this enzyme were constructed and characterized, both alone and as mixtures of adjacent polypeptide pairs, to assess the proposed domain architecture, the function of the insertional element, and the role of presumptive interdomain contacts. These studies indicated a requirement for the insertional element in functional folding and allowed definition of the minimum primary structure of N- and C-terminal active site peptides. Most importantly, the results showed that, although the two active sites of abietadiene synthase are catalytically independent, substantial contact between the two regions is essential for the functional competence of this enzyme. Thus, the two cyclization sites of abietadiene synthase cannot be dissected into catalytically distinct domains, and, therefore, abietadiene synthase is unlikely to have arisen by fusion of two previously independent genes.     

The porphobilinogen synthase catalyzed reaction mechanism

Porphobilinogen synthase (PBGS) catalyzes the first common reaction in the biosynthesis of the tetrapyrroles, the asymmetric condensation of two molecules of delta-aminolevulinic acid to form porphobilinogen. There is a variable requirement for an essential active site zinc that necessitates consideration of PBGS as an enzyme that may exhibit phylogenetic diversity in its chemical reaction mechanism. Recent crystal structures suggest reaction mechanisms that involve two covalent Schiff base linkages between adjacent active site lysine residues and each of the two substrate molecules. The reaction appears to stall at a covalently bound almost-product intermediate that is poised for breakdown to product upon binding of a substrate molecule to an adjacent active site and a subsequent conformational change.