Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Functional Relationships of the Sialyltransferases Involved in Expression of the Polysialic Acid Capsules ofEscherichia coli K1 and K92 and Neisseria meningitidis Groups B or C

2003; Elsevier BV; Volume: 278; Issue: 17 Linguagem: Inglês

10.1074/jbc.m208837200

ISSN

1083-351X

Autores

Susan M. Steenbergen, Eric R. Vimr,

Tópico(s)

Bacterial Genetics and Biotechnology

Resumo

Polysialic acid (PSA) capsules are cell-associated homopolymers of α2,8-, α2,9-, or alternating α2,8/2,9-linked sialic acid residues that function as essential virulence factors in neuroinvasive diseases caused by certain strains of Escherichia coli and Neisseria meningitidis. PSA chains structurally identical to the bacterial α2,8-linked capsular polysaccharides are also synthesized by the mammalian central nervous system, where they regulate neuronal function in association with the neural cell adhesion molecule (NCAM). Despite the structural identity between bacterial and NCAM PSAs, the respective polysialyltransferases (polySTs) responsible for polymerizing sialyl residues from donor CMP-sialic acid are not homologous glycosyltransferases. To better define the mechanism of capsule biosynthesis, we established the functional interchangeability of bacterial polySTs by complementation of a polymerase-deficient E. coli K1 mutant with the polyST genes from groups B or C N. meningitidis and the control E. coli K92 polymerase gene. The biochemical and immunochemical results demonstrated that linkage specificity is dictated solely by the source of the polymerase structural gene. To determine the molecular basis for linkage specificity, we created chimeras of the K1 and K92 polySTs by overlap extension PCR. Exchanging the first 52 N-terminal amino acids of the K1 NeuS with the C terminus of the K92 homologue did not alter specificity of the resulting chimera, whereas exchanging the first 85 or reciprocally exchanging the first 100 residues did. These results demonstrated that linkage specificity is dependent on residues located between positions 53 and 85 from the N terminus. Site-directed mutagenesis of the K92 polyST N terminus indicated that no single residue alteration was sufficient to affect specificity, consistent with the proposed function of this domain in orienting the acceptor. The combined results provide the first evidence for residues critical to acceptor binding and elongation in polysialyltransferase. Polysialic acid (PSA) capsules are cell-associated homopolymers of α2,8-, α2,9-, or alternating α2,8/2,9-linked sialic acid residues that function as essential virulence factors in neuroinvasive diseases caused by certain strains of Escherichia coli and Neisseria meningitidis. PSA chains structurally identical to the bacterial α2,8-linked capsular polysaccharides are also synthesized by the mammalian central nervous system, where they regulate neuronal function in association with the neural cell adhesion molecule (NCAM). Despite the structural identity between bacterial and NCAM PSAs, the respective polysialyltransferases (polySTs) responsible for polymerizing sialyl residues from donor CMP-sialic acid are not homologous glycosyltransferases. To better define the mechanism of capsule biosynthesis, we established the functional interchangeability of bacterial polySTs by complementation of a polymerase-deficient E. coli K1 mutant with the polyST genes from groups B or C N. meningitidis and the control E. coli K92 polymerase gene. The biochemical and immunochemical results demonstrated that linkage specificity is dictated solely by the source of the polymerase structural gene. To determine the molecular basis for linkage specificity, we created chimeras of the K1 and K92 polySTs by overlap extension PCR. Exchanging the first 52 N-terminal amino acids of the K1 NeuS with the C terminus of the K92 homologue did not alter specificity of the resulting chimera, whereas exchanging the first 85 or reciprocally exchanging the first 100 residues did. These results demonstrated that linkage specificity is dependent on residues located between positions 53 and 85 from the N terminus. Site-directed mutagenesis of the K92 polyST N terminus indicated that no single residue alteration was sufficient to affect specificity, consistent with the proposed function of this domain in orienting the acceptor. The combined results provide the first evidence for residues critical to acceptor binding and elongation in polysialyltransferase. N-acetylneuraminic acid N-acetylglucosamine polysialic acid polysialyltransferase anti-polyα2,8-linked sialic acid antiserum anti-polyα2,9-linked sialic acid antiserum endo-N-acylneuraminidase (polyα2,8-linked sialic acid depolymerase, endosialidase) PSA purified fromE. coli K92 strain Bos-12 PSA purified from group C meningococci neural cell adhesion molecule Homopolymers of N-acetylneuraminic acid (Neu5Ac,1 the most common sialic acid), also known as polysialic acids (PSAs), are essential virulence factors in neuroinvasive diseases caused by Escherichia coli K1 and K92 and certain strains of Neisseria meningitidis, Moraxella nonliquefaciens, andMannheimia (Pasteurella) hemolytica(1Silver R.P. Vimr E.R. Iglewski B. Miller V. The Bacteria 11, Molecular Basis of Bacterial Pathogenesis. Academic Press, New York1990: 39-60Google Scholar). PSA also regulates cell-cell apposition when expressed in higher eukaryotes on the neural cell adhesion molecule (NCAM) (2Rutishauser U. Watanabe M. Silver J. Troy F.A. Vimr E.R. J. Cell Biol. 1985; 101: 1842-1849Crossref PubMed Scopus (216) Google Scholar). In E. coli K1 or N. meningitidis group B, the polymeric sialyl residues are connected by α2,8 linkages, whereas group CN. meningitidis and E. coli K92 express PSA homopolymers with α2,9 or alternating α2,8/2,9 linkages, respectively (3Bhattacharjee A.K. Jennings H.J. Denny C.P. Martin A. Smith I.C.P. J. Biol. Chem. 1975; 250: 1926-1932Abstract Full Text PDF PubMed Google Scholar, 4Egan W. Liu T.-Y. Dorow D. Cohen J.S. Robbins J.D. Gotschlich E.C. Robbins J.B. Biochemistry. 1977; 16: 3687-3692Crossref PubMed Scopus (92) Google Scholar). Despite extensive investigation of PSA biosynthesis, including identification of the polySTs catalyzing the polymerization of sialyl residues, we do not know how PSA synthesis is initiated, how chain extension is terminated, or how polySTs interact with donor (CMP-Neu5Ac) and acceptor (nascent PSA) substrates. In contrast, all mammalian sialyltransferases investigated to date include a common primary structural motif (the L, or large sialyl motif consisting of 48 or 49 amino acid residues), which is thought to function in donor binding, and an S (small)-sialyl motif of 23 amino acid residues believed to bind both donor and acceptor substrates (5Datta A.K. Paulson J.C. J. Biol. Chem. 1995; 270: 1497-1500Abstract Full Text Full Text PDF PubMed Scopus (180) Google Scholar, 6Datta A.K. Sinha A. Paulson J.C. J. Biol. Chem. 1998; 273: 9608-9614Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 7Datta A.K. Paulson J.C. Indian J. Biochem. Biophys. 1997; 34: 157-165PubMed Google Scholar). Despite the structural identity between bacterial and mammalian PSA, the respective polySTs are not homologous glycosyltransferases, indicating the existence of at least two origins if not completely different mechanisms of sialyl homopolymer biosynthesis. The biosynthesis of bacterial PSA is thought to take place on the inner surface of the cytoplasmic membrane through the polyST-catalyzed addition of Neu5Ac residues from CMP-Neu5Ac donor to the nonreducing ends of nascent (acceptor) PSA chains. The functions of most of the proteins required for biosynthesis (encoded by the neugenes) have been identified (8Vimr E.R. Steenbergen S. Cieslewicz M. J. Industrial Microbiol. 1995; 15: 352-360Crossref PubMed Scopus (50) Google Scholar), including those needed for Neu5Ac synthesis, activation, and polymerization (see Fig. 1). Epistasis analysis and complementation experiments showed that polymerase activity is dependent on the structural integrity of the polysaccharide translocation and capsule assembly apparatus (9Vimr E.R. Aaronson W. Silver R.P. J. Bacteriol. 1989; 172: 1106-1117Crossref Google Scholar, 10Vimr E.R. Steenbergen S.M. Roth J. Rutishauser U. Troy II, F.A. Polysialic Acid, from Microbes to Man. Birkhauser Verlag, Basel1993: 73-91Google Scholar), suggesting the multiprotein complex diagramed below in Fig. 1. Using an immunological approach, a similar conclusion was reached for biosynthesis of the K5 capsule (11Hodson 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, 12Arrecubieta C. Hammarton T. Barrett B. Charesonsudjai S. Hodson N. Rainey D. Roberts I.S. J. Biol. Chem. 2001; 276: 4245-4250Abstract Full Text Full Text PDF PubMed Scopus (36) Google Scholar). Thus, K1 or K5 mutants with defects in conserved polypeptides encoded by the kps genes exhibit reduced rates of polymerization and accumulate intracellular (untranslocated) polysaccharides. Our working hypothesis, that polymerization is closely linked to translocation through protein-protein and protein-polysaccharide interactions, is supported by the apparent processivity of polyST and the loss or reduction of activity when polyST is expressed in the absence of other neu orkps gene products. Understanding how capsular polysaccharide production is regulated requires knowledge of polyST function. To extend our analysis of polyST function, we carried out heterologous complementation experiments, establishing the interchangeability of polymerases for PSA expression in an E. coli K1 polyST-deficient recipient and providing direct evidence for the exclusive role of the polymerases in specifying all known PSA linkages. Then, on the basis of primary structural similarities, we created reciprocal chimeras between the K1 and K92 polySTs to identify the enzymic domains conferring linkage specificity. Site-directed alteration of selected K1 or K92 polyST residues provided additional functional information. The results extend our earlier conclusion that the K92 polyST is a bifunctional enzyme (13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar), demonstrate for the first time cross-species complementation of neuS, and establish the domain organization of the K1 and K92 enzymes responsible for specifying product linkage. These conclusions provide an experimental scaffold for future studies aimed at better understanding the functions of polySTs and the regulation of PSA biosynthesis in bacteria, neurons, and tumor cells. Laboratory E. coli K-12 strain DH5α was used as the recipient for all molecular cloning of the various polyST structural genes. E. coli K1 hybrid strains EV136 (neuS::Tn10) and EV240 (neuS::Tn10 nanA4) are derivatives of the wild type strain EV36 (13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar, 14Steenbergen S.M. Vimr E.R. Mol. Microbiol. 1990; 4: 603-611Crossref PubMed Scopus (44) Google Scholar), whereas RS218 is a clinical K1 isolate kindly provided by Dr. R. Silver of the University of Rochester School of Medicine, Rochester, NY. These strains do not synthesize a PSA capsule due to insertional inactivation of the polyST structural gene, neuS, but retain all other genetic information for precursor biosynthesis, polymer assembly, and PSA translocation. The nonpolar nanA4 mutation inactivates sialate lyase (aldolase), thus the strain cannot degrade free sialic acid. EV239 is aneuB25 relative of EV240 and will not synthesize PSA even with neuS+ expressed in trans, unless provided with an exogenous source of free Neu5Ac, because of the loss of sialate synthase activity encoded by neuB (13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar). EV136 and EV240 synthesize PSA when neuS+ is provided intrans by transformation with either pSX92 or pSX305, encoding polySTs from E. coli K1 or K92 wild type strains (Fig. 2), respectively (13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar). EV138 harboring pLysS is ananA4, neuB2 double mutant whose construction has been previously described (15Vimr E.R. J. Bacteriol. 1992; 174: 6191-6197Crossref PubMed Google Scholar). Plasmids pUE3 and pUE14 (Fig. 2) contain the synD (polyST) or synB (CMP-Neu5Ac synthetase) structural genes from group B N. meningitidisand were the kind gift of Matthias Frosch (16Edwards U. Muller A. Hammerschmidt S. Gerardy-Schahn R. Frosch M. Mol. Microbiol. 1994; 14: 141-149Crossref PubMed Scopus (85) Google Scholar). The plasmid pSX306 harboring synE-encoding polyST from group C N. meningitidis was constructed from the strain Fam18 PCR product kindly provided by David Stephens (17Swartley J.S. Marfin A.A. Edupuganti S. Liu L.-J. Cieslak P. Perkins B. Wenger J.D. Stephens D.S. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 271-276Crossref PubMed Scopus (342) Google Scholar), which was subsequently ligated into the Promega vector pGEM-T Easy following the manufacturer's suggestions for direct cloning of PCR amplicons (see Fig. 2). Unless indicated otherwise, all experiments were carried out with bacteria grown in Luria-Bertani medium with vigorous aeration at 37 °C. Minimal medium for reconstitution of PSA expression in strain EV138 was carried out as previously described (15Vimr E.R. J. Bacteriol. 1992; 174: 6191-6197Crossref PubMed Google Scholar). Plasmids were maintained by cultivating bacteria with 100 μg ml−1 ampicillin. Cells were harvested by centrifugation during the late exponential or early stationary growth phase. Immunochemical determination of capsule chemotype was accomplished by the halo test (13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar, 14Steenbergen S.M. Vimr E.R. Mol. Microbiol. 1990; 4: 603-611Crossref PubMed Scopus (44) Google Scholar) using H46 or αGpC antiserum kindly donated by Willie Vann. H46 is specific for a minimal epitope of approximately eight α2,8-linked Neu5Ac residues, whereas αGpC recognizes the α2,9 linkages in PSAs from E. coli K92 or group C meningococci. Bacterial colonies that synthesize and export capsular polysaccharides on agar plates supplemented with 10% (v/v) antiserum produce precipitin halos that are indicative of PSA linkage (chemotype). endo-N was purified from phage K1F lysates as previously described (18Petter J.G. Vimr E.R. J. Bacteriol. 1993; 175: 4354-4363Crossref PubMed Google Scholar). All other chemicals were obtained from standard commercial sources and were of the highest available purity. Endogenous or exogenous polyST activity was determined in a total reaction volume of 15 μl in 10 mm Tris-HCl (pH 8.0), 5 mm magnesium acetate, and 5 mm dithiothreitol as described previously (9Vimr E.R. Aaronson W. Silver R.P. J. Bacteriol. 1989; 172: 1106-1117Crossref Google Scholar, 13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar,14Steenbergen S.M. Vimr E.R. Mol. Microbiol. 1990; 4: 603-611Crossref PubMed Scopus (44) Google Scholar), except that CMP-[4-14C]Neu5Ac (55 mCi mmol−1) was purchased from American Radiolabeled Chemicals, Inc. Data were expressed as picomoles of Neu5Ac transferred per unit of time per milligram of protein. Transfer to exogenous acceptors was normalized per milligram of acceptor. Neu5Ac tritiated on carbon-6 was obtained from CMP-[6-3H]Neu5Ac (20 Ci mol−1) by acid hydrolysis as previously described (18Petter J.G. Vimr E.R. J. Bacteriol. 1993; 175: 4354-4363Crossref PubMed Google Scholar). Overlap extension PCR for the construction of genes encoding chimeric polypeptides was carried out as originally described (19Horton R.M. Cai Z.L. Ho S.N. Pease L.R. BioTechniques. 1990; 8: 528-535PubMed Google Scholar). In two separate PCR amplifications, double-stranded fragments are generated and subsequently fused in PCR reaction 3. PCR reaction 1 uses a common 5′-far primer located 152 bases from the ATG start codons of either the K92 or K1neuS genes (5′-GCCGCCAAATGTTAATGTTAGGAC-3′), and the unique reverse overlap primer (Table I) that carries information for constructing the intended chimera. In PCR reaction 2, a 3′-far primer located 175 bases from the stop codon of either the K92 or K1 neuS genes (5′-CGGTTTATTATGGGGGGAACACAA-3′) and a primer that is the perfect complement of the overlap primer used in PCR reaction 1 (forward primers, Table I) are used to generate the second, or C-terminal fragment. When these two PCR products are mixed, denatured, and cooled the complementary overlaps anneal and the junction is extended by the template-dependent DNA polymerase. Finally, nested primers (5′-near and 3′-near) amplify the fragment bearing the chimeric construction and are cloned in pGEM-T Easy. The near primers are as follows: K1–5′-near, 5′-GGACTTTTGGAATTAAAAGATCTAC-3′; K92–5′-near, 5′-GGACTTTTGGAATTAAAAGATCGA-3′; K1–3′-near, 5′-CCATCCTCTTCAAAGAAAAGTAAC-3′; and K92–3′-near, 5′-GCATCCTCTTCAAAGAAA- AGTTAC-3′.Table ISpecific primers used for chimeric enzyme synthesis and mutagenesis by overlap extension PCRConstructForward primer1-aSlashes in parentheses indicate the position of overlap junctions between neuS homologues.,1-bResidues in parentheses indicate specific base change included in the primer for site-directed mutagenesis.Reverse primerpSX611, pSX612, and pSX6145′-AATAGTAGTTTGTATGAAGAAAA(/)TAA TTTATTATTTATTTGTTCA-3′5′-TGAACAAATAAATAATAAATTA(/)TTT TCTTCATACAAACTACTATT-3′pSX613 and pSX6155′-AATAGTACCTTGTATGAAGAAAA CAAC(/)TTTTTATTCATTTGCTCA-3′5′-TGAGCAAATGAATAAAAAGTTG(/)TTT TCTTCATACAAGGTACTATT-3′pSX6165′-AATCATTTGCTCAAGAATTAAT(T)TT GGCATTTGAGTGTTTTTTTT-3′5′-AAAAAAAACACTGAAATGCCAA(A) ATTAATTCTTGAGCAAATGAAT-3′pSX6175′-TGACGATTTAACTAATGGAATA(CC)T TTAAACTCAAAAATTTTTTT-3′5′-AAAAAAATTTTTGAGTTTAAA(GG) TATTCCATTAGTTAAATCGTCA-3′pSX6185′-TATAAAAACTCTTGCGTTATCAG(G)T TTATCTATTCTCATCAAGGT-3′5′-ACCTTGATGAGAATAGATAAA(C)C TGATAACGCAAGAGTTTTTATA-3′pSX6195′-TTTGTATGAAGAAAACAACTTATT(A) TTCATTTGCTCAAGAATTAA-3′5′-TTAATTCTTGAGCAAATGAATAA(T) AAGTTGTTTTCTTCATACAAA-3′pSX6205′-CTTGTATGAAGAAAATAATTT(T)TTA TTTATTTGTTCAACAATAAT-3′5′-ATTATTGTTGAACAAATAAATAA(A) AAATTATTTTCTTCATACAAG-3′pSX6215′-CATTTGCTCAAGAATTAATCTT(/)GG CATCTCAGTATTTTTTTCAGT-3′5′-ACTGAAAAAAATACTGAGATGCC(/) AAGATTAATTCTTGAGCAAATG-3′pSX6225′-CTCTGCTTTCTTTAATTGCCCTAAATG (/)ACTAATTATAAAAATATT-3′5′-AATATTTTTATAATTAGT(/)CAT TTAGGGCAATTAAAGAAAGCAGAG-3′pSX6235′-TCTGCTTTCTTTAATTGCCCTAAAT(T) AGCAACAATAAAAATATTT-3′5′-AAATATTTTTATTGTTGCT(A)ATT TAGGGCAATTAAAGAAAGCAGA-3′1-a Slashes in parentheses indicate the position of overlap junctions between neuS homologues.1-b Residues in parentheses indicate specific base change included in the primer for site-directed mutagenesis. Open table in a new tab Sufficient flanking DNA sequence was available for constructions involving K1 neuS (accession number M76370), but there was insufficient 3′-flanking data to construct chimeras fusing the C terminus of K92 NeuS (accession number M88479) to the N terminus of the K1 enzyme. Therefore, we sequenced pSX305 with a primer (5′-ATTAGCATCATCTTCTTTGGT-3′) identical to bp 1002–1022 of K92neuS to yield the 529 bp of new sequence 3′ to the K92neuS stop codon. These new sequence data are deposited under accession number AF318310. The conditions used for the first round of PCR were 35 cycles at 95 °C for 30 s interspersed at 60–65 °C for 30 s for annealing (this annealing temperature was optimized within the indicated range as determined individually for each primer set) and extension at 72 °C for 1 min. PCR reaction 3 was 35 cycles at 95 °C for 30 s, 65 °C for 30 s, and 72 °C for 90 s. All primers (synthesized and PAGE-purified by Integrated DNA Technologies, Inc.) were used at a concentration of 50 pmol reaction−1. Templates used for the first round of PCR were either the K1 (pSX92) or K92 (pSX305) neuS genes. In the second round of amplification, PCR reaction 3, 1 μl of each first round reaction was added as template. Taq polymerase was used in the PCR reactions for the initial set of chimeric polySTs. DNA sequencing of these constructs revealed unintended base changes introduced by the polymerase. Therefore, all subsequent amplifications were carried out using the PCR SuperMix High Fidelity purchased from Invitrogen. This PCR mix is a combination of the proofreading polymerase from Pyrococcus sp. GB-D andTaq DNA polymerase. After verification of product formation the PCR product was purified using the Wizard PCR Prep DNA purification system from Promega. After purification the products were A-tailed and cloned into the pGEM-T Easy Vector following the manufacturer's directions and transformed into DH5α for further analysis or storage. All constructs were sequenced using the pUC/M13 Forward and Reverse primers to verify the junction formed and to ensure no unplanned mutations were introduced during PCR amplifications. Any construct with an altered phenotype was completely sequenced on both strands using the previously named universal primers as well as threeneuS-specific primers that allowed complete double-stranded sequencing in two to three more reactions. DNA sequencing was carried out at the University of Illinois Keck Center for Biotechnology using standard automated sequencing and data retrieval systems. The plasmids were transformed into EV240 to test for polyST activity. Mutagenesis of the K1 or K92neuS genes was performed either by overlap extension (20Ho S.N. Hunt H.D. Horton R.M. Pullen J.K. Pease L.R. Gene (Amst.). 1989; 77: 51-59Crossref PubMed Scopus (6851) Google Scholar) or using the Altered Sites II in vitro Mutagenesis System fromPromega, in which the BglII-BamHI fragments of pSX92 or pSX305 were cloned into the mutagenesis vector. The primers used for the mutagenesis reactions by overlap extension are listed in Table I. The other mutagenic primers with their respective plasmids are as follows: pSX624, 5′-TCTTGGCATTTCAGTGTTTTTTTCAGTTGCCAAGACGATAAGAAA-3′; pSX625, 5′-CCTACTAAATTTTTGTATAAATGACTCTGCTTTCTTTAATTGCCC-3′; pSX626, 5′-GACGATAAGAAAATTACTACGCTTACTAAATTTTTGTATAAAAAG-3′; where the underlined residues indicate the base(s) altered to make the desired mutant polySTs. All mutant neuS copies were sequenced as described above to ensure that only the intended alteration was introduced by the mutagenesis. In addition to the immunological analysis of PSA chemotypes with linkage-specific antisera described above, sensitivity to K1-specific phage was used to confirm the synthesis of α2,8 sialyl linkages by plaque assay. Strains synthesizing K1 or K92 capsules are sensitive to bacteriophage infection and produce a characteristic plaque phenotype (21Cieslewicz M. Vimr E. Mol. Microbiol. 1997; 26: 237-249Crossref PubMed Scopus (65) Google Scholar). Where indicated, linkage assignments were confirmed by endo-N digestion and TLC analysis on silica gel G (Merck Research Laboratories) in an isopropanol, ammonium hydroxide, water (6:1:2) solvent system as previously described (18Petter J.G. Vimr E.R. J. Bacteriol. 1993; 175: 4354-4363Crossref PubMed Google Scholar). Sialyl oligomers of defined length for use as chromatography standards were produced by partial acid hydrolysis of colominic acid as previously described (22Vimr E.R. McCoy R.D. Vollger H.F. Wilkison N.C. Troy F.A. Proc. Natl. Acad. Sci. U. S. A. 1984; 81: 1971-1975Crossref PubMed Scopus (146) Google Scholar). Oligomers were identified colorimetrically with orcinol or by autoradiography using a Kodak BioMax Transcreeen LE intensifying screen for 14C-labeled sialic acids or Dupont Cronex Quanta-III intensifying screen after autoradiographic enhancement with Fluoro-Hance (rpi Research Products, Mount Prospect, IL). We previously showed that when neuS is expressed in an E. coli K-12 strain such as DH5α, a genetic background lacking the kps and neu genes for capsule biosynthesis (Fig. 1), polyST is unable to initiate PSA synthesis in vitro but transfers Neu5Ac from CMP-Neu5Ac to the exogenous acceptor colominic acid (13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar,14Steenbergen S.M. Vimr E.R. Mol. Microbiol. 1990; 4: 603-611Crossref PubMed Scopus (44) Google Scholar). As expected from these previous results, we observed that neither K92 neuS nor synD or synE expressed a polyST that was capable of initiating PSA synthesis in the K-12 background when provided with an in vitro source of the nucleotide sugar donor, CMP-Neu5Ac (TableII). A control plasmid (pUE14) expressing the group B meningococcal CMP-Neu5Ac synthetase encoded bysynB (Fig. 2), as expected, also had no polyST activity (Table II). In contrast, each of the recombinant polySTs expressed in E. coli K-12 transferred sialyl residues from CMP-Neu5Ac to both the cognate and heterologous exogenous acceptors, although with different relative efficiencies (Table II), indicating that at least for exogenous acceptors the mechanism of acceptor binding does not absolutely discriminate on the basis of sialyl linkage. Addition of monomeric Neu5Ac to the extracts as a potential exogenous acceptor did not result in polyST activities above the no-acceptor controls, indicating that none of the polySTs tested could use free sialic acid as exogenous acceptor. We concluded that all bacterial polySTs possess structurally similar acceptor binding sites. However, exogenous sialyl oligomers are poorly elongated (23Vimr E.R. Bassler B. Troy F.A. Abst. Ann. Meet. Am. Soc. Microbiol. 1985; K152: 154Google Scholar, 24Ferrero M.A. Luengo J.M. Reglero A. Biochem. J. 1991; 280: 575-579Crossref PubMed Scopus (16) Google Scholar, 25Chao C.-F. Chaung H.-C. Chiou S.-T. Liu T.-Y. J. Biol. Chem. 1999; 274: 18206-18212Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar), with most acting as acceptor for a single sialyl addition unless tethered to the membrane by a lipid anchor (26Cho J.-W. Troy F.A. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 11427-11431Crossref PubMed Scopus (34) Google Scholar). Therefore, broadly extrapolating results derived from the use of exogenous acceptors to natural PSA elongation may be suspect to misinterpretation because of the failure to duplicate the in vivo process. To overcome the problems inherent with these in vitro systems we developed an in vivo complementation system that permits the detection of polyST activity under isogenic conditions.Table IIExogenous polyST activity expressed in E. coli K-12PlasmidRelevant genotypePolyST activity with2-aData are normalized (except the no acceptor control) to 1 mg of acceptor per assay.No acceptor2-bActivities are comparable to E. coli K-12 strain DH5α with no plasmids.CABos-12GpCpmol/mg proteinpUE14synB1.20.4ND2-cND, not determined.NDpUE3synD0.6471365NDpSX92K1neuS1.612001694 ± 1062-dS.D. between independent measurements.NDpSX305K92 neuS0.9671308100pSX316synE4.2662878039802-a Data are normalized (except the no acceptor control) to 1 mg of acceptor per assay.2-b Activities are comparable to E. coli K-12 strain DH5α with no plasmids.2-c ND, not determined.2-d S.D. between independent measurements. Open table in a new tab PolyST activity can be detected by expressing an exogenous polymerase gene in trans to an E. coliK1 genetic background lacking endogenous polymerase due to transposon (insertional) inactivation of neuS (13Steenbergen S.M. Wrona T.J. Vimr E.R. J. Bacteriol. 1992; 174: 1099-1108Crossref PubMed Google Scholar, 14Steenbergen S.M. Vimr E.R. Mol. Microbiol. 1990; 4: 603-611Crossref PubMed Scopus (44) Google Scholar). Similarly, a double mutant with defects in neuB and nanA is dependent on an exogenous source of free Neu5Ac for PSA biosynthesis due to lack of an active synthase encoded by neuB (15Vimr E.R. J. Bacteriol. 1992; 174: 6191-6197Crossref PubMed Google Scholar). Fig.3 shows the time course of PSA synthesis when strain EV138 (nanA4 neuB25) is continuously pulsed with exogenous radiolabeled Neu5Ac. The sialic acid internalized by the sialate permease encoded by nanT is spared from the catabolic pathway by inactivation of sialate lyase (nanA) and thus can only participate as an intermediate in PSA biosynthesis (27Vimr E.R. Troy F.A. J. Bacteriol. 1985; 164: 854-860Crossref PubMed Google Scholar, 28Ringenberg M.C. Lichtensteiger C. Vimr E.R. Glycobiology. 2001; 11: 533-539Crossref PubMed Scopus (33) Google Scholar). The internalized Neu5Ac is activated for glycosyltransfer by the synthetase encoded by neuA, which couples Neu5Ac with CTP to produce intracellular CMP-Neu5Ac detectable by 5 min after beginning the pulse (Fig. 3). PSA steadily accumulated at the origin of the chromatogram without observable intermediates. The solvent system used for the TLC analysis separates Neu5Ac monomers through oligomers of about 10 sialyl units in length. The absence of detectable oligomeric sialic acid in vivo is consistent with the processive activity of polyST inferred previously from in vitro results (23Vimr E.R. Bassler B. Troy F.A. Abst. Ann. Meet. Am. Soc. Microbiol. 1985; K152: 154Google Scholar, 24Ferrero M.A. Luengo J.M. Reglero A. Biochem. J. 1991; 280: 575-579Crossref PubMed Scopus (16) Google Scholar, 25Chao C.-F. Chaung H.-C. Chiou S.-T. Liu T.-Y. J. Biol. Chem. 1999; 274: 18206-18212Abstract Full Text Full Text PDF PubMed Scopus (16) Google Scholar). The simplest interpretation of the current results is that PSA is elongated processively in vivo by a single polyST before chain termination by an unknown mechanism. To determine if all extant bacterial PSA chemotypes were dependent on the genetic origin of their cognate polymerases, we carried out complementation experiments with the neuS homologuessynD or synE from N. meningitidisgroups B and C, respectively (Fig. 2). The immunoreactivities of the resulting polysaccharide products confirmed that polyST linkage specificity is retained when these enzymes catalyzed PSA synthesis in the heterologous mutant K1 background (TableIII). Structural linkage assi

Referência(s)