Artigo Acesso aberto Revisado por pares

3-Deoxy-d-manno-oct-2-ulosonic Acid (Kdo) Transferase (WaaA) and Kdo Kinase (KdkA) of Haemophilus influenzae Are Both Required to Complement a waaAKnockout Mutation of Escherichia coli

2000; Elsevier BV; Volume: 275; Issue: 45 Linguagem: Inglês

10.1074/jbc.m005204200

ISSN

1083-351X

Autores

Werner Brabetz, Sven Müller‐Loennies, Helmut Brade,

Tópico(s)

Bacterial Genetics and Biotechnology

Resumo

The lipopolysaccharide (LPS) of the deep rough mutant Haemophilus influenzae I69 consists of lipid A and a single 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) residue substituted with one phosphate at position 4 or 5 (Helander, I. M., Lindner, B., Brade, H., Altmann, K., Lindberg, A. A., Rietschel, E. T., and Zähringer, U. (1988) Eur. J. Biochem. 177, 483–492). ThewaaA gene encoding the essential LPS-specific Kdo transferase was cloned from this strain, and its nucleotide sequence was identical to H. influenzae DSM11121. The gene was expressed in the Gram-positive host Corynebacterium glutamicum and characterized in vitro to encode a monofunctional Kdo transferase. waaA of H. influenzae could not complement a knockout mutation in the corresponding gene of an Re-type Escherichia coli strain. However, complementation was possible by coexpressing the recombinantwaaA together with the LPS-specific Kdo kinase gene (kdkA) of H. influenzae DSM11121 or I69, respectively. The sequences of both kdkA genes were determined and differed in 25 nucleotides, giving rise to six amino acid exchanges between the deduced proteins. Both E. colistrains which expressed waaA and kdkA fromH. influenzae synthesized an LPS containing a single Kdo residue that was exclusively phosphorylated at position 4. The structure was determined by nuclear magnetic resonance spectroscopy of deacylated LPS. Therefore, the reaction products of both cloned Kdo kinases represent only one of the two chemical structures synthesized by H. influenzae I69. The lipopolysaccharide (LPS) of the deep rough mutant Haemophilus influenzae I69 consists of lipid A and a single 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) residue substituted with one phosphate at position 4 or 5 (Helander, I. M., Lindner, B., Brade, H., Altmann, K., Lindberg, A. A., Rietschel, E. T., and Zähringer, U. (1988) Eur. J. Biochem. 177, 483–492). ThewaaA gene encoding the essential LPS-specific Kdo transferase was cloned from this strain, and its nucleotide sequence was identical to H. influenzae DSM11121. The gene was expressed in the Gram-positive host Corynebacterium glutamicum and characterized in vitro to encode a monofunctional Kdo transferase. waaA of H. influenzae could not complement a knockout mutation in the corresponding gene of an Re-type Escherichia coli strain. However, complementation was possible by coexpressing the recombinantwaaA together with the LPS-specific Kdo kinase gene (kdkA) of H. influenzae DSM11121 or I69, respectively. The sequences of both kdkA genes were determined and differed in 25 nucleotides, giving rise to six amino acid exchanges between the deduced proteins. Both E. colistrains which expressed waaA and kdkA fromH. influenzae synthesized an LPS containing a single Kdo residue that was exclusively phosphorylated at position 4. The structure was determined by nuclear magnetic resonance spectroscopy of deacylated LPS. Therefore, the reaction products of both cloned Kdo kinases represent only one of the two chemical structures synthesized by H. influenzae I69. lipopolysaccharide(s) 3-deoxy-d-manno-oct-2-ulosonic acid monoclonal antibody correlation spectroscopy double quantum-filtered rotating frame Overhauser effect rotating frame Overhauser effect spectroscopy heteronuclear multiple quantum coherence high performance anion exchange chromatography polymerase chain reaction base pair(s) Haemophilus influenzae is a nonenteric Gram-negative bacterium that is found in the human respiratory tract and may cause severe diseases, in particular septicemia and meningitis in children. One major virulence factor of this pathogen is the type b capsular polysaccharide, which is also the basis of the presently used vaccine (1Moxon E.R. J. Infect. Dis. 1992; 165 (suppl.): 77-81Crossref Scopus (19) Google Scholar). In addition, lipopolysaccharides (LPS)1 play a crucial role in the interaction of this microorganism with the host's immune system. LPS contribute to each stage of pathogenesis of H. influenzae infection including colonization of the upper respiratory tract, systemic dissemination, and the invasion of the central nervous system (2Hood D.R. Moxon E.R. Brade H. Opal S.M. Vogel S.N. Morrison D.C. Endotoxin in Health and Diseases. Marcel Dekker, Inc., New York1999: 39-54Google Scholar). During these processes, several surface exposed epitopes of the molecule are the subject of high frequency phase variation. LPS are the major amphiphilic constituents of the outer leaflet of the outer membrane of Gram-negative bacteria. They share a common architecture composed of a membrane-anchored phosphorylated and acylated β(1→6) linked glucosamine disaccharide, termed lipid A, to which a carbohydrate moiety of varying size is attached. The latter always contains 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) linked to lipid A. Mutants that are defective in biosynthetic enzymes of the Kdo-lipid A region are conditionally thermosensitive, indicating that this minimal structure is absolutely required for the integrity of the outer membrane and microbial cell growth (3Rick P.D. Raetz C.R.H. Brade H. Opal S.M. Vogel S.N. Morrison D.C. Endotoxin in Health and Diseases. Marcel Dekker, Inc., New York1999: 283-304Google Scholar, 4Belunis C.J. Clementz T. Carty S.M. Raetz C.R.H. J. Biol. Chem. 1995; 270: 27646-27652Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). Based on these findings, the corresponding enzymes evoked increasing interest as potential targets for new antibiotics against Gram-negative bacteria (5Onishi H.R. Pelak B.A. Gerckens L.S. Silver L.L. Kahan F.M. Chen M.H. Patchett A.A. Galloway S.M. Hyland S.A. Anderson M.S. Raetz C.H.R. Science. 1996; 274: 980-982Crossref PubMed Scopus (359) Google Scholar). The lipid A-linked Kdo residue is often substituted with a second Kdo, forming the disaccharide α-Kdo-(2→4)-α-Kdo, which also has been identified as the minimal core structure in enterobacterial deep rough mutants of the Re-type. Further core sugars are always attached to position 5 of the inner Kdo in wild-type bacteria. In contrast, LPS from H. influenzaecontain only a single Kdo residue substituted with phosphate or 2-aminoethanol pyrophosphate in position 4 (6Holst Brade H. Opal S.M. Vogel S.N. Morrison D.C. Endotoxin in Health and Diseases. Marcel Dekker, Inc., New York1999: 115-154Google Scholar). A deep rough mutant, termed I69, has been obtained from H. influenzaeRd−/b+ (7Zwalen A. Rubin L.G. Connelly C.J. Inzana T.J. Moxon E.R. J. Infect. Dis. 1985; 152: 485-492Crossref PubMed Scopus (44) Google Scholar), and chemical analyses of its LPS revealed two molecules composed of lipid A and a single Kdo residue phosphorylated either at position 4 or 5 (8Zamze S.E. Ferguson M.A.J. Moxon E.R. Dwek R.A. Rademacher T.W. Biochem. J. 1987; 245: 583-587Crossref PubMed Scopus (34) Google Scholar, 9Helander I.M. Lindner B. Brade H. Altman K. Lindberg A.A. Rietschel E.T. Zähringer U. Eur. J. Biochem. 1988; 177: 483-492Crossref PubMed Scopus (171) Google Scholar). Immunofluorescence with monoclonal antibodies (mAbs) identified both molecules in living bacteria of this strain (10Rozalski A. Brade L. Kosma P. Moxon R. Kusumoto S. Brade H. Mol. Microbiol. 1997; 23: 569-577Crossref PubMed Scopus (14) Google Scholar). Kdo transferases (WaaA) have been described as multifunctional enzymes that are able to transfer several Kdo residues from CMP-Kdo to different precursor molecules forming different linkages, which has not been observed for other glycosyltransferases (11Kleene R. Berger E.G. Biochim. Biophys. Acta. 1993; 1154: 283-325Crossref PubMed Scopus (202) Google Scholar). Bifunctional enzymes capable of synthesizing an α(2→4)-linked Kdo disaccharide have been cloned and characterized from Escherichia coli (12Belunis C.J. Raetz C.R.H. J. Biol. Chem. 1992; 267: 9988-9997Abstract Full Text PDF PubMed Google Scholar),Acinetobacter baumannii, and Acinetobacter haemolyticus (13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar). Furthermore, in Chlamydiaceae, even tri- and tetrafunctional WaaA have been identified that are responsible for the biosynthesis of a family-specific surface epitope comprising an α-Kdo-(2→8)-α-Kdo-(2→4)-α-Kdo trisaccharide of diagnostic value (12Belunis C.J. Raetz C.R.H. J. Biol. Chem. 1992; 267: 9988-9997Abstract Full Text PDF PubMed Google Scholar, 14Mamat U. Baumann M. Schmidt G. Brade H. Mol. Microbiol. 1993; 10: 935-941Crossref PubMed Scopus (43) Google Scholar, 15Löbau S. Mamat U. Brabetz W. Brade H. Mol. Microbiol. 1995; 18: 391-399Crossref PubMed Scopus (48) Google Scholar). In contrast, a monofunctional Kdo transferase activity has been demonstrated together with an LPS-specific, ATP-dependent Kdo kinase from membrane extracts of H. influenzae (16White K.A. Kalashov I.A. Cotter R.J. Raetz C.R.H. J. Biol. Chem. 1997; 272: 16555-16563Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar). These data have been recently confirmed by the cloning and in vitro characterization of a Kdo kinase gene (kdkA) (17White K.A. Lin S. Cotter R.J. Raetz C.R.H. J. Biol. Chem. 1999; 274: 31391-31400Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar). However, due to limiting amounts of thein vitro reaction products of KdkA, the position to which phosphate had been transferred has not been determined. We have established a cloning system based on a defined Re-typeE. coli strain that is devoid of the host's Kdo transferase activity and additionally harbors a ΔwaaCF mutation within the heptosyltransferase I and II genes involved in the consecutive transfer ofl-glycero-d-manno-heptose residues to Kdo. This strategy allowed us to characterize LPS that were synthesized in vivo by cloned Kdo transferases without interfering activity of the essential host-specific enzyme (18Brabetz W. Lindner B. Brade H. Eur. J. Biochem. 2000; 167: 5458-5465Crossref Scopus (30) Google Scholar). Using this approach, we now show that the cloned monofunctionalwaaA from H. influenzae is not able to complement a knockout mutation within the corresponding gene of E. coli; however, cloning of both waaA and kdkAfrom a wild type or the I69 strain of H. influenzae, respectively, we are able to complement the mutation. [γ-32P]ATP (4 × 1015 Bq/mol) was obtained from ICN Biochemicals. Restriction enzymes and T4 DNA ligase were purchased from New England Biolabs. CTP, NAD, isopropyl-β-d-thiogalactoside, hemin, and antibiotics were obtained from Sigma. Silica Gel 60 thin layer chromatography (TLC) plates were from Merck. Kdo and synthetic tetraacyl lipid A precursor compound 405 (1-monophosphoryl) and 406 (bisphosphoryl) (19Imoto M. Yoshimura H. Yamamoto M. Kusumoto S. Shiba T. Tetrahedron Lett. 1984; 26: 1545-1548Crossref Scopus (156) Google Scholar) were gifts of P. Kosma (University of Agricultural Sciences, Vienna, Austria) and S. Kusumoto (Osaka University, Osaka, Japan), respectively. CMP-Kdo synthetase was partially purified from Corynebacterium glutamicum R163/pJKB14 as described (13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar), and lipid A 4′-kinase was prepared from E. coli BLR(DE3)/pLysS/pJK2 according to Garrett et al. (20Garrett T.A. Kadrmas J.L. Raetz C.R.H. J. Biol. Chem. 1997; 272: 21855-21864Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar). Plasmids and bacterial strains used in this study are listed in TableI. H. influenzaestrains were cultivated at 37 °C in brain-heart-infusion (Life Technologies, Inc.) supplemented with 10 mg/liter hemin and 10 mg/liter NAD as described (21Hoiseth S.K. Balows A. Trüper H.G. Dworkin M. Harder W. Schleifer K.-H. The Prokaryotes. 2nd Ed. Springer-Verlag, New York1992: 3312-3317Google Scholar). E. coli and C. glutamicum strains were cultivated at 37 and 30 °C, respectively, in Luria-Bertani medium (10 g/liter casein peptone, 5 g/liter yeast extract, 5 g/liter NaCl, pH 7.2; compounds were from Life Technologies) supplemented with the appropriate antibiotics (both strains: 20 mg/liter kanamycin sulfate; E. coli only: 50 mg/liter streptomycin sulfate, 12.5 mg/liter tetracycline/HCl, 100 mg/liter ampicillin). Isopropyl-β-d-thiogalactoside (1 mm) was added to induce recombinant genes that had been cloned under transcriptional control of the tac ortrc promoter.Table IStrains and plasmids used in this studyStrain/plasmidDescriptionSourceStrain H. influenzae DSM11121Rd, strain used for genome sequencing, identical to ATCC 51907Ref. 36Fleischman R.D. Adams M.D. White O. Clayton R.A. Kirkness E.F. Kerlavage A.R. Butt C.J. Tomb J.-F. Dougherty B.A. Merrick J.M. McKenney K. Sutton G. FitzHugh W. Fields C. Gocayne J.D. Scott J. Shirley R. Lui L.-I. Glodek A. Kelley J.M. Weidman J.F. Phillips C.A. Spriggs T. Hedblom E. Cotton M.D. Utterback T.R. Hanna M.C. Nguyen D.T. Saudek D.M. Brandon R.C. Fine L.D. Fritchman J.L. Fuhrman J.L. Geoghagen N.S.M. Gnehm C.L. McDonald L.A. Small K.V. Fraser C.M. Smith H.O. Venter J.C. Science. 1995; 269: 496-512Crossref PubMed Scopus (4694) Google Scholar H. influenzae I69Type b strain RM4066 butisn (I69 = gmhA), LPS of deep rough chemotypeRefs. 7Zwalen A. Rubin L.G. Connelly C.J. Inzana T.J. Moxon E.R. J. Infect. Dis. 1985; 152: 485-492Crossref PubMed Scopus (44) Google Scholar and 42Preston A. Maskell D. Johnson A. Moxon E.R. J. Bacteriol. 1996; 178: 396-402Crossref PubMed Google Scholar C. glutamicumR163Restriction-deficient mutant of C. glutamicumAS019Ref. 43Liebl W. Schein B. Behrens D. Krämer P. Dechema Biotechnology Conferences. 4. VCH Verlagsgemeinschaft, Weinheim, Germany1990: 323-327Google Scholar C. glutamicumR163/pJKB14Gene expression vector with the kdsB gene encoding CMP-Kdo synthethase of E. coliRef.13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar C. glutamicum R163/pJKB16pCB20 derivative, gene expression vector encoding waaA of E. coli W3110 (K-12 strain)Ref. 13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar C. glutamicum R163/pCB20Source of pCB20, cell extracts of the strain were used as a control in in vitroenzyme assaysRef. 13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar C. glutamicum R163/pCB23Gene expression vector encodingwaaA of H. influenzae I69 (waaA +HI-169)This work E. coli XL1BluerecA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [F′ proAB lacIqZΔM15 Tn10(tet′)] cloning host for plasmid constructionsRef.24Bullock W.O. Fernandez J.M. Short J.M. BioTechniques. 1987; 5: 376-379Google Scholar E. coli JC7623recB21 recC22 sbcC201 sbcB15 rpsL31 rfbD1Ref. 28Brozek K.A. Hosaka K. Robertson A.D. Raetz C.R.H. J. Biol. Chem. 1989; 264: 6956-6966Abstract Full Text PDF PubMed Google Scholar E. coli BLR(DE3)/pLysS/pJK2DE3 lysogen cmrtetr pJK2 (T7-RNA polymerase expression cassette withlpxK of E. coli)Ref. 20Garrett T.A. Kadrmas J.L. Raetz C.R.H. J. Biol. Chem. 1997; 272: 21855-21864Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar E. coli WBB01JC7623 but ΔwaaCF∷tet6Ref. 25Brabetz W. Müller-Loennies S. Holst O. Brade H. Eur. J. Biochem. 1997; 247: 716-724Crossref PubMed Scopus (107) Google Scholar E. coli WBB21JC7623 but ΔwaaCF∷(tet6 kdkA +HI-DSM11121)This work E. coli WBB22JC7623 but ΔwaaCF∷(tet6 kdkA +HI-DSM11121)waaA∷(waaA +HI-I69 kan)This work E. coli WBB32JC7623 but ΔwaaCF∷(tet6 kdkA +HI-I69)This work E. coliWBB34JC7623 but ΔwaaCF∷(tet6 kdkA +HI-I69)waaA∷(waaA +HI-I69 kan)This workPlasmid pCB20ori(p15A), ori(pSR1),tac promoter, kan r, E. coli/C. glutamicum shuttle plasmidRef.13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar pJSC2ori ts(pMAK705), insert waaA +E. coliRef.4Belunis C.J. Clementz T. Carty S.M. Raetz C.R.H. J. Biol. Chem. 1995; 270: 27646-27652Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar pCF-TETori(pMB1),bla, insert from E. coli with ΔwaaCF∷tet6Ref. 25Brabetz W. Müller-Loennies S. Holst O. Brade H. Eur. J. Biochem. 1997; 247: 716-724Crossref PubMed Scopus (107) Google Scholar pCB23pCB20 derivative with Ptac waaA +HI-I69This work pJKB49pJSC2 derivative waaA∷(Ptac waaA +HI-I69 kan)This work pJKB113pCF-TET derivative, insert from E. coliwith ΔwaaCF∷(tet6 kdkA +HI-DSM11121)This work pJKB113ApCF-TET derivative, insert from E. coliwith ΔwaaCF∷(tet6 kdkA +HI-I69)This work Open table in a new tab Most DNA procedures were done according to standard techniques (22Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Sedman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley & Sons, Inc., New York1987Google Scholar). All polymerase chain reactions (PCRs) used for cloning were performed withPfu DNA polymerase (Stratagene), which exhibits proofreading activity. The digoxygenin-11-dUTP system (Roche Diagnostics) was used for DNA labeling and detection in Southern experiments according to the instructions of the manufacturer. DNA sequence analysis of both strands of the cloned genes was done by cycle sequencing with fluorescent dye terminators and sequence-specific primers on an ABI 377 sequencing automat (Perkin-Elmer). The computer programs genworks (Intelligenetics), custalx1.8 (23Thompson J.D. Gibson T.J. Plewniak F. Jeanmougin F. Higgins D.G. Nucleic Acids Res. 1997; 24: 4876-4882Crossref Scopus (35602) Google Scholar), genedoc (K. B. Nicholas and H. B. Nicholas, Jr.), 2Genedoc: Analysis and Visualization of Genetic Variation is available on the World Wide Web. and blast (servers at the National Center for Biotechnology Information (NCBI)) were used to analyze DNA and deduced amino acid sequences. The primers HinfI (5′-ATATGGATCCGAATTTCTTTATGTGGCG-3′,BamHI site underlined and start codon in boldface type) andHinfII (5′-ATATCTGCAGCGTCATACATTGCGCTCC-3′,PstI site underlined and stop codon in boldface type) were used to amplify by PCR a 1297-bp waaA-encoding fragment from chromosomal DNA of H. influenzae I69. The amplificate was cut with BamHI and PstI and ligated with pCB20 (13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar), which had been linearized with the same restriction enzymes. The resultant construct, termed pCB23, encoded the Kdo transferase under transcriptional control of the tac promoter (Fig.1 A). The expression cassette with the tacpromoter, waaA of H. influenzae I69, and the kanamycin resistance gene (kan) of pCB20 was amplified from pCB23 with the primers W151 (5′-TTTTTCGTCGACGGTACCCGG-3′,SalI site underlined) and W152 (5′-AGAAAGTGGTCGACCCACGGTTGATG-3′, SalI site underlined), cut with SalI, and ligated into theSalI site of pJSC2 (4Belunis C.J. Clementz T. Carty S.M. Raetz C.R.H. J. Biol. Chem. 1995; 270: 27646-27652Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). The construct with the orientation of the inset shown in Fig. 1 B was selected and termed pJKB49. The recombinant strain was cultivated at 30 °C due to the temperature-sensitive origin of replication of pJSC2 (4Belunis C.J. Clementz T. Carty S.M. Raetz C.R.H. J. Biol. Chem. 1995; 270: 27646-27652Abstract Full Text Full Text PDF PubMed Scopus (79) Google Scholar). ThekdkA genes were amplified by PCR from chromosomal DNA ofH. influenzae DSM11121 and H. influenzae I69 using the primers HIK3 (5′-ATAACAACGTCATGACCCACCAATTCCAACAAG-3′,BspHI site underlined and start codon in boldface type) and HIK4 (5′-TCATAATTAGGTACC TTATTGATGATAAGCTGACGT-3′,KpnI site underlined and stop codon in boldface type). The 755-bp amplificates were cut with BspHI and KpnI and ligated with pCF-TET (25Brabetz W. Müller-Loennies S. Holst O. Brade H. Eur. J. Biochem. 1997; 247: 716-724Crossref PubMed Scopus (107) Google Scholar), which had been linearized withNcoI and KpnI. The resultant plasmids that encoded the kdkA genes of H. influenzae DSM11121 or H. influenzae I69 under transcriptional control of thetrc promoter were termed pJKB113 and pJKB113A, respectively (Fig. 1 C). In addition, PCR fragments that encodedwaaA or kdkA of H. influenzae DSM11121 and H. influenzae I69 with additional sequences around the open reading frames were amplified from chromosomal DNA and sequenced to determine the 5′- and 3′-ends of the genes including the binding sites of the cloning primers. For this purpose, the primer pairs HIWAAA1 (5′-AATGACCGTTCGTAATCGG-3′) and HIWAAA2 (5′-ATTGCAACGCTAATTGTAGGC-3′) were used in the case of waaA, and HIKDKA1 (5′-CTTCTGGGCTTTCAATGCC-3′) and HIKDKA2 (5′-TAAGGCATGACAGACATCGC-3′) were used in the case ofkdkA. The plasmids pJKB113 and pJKB113A were linearized withPvuII and ScaI and transformed into E. coli JC7623 (26Kushner S.R. Nagaishi H. Clark A.J. Proc. Natl. Acad. Sci. U. S. A. 1972; 69: 824-827Crossref PubMed Scopus (88) Google Scholar), which is recBC sbcBC and, thus, could be used to transfer the expression cassettes with the kdkAgenes of the H. influenzae strains DSM11121 and I69 together with the tetracycline resistance gene of pCF-TET to the chromosomalwaaCF locus by homologous recombination in one step (27Winans S.C. Elledge S.J. Krueger J.H. Walker G.C. J. Bacteriol. 1985; 161: 1219-1221Crossref PubMed Google Scholar). The corresponding strains derived from pJKB113 and pJKB113A were termedE. coli WBB21 and E. coli WBB32, respectively. Both recombinant strains were further transformed with pJKB49 that had been linearized with EcoRI and AflII to integrate the waaA gene from H. influenzae together with the kanamycin resistance gene (kan) into the chromosomalwaaA locus. The corresponding derivatives of E. coli WBB21 and E. coli WBB32 were termed E. coli WBB22 and E. coli WBB34, respectively. E. coli WBB01 (25Brabetz W. Müller-Loennies S. Holst O. Brade H. Eur. J. Biochem. 1997; 247: 716-724Crossref PubMed Scopus (107) Google Scholar), which is a ΔwaaCF derivative of strain E. coli JC7623, was used as a control recipient of linearized pJKB49 DNA. All chromosomal knockout mutants were characterized by Southern experiments, PCR, and DNA sequence analysis of the recombinant genes reamplified from the chromosomes. For this purpose, the primers LACP (5′-CGGCTCGTATAATGTGTGGA-3′) and WSB10R (5′-CGATTGTAAAACAGGCTGGC-3′) were used to amplify by PCR an 824-bpkdkA encoding fragment from chromosomal DNA of the recombinant strains that was not present in the wild type strain. The primers EC49A (5′-TGATCTGGATACGGCTCTGG-3′) and EC49B (5′-GCAGCAATCAGCGTAATACG-3′) were used to amplify from E. coli K-12 wild type DNA a 499-bp fragment around the singleSalI site that is located within the waaA gene (see Fig. 1 B). The same PCR revealed a single product of 3057 bp in the cases of E. coli WBB22 and E. coliWBB34. The preparation of cell extracts from recombinant strains and standard conditions for the Kdo transferase assay were performed as described (13Bode C.E. Brabetz W. Brade H. Eur. J. Biochem. 1998; 254: 404-412Crossref PubMed Scopus (28) Google Scholar). The protein content in cell extracts was determined using a commercial reagent (Bradford kit; Bio-Rad) and bovine serum albumin as a standard. The enzyme reaction mixture contained in a total volume of 20 μl of 50 mm Tris/HCl, pH 7.5, 10 mm MgCl2, 3.2 mm Triton X-100, 2 mm Kdo, 5 mmCTP, 0.1 mm synthetic bisphosphorylated tetraacyl lipid A precursor 406 (18Brabetz W. Lindner B. Brade H. Eur. J. Biochem. 2000; 167: 5458-5465Crossref Scopus (30) Google Scholar), 1.67 picokatal of CMP-Kdo synthetase and cell extracts from recombinant C. glutamicum strains (40 μg of protein). The in vitro tests were incubated up to 60 min at 37 °C and stopped by spotting 5 μl onto a TLC plate. Kinetic experiments were performed with 40-μl reaction mixtures from which 5-μl samples were withdrawn at different time points and stopped with 10 μl ice-cold ethanol before spotting onto TLC plates. TLC plates were developed with chloroform/pyridine/88% formic acid/water in a ratio of 30:70:16:10 (by volume). Radioactive [4′-32P]406 was synthesized from monophosphorylated lipid A precursor 405 as described (27Winans S.C. Elledge S.J. Krueger J.H. Walker G.C. J. Bacteriol. 1985; 161: 1219-1221Crossref PubMed Google Scholar) using recombinant lipid A 4′-kinase prepared from E. coli BLR(DE3)/pLysS/pJK2 (19Imoto M. Yoshimura H. Yamamoto M. Kusumoto S. Shiba T. Tetrahedron Lett. 1984; 26: 1545-1548Crossref Scopus (156) Google Scholar). The specific activity was adjusted with unlabeled 406 to approximately 15,000–20,000 cpm/nmol. Radioactive products were detected and quantified with a PhosphorImager equipped with the software ImageQuant (Molecular Dynamics). A standard enzyme reaction with radioactively labeled acceptor 406 and cell extract of C. glutamicum R163/pCB23 was scaled up to 800 μl and incubated for 1 h at 37 °C. The reaction products were separated by TLC as described above. The product Kdo-406 was identified by autoradiography and isolated from the TLC plate according to Brozek et al. (28Brozek K.A. Hosaka K. Robertson A.D. Raetz C.R.H. J. Biol. Chem. 1989; 264: 6956-6966Abstract Full Text PDF PubMed Google Scholar). The yield of Kdo-406 (39.7%) was calculated from its glucosamine content, which was determined according to published procedures (29Kaca W. de Jongh-Leuvenink J. Zähringer U. Brade H. Verhoef J. Sinnwell V. Carbohydr. Res. 1988; 179: 289-299Crossref PubMed Scopus (63) Google Scholar). Protein-free whole cell lysates were prepared as described (25Brabetz W. Müller-Loennies S. Holst O. Brade H. Eur. J. Biochem. 1997; 247: 716-724Crossref PubMed Scopus (107) Google Scholar). The samples were separated by SDS polyacrylamide gel electrophoresis (15% total acrylamide, 3.3% cross-linker) and silver-stained according to Tsai and Frasch (30Tsai C.-M. Frasch C.E. Anal. Biochem. 1982; 119: 115-119Crossref PubMed Scopus (2317) Google Scholar). Immunostaining of LPS after blotting onto nitrocellulose membranes was done according to Löbau et al. (15Löbau S. Mamat U. Brabetz W. Brade H. Mol. Microbiol. 1995; 18: 391-399Crossref PubMed Scopus (48) Google Scholar). Colony immunoblots were performed as described (25Brabetz W. Müller-Loennies S. Holst O. Brade H. Eur. J. Biochem. 1997; 247: 716-724Crossref PubMed Scopus (107) Google Scholar). The specificities of mAbs used in this study were described elsewhere (mAb A20 (31Rozalski A. Brade L. Kosma P. Appelmelck B.J. Krogmann C. Brade H. Infect. Immun. 1989; 57: 2645-2652Crossref PubMed Google Scholar); mAbs S42–16 and S42–21 (10Rozalski A. Brade L. Kosma P. Moxon R. Kusumoto S. Brade H. Mol. Microbiol. 1997; 23: 569-577Crossref PubMed Scopus (14) Google Scholar)) and are indicated in the legend to Fig. 5 and under “Results.” E. coli WBB22 was cultivated for 24 h at 30 °C in 6 liters of LB supplemented with tetracycline and kanamycin sulfate. The cells were centrifuged at 4 °C; washed 10 min with 200 ml of ice-cold 0.9% NaCl, 16 h with 200 ml of 96% ethanol, and twice for 3 h with 200 ml of acetone; and air-dried (yield: 1.1 g of dry cells/liter). Bacterial LPS was obtained by extraction of the dried bacteria with phenol-chloroform-petrolether as described (32Galanos C. Lüderitz O. Westphal O. Eur. J. Biochem. 1969; 9: 245-249Crossref PubMed Scopus (1369) Google Scholar) (yield: 4.1%). Analyses of GlcN, Kdo, and phosphate were performed as described (29Kaca W. de Jongh-Leuvenink J. Zähringer U. Brade H. Verhoef J. Sinnwell V. Carbohydr. Res. 1988; 179: 289-299Crossref PubMed Scopus (63) Google Scholar). LPS (50 mg, containing 30 μmol of GlcN) was de-O-acylated by mild hydrazinolysis as described (33Holst O. Thomas-Oates J.E. Brade H. Eur. J. Biochem. 1994; 222: 183-194Crossref PubMed Scopus (81) Google Scholar) (yield: 34 mg, 21 μmol of GlcN, 70%). De-O-acylated LPS (30 mg) was subjected to de-N-acylation by strong alkaline treatment as described (33Holst O. Thomas-Oates J.E. Brade H. Eur. J. Biochem. 1994; 222: 183-194Crossref PubMed Scopus (81) Google Scholar). After neutralization with hydrochloric acid and desalting on Sephacryl G-10 (Amersham Pharmacia Biotech) in pyridinium actetate, pH 4.0, a single fraction was obtained, which was analyzed by analytical high performance anion exchange chromatography (HPAEC) on a Dionex DX 300 chromatography system equipped with a Dionex CarboPac PA 1 column (4.5 × 250 mm), eluted at 1 ml/min using eluents A (water) and B (1 m sodium acetate, pH 6.0) and a linear gradient of 0–60% B in 70 min. NMR spectra were recorded on a solution of 5 mg of oligosaccharide (in 500 μl of D2O, 99.99%; Sigma) at pD 2.8 after three deuterium exchanges by evaporation at reduced pressure. All spectra were recorded at a temperature of 300 K using standard Bruker pulse programs. 1H NMR and1H,13C COSY spectra were recorded on a Bruker DRX600 spectrometer (600.13 MHz, measured relative to acetone, 2.225 ppm) equipped with a 5-mm multinuclear inverse probe head with Z-gradient. One-dimensional 13C NMR (90.6 MHz, relative to acetone, 31.07 ppm) and 31P spectra (145.8 MHz, relative to 85% H3PO4, 0.00 ppm) were recorded on a Bruker DPX360 spectrometer equipped with a 5-mm multinuclear inverse Z-gradient probe head. 13C and 31P NMR chemical shift assignments were achieved by inverse heteronuclear multiple quantum coherence (HMQC) experiments (34Bax A. Subaramaniam S. J. Magn. Reson. 1986; 67: 565-569Google Scholar) in phase-sensitive mode using states-time proportional phase incrementation (35Marion D. Ikura M. Tschudin R. Bax A. J. Magn. Reson. 1989; 85: 393-399Google Scholar). The1H,13C HMQC spectrum was recorded by sampling 2048 data points in t2 and 256 increments of 32 scans in t1. Garp decoupling was applied during t2. The spectral width was 110 ppm in F1 and 10 ppm in F2. Prior to Fourier transformation, a qsine window function was ap

Referência(s)