FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1993977508> ?p ?o ?g. }

• W1993977508 endingPage "30642" @default.
• W1993977508 startingPage "30634" @default.
• W1993977508 abstract "Polyketide-associated protein A5 (PapA5) is an acyltransferase that is involved in production of phthiocerol and phthiodiolone dimycocerosate esters, a class of virulence-enhancing lipids produced by Mycobacterium tuberculosis. Structural analysis of PapA5 at 2.75-Å resolution reveals a two-domain structure that shares unexpected similarity to structures of chloramphenicol acetyltransferase, dihydrolipoyl transacetylase, carnitine acetyltransferase, and VibH, a non-ribosomal peptide synthesis condensation enzyme. The PapA5 active site includes conserved histidine and aspartic acid residues that are critical to PapA5 acyltransferase activity. PapA5 catalyzes acyl transfer reactions on model substrates that contain long aliphatic carbon chains, and two hydrophobic channels were observed linking the PapA5 surface to the active site with properties consistent with these biochemical activities and substrate preferences. An additional α helix not observed in other acyltransferase structures blocks the putative entrance into the PapA5 active site, indicating that conformational changes may be associated with PapA5 activity. PapA5 represents the first structure solved for a protein involved in polyketide synthesis in Mycobacteria. Polyketide-associated protein A5 (PapA5) is an acyltransferase that is involved in production of phthiocerol and phthiodiolone dimycocerosate esters, a class of virulence-enhancing lipids produced by Mycobacterium tuberculosis. Structural analysis of PapA5 at 2.75-Å resolution reveals a two-domain structure that shares unexpected similarity to structures of chloramphenicol acetyltransferase, dihydrolipoyl transacetylase, carnitine acetyltransferase, and VibH, a non-ribosomal peptide synthesis condensation enzyme. The PapA5 active site includes conserved histidine and aspartic acid residues that are critical to PapA5 acyltransferase activity. PapA5 catalyzes acyl transfer reactions on model substrates that contain long aliphatic carbon chains, and two hydrophobic channels were observed linking the PapA5 surface to the active site with properties consistent with these biochemical activities and substrate preferences. An additional α helix not observed in other acyltransferase structures blocks the putative entrance into the PapA5 active site, indicating that conformational changes may be associated with PapA5 activity. PapA5 represents the first structure solved for a protein involved in polyketide synthesis in Mycobacteria. Mycobacterium tuberculosis produces polyketides, a complex family of lipids (1Bentley R. Bennett J.W. Annu. Rev. Microbiol. 1999; 53: 411-446Crossref PubMed Scopus (76) Google Scholar) that includes compounds associated with mycobacterial virulence. Up to 24 genes are predicted to encode proteins with polyketide synthase activity in the M. tuberculosis genome (H37Rv) (2Cole S.T. Brosch R. Parkhill J. Garnier T. Churcher C. Harris D. Gordon S.V. Eiglmeier K. Gas S. Barry III, C.E. Tekaia F. Badcock K. Basham D. Brown D. Chillingworth T. Connor R. Davies R. Devlin K. Feltwell T. Gentles S. Hamlin N. Holroyd S. Hornsby T. Jagels K. Krogh A. McLean J. Moule S. Murphy L. Oliver K. Osborne J. Quail M.A. Rajandream M.-A. Rogers J. Rutter S. Seeger K. Skelton J. Squares R. Squares S. Sulston J.E. Taylor K. Whitehead S. Barrell B.G. Nature. 1998; 393: 537-544Crossref PubMed Scopus (6502) Google Scholar). Although several of these and associated genes have been directly implicated with Mycobacterium virulence (3Cox J.S. Chen B. McNeil M. Jacobs W.R. Nature. 1999; 402: 79-83Crossref PubMed Scopus (611) Google Scholar, 4Camacho L.R. Ensergueix D. Perez E. Gicquel B. Guilhot C. Mol. Microbiol. 1999; 34: 257-267Crossref PubMed Scopus (517) Google Scholar, 5Dubey V.S. Sirakova T.D. Cynamon M.H. Kolattukudy P.E. J. Bacteriol. 2003; 185: 4620-4625Crossref PubMed Scopus (30) Google Scholar, 6Sirakova T.D. Dubey V.S. Cynamon M.H. Kolattukudy P.E. J. Bacteriol. 2003; 185: 2999-3008Crossref PubMed Scopus (52) Google Scholar, 7Sirakova T.D. Dubey V.S. Kim H.-J. Cynamon M.H. Kolattukudy P.E. Infect. Immun. 2003; 71: 3794-3801Crossref PubMed Scopus (48) Google Scholar, 8Rousseau C. Sirakova T.D. Dubey V.S. Bordat Y. Kolattukudy P.E. Gicquel B. Jackson M. Microbiology. 2003; 149: 1837-1847Crossref PubMed Scopus (58) Google Scholar, 9Mathur M. Kolattukudy P.E. J. Biol. Chem. 1992; 267: 19388-19395Abstract Full Text PDF PubMed Google Scholar, 10Kolattukudy P.E. Fernandes N.D. Azad A.K. Fitzmaurice A.M. Sirakova T.D. Mol. Microbiol. 1997; 24: 263-270Crossref PubMed Scopus (163) Google Scholar, 11Azad A.K. Sirakova T.D. Fernandes N.D. Kolattukudy P.E. J. Biol. Chem. 1997; 272: 16741-16745Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar), several others remain less well characterized. One such group includes a family of five polyketide-associated proteins (Paps) 1The abbreviations used are: Pap, polyketide-associated protein; PDIM, phthiocerol dimycocerosate ester and its congener; NCS, non-crystallographic symmetry; r.m.s.d., root mean square deviation; CAT, chloramphenicol acetyltransferase.1The abbreviations used are: Pap, polyketide-associated protein; PDIM, phthiocerol dimycocerosate ester and its congener; NCS, non-crystallographic symmetry; r.m.s.d., root mean square deviation; CAT, chloramphenicol acetyltransferase. that were suspected to encode activities associated with polyketide biosynthesis or transport (12Garnier T. Eiglmeier K. Camus J.C. Medina N. Mansoor H. Pryor M. Duthoy S. Grondin S. Lacroix C. Monsempe C. Simon S. Harris B. Atkin R. Doggett J. Mayes R. Keating L. Wheeler P.R. Parkhill J. Barrell B.G. Cole S.T. Gordon S.V. Hewinson R.G. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 7877-7882Crossref PubMed Scopus (748) Google Scholar, 13Cole S.T. Eiglmeier K. Parkhill J. James K.D. Thomson N.R. Wheeler P.R. Honore N. Garnier T. Churcher C. Harris D. Mungall K. Basham D. Brown D. Chillingworth T. Connor R. Davies R.M. Devlin K. Duthoy S. Feltwell T. Fraser A. Hamlin N. Holroyd S. Hornsby T. Jagels K. Lacroix C. Maclean J. Moule S. Murphy L. Oliver K. Quail M.A. Rajandream M.A. Rutherford K.M. Rutter S. Seeger K. Simon S. Simmonds M. Skelton J. Squares R. Squares S. Stevens K. Taylor K. Whitehead S. Woodward J.R. Barrell B.G. Nature. 2001; 409: 1007-1011Crossref PubMed Scopus (1366) Google Scholar, 14Brennan P.J. Vissa V.D. Lepr. Rev. 2001; 72: 415-428PubMed Google Scholar, 15de Crécy-Lagard V. Kelly J.W. Amino Acids, Peptides, Porphyrins, and Alkaloids. 4. Elsevier Science Publishers B.V., Amsterdam1999: 221-238Google Scholar). Additional Pap orthologs have been uncovered in polyketide synthase loci in Mycobacterium leprae, Mycobacterium bovis, and other Mycobacterium species (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar), suggesting that Paps may contribute activities to conserved pathways across Mycobacterium species. Phthiocerol dimycocerosate esters and their congeners, otherwise known as PDIMs, comprise a polyketide family that has been shown to be directly involved in mycobacterial virulence (3Cox J.S. Chen B. McNeil M. Jacobs W.R. Nature. 1999; 402: 79-83Crossref PubMed Scopus (611) Google Scholar, 4Camacho L.R. Ensergueix D. Perez E. Gicquel B. Guilhot C. Mol. Microbiol. 1999; 34: 257-267Crossref PubMed Scopus (517) Google Scholar). M. tuberculosis mutants deficient in PDIM production are attenuated in mice and PDIMs produced by M. leprae promote Schwann cell tropism (17Ng V. Zanazzi G. Timpl R. Talts J.F. Salzer J.L. Brennan P.J. Rambukkana A. Cell. 2000; 103: 511-524Abstract Full Text Full Text PDF PubMed Scopus (161) Google Scholar, 18Rambukkana A. Zanazzi G. Tapinos N. Salzer J.L. Science. 2002; 296: 927-931Crossref PubMed Scopus (164) Google Scholar). PDIM biosynthesis has been proposed to involve the activities of at least three polyketide synthase gene systems. The first and second systems include ppsA-E and mas, genes involved in phthiocerol/phthiodiolone synthesis and mycocerosic acid synthesis, respectively (10Kolattukudy P.E. Fernandes N.D. Azad A.K. Fitzmaurice A.M. Sirakova T.D. Mol. Microbiol. 1997; 24: 263-270Crossref PubMed Scopus (163) Google Scholar). The third includes pks15/1, a gene that has been associated with the incorporation of the phenolic group into mycoside PDIM variants (19Constant P. Perez E. Malaga W. Laneelle M.A. Saurel O. Daffe M. Guilhot C. J. Biol. Chem. 2002; 277: 38148-38158Abstract Full Text Full Text PDF PubMed Scopus (206) Google Scholar), and possibly other pks genes to produce early biosynthetic precursors (6Sirakova T.D. Dubey V.S. Cynamon M.H. Kolattukudy P.E. J. Bacteriol. 2003; 185: 2999-3008Crossref PubMed Scopus (52) Google Scholar, 8Rousseau C. Sirakova T.D. Dubey V.S. Bordat Y. Kolattukudy P.E. Gicquel B. Jackson M. Microbiology. 2003; 149: 1837-1847Crossref PubMed Scopus (58) Google Scholar, 21Sirakova T.D. Thirumala A.J. Dubey V.S. Sprecher H. Kolattukudy P.E. J. Biol. Chem. 2001; 276: 16833-16839Abstract Full Text Full Text PDF PubMed Scopus (140) Google Scholar). The genes associated with diesterification of phthiocerol and phthiodiolone to mycocerosic acid have remained unclear, but recent progress has been made through the genetic and functional characterization of M. tuberculosis PapA5, a Pap family member located within the PDIM synthesis gene cluster (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar). Deletion and complementation of the gene encoding PapA5 indicated that PapA5 was essential for PDIM production in M. tuberculosis, and although the involvement of PapA5 in the diesterification of phthiocerol and phthiodiolone could not be tested directly due to the unavailability of these compounds, PapA5 was assayed for CoA-dependent acyltransferase activities in an effort to define PapA5 substrate specificities for a variety of model lipid compounds. Taken together, these results suggested that papA5 encoded a protein capable of catalyzing acyl transfer chemistry. Although protein sequences of Pap family members include several amino acid motifs associated with other proteins that catalyze acyltransferase activities, the Pap family shares little sequence identity with acyltransferases outside of these regions. To elucidate structure-activity relationships for the Pap family of proteins, we determined the 2.75-Å crystal structure of PapA5 from M. tuberculosis. The structure reveals that PapA5 is related to the family of CoA-dependent acyltransferases. Further structural analysis combined with previously reported acyltransferase activities on defined lipid substrates suggests a model for PapA5 function. Purification of M. tuberculosis PapA5—PapA5-(1–422) was cloned, expressed, and purified from Escherichia coli as a N-terminal His6-Smt3 fusion protein (22Mossessova E. Lima C.D. Mol. Cell. 2000; 5: 865-876Abstract Full Text Full Text PDF PubMed Scopus (569) Google Scholar). The pET-based plasmid was transformed into E. coli BL21(DE3) CodonPlus RIL (Stratagene). A 5-liter culture was grown by fermentation at 37 °C to an A600 of 2, adjusted to 30 °C and 1 mm isopropyl-1-thio-β-d-galactopyranoside, and incubated for 4 h. Cells were harvested by centrifugation and resuspended in 20 mm Tris-HCl (pH 8.0), 350 mm NaCl, 10 mm imidazole, 20% sucrose, 1 mm β-mercaptoethanol, and 20 μg/ml lysozyme and sonicated. After insoluble material was removed by centrifugation, His6-Smt3-PapA5 was purified by metal affinity and gel filtration chromatography (Superdex 200). The His-Smt3 tag was removed by the Smt3-specific protease Ulp1, and PapA5 was further purified by gel filtration (Superdex 75). PapA5 was obtained at 10 mg/L of E. coli culture and appeared homogeneous by SDS-PAGE and Coommassie Blue staining. PapA5 was concentrated to 10 mg/ml, flash-frozen in liquid nitrogen, and stored at -80 °C. Crystallographic Analysis—PapA5 crystals were obtained by vapor diffusion against a well solution containing 5–10% polyethylene glycol 4000, 0.2 m ammonium acetate, 5% glycerol, 0.1 m sodium acetate (pH 5), and 20 mm dithiothreitol. Crystals were cryo-protected by addition of 15% glycerol. Crystals of native protein diffracted X-rays to 2.5 Å, although data were only processed to 2.75 Å due to problems associated with crystal mosaicity (P3121 a = b = 172.98 Å, c = 80.54 Å, α = β = 90°, γ = 120°). Native data were collected at National Synchrotron Light Source beamline X4A (Brookhaven, NY), and mercury data were collected from a crystal using a laboratory copper Kα source (Rigaku RU200) equipped with a confocal Osmic multilayer system and a Raxis-IV imaging plate detector. Data were reduced with DENZO, SCALEPACK (23Otwinowski Z. Minor W. Methods Enzymol. 1997; 276: 307-326Crossref PubMed Scopus (38526) Google Scholar), and CCP4 (24Project Collaborative Computational Acta Crystallogr. Sect. D Biol. Crystallogr. 1994; 50: 760-763Crossref PubMed Scopus (19748) Google Scholar). 2.75 Å phases were calculated with SOLVE and RESOLVE (25Terwilliger T.C. Berendzen J. Acta Crystallogr. Sect. D Biol. Crystallogr. 1999; 55: 849-861Crossref PubMed Scopus (3219) Google Scholar) using three mercury positions and non-crystallographic symmetry (NCS) averaging using the two molecules of PapA5 in the asymmetric unit. Subsequent electron density maps were traced by hand with the program O to produce atomic models for each PapA5 protomer (26Jones T.A. Zou J.Y. Cowan S.W. Kjeldgaard M. Acta Crystallogr. Sect. A. 1991; 47: 110-118Crossref PubMed Scopus (13009) Google Scholar). The model was released from NCS restraints, each monomer was rebuilt, and the model was refined to 2.75 Å with CNS (27Brunger A.T. Adams P.D. Clore G.M. DeLano W.L. Gros P. Grosse-Kunstleve R.W. Jiang J.S. Kuszewski J. Nilges M. Pannu N.S. Read R.J. Rice L.M. Simonson T. Warren G.L. Acta Crystallogr. Sect. D Biol. Crystallogr. 1998; 54: 905-921Crossref PubMed Scopus (16957) Google Scholar). The deposited model consists of resides 1–81, 94–175, 181–191, and 205–418 for monomer A and residues 3–81, 94–175, and 181–418 for monomer B. The model has excellent geometry with no Ramachandran outliers in disallowed regions (See Table I for crystallographic statistics).Table ICrystallographic data and refinement statistics Data indicated within parentheses indicate statistics for data in the highest resolution bin. mc, main chain; sc, side chain.NativeHg (thimerosal)Protein Data Bank ID1Q9JSourceNSLS X4ACuαWavelength (Å)0.97981.5418Resolution limits (Å)20–2.7520–2.8Unit cell (Å)a = b = 172.98 c = 80.54a = b = 172.27 c = 80.03Number of observations189,990236,093Number of reflections34,79859,221aData completeness treats Bijvoët mates independentlyCompleteness (%)96.9 (85.3)91.5 (81.8)Mean I/σI14.9 (1.8)9.9 (1.0)Rmerge on IbRmerge = Σhkl ΣI|I(hkl)i – 〈I(hkl)〉|/ΣhklΣi 〈I(hkl)I〉7.3 (47.2)9.0 (57.2)Cut-off criteria–2.0–0.5Figure of merit0.26/0.47(SOLVE/RESOLVE)Refinement statisticsResolution limits (Å)20–2.75Number of reflections33,906Completeness93.7Cut-off criteria0PapA5 atoms6011Number of water atoms226RcrystcRcryst = Σhkl|Fo(hkl) – Fc(hkl)|/Σhkl|Fo(hkl)|, where Fo and Fc are observed and calculated structure factors, respectively0.236 (0.364)Rfree (5% of data)0.295 (0.405)Bonds (Å)0.008Angles (°)1.30Bfactor (mc/sc in Å2)2.66/3.58a Data completeness treats Bijvoët mates independentlyb Rmerge = Σhkl ΣI|I(hkl)i – 〈I(hkl)〉|/ΣhklΣi 〈I(hkl)I〉c Rcryst = Σhkl|Fo(hkl) – Fc(hkl)|/Σhkl|Fo(hkl)|, where Fo and Fc are observed and calculated structure factors, respectively Open table in a new tab Overview of the PapA5 Structure—Purified recombinant PapA5 containing amino acids 1–422 was purified and crystallized. Crystals of PapA5 belonged to space group P3121 and contained two independent monomers per asymmetric unit. Phases were calculated to 2.75 Å using 2-fold NCS averaging with data obtained from a native protein crystal and a native crystal into which thimerosal was soaked (see “Experimental Procedures”; Table I). Both PapA5 monomers contain segments of polypeptide that did not have sufficient electron density to permit model building. These regions include residues 82–93, 176–180, 192–204, and 419–422 for monomer A and 1–2, 82–93, 176–180, and 419–422 for monomer B. The overall average Bfactor for the final coordinate set was ∼68 Å2. The termini that demarcate the disordered segments (marked by asterisks in Fig. 1) had Bfactor values of nearly twice that value, suggesting that segments not observed in the electron density maps were due to thermal motion in the crystal lattice. Despite the high overall Bfactor and disordered segments, the final model was successfully refined without NCS restraints to an Rfactor of 23.6 and Rfree of 29.5 with excellent geometry and no outliers in the Ramachandran plot. Monomer B will be referred to in subsequent discussions as few differences were observed between monomers, and monomer B contained a larger number of ordered amino acids. The PapA5 structure can most easily be described by dividing the protein into two domains (Fig. 1). Domain 1 is composed of secondary structural elements that include β-strands 1–8 and 13 and α-helices A through D. Domain 2 includes β-strains 9–12 and 14–15 and α-helices E through I. Domains 1 and 2 are self-contained with a few noted exceptions. Domain 1 includes β13, a strand that emanates from domain 2 to complete the four-stranded anti-parallel β-sheet in domain 1 (β6, β7, β2, and β13). In addition, a loop from domain 2 between β-strands 10 and 11 extends into domain 1 and contacts portions of helix C and D. A large crossover loop is also observed between helix D and β-strand 8 that spans nearly 50 Å between the two domains. Monomer B amino acids 192–204 within the crossover loop have Bfactor values nearly twice that of the average model Bfactor, while the same region in monomer A is disordered and not present in the electron density maps. Despite these interdomain contacts, the connectivity between domains would not restrict movements of domain 1 and 2 with respect to one another. Domains 1 and 2 are structurally related and can be aligned to within 4.1 Å r.m.s.d. over 104 amino acids with 7% sequence identity. While similar, domain 1 contains the only known catalytic amino acid residues (His124 and Asp128) that have been directly implicated in PapA5 activity (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar). His124 and Asp128 are located between strand 7 and helix C in the interface between domains 1 and 2 (Figs. 1 and 2). PapA5 Is Related to CoA-dependent Acyltransferases—A structural alignment between PapA5 and the Protein Data Bank using DALI shows that PapA5 contains structural and sequence motifs that are characteristic of the CoA-dependent acyltransferase family (Fig. 2) (28Holm L. Sander C. J. Mol. Biol. 1993; 233: 123-138Crossref PubMed Scopus (3561) Google Scholar). In rank order, proteins that could be aligned to PapA5 include the condensation domain from VibH (Protein Data Bank code 1l5a; 336 amino acids aligned to 4.0 Å r.m.s.d. with 11% sequence identity; Z-score 23.1), carnitine acetyltransferase (Protein Data Bank code 1ndf; 285 amino acids aligned to 3.8 Å r.m.s.d. with 10% sequence identity; Z-score 12.1), chloramphenicol acetyltransferase (CAT) (Protein Data Bank code 3cla; 127 amino acids aligned to 3.9 Å r.m.s.d. with 12% sequence identity; Z-score 6.9), and dihydrolipoyl transacetylase (Protein Data Bank code 1eaf; 108 amino acids aligned to 2.9 Å r.m.s.d. with 17% sequence identity; Z-score 5.6). As stated above, domains 1 and 2 from Papa5 are related to each other, so each can be aligned to a single protomer from chloramphenicol acetyltransferase. In addition, PapA5 domains 1 and 2 are oriented in a similar manner to that observed for two of the three chloramphenicol acetyltransferase protomers as observed in the intact chloramphenicol acetyltransferase trimer (29Leslie A.G.W. Moody P.C. Shaw W.V. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4133-4137Crossref PubMed Scopus (133) Google Scholar). Although structural alignments show PapA5 domain 2 to be more similar based on the number of amino acids that could be aligned, the alignments to domain 1 reveal higher sequence identity between structures, including the known HHX3DG catalytic amino acid motif that is conserved and observed in many CoA-dependent acyltransferase family members (Fig. 2). Chloramphenicol acetyltransferase catalyzes CoA-dependent acetyl transfer in a reaction that is dependent on the second conserved histidine (His195) in the active site HHX3DG sequence motif (29Leslie A.G.W. Moody P.C. Shaw W.V. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4133-4137Crossref PubMed Scopus (133) Google Scholar). His195 has been proposed to be a general base that promotes deprotonation of the chloramphenicol hydroxyl prior to the transfer of the acetyl group, and mutation of this residue results in a severely defective enzyme. While the aspartic acid has also been shown to be critical for activity, it appears to play a structural role in the organization of the active site. His124 and Asp128 are the corresponding histidine and aspartic acid in the PapA5 sequence and structure (Figs. 2 and 3), and as observed for chloramphenicol acetyltransferase, His124 is essential for PapA5 CoA-dependent acyltransferase activity (see below; Ref. 16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar). Although conserved, the second histidine in the HHX3DG motif does not perform similar catalytic roles in all CAT family members insomuch as mutation of this residue in the yeast dihydrolipoyl transacetylase results in no observable catalytic defect (30Niu X.D. Stoops J.K. Reed L.J. Biochemistry. 1990; 29: 8614-8619Crossref PubMed Scopus (31) Google Scholar). PapA5 exhibits more structural similarity to VibH and carnitine acetyltransferase, since PapA5, VibH, and carnitine acetyltransferase each contain two tandem CAT-like domains with one active site located within the N-terminal CAT-like domain (31Keating T.A. Marshall C.G. Walsh C.T. Keating A.E. Nat. Struct. Biol. 2002; 9: 522-526PubMed Google Scholar, 32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). Carnitine acetyltransferase catalyzes the acyl transfer between carnitine and acetyl-CoA utilizing a similar mechanism to that proposed for CAT. In addition, the structural analysis of carnitine acetyltransferase and respective ligand complexes combined with previous biochemical data support a general base mechanism for His343 in deprotonation of the carnitine hydroxyl group prior to acyl transfer (32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). VibH belongs to a large family of condensation domains that are associated with non-ribosomal peptide synthetases, large multidomain enzymes that catalyze the synthesis of a number of compounds such as antibiotics and virulence factors (33Marahiel M.A. Stachelhaus T. Mootz H.D. Chem. Rev. 1997; 97: 2651-2674Crossref PubMed Scopus (907) Google Scholar, 34Quadri L.E. Keating T.A. Patel H.M. Walsh C.T. Biochemistry. 1999; 38: 14941-14954Crossref PubMed Scopus (116) Google Scholar, 35Quadri L.E. Mol. Microbiol. 2000; 37: 1-12Crossref PubMed Scopus (81) Google Scholar). VibH catalyzes amide bond formation in the synthesis of vibriobactin (36Keating T.A. Marshall C.G. Walsh C.T. Biochemistry. 2000; 39: 15513-15521Crossref PubMed Scopus (95) Google Scholar, 37Keating T.A. Marshall C.G. Walsh C.T. Biochemistry. 2000; 39: 15522-15530Crossref PubMed Scopus (106) Google Scholar). VibH also contains a conserved HHX3DG catalytic amino acid motif, although structural and mutational analysis revealed that VibH utilizes a mechanism distinct from CAT and carnitine acetyltransferase in that mutation of the second histidine did not result in severe catalytic defects but mutation of the Asp residue did. These data suggest that the tandem CAT-like domain architecture can be utilized in alternative ways to achieve a variety of chemical reactions. CoA-dependent acyltransferase family members have been categorized on the structural level in the SCOP data base by virtue of their oligomeric state (38Hubbard T.J. Murzin A.G. Brenner S.E. Chothia C. Nucleic Acids Res. 1997; 25: 236-239Crossref PubMed Scopus (222) Google Scholar). Some CAT family members such as chloramphenicol acetyltransferase and dihydrolipoyl transacetylase are oligomeric and form their respective active sites in the intersubunit interfaces between protomers. Carnitine acetyltransferase and VibH are monomeric, but contain two tandem CAT-like domains. In both instances, VibH and carnitine acetyltransferase share similarity with the CAT intersubunit active site organization insomuch as each has its active site positioned within the interface between the two tandem CAT-like domains. PapA5 also contains two CAT-like domains and is organized in a similar manner to that observed for VibH and carnitine acetyltransferase (Fig. 1). Analysis of the aligned structures for PapA5 and carnitine acetyltransferase revealed similar locations for the active site histidine and aspartic acid residues located within domain 1 between domains 1 and 2 (Fig. 4). The carnitine acetyltransferase structure utilized for this alignment also included the substrate carnitine (32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar), and comparison of the PapA5 and carnitine acetyltransferase active site clefts revealed a deep substrate cleft for carnitine acetyltransferase (Fig. 4D), whereas the analogous PapA5 cleft is occluded by helix H (Fig. 4C). Helix H is unique to the Pap family (Fig. 2), and its potential role in substrate coordination is discussed further below. PapA5 Active Site and Substrate Selectivity—PapA5 has been proposed to catalyze the diesterification of phthiocerol and phthiodiolone with mycocerosate, possibly through a mechanism that is dependent on the activation of mycocerosic acids as thioesters (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar). The possible mechanisms by which PapA5 might participate in this reaction were previously explored by measuring PapA5 acyltransferase activity using palmitoyl-CoA and several model substrates that included short-, medium-, and long-chain alcohols, diols, hydroxy esters, acids, amines, and thiols (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar). A subset of those substrates tested is depicted in Fig. 5. PapA5 exhibited a preference for saturated medium chain alcohols in reactions that were dependent on the presence of amino acid residues His124 and Asp128, suggesting that PapA5 utilizes a similar acyl transfer mechanism to that observed for several CAT family members including carnitine acetyltransferase. While CAT and dihydrolipoyl transacetylase utilize intersubunit interfaces to interact with respective ligands, large solvent-exposed channels were observed between the two tandem CAT-like domains in both VibH and carnitine acetyltransferase (Fig. 4, C and D) (31Keating T.A. Marshall C.G. Walsh C.T. Keating A.E. Nat. Struct. Biol. 2002; 9: 522-526PubMed Google Scholar, 32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). To gain insight into how PapA5 might organize its active site with respect to these other CAT family members, the PapA5 active site was superimposed to crystal structures of CAT, dihydrolipoyl transacetylase, and carnitine acetyltransferase to enable modeling of chloramphenicol (29Leslie A.G.W. Moody P.C. Shaw W.V. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4133-4137Crossref PubMed Scopus (133) Google Scholar), carnitine (32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar), and CoA (32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar, 39Mattevi A. Obmolova G. Schulze E. Kalk K.H. Westphal A.H. de Kok A. Hol W.G.J. Science. 1992; 255: 1544-1550Crossref PubMed Scopus (228) Google Scholar) into the PapA5 active site (Fig. 3B). Only CoA and chloramphenicol are depicted in Fig. 3B, since the ligand positions observed in carnitine acetyltransferase superimpose well to a first approximation with those ligands observed in CAT and dihydrolipoyl transacetylase (32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). A similar modeling exercise was undertaken using VibH and respective ligands from CAT and dihydrolipoyl transacetylase (31Keating T.A. Marshall C.G. Walsh C.T. Keating A.E. Nat. Struct. Biol. 2002; 9: 522-526PubMed Google Scholar). These alignments revealed both ligand binding sites to be accessible to solvent within the VibH structure, suggesting that VibH utilizes similar surfaces and substrate clefts to interact with its substrates. Experimental structures of carnitine acetyltransferase in complex with carnitine and CoA also showed a similar arrangement of ligands, suggesting that it too utilizes the same solvent-exposed channels to bind and coordinate respective ligands (32Jogl G. Tong L. Cell. 2003; 112: 113-122Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar). Inspection of the modeled ligands in the PapA5 active site shows the CoA ligand coordinated between β-strands 10 and 12 in an orientation similar to that observed in all CAT-CoA complexes, suggesting that PapA5 would interact with CoA ligands in a similar manner (Figs. 1 and 3B). The modeled positions of carnitine and chloramphenicol within the PapA5 active site show the respective hydroxyl moieties directly over catalytic His124 in PapA5, suggesting that the basic mechanisms employed in CAT activity are likely conserved in the PapA5 structure (Fig. 3B, carnitine shown in Fig. 4). Despite proper placement of the hydroxyl group, both carnitine and chloramphenicol encounter significant steric clashes with amino acid residues emanating from helix H, namely Phe327 and Phe331 (Fig. 3B). Helix H is unique to PapA5 and is not observed in either VibH, carnitine acetyltransferase, or other CAT family members (Fig. 2). Helix H effectively blocks access to one of the solvent exposed channels into the active sites observed in either VibH or carnitine acetyltransferase (Figs. 1, 3, and 4). Further inspection of the PapA5 molecular surface reveals two channels that lead into the PapA5 active site (Fig. 6). Channel 1 is ∼20 Å deep and is fully exposed via a solvent channel to the exterior of PapA5 (Figs. 3 and 6), whereas channel 2 is constricted by helix H and is ∼15 Å deep, indicating that if PapA5 were to utilize channel 2 in a manner reminiscent to that observed in carnitine acetyltransferase, it must undergo conformational changes to move helix H away from the channel to enable substrate interaction (Fig. 4). Interestingly, PapA5 exhibited substrate preference for octanol in acyltransferase activity assays in addition to other observed activities on substrates with aliphatic chains of ∼10 Å in length (Fig. 5; Ref. 16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar). While the measured depth of either channel would accommodate the preferred substrate lengths, they differ with respect to hydrophobicity and sequence conservation (Fig. 6, C and D). Channel 1 includes few conserved residues between Pap family members but is very hydrophobic in nature, while channel 2 includes several conserved residues but is slightly more hydrophilic in character. The preferred substrate lengths exhibited by PapA5 in vitro suggest a possible mode of interaction with phthiocerol and phthiodiolone, the proposed in vivo substrates for PapA5 (Fig. 5). Phthiocerol and phthiodiolone include two hydroxyl moieties located at positions 9 and 11 along the aliphatic chain (Fig. 5C). The approximate length (9–11 carbon units) of the remaining aliphatic chain is architecturally similar in many respects to that observed for several of the preferred model substrates, namely octanol (Fig. 5, compare A and C). If analogous to carnitine acetyltransferase and chloramphenicol acetyltransferase, channel 2 would provide the most likely binding site for octanol and the analogous portions of either phthiocerol and phthiodiolone. In addition, the long aliphatic chains associated with the remaining portions of either phthiocerol and phthiodiolone could be accommodated in channel 1 (17–21 carbon units; Figs. 5C and 6). Phthiocerol or phthiodiolone are not commercially available or easily purified from natural sources, so it is currently implausible to obtain the relevant complexes to enable a more detailed study of PapA5 in complex with its physiological ligands, a necessary step to provide the basis for development of structure-based inhibitors of this enzyme. Although natural ligand complexes are currently beyond the scope of this study, we do plan to obtain complexes between PapA5 and some of the model compounds previously reported (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar) and represented in Fig. 5. However, the structural and biochemical characterization of PapA5 combined with the structures of carnitine acetyltransferase and VibH suggest putative roles for the catalytic residues observed in PapA5. In addition, structural and functional similarity between these protein families has likely identified the surfaces and channels utilized by PapA5 in its interactions with respective substrates, the exact details of which await further investigation. It has been previously noted that the Pap family shares weak similarity to Rif20 (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar), a gene encoded within the rifamycin gene cluster (40Floss H.G. Yu T.W. Curr. Opin. Chem. Biol. 1999; 3: 592-597Crossref PubMed Scopus (63) Google Scholar). The PapA5 structure and the structure-based sequence alignment between PapA5 and Rif20 and the conserved catalytic elements observed in these proteins support our earlier speculation (16Onwueme K.C. Ferreras J.A. Buglino J. Lima C.D. Quadri L.E. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 4608-4613Crossref PubMed Scopus (79) Google Scholar) that Rif20 encodes a protein with similar catalytic properties to that observed for PapA5 and is responsible for catalyzing the as yet unidentified acyltransferase activity required for C25 O-acetylation during rifamycin biosynthesis. We thank the staff of beamline X4A at the National Synchrotron Light Source." @default.
• W1993977508 created "2016-06-24" @default.
• W1993977508 creator A5013774625 @default.
• W1993977508 creator A5015599531 @default.
• W1993977508 creator A5060190581 @default.
• W1993977508 creator A5060714280 @default.
• W1993977508 creator A5078363786 @default.
• W1993977508 date "2004-07-01" @default.
• W1993977508 modified "2023-10-16" @default.
• W1993977508 title "Crystal Structure of PapA5, a Phthiocerol Dimycocerosyl Transferase from Mycobacterium tuberculosis" @default.
• W1993977508 cites W1514341457 @default.
• W1993977508 cites W1539796472 @default.
• W1993977508 cites W1614868353 @default.
• W1993977508 cites W1965277349 @default.
• W1993977508 cites W1971337268 @default.
• W1993977508 cites W1983544779 @default.
• W1993977508 cites W1985674001 @default.
• W1993977508 cites W1995017064 @default.
• W1993977508 cites W2001641653 @default.
• W1993977508 cites W2005352397 @default.
• W1993977508 cites W2013083986 @default.
• W1993977508 cites W2014714736 @default.
• W1993977508 cites W2018990299 @default.
• W1993977508 cites W2022058405 @default.
• W1993977508 cites W2030843417 @default.
• W1993977508 cites W2035214478 @default.
• W1993977508 cites W2044690664 @default.
• W1993977508 cites W2049521571 @default.
• W1993977508 cites W2049726411 @default.
• W1993977508 cites W2052791500 @default.
• W1993977508 cites W2073790895 @default.
• W1993977508 cites W2079901091 @default.
• W1993977508 cites W2085774588 @default.
• W1993977508 cites W2094761728 @default.
• W1993977508 cites W2096520521 @default.
• W1993977508 cites W2106315897 @default.
• W1993977508 cites W2106457742 @default.
• W1993977508 cites W2108527886 @default.
• W1993977508 cites W2117077088 @default.
• W1993977508 cites W2128026302 @default.
• W1993977508 cites W2131256962 @default.
• W1993977508 cites W2137631963 @default.
• W1993977508 cites W2152400582 @default.
• W1993977508 cites W2153390332 @default.
• W1993977508 cites W2155655994 @default.
• W1993977508 cites W2158023121 @default.
• W1993977508 cites W2161998885 @default.
• W1993977508 cites W2164217177 @default.
• W1993977508 doi "https://doi.org/10.1074/jbc.m404011200" @default.
• W1993977508 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/15123643" @default.
• W1993977508 hasPublicationYear "2004" @default.
• W1993977508 type Work @default.
• W1993977508 sameAs 1993977508 @default.
• W1993977508 citedByCount "64" @default.
• W1993977508 countsByYear W19939775082012 @default.
• W1993977508 countsByYear W19939775082013 @default.
• W1993977508 countsByYear W19939775082014 @default.
• W1993977508 countsByYear W19939775082015 @default.
• W1993977508 countsByYear W19939775082016 @default.
• W1993977508 countsByYear W19939775082018 @default.
• W1993977508 countsByYear W19939775082019 @default.
• W1993977508 countsByYear W19939775082020 @default.
• W1993977508 countsByYear W19939775082021 @default.
• W1993977508 countsByYear W19939775082022 @default.
• W1993977508 countsByYear W19939775082023 @default.
• W1993977508 crossrefType "journal-article" @default.
• W1993977508 hasAuthorship W1993977508A5013774625 @default.
• W1993977508 hasAuthorship W1993977508A5015599531 @default.
• W1993977508 hasAuthorship W1993977508A5060190581 @default.
• W1993977508 hasAuthorship W1993977508A5060714280 @default.
• W1993977508 hasAuthorship W1993977508A5078363786 @default.
• W1993977508 hasBestOaLocation W19939775081 @default.
• W1993977508 hasConcept C104317684 @default.
• W1993977508 hasConcept C115624301 @default.
• W1993977508 hasConcept C142724271 @default.
• W1993977508 hasConcept C181199279 @default.
• W1993977508 hasConcept C185592680 @default.
• W1993977508 hasConcept C2776376580 @default.
• W1993977508 hasConcept C2777975735 @default.
• W1993977508 hasConcept C2781069245 @default.
• W1993977508 hasConcept C33288867 @default.
• W1993977508 hasConcept C55493867 @default.
• W1993977508 hasConcept C71924100 @default.
• W1993977508 hasConcept C8010536 @default.
• W1993977508 hasConcept C86803240 @default.
• W1993977508 hasConcept C89423630 @default.
• W1993977508 hasConcept C91965660 @default.
• W1993977508 hasConceptScore W1993977508C104317684 @default.
• W1993977508 hasConceptScore W1993977508C115624301 @default.
• W1993977508 hasConceptScore W1993977508C142724271 @default.
• W1993977508 hasConceptScore W1993977508C181199279 @default.
• W1993977508 hasConceptScore W1993977508C185592680 @default.
• W1993977508 hasConceptScore W1993977508C2776376580 @default.
• W1993977508 hasConceptScore W1993977508C2777975735 @default.
• W1993977508 hasConceptScore W1993977508C2781069245 @default.
• W1993977508 hasConceptScore W1993977508C33288867 @default.
• W1993977508 hasConceptScore W1993977508C55493867 @default.
• W1993977508 hasConceptScore W1993977508C71924100 @default.

• next